THE KNIGHTS OF PETERBOROUGH

(Temp. Henry I)

The interesting 'Descriptio militum de Abbatia de Burgo' is found in the same MS. as the Northamptonshire Geld-roll.[¹]

It was printed by Stapleton in the appendix to his Chronicon Petroburgense (pp. 168-75),[²] but no attempt was made to date it. The name of Eudo Dapifer proves that it cannot have been compiled later than 1120. On the other hand, it cannot well be earlier than 1100, for some of the Domesday tenants had been succeeded by their sons—Robert (?) Marmion, for instance, by Roger, and Coleswegen by Picot—while the mention of 'Gislebertus filius Ricardi', possibly the son of Richard of 'Wodeford' (i. 224b), points in the same direction. As the majority of names, however, seem to be those of Domesday tenants, it is probable that the list is not later than the Lindsey survey itself, if, indeed, it is not earlier. The first entry it contains is a good specimen of its value:

Asketillus de Sancto Medardo tenet de abbatia de Burch in Hamtonascira x. hidas et iii. partes i. virgæ, et in Lincolnescira iii. carrucatas et inde servit se vi. milite. Et de feudo hujus militis dedit rex Willelmus senior Eudoni Dapifero in Estona hidam et dimidiam et mandavit de Normannia in Angliam Episcopo Constantiarum et R. de Oilli per breves suos ut inde darent ei excambium ad valens in quocumque vellet de iii. vicinis comitatibus; sed abbas noluit.

We duly find 'Anschitillus' in Domesday, holding 'Witheringham', Northants and 'Osgodeby', Linc., of the Abbot (i. 221b, 345b). In the same way we are enabled to identify the 'Rogerius Infans' of our list with 'Rogerius' who held 'Pilchetone', according to Domesday (i. 221b), of the Abbot, 'Ascelinus de Waltervilla' with the 'Azelinus' of Domesday (Ibid.), 'Gosfridus nepos Abbatis', with 'Goisfridus' who held in 'Sudtorp' (Ibid.), and 'Rogerius Malfed' with that 'Rogerius' who held of the Abbot at Woodford (i. 222). 'Rogerus', on the other hand, who held in Domesday two hides at Milton, Northants (i. 221b), and seven bovates at Cleatham, Linc. (i. 346), is represented in our list by the entry:

Turoldus de Meletona ii. hidas in Hamtonascira, et in Lindeseia vi. bovatas, et inde servit se altero milite (p. 171).

The chief lesson taught us here is the rashness of assuming the identity of tenants happening to bear the same name. For even among the few who are named as holding of the Abbot of Peterborough, we have found three Rogers quite distinct from one another.

The entries which follow are of value as absolute proofs of succession:

Domesday	Descriptio Militum
In Dailintone tenet Ricardus de abbate iiii^or. hidas (i. 222). In Risun habuit Elnod iiii. bovatas terre ad geldum ... Nunc habet Colsuan de abbate Turoldo (i. 345b).	Rodbertus filius Ricardi iiii. hidas in Hamtonascira, et inde servit se altero milite (p. 175). Picotus filius Colsuaini habet dimidiam carrucatam in Rison, quam abbas dedit patri suo tali servicio quod esset ad placita abbatis et manuteneret res suas et homines suos in scira et in aliis locis (p. 175).

This second entry not only records a peculiarly interesting enfeoffment, but identifies 'Colsuan', the Abbot's under-tenant at Riseholme, with no less a person than the conqueror's 'English favourite Coleswegen, ... an Englishman who, by whatever means, contrived to hold up his head among the conquerors of England'.[³]

As sons, in such cases as these, have succeeded their fathers, it need not surprise us that our list comprises some names that are found in the Liber Niger survey of 1125.[⁴] Vivian, whom, it tells us, Abbot Turold had enfeoffed at Oundle (p. 175) occurs there in that survey (p. 158), as does Robert d'Oilli at Cottingham (pp. 159-73).[⁵] Vitalis ('Viel') Engaine had succeeded William (Engaine) at Pytchley both in our list and in the survey of 1125 (cf. ante, p. 129).

One of the most interesting and important points in this list of knights is the gleam of new light it throws on Hereward 'the Wake'. In it we read:

Hugo de Euremou iii. hidas in dominio et vii. bovatas in Lincolneshira, et servit pro ii. militibus.

Ansford iii. carucatas et servit pro dimidia hida [sic].

Now Hugh de Euremou is the name of the man who, according to the pseudo-Ingulf, married Hereward's daughter. Here we have proof of his real existence, and are enabled moreover to detect him, I claim, in that Hugh who, as a 'miles' of the Abbot, held three hides at 'Edintone' [Etton, Northants] in Domesday (i. 222). Mr Freeman speaking of the vacancy at Bayeux in 1908, wrote:

William at once bestowed the staff on Turold, the brother of Hugh of Evermont [sic], seemingly the same Hugh who figures in the legend of Hereward as his son-in-law and successor.[⁶]

But the French editors of Ordericus, in a note to the passage from which this statement was taken (iv. 18), speak of our man as 'Hugue d'Envermeu, donateur du prieuré de St. Laurent d'Envermeu à l'Abbaye de Bec'.[⁷]

Turning for a moment from Hugh to Ansford, we read in the Lincolnshire 'Clamores':

Terram Asford in Bercham hund' dicit Wapentac non habuisse Herewardum die quo aufugiit (D.B., i. 376b).

About this entry, as Mr Freeman observed, 'there can be no doubt'. But as the result of his careful inquiry,[⁸] he limited 'our positive knowledge', from Domesday, to this entry and to two in the text of the Lincolnshire survey (364b-377). It is strange that he did not follow up the clue the 'Clamores' gave him. The relevant entry in the text of the Survey is duly found under the Peterborough fief:

In Witham et Mannetorp et Toftlund habuit Hereward xii. bovatas terræ ad geldum.... Ibi Asuert [sic] homo abbatis Turoldi habet, etc....

Berew[ita] hujus M. in Bercaham et Estou i. carucata terræ ad geldum. ... Ibi Asford habet, etc....

In Estov Soca in Witham iiii. bovatæ terræ et dimidia ad geldum.... Ibi Asfort de abbate habet, etc.... (i. 346).

This is the 'terra Asford' referred to in the 'Clamores', and, as amounting to 3¹⁄16 carucates, it is clearly the 'iii. carucatas' assigned in our list to 'Ansford'. Thus, through his successor Ansford, we have at last run down our man; Hereward was, exactly as is stated by Hugh 'Candidus', a 'man' of the Abbot of Peterborough; and his holding was situated at Witham on the Hill,[⁹] not far from Bourne, and, at Barholme-with-Stow a few miles off, all in the extreme south-west of the county. This is the fact for which Mr Freeman sought in vain, and which has eluded Professor Tout, in his careful life of the outlaw for the Dictionary of National Biography.

We are now in a position to examine the gloss of Hugh 'Candidus', showing how 'Baldwin Wake' possessed the holdings both of Hugh and of Ansford:[¹⁰]

Primus Hugo de Euremu. Baldwinus Wake tenet in Depinge, Plumtre, et Stove feoda duorum militum.... Et præterea dictus Baldewinus tenet feodum unius militis in Wytham et Bergham de terra Affordi. Et prædictus Baldewinus de predictis feodis abbati de Burgo debet plenarie respondere de omni forensi [servitio].

Here we see how the legendary name and legendary position of Hereward were evolved. The Wakes, Lords of Bourne, held among their lands some, not far from Bourne, which had once been held by Hereward. Thus arose the story that Hereward had been Lord of Bourne; and it was but a step further to connect him directly with the Wakes, by giving him a daughter and heir married to Hugh de Evermou, whose lands had similarly passed to the Lords of Bourne. The pedigree-maker's crowning stroke was to make Hereward himself a Wake,[¹¹] just as Baldwin fitz Gilbert (de Clare) is in one place transformed into a Wake.[¹²] The climax was reached when the modern Wakes revived the name of Hereward, just as 'Sir Brian Newcome of Newcome' set the seal to his family legend by giving his children 'names out of the Saxon calendar'.

Returning to Hereward himself, we find Mr Freeman writing (of the spring of 1070):

At this moment we hear for the first time of one whose mythical fame outshines all the names of his generation, and of whom the few historical notices make us wish that details could be filled in from some other source than legend.... Both the voice of legend and the witness of the great Survey agree in connecting Hereward with Lincolnshire, but they differ as to the particular spot in the shire in which he is to be quartered. Legend also has forgotten a fact which the document has preserved, namely, that the hero of the fenland did not belong wholly to Lincolnshire, but that he was also a landholder in the distant shire of Warwick. But the Survey has preserved another fact with which the legendary versions of his life have been specially busy. Hereward, at some time it would seem, before the period of his exploits, had fled from his country.[¹³]

Let us first dismiss from our minds the alleged fact as to Warwickshire. There is absolutely nothing to connect the Count of Meulan's tenant there with the Lincolnshire hero; indeed Mr Freeman admits in his appendix 'that the Hereward of these entries may be some other person' (p. 805). Legend had an excellent reason for ignoring this alleged 'fact' as had 'romances' for having 'perversely forgotten' to mention the deeds or the fate of William Malet in the Isle (Ibid., p. 473). We must also dismiss the 'fact'—'undoubted history' though it be (Ibid., p. 805)—of Hereward's 'banishment' at some time between 1062 and 1070. For the Survey gives no date; it merely speaks of 'die quâ aufugiit' (i. 376b), which phrase, in the absence of evidence to the contrary, must be referred to his escape from the 'Isle',[¹⁴] when (1071) in the words of Florence, 'cum paucis evasit'. This at once explains the Domesday entry (ante, p. 160), for he would, of course, have forfeited his holding before that date.

'But leaving fables and guesses aside,' in Mr Freeman's words, 'we know enough of Hereward to make us earnestly long to know more' (p. 456). My proof that the English hero was a 'man' of the Abbot of Peterborough explains why 'Hereward and his gang', as they are termed in the Peterborough Chronicle, 'seem', Mr Freeman is forced to admit, 'to be specially the rebellious tenants of the Abbey', as distinct from the Danes and the outlaws (p. 459). And the vindication, on this point, of Hugh Candidus' accuracy makes one regret that Mr Freeman, though eager for information as to Hereward, ignored so completely that writer's narrative. It is in absolute agreement with the Peterborough Chronicle, Mr Freeman's own authority, but records some interesting details which the Chronicle omits.[¹⁵] These place Hereward's conduct in a somewhat different light, and suggest that he may really have been loyal to the Abbey whose 'man' he was. His plea for bringing the Danes to Peterborough was that he honestly believed that they would overthrow the Normans, and that the treasures of the church would, therefore, be safer in their hands. He may perfectly well have been hostile to the Normans, and yet faithful to the Abbey so long as Brand held it; but the news that Turold and his knights were coming to make the Abbey a centre of Norman rule against him [¹⁶] would drive him to extreme courses. Professor Tout has made some use of Hugh, but says, strangely, that 'the stern rule of the new Abbot Turold drove into revolt the tenants', when his rule had not yet begun.

Again, there is now no doubt where Hereward ought 'to be quartered'. Two other places with which the Domesday survey connects him are Rippingale and, possibly, Laughton to the north of Bourne. Living thus on the edge of the fenland, he may well have been a leader among 'that English folk of the fenlands' who rose, says the Peterborough Chronicle, in the spring of 1070, to join the Danish fleet and throw off the Norman yoke. And the prospect of being ousted from his Peterborough lands by a follower of the new French abbot would have added a personal zest to his patriotic zeal.

Mr Freeman, followed by Professor Tout,[¹⁷] holds that the story in the false Ingulf is not to be wholly cast aside, as it may contain some genuine Crowland tradition;[¹⁸] but he has not accurately given that story. It might hastily be gathered, as it was by him, that it was Hereward's mother-in-law who 'very considerately takes the veil at the hands of Abbot Ulfcytel', whereas it was, according to the Gesta, his wife who did this. The Gesta version, he writes, 'of Turfrida going into a monastery to make way for Ælfthryth is plainly another form of the story in Ingulf, which makes not herself but her mother do so'. But if the Historia Ingulphi (pp. 67-8) be read with care, it will be seen that 'mater Turfridæ' should clearly be 'mater Turfrida', the reading that the sense requires. So there is here no opposition, and Ingulf merely follows the Gesta version.

As for the honour of Bourne, it can be shown from the carta of Hugh Wac in 1166, from our list of knights, and from the Pipe-Roll of 1130, to have been formed from separate holdings and to have descended as follows:

1138 [¹⁹]
(see p. [359])

The Psuedo-Ingulf's version runs:

It will be seen how skilfully the author of this famous forgery brings in the names of real people while confusing their connection and their dates. Richard de Rullos, for instance, was living shortly before 1130, yet is here described as living under the Conqueror, though represented as marrying the great granddaughter of a man who was himself in the prime of life in 1062. The whole account of him as an ardent agriculturist, devoted to the improvement of live-stock and the reclamation of waste, is quaintly anachronistic; but the fact of his being a friend and benefactor to Crowland is one for which the writer had probably some ground. For my part, I attach most importance to his incidental statement that the daring deeds of Hereward the outlaw, 'adhuc in triviis canuntur', an allusion, perhaps unnoticed, to a ballad history surviving, it may be, so late as the days when the forgery was compiled.

But, leaving Hereward, no entries in this list are more deserving of notice than those which bring before us the famous name of Nevile:

Gislebertus de Nevila [tenet] ii. carrucatas in Lincolnescira, et servit Abbatiæ pro ii. hidis et inde inventi i. militem (p. 171).

Radulfus de Nevila [tenet] x. carrucatas in Lincolnescira et i. hidam et dimidiam in Hamtonascira et servit se tercio milite (p. 175).

Hugh Candidus wrote of the former:

Heres Galfridi de Nevile tenet in Lincolnescire, scilicet in Waletone [sic] justa Folkingham, et Yoltorpe duas carrucatas terra et inde facit plenum servitium unius militis (p. 59).

With this clue we are enabled to detect Gilbert de Nevile in that 'Gislebertus homo Abbatis', who held of the Abbot (D.B., i. 345b) at 'Walecote' (Walcot near Folkingham). So also Hugh 'Candidus' writes of the other Nevile fee:

Heres Radulfi de Nevile tenet decem carrucatas terræ in Lincolnshire, scilicet in Scottone Malmetone; et in Norhamtonscire unam hidam et dimidiam, scilicet in Holme, Rayniltorp, et inde facit plenum servitium trium militum (p. 55).

It is, then, Ralf de Nevile that we have in that 'Radulfus homo Abbatis', who held of him at 'Mameltune', and 'Rageneltorp' with 'Holm' in Domesday (i. 345b, 346)—Manton, with Raventhorpe and Holme (near Bottesford, co. Linc.)—for Hugh, of course, has blundered in placing the two latter places in Northamptonshire.[²⁰] The Testa, more exact, enables us to add Ashby to Holme and Raventhorpe as part of one estate, held as a single knight's fee. Scotton, in the same neighbourhood, was held by 'Ricardus' in Domesday, but, in the hands of Nevile's heirs, represented a fee and a third.

Between Ralf and Gilbert de Nevile on fo. 346 we find 'Gislebertus homo Abbatis' holding ten bovates at Hibaldstow. This was the 'Gislebertus Falvel' of our return, not Gilbert de Nevile.

The last Domesday name I shall identify is that of the Abbot's under-tenant 'Eustacius', who held of him at Polebrook, Clapton (Northants), and Catworth (Hunts). He was, I believe, the same as that Eustace who held land, as a tenant-in-chief, at Polebrook, Northants, and with that Eustace the sheriff ('Vice-comes') who held (at Catworth, Hunts) also in capite. Indeed the abbot's tenant is identified with the latter in the story of the foundation of Huntingdon Priory (Mon. Ang., vi. 78), where, as in our list, we find that his two knights' fees soon passed to Lovetot.[²¹]

We may learn from this identification that two different tenants-in-chief and at least one under-tenant may prove to be all one man, just as, on the other hand, we found three distinct Rogers among the Domesday under-tenants of the Abbot. An additional conclusion is suggested by the name 'Eustachius de Huntendune', given to this sheriff in the Inquisitio Eliensis.[²²] For we find Picot, the Sheriff of Cambridgeshire, similarly styled in Domesday (i. 200), 'Picot de Grentebrige'. 'Ilbert de Hertford', I think, was the Sheriff of Hertfordshire,[²³] and Hamo, a contemporary sheriff of Kent, attests a charter as 'Hamo de Cantuaria'. Turold, sheriff of Lincolnshire, is found as Turold 'of Lincoln' (see p. [255]), and Hugh, sheriff of Dorset, as Hugh of 'Wareham', while Walter and Miles 'of Gloucester', Edward and Walter 'of Salisbury', are also cases in point. Hugh 'of Leicester' was sheriff of Leicestershire temp. Henry I, while Turchil 'de Warwic' (D.B., i. 240b) may possibly have owed that appellation to the fact that his father Ælfwine was sheriff of Warwickshire. Enough, in any case, has been said to show that it was a regular practice for sheriffs to derive, as often did earls, their styles from the capital town of their shire.

[¹] Society of Antiquaries' MS. 60.

[²] Ed. Camden Society.

[³] Norman Conquest, iv. 219. We know aliunde that 'Picot filius Colsuani' was the son of Colswegen of Lincoln. It would seem to be of this estate that we read in the 'Clamores': 'Abbas de Burg clamat iiii. bov. terræ in Risun terra Colsuani, et Wap' testatur quod T.R.E. jacuerunt in æcclesia Omnium Sanctorum in Lincolia.'

[⁴] Society of Antiquaries' MS. 60. Printed by Stapleton ut supra.

[⁵] But possibly the Robert d'Oilli of our list may be the first Robert (who, as 'Robertus' in Domesday, held Cranford of the Abbot), while the tenant of that name in 1125 may be the second Robert, entered in the Pipe-Roll of 1130, and living temp. Stephen.

[⁶] William Rufus, i. 571. He makes it 'Evermouth' in the Norman Conquest.

[⁷] Envermeu lay on the coast some 19 miles to the east of Dieppe.

[⁸] 'The legend of Hereward' (Norman Conquest, iv. [1st Ed., 805).

[⁹] With its hamlet of Manthorpe and Toft with Lound.

[¹⁰] Ed. Sparke Historiæ Anglicanæ Scriptores [1723].

[¹¹] Professor Tout throws out the unlucky suggestion: 'the Wake, i.e. apparently the watchful one'.

[¹²] See the new Monasticon on Deeping Priory, and the rubric to Baldwin's charter. The true parentage of Baldwin fitz Gilbert will be shown infra in the paper on 'Walter Tirel and his wife'.

[¹³] Norman Conquest (1st Ed.), iv. 455-6.

[¹⁴] Norman Conquest (1st Ed.), iv. 484. Professor Tout, however, follows Mr Freeman, and accepts an earlier 'flight from England' as a fact. One must therefore insist that 'the whole story has no historical basis'.

[¹⁵] I am tempted, indeed, to suggest that Hugh may have had before him that lost local 'account of Hereward's doings', which was inserted (but, according to my own view, in an abbreviated form) into the earlier chronicle, according to Professor Earle (see Norm. Conq., iv. 461, note 3). This solution would explain everything, and would, if accepted, greatly increase the importance of Hugh's chronicle.

[¹⁶] Cf. William of Malmesbury in loco.

[¹⁷] Dictionary of National Biography.

[¹⁸] Appendix on 'the Legend of Hereward', ut supra.

[¹⁹] The names of the churches he bestowed on the Priory illustrate the constituents of the Honour of Bourne.

[²⁰] The name of Ralf de Nevilla occurs in full in the Lincolnshire 'Clamores' (i. 376b), annihilating the old assertion that this famous surname is nowhere found in Domesday. (See my letter in Academy, xxxvii. 373.)

[²¹] It is specially interesting to trace his holding at Winwick, Hunts, which then lay partly in Northants. As 'Eustachius' he held in capite at 'Winewincle' (i. 228), as 'Eustachius Vicecomes' at 'Winewiche' (i. 206), and as 'Eustacius', a tenant of the Abbot, at 'Winewiche' (i. 221). In the first two cases his under-tenants are given as 'Widelard[us]' and 'Oilard[us]', doubtless the same man. For 'Winewincle' we should probably read 'Winewicke'. See also p. 222, infra.

[²²] Inq. Com. Cant., Ed. Hamilton, p. 111.

[²³] Ibid., 56, 192.