Feature of the month: No.10 March 2010

'Absent' and dead charter-witnesses

Professor Dauvit Broun, PoMS Principal Investigator ¹

Witness-lists

A crucial source for this project's database of information on everyone in Scotland mentioned in documents between 1093 and the death of Alexander III on 19 March 1286 is the list of witnesses that are typically found at the end of a charter. It seems natural to assume that everyone mentioned there was present when the charter was made. It is true that we must be aware of a very few charters from early in the period, when such documents were still something of a novelty, in which the record of more than one donation was combined. As a result, different witness-lists were merged, so that it is possible to find the dead rubbing shoulders with the living. ² There is also a type of document where, instead of listing the names of witnesses, each was represented by the sign of the cross with their name written around it or beside it. These have been found elsewhere, and before our period, to include names which have been added much later. ³ No instances of this have yet been identified in Scotland. Such highly unusual curiosities do not threaten to undermine our confidence that, as far as the typical types of charter in our period are concerned, those listed as witnesses were present on the same occasion.

'Absent' witnesses: a classic case

Historians have been haunted for over a century by doubts about whether this is really the case. The original cause of this anxiety was the discovery of instances where a donor has written to individuals to ask if they would witness a charter which has already been drawn up. The classic case of this, published in facsimile in 1903, is from England in the late twelfth century. ⁴ Both the charter and the letter of the donor inviting his friends to be witnesses survive as 'originals'—that is, on the very sheets of parchment on which they were written, rather than as copies in a cartulary (i.e., a manuscript-book of charters). The first thing that strikes you is that the witnesses appear in exactly the same order in the charter and in the letter of invitation. If you look closely at the facsimile reproductions of the originals it also seems that both were written by the same scribe—albeit that the writing in the letter is in a lower register than the charter. This impression is confirmed if the originals are inspected 'live'. There seems little doubt, therefore, that both the charter and the letter of invitation were written at the same time, and that none of the witnesses would have been present when the charter was drawn up.

'Absent' witnesses: Scottish examples

Evidence has also been found for this practice in the cartulary of Newbattle Abbey (a Cistercian monastery which celebrated 1140 as the year of its foundation by David I). ⁵ There are a number of occasions where the scribe, after a charter (or at the end of a batch of charters relating to a particular property) has noted that there are some documents which he has decided not to copy out for one reason or another. More often than not it is because they were 'repeats' of an earlier charter (i.e., it had originally been produced not simply as a single sheet of parchment, but as two sheets with identical text, each presumably with the donor's seal attached). Nonetheless he thought it appropriate to note their existence for future readers in his monastery. In two cases the documents he left uncopied are letters to witnesses who had, it seems, been named as witnesses of a charter in their absence. In the first instance he wrote:

'Memorandum that Mary of Hailes made the charter of her donation three times with the same contents, and moreover that she wrote to those who she used as witnesses supposing that they would have agreed to it if their names were put in the charter.' ⁶

In the second instance we are told that a charter by William Noble had been made twice, and that along with it there was 'a letter of humble request by the said William Noble to all the individuals who would be willing to be witnesses in his charter'. ⁷ Here, it seems, we have two cases of letters written to people asking them to be witnesses in a charter that has already been produced with them named as witnesses. There may perhaps be a third case: we are informed of a charter in duplicate, and then of 'a certain letter of humble request whose value expired long ago'. ⁸ The fact that the scribe was more forthcoming about the others, however, suggests that this letter may have been different. We are told nothing about its contents, or who wrote it (unless it may be inferred that it was connected with the duplicate charter). It was known, for example, for a donor to write a letter asking someone to append their seal to his charter. ⁹

Fortunately not every cartulary scribe in Scotland chose to note the existence of such letters rather than copy them. The only other known example survives because it was copied into the oldest of the two cartularies of Melrose Abbey (a Cistercian monastery which celebrated 1136 as the year of its foundation by David I). It relates to a gift by the lord of Wooler in Northumberland. ¹⁰ Both the charter of donation and the letter to the charter-witnesses have been included. ¹¹

The archival perspective

It is very striking that this evidence that witnesses were not, in fact, present has been so diligently preserved. All cases of this kind in England as well as Scotland involve records of donations to religious houses, and they survive only because the evidence was retained by them. It is also very likely that both the charters and the letters inviting people to be witnesses were written by the religious houses themselves. This suggests that they were keen to show how the witnesses would have known about the donations, even though they were absent when the charter was drawn up and sealed. The hope, presumably, was that this would be sufficient if they were called upon to testify to the charter. It would certainly have been fatal for the charter's credibility if, in the event of a dispute going to trial, the witnesses declared that they had no knowledge of it or its contents. Perhaps the beneficiary might also have hoped that a connection between charter and letter would be demonstrable by the fact that, in the letter's address clause, the recipients were identical with the charter-witnesses.

How common was this use of 'absent' witnesses?

Overall, three examples (with a possible fourth) can be identified among the many hundreds of charters of donation that survive from Scotland in our period. All can be dated probably or certainly to the early decades of the thirteenth century. We can be fairly confident, in the cases of Melrose and Newbattle, that had any more such letters existed they would have been noted in the Newbattle cartulary or copied into the thirteenth-century Melrose cartulary. ¹² The helpful habit of the Newbattle scribe of telling us when documents have not been copied—even when they were originally produced as two or three identical charters—allows us to see that in every case of 'absent' witnesses the charters were produced in multiple copies. Although charters were not regularly created in duplicate (never mind triplicate), the practice was not unusual. If you had two copies (or more) of the original charter, each sealed by the donor, then you were less vulnerable to losing a written record of the donation (or whatever else that was recalled in the charter). In the case of Newbattle, it seems that this was not routine practice. This safeguard was applied to only a minority of charters of donation, even in the limited period of Alexander II's reign (1214–49). ¹³ This suggests that they may have had cause to be anxious about these particular gifts. The point of obliging the donor to put their seal to repeated copies of their charter may have been to impress on him or her (and their potential heirs) that Newbattle were never likely to lose it. It may have served to reinforce the hoped-for permanence of the gift. ¹⁴

It is noteworthy that the two Newbattle cases of charters with 'absent' witnesses (and possibly also the much less certain example) were produced in multiple copies, including the only example of a charter produced in triplicate. Was the concern here simply that the witnesses were not present? This was hardly going to be improved by obtaining more than one copy. Instead of supposing a direct connection, it might be more plausible to envisage that these two phenomena—the 'absent' charter-witnesses and the multiple copies—were independent symptoms of a single problem. If 'absent' witnesses were an irregular occurrence, then they could be an indication that the charter was produced in a hurry. And if it was produced in a hurry, that could well be because the circumstances behind the donation were unusual, which could in turn explain the unease indicated by the creation of multiple copies. It is possible to take this even further, and ask if these charters and their accompanying letters to witnesses should be accepted as genuine statements by the donor. ¹⁵

Does this evidence for 'absent' witnesses undermine the general sense that witnesses were usually present when a charter was drawn up? All the indications point to 'no' as the answer. The evidence suggests that 'absent' witnesses were an unusual measure that may only have arisen in peculiar circumstances. It would, almost certainly, have made a charter more secure if its witnesses were, in fact, present when it was produced.

Charter-witnesses who witnessed the transaction, not the charter?

If we think closely about how charters were produced, a distinction can be drawn between the donation itself and the charter which recorded the donation. ¹⁶ The two could be separated by a considerable period—months, or even years. The donation was regarded as a ceremony in which something representing the gift was publicly transferred by the donor to the beneficiary. (This is known technically as 'livery of sasine'.) The charter was drawn up later as a record that this had occurred. There was, in fact, no necessity to have a charter at all (although, as time went on, it became increasingly common for a charter to be produced). It is conceivable, in theory at least, that the witnesses in the charter—who are, after all, giving an assurance that the donation occurred—were present only at the ceremony when the gift was formally transferred from donor to beneficiary, rather than at the subsequent writing and sealing of the charter itself. If this ever happened, however, some kind of record of the ceremonial transfer must have been kept. There are, in fact, some remarkable examples of charters that survive as drafts. ¹⁷ It is possible, therefore, that the ceremony of donation and the production of a charter, duly finished and sealed, could have been connected through a draft, even if the event and the sealing were separated by a significant period. On the other hand, both ceremony and charter could, of course, occur on the same occasion.

If ceremony and charter were separated by a significant gap, it would be difficult to find enough information in this period to show that the witnesses could only have been present at the symbolic transfer. This could be important if the finished charter was dated according to where and when it was sealed, because we would naturally read it as meaning that the witnesses were present then, rather than (say) months earlier at the ceremony of donation (or whatever the transaction was). But is there any hard evidence that this is any more than a hypothetical problem?

A charter with a date later than the death of a witness?

Professor Barrow thought that he had identified among the charters of David I a clear-cut example of witnesses who were present not when the charter was finally produced and sealed but at the transaction itself a couple of months earlier. ¹⁸ The dating clause (an unusual feature in David I's charters) reads: 'On the day of the Invention of the Holy Cross in the year 1147 from the incarnation of the Lord, that is, in the year that the king of France and many Christians went forth (perrexerunt) to Jerusalem'. The Invention (i.e., finding) of the Holy Cross was celebrated on 3 May, but King Louis VII did not leave France on the Second Crusade until June 1147. Professor Barrow argued that it would have taken until July or August for news of Louis' departure to have reached Scotland, so the charter cannot have been written until months after 3 May. The date of 3 May must, therefore, refer to the transaction itself (a confirmation by David I of a gift to Coldingham Priory which also features a record of arrangements for the performance of services due to the king from the lands in question). One of the witnesses, Bishop John of Glasgow, had died in the interim, so he (and presumably the other witnesses) must have been present at the transaction rather than at the writing and sealing of the charter.

This all hinges on the assumption that no-one in Scotland on 3 May could have described the crusade as already underway. As Professor Richard Sharpe has pointed out to me, however, by 3 May anyone from Scotland taking part in the crusade must either have departed, or been very close to doing so. From a Scottish point of view it could, therefore, have been appropriate to regard the crusade as already under way. It is true that the charter also refers to the king of France's departure, which plainly had not yet occurred. If, however, the crusaders from Scotland regarded him as the preeminent leader of the crusade, then the statement overall might still be regarded as feasible from a Scottish point of view, regardless of the actual date of departure from France. Although some might regard this as special pleading, it at least means that the case highlighted by Professor Barrow may not be quite so clear-cut after all.

A clear-cut case of a dead witness

There is one charter of donation of standard type in which one of the witnesses was already dead by the time the charter was drawn up. In this case the witness—if he witnessed anything at all—must have been present at the ceremonial transfer of the donated land, rather than at the drawing up and sealing of the charter. The charter will be given in translation in April's Feature of the Month. It records the gift of Balfeith (in the parish of Fordoun in the Mearns) by Humphrey of Berkeley, who in an associated document (also to be translated in April's Feature) is referred to as sheriff of the Mearns at that time. In the charter the gift is said to be for the souls of generations of royalty, from David I to Alexander, son of William the Lion. The charter, therefore, cannot have been written before Alexander's birth on 24 August 1198. One of the witnesses, however, is Walter Scot the elder (whose son, Walter, is also a witness). Walter Scot the elder is recorded elsewhere as giving a death-bed testimonial to Isaac, a clerk of the bishop of St Andrews, when Isaac stayed at Arbuthnott on his way north to invite the bishops of Moray, Ross and Caithness to attend the consecration of Roger, bishop of St Andrews, which took place on 15 February 1198. ¹⁹ There is no way round this without indulging in some very special pleading! It has to be admitted that Walter Scot was surely dead and buried by the time the charter was drawn up.

A dead witness was of little use as far as people in this period were concerned. They would be unable to vouch for the charter and its contents. Once all witnesses had died, a charter would be less effective as a means of safeguarding a gift or other transaction—although it would still retain some value. It was not uncommon, for example, for scribes of cartularies in the thirteenth century to omit the witnesses when copying charters. If all the witnesses named in a charter were dead, the charter's validity as evidence could be called into question. For example, in a dispute between the abbeys of Cambuskenneth and Dunfermline which went to arbitration in 1207, a charter of David I in favour of Cambuskenneth Abbey was produced in evidence. ²⁰ The arbitrators (the bishop of St Andrews, abbot of Arbroath, archdeacon of St Andrews and Official of St Andrews) were unsure about whether it could be accepted because the witnesses were no longer alive. They applied to Pope Innocent III for advice. He replied that the charter could be used in evidence if it was accepted practice in Scotland to have faith in charters of this king (a statement that found its way into the Decretals, the compendium of papal pronouncements which constituted church law). This allowed the issue of the lack of a living witness to be sidestepped. Without this manoeuvre, it is unlikely that Cambuskenneth would have derived any benefit from the charter in this case.

If no-one would deliberately name a dead witness in a charter, how had Walter Scot the elder come to be included in the witness-list of Humphrey of Berkeley's charter? This can only have happened inadvertently. The simplest explanation is that the witness-list was copied in its entirety from an earlier record of the donation. What might this earlier record have been? This question will be addressed head-on in the Feature for April. Suffice to say at this stage that the explanation could, at worst, imply that a charter of donation which includes conspicuous detail about who carried out a perambulation may have included witnesses to a ceremony of donation rather than witnesses to the charter itself. Only a very small proportion of charters fall into that category. In some that do, it is made clear what is going on—for example, that witnesses and perambulators are one-and-the-same. ²¹ Overall, it seems likely that, for the vast majority of charters, the witnesses who are listed were indeed present when the document was produced. This is important, because for the majority of charters the witnesses are usually the only evidence there is for when they were produced. In only a minority of cases is a date given in the charter itself. The witness-lists are also crucial evidence for social interactions and relationships. As far as that goes, however, the chief interest is seeing who the donor and beneficiary thought should be named—and that would work just as well if the witnesses were present or were only made aware of the charter by letter afterwards. A witness-list is never simply a haphazard list of those who happened to be present at the time.