Redactions of many stages

I finished reading Professor Benoit-Michel Tock‘s book on subscriptions to charters about which I’ve been raving here lately over the month of November 2013, so those posts will shortly cease, but here’s just one more, because in the ninth chapter he hits at an issue that is kind of crucial to how we understand what medieval documents are telling us, the stories they tell about their own creation. I could go on about this for a while, and indeed have before and will again a bit below, but M. le Prof. Tock is briefer than I would be and also more stylish (though I have added emphasis to bring out things that are more ambiguous in English than in the French):

“The question is clearly important, since it concerns, in the most direct manner, the conditions in which the acts were made and, thereby, the links between the act and the judicial action. Is the narrative of the action as it appears in the act sincere? If the act was redacted before the action proper, is it not possible that last-minute changes in the action may not be reflected in the text? Most of all, if the act is prepared before the action, it can play a part in it, either as an act of title or as a symbol of the action.”¹

There are lots of ways this can play out, and M. le Prof. Tock goes through most if not all of them, finding all the weird cases where we can see, or occasionally are told, what happened to get the text and names onto a parchment and into an archive. There are more possibilities covered than he mentions in that paragraph: the document may indeed be prepared before the ceremony, and then be signed at the ceremony, or at least have the actors and witnesses mark crosses next to their names, or maybe just touch where the scribe has done it for them or sign the cross in the air above the page. In a few cases, however, it seems that the crosses were put on first and then the entire text written in above them later, presumably once it was back in the recipient’s scriptorium. (And at that rate, something Tock does not consider, who could stop the scribe writing an entirely different version of events?) It has been suggested before, to resolve the paradox of documents which claim that they were placed on an altar to complete the donation, but do so in the past tense as if to indicate that this happened before the document was actually written, that blank sheets might be so deposited and then have the charter written onto them afterwards; Tock’s cases (oddly, as with much of the rest of the book, more from the relatively slight archive of Nouaillé in Poitou than elsewhere²) may be the only evidence of this actually happening I’ve ever seen. He does also raise the rather worrying possibility that once that had been done, at the very bottom of the parchment, since autograph signatures seem to have thought desirable in the ninth century but much less so as time goes and even then not judicially necessary,³ the scribe writing up the fine version of the document on a parchment daubed with out-of-place crosses (“si on ose dire, comme un cheveu dans le soupe…”) might just clip the untidy bottom edge off once he’d written up proper formal versions of the signatures…⁴) If that possibility makes you as uneasy as it does me, our subjects themselves destroying our evidence of them, you must be a proper charter geek, congratulations!

Here’s one of Tock’s two examples, ARTEM 780, otherwise known as Niort, Archive Départementale de Deux-Sèvres, H 85, edited as Cédric Giraud, Jean-Baptiste Renault & Benoît-Michel Tock (edd.), Chartes originales antérieures à 1121 conservées en France (Nancy 2010), no. 780, unprinted. It’s a donation to the abbey of Saint-Maixent de Poitiers from 1108.

A much clearer example albeit in a worse image (my fault), Poitiers, Archive Départementale de Vienne, Carton 10, n^o 126, edited as P. de Monsabert (ed.), Chartes de l’abbaye de Nouaillé : de 678 à 1200, Archives historiques du Poitou, 49 (Poitiers 1936), no. 143, and now Giraud, Renault & Tock, Chartes originales, no. 1291, online here. It’s a donation of vineyards to St-Junien de Nouaillé of 1091. You can see how that bottom half-inch was lucky to survive. What’s it telling anyone after a century? This is a formal document we’re keeping here! Snip snip…

I especially love this stuff, however, because it doesn’t surprise me at all. Round about the same time Tock must have been dealing with the proofs of his book, I was checking over the text of this:

“To take an elaborate but forceful example, let us take the act recording the consecration of Santa Maria de Ripoll’s new church in 977. This was almost certainly composed by the learned Bishop of Girona, also Count of Besalú, Miró Bonfill, and it is partly past tense and partly present. The point at which it changes is after a description of the consecration of the several separate altars, as the assembled prelates seek to impress upon the local counts that the monastery’s possession of its various goods, ‘just as the schedule already prefaced teaches’, are placed beyond secular intervention. From this, and from the slightly different final witness list, it is clear that this change of tense reflects a move from an occasion past to a different, notionally present one. The elaborate and lengthy document thus stands between the two occasions grammatically; recording one as already past, the second, the proclamation of the quasi-royal immunity, unfolds in the text. This part of the text was presumably written before the second gathering; the former, given its detail, was probably composed after the ceremony, as beforehand the final order of events could not have been known. The act as it stands is thus not from either of the ceremonies it describes, or the date which it gives, and it adopts a narrative construction of the ceremony. Especially since consecrations were formal occasions whose rules were laid down, this narrative was under a strong impulse to say the correct thing. Santa Maria’s case, with numerous altars whose different patrons made different collaborations of bishops appropriate, bent this pattern but if events had in fact broken the convention, we cannot expect that this document would record this for us.”

This is from the first chapter of my doctoral thesis, and it was one of a large number of examples I was collecting, just as does Tock in his chapter, of procedural oddities that force us to question stereotypes about how documents were produced.⁵ We hit many of the same concerns with our examples. I’m not trying to claim massive cleverness for myself here or anything: firstly, I wouldn’t have got far along these lines without the truly excellent palaeographical analyses of the Sant Joan de Ripoll charter material by its editor, Federico Udina i Martorell, but more basically, as that suggests, these things become unmissable when you’ve just got enough original documents that they are preserved, and this is what Tock found once he had a dataset of 3658 original documents, just as I did with slightly fewer documents more deeply studied.⁶ Also I have to thank some of my scribes: Tock doesn’t have anyone like the famous judge Bonhom explaining that he wrote a given charter in two stints with two different inks, apparently afraid that someone might therefore think it forged.⁷ And when you have something like the hearing over the Vall de Sant Joan de les Abadesses, about which I have gone on so often, but where three writing stints at least are basically visible straight away, it’s hard not to come to these conclusions, like it or not.⁸ I love my source base!

Reduced-quality facsimile of the Vall de Sant Joan hearing charter, Arxiu de la Corona d’Aragó, Pergamins, Cancilleria, Miron 3, some enhancement applied

But it is possible that this source base means that I can answer one of Tock’s rhetorical questions. Considering the possibility that a charter might be entirely written up beforehand, he asks how a scribe could know who would turn up, and considers this a major problem with the concept. I’m not sure that it is, as that rather depends on anyone else ever noticing the charter didn’t say the right thing or people not being well aware that those people should have been there, even if in fact they weren’t, and that the document ought to say as much: this is, I assume, what explains the occasional Catalan cases where a dead man’s signature is given to emphasise that he consented to the act even if he died before it happened.⁹ I’m more convinced, however, by the cases where we can see that such information was available beforehand precisely because it turned out to be wrong, and the Vall de Sant Joan hearing is one of those, because both its list of those present for the oath it records and the list of those actually swearing it differ slightly from the names in the signatures. I’ve argued already that this is because the whole thing happened over maybe weeks, and there were probably two ceremonies involved, but it may also be Tock’s suggested pattern: what actually happened was not quite what had been anticipated.

Where have you argued this, I hear you ask? Why, in my Rulers and Ruled in Frontier Catalonia, 880-1010: pathways of power (Woodbridge 2010), available from at least some good booksellers!

At the very least, we can see that information was collected beforehand, because as I argue in my book, there are mistakes between the two sets of oath-swearers that would be hard to explain without written exemplars in the process. A woman signing for one hamlet who is called Marcia gets recorded as Imitara in the initial list, not exactly similar in pronunciation you might think, and it’s only when one realises that she’s the second woman called Marcia in that list that one realises one’s probably looking at a mistranscription of “item Marcia”, ‘another Marcia’, perhaps abbreviated in a way that the scribe didn’t recognise because it had been done by someone else. (I hypothesize that at the signature phase of writing, the people who’d written the lists were there to tell him what they said, which implies that he did the initial drafting elsewhere.) When I worked all this out, I felt really very clever indeed—ah me, so long ago now…—but it’s not really my brainpower so much as a really fabulous source and a willingness to tabulate long lists.¹⁰ Tock didn’t have the time of scope to do that work here, but his other writing shows that he’s perfectly capable of it; this book was painting a wider picture, is all.¹¹

I cannot find any pictures of the verso of a St Gallen charter! On the other hand, some of the Vorakte are pretty nearly as fine as the recto versions… so here’s a recto image of one from 825? Stiftsarchiv St. Gallen, II 65.

Nonetheless, the conclusion, that charter scribes worked from notes, is not present in Tock’s suggestions here, despite other evidence that could be brought into play like the apparent dorsal drafts versions made on the backs of eventual charters at St Gallen in Switzerland (which fall outside Tock’s sample and weren’t then digitised and studied as they now are), and I think it probably explains a lot of the difficulties he sees with the conclusions to which the documents force him.¹² They didn’t get things wrong that often because the information was being recorded elsewhere; the cases we have where things have gone wrong, as with Sant Joan or Sant Pere de Casserres or any of the others so dear to me, are where something so special or rushed was being done that it couldn’t be gone back on or was too much effort to consider rewriting. So we’re trying to reconstruct the process of charter redaction from its mistakes; not impossible, but challenging, and worth saying explicitly!

“La question est évidemment importante, puisqu’elle concerne, de la manière la plus directe, les conditions d’élaboration des actes, et par là, les liens entre l’acte et l’action juridique. Le récit de l’action tel qu’il figure dans l’acte est-il sincère ? Si l’acte a été rédigé avant l’action elle-même, n’est-il pas possible que des changements intervenus en dernière minute dans l’action ne soient pas reflétés dans le texte ? Surtout, si l’acte est préparé avant l’action, il peut jouer un rôle au course de celle-ci, soit comme acte dispositif, soit comme symbole de l’action.”

2. First printed as P. de Monsabert (ed.), Chartes de l’abbaye de Nouaillé : de 678 à 1200, Archives historiques du Poitou, 49 (Poitiers 1936)

3. Tock, Scribes, souscripteurs et témoins, pp. 193-223, esp. pp. 221-223.

4. Ibid. pp. 392-397, quote at p. 394.

5. J. Jarrett, “Pathways of Power in late-Carolingian Catalonia”, unpublished doctoral thesis (University of London 2005), online here, pp. 35-36, some day probably to be repeated in the paper for which I was reading Tock’s book…

6. Federico Udina Martorell, El Archivo Condal de Barcelona en los siglos IX-X: estudio crítico de sus fondos, Textos 18 (Barcelona 1951).

7. Anscari Manuel Mundó i Marcet, “El jutge Bonsom de Barcelona, cal·lígraf i copista del 979 al 1024″ in Emma Condello & Guiseppe De Gregorio (edd.), Scribi e colofoni: Le sottoscrizioni di copisti dalle origini all’avvento della stampa. Atti del seminario di Erice/X Colloquio del Comité international de paléographie latine (23-28 ottobre 1993), Biblioteca del «Centro per il collegamento degli studi medievali e umanistici in Umbria» 14 (Spoleto 1995) pp. 269-288; Jeffrey A. Bowman, Shifting Landmarks: Property, Proof, and Dispute in Catalonia around the Year 1000 (Ithaca 2004), pp. 84-99. The charter in question is the sacramental testament of Bishop Vives of Barcelona, printed as Àngel Fabrega i Grau (ed.), Diplomatari de la Catedral de Barcelona: documents dels anys 844-1260. Volum I: documents dels anys 844-1000, Fonts Documentals 1 (Barcelona 1995), doc. no. 265, and it’s huge, so the two-stint thing is understandable. Given it’s Bonhom I’m just faintly surprised he didn’t also tell us what he had for lunch in between times.

8. Jarrett, Rulers and Ruled, pp. 37-38, which took a lot longer figuring out than those two paragraphs give away…

9. An example of this is Ramon Ordeig i Mata (ed.), Catalunya Carolíngia IV: els comtats d’Osona i Manresa, Memòries de la Secció Històrico-arqueològica LIII (Barcelona 1999), 3 vols, doc. no. 1380, which is also a rule-breaker in so far as it’s the only private transaction I know that still threatens its infringers with a portion in Hell with Judas. Cf. Tock, Scribes, souscripteurs et témoins, pp. 194-197: such concerns do not apparently manifest themselves in the French documents.

10. See n. 8 above.

11. For example, B.-M. Tock, “Auteur ou impétrant ? Réflexions sur les chartes des évêques d’Arras au XII^e siècle” in Bibliothèque de l’École des Chartes Vol. 149 (Paris 1991), pp. 215-248, to name but one of many.

12. Albert Bruckner, Die Vorakte der älteren St. Galler Urkunden, Urkundenbuch der Abtei Sanct Gallen Ergänzungsheft 1 (St. Gallen 1931); see now Bernhard Zeller, “Writing Charters as a Public Activity: The Example of the Carolingian Charters of St Gall” in Marco Mostert & Paul Barnwell (edd.), Medieval Legal Process: physical, spoken and written performance in the Middle Ages, Utrecht Studies in Medieval Literacy 22 (Turhnout 2011), pp. 27-37.