FEUDAL
ENGLAND

HISTORICAL STUDIES ON
THE ELEVENTH AND TWELFTH
CENTURIES

BY

J. H. ROUND

FIRST PUBLISHED 1895
Second impression 1909
Third impression 1909

PREFACE

The present work is the outcome of a wish expressed to me from more than one quarter that I would reprint in a collected form, for the convenience of historical students, some more results of my researches in the history of the eleventh and twelfth centuries. But to these I have added, especially on Domesday, so much which has not yet seen the light, that the greater portion of the work is new, while the rest has been in part re-written. The object I have set before myself throughout is either to add to or correct our existing knowledge of facts. And for this I have gone in the main to records, whether in manuscript or in print. It is my hope that the papers in this volume may further illustrate the value of such evidence as supplementing and checking the chroniclers for what is still, in many respects, an obscure period of our history.

As a foreign scholar has felicitously observed:

Je lis avec plaisir le chroniqueur qui nous raconte les événements de son époque. Les détails anecdotiques, les traits piquants dont son œuvre est parsémée font mes délices. Mais comment saurai-je s'il dit la vérité si les pages qu'il me présente ne sont pas un roman de pure imagination? Dans les chartes, au contraire, tout est authentique, certain, précis, indubitable. Leur témoignage est contradictoirement établi, sous le contrôle de la partie adverse, avec l'approbation et la reconaissance de l'autorité souveraine, en présence d'une imposante assemblée de notables qui apposent leur signature. C'est la plus pure de toutes les sources où il soit possible de puiser un renseignement historique.¹

An instance in point will be found in the paper on 'Richard the First's change of seal'.

A collective title for a series of studies covering the period 1050-1200, is not by any means easy to find. But dealing as they do so largely with the origins of 'Feudal England', I have ventured to give them this title, which may serve, I hope, to emphasize my point that the feudal element introduced at the Conquest had a greater influence on our national institutions than recent historians admit.² Even Domesday Book has its place in the study of feudalism, rearranging, as it does, the Hundred and the Vill under Fiefs and 'Manors'.

To those in search of new light on our early mediaeval history, I commend the first portion of this work, as setting forth, for their careful consideration, views as evolutionary on the Domesday hide and the whole system of land assessment as on the actual introduction [pg 10] of the feudal system into England. Although I have here brought into conjunction my discovery that the assessment of knight-service was based on a five-knights unit, irrespective of area or value, and my theory that the original assessment of land was based on a five-hides unit, not calculated on area or value, yet the two, one need hardly add, are, of course, unconnected. The one was an Anglo-Saxon system, and, as I maintain, of early date; the other was of Norman introduction, and of independent origin. My theories were formed at different times, as the result of wholly separate investigations. That of the five-hides unit was arrived at several years ago, but was kept back in the hope that I might light on some really satisfactory explanation of the phenomena presented. The solution I now propound can only be deemed tentative. I would hope, however, that the theories I advance may stimulate others to approach the subject, and, above all, that they may indicate to local students, in the future, the lines on which they should work and the absolute need of their assistance.

Perhaps the most important conclusion to which my researches point is that Domesday reveals the existence of two separate systems in England, co-extensive with two nationalities, the original five hides of the 'Anglo-Saxon' in the south, and the later six carucates of the 'Danish' invaders in the north.³

No one, I may add, is better qualified to carry further these inquiries than Prof Maitland, whose brilliant pen has illumined for us the origins of English law. Himself engaged on the study of Domesday, he kindly offered to withhold his conclusions until my work should have appeared.⁴

Among the fresh points here discussed in connection with Domesday Book will be found the composition of the juries by whom the returns were made, the origin and true character of the Inquisitio Eliensis, and the marked difference of the two volumes compiled from the Domesday returns.

Of the six early surveys dealt with in conjunction with Domesday, I would call attention to that of Leicestershire as having, it would seem, till now remained absolutely unknown. It has long been a wish of mine to deal with these surveys,⁵ not only as belonging to a period for which we have no records, but also as illustrating Domesday Book. In 'The Knights of Peterborough' will be found some facts relating to Hereward 'the Wake', which seem to have eluded Mr Freeman's investigations, and even those of Mr Tout.

In case it should suggest itself that these papers, and some in the [pg 11] other portion of the work dwell at undue length on unimportant points, I would observe that apart from the fact that even small points acquire a relative importance from our scanty knowledge of the time, there are cases in which their careful investigation may lead to unforeseen results. At the last anniversary of the Royal Society, Lord Kelvin quoted these words from his own presidential address in 1871:

Accurate and minute measurement seems to the non-scientific imagination a less lofty and dignified work than looking for something new. But nearly all the grandest discoveries of science have been but the rewards of accurate measurement and patient, long continued labour in the minute sifting of numerical results.

The same principle applies to the study of institutional history. Whether we are dealing with military service, with the land, with finance, or with the king's court, 'the minute sifting' of facts and figures is the only sure method by which we can extend knowledge.

To those who know how few are the original authorities for the period, and how diligently these have been explored and their information exhausted, the wonder will be not so much that there is little, as that there was anything at all yet left to discover.

In a work dealing with the history of the eleventh and twelfth centuries, a writer must inevitably find himself at times dealing with the same subjects as the late Professor Freeman. Without in any way disparaging the genius of that eminent man, one may deem it a duty to correct the errors into which he fell, and conscientiously to combat, as an obstinate and mischievous superstition, the conviction of his pre-eminent accuracy and authority on matters of fact. It would be far pleasanter to dwell only on his merits; but when one finds that, in spite of the proofs I have been producing for years, Mr Herbert Fisher, representing the Oxford school of history, can still declare Mr Freeman to have reached 'the highest standard of scholarly exactitude',⁶ it is evident that the works of the Regius Professor are still surrounded by a false glamour, and that one must further expose his grave liability to error. I cannot suppose that any competent scholar who may carefully peruse this work will in future venture to deny that, in spite of his many and his splendid gifts, Mr Freeman was as liable as any of us to error, or that however laudable his intentions, he was capable of precisely the same inaccuracy and occasionally of the same confusion as he denounced so bitterly in others.

It is, indeed, my contention, as I have already explained,⁷ that to these denunciations of the errors of others is largely due the [pg 12] conviction of Mr Freeman's supreme accuracy. The question raised may seem to affect the whole method of history, for if, as has been said, it is the argument of the scientific historian that we ought to prefer accuracy of fact to charm of presentment and to literary style, the proof that his method fails to save him from erring like any 'literary' historian strikes at the root of his whole contention.

Yet it is not the scientific method, but its prophet himself that was at fault.

Although I am here only concerned with inaccuracy in matters of fact, I would guard myself against the retort that, at least, Mr Freeman's errors are of little consequence as compared with that obliquity of vision which led Mr Froude, at all hazards, to vindicate Henry the Eighth. Without insisting on an absolute parallel, I trace a resemblance even here. Just as his bias against the Roman church led Mr Froude to vindicate Henry in order to justify the breach with Rome, so Mr Freeman's passion for democracy made him an advocate on behalf of Harold, as 'one whose claim was not drawn only from the winding-sheet of his fathers'. I have elsewhere maintained, as to Harold's election 'by the free choice of a free people', that Mr Freeman's undoubted perversion of the case at this 'the central point' of his history, gravely impairs his narrative of the Conquest, because its success, and even its undertaking, can actually be traced to that election.⁸ Unless we realize its disastrous effect on the situation both at home and abroad, we cannot rightly understand the triumph of the Duke's enterprise.

It had been my hope, in the present work, to have avoided acute controversy, but the attitude adopted, unfortunately, by the late Professor's champions has rendered that course impossible. One can but rejoice that his accuracy should find strenuous defenders, as it removes the reluctance one would otherwise feel in continuing to criticize it now. A case is doubly proved when proved in the teeth of opposition. But one expects that opposition to be fair, and the line my opponents have taken throughout cannot, by any stretch of courtesy, be so described. My difficulty, indeed, in dealing with their arguments on the Battle of Hastings, is that they do not affect or even touch my case. In spite of their persistent efforts to obscure a plain issue, there is not, and there cannot be, any 'controversy' as to Mr Freeman and the 'palisade'. For, while fully recognizing that the onus probandi lay on those who assert its existence, he failed, on his own showing, to produce any proof of it whatever.⁹ Mr Archer has ended,¹⁰ as he began,¹¹ by deliberately ignoring Mr Freeman's words,¹² on which my case avowedly rests, and without [pg 13] suppressing which he could not even enter the field. This, indeed, I have explained so often, that I need not again have disposed of his arguments had not Mr Gardiner, in the exercise of his editorial discretion, allowed him to make certain statements,¹³ and refused me the right of exposing them. A typical example will be found on p. 273.¹⁴

It is not only demonstrable error that justifies critical treatment; no less dangerous, if not more so, is that subtle commixture of guess-work and fact, which leaves us in doubt as to what is proved and what is merely hypothesis. In his lecture on 'The Nature of Historical Evidence', the late Professor himself well brought out the point:

Many people seem to think that a position is proved if it can not be disproved.... Very few see with Sir George Lewis—though Sir George Lewis perhaps carried his own doctrine a little too far—that in a great many cases we ought to be satisfied with a negative result, that we must often put up with knowing that a thing did not happen in a particular way, or did not happen at all, without being furnished with any counter-statement to put in the place of that which we reject.¹⁵

The question is whether a statement can be proved, not whether it can be disproved. Cases in point will be found on pp. 291, 298, 331-3.

It may, in view of certain comments, be desirable, perhaps, to explain that the study on the origin of knight-service appeared in Mr Freeman's lifetime,¹⁶ and that my open criticism of his work began so far back as 1882. It will be seen, therefore, that I challenged its accuracy when he was himself able to reply.

To those who may hold that in these studies excessive attention is bestowed on Anglo-Norman genealogy, I commend the words, not of a genealogist, but of the historian Kemble:

It is indispensable to a clear view of the constitutional law and governmental institutions of this country, that we should not lose sight of the distribution of landed estates among the great families, and that the rise and fall of these houses should be carefully traced and steadily borne in mind....

Amidst all the tumult and confusions of civil and foreign wars; throughout religious and political revolutions; from the days of Arminius to those of Harald; from the days of Harald to our own; the successions of the landowners and the relations arising out of these successions, are the running comment upon the events in our national history: they are at once the causes and the criteria of facts, and upon them has depended the development and settlement of principles, in laws which still survive, in institutions which we [pg 14] cling to with reverence, in feelings which make up the complex of our national character.¹⁷

The paper on 'Walter Tirel and his wife' may serve to show that in this department there is still needed much labour before we can hope for a perfect record of the great houses of the Conquest.

I have to thank Mr Murray for his kind permission to make use of two of the articles I have contributed to the Quarterly Review.. Some of the studies have previously appeared in the English Historical Review, and these are now republished with Messrs Longmans' consent. Lastly, I would take the opportunity afforded by this preface of acknowledging the encouragement my researches have derived from the approval not only of our supreme authority—I mean the Bishop of Oxford—but also of that eminent scholar, Dr Liebermann, whose name one is proud to associate with a work on mediaeval history.

J. H. ROUND

[Note: I have not thought it needful to include in the index names of persons or places only introduced incidentally in illustration of arguments. The prefix 'Fitz', as in Geoffrey de Mandeville, has been retained as a useful convention, whatever the actual name may have been.]

¹ Table chronologique des chartes et diplômes imprimés concernant l'histoire de la Belgique. Par Alphonse Wauters, vol. i, p. xxxi.

² See pp. 198, 208, 404-5.

³ See p. 430.

⁴ Prof Maitland informs me that since the appearance of his Select Pleas in Manorial Courts, he has discovered the earlier occurrence of the word 'leet' (see p. 90).

⁵ See Domesday Studies.

⁶ Fortnightly Review, December 1894, pp. 804-5.

⁷ Quarterly Review., July 1892.

⁸ See Quarterly Review. as above.

⁹ See pp. 263-9.

¹⁰ English Historical Review, July 1894.

¹¹ Contemporary Review, March 1893, pp. 335-55.

¹² Norman Conquest (2nd Ed.), iii, 763-4.

¹³ English Historical Review, as above.

¹⁴ I have, therefore, been obliged to refer in some detail to these statements, while for those I have already disposed of I have given the references to the Q.R. and E.H.R.

¹⁵ Methods of Historical Study, p. 141.

¹⁶ English Historical Review, July 1891-January 1892.

¹⁷ The Names, Surnames, and Nicknames of the Anglo-Saxons. Read at Winchester, September 11, 1845.

[pg 15]

CONTENTS

[pg 17]

PART I

TERRITORIAL STUDIES

DOMESDAY BOOK

The true key to the Domesday Survey, and to the system of land assessment it records, is found in the Inquisitio Comitatus Cantabrigiensis. Although the document so styled is one of cardinal importance, it has, from accident, been known to few, and has consequently never succeeded in obtaining the attention and scientific treatment it deserved. The merit of its identification belongs to Mr Philip Carteret Webb, who published in 1756 a paper originally read before the Society of Antiquaries, entitled, A Short Account of Danegeld, with some further particulars relating to William the Conqueror's Survey. It is difficult to speak too highly of this production, remembering the date at which it was composed. Many years were yet to elapse before the printing of Domesday was even begun, and historical evidences were largely inaccessible as compared with the condition of things today. Yet the ability shown by Mr Webb in this careful and conscientious piece of work is well seen in his interesting discovery, which he announced in these words:

In searching for the Liber Eliensis, I have had the good fortune to discover in the Cotton Library a MS. copy of the Inquisition of the jury, containing their survey for most of the hundreds in Cambridgeshire. This MS. is written on vellum in double columns and on both sides of the page. It is bound up with the Liber Eliensis, and begins at p. 76a and ends at p. 113. It is written in a very fair but ancient character, not coeval with the Survey, but of about the time of Henry II. It was given by Mr Arthur Agard to Sir Robert Cotton, and is marked Tiberius A. VI 4. Your lordship and the Society will be of opinion that this is a discovery of importance, and what had escaped the observation of Sir H. Spelman, Mr Selden, and other antiquarians. A part of this valuable morsel of antiquity is already transcribed, and in a few weeks I hope to be able to communicate the whole of it to the Society (p. 26).

Mr Webb's discovery was known to Kelham, and duly referred to by him in his Domesday Book Illustrated (1788). It was also known to Sir Francis Palgrave, strong in his acquaintance with manuscript authorities, who alluded (1832) to the fact that 'fragments of the [pg 18] original inquisitions have been preserved',¹ and described the MS. Tib. A. VI, of which 'the first portion consists of the Inquisitio Eliensis, extending, as above mentioned, into five counties; it is followed by the inedited Inquisitio', etc.² It is, however, undoubtedly ignored in Ellis's Introduction to Domesday Book (1833), and 'even the indefatigable Sir Thomas Duffus Hardy', writes Mr Birch,³ 'has omitted all notice of this manuscript in his Descriptive Catalogue of Manuscripts relating to the History of Great Britain and Ireland, vol. ii. (1865)'. This, however, is not strictly the case, for in his notice of the Domesday MSS. he observes in a footnoteid:

The Cottonian MS. [Tib. A. VI] has also a second and unique portion of this survey, which was not printed in the edition published by the Record Commission in 1816. It commences 'in Grantebriggesira, in Staplehouhund', and ends imperfectly 'et vicecomiti regis v. auras'.

These words prove that Sir Thomas had inspected the MS., which duly begins and ends with the words here given.

It is certain, however, that Mr Freeman, most ardent of Domesday students, knew nothing of this precious evidence, and remained therefore virtually unacquainted with the modus operandi of the Great Survey. The pages, we shall find, of the Inquisitio afford information that no one would have welcomed more eagerly than himself. Perhaps, therefore, it is not surprising that Mr N. E. S. A. Hamilton, when editing this document for the Royal Society of Literature (1876), should have supposed that it had been overlooked till then, or that he was 'the first to bring its importance to light' (p. vi). It is, however, much to be regretted that Mr De Gray Birch should have strenuously insisted that Webb (whose paper he actually names) and Kelham 'appear to have been strangely ignorant of the true and important nature of this manuscript',⁴ and should have repeated this assertion⁵ after I had shown at the Domesday Commemoration (1886) that the honour of the discovery really belonged to Mr P. C. Webb. One may claim that Webb should have his due, while gladly expressing gratitude to Mr Hamilton for his noble edition of the Inquisitio, which has conferred on Domesday students an inestimable boon.⁶

The printing of the document in record type, the collation throughout with Domesday Book, and the appending of the Inquisitio Eliensis, edited from three different texts, represent an extraordinary amount of minute and wearisome labour. The result is a volume as helpful as it is indispensable to the scholar.

[pg 19]

I propose in this paper to take up anew the subject, at the point where Mr Hamilton has left it, to submit the text to scientific criticism, to assign it its weight in the scale of authority, and to explain its glossarial and its illustrative value for the construction and the contents of Domesday Book.

I. NATURE OF THE 'INQ. COM. CANT.'

Exact definition is needful at the outset in dealing with this document. The Inquisitio Comitatus Cantabrigiensi, which is entered on fos. 76-113 of Tib. A. VI, must be carefully distinguished from the Inquisitio Eliensis on fos. 38-68. Mr Hamilton doubted whether any one before him 'had distinguished between' the two, but this, we have seen, was a mistake. The distinction however is all-important, the two documents differing altogether in character. One would not think it necessary to distinguish them also from the so-called Liber Eliensis (which is not a survey at all) had not Mr Eyton inadvertently stated that our document has been printed under the title of Liber Eliensis.⁷

The Inquisitio Comitatus Cantabrigiensi (hereafter styled 'the I.C.C.') deals with the county of Cambridge alone, but, in that county, with the lands of all holders. The Inquisitio Eliensis (which I propose to style 'the I.E.') deals with several counties, but, in these counties, with the lands of the abbey alone. The latter was duly printed, with Domesday Book, by the Record Commission; the former remained in manuscript till printed by Mr Hamilton.

Mr Hamilton describes his record at the outset as 'the Original Return made by the Juratores of the county of Cambridge in obedience to the Conqueror's mandate, from which the Exchequer Domesday for that county was afterwards compiled by the King's secretaries', and as 'the original source from which the Exchequer Domesday for that county was derived'. Mr Birch here again repeats the words, insisting 'that we have in this very precious Cottonian MS. the original source from which the Exchequer Domesday of Cambridgeshire was compiled'.⁸

Such a description is most unfortunate being not only inaccurate but misleading. All that we are entitled to predicate of the document is that it is apparently a copy of the original returns from which Domesday Book was compiled. For 'the original source' of both we must look to the now missing returns of the jurors, the primary authority from which Domesday Book and the Inquisitio Com. Cant. are independently derived. This distinction is all-important, reducing, as it does, the Inquisitio from the rank of an 'original' to that [pg 20] of a secondary authority on the same level with Domesday Book.⁹ Mr Hamilton, like Mr Webb before him, assigned the handwriting of the Inquisitio to about the close of the twelfth century. The copy of the returns which it contains, therefore, was made about a century later than the returns themselves.

The problem then that we have to solve is this: 'Is the I.C.C. an actual transcript of these original returns, and if so, is it faithful?' I will not, like Mr Hamilton, assume an affirmative, but will attempt an impartial inquiry.

The two paths which we must follow in turn to arrive at a just conclusion are (1) the construction of the I.C.C., (2) collation with the Inq. Eliensis. For I hope to show that the latter record must have been derived from the same source as the Inq. Com. Cant.

Following the first of these paths, we note at once that while Domesday Book arranges the Manors according to fiefs, the Inq. Com. Cant., on the contrary, arranges them by hundreds and townships. Its system is regular and simple. For every hundred it first enumerates the principal jurors who made the return, and then gives the return itself, arranged according to townships (villæ). These townships are thus the units of which the Manors they contain are merely the component fractions. This is precisely what we should expect to find in the original returns, but it only creates a presumption; it does not afford a proof. For instance, it might be reasonably urged that these copies may have omitted certain items in the returns, just as Domesday Book omitted others.

To reply to this objection, we must turn to the second path; that is to say, we must collate the Inquisitio Eliensis with the Inq. Com. Cant. I shall prove below that the latter cannot have been taken from the former, which only covers a portion of its field, and that, on the other hand, the former cannot have been taken from the latter, because the Inquisitio Eliensis is accurate in places where the Inq. Com. Cant. is in error. Consequently they must both have been derived independently from some third document. This being so, if we should find that their versions agree closely, we may fairly infer that each is intended to be a faithful reproduction of the above 'third document'. In other words, if neither version omits items which are given in the other, we are entitled to assume that the copy is in each case exhaustive, for two scribes working independently are not likely to have systematically omitted the same items from the document before them.

What then was the 'third document' from which they both copied? Obviously it was either the original returns of the Domesday jurors, or a copy (exhaustive or not) of these returns. Now we cannot suppose that two scribes, working, as I have said, independently, [pg 21] would both have worked, not from the original returns themselves, but from a copy, and that the same copy of these returns—a copy, moreover, of the existence of which we have no evidence whatever. Moreover, in this hypothetical copy, there would, we may safely assert, have been some clerical errors. These would have duly re-appeared in both the Inquisitiones, and collation with Domesday Book would enable us to detect them. Yet in no single instance, though each of them contains errors, have I found a clerical error common to both. We are thus driven to the conclusion that in both these Inquisitiones we have copies of the actual returns made by the Domesday jurors.

One of the postulates in the above argument is that the Inq. Com. Cant. and the Inq. Eliensis 'agree closely' in their versions. Here is an instance in illustration:¹⁰

These extracts are typical and instructive. They leave, in the first place, no doubt upon the mind that both are versions of the same original. This, which proves my postulate, will be shown below to possess a further and important bearing. But while these versions closely agree, we notice (1) independent blunders, (2) slight variants in diction. As to blunders, we see that the I.C.C. has 'sol[idis]' where the I.E. has the correct 'hidis', while, conversely, the I.E. reads 'hun[dredis]' where the I.C.C. has, rightly, 'hidis'. Again the I.C.C. allots to demesne an assessment of a hide and a virgate, but I.E. a hide and a half (i.e. two virgates). Collation with [pg 22] Domesday Book confirms the former version. Conversely, the I.C.C. assigns to the mill the value of three shillings and eightpence, but the I.E. of two shillings and eightpence. Collation with Domesday Book confirms the latter. Turning now to the variants, we may express them more clearly thus:

These prove that verbal accuracy was not aimed at by the transcribers. The same freedom from its trammels is seen in the transposition of the 'mill' and 'meadow' passages, and, indeed, in the highly abbreviated form of the I.E. entries (in which a single letter, mostly, does duty for a word), which shows that the original version must have been either extended in the I.C.C., or (more probably) abbreviated in the I.E.

We are now in a position to advance to the criticism of the text of the Inq. Com. Cant., and to inquire how far it can be trusted as a reproduction of the original returns. In other words, are its contents more or less trustworthy than those of Domesday Book?

It might, no doubt, be fairly presumed that a simple transcript of the original returns was less likely to contain error than such a compilation as Domesday Book, in which their contents were (1) rearranged on a different system, (2) epitomized and partly omitted, (3) altered in wording. Mr Hamilton, indeed, who was naturally tempted to make the most of his MS., appears to have jumped at this conclusion; for he speaks in his preface (p. xii) of its 'superior exactness', and gives us no hint of omissions or of blunders. There are, however, plenty of both, as will be seen from the lists below, which do not profess to be exhaustive.

But we will first examine the instances adduced by Mr Hamilton. Out of ten examples in proof of its value, five are cases in which 'the want of precision in Domesday' leaves the identity of the tenant-in-chief 'undefined'. It is difficult to comment on these statements, because in all five cases the name is as carefully recorded in Domesday as in the I.C.C. Mr Hamilton's error can only, it will be found, have arisen from comparing the I.C.C. not with Domesday Book, but with the extracts therefrom printed in his work, which, being torn from their place, do not, of course, contain the tenant's full name, which in Domesday itself is given at the head of the list from which they are taken. Moreover, as it happens, this [pg 23] test demonstrates not the inferiority, but (in one instance at least) the superiority of Domesday, the I.C.C. (fo. 97, col. 2) reading 'Hanc terram tenuit comes alanus' [sic], where Domesday has (rightly) 'Hanc terram tenuit Algar comes'. The former must have wrongly extended the abbreviated original entry.¹¹

Another of Mr Hamilton's examples is this:

'Hæc terra fuit et est de dominio æcclesiæ' (Domesday) is abbreviated from a long account of the holdings of Harduuinus de Scalariis and Turcus homo abbatis de Rameseio in the Cotton MS.

But, on referring to the passage in question, we find that the Domesday passage: 'Hæc terra fuit et est de dominio æcclesiæ' has nothing to do with that 'long account', but corresponds to the simple formula in the I.C.C., 'Hanc terram tenuerunt monache de cet'ero T.R.E. et modo tenent'. The example which follows it is this:

At pp. 38, 39 we see a curious alteration in the value of the land, which had risen from xv. lib. 'quando recepit' and T.R.E. to xvii. lib. at the time the return was made, and dropped again to xvi. lib. in the Domesday Survey.

This strange comment implies the supposition that the I.C.C. records an earlier survey than Domesday Book, whereas, of course, they are derived from the same returns, so that the discrepancy of xvi. and xvii. is merely a clerical error. One more instance, the 'curious reading' Harlestone in the I.C.C., is shown below to be merely an error in that MS. Such are eight of the examples adduced by Mr Hamilton. The remaining two merely illustrate not the superior accuracy, but the greater elaboration of the I.C.C. It has been absolutely necessary to dispose of these examples, in order to show that a critical estimate of the value of the I.C.C. has yet to be made.

Taking the omissions in the MS. first, we find some really bad ones. On fo. 79a (2), collation with Domesday gives this result:

[pg 24]

A similar 'run on' omission is found on fo. 109a (1):

Another instance of 'running on' occurs on fo. 105a (1), where 'xviii. cotarii' (p. 67) is proved by Domesday to stand for 'xviii. [bordarii x.] cotarii'. Again on fo. 79b (2) we have this:

So, too, on fo. 100b(1):

The importance of such an omission as this lies in the proof of unintelligent clerkship and want of revision which so unmeaning an entry as 'xv. xvi. carrucis' supplies.

Omissions of another character are not infrequent. On fo. 95b (1) an entire holding of a virgate (held by a sokeman of Earl Alan) is omitted (p. 34). Another sokeman of Earl Alan (p. 32) has his [pg 25] holding (¼ virgate) omitted on the same folio (95a, 1), so is an entire holding of Hardwin's (p. 36) on fo. 96a (2). A demesne plough ('i. caruca') of Hugh de Port (p. 8) is omitted (78a, 1), and so are the ploughs ('et iiii. villanis') of Aubrey's villeins (p. 9) a few lines lower down. On fo. 90a (1) the words 'ibi est terra' are wanting (p. 15),¹⁵ and so are 'non potuit' on fo. 100 (A) 1.¹⁶ The word 'recedere' is left out on fo. 103b (2),¹⁷ and 'soca' just before (103 (B) 1).¹⁸ 'Odo' is similarly wanting on fo. 90a (1).¹⁹ The note also on the Abbot of Ely's sokeman at Lollesworth (p. 95), is wholly omitted (fo. 113, B, 2), though found both in Domesday Book and in the Inquisitio Eliensis.²⁰

Turning now to the clerical blunders, we find an abundant crop. We may express them conveniently in tabular form:

Besides these, Ralf 'de bans' is often entered as Ralf 'de scannis'. Again, we find such blunders as this:

In all these three cases the italicized words are misplaced, and in all three the explanation is the same, the scribe having first omitted them, and then inserted them later out of place. Having now criticized the text of the I.C.C., and shown that it presents no small traces of unintelligent clerkship, if not of actual ignorance of the terms and formulæ of Domesday, I turn to the text of Domesday Book, to test it by comparison with that of the I.C.C.

II. CRITICISM OF THE DOMESDAY TEXT

Among the omissions are, on i, 195 (b) 1, 'Item et reddebat viii. den. vel aueram si rex in vicecomitatu venit' (p. 5). At Kirtling [pg 27] (p. 11), 'et v^ta. caruca potest fieri [in dominio]' is omitted (i. 202 a). So is (p. 25) a potential demesne plough of John fitz Waleran (i. 201 b). The Countess Judith's sokemen at Carlton (pp. 20, 21) have their values omitted²⁵ (i. 202, a, 2). 'Habuerunt dimidiam hidam, et,' is omitted (p. 28) in the entry of two sokemen of Godwine (201, b, 2). On i. 196 (a) 1, 'Terra est i. bovi' is wanting (p. 79). More important, however, are the omissions of whole entries. These are by no means difficult to account for, the process of extracting from the original returns, the various entries relating to each particular fief being one which was almost certain to result in such omissions.²⁶

Moreover, two entries were occasionally thrown into one, a dangerous plan for the clerks themselves, and one which may sometimes lead us to think that an entry is omitted when it is duly to be found under another head. Lastly, the compilers of Domesday Book had no such invaluable check for their work as was afforded in the original by entering first the assessment of the whole township, and then that of each of its component Manors separately. But of this more below.²⁷ The only wonder is that the omissions are, after all, so few. Perhaps even of these some may be only apparent. Hardwin's half-hide in Burwell (p. 6) is wanting; so is Aubrey's half-virgate in Badburgham, according to Mr Hamilton (p. 36), but the oversight is his. A virgate held in Trumpington by a burgess of Cambridge (p. 51) would seem to be not forthcoming, but its position was somewhat anomalous.²⁸ Guy de Rembercurt held a hide and a virgate in Haslingefield (p. 73), though we cannot find it in Domesday; and in Witewelle (Outwell) two hides which were held by Robert, a tenant [pg 28] of Hardwin (p. 81), are similarly omitted, according to Mr Hamilton but will be found under 'Wateuuelle' (198, b, 2).

There are cases in which the I.C.C. corrects D.B., cases in which D.B. corrects the I.C.C., and cases in which the I.C.C. corrects itself. There are also several cases of discrepancy between the two, in which we cannot positively pronounce which, if either, is right. A singular instance of both being wrong is found in the case of Soham. The assessment of this township was actually eleven hides, its four component holdings being severally assessed at nine and a half hides less six acres, half a hide, one hide, and six acres. The I.C.C. at first gives the total assessment as eleven hides and a half, while D.B. erroneously assesses the first of the four holdings at six hides and forty acres in one place, and nine hides and a half in the other, both figures being wrong. A most remarkable case of yet another kind is found in Scelford (Shelford). Here the entry in I.C.C. agrees exactly with the duplicate entries found in D.B. Yet they both make nonsense.²⁹ But on turning to the Inquisitio Eliensis we obtain the correct version. As this is a very important and probably unique instance, the entries are here given in parallel columns:

Here the Inquisitio Eliensis version shows us that the estate had two divisions held by different tenures. Three virgates the sokemen were not free to sell; the other three they might sell, but if they did, [pg 29] 'predictus abbas semper socham habuit'.³¹ The two divisions of the estate are confused in the other versions. But all three of these correspond so exactly that we are driven to assign the error to the original returns themselves. In that case the compiler (or compilers) of the I.E. will have corrected the original return from his own knowledge of the facts, which knowledge, I shall show, he certainly possessed.

This brings us to the errors of Domesday. For comparison's sake, I here tabulate them like those of the I.C.C.:

Comparing the omissions and errors, as a whole, in these two versions of the original returns, it may be said that the comparison is in favour of the Domesday Book text, although, from the process of its compilation, it was far the most exposed to error. No one who has not analysed and collated such texts for himself can realize the extreme difficulty of avoiding occasional error. The abbreviations and the formulæ employed in these surveys are so many pitfalls for [pg 30] the transcriber, and the use of Roman numerals is almost fatal to accuracy. The insertion or omission of an 'x' or an 'i' was probably the cause of half the errors of which the Domesday scribes were guilty. Remembering that they had, in Mr Eyton's words,³⁴ to perform 'a task, not of mere manual labour and imitative accuracy, but a task requiring intellect—intellect, clear, well-balanced, apprehensive, comprehensive, and trained withal', we can really only wonder that they performed it so well as they did.

Still, the fact remains that on a few pages of Domesday we have been able to detect a considerable number of inaccuracies and omissions. The sacrosanct status of the Great Survey is thus gravely modified. I desire to lay stress on this fact, which is worthy of the labour it has cost to establish. For two important conclusions follow. Firstly, it is neither safe nor legitimate to make general inferences from a single entry in Domesday. All conclusions as to the interpretation of its formulae should be based on data sufficiently numerous to exclude the influence of error. Secondly, if we find that a rule of interpretation can be established in an overwhelming majority of the cases examined, we are justified, conversely, in claiming that the apparent exceptions may be due to errors in the text.

The first of these conclusions has a special bearing on the theories propounded by Mr Pell with so much ingenuity and learning.³⁵ I have shown, in an essay criticizing these theories,³⁶ that the case of Clifton, to which Mr Pell attached so much importance,³⁷ is nothing, in all probability, but one of Domesday's blunders, of which I gave, in that essay, other instances. So, too, in the case of his own Manor of Wilburton, Mr Pell accepted without question the reading 'six ploughlands', as representing the 'primary return',³⁸ although that reading is only found in the most corrupt of the three versions of the Inquisitio Eliensis, while the two better versions (B and C texts) agree with Domesday Book, and with the abbreviated return at the end of the A text itself (Tib. A. VI fo. 67, b, 1), in giving the ploughlands as seven. Really it is nothing but waste of time to argue from a reading which is only found in one out of five MSS., and that one the most corrupt.

This brings me to the existence and the value of duplicate entries in Domesday. Mr Hamilton describes as 'a curious reading' the words in the I.C.C., 'sed soca remanebat Harlestone'. Now it so happens that in this case we have five separate versions of the original entry: one in the I.C.C., one in the I.E., and three in Domesday Book. Here they are side by side:

[pg 31]

The value of such collation as this ought to be self-evident. It is not only that we thus find four out of five MSS. to be against the reading 'Harlestone' (which, indeed, to any one familiar with the survey is obviously a clerical error), but that here and elsewhere we are thus afforded what might almost be termed a bilingual inscription. We learn, for instance, that the Domesday scribe deemed it quite immaterial whether he wrote 'recedere cum terra ejus', or 'vendere' or 'recedere sine licentia'. Consequently, these phrases were all identical in meaning.³⁹

Considerable light is thrown by the I.C.C. on the origin of these little known duplicate entries in Domesday. In every instance of their occurrence within the limits of its province they are due to a conflict of title recorded in the original return. They appear further to be confined to the estates of two landowners, Picot, the sheriff, and Hardwin d'Eschalers, the titles of both being frequently contested by the injured Abbot of Ely. Why the third local offender, Guy de Raimbercurt, does not similarly appear, it is difficult to say. He was the smallest offender of the three, and Picot the worst; but it is Hardwin's name which occurs most frequently in these duplicate entries.⁴⁰ The principle which guided the Domesday [pg 32] scribes cannot be certainly decided, for they duplicated entries in the original return which (according to the I.C.C.) varied greatly in their statements of tenure. Thus, to take the first three:

I.C.C.		D.B.
fo. 79 (b) 1, 'Tenet Harduuinus descalariis'.41		I. 190 (b) 2, 'Tenet Harduinus sub abbate'.
		I. 199 (a) 2, 'Tenet Harduinus'.
fo. 90 (b) 2, 'Tenet Harduuinus de abbate'.		I. 190 (b) 1, 'Tenet Harduinus de Escalers de abbate'.
		I. 199 (a) 2, 'Tenet Harduinus'.
fo. 92 (a) 2, 'Tenet Harduuinus de rege'.		I. 199 (b) 2, 'Tenet Harduinus de abbate'.
		I. 199 (a) 2, 'Tenet Harduinus'.

Here, whether the original return states Hardwin to hold (1) of the abbot, (2) of the king, or (3) of neither, the scribes, in each of the three cases, enter the estates (A) under the Abbot's land, as held of the Abbot, (B) under Hardwin's land, as held in capite. And it is singular that in all these three cases the entry of the estate under the Abbot's land is the fuller of the two.⁴²

On the whole it would appear that the Domesday scribes did not consistently carry out a system of duplicate entry, though, on the other hand, these entries were by no means due to mere clerical inadvertence, but were prompted by a doubt as to the title, which led to the precaution of entering them under the names of both the claimants.

But the chief point of interest in these same entries is that they give us, when we add the versions of the I.C.C. and the I.E., four parallel texts. At some of the results of their collation we will now glance.

[pg 33]

These extracts illustrate the use of the terms dare, vendere, recedere, etc. They are supplemented by those given below:

I.C.C.	D.B.		I.E.
(76, a, 1)	(I. 196, b, 1)
Potuit dare sine licentia domini sui terram suam.	Terram suam tamen dare et vendere potuit.
(76, b, 2)	(I. 199, a, 2)		(p. 101)
Absque eius licentia dare terram suam potuerunt, sed socham eorum habuit archiepiscopus.	Sine ejus licentia poterant recedere et terram suam dare vel vendere, sed soca remansit Archiepiscopo.		Potuerunt dare vel vendere terram suam. Saca remansit abbati.
(76, b, 2)	(I. 196, b, 1)
Potuit dare cui voluit.	Potuit absque43 ejus licentia recedere.
(77, b, 2)	(I. 195, b, 1)
Potuerunt recedere cum terra ad quem dominum voluerunt.	Potuerunt receder sine licentia eorum.
(78, a, 1)	(I. 190, b, 1)
Potuerunt recedere cum terra sua absque licentia domini sui.	Dare et vendere potuerunt.
(90, a, 2)	(I. 190, b, 2)		(p. 102)
Non potuerunt recedere sine licentia abbatis.	Non potuerunt recedere sine ejus licentia.		Non potuerunt recedere vel vendere absque eius licentia.
	(I. 200, a, 2)
	Non potuerunt vendere sine ejus licentia.
[pg 34] (105, a, 2)	(I. 200, a, 1)		(p. 109)
Potuerunt dare et vendere sine soca.	Terras suas vendere potuerunt. Soca de viii. sochemannis remansit in abbatia de ely.		Potuerunt dare vel vendere cui voluerunt, sed saca eorum remansit eidem abbati.
(113, b, 1)	(201, a, 1)		(p. 112)
Potuerunt recedere sine soca.	Terram suam vendere potuerunt. Soca vero remansit abbati.		Potuerunt dare preter licentiam abbatis et sine soca.

No one can glance at these passages without perceiving that dare, vendere, and recedere are all interchangeably used, and that even any two of them (whether they have the conjunction 'et' or the disjunction 'vel' between them) are identical with any one. It would be possible to collect almost any number of instances in point. Further, the insertion or omission of the phrase 'sine' (or 'absque') 'ejus licentia' is immaterial, it being understood where not expressed. So too with the words 'cui voluit'. In short, like the translators to whom we owe the Authorized Version, the Domesday scribes appear to have revelled in the use of synonym and paraphrase.⁴⁴ Our own conceptions of the sacredness of a text and of the need for verbal accuracy were evidently foreign to their minds.

Glancing for a moment at another county, we have in the Survey of Leicestershire a remarkable instance of a whole fief being entered twice over. It is that of Robert Hostiarius:

[pg 35]

Here the last two entries (both relating to Claxton) have been boldly thrown into one in the second version, which also (though omitting the number of ploughlands) gives additional information in the name of Robert's father, and in those of his predecessors T.R.E. This is thus an excellent illustration of the liberty allowed themselves by the compilers of Domesday.

An instance on a smaller scale is found in the Survey of Cambridgeshire, where we read on opposite pages:

The second entry has been deleted as a duplicate, but it serves to show us that the scribes, even when free from error, were no mere copyists.⁴⁵

III. 'SOCA' AND 'THEINLAND'

The extracts I have given above establish beyond a doubt the existence among the 'sochemanni' of two kinds of tenure. We have (1) those who were free to part with (vendere) and leave (recedere) their land, (2) those who were not, i.e. who could not do so without the abbot's licence. This distinction is reproduced in two terms which I will now examine.

In the Inquisitio Eliensis and the documents connected with it there is much mention of the 'thegnlands' of the Abbey. These lands are specially distinguished from 'sokeland' (terra de soca). Both, of course, are distinct from the 'dominium'. Thus in one of the Conqueror's writs we read:

Restituantur ecclesiæ terræ que in dominio suo erant die obitus Æduardi.... Qui autem tenent theinlandes que procul dubio debent teneri de ecclesia faciant concordiam cum abbate quam meliorem poterint,... Hoc quoque de tenentibus socam et sacam fiat.⁴⁶

Now this distinction between 'thegnland' and 'sokeland' will be found to fit in exactly with the difference in tenure we have examined above. Here is an instance from the 'breve abbatis' in the record of Guy de Raimbercurt's aggressions:

[pg 36]

In melreda ii. hidas et dim. virg.

In meldeburne ii. hidas et dim.⁴⁷ et dim. virg.

Hoc est iiii. hidas et iii. virg. Ex his sunt i. virg. et dim. thainlande et iiii. hidas et dim.⁴⁸ de soca.

On reference to the two Manors in question, there is, at first sight, nothing in the I.C.C., the I.E., or Domesday to distinguish the 'thegnland' from the 'sokeland'. Of the first holding we read that it had been held T.R.E. by 10 sochemanni 'de soca S. Edelride'; of the second, that it was held by 'viii. sochemanni ... homines abbatis de Ely'. But closer examination of the I.C.C. reveals, in the former case, this distinction:

De his ii. hidis et dimidia virga tenuit i. istorum unam virgam et dimidiam. Non potuit dare nec vendere absque licentia abbatis. Sed alii novem potuerunt recedere et vendere cui voluerunt.⁴⁹

Here then we identify the virgate and a half of 'theinland'—though held by a sochemannus—and this same distinction of tenure proves to be the key throughout. Thus, for instance, in the same document 'Herchenger pistor' is recorded to have seized 'in Hardwic i. hidam thainlande et dim. hidam et vi. acras de soca' (p. 177). Reference to the I.C.C., D.B., and the I.E. reveals that the former holding had belonged to 'v. sochemanni homines abbatis de ely', and that 'isti non potuerunt dare neque vendere alicui extra ecclesiam S. Edeldride ely'.⁵⁰ But the latter holding had belonged to a sochemannus, of whom it is said—'homo abbatis de ely fuit: potuit recedere, sed socam ejus abbas habuit'.⁵¹

This enables us to understand the distinctions found in the summaries appended to the Cambridgeshire portion of the I.E., and recorded in the Breve Abbatis. Indeed they confirm the above distinction, for the formula they apply to holders 'de soca abbatie ely' is: 'Illi qui hanc terram tenuerunt de soca T.R.E. vendere potuerunt, sed saca et soca et commendatio et servitium semper remanebat ecclesia de ely.'

These terms are valuable for their definition of rights. Over the holder of land 'de soco' the lord had (1) 'saca et soca', (2) 'commendatio et (3) servitium'. If the land was thegnland then the [pg 37] Abbot received 'omnem consuetudinem' as well.⁵² We will first deal with the latter class, those from whom the Abbot received 'consuetudo', and then those who held 'de soca'.

For contemporary (indeed, slightly earlier) evidence, we must turn to the Ely placitum of 1072-75.⁵³ The special value which this placitum possesses is found in its record of the services due from sochemanni, and even from freemen. It thus helps to interpret the bald figures of Domesday, to which it is actually anterior. The first two instances it affords are these:

In breuessan tenet isdem W. terram Elfrici supradicte consuetudinis. In brucge tenet ipse W. terram etfled ejusdem modi.

The consuetudo referred to was this:

Ita proprie sunt abbati ut quotienscunque preceperit prepositus monasterii ire et omnem rei emendationem persolvere. Et si quid de suo voluerint venundare, a preposito prius licentiam debent accipere.

The corresponding entries in the I.E. run thus:

'In Brugge una libera femina commend' S. Ædel. de lxxx. ac. pro manerio.

In Beuresham ten[uit] Ælfricus i. liber homo commed' S. Ædel.⁵⁴ lx. acras pro manerio' (p. 165).

Thus we obtain direct evidence of the services due from commended freemen owing 'consuetudines'. Turning now to those of sochemanni, we have this important passage:

Willelmus de Warena tenet quadraginta quinque socamans in predicta felteuuella qui quotiens abbas preceperit in anno arabunt suam terram, colligent et purgabunt segetes, adducent et mittent in horrea, portabunt victum monachorum ad monasterium, et quotiens eorum equos voluerit, et ubicunque sibi placuerit, totiens habebit, et ubicunque forsfecerint abbas forsfacturam habebit, et de illis similiter qui in eorum terram forsfecerint.

Item Willelmus de uuarenna tenet triginta tres socamans, istius consuetudinis in Nortuuolda.

Item W. tenet quinque socamans istius modi in Muddaforda.

Supradictus Walterus et cum eo Durandus, homines hugonis de monte forti, tenent xxvi. socamans supradicte consuetudinis in Maraham.

Collating as before from the I.E. the relative entries, we find they run thus:

[pg 38]

Felteuuelle ... Huic manerio adjacebant T.R.E. xxxiiii. homines cum omni consuetudine, et alii vii. erant liberi homines,⁵⁵ qui poterant vendere terras, sed soca et commendatio remansit S. Ædel. (p. 132).

In felteuuella tenet W. de uuarenna xli. sochemannos ... Super hos omnes habebat S. Ædel. socam et commendationem et omnem consuetudinem. Illorum vii. liberi erant cum terris suis, sed soca et commendatio remanebat S. Ædel. (p. 139).

IIII. sochemanni adjacent [sic] huic manerio [felteuuella] T.R.E. Et modo habet eos W. de Warenna (p. 138).

Nortuualde ... Huic manerio adjacebant T.R.E. xxx. sochemanni cum omni consuetudine. Et alii iiii. liberi homines qui poterant vendere terras, sed saca et commendatio remanebat S. Ædel. (p. 132).

In Nortuualde S. Ædel. xxxiiii. sochem [annos] ... S. Ædel. [habuit] socam et commendationem et omnem consuetudinem de illis xxx. tantum; et iiii. erant liberi homines, socam et sacam et commendationem [super hos] S. Ædel. habebat⁵⁶ (p. 139).

Mundeforde ... Huic manerio adjacebant T.R.E. septem sochemanni cum omni consuetudine (p. 132).

In Mundeforde S. Ædel. vii. sochemannos cum omni consuetudine (p. 139).

Huic manerio [Mareham] T.R.E. adjacebant viginti vii. sochemanni cum omni consuetudine, sed postquam Rex W. advenit, habuit eos hugo de Munfort preter unum (p. 130).

[Terre hugo de Munford.] In mareham xxvi. sochemanni quos tenet [sic] S. Ædel. T.R.E.⁵⁷ ... hanc terram receperunt⁵⁸ pro escangio, et mensurata est in brevi S. Ædel. (p. 137).

Here then we identify these four cases: Feltwell, with its 41 sochemanni (more accurately described as 34 s. and 7 liberi homines) attached to one Manor and four to another—45 in all; Northwold, with its 33 or 34;⁵⁹ Muddiford with 5 or 7;⁶⁰ and Marham with its 26.

The three former Manors lay in the Hundred of Grimeshoe, the fourth northwards, towards the Wash. Just to the south of the three Manors, over the borders of Suffolk, lay Brandon, where Lisois de Moustiers had usurped the rights of Ely over six sochemanni.

In Lakincgeheda et in Brandona vi. sochemanni S. Ædel. ita quod non potuerunt vendere terras liberati liseie antecessori eudo[nis] dapif[eri] ... Post eum tenuit eos eudo et tenet cum saca et soca (p. 142).

The record of the placitum, drawn up during the tenure of Lisois, shows us their limited services: 'Isti solummodo arabant et c'terent [pg 39] [sic] messes ejusdem loci quotienscunque abbas præceperit.' The difference between these services and the others we have seen recorded is considerable.

Yet another group of sokemen on Suffolk Manors rendered these services:

Ita proprie sunt abbati ut quotienscunque ipse præceperit in anno arabunt suam terram, purgabunt et colligent segetes, portabunt victum monachorum ad monasterium, equos eorum in suis necessitatibus habebit [abbas], et ubicunque deliquerint emendationem habebit semper et de omnibus illis qui in terris eorum deliquerint.

This is practically the same definition as we had for the other group, and suggests that it was of wide prevalence. A notable contrast is afforded by the entry: 'In villa que vocatur Blot tenet ipse R. iiii. homines qui tantum debent servire abbati cum propriis equis in omnibus necessitatibus suis.'

We have now examined the consuetudines due from those 'qui vendere non potuerunt', and may turn to the rights exercised over the other class. Excluding 'servitium' (which is usually omitted as subordinate or comprised in the others), these are: (1) 'commendatio' (2) 'saca et soca'. The distinction between the two meets us throughout the survey of the eastern counties. A man might be 'commended' to one lord while another held his soca. Thus we read of Eadwine, a 'man' of the Abbot of Ely: 'Potuit dare absque eius licentia, sed socam comes Algarus habuit.'⁶¹ That is to say, he was 'commended to the Abbot of Ely', but Earl Ælfgar had the right of 'sac and soc' over him.⁶²

So too in the case of three 'liberi homines', commended to the Abbot in Norfolk. He had no right over them, but such as commendation conferred 'non habebat nisi commendationem', while their 'soca' belonged to the King's Manor of Keninghall.⁶³ Conversely, the Abbot of Ely had the 'soca' of a 'man' of Earl Waltheof,⁶⁴ and a 'man' of John, Waleran's nephew.⁶⁵ 'Commendatio', of course, took precedence as a right. Thus we read of the above three 'liberi homines'—'Hos liberos homines tenet [tenuit] Ratfridus, postea W. de Scodies, et abbas saisivit eos propter commendationem suam' (p. 133).

[pg 40]

In the above extracts we saw 'liberi homines qui vendere poterant' distinguished from 'Sochemanni', who could not sell. But we also saw that the two classes were not always carefully distinguished. We find, moreover, that the 'liberi homines' were themselves, sometimes, 'not free to sell'. Thus 'tenuit anant unus liber homo sub S. Ædel. T.R.E. pro manerio ii. carucatas terræ sed non potuit vendere' (p. 142). Some light may be thrown on this by the case of the estate held by Godmund, an abbot's brother:

Totam terram quam tenebat Gudmundus in dominio, id est Nectuna, sic tenebat T.R.E. de S. Ædel. quod nullo modo poterat vendere, nec dare; sed post mortem suam debebat manerium redire in dominio ecclesiæ; quia tali pacto tenuit Gudmundus de Abbate (p. 144).

With this we may compare these entries:

In Cloptuna ... Ædmundus commendatus S. Ædel. unam carucatam ... quam non potuit vendere nec dare (p. 150).

In Brandestuna Ædmundus presbyter terram quam accepit cum femina sua dedit S. Ædel. concedente femina T.R.E. ea conventione quod non posset eam dare nec vendere. Similiter de Clopetona' (p. 152).

In these cases the holder had only a life interest. Exactly parallel with the second is the case of 'Eadward', citizen of London, who gave lands to St. Paul's, reserving a life interest for himself and his wife—'et mortua illa Sanctus Paulus hereditare debuit'.⁶⁶

The above commendation of Edmund the priest ought to be compared with that of 'unus liber homo S. Ædel. commendatus ita quod non poterat vendere terram suam sine licentia abbatis', and of 'i. liber homo S. Ædel. Commendatus ita quod non poterat vendere terram suam extra ecclesiam (sed sacam et socam habuit stigandus in hersham)'.⁶⁷ Thus both those who were free to sell and those who were not, might belong to the class of 'liberi homines'. The essential distinction was one, not of status, but of tenure.

IV. THE DOMESDAY CARUCA

Yet more definite and striking, however, is the information on the Domesday caruca afforded by collating D.B. with the I.C.C. I referred at the Domesday Commemoration (1886) to the problem raised by the caruca,⁶⁸ and recorded my belief that in Domesday the word must always mean a plough team of eight oxen. The eight oxen, as Mr Seebohm has shown, are the key to the whole system of the carucate and the bovate. In Domesday, as I argued, the formula [pg 41] employed involves of necessity the conclusion that the caruca was a fixed quantity. Such entries, moreover, as 'terra i. bovi', 'terra ad iii. boves', etc., can only be explained on the hypothesis that the relation of the bos to the caruca was constant. But as the question is one of undoubted perplexity, and as some, like Mr Pell, have strenuously denied that the number of oxen in the Domesday caruca was fixed,⁶⁹ the evidence given below is as welcome as it is conclusive:

It is absolutely certain from these entries that the scribes must have deemed it quite immaterial whether they wrote 'dimidia caruca' or 'iiii. boves'; as immaterial as it would be to us whether we wrote 'half a sovereign' or 'ten shillings'. It is, consequently, as absolutely certain that the Domesday caruca was composed of eight oxen as that our own sovereign is composed of twenty shillings. And from this conclusion there is no escape.⁷⁰

Another point in connection with the caruca on which the I.C.C. gives us the light we need is this:

Here the Domesday text is utterly misleading as it stands. But the I.C.C., by supplying the omitted 'sex', gives us at once the right sense.

V. THE DOMESDAY HIDE

Similar to its evidence on the Domesday 'plough' is that which the I.C.C. affords as to the hide and virgate. In my criticism of Mr Pell's [pg 42] learned paper, I strenuously opposed his view that the hida of Domesday was composed of a variable number of virgates, and I insisted on the fact that the Domesday 'virgate' was essentially and always the quarter of the geldable 'hide'.⁷¹ The following parallel passages will amply prove the fact:

These are only some of the passages of direct glossarial value.⁷² Indirectly, that is to say by analysis of the township assessments, we obtain the same result throughout the survey passim.⁷³ Here, again, we are able to assert that two virgates must have been to the scribes as obviously equivalent to half a hide as ten shillings with us are equivalent to half a sovereign. For here, again, the point is that these scribes had no knowledge of the varying circumstances of each locality. They had nothing to guide them but the return itself, so that the rule, in Domesday, of 'four virgates to a hide' must have been of universal application.

But not only were there thus, in Domesday, four virgates to a hide; there were also in the Domesday virgate thirty Domesday acres. Mr Eyton, though perhaps unrivalled in the study he has bestowed on the subject, believed that there were only twelve such acres, of which, therefore, forty-eight composed the Domesday hide.⁷⁴ It is, perhaps, the most important information to be derived from the I.C.C. that a hundred and twenty Domesday acres composed the Domesday hide.⁷⁵

[pg 43]

We have the following direct statements:

If 20 acres were identical with two-thirds of a virgate, there must, in a whole virgate, have been 30 acres; and if a virgate, plus 10 acres, was equivalent to half a hide minus 20 acres, we have again a virgate of thirty, and a hide of 120 acres. But the conclusion I uphold will be found to rest on no isolated facts. It is based on a careful analysis of the Inquisitio throughout. Here are some striking examples:

fo. 92 (b) 1. 'Belesham pro x. hidis se defendit.'

fo. 99 (b) 1: 'tenet hardeuuinus de scal' vi. hidas et i. virgam et vii. acras de rege.'

	H.	V.	A.
Ely Abbey	2½	0	9
7 Sokemen	1½	0	6
3 Sokemen	½	0	0
'Alsi'	½	0	0		T.R.E.
2 Sokemen		1	7
5 Sokemen		3½	0
	——	——	——
	6	1	7

fo. 79 (a) 2: 'Suafham pro x. hidis se defendit.'

[pg 44]

fo. 90 (a) 'choeie et stoua pro x. hidis se defenderunt.'

fo. 96 (a) 2: 'Pampeswrda pro v. hidis et xxii. acris se defendit.'

fo. 107 (a) 2: 'Barentona pro x. hidis se defendit.'

[pg 45]

fo. 108 (a) 2: 'Oreuuella pro iiii. hidis se defendit.'

This last example is, perhaps, the most remarkable of all, in the accuracy with which the virgates and their fractions, by the help of the five acres, combine to give us the required total.

But, it may be asked, how far does the Inquisitio, as a whole, confirm this conclusion? In order to reply to this inquiry, I have analysed every one of the Manors it contains. The result of that analysis has been that of the ninety-four townships which the fragment includes (not counting 'Matingeleia', of which the account is imperfect) there are only fifteen cases in which my calculation does not hold good, that is to say, in which the constituents as given do not equal the total assessment when we add them up on the above hypothesis of thirty acres to the virgate, and four virgates to the hide. This number, however, would be considerably larger if we had to work only from D.B., or only from the I.C.C. But as each of these, in several cases, corrects the errors of the other, the total of apparent exceptions is thus reduced. Hence I contend that if we could only get a really perfect return, the remaining apparent exceptions would largely disappear.

In some of these exceptions the discrepancy is trifling. Thus, at Triplow, we have 2 acres in excess of the 8 hide assessment—a discrepancy of ¹⁄240. At 'Burch and Weslai' we have a deficit of 5 acres on 10 hides, that is ¹⁄240. At 'Scelforda' the figures of D.B. give us an excess of 7 acres on the 20 hide assessment, that is ⁷⁄2400. The I.C.C. figures make the excess to be 12 acres.

Another class of exceptions is accounted for by the tendency of both texts, as we have seen, to enter a virgate too much or too little, and to confuse virgates with their fractions. Thus at 'Litlingetona' our figures give us a virgate in excess of the assessment, while at 'Bercheham'⁸⁰ and again at 'Witlesforde' we have a virgate short of [pg 46] the amount. At 'Herlestona' we have, similarly, half a virgate too much, and 'Kingestona' half a virgate (15 acres) too little. Lastly, at 'Wicheham', the aggregate of the figures is a quarter of a virgate short of the amount.

A third class of these exceptions is due to the frequent omission in the I.C.C. of estates belonging to the king. Thus at Wilbraham it records an assessment of 10 hides represented only by two estates of four hides apiece. But on turning to Domesday (i. 189 b) we read: 'Wilborham dominica villa regis est. Ibi ii. hidæ.' The missing factor is thus supplied, and the apparent discrepancy disposed of. So, too, at 'Haslingefelda' (Haslingfield), where the I.C.C. accounts only for twelve hides and three virgates out of an assessment of twenty hides. Domesday here, again, supplies the missing factor in a royal Manor of seven hides and a virgate. We thus obtain, instead of an exception, a fresh illustration of our rule.

Haslingfield

Domesday omits altogether, so far as I can find, the holding of Guy, an omission which would upset the whole calculation. But, in the case of Isleham, the apparent exception is due to the I.C.C., not to Domesday Book. Its assessment, in that document, is given as four hides. But the aggregate of its Manors, as there recorded, gives us an assessment of three hides plus eighty acres. Here any one who was rash enough to argue from a single instance (as Mr Eyton and Mr Pell were too apt to do) might jump at the conclusion that the hide must here have been of eighty acres. Yet Domesday enables us to collect all the constituents of the 'Vill', among them the king's estate, here again omitted. The real figures, therefore, were these:

[pg 47]

Isleham, then, was a normal ten-hide township, and confirms, instead of rebutting, the rule that the geldable hide contained 120 acres.⁸¹

The remaining exceptions are 'Somm[er]tona' partly explained by the omission of terra Regis, 'Bathburgeham' (Babraham) with 21 acres short of an assessment of 7 hides, and Carlton, which fitly closes the list of these exceptions. For here, on an assessment of 10 hides, we have, according to the I.C.C., 27 acres short, but, according to D.B., 53½ (27 + 20 + 6½). A demonstrable blunder in Domesday Book and a discrepancy between it and the I.C.C. are responsible, together, for the difference.⁸² Thus we see how wide a margin should be allowed, in these calculations, for textual error.

It is necessary to remember that there were three processes, in each one of which error might arise:

I. In the actual survey and its returns, 'by reason of the insignificance of some estates, or by reason of forgetfulness, or inaccuracy, or confusion, or doubt on the part of local jurors and witnesses, or of the clerks who indited their statements'.⁸³

II. In the collection and transmission of the returns, by the loss of a 'leaflet or rotulet of the commissioners' work'.⁸⁴

III. In the transcription of the returns into D.B., or into the I.C.C., plus, in the case of the former, the rearrangement and abridgment of the materials.

We may now quit this part of our subject, claiming to have settled, by the aid of the I.C.C., a problem which has puzzled generations of antiquaries, namely: 'What was the Domesday hide?'⁸⁵ We have shown that it denoted a measure of assessment composed of four (geld) virgates or a hundred and twenty (geld) acres. What relation, if any, it bore to area and to value is a question wholly distinct, on which the next portion of this essay may throw quite a new light.

VI. THE FIVE-HIDE UNIT

It is one of the distinctive and valuable features of the Inq. Com. Cant. that it gives us the total assessment for each Vill of which it [pg 48] treats before recording the several Manors of which that Vill is composed, the aggregate assessments of which Manors make up the total assessment for the Vill. In this feature we have something which Domesday does not contain, and which (independently of its checking value),⁸⁶ gives us at once those Vill assessments which we could only extract from the Domesday entries by great labour and with much uncertainty. Let us see then if these Vill assessments lead us to any new conclusions on the whole assessment system.

The first point that we notice is this. The five-hide unit is brought into startling prominence. No careful student, one would suppose, of Domesday, can have failed to be struck by the singular number of Manors in the hidated portion of the realm, which are assessed in terms of the five-hide unit, that is to say, which are entered as of five hides or some multiple of five hides. This is specially the case with towns, and some years ago, in one of my earliest essays, I called attention to the fact, and explained its bearing in connection with the unit of military service.⁸⁷ Yet no one, it would seem, has been struck by the fact, or has seen that there must be some significance in this singular preponderance of five-hide Manors. Now what the Inquisitio here does for us is to show us that this preponderance is infinitely greater than we should gather from the pages of Domesday, and that when the scattered Manors are pieced together in their Vills, the aggregate of their assessments generally amounts either to five hides or to a multiple of the five-hide unit. Thus the rural townships are brought into line with towns, and we learn that in both the assessment was based on the five-hide unit.

Let us now take a typical Hundred and test this theory in practice:

Here we have five Vills varying in area from eleven ploughlands to twenty, and in value T.R.E., from £11 10s to £20, all assessed alike at ten hides each. What is the meaning of it? Simply that ASSESSMENT BORE NO RATIO TO AREA OR TO VALUE in a Vill, and still less in a Manor.

[pg 49]

Assessment was not objective, but subjective; it was not fixed relatively to area or to value, but to the five-hide unit. The aim of the assessors was clearly to arrange the assessment in sums of five hides, ten hides, etc.

Take now the next Hundred in the Inq. Com. Cant.:

Here again we have seven Vills varying in area from thirteen and a quarter ploughlands to twenty, and in value from £8 8s to £19 5s, all uniformly assessed at ten hides each. The thing speaks for itself. Had the hidation in these two Hundreds been dependent on area or value, the assessments would have varied infinitely. As it is, there is for each Vill but one and the same assessment.

Note further that the I.C.C. enables us to localize holdings the locality of which is unnamed in Domesday: also, that it shows us how certain Vills were combined for the purpose of assessment. Thus Borough Green and Westley are treated in Domesday as distinct, but here we find that they were assessed together as a ten-hide block. By this means we are enabled to see how the five-hide system could be traced further still if we had in other districts the same means of learning how two or three Vills were thus grouped together.

We may now take a step in advance, and pass to the Hundred of Whittlesford.

Hundred of Whittlesford
(Inq. Com. Cant., pp. 38-43)
	Hides			Ploughlands							Valets
Whittlesford	12		20	11		20	£15	2	0		£34	2	0
Sawston	8			9			19	0	0
Hinxton			20			16					20	10	0
Icklington			20			24½					24	5	0
Duxford			20			20¼					27	5	0
			—			——					——	—	—
			80			80¾					£106	2	0

[pg 50] Here we are left to discover for ourselves that Whittlesford and Sawston were grouped together to form a twenty-hide block. And on turning from the above figures to the map we find the discovery verified, these two Vills jointly occupying the northern portion of the hundred. Thus, this hundred, instead of being divided like its two predecessors into ten-hide blocks, was assessed in four blocks of twenty hides each, each of them representing one of those quarters so dear to the Anglo-Saxon mind (virgata, etc.), and lying respectively in the north, south, east and west of the district. Proceeding on the lines of this discovery, we come to the Hundred of Wetherley, which carries us a step further.

Hundred of Wetherley
(Inq. Com. Cant., pp. 68-83)
	Hides			Ploughlands
Comberton	6		20	7		32
Barton	7			12
Grantchester	7			13
Haslingfield			20			2288
Harlton	5		20	7		27⅞
Barrington	10			15⅜
Shepreth	5			5½
Ordwell	4		20	55⁄16		293⁄16
Wratworth	4			5⅜
Whitwell	4			5
Wimpole	4			5
Arrington	4			8½
			—			——
			80			1111⁄16

It is important to observe that, though the grouping is my own, the order of the Vills is exactly that which is given in the Inq. Com. Cant., and by that order the grouping is confirmed. Note also how, without such grouping, we should have but a chaos of Vills, whereas, by its aid, from this chaos is evolved perfect symmetry. Lastly, glance at the four 'quarters' and see how variously they are subdivided.

Advancing still on the same lines, we approach the very remarkable case of the adjoining Hundred of Long Stow.

Now it is necessary to explain at the outset that, the Inq. Com. Cant. being here imperfect, it only gives us the first two of the above 'quarters', its evidence ending with Bourne. But, by good fortune, it is possible to reconstruct from Domesday alone the remaining half of the Hundred, and thus to obtain the most valuable example of the system we are engaged in tracing that we have yet met with. The grouping I have adopted is based on the figures, but in some [pg 51] cases it is obvious from the map: Eltisley and Croxton, for instance, which form a ten-hide block, occupy a projecting portion of the county all to themselves, while Caxton adjoins them.

Hundred of Longstow
(Inq. Com. Cant., pp. 83-89)
	Hides					Ploughlands
Eversden	8⅓				25	13⅜		381⁄16
Kingston	8⅓					89⁄16
Toft and Hardwick	8⅓					16⅛
Grandsen	5				25	9		32½
Bourne	20					[23
Gamlingay			20		25
Hatley	4¼		5
[Unnamed]	¾
Croxton	7		10		25
Eltisley	3
Caxton			10
Caldecot	1¾		5
Long Stow	3¼
					——
					100

Several points are here noticeable. Observe, in the first place, how the twenty-five hide 'quarter' which heads the list is divided into three equal blocks of 8⅓ hides each, just as we found in Wetherley Hundred that one of the twenty-hide 'quarters' was divided into five equal blocks of four hides each. In these cases the same principle of simple equal division was applied to the quarter hundred as we saw applied to the whole hundred in the first two cases we studied—the Hundreds of Staines and of Radfield. Notice next how the two Vills of Toft and Hardwick, which are separately surveyed in Domesday under their respective names, are found from the Inq. Com. Cant. to have combined (under the name of 'Toft') in a block of 8⅓ hides. Lastly, it should not be overlooked that the ¾ hide not localized in Domesday fits in exactly with Hatley to complete its five hides.

The chase now becomes exciting: it can no longer be doubted that we are well on the track of a vast system of artificial hidation, of which the very existence has been hitherto unsuspected. Let us see what further light can be thrown by research on its nature.

On looking back at the evidence I have collected, one is struck, surely, by the thought that the system of assessment seems to work, not as is supposed, up from, but down to the Manor. Can it be possible that what was really assessed was not the Manor, nor even the Vill, but the Hundred as a whole? This view is so revolutionary, so subversive of all that has ever been written on the subject, that it cannot be answered off-hand. We will therefore begin by examining [pg 52] the case of the Hundred of Erningford, which introduces us to a further phenomenon, the reduction of assessment.

Here we have, as in the last instance, a Hundred of exactly a hundred hides (assessment). But we are confronted with a new problem, that of reduction. Before we form any conclusions, it is important to explain that this problem can only be studied by the aid of the Inq. Com. Cant., for the evidence both of Domesday and of the Inq. El. is distinctly misleading. Reduction of assessment is only recorded in these two documents when the Manor is identical with the Vill. In cases where the Vill contains two or more Manors, the Vill is not entered as a whole, and consequently the reduction on the assessment of that Vill as a whole is not entered at all.

After this explanation I pass to the case of the above Hundred, in which the evidence on the reduction is fortunately perfect. The first point to be noticed is that in four out of the five Hundreds that we have as yet examined, there is not a single instance of reduction, whereas here, on the contrary, the assessment is reduced in every Vill throughout the Hundred. That is to say, the reduction is conterminous with the Hundred. Cross its border into the Hundred of Wetherley, or of Triplow, and in neither district will you find a trace of reduction. Observe next that the reduction is uniform throughout the whole, being 20 per cent in every instance. Now what is the inevitable conclusion from the data thus afforded? Obviously that the reduction was made on the assessment of the Hundred as a whole, and that this reduction was distributed among its several Vills [pg 53] pro rata.⁸⁹ Further research confirms the conclusion that these reductions were systematically made on Hundreds, not on Vills. There is a well defined belt, or rather crescent, of Hundreds, in all of which the assessment is reduced. They follow one another on the map in this order: Erningford, Long Stow, Papworth, North Stow, Staplehow, and Cheveley. Within this crescent there lies a compact block of Hundreds, in no one of which has a single assessment been reduced. They are Triplow, Wetherley (? Cambridge⁹⁰), Flendish, Staines, Radfield, Chilford and Whittlesford. Beyond the crescent there lie 'the two Hundreds of Ely', in which, so far as our evidence goes, there would seem to have been similarly no reduction. As the two horns of the crescent, so to speak, are the Hundreds of Erningford and Cheveley, we will now glance at the latter, and compare the evidence of the two.

Hundred of Cheveley
(Inq. Com. Cant., pp. 9-11)
	Hides
	T.R.E.			T.R.W.			Ploughlands
Silverley	6½		10	4		6	8		12
Ashley	3½			2			4
Saxon Street	5			3			793
Ditton	5			392		(or 4)	10
Ditton	10			1			16
Kirtling	10			6			21
Cheveley	1091
	—
	50

As a preliminary point, attention may be called to the fact that the grouping of Ashley and Silverley, although they are surveyed separately in the Inq. Com. Cant., is justified by their forming, as 'Ashley-cum-Silverley' a single parish. So too, Saxon Street may be safely combined with Ditton, in which it is actually situate. We thus have a Hundred of fifty hides divided into five blocks of ten hides each, and thus presenting a precise parallel to the Hundred of Staines, the first that we examined.

[pg 54]

And now for the reductions. As the Vill of Cheveley, unluckily, is nowhere surveyed as a whole, we have in its case no evidence. But of the five remaining Vills above (counting Ashley-cum-Silverley as one), four we see had had their assessments reduced on a uniform scale, just as in the Hundred of Long Stow. Now this is a singular circumstance, and it leads me to this conclusion. I believe that, precisely as in the latter case, the assessment of the Hundred as a whole was reduced by twenty hides. This was equivalent to 40 per cent, which was accordingly knocked off from the assessment of each of its constituent Vills. One of the Dittons is clearly an exception, having nine hides, not four, thus knocked off. I would suggest, as the reason for this exception, that Ditton having now become a 'dominica villa regis' (Inq. Com. Cant., p. 10), was specially favoured by having a five-hide unit further knocked off its assessment, just as in the case of Chippenham (Ibid., p. 2).⁹⁴

It has been my object in the above argument to recall attention to the corporate character, the solidarité of the Hundred. This character, of which the traces are preserved in its collective responsibility, even now, for damages caused by riot, strongly favours the view which I am here bringing forward, that it was the Hundred itself which was assessed for geld, and which was held responsible for its payment. Although this view is absolutely novel, and indeed destructive of the accepted belief, it is in complete harmony with the general principle enunciated by Dr Stubbs, and is a further proof of the confirmation which his views often obtain from research and discovery. Treating of 'the Hundred as an area for rating', he writes thus:

There can be no doubt that the organization of the Hundred had a fiscal importance, not merely as furnishing the profits of fines and the produce of demesne or folkland, but as forming a rateable division of the county.⁹⁵

Now there are several circumstances which undoubtedly point to my own conclusion. We know from the Inq. Com. Cant., that the Domesday Commissioners held their inquiry in the Court of each Hundred, and had for jurors the men of that Hundred. Now if the Hundred, as I suggest, was assessed for geld as a whole, its representatives would be clearly the parties most interested in seeing that each Vill or Manor was debited with its correct share of the general liability. Again we know from the Inquisitio Geldi that the geld was collected and paid through the machinery of the Hundred; and its collectors, in Devonshire, are 'Hundremanni'. The Hundred, in fact, [pg 55] was the unit for the purpose.⁹⁶ Further, we have testimony to the same effect in the survey of East Anglia. But as that survey stands by itself, it must have separate treatment.⁹⁷

I need not further discuss the collective liability of the Hundred, having already shown in my 'Danegeld' paper how many allusions to it are to be found in Domesday in the case of urban 'Hundreds'.⁹⁸ It is only necessary here to add, as a corollary of this conclusion, that the assessment of a single Manor could not be reduced by the Crown without the amount of that reduction falling upon the rest of the Hundred. Either therefore, that amount must have been allowed ('computatum') to the local collector as were terræ datæ to the sheriff, or (which came to the same thing) the assessment on the Hundred must have been reduced pro tanto.

I now proceed to apply my theory that the Hundreds themselves were first assessed, and that such assessments were multiples of the five-hide unit.

We are enabled from the Inq. Com. Cant., to determine the assessments of eleven Hundreds.⁹⁹ Nine out of these eleven Hundreds prove to have been assessed as follows:

This list speaks for itself, but it may be as well to point out how convenient for the Treasury was this system. At the normal Danegeld rate of two shillings on the hide, an assessment of fifty hides would represent £5, one hundred hides £10, and so on.

Can we discover in other counties traces of this same system? Let us first take the adjacent county of Bedfordshire.

I am anxious to explain that for the means of utilizing the Bedfordshire evidence I am entirely indebted to the Digest of the Domesday of Bedfordshire by the late Rev. William Airy (edited by his son, the Rev. B. R. Airy¹⁰¹). It was, most happily, pointed out to the author [pg 56] by the Rev. Joseph Hunter 'that what we want is not translations but analyses of the surveys of the several counties' (p. viii). To this most true remark we owe it that Mr Airy resolved to give us a 'digest' instead of that usual 'extension and translation', which is perfectly useless to the Domesday student. It is easy to take from the record itself such an instance as these Beauchamp Manors entered in succession (213): Willington 10 hides, Stotford 15; 'Houstone' 5, Hawnes 5, 'Salchou' 5, Aspley 10, Salford 5; but it is only Mr Airy's work that enables us to reconstruct the townships, and to show how fractions—apparently meaningless—fit in, exactly as in Cambridgeshire, with one another. His work is all the more valuable from the fact that he had no theory to prove, and did not even add together the factors he had ascertained. His figures therefore are absolutely free from the suspicion that always attaches to those adduced to prove a case.

[pg 57]

Of these fifteen ten-hide townships, the last is selected as an instance of those slight discrepancies which creep in so easily and which account for many apparent exceptions to the rule. Passing to other multiples of the five-hide unit we have:

Marston					Roxton			Dunton
H.			V.		H.	V.		H.			V.
10		2	(less ½ virg.)		1	1		10		8	1
		8	(plus ½ virg.)		0	4				1	3
5		1			1	1		10		5	0
		½			7½	1				4½	0
		3			8	3				½	0
		½
—————————					——————			——————
15			0		20	0		20			0

I now give three illustrations of slight discrepancies:

Streatley			Sutton					Eaton Socon
H.	V.		H.			V.		H.	V.
1	0		5		0	3		20	0
4	1				1	0		6	3
4⅓	0				1½	0		0	1½
0	⅔				½	0		0	½
0	⅔				0	3½		9	1
					0	1½		0	5½
					2	0		2	½
					0	3		0	1
					½	0
					0	1½
					1	0
——————					——————			——————
9	3⅔				9	0½		40	1

[pg 58]

In the first case there is a deficiency of ¹⁄120, and in the second of ⁷⁄80, while in the third we find an excess of ¹⁄160. No one can doubt that these were really ten-hide, ten-hide, and forty-hide townships. We have to allow, in the first place, for trivial slips, and in the second for possible errors in the baffling work of identification at the present day. One can hardly doubt that if a student with the requisite local knowledge set himself to reconstruct, according to Hundreds, the Bedfordshire Domesday, he would find, as in Cambridgeshire, that even where a township was not assessed in terms of the five-hide unit, it was combined in an adjacent one in such an assessment.

We will now cross the border into Huntingdonshire, and enter the great Hundred of Hurstingston. This, which may be described as a double Hundred, was assessed, Domesday implies, at 200 hides. Quartering this total, on the Cambridgeshire system, we obtain fifty hides, and this quarter was the assessment allotted to the borough of Huntingdon.¹⁰² The total assessment of the Hundred was thus accounted for:

			Hides
Huntingdon			50
St. Ives (Slepe)			20
Hartford			15
Spaldwick			15
Stukeley			10
Abbots Ripton			10
Upwood			10
Warboys			10
Calne	6
Bluntisham	6½		20½103
Somersham	8
Wistow104			9
Holywell			9
Houghton			7
Wyton			7
Broughton			4
Catworth			4105
			———
			200½

[pg 59]

Passing on into Northamptonshire, we come to that most curious document, which I shall discuss below (see p. 124), and which was printed by Ellis (Introduction to Domesday, i. 187 et seq.). Ellis, however, can scarcely have read his own document, for he speaks of it as a list 'in which every Hundred is made to consist of a hundred hides'.¹⁰⁶ This extraordinary assertion has completely misled Dr Stubbs, who writes:

The document given by Ellis as showing that the Hundreds of Northampton each contained a hundred hides seems to be a mere attempt of an early scribe to force them into symmetry.¹⁰⁷

It is greatly to be wished that some one with the requisite local knowledge should take this list in hand and work out its details thoroughly. In capable hands it should prove a record of the highest interest. For the present I will only point out that its contents are in complete harmony with the results that I obtained on the Hundred in Cambridgeshire; for it gives us Hundreds assessed at 150 (four), 100 (nine), 90 (two), 80 (four), 60 (one), and 40 (one) hides, with a small minority of odd numbers. This list throws further light on the institution of the Hundred by its recognition of 'double' and 'half' Hundreds. Note also in this connection the preference for 100-hide and fifty-hide assessments, which here amount to thirteen out of the twenty instances above, and in Cambridgeshire to four out of nine. These signs of an endeavour to force such assessments into terms of a fifty-hide unit will be dealt with below.¹⁰⁸

In Hertfordshire, as indeed in other counties, there is great need for that local research which alone can identify and group the Domesday holdings. So far as single Vills are concerned, Bengeo affords a good illustration of the way in which scattered fractions work out in combination.

	H.		V.
Count Alan		0	1
Hugh de Beauchamp		6	0
Geoffrey de Mandeville		3	1
Geoffrey de Bech		5	1
		6½	0
		1	1½
		0	5½
		0	3½
Peter de Valognes		0	½
		—————
		25	0

[pg 60]

If we now push on to Worcestershire, we find a striking case in the Hundred (or rather the triple Hundred¹⁰⁹) of Oswaldslow. Its assessment was 300 hides;¹¹⁰ and I am able to assert that of these we can account for 299, and that it contained Manors of 50, 40, 35, 25 (two), and 15 hides.¹¹¹ We have also, in this county, the case of the Hundred of Fishborough, made up to 100 hides, and remarkable for including in this total the fifteen hides at which Worcester itself was assessed. The special value of this and of the Huntingdon instances lies in its placing the assessments of a borough on all fours with the assessment of a rural Manor, as a mere factor in the assessment of a rural Hundred. By thus combining town and country it shows us that the assessments of both were part of the same general system. This is a point of great importance.

This case of the Hundred of Fishborough is, however, peculiar. The entry, which was prominently quoted by Ellis (who failed to see its true significance), is this:

In Fisseberge hundred habet æcclesia de Euesham lxv. hidæ. Ex his xii. hidæ sunt liberæ. In illo Hundredo jacent xx. hidæ de dodentreu. et xv. hidæ de Wircecestre perficiunt hundred.¹¹²

Now this entry is purely incidental, and its real meaning is this. In the true Hundred of Fishborough (adjoining Evesham on the east), Evesham Abbey held sixty-five hides (assessed value), of which twelve were exempted from payment of geld, a statement which can be absolutely verified from the details given. To this aggregate was added the fifteen hides of Worcester (though in another part of the county), together with twenty hides of the distant Hundred of Doddentree. A total of 100 hides was that arrived at. Now the Hundred of Doddentree had itself made up to about 120 hides,¹¹³ by the addition of eighteen hides, which belonged to Hertford as to 'firma'.¹¹⁴ A reduction, therefore, of twenty hides suggests a complicated process of levelling the local Hundreds, which may remind us how large a margin must be allowed for these arrangements.

Before leaving Worcestershire, attention should be called to the great Manor of Pershore, which Westminster Abbey held for [pg 61] 200 hides, and to the 100 hides connected therewith under the heading 'Terra sanctæ Mariæ de Persore'.

In Somerset we find some good instances, with the help of Mr Eyton's analyses.

Hundred of Crewkerne
Merriott (5 + 7)	12		15
Seaborough (1½ + 1½)	3
Hinton St. George	13		25
In Crewkerne	12
			—
			40
Hundred of Whitstone
East Pennard (19 + 1)			20
Baltonsborough			5
Doulting (14 + 3¼ + 2¾)			20
Batcombe (10¼ + 2 + 7¾)			20
Ditcheat (5 + 5½ + 6½ + 5½ + 1 + 7)			30½
Pilton (6½ + 3 + 5 + 5 + 2)			21½
Stoke St. Michael			3
			——
			120

There are also abundant cases of Manors which work out similarly such as Walton and its group (4½ + 5 + 3 + 2 + 3 + 2½ = 20), Butleigh (7½ + 8 + 2 + ½ + 2 = 20). Again, in the Hundred of Frome we find eight Manors (Camerton, Englishcombe, Charterhouse Hinton, Norton St Philip, Corston, Beckington, Cloford, and Laverton), assessed at ten hides each, in addition to divided Manors, such as Road (9 + 1), and Tiverton (7½ + 2½).¹¹⁵

We will now pass to Devon and examine the assessments of its Hundreds. Of these thirty-one are entered in the Inquisitio Geldi. Now, as four virgates went to the hide, such assessments as 25¾, 9¼ hides, show us that the simple doctrine of probability is in favour of only one Hundred in every twenty proving to be assessed in multiples of the five-hide unit. Yet we find that those so assessed form an absolute majority of the whole. When classified, they run thus—50 (four), 40 (one), 30 (two), 25 (four), 20 (five): total, 16 Hundreds.

It will at once be observed that these assessments are, as nearly as possible, on one half the scale of those we met with in Cambridgeshire and Northamptonshire. But this must be taken in conjunction with the fact that the Devon and Cornwall assessments are altogether peculiar. 'In Devon and Cornwall, where the scope of the gheld-hide was enormous, it was necessary to introduce another quantity, intermediate between the virgate and the acre. This was the [pg 62] Ferndel or Ferdingdel, to wit, the fourth part of the next superior denomination, the fourth part of the virgate.'¹¹⁶ One might at first sight be tempted to suggest that the hide was in these two counties a term of higher denomination when we find Manor after Manor assessed at a fraction of a hide, while in Cornwall the 'acra terræ' was clearly a peculiar measure.¹¹⁷ Yet in some Manors adjacent to Exeter or to the neighbouring coast the assessment is much less abnormally low, though even there moderate. There is much scope, here also, for intelligent local research, although we may conclude, from the evidence of the Pipe Rolls, that the hide represented the same unit here as elsewhere, as it would seem did the Devonshire Hundred, in spite of its singularly low average assessment. Indeed, it represented a larger, not a smaller, area than usual. I shall deal with this phenomenon below, and endeavour to explain its significance. For the present it is only necessary to insist on the evidence that the Hundreds afford of assessment on the five-hide system.

Indeed, though I definitely advance the suggestion that the assessment was, in the first instance, laid upon the Hundred itself, and that the subsequent assessment of its Vills and Manors was arrived at by division and subdivision, the truth or falsehood of this theory in no way affects the indisputable phenomenon of the five-hide unit. On the prominence of that unit I take my stand as absolute proof that the hide assessment was fixed independently of area or value, and that, consequently, all the attempts that have been made by ingenious men to discover and establish the relation which that assessment bore to area, whether in Vill or Manor, have proved not only contradictory among themselves, but, as was inevitable, vain.

The late Mr Eyton did much to destroy the old belief held by Kemble and other well-known writers that the Domesday hide was an areal measure and to substitute the sounder view that it was used as a term of assessment, and Mr Chester Waters, in his Survey of Lindsey (1883), claimed that the 'key to the puzzle' had been thus finally discovered. Canon Taylor, on the other hand, at the Domesday Commemoration (1886), claimed that if his own most ingenious theory of the relation of the geld-carucate to area could be more generally extended, 'many volumes of Domesday exposition, including, among others, Mr Eyton's Key to Domesday, may be finally consigned al limbo dei bambini'.¹¹⁸ Mr Pell's theories—the inclusion of which at enormous length in Domesday Studies¹¹⁹ cannot be too[pg 63] deeply regretted—require a passing notice. According to him, the Domesday hide was virtually an areal term; but the interests of truth and of historical research require, as to his confident calculations, very plain speaking. Although I devoted to the investigation of Mr Pell's theories a deplorable amount of time and labour,¹²⁰ I would rather state the inevitable conclusion in the words of that sound scholar, Mr W. H. Stevenson:

All the fanciful calculations that Mr Pell has based upon this assumption, including his delicious 'Ready Reckoner', may be safely left to slumber in oblivion by the Domesday student who does not wish to waste his time.

The only abiding principle underlying Mr Pell's calculations is that the figures in Domesday, or wherever found, have to produce a certain total that Mr Pell has already fixed upon. To do this, virgates may mean hides, carucates may mean virgates, and, in short, anything may mean anything else.¹²¹

Although Mr Eyton also indulged in 'fanciful calculations', and committed the fatal error of combining facts and fancies, he was at least on the right track in discarding the notion that the Domesday hide denoted a fixed area, and in treating it as a term of assessment. At the same time, the acceptance of my theory that this assessment was not determined by the real value of the Manor or Vill, but was unconnected with it, would be, of course, destructive of all his calculations.

The five-hide unit which lies at the root of my theory is found ever to the front, turn where we will. In Oxon¹²² we find entered in succession the Bishop of Lincoln's Manors 90, 60, 40, 50, 50 hides, while if we work through the southern extremity of the county (lying south of Ewelme), following the bend of the Thames, we find the assessments are as follows: Preston Crowmarsh, 5; Crowmarsh Gifford, 10; Newnham Murren, 10; Mongewell, 10; Ipsden, 5; North and South Stoke, 20¼; Checkenden, 5; Goring, 20; Gethampton, 6½; Whitchurch, 10; Mapledurham, 10; Caversham, 20; Dunsden, 20; Bolney (8) and Lashbrook (12) 20; Harpsden, 5; Rotherfield, 10; Badgemoor, 5; Bix 5. So too on the western border we have in succession Churchill, 20; Kingham, 10; Foxcote (1) and Tilbury (14), 15; Lyneham, 10; Fyfield, 5; Tainton, 10; Upton, 5; Burford (8) and Widford (2), 10; Westwell, 5.¹²³

Berkshire undoubtedly offers a fruitful sphere of study. On the one hand, we have so large a proportion of Manors assessed at [pg 64] 5, 10, 15, 20 hides, and so forth as to strike the reader at once without special research; on the other, we find these archaic assessments reduced under the Conqueror in the most sweeping manner, and the old system thus effaced. Fortunately for us in this case its existence is recorded in the Domesday entries of the previous assessments. What is here, as elsewhere, wanted is a thorough local analysis of the hidage, Hundred by Hundred. For no county is such an analysis more urgently needed.

In Bucks the Primate's three Manors are of 40, 5, 30 hides, while nine Manors of Walter Giffard follow one another with these assessments: 20, 10, 10, 20, 3½, 10, 5, 5, 10; and in Gloucestershire we are met on every side by Manors of 5, 10, 15, 20 hides, and so on. In Surrey, the Primate's six Manors are assessed at 30, 20 80, 5, 20, 14 hides. As a proof that this feature is in no way of my own creation, I will take the Wiltshire Manors selected by Mr Pell for his tables. Seven out of the eleven selected by him are five-hide assessments, being 5, 10, 20, 40, 20, 5, 10. The marvel is that any one can have failed to observe the general occurrence of the fact.

In Middlesex the five-hide unit is peculiarly prominent. We have only to glance at the pages of Domesday to be struck by such assessments as Harrow (100 hides), Fulham (50 hides¹²⁴), Isleworth (70 hides), Harmondsworth (30 hides), while on folios 129b-130, we have seven Manors in succession of which the assessments are 15, 35, 30, 30, 7½, 15, 10, representing 3, 7, 6, 6, 1½, 3, 2, multiples of the five-hide unit. But, here again, conspicuous as is this unit even in the case of Manors, its prevalence would be still more apparent, if we could reconstruct the Vills. Thus, for instance, in the Hundred of Spelthorne we find these assessments:

[pg 65]

'Exeforde' is Ashford, which 'appears from a very early period till after the dissolution of the monasteries to have been an appendage of Stains'.¹²⁷ Thus we obtain an assessment of 20 hides for Staines cum Ashford. So too we have at once for Laleham an assessment of ten hides, while that of East and West Bedfont was, we see, twenty hides. The most striking case, however, is that of Hatton; for, if we add to its two named Manors the nameless estates in the above list, the four fit in like a puzzle, giving us an aggregate assessment of exactly five hides.

The hundred, therefore, was assessed thus:

Let us now connect the territorial with the institutional unit. Dealing in my 'Danegeld' essay with the evident assessment of towns in terms of the five-hide unit, I traced it to the fact that 'five hides were the unit of assessment for the purpose of military service'.¹²⁸ The evidence I have adduced in the present paper carries further its significance; but we must not allow its financial to obscure its military importance. I appealed, at that time, to the Exeter instance:

Quando expeditio ibat per terram aut per mare serviebat hæc civitas quantum v. hidæ terræ;

and to the service of Malmesbury:

Quando rex ibat in expeditione vel terra vel mari habebat de hoc burgo aut xx. solidos ad pascendos suos buzecarlos aut unum hominem ducebat secum pro honore v. hidarum.¹²⁹

Of course this brings us to the notoriously difficult question of the thegn and his qualification. With this I am only concerned here so far as it illustrates the prevalence of a five-hide unit. Mr Little, who holds that Maurer, followed by Dr Stubbs, has gone too far, and that 'there is no proof of any general law or widely prevalent custom which conferred on the owner of five hides pure and simple the title, duties, and rights of a thegn',¹³⁰ sets forth his view thus:

What then is the meaning of the frequent recurrence in the laws of possession of five hides of land as the distinctive mark of a particular rank?

[pg 66]

An explanation may be hazarded: at the end of the seventh century it was the normal and traditional holding of a royal thegn.... It is not too much to infer from the parallelism of the two wergelds, that five hides formed also the regular endowment of a Saxon king's thegn.¹³¹

Dr Stubbs' views will be found in his Constitutional History (1874), i. 155-6, 190-2, and those of Gneist in his Constitutional History (1886), i. 13, 90, 94. The latter writer follows Schmidt rather than Maurer, but sums up his position in the words: 'Since under Ælfred and his successors every estate of five hides is reckoned in the militia system as one heavy-armed man, the rank of a thane becomes the right (as such) of a possessor of five hides.'

Lastly, it is an interesting and curious fact that we owe to the five-hide unit such place-names as Fivehead, Somerset; Fifehead, Dorset; Fifield, Oxon; Fifield and Fyfield, Wilts; Fyfield, Hants; and Fyfield, Essex—all of them in Domesday 'Fifhide' or 'Fifehide'—as well as Fyfield, Berks, which occurs in Domesday as 'Fivehide'. Philologists will note the corruption and its bearing on the original pronunciation.

To the probable antiquity and origin of the five-hide system I must recur, after glancing at the evidence for the northern and eastern districts of England.

VII. THE SIX-CARUCATE UNIT

The subject that I now approach is one of the highest interest. I propose to adduce for my theory convincing corroborative evidence by showing that the part which is played in the hidated district of England by the five-hide unit is played in the Danish districts by a unit of six carucates. In other words, where we look in the former for 'v. hidæ', we must learn to look in the latter for 'vi. carucatæ terræ'.

One must dissociate at the outset this six-carucate unit from the 'long hundred', or Angelicus numerus, with which Mr Pell confused it. In Mr Stevenson's instructive article on 'The Long Hundred and its use in England',¹³² he has clearly explained that this reckoning only applied to a whole hundred, which, if a 'long' hundred, was really 120. Any lesser number was reckoned in our usual manner. This is seen at once in the test passage at Lincoln (D.B., i. 336a), where 1,150 houses are reckoned as 'novies centum et lxx.', because 'hic numerus Anglice computatur, id est centum pro cxx'.¹³³ The persistence, [pg 67] in Lincolnshire, of the long hundred is well shown in the Inquisitiones post mortem on Robert de Ros, 1311, among those printed by Mr Vincent.¹³⁴ We there read of 'c. acre terra arrabilis per majorem centenam que valent per annum lx. s. prec' acre vj. den.', at Wyville and Hungerton (on the border of Leicestershire); while at Claxby and Normanby (in the north of the county) we have 'cc. acras per minorem centenam et valent c. s. prec' acre vj. d.' Again, at Gedney (in the south-east), we have 'cc. acre terre arrabilis per majus centum et valent per annum xxiiij. li'. prec' acre ij. s. et iiij^xx. acre prati et valet per annum viij. li., prec' acre ij. s. Et cxiij. acre pasture per majus centum et valent per annum ix. li. xix. s. vi. d. prec' acre xviij. d.' On the same property there were due 'ccciiij^xx. opera autumpnalia cum falcis, et valent xxxvj. s. viij. d., prec' operis j. den.', so that these also were reckoned by the long hundred.

Mr Stevenson was not aware of this evidence, but admitted that as the Domesday passage refers to 'such a Danish stronghold as Lincolnshire, it is not free from the suspicion of Danish influence'. His own evidence from a sixteenth-century rental¹³⁵ is subject to a similar criticism. For the general use, therefore, of the 'long hundred' in England he is compelled to rely on the Dialogus de Scaccario and Howden's description of the new survey of 1198, the 'hide or ploughland' being described in both cases as of a hundred acres, where the 'hundred' must have meant 120. But I venture to think that the use of this reckoning for the ploughland, or archaic 'hide', does not establish its general employment. In Domesday, certainly, it is only at Lincoln that we find it actually recognized, houses being reckoned everywhere else on the usual system.

I think, therefore, that we fairly may hold the Anglicus numerus, or long hundred, to have specially prevailed in the 'Danish' districts, which were also assessed, we shall find, in sums of six and twelve. But what was the boundary of this Danish district? It was not the border between Mercia and Wessex, for Mercia was itself divided between the 'six' and the 'five' systems.¹³⁶ Of the two adjacent [pg 68] Mercian shires, for instance, of Leicester and Warwick (afterwards united under one sheriff), we find the latter decimal and the former duodecimal. The military service of Warwick and Leicester was arranged on the same method, yet Leicester sent twelve 'burgesses' to the fyrd where Warwick sent ten. But, it may be urged, the two shires were divided by the Watling Street, the boundary (under the peace of Wedmore) of Danelaw. Was then the Danelaw the district within which the systems prevailed? No, for the Danelaw, under this treaty, included all Cambridgeshire and other hidated districts. The answer, therefore, which I propound is this: The district in which men measured by carucates, and counted by twelves and sixes, was not the district which the Danes conquered, but the district which the Danes settled, the district of 'the Five Boroughs'.

Dependent on these 'Five Boroughs' were the four shires of Leicester, Derby, Nottingham, and Lincoln. For two of the Boroughs, Lincoln and Stamford, both belonged to this last shire, which was, indeed, the stronghold of the system.¹³⁷ Between Stamford and Cambridge we have the same contrast as between Warwick and Leicester, for while Cambridge was divided into ten wards ('custodiæ'), Stamford was divided into six. Lincolnshire, as I have said, was the stronghold of the system, and it is in Lincoln itself that we find Domesday alluding eo nomine to the Anglicus numerus, the practice of counting 120 as 100.

Now in the peculiar district of which I am treating there occurs an important formula which covers Lincolnshire, Yorkshire, Derbyshire, and Notts. Domesday has nothing like it for the other parts of England. Here are the three passages in which we find it recorded:

[pg 69]

For comparison with these three passages we may turn to the charter of immunities confirmed to York Cathedral by Henry I, Stephen, and Henry II. We there read:

Si quis enim quemlibet cujuscumque facinoris aut flagitii reum et convictum infra atrium ecclesiæ caperet et retineret, universali judicio vi. hundreth emendabit; si vero infra ecclesiam xii. hundreth infra chorum xviii. ... In hundreth viii. libræ continentur.¹³⁸

As there were twelve carucates in the 'Hundred', so it paid twelve marcs, which, if we can trust the above explanation, themselves came to be termed a 'Hundred'. Moreover, the 'Hundreds' themselves were grouped in multiples of six. So too the Yorkshire thegn who held six Manors or less paid three marcs to the sheriff; if he held more than six, twelve marcs to the king (Domesday, i. 289b).

It is a special feature of the 'Danish' district that each territorial 'Hundred' contained twelve 'carucatæ terræ'. This point is all-important. Just as a 'Hundred' to an Anglo-Saxon suggested one hundred 'hides', so to the Danes of this district it suggested twelve 'carucates'. Nay, to the men of Lincolnshire there could be no more question that twelve carucates made a 'Hundred' than there could be now, among ourselves that twelve pence make a shilling. If we turn to the Lindsey Survey,¹³⁹ a generation later than Domesday, we obtain proof to that effect. We find that Survey, in three instances, adding up all the estates of a tenant within a Wapentake, and giving us the result in 'Hundreds' and 'carucates'. Here are the actual figures:

[pg 70]

Now we must observe that these 'Hundreds' are not districts with 'a local habitation and a name'; they are merely sums of twelve carucates produced by compound addition. We further find, at the head of the survey of each Wapentake, a note that it is reckoned to contain so many 'Hundreds', with the explanation, in some instances that in each 'Hundred' were 'xii. carucatæ terræ'.¹⁴³ But even here the real unit is shown to be 'six carucates', for several Wapentakes contain an odd 'half-hundred', while in that of Horncastle this is actually entered as 'six carucates'.

Here are the nineteen Wapentakes, with the number of Hundreds assigned to each, and the number of 'carucatæ terræ' that such Hundreds would imply:

All the above, it will be seen, are multiples of the six-carucate unit. That the aggregate of recorded 'carucatæ terræ' appears to differ, though slightly, from the totals here given only shows how vain is the [pg 71] argument that, because the recorded aggregates of Hundreds may often be uneven figures, there could therefore have been no system at work such as I contend there was. Clerical error and special alterations have both to be allowed for.

It has never, so far as I know, been pointed out that these Lindsey Trithings were so arranged as to contain an approximately equal number of 'Hundreds'. So far as it is possible now to reckon them, the South Trithing contained 51½, the North Trithing 51½, and the West Trithing 49½. Fifty 'Hundreds' would represent 600 carucatæ; and it is, to say the least, a singular coincidence that, in the archaic territorial list that has hitherto baffled investigation, the North Gyrwa, South Gyrwa, and Spalda are reckoned each at 600 hides.¹⁴⁵

I shall now give some instances of Lindsey townships assessed on the basis of the six-carucate unit:

[pg 72]

These instances will illustrate the value of the Lindsey Survey in enabling us to group the fractional assessments which appear in Domesday Book. Here are some other varieties:

Lastly, to complete the parallel with the Leicestershire Hundreds infra, we may take this case (cf. p. 63, note 122):

[pg 73]

Precisely the same system prevailed in Holland as in Lindsey, for the 'Testa de Nevill' preserves for us the constituents of a Holland Wapentake, that of 'Elhou':

Pinchbeck			12
Spalding			12
Weston			6
Moulton			6
Whaplode and Holbeach			18
Fleet			6
Gedney	8		12
Lutton	4
Sutton	9¾		12
Tydd	2¼
			—
			84

The Lindsey Survey would describe such a Wapentake as containing 'Seven Hundreds'.

Crossing the border from Lincolnshire into Rutland (i.e. the Rutland of Domesday), we find the same system at work that meets us in the Lindsey Survey. We read:

In Alfnodestou Wapent' sunt ii. Hundrez. In unoquoque [sunt] xii. carucatæ ad geldum.... In Martinesleie Wap' est i. hundret, in quo xii. carucatæ ad geldum.—D.B., i. 293b.

On analysing the contents of these Wapentakes, we find this statement fully borne out, the former containing twenty-four, and the latter twelve, 'carucatæ terræ'. These are carefully contrasted throughout with the 'terra carucæ' or areal measure.¹⁴⁶

In Yorkshire, Notts and Derby, we have less direct evidence. Sawley, in Derbyshire, has indeed been alleged to be entered in Domesday as a Hundred of twelve carucates, but Domesday does not justify this assertion being made.¹⁴⁷ I would rather trust to the notable formula, which, as I explained at the outset, is common to these counties for proof that they also were arranged in 'Hundreds' of twelve carucates.

The prevalence, however, of assessment by sixes, threes, and twelves, meets us on every side, as does, in hidated districts, the assessments by fives and tens. At the outset, for instance, of the survey of Yorkshire we have the district 'gelding' with the city assessed at eighty-four (12×7) carucates (which would be described [pg 74] in Lincolnshire as seven 'Hundreds'). We have two lists of the details, which are given here.¹⁴⁸

These lists have a value independent of their illustration of the arrangement in threes and sixes. They show how Domesday breaks down, when it supplies a check upon its own evidence, by failing to make its details agree with its total; and they further show by the discrepancy between them how easily error may arise, and how rash it must be to argue from a single case.¹⁴⁹

Yorkshire presents other traces, in its Hundreds, of the same system. Thus the townships in the Hundred of 'Toreshou' follow one another in this order: 18, 18, 20, 6, 18, 8, 12, 12 (8+4), 6, 18, 8, 18, etc. (infra, p. 80).

But my strong evidence is found in an invaluable survey of Leicestershire, unknown till now to historians,¹⁵⁰ which does for the carucated districts just what the Inq. Com. Cant. does for the hidated ones. Here we find the townships grouped in small blocks of from six to twenty-four 'carucatæ terræ', as a rule with almost monotonous regularity. And these blocks are further combined in small local Hundreds, of which the very existence is unknown to historians and antiquaries,¹⁵¹ and which are usually multiples, like the Lincolnshire Wapentake, of the six-carucate unit.

It will be remembered that in the case of Cambridgeshire, I selected for my first two examples a Hundred of 50 hides, composed of 5 Vills assessed at 10 hides each, and a Hundred of 70 hides, composed [pg 75] of 7 Vills, assessed at 10 hides each. In Leicestershire, precisely in the same manner, I shall begin with the simplest forms and select Hundreds of 36 and 48 carucates, composed of Vills uniformly assessed at 12 carucates each.

From these we may advance to other combinations:

Hundred of Harby
Harby and Plungar				18
Stathern				18
				—
				36
Hundred of Tong
Tong				12
Kegworth				15		18
Worthington				3
'Dominicum'				12
				—
				42
Hundred of Langton
Langton (1)		24		14½		(11¼ + 3¼)
Thorp (Langton)				3¾
Langton (2)				5¾
Tur Langton		24		12
Shangton				12		(10 + 2)
				—
				48

With these types as clues we are in a position to assert that where the total assessment of a Hundred varies but slightly from a multiple of six, there must have been some slight error in one of the figures. Thus Hundreds of 35½, 34¹³⁄16 carucates, etc., may be safely assumed to have been Hundreds of 36 carucates; those of 41, 43⅞, etc., [pg 76] would be of 42 carucates; those of 48⅞, 50, etc., would be of 48 carucates. These slight discrepancies, precisely as in Lincolnshire, are accounted for by Vills of 6 or 12 carucates, being entered as of 5⅞, 5¹³⁄16, 6¾, or 11⅞, 13, etc. Thus:

The most usual Leicestershire Hundreds are those of 36, 42, and 48 carucates, which, be it observed, would be described in the language of the Lindsey Survey as 'Wapentakes' of 3, 3½, and 4 'Hundreds' respectively. The name may be different: the thing is the same.¹⁵²

It will have been seen by this Survey that the 'Vills', single or grouped, were assessed precisely as in Cambridgeshire, save that there the assessment was reckoned in fives and tens, while here it was in sixes and twelves.

VIII. THE LEICESTERSHIRE 'HIDA'

The case of Leicestershire introduces us to a very curious point. Leicestershire is not one of those counties to which the singular formula that I discussed above refers. This suggests that it was not arranged in 'Hundreds' of twelve carucates. The above Survey confirms this, for it shows us Hundreds resembling in character those found in the hidated districts. But although the twelve-carucate unit of the 'Hundred' is not found in Leicestershire, we do find in it a group-unit, and that unit is the hida. Just as we have seen the Hundred used in two wholly different senses, so also was the 'hida'. The quite peculiar way in which 'hida' occurs in Leicestershire (which was not a hidated but carucated district) completely baffled Mr Eyton, and was misunderstood by Mr Pell.¹⁵³ Both writers failed to observe not only that the use of 'hida' is here of a peculiar character, but also that the normal 'hida' of Domesday (from which they could not emancipate themselves) would be quite out of place in this carucated district.

[pg 77]

The first point to grasp is that this Leicestershire 'hida' was a term which, locally I mean, explained itself. It is used at least a dozen times in the Survey of Leicestershire without any mention of its contents. Those contents must have been, therefore, familiar and fixed. But what were those contents? Three incidental notices enable us to determine them:

231 (a), 2: 'Ibi est i. hida et iiii^ta. pars i. hidæ. Ibi sunt xxii. car' terræ et dimidia.'

236 (a), 1: 'II. partes unius hidæ, id est xii. car' terræ.'

237 (a), 2: 'II. partes unius hidæ, id est xii. car' terræ.'

Just as the 'Hundred' of Lincolnshire was a sum of twelve carucates, so the 'Hide' of Leicestershire was a sum of eighteen carucates.¹⁵⁴ Working in the light of this discovery (for as such I claim it), we find that the other 'hides', thus interpreted, give us an aggregate of 'carucates' obviously suitable to the recorded ploughlands.¹⁵⁵ It may, however, be fairly asked why Domesday should speak in one place of half a 'hide', and in another of nine 'carucates'; in one place of a hide and a third, and in another of twenty-four carucates. The answer is that the singular love of variety which distinguishes Domesday in Cambridgeshire (as we saw) is at work here also. For instance, two equal estates are thus described: 'Willelmus iiii. car' terræ et dimidiam et iii. bovatas, et Rogerus iiii. car' terræ et vii. bovatas' (fo. 234a). The same instinct which led the scribe to enter these seven bovates as half a carucate plus three bovates, led him also to enter ten and a half carucates as half a hide plus a carucate and a half (fo. 237a).

But to the rule I have established there is a single exception. We read of 'Medeltone' in this shire: 'Ibi sunt vii. hidæ et una carucata terræ et una bovata. In unaquaque hida sunt xiiii. carucatæ terræ et dimidia' (fo. 235b). The actual formula employed is unique for the shire, and the figures are specially given as an exception. But, with singular perversity, Domesday students have always been inclined to pitch upon the exceptions as representing the rule, forgetting that it was precisely in exceptional cases that figures had to be given. In normal cases they would have been superfluous.

Several years have elapsed since I wrote the above explanation, but I have decided to publish it exactly as it originally stood. In the meanwhile, however, Mr Stevenson has dealt with the subject in an article on 'The Hundreds of Domesday: the Hundred of Land' (1890).¹⁵⁶ He has advanced the ingenious theory that the Leicestershire [pg 78] 'hida' was only a clerical error for H[undred], and that it was really that 'Hundred' of twelve carucates which we meet with in the Lindsey Survey. To prove this, he reads an entry on 236a, 1, as 'Ogerus Brito tenet in Cilebe de rege ii. partes unius hidæ, id est xii. car[ucatæ] terræ', and claims that this gloss defines the 'hida' as a 'hundred' of twelve carucates. I confess that to me such a rendering is in the highest degree non-natural. If we speak of 'two-thirds of a yard, that is twenty-four inches', we should clearly imply that the yard itself was thirty-six inches, not twenty-four. Similarly, I claim to render the 'gloss' as implying that the 'hida' itself contained eighteen carucatæ, not twelve.¹⁵⁷ If I am right, Mr Stevenson's suggestion that this 'hida' was really a 'Hundred' also falls to the ground.

After careful study of the Domesday Survey of Leicestershire, I definitely hold that in that county 'carucata terræ' was the geld-carucate and 'terra x car[ucis]' the actual ploughlands.¹⁵⁸ Now there are only three instances in which the Survey records the assessment both in terms of the 'hida' and in 'carucatæ terræ', and in all three the figures support my own theory. The Abbot of Coventry's Burbage estate (231a, 2), where a 'hide' and a quarter equates 22½ 'carucatæ terræ', is a test-case, and Mr Stevenson there takes refuge in a suggested 'beneficial hidation'. The exact formula, no doubt, is peculiar, but reference to the text shows that 's[un]t' has been interpolated between 'ibi' and 'xxii.' I suspect that the scribe had written 'ibi' (from the force of habit) when he ought to have written 'id est'.

I close this portion of my essay by applying my own theory to the case of 'Erendesbi' (Arnesby). The relative entries are:

'Episcopus Constantiensis tenet in Erendesber iii^as. car[ucatas] terræ et dim. et unam bovatam (231).'

'W[illelmus] Pevrel tenet dim. hidam et iii. bovatas terræ in Erendesbi (235).'

Put into figures they work out:

So that Arnesby was a typical Vill assessed at twelve carucates.¹⁵⁹

[pg 79]

IX. THE LANCASHIRE 'HIDA'

There is one other case of a peculiar 'hide' in Domesday. This is that which is found in the land 'between Ribble and Mersey', that district of which the description offers so many peculiarities. We find it divided into six hundreds, and of the 'hides' in the first, that of (West) Derby, we read: 'In unaquaque hida sunt vi. carucatæ terræ' (i. 269b). Whether or not that explanation applies, as is believed, to the whole district, we have here again a 'Danish' place-name brought into direct relation with the six-carucate unit. On the opposite bank of the Mersey lay the Wirral peninsula, in which this system of assessment cannot be traced.

Mr Green alluded to the Danish 'byes' as found, by exception, 'about Wirral in Cheshire',¹⁶⁰ and held that Norsemen from the Isle of Man had founded 'the little group of northern villages which we find in the Cheshire peninsula of the Wirral'.¹⁶¹ I cannot find them myself. In his 'Notes on the Domesday Survey, so far as it relates to the Hundred of Wirral'¹⁶² (1893), Mr Fergusson Irvine, in a paper which shows, though somewhat discursive, how much can only be done by intelligent local research, has collated all the Domesday entries. 'Raby' is the one place I can there find in the peninsula with the 'bye' termination; while out of fifty-one entries twenty refer to places with the English termination 'tone', and the Anglo-Saxon test-words 'ham' and 'ford' are found in four others. There were, doubtless, Norse elements in the peninsula, but they were not strong enough to change the place-names or divide the land on their own system. In the same way, Chester had its 'lawmen', though it was not one of the Five Boroughs, nor is what I have termed the Scandinavian formula applied to Cheshire in Domesday. So, too, there were lawmen at Cambridge, and their heriot included eight pounds,¹⁶³ which occur in the above formula as the twelve marcs of the Danish 'Hundred'. Yet the whole system of Cambridgeshire was non-Danish. It was only, in short, where the northern invaders had settled down as a people that they were strong enough to divide the land anew and organize the whole assessment on their own system.

X. THE YORKSHIRE UNIT

We have seen that the unit of assessment for the carucated districts of England was 'vi. carucatæ terræ', just as five hides was the old unit in the south. We have also seen that the former reckoning extended over those districts which the Danish immigrants had settled. There remains the question whether the Danes had merely [pg 80] substituted six for five in the pre-existing arrangement, or had made a wholly new one for themselves based on actual area.

It is primâ facie not probable that they can have adopted the latter course, for the uniformity of their assessment proves its artificial character. Yet, in his remarkable paper on 'The Ploughland and the Plough',¹⁶⁴ Canon Taylor has arrived at the conclusion that:

The geldable carucate of Domesday does not signify what the carucate usually signifies in other early documents. The 'carucata ad geldum' is not as commonly stated by Domesday commentators, the quantity of land ploughed in each year by one plough, but it is the quantity tilled in one year in one arable field by one plough.¹⁶⁵

This 'novel and important proposition', as its author truly described it, was probably the most notable contribution to our knowledge that the Domesday Commemoration produced. The Canon's theory, which (so far as his own East Riding is concerned) he certainly seems to have established, is, at first sight, fatal to mine. But, on the other hand, my own theory can be proved no less clearly for Leicestershire, where the 'carucata terræ' and the ploughs are often connected in about the same ratio as in Yorkshire.¹⁶⁶ This leads us to inquire whether, even in the East Riding (where his theory works best), we may not find traces of that assessment by the six-carucate unit which I advocate myself. Such traces in Yorkshire we have already seen,¹⁶⁷ but there is other and stronger evidence.

If we take the modern Wapentake of Dickering (the first on Canon Taylor's list) and examine its three Domesday Hundreds of Turbar, Hunton, and Burton, we obtain these results:¹⁶⁸

Turbar Hundred
Hundemanebi			24
Ricstorp, Mustone, Scloftone, and Neuton			18
Flotemanebi			6
Muston and Neuton			6
Fordun and Ledemare			6
Burton, Fulcheton, and Chelc			30
Chelc (2), Ergone, Bringeham, Estolf, Fodstone, and Chemelinge			19
Nadfartone			23¾
Pochetorp			6
Helmeswelle and Gartune			44
[pg 81] Hunton Hundred
Flaneburg and Siwardbi	24½
Marton	9
Bredinton	18
Hilgertorp	6
Wivlestorp and Basingebi	12
Frestintorp	9		29½
Eleburne	½
Eston	6
Bovintorp	14
Gerendele	12
Ricton, Benton and Spetton	24
Bocheton	12
Fleuston	14		27
Stactone	6
Foxhole	7
Burton Hundred
Burton	12
Grenzmore (4+2)	6
Arpen (4+8)	12
Chillon (30+11+7)	48
Roreston (9+3)	12
Logetorp (1½+5½)	7		36
Thirnon	7
Ascheltorp (4+2)	6
Torp	3
Cherendebi	13
Caretorp (5+4+3)	12
Rodestain (8+8+8)	24
Twenc	17¼
Suauetorp	9
Fornetorp (4+14)	18
Butruid	12
Langetou (9+6)	15		42
Buitorp	5
Bruneton	3
Galmeton	8
Binneton	6
Widlaueston	5

The evidence of this last Hundred is so overwhelming that it cannot be gainsaid.¹⁶⁹

I claim, therefore, that my theory holds good even in Canon Taylor's stronghold, but I do so without venturing to dispute the accuracy of his own. How far they can be reconciled I leave to others to decide.

There are certain difficulties, however, which his brilliant suggestion must raise. It is the essence of his theory that in a two-field Manor the ploughland of 160 acres (half fallow) was assessed at one 'carucata terræ', while in the three-field Manor the ploughland of 180 acres (a third fallow) was assessed at two. This would be an obvious and gross injustice. Again, remembering that, according to the Canon, the proportion of 'carucatæ' to ploughlands should be either 2 to 1 or 1 to 1, what are we to make of such figures as these, taken at a venture from a page of the Leicestershire Survey (232a, 1):

[pg 82]

It is certainly difficult to discover any regular or consistent assessment in a system where the ploughland was represented by anything from ½ carucata to 2¼ carucatæ. There is, however, in so many cases an approximation to an assessment of three carucatæ for two ploughlands, that there seems to have been some underlying idea, if we could only trace it out. But for this there is needed a special investigation of all the carucated counties, a work of great labour and requiring local co-operation. If we could have tables for each county, arranged Hundred by Hundred and Vill by Vill, showing in parallel columns the ploughland and the carucatæ ad geldum, we could then, and only then, venture to speak positively. Till that is accomplished we are not in a position to explain how a system of assessment, based on actual area, could result in aggregate assessments uniformly expressed in terms of the six-carucate unit.

XI. GENERAL CONCLUSIONS

In seeking a clue to the origin of that artificial assessment, of which the traces, whether more or less apparent, linger on the pages of Domesday, I propose to exclude the carucated district, because we require, as I have said, more complete evidence as to the system pursued within it, and because, being associated with the settlement of the Danes it represents a later introduction, while the very name 'carucate', as I observed in Domesday Studies, has, unlike the mysterious 'hide', an obvious connection with the ploughland. Confining ourselves to the district assessed in terms of the 'hide', we seek to learn the origin of the system by which, as I contend, it was divided for the purpose of taxation into blocks, each of which was expressed in terms of the five-hide unit.

Now if we follow the clue afforded by the Cambridgeshire evidence, and hold that the assessment was originally laid not on the Manor, nor even on the Vill, but on the Hundred as a whole,¹⁷⁰ [pg 83] it might be suggested that the Hundred itself subdivided the amount among its constituent elements. In practice, indeed, from the nature of the case, this principle must have prevailed in every town assessed at a Hundred or Half-Hundred, for where an urban community was assessed in 'hides' the burgesses must, as in later days, have settled among themselves the proportion to be borne by individuals or individual properties. If, then, they were able to do this, and if, as I hold, town and country were assessed on the same principle, as part of the same system, what was to prevent their neighbours, in the court of the rural Hundred, similarly distributing among its constituents their respective shares of the common burden?

We might even be tempted to go far further than this, and to carry our discoveries to a logical conclusion. If, as is asserted, direct taxation ('geld') began in England with the need for raising money to buy off the Danes, let us ask ourselves how the Witan would proceed when confronted with a demand, let us say, for £10,000. As there had been hitherto, ex hypothesi, no direct taxation, there would be no statistical information at their disposal, enabling them to raise by a direct levy the sum required. Their only possible resource, we might hold, would be to apportionate it in round sums among the contributory shires. Proceeding on precisely the same lines, the county court, in its turn, would distribute the quota of the shire among its constituent Hundreds, and the Hundred court would then assign to each Vill its share. As the Vills were represented in the Hundred court, and the Hundreds in the Shire court, the just apportionment of the Shire's quota would be thus practically secured. The arrangement would, moreover, be as satisfactory to the Witan as it was fair to the contributors inter se; for, by this gradation of responsibility, the payment of the whole was absolutely secured. This explanation is very tempting, and, indeed, such a system of apportioning liability is to be traced from time immemorial in the Indian village community.¹⁷¹ Moreover, if the ratio of 'hides' to ploughlands were found to vary to any marked extent, according to county, the hypothesis that the quota, in the first instance, was laid upon each county would duly explain the ratio assessment being higher or lower in one county than in another.

But such an hypothesis would imply that this assessment dated only from the days of Æthelred, or circ. 1000. Now the five-hide unit, on the contrary, was undoubtedly an old institution. Church lordships, the easiest to trace, appear to have retained their hidation unchanged from early times, and the 'possessio decem familiarum' of Bede seems to carry the decimal system back to very early days. Mr Seebohm, indeed—though, like others, he had failed to [pg 84] discover the existence of the five-hide system—saw in this 'possessio' of Bede a connecting link with the Roman decuria, just as he saw in the Roman jugatio the possible origin of English hidation. And we must, of course, trace its artificial arrangement (1) either to the Romans, (2) or to the Britons—assuming them to have had the same system as existed in Wales for the food-rents, (3) or to the English invaders.

Arrested at this point by the difficulty of assigning to the system I have described its real origin, I dropped these studies for some years in the hope that there might come from some quarter fresh light upon the problem. As I cannot, however, for lack of evidence, propound a solution capable of proof, I will content myself with indicating the line of research that offers, I venture to think, the most likelihood of success.

The proportionate sums contributed by the several counties to the Danegeld present a fruitful field of inquiry, but one, it would seem, as yet unworked. Mr Eyton, it is true, observed that 'in Devon and Cornwall the scope of the gheld-hide was enormous',¹⁷² that is, in other words, the assessment was strangely low, but it did not occur to him to seek the cause of the phenomenon he observed. If, as was the case, West Wales was assessed on quite a different scale to the counties adjoining it on the east, it may suggest a conclusion no less important than that, when the latter were originally assessed, West Wales was not yet a portion of the English realm. But, before concluding that the hide assessment is proved to be as ancient as this, we must see whether it is possible to detect any principle at work in the total assessments of the several counties, any relation between their area and the sums they contributed to the geld as entered in the Pipe Roll of 1130, our first evidence on the subject.

For such an enquiry it is especially needful to insist on breadth of treatment. In the first place, the modern area of the counties may vary more or less from the original extent;¹⁷³ in the second we have no proof that the assessment had always been the same, though the tendency in early days, no doubt, was to stereotype such figures. We must not, therefore attempt close or detailed investigation but if, on a review of the whole evidence, we detect certain broad features, uneffaced by the hand of time, we may fairly claim that we have in these the traces of a principle at work, the witness to a state of things prevailing in the distant past.

On comparing the contributions to a 'geld' at two shillings on the hide with the (modern) area of counties, we find that a rate of about a pound for every seven square miles prevailed widely enough to be almost described as normal.

[pg 85]

The three eastern counties work out thus:

In all three cases the proportion to the square mile is between a sixth and a seventh of a pound. In Cambridgeshire it is just under, in Sussex, just over, a seventh:

Most remarkable, however, is this Midland group:

It is remarkable, not only for this agreement inter se, but also for the sharp contrast it presents to the groups of counties, lying respectively to the south-east and the north-west of it. The former approximates a rate twice as high, namely, two-sevenths of a pound to the square mile:

Taking this group as a whole, it paid £1,032 18s 1d, a curiously close approximation to the £1,021⁴⁄7 which my suggested rate of ²⁄7 would give. Middlesex was so exceptional a county, that one hardly likes to include it, but there also the rate was a little over two-sevenths.

On the other hand, the counties to the north-west of what I have termed the Midland group are assessed at a rate singularly low. Nottingham and Derby, with a joint area of 1,855 miles, contributed only £108 8s 6d, representing one-seventeenth;¹⁷⁴ while Staffordshire, with its 1,169 miles, is found paying £44 0s 11d, [pg 86] a rate scarcely more than one twenty-seventh. Passing to the opposite corner of the realm, we have Kent, always a wealthy county, assessed at the phenomenally low rate of about one-fifteenth (£105 2s 10d, as against 1,555 miles), rather less than half that of Essex to its north, and Sussex to its west.

It would seem impossible to resist the conclusion that in these widely differing rates we have traces of a polity as yet divided, of those independent kingdoms from which had been formed the realm. Kent, for instance, which had so steadily maintained, first, its independent existence, and then its local institutions, had succeeded in preserving an assessment that its neighbours had cause to envy. In the west, Cornwall similarly enjoyed a low, indeed a nominal assessment while that of Devon, though higher than this, was so significantly lower than those of Somerset and Dorset¹⁷⁵ as to remind us that here, in part at least, the 'Welsh' long held their own. If the incidence of geld were shown by shading a map of England, on the plan so successfully adopted in Mr Seebohm's great work, it would show that the heavily assessed counties were those which formed the nucleus of the old West-Saxon realm.¹⁷⁶ All round this nucleus the map would shade off sharply, another sudden change marking the Danish counties on the north, the Jutish kingdom on the east, and the British district in the south-west. It is, perhaps, worthy of remark that Shropshire was assessed twice as heavily as the adjoining county of Stafford, possibly because part of it was added, at a very early period, to the kingdom of the West Saxons. If Mr Eyton was right in his reckoning that Kesteven was assessed twice as heavily as Lindsey, and Lindsey, in turn, twice as heavily as Holland, it would illustrate the survival of local distinctions even within the compass of a modern county, as well as the 'shading off' tendency of which I have already spoken.

The point I have here endeavoured to bring out is that if the system of artificial assessment were of Roman or British origin, we should expect to find it fairly uniform over the whole country, whereas we find, on the contrary, the very widest discrepancies. It might be urged, perhaps, that these were due to the differing conditions of particular counties, to their more or less partial reclamation, for instance, of the date when they were assessed. But this would not account for the grouping I have traced, and would imply that each county ought to differ indefinitely. Nor would it explain the case of Kent, where a county that must have been foremost in early development and prosperity enjoyed a phenomenally low assessment.

[pg 87]

Another objection that may be raised to my hypothesis is that the Hundred, as an area for police and rating, was a comparatively late institution, and that if the artificial system of assessment were as ancient as I suggest, it could not have operated, as we saw, in Cambridgeshire, it did operate, through the 'Hundred'. It is, however, admitted that the thing represented by the 'Hundred' was, whatever its original name, of immemorial antiquity, as the intermediate division between the Vill and the Shire or kingdom. Approaching the subject from the legal standpoint, Professor Maitland has pointed out that the Hundred having a proper court, which the Vill had not, was the older institution of the two, and has skilfully seized on the differentiation of villages originally possessing one name in common as a hint that some such subdivision may have been going on more widely than is known. It seems to me to be at least possible that the district originally representing a Hundred, and named, as we are learning, in most cases from the primitive meeting-place of its settlers, was reckoned as so many multiples of five or ten hides, and that this aggregate was subsequently distributed by its community among themselves.¹⁷⁷

If it be not presumption to touch on the controversies as to the Hundred,¹⁷⁸ I would suggest that while agreeing with Dr Stubbs, that the name of 'Hundred' may be traced to the ordinance of Edgar¹⁷⁹ —which did not, however, create the district itself—I cannot reconcile it with the view to which he leans in his Constitutional History, that 'under the name of geographical hundreds we have the variously sized pagi or districts in which the hundred warriors settled'; and that we should 'recognize in the name the vestige of the primitive settlement, and in the district itself an earlier or a later subdivision of the kingdom to which it belonged'.¹⁸⁰ For my part, I have never been able to understand the anxiety to identify the district known, in later days, as a 'Hundred' with an original hundred warriors, families, or hides. The significant remark on the 'centeni' by Tacitus, that 'quod primo numerus fuit, jam nomen et honor est', would surely lead us to expect that by the time of the migration the 'Hundred' had become, like the 'hide' of Domesday, a term even more at variance with fact. Indeed, in his masterly 'Introductory sketch', Dr Stubbs observed that the 'superior divisions' made by the 'new-comers' would 'have that indefiniteness [pg 88] which even in the days of Tacitus belonged to the Hundreds, the centeni of the Germans', and that their 'system' would be 'transported whole, at the point of development which it has reached at home'.¹⁸¹

The suggestion I have made as to the origin of the five-hide system is tentative only, and must remain so until we have at our disposal for the whole hidated region that complete and trustworthy analysis of assessment, on the need of which I again insist, at the risk of wearisome iteration.

XII. THE EAST ANGLIAN 'LEET'

In Norfolk and Suffolk we find Domesday recording assessed values not, as everywhere else, at the outset of an entry, but at its close; not in terms of hides and carucates, but in terms of shillings and pence. Instead of saying that a Manor paid on so many 'hidæ' or 'carucatæ terræ', Domesday, in the case of these counties, normally employs the phrase: 'x denarii de gelto'. Its meaning is that to every pound paid by the Hundred as geld the Manor contributed x pence.¹⁸² Thus, in the case of a Hundred assessed at a hundred hides, the formula for a five-hide Manor would be here 'xii. denarii de gelto', instead of the usual 'defendit se pro v. hidis', or some such phrase as that. There is an exact parallel to this method of recording assessed values in the case of fractions of knights' fees where portions of land are entered as paying so much 'when the scutage is forty shillings', instead of being assessed in terms of the knight's fee.¹⁸³ This system would seem, however, to have been understood imperfectly if at all. I may, therefore, point out that its nature is clear from the case of the Suffolk Hundred of Thingoe.

The case of this Hundred is singularly instructive. We find its twenty 'Vills' grouped in blocks, precisely as in the Cambridgeshire Hundreds, and these blocks are all equal units of assessment, like the ten-hide groups of the hidated districts. But in this case we can go further still, for we are not dependent on Domesday alone. The portion of a special Survey executed about a century later (circ. 1185) for Abbot Sampson of St Edmund's, which relates to its Hundred, is [pg 89] fortunately preserved, and gives us the name of the twelve 'leets' into which this Hundred was divided.¹⁸⁴

Here are the divisions recorded in it, with the Domesday assessment (in pence) of each Vill placed against its name.

				£	s	d
I.		Barrow	7
		Flemington	6
		Lackford	6
			—
			19	0	1	7
II.		Risby	20	0	1	8
III.		Saxham (A)	7
		Saxham (B)	7
		Westley	6½
			—
			20½	0	1	8½
IV.		Hengrave	10
		Fornham	10
			—
			20	0	1	8
V.		Ickworth	7½
		Chevington	6½
		Hargrave	7
			—
			21	0	1	9
VI.		Brockley	7
		Rede	7
		Manston	6
			—
			20	0	1	8
VII.		Whepstead	20	0	1	8
VIII.		Hawstead	13½
		Newton	6½
			—
			20	0	1	8
IX.		Horningsheath	20	0	1	8
X., XI., XII.		Sudbury	60	0	5	0
				————
				£1	0	0½

[pg 90]

The two records—Domesday and the Inquest—thus confirm one another, and their concurrent testimony establishes the fact not only that the Suffolk Hundred was divided into blocks of equal assessment, but that these blocks were known by the name of 'leets'.

Now Professor Maitland, in his Dissertation on the 'History of the Word Leet',¹⁸⁵ pronounces this 'the earliest occurrence of the word' that he has seen. But I can carry it back to Domesday itself. Though not entered in the Index Rerum, we find it in such instances as these:

'H[undredum] de Grenehou de xiv. letis' (ii. 119b).

'Hund[redum] et dim[idium] de Clakelosa de x. leitis' (ii. 212b).

I think it probable that in these cases the entry happened to stand first on the original return for the Hundred, and so—as in the I.E., where it is derived from the original returns—the general heading crept in. Though Professor Maitland has to leave the origin of the word unexplained, it seems to me impossible to overlook the analogy between the Danish lægd, described by Dr Skeat as a division of the country (in Denmark) for military conscription,¹⁸⁶ and the East Anglian leet, a division of the country (as we have seen) for purposes of taxation.

Sudbury, it will be observed, was a quarter of the Hundred of Thingoe,¹⁸⁷ just as Huntingdon was a quarter of a Hundred,¹⁸⁸ and Wisbech a quarter of a Hundred.¹⁸⁹

Having thus obtained from the Hundred of Thingoe the clue to this peculiar system, we can advance to more difficult types. The Hundred of Thedwastre, for instance, was divided not into twelve blocks, each paying twenty pence in the pound, but into nine blocks, each paying twenty-seven. This assessment allowed a margin of 3d for every pound (i.e. £1 0s 3d); but in the case of Thedwastre the total excess was only 1½d on the pound (i.e. £1 0s 1½d). I group the Vills tentatively, thus:

[pg 91]

					d
I.		Barton			27
II.		Fornham	6½		26½
		Rougham	20
III.		Peckenham	13½		26½
		Bradfield	5
		Fornham St Genevieve	8
IV.		Thurston	16		27
		Woolpit	11
V.		Rushbrook	7		27
		Ratlesden	20
VI.		Hessett	18		28
		Felsham	5
		Bradfield	5
VII.		Gedding	5		26
		Whelnetham	10
		Drinkston	11
VIII.		Ampton	7		27½
		Tostock	10½
		Staningfield	10
IX.		Tinworth	14		26
		Livermere	12
					——
					241½   (£1 0s 1½d)

The same unit of 27 (x9)—or, which comes to the same thing, 13½ (x18)—was adopted in Risbridge Hundred. In this case no less than five Manors are assessed at the same unit—13½d. So, again, in the Hundred of Blackbourn the units are 34½d and 17¼d, one Manor being assessed at the former, and five at the latter sum. Such is the key to the peculiar system of East Anglian assessment.

It is to be noted that 'twenty shillings'¹⁹⁰ represents ten hides at two shillings on the hide (the normal Danegeld rate), and thus suggests that in Norfolk, as in Cambridgeshire, the Hundreds were normally assessed in multiples of ten hides. The point, however, that I want to bring out is that the Hundred, not the Manor, nor even the Vill, is here treated as 'the fiscal unit for the collection of Danegeld'.¹⁹¹

XIII. THE WORDS 'SOLINUM' AND 'SOLANDA'¹⁹²

Several years ago I arrived at the conclusion that the identity of these two words was an unsupported conjecture. So long as it remained a conjecture only, its correction was not urgent; but since then, as is so often the case, the result of leaving it unassailed has been [pg 92] that arguments are based upon it. There appeared in the English Historical Review for July 1892 a paper by Mr Seebohm, in which that distinguished scholar took the identity for granted, as his no less distinguished opponent, Professor Vinogradoff, has done in his masterly work on Villainage in England.

I believe the alleged identity was first asserted by Archdeacon Hale, who wrote in his Domesday of St. Paul's (1858), p. xiv:

The word solanda, or, as it is written at p. 142, scolanda, is so evidently a Latinized form of the Anglo-Saxon sulung, or ploughland, and approaches so near to the Kentish solinus, that we need scarcely hesitate to consider them identical.

Let us start from the facts. In the Domesday of Kent we find the form solin, or its Latin equivalent solinum, used for the unit of assessment, like the hide and the carucate in other counties. In the Kent monastic surveys it is found as sullung or suolinga. But when we turn to the Domesday of St. Paul's, we find—first, that instead of being universal, as in Kent, it occurs only in three cases; secondly, that the form is solande, solanda, scholanda, scolanda, or even (we shall see) Scotlande; thirdly, that it is not employed as a unit of assessment at all.

The three places where the term occurs in the Domesday of St. Paul's are Drayton and Sutton in Middlesex, and Tillingham in Essex. Hale would seem to have arrived at no clear idea of what the word meant. At p. xiv he wrote that 'a solanda consisted of two hides, but probably in this case the hide was not of the ordinary dimension'. At p. lxxviii he inferred, from a reference to 'la Scoland' in a survey of Drayton, that '"ploughed land" would seem to be opposed to "Scoland"'. At p. cx he was led by the important passage—'De hydis hiis decem, due fuerunt in dominio, una in scolanda, et vii. assisæ'—to suggest that it 'appears to denote some difference in the tenure'. This last conjecture seems the most probable. If we take the case of Sutton and Chiswick, we read in the survey of 1222:

Juratores dicunt quod manerium istud defendit se versus regem pro tribus hidis preter solandam de Chesewich que per se habet duas hidas, et sunt geldabiles cum hidis de Sutton.

Hale (p. 119) believed that this Solande de Chesewich was no other than the Scotlande thesaurarii of 1181, namely the prebend of Chiswick. The above passage should further be compared with the survey of Caddington (1222):

Dicunt juratores quod manerium istud defendit se versus regem pro x. hidis ... preter duas prebendas quæ sunt in eadem parochia.

The formula is the same in both cases, and a solanda was clearly land held on some special terms, and was not a measure or unit of assessment at all. Indeed Hale himself admitted that it could not be identified with one or with two hides.

Fortunately I have discovered an occurrence of the word solanda [pg 93] which conclusively proves that it meant an estate, such as a prebend, and was not a unit of measurement. We have, in 1183, a 'grant by William de Belmes, canon of St. Paul's, to the chapter of that church, of the Church of St. Pancras, situate in his solanda near London' (i.e. his prebend of St. Pancras), etc.¹⁹³ This solves the mystery. The three solandæ at Tillingham were no other than the three prebends—Ealdland, Weldland, and Reculverland—which that parish actually contained.¹⁹⁴

Hale, however, misled Mr Seebohm, who in his great work on the English Village Community (p. 54), wrote of Tillingham:

There was further in this Manor a double hide, called a solanda, presumably of 240 acres. This double hide, called a solanda, is also mentioned in a Manor in Middlesex [Sutton], and in another in Surrey [Drayton]¹⁹⁵; and the term solanda is probably the same as the well-known 'Sollung' or 'solin' of Kent, meaning a 'ploughland'.

Proceeding further (p. 395), Mr Seebohm wrote:

Generally in Kent, and sometimes in Sussex, Berks and Essex, we found, in addition to, or instead of, the hide or carucate, or 'terra unius aratri', solins, sullungs, or swullungs, the land pertaining to a 'suhl', the Anglo-Saxon word for plough.

Unfortunately no reference is given for the cases of Sussex and Berks, and I know of none myself.

Turning now to the learned work of Professor Vinogradoff, we find him equally misled:

Of the sulung I have spoken already. It is a full ploughland, and 200 acres are commonly reckoned to belong to it. The name is sometimes found out of Kent, in Essex for instance. In Tillingham, a Manor of St. Paul's, of London, we come across six hides 'trium solandarum'. The most probable explanation seems to be that the hide or unit of assessment is contrasted with the solanda or sulland¹⁹⁶ (sulung), that is with the actual ploughland, and two hides are reckoned as a single solanda (p. 255).

Lastly, we come to Mr Seebohm's reply to Professor Vinogradoff (ante, pp. 444-465). Here the identity is again assumed:

Along with parts of Essex, the Kentish records differ in phraseology from those of the rest of England. Their sullungs of 240 acres occur also in the Manors of Essex belonging to St. Paul's, and the custom of gavelkind and succession of the youngest child mark it off as exceptional. Mr Vinogradoff ... shows that in the Kentish district, and in Essex, where the sullung or solanda takes the place of the hide, and where gavelkind prevailed, the unity of the hides and virgates was preserved only for the purposes of taxation and the services; whilst in reality the holdings clustered under the nominal unit were many and irregular.

[pg 94]

I yield to no one in admiration for Mr Seebohm's work, but the question raised is so important that accuracy as to the fact is here essential. (1) Sullung is nowhere found in Essex, but only solanda; (2) Solanda does not occur 'in the Manors' referred to, but at Tillingham alone; (3) In Essex it nowhere 'takes the places of the hide', as it does in Kent; (4) The Essex instance adduced by Professor Vinogradoff is taken from a Manor where solanda does not occur.

Two issues—quite distinct—are involved. In the first place, Mr Seebohm contends that Professor Vinogradoff must not argue from 'the custom of Kent' to the rest of England, because (inter alia) Kent, unlike the rest of England, was divided into sulungs, which points to some difference in its organization.¹⁹⁷ This contention is sound, and is actually strengthened if we reject the identity of sulung and solanda. But, in the second place, he endeavours to explain away the Essex case of subdivision at Eadwulfsness, to which the Professor appeals, by connecting it with the Kentish system through the term solanda. This, as I have shown above, is based on a misreading of the evidence, and is contrary to the facts of the case.

Let us then look more closely at the Essex instance of subdivision. It is taken from one Manor alone, the great 'soke' of Eadwulfsness, in the north-east corner of the county. This 'soke' comprised the townships of Thorpe 'le soken', Kirby 'le soken', and Walton 'le soken' (better known as Walton-on-the-Naze). Such names proclaim the Danish origin of the community, and it is noteworthy that the 'hidarii', on whom the argument turns, are found only at Thorpe and Kirby, the very two townships which bear Danish names. This circumstance points to quite another track. That the system in this little corner of Essex was wholly peculiar had been pointed out by Hale, and it might perhaps have originated in the superimposition of hides on a previous system, instead of in the breaking up of the hide and virgate system. But this is only a conjecture. The two facts on which I would lay stress are that at Thorpe, according to Hale, 'the holders of the nine hides (in 1279) possessed also among them seventy-two messuages', which, by its proportion of eight to the hide, favours Mr Seebohm's views; and that the holdings of the 'hidarii' were rigidly formed on the decimal system (such as 60, 30, 15, 7½ acres, or 40, 20, 10, 5 acres),¹⁹⁸ unlike the holdings of an odd number of acres on the Kentish Manors of St. Augustine's. The reason for the Essex system was clearly the necessity of keeping the holdings in a fixed relation to the hide, that their proportion of the hide's service might be easily determined. These two points have, perhaps, I think, been overlooked by both of the eminent scholars in their controversy.

[pg 95]

Before leaving the subject of the sulung, one should mention perhaps that it was divided (as Mr Seebohm has explained) into four quarters known as juga, just as the hide was divided into four virgates. Mr Seebohm bases this statement on Anglo-Saxon evidence,¹⁹⁹ but it is abundantly confirmed by Domesday, where we read of Eastwell (in Kent): 'pro uno solin se defendit. Tria juga sunt infra divisionem Hugonis, et quartum jugum est extra' (i. 13). So far all is clear; but Professor Vinogradoff, on the contrary, asserts that 'the yokes (juga) of Battle Abbey (in Kent) are not virgates, but carucates, full ploughlands' (p. 225). This assertion is based on a very natural misapprehension. In the Battle Manor of Wye (Kent) we find that the jugum itself was divided into four quarters, called 'virgates' which were each, consequently, the sixteenth, not, as in the hidated district, the fourth of a ploughland. Professor Vinogradoff, naturally assuming that the 'virgate' meant the same here as elsewhere, inferred that four 'virgates' (that is, a jugum) must constitute a full ploughland. But this change of denotation goes further still. The Battle Cartulary records yet another 'virgate', namely, the fourth (not of a ploughland, but) of an acre! This led me, on its publication, to wonder whether we have here the clue to the origin of the somewhat mysterious term 'virgate'. Starting from the acre, we should have in the virgata (rood) its quarter, with a name derived from the virga (rod) which formed its base in mensuration. The sense of 'quarter' once established, it might be transferred to the quarter of a jugum, or the quarter of a hide. This is a suggestion which, of course, I advance with all diffidence, but which would solve an otherwise insoluble problem. The relation of the bovate to the carucate, and of the jugum to the sulung, are both so obviously based upon the unit of the plough-team that they raise no difficulty. But the term 'virgate' does not, like them, speak for itself. If we might take it to denote merely a 'quarter' of the hide, it would become a term of relation only, leaving the 'hide' as the original unit. Should this suggestion meet with acceptance, it might obviously lead to rather important results.

Mr Elton, in his well-known Tenures of Kent, attaches considerable importance to a list, 'De Suylingis Comitatus Kanciæ et qui eas tenent', in the Cottonian MS., Claud. C. IV, which he placed little subsequent to Domesday. Having transcribed it for collation with the Survey, I came to the conclusion that it was not sufficiently trustworthy for publication, for the names, in my opinion, involve some anachronism. The feature of the list is that it shows us as tenants-in-chief, the leading tenants of Bishop Odo; and the change of most interest to genealogists is the succession of Patrick 'de Caurcio' to the holding of Ernulf de Hesdin.

[pg 96]

XIV. THE 'FIRMA UNIUS NOCTIS'

The curious and evidently archaic institution of the firma unius noctis was clearly connected with the problem of hidation. In Somerset the formula for a Manor contributing to this firma was:

Nunquam geldavit nec scitur quot hidæ sint ibi (i. 85).

In Dorset it ran:

Nescitur quot hidæ sint ibi quia non geldabat T.R.E. (i. 75).

In Wiltshire we read:

Nunquam geldavit nec hidata fuit, or nunquam geldavit: ideo nescitur quot hidæ sint ibi.²⁰⁰

In all these entries the 'hide' is recognized as merely a measure of assessment quite independent of area.

Hampshire affords us, in a group of Manors, a peculiarly good instance in point. Of Basingstoke, Kingsclere, and 'Esseborne', we read:

Rex tenet in dominio Basingestoches. Regale manerium fuit semper. Numquam geldum dedit, nec hida ibi distributa fuit....

Clere tenet rex in dominio. De firma Regis Edwardi fuit, et pertinet ad firmam diei de Basingestoches. Numerum hidarum nescierunt....

Esseborne tenet rex in dominio. De firma Regis Edwardi fuit. Numerum hidarum non habent....

Hæc tria maneria, Basingestoches, Clere, Esseborne, reddunt firmam unius diei (39).

Other Manors are found about the county displaying the same peculiarity.

Ipse rex tenet Bertune. De firmâ Regis E. fuit, et dimidiam diem firmæ reddidit in omnibus rebus.... Nunquam in hid(is) numeratum fuit.... Numerum hidarum non dixerunt.

Ipse rex tenet Edlinges in dominio. Hoc manerium reddidit dimidiam diem firmæ tempore Regis E. Numerum hidarum nesciunt (38).

Manors, such as Andover, not hidated, clearly belonged to the same system, though neither their value nor their render is given.

Thus, then, within the limits of Wessex, in the four adjacent counties of Dorset, Somerset, Wiltshire, and Hants, we find surviving, at the time of the Conquest, an archaic but uniform system of provision for the needs of the Crown by the assignment of certain [pg 97] estates or groups of estates, the render of which was expressed in terms of the 'firma noctis' or 'firma diei', and which, unlike the country around them, had never been assessed in 'hides'.

Mr Seebohm hints slightly at this firma system,²⁰¹ but only speaks of it as existing in Dorset. Nor does he allude to the significant fact of such Manors having never been hidated. It would lead us far afield to speculate on the origin of this system, or to trace its possible connection with the Welsh gwestva.²⁰² Nor can we here concern ourselves with the few scattered traces of it that we meet with elsewhere in Domesday. Its existence in four adjacent counties, with non-hidation as a common feature, is the point I wish to emphasize.

The system of grouping townships in the west for the payment of a food-rent (firma unius noctis) was exactly parallel to the grouping in the east for the payment, not of rent but of 'geld'. We can best trace this parallel in Somerset, because the firma unius noctis of the days before the Conquest had been there commuted for a money payment at the time of Domesday. Turning to the Cambridgeshire hundred of Long Stow, we find one of its 'blocks' (of twenty-five hides) divided into three equal parts, while another is divided into three parts, of which one is half the size of the two others. And so in Somerset we have Frome and Bedminster combined in one group for the payment of this firma, and the two Perrotts similarly combined with Curry. Frome and Bedminster are each assigned the same payment, but in the other group the contribution of one is half that of the two others.

Here are the Somerset groups of demesne, each charged with the render of a firma unius noctis.

	Commutation				£	s	d
Somerton (with Borough of Langport)	79	10	7		100	10	9½
Chedder (with borough of Axbridge)	21	0	2½
North Petherton	42	8	4		106	0	10
South Petherton	42	8	4
Curry Rivell	21	4	2
Williton					105	17	4½
Carhampton
Cannington
Frome	53	0	5		106	0	10
Bruton	53	0	5
Milborne Port (with Ilchester)					79	10	7
[Bedminster203					21	0	2½]

[pg 98]

Of these two last, Milborne Port is entered as having paid three-quarters of a firma noctis under the Confessor, while Bedminster—though in the midst of this group of firma Manors—is alone in having no render T.R.E. assigned to it. One is tempted to look on the two as originally combined in one firma (like Somerton and Chedder), save that the whole width of the county divides them, while in the other cases the constituents are grouped geographically.

The Wiltshire Manors, each of which rendered a firma unius noctis, were:

From the figures given for Somerset and Wilts, it may fairly be concluded that, in this district, the value of the 'firma' was about £105. In Somerset, however, there was clearly a special sum, £106 0s 10d, on which calculations were based.

An examination of Mr Eyton's statements on the firma unius noctis in Somerset and Dorset would prove a peculiarly conclusive test of his whole system.

In the case of Somerset one need not dwell on his giving its amount for the Williton group as £105 16s 6½d, when the sum named is £105 17s 4½d, although absolute accuracy is, in these matters, essential. We will pass at once to the bottom of the page (ii. 2), and collate his rendering of Domesday with the original:

Domesday gives the payment (in a characteristic phrase), as three-quarters [a half and a quarter] of a firma noctis. Mr Eyton first interpolates a 'unum', and then overlooks the 'quadrantem', with the result that he represents the due T.R.E. as a firma dimidiæ noctis (i. 77). So far, this is only a matter of error per se. But Domesday records the commutation of the due T.R.W. at £79 10s 7d. This proves to be three-quarters of the commutation, in two other cases, for a whole firma noctis (£106 0s 10d). Mr Eyton, however, imagining the due to have been only half a firma set himself to account for its commutation at so high a figure (i. 77-8). This he found no difficulty in doing. He explained that 'this was not a mere commutation', but 'was doubtless a change which took into consideration the extra means and enhanced value of Meleborne'.

[pg 99]

The probability is, then, that what we have called the enhanced ferm, was enhanced by something less than the gross profits we have instanced; that is, that a part of those profits, say the Burgage rents, or some of them, had contributed to the dimidia firma noctis before the commutation.

All these ready assumptions, we must remember, are introduced to account for a discrepancy which does not exist.

Great masses of Mr Eyton's work consist of similar guesses and assumptions. Now, if these were kept scrupulously apart from the facts, they would not much matter; but they are so inextricably confused with the real facts of Domesday that, virtually, one can never be sure if one is dealing with facts or fancies.

And far more startling than the case of Somerset is that of Dorset, the 'Key to Domesday'. Mr Eyton here held that Dorchester, Bridport, and Wareham paid a full firma unius noctis each, the total amount being reckoned by him at the astounding figure of £312 (p. 70)! Exeter, which affords a good comparison, paid only £18 (as render), though the king had 285 houses there: the three Dorset towns in which, says Mr Eyton, the Crown had 323 houses, paid in all, according to him, £312. The mere comparison of these figures is sufficient. But further, Mr Eyton observes (p. 93), that in 1156 'Fordington, Dorchester, and Bridport' were granted by Henry II to his uncle, 'as representing Royal Demesne to the annual value of £60'. This is an instructive commentary on his view that Dorchester and Bridport alone rendered £208 per annum. Our doubts being thus aroused, we turn to Domesday and find that it does not speak of any of these towns as paying that preposterous firma. The right formula for that would be 'reddit firmam unius noctis' (p. 84). Instead of that, we only have 'exceptis consuetudinibus quæ pertinent ad firmam unius noctis' (p. 70). The explanation is quite simple. Just as in Somerset, Mr Eyton admits, Langport and Ilchester, although boroughs, were 'interned' in groups of Royal demesne, paying the firma unius noctis, so in Dorset the boroughs were 'interned' in groups of Royal demesne. Indeed one of these groups was headed by Dorchester, and is styled by Mr Eyton the 'Dorchester group'. But he boldly assumed that 'Dorchester' must have two different meanings:

[A] We assume about 100 acres to have belonged to the Domesday Burgh, and perhaps 882 acres to represent land, subinfeuded at Domesday, and annexed to Dorchester Hundred. [B] It follows that we assume about 429 acres, [to be that] ... which here figures [fo. 75] under the name Dorchester.

It is not too much to say that any one, who refers to pp. 70-3, 78-101 of the Key to Domesday, will find that the singular misconception as to the Dorset Boroughs makes havoc of the whole calculation. But here again the point to be insisted on is not the [pg 100] mere mistake per se, but the elaborate assumptions based upon it and permeating the whole work.²⁰⁴

Apart from the Manors grouped for a firma unius noctis, if we take the comital Manors (mansiones de comitatu) of Somerset, which were in the King's hands in 1086, we find their rentals given on quite a different principle to those of the Manors in private hands.

(1) They are entered as renders ('reddit'), not as values ('valet').

(2) The sums rendered are 'de albo argento'.

(3) In at least ten out of the fifteen cases, they are multiples of the strange unit £1 3s.

As this fact seems to have escaped Mr Eyton's notice, I append a list of these Manors, showing the multiples of this unit that their renders represent:

Whatever this strange unit represented, it formed the basis in these Manors of a reckoning wholly independent of the 'hides' or ploughlands of the Manor, and as clearly artificial as the system of hidation I have made it my business to explain.

XV. 'WARA'

The meaning of 'Wara' is made indisputable by the I.C.C. When land was an appurtenance, quoad ownership, of a Manor in one township, but was assessed in another in which it actually lay, the land was said to be in the former, but its 'wara' in the latter. As this 'wara' was an integral part of the total assessment of the township, it had to be recorded, under its township, in the I.C.C. Here are the three examples in point:

[Histon.] De his xx. hidis jacet Warra de una hida et dimidia in hestitone de manerio cestreford. Hanc terram tenuit comes alanus [sic] et est appretiata in essexia (p. 40).

[pg 101]

[Shelford.] De his xx. hidis tenet petrus valonensis iii. hidas de firma regis in neueport.... Hæc terra est berewica in neueport, sed Wara jacet in grantebrigge syra (p. 49).

[Trumpington.] De his vii. hidis [tenet] unus burgensis de grenteburga i. virgam. Et Warra jacet in trompintona, et terra in grantebrigga (p. 51).

To these I may add a fourth instance, although in this case the name wara does not occur:

[Bathburgam.] De his vii. hidis tenet Picotus in manu regis dimidiam hidam et dimidiam virgam. Hæc terra jacet in cestreforda et ibi est appretiata xxx. sol. in essexia (p. 36).

The lands at Histon and 'Bathburgam' were mere outlying portions of the royal Manor of Chesterford in Essex, and those at Shelford were a 'berewick' of the royal Manor of Newport, also in Essex. But they were all assessed in Cambridgeshire, where they actually lay.

So also we read under Berkshire (61b): 'Hæc terra jacet et appreciata est in Gratentun quod est in Oxenefordscire, et tamen dat scotum in Berchesire'. Again (203b) we read under Pertenhall: 'Hec terra sita est in Bedefordsire, set geldum et servitium reddit in Hontedunscyre'. A good instance of the same arrangement in another part of England is found in those Worcestershire Manors which were annexed as estates to Hereford, but which were assessed in those Worcestershire Hundreds where they actually lay (see p. 60).

A similar expression is applied to the possession of 'soca'. Thus under Shelford we read:

De hac terra adhuc tenuerunt iii. sochemanni dimidiam hidam sub gurdo comite. Non potuerunt recedere sine licentia comitis gurdi. Et soca jacebat in Witlesforda (p. 48).

Here the land was in Shelford, but the jurisdiction (soca) was attached to Earl Gyrth's Manor of Whittlesford.

Prof Vinogradoff has dealt with 'the word wara' in his Villainage in England (i. 241-4), and asserts that the 'origin and use of the term is of considerable importance'. But he does not allude to the above evidence, and I cannot follow him in his argument. While rightly disregarding Mr Pell's fanciful derivation from 'warectum', he asserts that:

We often find the expression 'ad inwaram' in Domesday, and it corresponds to the plain 'ad gildam [sic] regis'. If a Manor is said to contain seven hides ad inwaram, it is meant that it pays to the king for seven hides.... The Burton cartulary, the earliest survey after Domesday, employed the word 'wara' in the same sense.

One cannot disprove the first proposition without reading through all Domesday for this purpose. I can only say that I do not remember [pg 102] ever meeting in Domesday Book with such an expression. The solitary instance of its use known to me is in the Liber Niger of Peterborough (p. 159), where we read: 'in Estona sunt iii. hidæ ad in Waram'; and there the relevant entry in Domesday has no such expression. Of the statement as to the Burton cartulary, one can positively say it is an error. Its 'waræ' have quite another meaning and are spoken of as virgates would elsewhere be.

Collation with what I have termed the Northamptonshire geld-roll renders it clear that 'wara', in Domesday, represents the old English word for 'defence', in the sense of assessment, the 'defendit se' formula of the great Survey leading even to the phrase of 'Defensio x. acrarum', for assessment to Danegeld, which is found in the first volume of Fines published by the Pipe-Roll Society.

XVI. THE DOMESDAY 'JURATORES'

I now approach the subject of the Domesday juratores.

The lists of these in the I.E. and in the I.C.C. afford priceless information. The latter gives us the names for all but three of the Cambridgeshire Hundreds, the former for all Cambridgeshire (one Hundred excepted) and for three Hertfordshire Hundreds as well. The opening paragraph of the I.E. tells us 'quomodo barones regis inquisierunt, videlicet per sacramentum vicecomitis scire et omnium baronum et eorum francigenarum et tocius centuriatus presbyteri prepositi vi. villani [sic] uniuscuiusque ville'.²⁰⁶ Careful reading of this phrase will show that the 'barones regis' must have been the Domesday Commissioners. The difficulty is caused by the statement as to the oaths of the sheriff, the tenants-in-chief (barones), and their foreign (? military) under-tenants (francigenæ). The lists of juratores contain the names of many francigenæ in their respective hundreds, but, so far as I can find, of no tenants-in-chief. The sheriff, of course, stands apart. His name indeed in the I.C.C. is appended to the list of jurors for the first Hundred on the list, but is not found in the I.E. Moreover, it should be noted that the above formula speaks of all the tenants-in-chief, but only of a single Hundred court. Two hypotheses suggest themselves. The one, that the sheriff and barones of the county made a circuit of the Hundreds, and then handed in, on their oaths, to the commissioners a return for the whole county; the other, that the circuit was made by the commissioners themselves, attended by the sheriff and barones. In the former case it is obvious that the commissioners would fail to obtain at first hand that direct local information which it was their object to elicit: and further, when we find the sheriff and barones [pg 103] charged with wrongdoing in these very returns, it is, to say the least, improbable that they were their own accusers, especially in the case of such a sheriff as Picot, at once dreaded and unscrupulous.

It seems, therefore, the best conclusion that the Domesday commissioners themselves attended every Hundred court, and heard the evidence, sometimes conflicting, of 'French' and 'English'.²⁰⁷

The order in which the Hundreds occur must not be passed over, because their sequence distinctly suggests a regular circuit of the country. Here is the sequence given in our three authorities: the I.C.C., the I.E., and the list of jurors prefixed to the latter:

On comparing the first two of these lists it will be found that (except in the case of three contiguous Hundreds, which does not affect the argument) the Hundreds are taken in a certain sequence, which is seen, on reference to the valuable map prefixed to Mr Hamilton's book, to represent a circuit of the southern portion of the county from north-east to north-west, followed by an inquest on the district to its north, the 'two Hundreds' of Ely.

The third list, on the other hand, misplaces the Hundreds of Triplow and Erningford altogether, and wholly omits that of Childeford. The transposition and omission are both notable evidence that the B and C texts, as I shall urge, were derived from some common original which contained these defects.

The essential point, however, is that a circuit was made of the county whether merely by the sheriff, or, as seems most probable, [pg 104] by the Domesday Commissioners themselves—the 'barones regis' of the record—who must have attended the several Hundred-courts in succession.

But when we speak of the Hundred-court it is necessary to explain at once that the body which gave evidence for the Domesday Inquest was of a special and most interesting character. It combined the old centuriatus—deputations of the priest, reeve, and six villeins from each township (villa)—with the new settlers in the Hundred, the francigenæ. A careful investigation of the lists will prove that half the juratores were selected from the former and half from the latter. This fact, which would seem to have been hitherto overlooked, throws a flood of light on the compilation of the Survey, and admirably illustrates the King's policy of combining the old with the new, and fusing his subjects, their rights and institutions, into one harmonious whole. Conquerors and conquered were alike bound by their common sworn verdicts.²⁰⁹

We have the lists, in all, for eighteen Hundreds, fifteen in Cambridgeshire and three in Herts, of which two were 'double'. There were, practically, for each Hundred exactly eight juratores, half of them 'French' and half 'English'. But the two 'double' Hundreds had sixteen each, half of them 'French' and half 'English'. Although it is recorded that 'alii omnes franci et angli de hoc hundredo juraverunt', it is obvious that the eight men always specially mentioned were, in a special degree, responsible for the verdict. Their position is illustrated, I think, by the record of a Cambridgeshire placitum found in the Rochester chronicles. This is the famous suit of Bishop Gundulf against Picot the sheriff in the County Court of Cambridgeshire,²¹⁰ which affords a valuable instance of a jury being elected to confirm by their oaths the (unsworn) verdict of the whole court:

Cum illis (i.e. omnes illius comitatus homines) Baiocensis episcopus, qui placito præerat, non bene crederet; præcepit ut, si verum esse quod dicebant scirent, ex seipsis duodecim eligerent, qui quod omnes dixerant jure jurando confirmarent.

Now we read of this jury:

Hi autem fuerunt Edwardus de Cipenham, Heruldus et Leofwine saca de Exninge, Eadric de Giselham, Wlfwine de Landwade, Ordmer de Berlincham, et alii sex de melioribus comitatus.

[pg 105]

Investigation shows that the names mentioned are local. The land in dispute was a holding in Isleham in the Hundred of Staplehoe. One juror, Eadric, came from Isleham itself, two from Exning, one from Chippenham, one from Landwade, while the sixth, Ordmer, was an under-tenant of Count Alan, in the Manor from which he took his name (Badlingham), and was a Domesday juror for the Hundred. These six, then, were clearly natives chosen for their local knowledge. The other six, chosen 'de melioribus comitatus', were probably, as at the Domesday inquest, Normans (Franci). Thus the double character of the jury would be here too preserved, and the principle of testimony from personal knowledge upheld.

So again in the Dorset suit of St. Stephen's, Caen (1122),²¹¹ the men of seven Hundreds are convened, but the suit is to be decided 'in affirmatione virorum de quatuor partibus vicinitatis illius villæ'.²¹² Accordingly, 'sexdecim homines, tres videlicet de Brideport, et tres de Bridetona, et decem de vicinis, juraverunt se veram affirmationem facturos de inquisitione terræ illius'. The names of the jurors are carefully given: 'Nomina vero illorum qui juraverunt, hæc sunt'. Again in the same Abbey's suit for lands in London, 'per commune consilium de Hustingo, secundum præceptum regis, elegerunt quatuordecim viros de civibus civitatis Londoniæ qui juraverunt'. And in this case also we read: 'Hæc sunt nomina illorum qui juraverunt.... Et hæc sunt nomina eorum in quorum præsentia juraverunt.'²¹³

This corresponds, it will be seen, exactly with the writ to which the Inquisitio Eliensis was, I hold, the return: 'Inquire ... qui eas (terras) juraverunt et qui jurationem audierunt' (infra, p. 114).

Enough has now been said to show that the names of the Domesday jurors recorded for each Hundred represent a jury of eight, elected to swear on behalf of the whole Hundred, and composed of four foreigners and four Englishmen, in accordance with the principle that the conflicting interests ought to be equally represented.²¹⁴

We may take, as a typical set of juratores, those for the Hundred of Erningford, the survey of which, in Mr Hamilton's book, occupies pp. 51-68. I give them in their order:

[pg 106]

All four francigenæ can be identified in the Hundred. Walter held a hide and a quarter in 'Hatelai' from the wife of Ralf Tailbois; Humfrey, a hide and a quarter in 'Hatelai', from Eudo dapifer;²¹⁵ Hugh, a hide and a half in 'Melrede', from Hardwin de Scalers; and Richard, three virgates in 'Mordune', from Geoffrey de Mandeville. Of the Angli, Colsuenus was clearly Count Alan's under-tenant at three townships within the Hundred, holding in all two hides; 'Ailmarus', his son, was, just possibly, the 'Almarus de Bronna', who was a tenant of Count Alan in two adjacent townships, holding two hides and three-eighths; 'Turolfus' and 'Alfuuinus' cannot be identified, and were probably lower in the social scale.

It will be observed that Colsweyn belongs to a special class, the English under-tenants. He is thus distinct at once from the Francigenæ, and from the villeins of the township. He and his peers, however, are classed with the latter as jurors, because they are both of English nationality. In the great majority of cases the English juratores cannot be identified as under-tenants, and may therefore be presumed to have belonged to the township deputations.

XVII. THE 'INQUISITIO ELIENSIS'

The record known by this name has long been familiar to Domesday students, but no one, so far as I know, has ever approached the questions: Why was it compiled? When was it compiled? From what sources was it compiled? These three questions I shall now endeavour to answer.

First printed by the Record Commission in their 'Additamenta' volume of Domesday (1816), its editor, Sir Henry Ellis, selected for his text the most familiar, but, as I shall show, the worst of its three transcripts (Cott. MS., Tib. A. VI), though he knew of what I believe to be the best, the Trin. Coll. MS., O. 2, 1, which seems to be the one styled by him 68 B 2.²¹⁶ In his introduction he thus described it:

[pg 107]

The Inquisitio Eliensis is a document of the same kind with the Exeter Domesday; relating to the property of the Monastery of Ely recorded afterwards in the two volumes of the Domesday Survey (p. xiv).

From this it would seem that Ellis believed the Inquisitio, at any rate, to be previous to Domesday Book, but he practically left its origin altogether in doubt.

Sixty years later (1876) the Inquisitio was published anew, but without any further solution of the points in question being offered.²¹⁷ For this edition three MSS. were collated, with praiseworthy and infinite pains, by Mr N. E. S. A. Hamilton. Taking for his text, like Ellis, the Cottonian MS. Tib. A. VI, which he distinguished as A, he gave in footnotes the variants found in the MSS. at Trinity College, Cambridge, viz.: O. 2, 41 (which he termed B), and O. 2, 1 (which he distinguished as C). In Mr Hamilton's opinion (p. xiv) the 'C' text 'appears to have been derived from the "B" MS. rather than the Cottonian' ('A'). From this opinion, it will be seen, I differ wholly.

A careful analysis of the three texts has satisfied me beyond question that while C is the most accurate in detail, it is marred by a peculiar tendency to omission on the part of its scribe. This, indeed, is its distinctive feature. Now B cannot be derived from C, because it supplies the latter's omissions. On the other hand, C cannot be derived from B, because it corrects, throughout, B's inaccuracies. Consequently they are independent. More difficult to determine is the genesis of A, the worst of the three texts; but as it virtually reproduces all the inaccuracies found in B (besides containing many fresh ones), without correcting any, it can only be inferred that B was its source. Thus we have on the one hand C, and, on the other B (with its offspring A), derived independently from some common source. And this conclusion agrees well with the fact that a long catalogue of lands abstracted from the House of Ely is found in C, but not in A or B,²¹⁸ and with the circumstance that the famous rubric ('Hic subscribitur inquisitio'), which heads the inquisition in A and B, is placed by C at the end of the lists of jurors.²¹⁹

Starting from this conclusion, let us now proceed to ask, what was the document from which B and C copied independently? Clearly, it was not Domesday Book, for outside the eastern counties they record the returns in full, like the Inq. Com. Cant. itself. Were they then taken from the original returns, or at least from the copy of those returns in the Inq. Com. Cant.? This point can only be determined by close analysis of the variants; if we find B and C containing occasionally the same errors and peculiarities, although copied [pg 108] independently, it follows that the document from which they both copied must have contained those same errors and peculiarities. Let us take the case of Papworth. The right reading, as given both in Domesday and the Inq. Com. Cant., I have placed on the left, and the wrong reading, in B and C, on the right:

Here are some further illustrations of errors in the I.E.:

Again, the clause 'Tost²²² pro viii. hidis et xl. acris', which ought to head the Hardwick entries, is wrongly appended in the I.E. (p. 110) to a Kingston entry with which it had nothing to do. So too, 'hoc manerium pro x. hidis se defendit [sic] T.R.E. et modo pro viii. hidis', which belongs to Whaddon, is erroneously thrown back by the I.E. (p. 107), into Trumpington, a Manor in another Hundred. It is singular also that all the MSS. of the I.E. read 'iii. cotarii' (p. 101), where D.B. and the I.C.C. have 'iii. bordarii' (p. 3), and 'x. cotarii' (p. 101), where they have 'x. bordarii' (p. 6): conversely, the former, in one place, read 'xv. bordarii' (p. 107), where the latter have 'xv. cotarii' (p. 63).

In comparing the text of the I.E. with that of the I.C.C., we shall find most striking and instructive variants in the lists of juratores for the several Hundreds. Take, for instance, the lists for the Hundreds of Cheveley and Staines, which follow one another in both MSS.

[pg 109]

The second name on these lists can be conclusively tested. For the relative entry in the I.C.C. is 'Esselei tenet euerard[us]²²⁹ filius brientii de Alberico'. This proves that the I.C.C. is right in reading 'Euerard[us]', while the I.E. is right in adding 'homo Alb[er]ici de uer'.

These are the lists for Staines Hundred.

In these two lists the points to strike us are that Harold is placed first on one list and seventh on another; Aleran third on one list and first on another; and 'Fareman' distinguished more clearly in the I.E. than in the I.C.C. as a separate individual.

[pg 110]

If we now collect from the other Hundreds some instances of instructive variants, we shall obtain important evidence.

It is impossible to examine the italicized variations in these parallel texts without coming to the conclusion that they must have been independently derived from some common original, an original containing more detail than either of them. On the other hand, the comparatively close agreement between the texts of the actual returns in the I.C.C. and the I.E. leads one to infer that these were copied with far more exactitude than the comparatively unimportant surnames of the jurors. For us the value of these variations in the jurors' lists lies in the evidence afforded to the origin of the existing MSS.

The object of this careful scrutiny has been to prove that as certain errors and peculiarities are found in two independent MSS., they must have existed in the original document from which both were [pg 111] copied, and which was neither the I.C.C. transcripts nor the original Domesday returns. What then was this document? It was, and can only have been, the true Inquisitio Eliensis, the date and origin of which I shall discuss below. Further, I should imagine this document to have probably been a roll or rolls, which—on its contents being subsequently transcribed into a book for convenience—was allowed, precisely as happened to the Domesday rolls themselves, to disappear. In perfect accordance with this view we find the whole contents of the Inquisitio arranged for a special purpose, and no mere transcript of the Domesday returns. Thus, after abstracting all the entries relating to the Cambridgeshire estates, and subjoining a list of houses held in Cambridge itself, it proceeds to add up all the items independently, and record their total values to the Abbey. This analysis is carried out for several counties (pp. 121-4), and is, of course, peculiar to the Inquisitio, although inserted between the abstracts of the Domesday returns for Cambridgeshire and Herts. So too the breviate or short abstract of the estates (pp. 168-173), which was part of the original document—for it is found in all the derived MSS.—must have been specially compiled for it, and so also was the Nomina Villarum (pp. 174-83).

Another peculiarity of the Inquisitio is the care with which it records the names of sokemen on the Abbey estates when omitted in the I.C.C. and D.B. This may lead us to ask whether its compilers supplied these names from their personal knowledge. We might think not, for in some cases they are recorded by the D.B. and the I.C.C., while in one (p. 106) the I.E. actually omits the name, reading only 'quidam sochemanus', where the other two documents (p. 46) supply his name ('Fridebertus'). From this we might infer that the names were probably recorded in the original returns, but deemed of too slight importance to be always copied by the transcriber. Yet the balance of evidence leads me to believe that the I.E. did supply names from independent knowledge. With the values, however, the case is clearer. The I.E. contains special and exclusive information on the value of socman-holdings, and must, I think, have derived it from some other source than the original Domesday returns. Here are some instances in point.