RICHARD THE FIRST'S CHANGE OF SEAL (1198)
With the superficial student and the empiric politician, it is too common to relegate the investigation of such changes to the domain of archæology. I shall not attempt to rebut the imputation; only, if such things are archæology, then archæology is history.—Stubbs, Preface to R. Hoveden, IV, lxxx.
Historical research is about to pass, if indeed it is not already passing, into a new sphere—the sphere of Archæology. The central idea of that great advance which the present generation has witnessed in the domain of history has been the rebuilding of the historical fabric on the relatively sure foundation of original and contemporary authorities, studied in the purest texts. Chronicles, however, are not inexhaustible: for many periods they are all too few. The reaper has almost done his work; the turn of the gleaner has come. The smaller quellen of history have now to be diligently examined and made to yield those fragments of information which will supplement, often where most needed, our existing stock of knowledge.
But this is not our only gain as we leave the broad highways trodden by so many before us. Those precious fragments which are to form our spoils will enable us to do more than supplement the statements of our standard chroniclers: they will afford the means of checking, of testing, by independent evidence, these statements, of submitting our witnesses to a cross-examination which may shake their testimony and their credit in a most unexpected manner.
As an instance of the results to be attained by archæological research, I have selected Richard the First's celebrated change of seal. Interesting as being the occasion on which the three lions first appear as the Royal arms of England—arms unchanged to the present day—it possesses exceptional historical importance from the circumstances by which it was accompanied, and which led, admittedly, to its adoption.
Historians have agreed, without the least hesitation, to refer this event to the year 1194, and to place it subsequent to the truce of Tillières or about the beginning of August. 'That Richard I,' writes a veteran student,[1] 'adopted a new seal upon his return from the Holy Land is a matter of notoriety.' Speed, in fact, had shown the way. We are told by him that 'the king caused [1194] a new broad seale to be made, requiring that all charters granted under his former seale should be confirmed under this, whereby he drew a great masse of money to his treasurie'.[2] The Bishop of Oxford, with his wonted accuracy, faithfully reproduces the statement of Hoveden (the original and sole authority we shall find for the story), telling us that 'Amongst other oppressive acts he [Richard] took the seal from his unscrupulous but faithful chancellor, and, having ordered a new one to be made, proclaimed the nullity of all charters which had been sealed with the old one.'[3] Mr Freeman similarly places the episode just before 'the licenses for the tournaments' (August 20, 1194), and consistently refers to Dr Stubbs's history.[4] Miss Norgate, in her valuable work, our latest authority on the period, assigns the event to the same date, and tells us that 'Rog. Hoveden's very confused account of the seals is made clear by Bishop Stubbs'.[5] Mr Maitland, in his noble edition of 'Bracton's Note-book', gives a case (ii. 69) in which a charter sealed 'secundo sigillo Regis Ricardi' was actually produced in court (1219), and explains that 'Richard had a new seal made in 1194', referring to Hoveden for his authority.[6]
It should be observed that all these writers rely merely on Hoveden, none of them throwing any light on the process of confirmation, or telling us how it was effected, and whether any traces of it remain. An independent writer, M. Boivin-Champeaux, in his monograph on William Longchamp, discusses the episode at some length, and asserts that the repudiated documents were 'assujettis, pour leur revalidation, à une nouvelle et coûteuse scellure'. Like the others, however, he relies on the authority of Hoveden, and consequently repeats the same date.
In the course of examining some ancient charters, I recognized one of them as nothing less than an actual instance of a confirmation consequent on this change of seal. But its incomprehensible feature was that the charter was confirmed on August 22, 1198, having originally been granted, 'sub primo sigillo', so recently as January 7th preceding. How could this be possible if the great seal had been changed so early as August 1194, and if the first seal, as stated by Dr Stubbs, was 'broken' on that occasion? Careful and prolonged research among the charters of the period (both in the original and in transcripts) has enabled me to answer the question, and to prove that (as, of course, the above charter implies) the change of seal did not take place in 1194, but 1198, and between January and May of that year.
Original charters under the second seal, confirming grants under the first, are distinctly rare. I have found, as yet, but one in the Public Record Office, and only two at the British Museum. But of originals and transcripts together I have noted twenty-eight. The dates of the original grants range from September 5, 1189, to January 7, 1198 (1197-8), and of the confirmations from May 27, 1198, to April 5, 1199.[7]
In a single instance there is fortunately preserved not only the text of the confirmation charter, but also that of the original grant.[8] From this we learn that the charter of confirmation did not necessarily give the wording, but only the gist ('tenor') of the original grant. We are thus brought to the instructive formula invariably used in these charters:
Is erat tenor carte nostre in primo sigillo nostro. Quod quia aliquando perditum fuit, et, dum capti essemus in alem[anniâ], in aliena potestate constitutum, mutatum est. Huius autem innovationis testes sunt Hii, etc., etc.
We may here turn to the passage in Hoveden [ed. Stubbs, iii. 267] on which historians have relied, and see how far the reasons for the change given in the charters themselves correspond with those alleged by the chronicler.
Fecit sibi novum sigillum fieri, et mandavit, per singulas terras suas, quod nihil ratum foret quod fuerat per vetus sigillum suum; tum quia cancellarius ille operatus fuerat inde minus discrete quam esset necesse, tum quia sigillum illud perditum erat, quando Rogerus Malus Catulus, vicecancellarius suus, submersus erat in mari ante insulam de Cipro, et præcepit rex quod omnes qui cartas habebant venirent ad novum sigillum ad cartas suas renovandas.
In both cases we find there are two reasons given; but while one of these is the same in both, namely the temporary loss of the seal when Roger Malchael was drowned, the other is wholly and essentially different. The whole aspect of the transaction is thus altered. To illustrate this I shall now place side by side the independent glosses of the Bishop of Oxford and of M. Boivin-Champeaux:
Thus both writers assume that there were two seals, one which remained in England with the chancellor, and one which accompanied the king to the east. They further (though Dr Stubbs is somewhat obscure) hold that the two excuses given refer respectively to the two seals, thus discrediting both. But when we turn to the charters themselves, we find but one seal mentioned, and to that one seal alone both the excuses refer. The king explains that on two occasions it was, so to speak, 'out on the loose'—(1) when his vice-chancellor was drowned; (2) when he himself was captured in Germany. This was, of course, the seal which accompanied him to the east.[9] The king makes no allusion to any other or to the chancellor. Such charters and grants as are known to us all proceed from the king himself, either before he left Messina or after he had reached Germany on his return. No charter or grant of Longchamp, as representing him, is known. In short, the whole of our record evidence points one way: the charters which the king proclaimed must be confirmed, and which we find brought to him for that purpose were those which he had himself granted, and no other. Lastly, even had we nothing before us but the passage in Hoveden which all have followed, I contend that it may, and indeed ought to be, read as referring to a single seal. But it is, as Miss Norgate justly observes, 'very confused', from its allusion to the chancellor's use of the seal. That allusion, however, would most naturally refer to the truce of Tillières, and not to the use of a separate seal in England. Therefore even if we accepted, which I do not, Hoveden's statement, it would not warrant the inference that has been drawn.
Again, when Miss Norgate writes of the 'withdrawal of the seal from William', and when Dr Stubbs tells us that the king 'took the seal from' him, these statements may have two meanings. But M. Boivin-Champeaux is more precise: 'L'emploi de ces procédés emportait le mépris et la violation non seulement de tous les actes étrangers au chancelier, mais encore de tous ceux où il avait mis la main. Il ne pouvait décemment conserver les sceaux. Le roi les lui enleva.' This is a distinct assertion that Longchamp was deprived of his office. Yet all our evidence points to the conclusion that he remained chancellor to the day of his death.
Dismissing Hoveden for the time, and returning to the testimony of the charters, we have seen that they point to the event we are discussing having taken place in 1198, between January 7, at which date the first seal was still in use, and May 27, when charters were already being brought for confirmation under the second seal. Passing now from the charters to the seals still in existence, we learn from Mr Wyon's magnificent work[10] (which has appeared since I completed my own investigation) that the first seal was still in use on April 1, 1198,[11] while an impression of the second is found as early as May 22, 1198.[12] Thus our limit of time for the change is narrowed to April 1-May 22, 1198.[13] The evidence of the charters and of the seals being thus in perfect harmony, let us see whether this limit of date corresponds with a time of financial difficulty. For, so desperate a device as that of the king's repudiation of his charters would only have been resorted to at a time of extreme pressure. What do we find? We find that the time of this change of seal corresponds with the great financial crisis of Richard's reign. The Church had at length lost patience, and had actually in the Council at Oxford (December 1197) raised a protest. The 'want of money', in Miss Norgate's words, was 'a difficulty which ... must have seemed well-nigh insurmountable'. Preparations were being made for a huge levy at five shillings on every ploughland. It was at this moment that the desperate king repudiated all the charters he had granted throughout his reign, and proclaimed that they must be 'brought to him for confirmation; in other words ... paid for a second time'.[14]
Let us now look at the other chroniclers. R. Coggeshall is independent and precise:
Accessit autem ad totius mali cumulum, juxta vitæ ejus terminum, prioris sigilli sui renovatio, quo exiit edictum per totum ejus regnum ut omnes cartæ, confirmationes, ac privilegiatæ libertates quæ prioris sigilli impressione roboraverat, irrita forent nec alicujus libertatis vigorem obtinerent, nisi posteriori sigillo roborarentur. In quibus renovandis et iterum comparandis innumerabilis pecunia congesta est (p. 93).
This is in complete accordance with the now ascertained fact that Richard changed his seal, and regranted the old charters, within the last year of his life. Similarly independent and precise evidence is afforded by the Annals of Waverley:
mcxcviii. Anno x. regis Ricardi præcepit idem rex omnes cartas in regno suo emptas reformari, et novo sigilli sui impressione roborari, vel omnes cassari, cujuscunque dignitatis aut ordinis essent, qui vellent sua protectione defensari, vel universa bona sua confiscari.[15]
Further, we read in the Annals of Worcester[16] and in the Historia Major of M. Paris (ii. 450-451)[17] that in 1198, 'circaque festum sancti Michaelis, mutatæ sunt carte quas prius fecerat rex Ricardus, novo sigillo suo'. Now this Michaelmas fell just in the heart of the period within which the process of confirmation is proved to have been going on.
We see, then, that the evidence (1) of the seals, (2) of the charters, (3) of the circumstances of the time, (4) of other chroniclers, all concur in pointing to the spring of 1198. And now we will lastly appeal to Hoveden against himself. After telling us of the king's proclamation on the refusal of the religious to contribute to the carucage in the spring of 1198, he adds:
Præterea præcepit idem rex ut omnes, tam clerici quam laici, qui cartas sive confirmationes habebant de sigillo suo veteri deferrent eas ad sigillum suum novum renovandas, et nisi fecerint, nihil quod actum fuerat per sigillum suum vetus ratum haberetur (iv. 66).
This passage, which ought to be compared with Coggeshall, is merely ignored by Dr Stubbs. Miss Norgate, however, boldly explains it as 'a renewal of the decree requiring all charters granted under the king's old seal to be brought up for confirmation under the new one' (ii. 356). But the passage stands by itself, as describing a new measure.[18]
The only conclusion to be drawn from this cumulative evidence is that the earlier passage in Hoveden (1194) which has been so universally accepted, must be rejected altogether. Against the facts I have adduced it cannot stand.
Incredible though it may seem that a court official, a chronicler so able and well informed, indeed, in the words of his editor, 'our primary authority for the period',[19] should have misstated so grossly an event, as it were, under his own eyes, we must remember that 'Hoveden's personality is to a certain degree vindicated by a sort of carelessness about exact dates'.[20] Yet even so, 'few are the points', our supreme authority assures us, 'in which a very close examination and collation with contemporary authors can detect chronological error in Hoveden'.[21] Nor, of the eight anachronisms laboriously established by Dr Stubbs, does any one approach in magnitude the error I have here exposed. The importance of every anachronism in its bearing on the authorship of the chronicle is by him clearly explained.
How far does the rejection of this statement on the change of seal affect the statement which precedes it as to the Truce of Tillières? Hoveden places the latter and the former in the relation of cause and effect:
Deinde veniens in Normanniam moleste tulit quicquid factum fuerat de supradictis treugis, et imputans cancellario suo hoc per eum fuisse factum, abstulit ab eo sigillum suum, et fecit, etc. (iii. 267).
This is rendered by Dr Stubbs in the margin: 'He annuls the truce and all the acts of the chancellor passed under the old seal.' The passage has also been so read by M. Boivin-Champeaux (p. 221); but if that is the meaning, which I think is by no means certain, Hoveden contradicts himself. For he speaks five months later of the truce ('Treuga quæ inter eos statuta fuerat duratura usque ad festum omnium sanctorum') as not having stopped private raids on either side.[22] R. de Diceto, mentioning the truce (ii. 120), says nothing of it being annulled, nor does R. Newburgh in his careful account. On the contrary, he implies that it held good, though the terms were thought dishonourable to Richard (ii. 420). I should, therefore, read Hoveden as stating simply that Richard was much annoyed at ('moleste tulit') its terms, and was wroth with the chancellor for accepting them.
In addition to correcting the received date for Richard the First's change of seal, the evidence I have collected enables us, for the first time, to learn how and to what extent the confirmation of the charters was effected. We find that it was no sweeping process, carried out on a single occasion, but that it was gradually and slowly proceeding during the last eleven months of the king's life. Here, then, is the explanation of another fact (also hitherto overlooked), namely that only a minority of the charters were ever confirmed under the second seal.[23] For the king's death abruptly stopped the operation of that oppressive decree which was being so reluctantly obeyed.
It should be superfluous for me to add that, in thus correcting previous statements, I have not impeached the accuracy of our greatest living historian, who could only form his judgment from the evidence before him. The result of my researches has been to show that the evidence itself breaks down when submitted to the test of fact.
| Granted | at | Confirmed | at | Grantee | Authority |
|---|---|---|---|---|---|
| 16 April, 1194[24] | Winchester | 27 May, 1198 | Lions | Robert fitz Roger | Cart. Ant. EE. 6 |
| 2 December, 1189 | Canterbury | 15 June, 1198 | Château Gaillard | Hugh Bardulf | Cart. Ant. EE. 10 |
| 10 October, 1189 | Westminster | 1 July, 1198 | Château Gaillard | Ely | Cart. Ant. JJ. 43 |
| 28 November, 1189 | Canterbury | 1 July, 1198 | Château Gaillard | Ely | Cart. Ant. NN. 26 |
| 1 July, 1190 | Dangu | 3 July, 1198 | Château Gaillard | William Longchamp | Cart. Ant. JJ. 46 |
| 5 September, 1189 | Westminster | 30 July, 1198 | Lire | Rievaulx Abbey | Rievaulx Cartulary (Surtees Soc.), p. 308 |
| 17 September, 1189 | Geddington | 30 July, 1198 | Lire | Rievaulx Abbey | Rievaulx Cartulary (Surtees Soc.), p. 308 |
| 25 April, 1194 | 22 August, 1198 | Thomas Basset | Hist. MSS., 9th Report, ii. 404 | ||
| 12 December, 1194 | Chinon | 22 August, 1198 | Roche d'Orival | Alan Basset | Cott. Cart. xvi. 1 (Rymer i. 67) |
| 7 January, 1198 | Vaudreuil | 22 August, 1198 | Roche d'Orival | Alan Basset | Anc. Deeds, Ser. A. No. 5924 |
| 8 December, 1189 | Dover | 10 September [1198] | Château Gaillard | Shaftesbury Abbey | Harl. MS. 61, fo. 26 |
| 6 December, 1189 | Dover | 15 September [1198] | Château Gaillard | Peterborough Abbey | Cart. Ant. EE. 21 |
| 14 March, 1190 | Nonancourt | 18 September, 1198 | Château Gaillard | Waltham Abbey | Cart. Ant. RR. 7 & 8 |
| 23 March, 1190 | Rouen | 19 September, 1198 | Château Gaillard | Roger de Sancto Manveo | Cart. Ant. BB. 6 |
| 29 November, 1189 | Canterbury | 9 October, 1198 | Château Gaillard | Fontevrault | Cart. Ant. F. 1 |
| 6 October, 1189 | Westminster | 20 October, 1198 | Lions | St Leonard's, Stratford | Add. MS. 6, 166, fo. 341 |
| 7 December, 1189 | Dover | 24 October, 1198 | Château Gaillard | Stratford Langthorne Abbey | Cart. Ant. E. 1 |
| 23 March, 1190 | Rouen | 5 November, 1198 | Château Gaillard | St Jacques de Boishallebout | Add. Cart. (Brit. Mus.) No. 3 |
| 7 December, 1189 | Dover | 10 November, 1198 | Château Gaillard | Boxley Abbey | Cart. Ant. Q. 8 |
| 17 September, 1189 | Geddington | 12 November [1198] | Château Gaillard | St Alban's Abbey | Ancient Deeds, A. 1050 |
| 28 November,[25] 1189 | Canterbury | 13 November, 1198 | Château Gaillard | Tynmouth Priory | Cart. Ant. BB. 18 |
| 27 July, 1197 | Isle d'Andely | 14 November, 1198 | Château Gaillard | Llanthony Abbey | Cart. Ant. B. 26 |
| 10 November, 1189 | Westminster | 30 November, 1198 | Lions | The Templars | Deville's Transcripts |
| 5 August, 1190 | Marseilles | 7 December, 1198 | Lions | Church of Durham | Surtees Soc., vol. IX. p. lvi. |
| September, 1197 | Rouen | 17 December, 1198 | 'Sanctum Ebruskum' | Domus Dei (Southampton) | Cart. Ant. D. 30 |
| 1189 | [No place] | 24 January, 1199 | Cahagnes | Spalding Priory | Add. MS. 5844, fo. 228 |
| 15 April, 1190 | Evreux | 3 March, 1199 | Château du Loir | Gilbert fitz Roger | Hist. MSS., 10th Report, 325 |
| 22 June, 1190 | Chinon | 11 March, 1199 | Chinon | W. Briwerre | Great Coucher II. 1, 67 IV. (1, 2) |
| 25 April, 1194 | Portsmouth | 5 April, 1199 | [No place] | Noel 'serviens' | Cart. Ant. D. 30 |
[1] Canon Raine, Historiæ Dunelmensis Scriptores Tres (Surtees Soc.), p. 379.
[2] Speed's History (1611).
[3] Const. Hist., i. 506.
[4] Norman Conquest, v. 693. Compare The Office of the Historical Professor, pp. 16, 17: 'In a long and careful study of the Bishop of Chester's writings ... I have never found a flaw in the statement of his evidence. If I have now and then lighted on something that looked like oversight, I have always found in the end that the oversight was mine and not his.'
[5] England under the Angevin Kings, ii. 343.
[6] I have been able to identify this very charter.
[7] This is the only confirmation I have found later than March 3. If the date can be relied on, it is of special interest as being the day before the king died.
[8] Charters to W. Briwere, June 22, 1190, and March 11, 1199 (1198-9), transcribed in the Great Coucher (Duchy of Lancaster)
[9] Dr Stubbs, indeed, writes, as we have seen, that 'the seal that was now broken must have been the one which the chancellor had used during the king's absence'. But Longchamp had been ejected from the chancellorship in October 1191, whereas Richard limits the period of abuse to the duration of his captivity, which did not begin till December 20, 1192.
[10] The Great Seals of England (Stock), p. 149.
[11] Its impression is attached to a charter tested at Tours, now at Lambeth Palace. If the date of this charter is correctly given, it is an important contribution to the Itinerary of Richard.
[12] ibid., p. 19.
[13] It is singular that Mr Wyon, while giving these data, should himself assign the change to 'circ. 1197', and still more singular that he should elsewhere (p. 20) accept the usual passage from Hoveden (iii. 267).]
[14] Miss Norgate (1194), ii. 343.
[15] Annales Monastici, ii. 251.
[16] ibid., iv. 389 (Vespasian E, iv.).
[17] Faust A. 8. fo. 136. It is a striking instance of the confusion and blundering to be met with even in our best chronicles that M. Paris (Chron. Maj., ii. 356) has an independent allusion to the king's change of seal (as a 'factum Ricardi regis enorme') in which he gives us a circumstantial account of the event and of the prior of St Alban's going over to France to secure the confirmation, 'cum effusione multæ pecuniæ et laboris', but assigns it to the year 1189. Hoveden's error pales before such a blunder as this, which has been accepted without question by the learned editor, Dr Luard.]
[18] Hoveden, by placing it wrongly (p. 66) after Hubert's resignation (p. 48), to which it was some two months previous, has misled Miss Norgate into the belief that it was the work of his successor, Geoffrey.]
[19] Stubbs' Hoveden, iv., xxxii.
[20] ibid., p. xxv.
[21] ibid., p. xxxi.
[22] iii. 276. This distinctly implies that the truce had been nominally in full force. Note that it is here spoken of as 'till All Saints', while in the document itself (iii. 259) it is made for a year from All Saints. Miss Norgate (ii. 367) speaks of it as 'till All Saints' (1195), but I think it was made from July 1194 to All Saints 1195.]
[23] I have not found a single charter of municipal liberties, though the reign was so rich in them, among these confirmations. Nor since this article first appeared, in 1888 (Arch. Rev., vol. i.), have I found more than four additional cases of resealed charters, raising the total to twenty-eight. Of these a detailed list is given on pp. 442-15.]
[24] 'Scilicet die secunda coronationis nostræ.'
[25] 'December' in Cart. Ant., which date is accepted in Gibson's 'Monastery of Tynmouth'.