Comité International des Sciences Historiques – International Committee of Historical Sciences XXIIème Congrès du CISH – XXIInd Congress of ICHS Jinan (China) 23-29 August /23-29 août 2015
Thème spécialisé 1 – Specialised Theme 1
The History of Writing Practices and Scribal Culture Wednesday 26 August 2015, 9 AM-12:15 PM/9 h-12 h 15
ANTONELLA GHIGNOLI (La Sapienza –University of Rome, Italy)
Scripts and Signs in Documents of the Early Medieval Europe: Origins, Transmission, Functions
The late Roman world transmitted to the élites of the 'new' barbarian kingdoms of the West a written language, Latin, and a set of alphabetical scripts as a medium both to fix narratives texts and to communicate their mutual political and economic relationships on documents on papyrus, and later on parchment. But an important legacy of the Late Roman Antiquity to the rarefied written world of the early medieval Europe was also a set of graphic signs and the practice of shorthand (the proper Tironiana script and various systems based on syllabic notes). The paper aims at pointing out some cases, which represent further aspects of a dynamic survival of 'Roman' written forms, seen in the evidences of legal documents as well practical texts like private letters: the writing process in this field, unlike in that of the manuscripted books, involved people of different social and professional status, literates and illiterates too.
In European history from 400 to 1000 «the first break, and the most momentous, remains the break-up of the western Roman empire. (...) Reactions to the old moralistic reading of the 'end' of ancient civilization have often in recent decades sought to stress continuities across the fifth century, particularly in cultural and religious practices, and partly in political aspiration too; these continuities were real. The old image of the sweeping away of Roman culture by vital Germanic barbarism (succeded by Roman- German 'fusion' under the aegis of Catholic churchmen) is irretrievably outdated as a result. But this does not mean that the fifth century in the West was not a major period of change»1. During that century the fiscal and economical basis of the Roman state
1 Ch. Wickham, The Inheritance of Rome. A History of Europe from 400 to 1000, London, 2009, p. 553.
1
changed indeed. The result was that the economic unity of the western Mediterranean was broken, the 'barbarian' aristocracies at the guide of the post-Roman kingdoms became localized and poorer, and the material culture much simpler in most places. The book of Chris Wickham – from which I have cited the passage above – is a brilliant example of history writing, in which Late Antiquity and Early Middle Ages are conceived together on the same continuum. The historical relationship between the 'end' of the Roman empire and the beginning of the Early Middle Ages represents one of the broadest and most pervasive themes ever discussed in western historiography: it is significant that a panel at this congress is itself is devoted to «Late Antiquity in Contemporary Debate» as a specialized theme2. Whatever the interpretations of the transition from late Antiquity to early Middle Ages in the West may be – break, trasformation or continuity –, at least two facts are incontrovertible: 1) the basis of the government and administration of the late Roman empire was founded on a greek-latin written koiné, both linguistic and graphic3; 2) the 'barbarian' aristocracies of the post-Roman kingdoms did not ignore the written medium as a tool both of government and of economic relations: They used it in much more simplifed and fragmented way, as far as we know from the rare survival of documents of that period. The scripts (book scripts and cursive) the 'Barbarians' used were somehow 'Roman', and the 'written' language they were capable to read – or to hear and understand – was Latin. Regarding scripts and handwritings, palaeographers have long since had a quite clear sense of the inheritance from late Rome4. Regarding language, however, our understanding of written early medieval Latin, in books as well in documents, is now
2 Specialised Theme 16 (Organizer Rita Lizzi Testa, Unversity of Perugia, with the support of the Italian National Committee): see the official Program of the 22th International Congress of Historical Sciences, p. 40. 3 For the greek-latin graphic koiné see M. Norsa, Analogie e coincidenze tra scritture greche e latine nei papiri, in Miscellanea G. Mercati, VI, Città del Vaticano 1946 pp. 105-121 (= Omaggio a Medea Norsa, a cura di M. Capasso, Napoli, 1993, 137-156; G. Cavallo, La ϰοινή scrittoria greco-romana nella prassi documentale di età bizantina, in «Jahrbuch der Österreichischen Byzantinistik», 19 (1970), pp. 1-31; Id., Ecriture grecque et écriture latine en situation de "multigrafismo assoluto", in L'écriture: le cerveau, l'oeil et la main, a cura di C. Sirat, J. Irigoin ed E. Poulle, Turnhout 1990 ("Bibliologia", 10), pp. 349-62. 4 Among the handbooks available is still a valid guide: B. Bischoff, Latin palaeography. Antiquity and Middle Ages, Cambridge 1990.
2
considerably altered by the relatively recent perspective of socio-linguistic investigations. Such scholars as Michael Richter and Michel Banniard, for example, have retarded considerably the emergence of Romance languages perceived as different from Latin by hearers and speakers5. For the period between late Roman antiquity and early Middle Ages legal documents offer beyond doubt an very interesting field of investigation, even though (or perhaps, because) it is a problematic. Documents are both subjects of historical investigation and sources themselves. On one hand, the textual structures of the legal documents preserved in the western post-Roman kingdoms from the sixth to the eighth centuries demonstrate an undoubted derivation from those of the late Roman documentary practice6. But while the former are produced in a latin written world, become poorer and fragmented in any kind of its human relationship (political, economic and religious), the latter had been part of a very large greek-latin koinè of writing practices within an empire. The question of the sources is not irrelevant either. The extremely rarefied status of sources available for the late Roman period makes it impossible to have a real idea of the 'late Roman document', especially in the western part of the empire7, so that the late
5 M. Richter, The Formation of the Medieval West: Studies in the Oral Culture of the Barbarians, Dublin, 1994; M. Banniard, Viva voce. Communication écrite et communication orale du IVe au IXe siècle en Occident latin, Paris, 1992. 6 P. Classen, Kaiserreskript und Königsurkunden, Thessaloniki, 1977; Id., Fortleben und Wandel spätrömischen Urkundenwesens im Frühmittelalter, in Recht und Schrift im Mittelalter, ed. P. Classen, Sigmaringen, 1977. For the 'private charters', a recent overview in A. Ghignoli, Koinè, influenze, importazioni transalpine nella documentazione ‘privata’ dei secoli VII-VIII: lo stato dell’arte, in Le Alpi porta d’Europa. Scritture, uomini, idee da Giustiniano al Barbarossa. Atti del Convegno internazionale di studio dell’Associazione Italiana dei Paleografi e Diplomatisti. Cividale del Friuli (5-7 ottobre 2006), a cura di L. Pani e C. Scalon, Spoleto 2009, pp. 83-110. 7 Available sources are: Egyptian papyri (from 500 to the late eight century), Italian papyri (from Ravenna, from sixth to early seventh century), the wooden tablets from Vandalic Africa (Tebessa, fifth century). For editions and classification of all the papyri, see the portal "Papyri.info" –
3
Roman 'archtype' of the post-Roman 'barbarians' documents stands remote and – to say in terms of the philology – the genealogical relationship between 'archetype' and its later transmission remains for many aspects obscure. It goes without saying, that the situation is the same, perhaps even worse, as regards the western post-Roman sources and documents8. The early medieval charter of the latin West does not consist of written text only: It is a complex system of written and graphic signs, as well as material ones, like for example seals: all these are elements of different nature, employed together as codes in a same entire communicative process. The new approach in studying the early medieval charters as such a complex system began in Germany with the leading works of Peter Rück and his school9. That means that diplomatics – the discipline traditionally devoted to the study of medieval documents – must involve concepts and results taken from other disciplines – archeology, numismatics, semiotics, antropology for examples – in explicating morphology, semantics, syntax, pragmatic function and changes over time of the elements of this system10. The main non-alphabetical signs as graphic symbols in early medieval charters are: cross (in a variety of forms), monogramms containing personal names (first of all, the name of Christ) and/or other messages, and certain signs with complex structure11. Comparative studies in the field of "diplomatic semiotics" have gained considerable results in understanding their morphology, though synchronic and diachronic analyses
Dioscore d’Aphrodité cent ans après leur découverte. Histoire et culture dans l'Égypte byzantine, ed. J.-L. Fournet, Paris, 2008. All the papyri in latin language are published in facsimile in Chartae Latinae Antiquiores. Facsimile-Edition of the Latin Charters prior to the Ninth Century, ed. by A. Bruckner-R. Marichal, Dietikon-Zürich, 1954-1998 (this fundamental series will be abbreviated here as ChLA). For the Vandalic tablets: Tablettes Albertini. Actes prive´s de l’e´poque vandale (fin du Ve siècle), ed. C. Courtois et al., Paris, 1952. 8 Documents in decent quantity and in original transmission are available for Frankish Gaul (from seventh century), Lombard Italy (from ealry eighth century), and Spain (rare fragments on parchment and documents on slate, from late sixth century). Not one document in original is preserved from Rome and from the papal curia until the late eight century. Facsimile-edition of all these sources, in ChLA. For the documents on slate: Documentos de época visigoda escritos en pizarra (siglos VI–VIII), ed. I. Velázquez Soriano, Turnhout, 2000. 9 See P. Worm, Ein neues Bild von der Urkunde: Peter Rück und seine Schüler, in «Archiv für Diplomatik», 52 (2006), pp. 335-370. 10 Graphische Symbole in mittelalterlichen Urkunden, ed. von P. Rück, Sigmaringen, 1996. 11 An overview in P. Rück, Beiträge zur diplomatischen Semiotik, in Graphische Symbole cit. (see note 8), pp. 13-47.
4
in this field have been conducted only on post-Roman western sources, substantially leaving out the late Roman sources. But I think it is worth trying to look at Roman late antiquity through the perspective of the presence and meaning of graphic signs drawn within the text of legal document as well private letters, and exploring the possibility to individuate further traces of the relationship between late antiquity and early middle ages. In this paper I will point out only few selected cases, which, I think, can show us further aspects of a dynamic survival of 'Roman' writing practices. In such texts as those of practical use transmitted by original documents and letters, unlike in those of manuscripted books, were involved people of different social and professional status, literates and illiterates too.
1. Sign of the cross and related symbols in the late Roman world: writing practices of daily life
In the third essay devoted to the history of the sign of cross Franz Joseph Dölger has demonstrated that from the end of the fourth century at least the greek cross (+) and the diagonal cross (×) – in the shape of the so called 'saltire' – were employed as a christian symbol in written texts12. Different functions can be observed according to the context, but the meaning was always the same: the "Name of Christ". The connection between sign and meaning is resolved in "Christ" as a personal name, and it can be said that it is a sort of 'nominal' connection. This is due also to the fact that on the one hand the diagonal cross had the same shape of the greek letter X (Chi), the initial of the name Χριστὸς; on the other, that the Greek cross (+) was used to indicate the greek letter X itself. The first evidence of a greek cross employed as christian sign in documentary texts seems to be in P. Oxy. LVI 3862, a letter of recommandation written by Philoxenus to his parents and uncle, of course in greek, dating to the fourth-fifth
12 J. F. Dölger, Beiträge zur Geschichte der Kreuzzeichens. III, in «Jahrbuch für Antike und Christentum», 3 (1960), pp. 5-16; a series of nine essays on the subject was published in this review from 1958 to 1967.
5
centuries: It is drawn in central position over the first line of the text. The presence of crosses in christian letters from the late antique Oxyrinchus increases during the fifth century and became general in the sixth and seventh centuries13. Its function is clear: it serves as a invocation to Christ, the very beginning of everything, but at the same time it works also as a special greeting blessing to whom would be reading the letter, the addressee. A cross in the so called 'latin' shape – or crux ordinaria, where the vertical stroke is longer than the horizontal one – is not attested in the papyri from the late Roman Egypt. The christian letters on papyrus of Oxyrinchus attest for the first time another important symbol related to the cross: the staurogramm. It appears sometimes marked with greek crosses but sometimes it is drawn alone. It is a monogramm of two greek letters, rho and tau, a sort of abbreviation for the greek nomen sacrum of the cross – stauros – and consequently of the crucified, Christ14. The staurogram is usually drawn in the left margin at either the beginning or the end of the letter. Its symbolic function is clearly evident and it is also confirmed by its position in the text15. The chi-rho monogramm is much less attested than staurogramm in these late antique sources16. It is important to stress that staurogramm and christogramm are two distinct signs, with a clear different morphology and that the staurogramm has developed indipendently within the christian tradition without any original connection with the Egyptian ankh (the sign meaning "life")17, which is currently called crux ansata. The
13 See L. H. Blumell, Lettered Christians. Christians, Letters, and Late Antique Oxyrhynchus, Leiden - Boston, 2012, pp. 43-46 and Appendix, Table 4. 14 See L. W. Hurtado, The Staurogram in Early Christian Manuscripts: The Eearliest Visual Rreference To The Crucified Jesus?, in New Testament Manuscripts: Their Text and Their World, ed. T. J. Kraus and T. Nicklas, Leiden, 2006, pp. 207-26. 15 For examples of 'practical' functions of the cross, as a non-christian symbol, in documentary texts, see Blumell Lettered Christian cit. (see note 11), p. 43, note 75. 16 Probably for the contextual secular usage of the sign as abbreviation: «(...) the chi-rho sign which Constantine placed at the head of his army can also befound in non-Christian papyrological contexts as an abbreviation. Secular use is frequently transparent as such, intergrated into an account or similar text, but care must still be taken; these secular resolutions were by no means totally subsumed by the Christian usage in the course of the fourth century. While the more ʻ ornamentalʼ function of the Christian chi-rho stands out against them, where nothing immediately suggests a Christian context, other possibilities may be preferable»: M. Choat, Belief and Cult in Fourth-Century Papyri, Studia Antiqua Australiensia I, 2006, p. 118. 17 See Hurtado, The Staurogram in Early Christian Manuscripts cit. in note 14.
6
best way to avoid the confusion about the identification of staurograms and their distinction from christograms is to not to use such terms as crux ansata or Enkelkreuz, whereas some scholars employ them to describe what actually is a staurogramm18. If it is true that the ancient roman epistola is the archetype of the documents issued by the 'german' rulers in the early middle ages19, then it cannot be stated that the symbolic invocation at the beginning of charters represent an innovation of the early medieval documentary tradition20. The practices of writing letters attested in the papyri from the late antique Oxyrinchus demonstrate the contrary. An extraordinary example of the link between late antiquity and early middle ages is provided in the most ancient preserved document issued by the visigothic king Reccaredo (586-601). The text is in latin, written on slate in a cursive roman minuscule – the so called 'new roman cursive' – and it is structured, of course, as a 'letter'. It opens – as expected – with a christian symbol of invocation: a staurogram21.
2. The sign of cross as writing practice of illiterates
Christograms and staurograms may appear signs of a simple shape; but they are not. Their final schape is the result of writing two greek alfabetical signs (respectively chi-rho and tau-rho) in monogrammatic composition. Certainly, one could have learned to imitate the whole figure; In any case, it would have involved holding the calamus (the 'pen' of that time) and managing it in order to trace a complex sign over the surface of a piece of papyrus, parchment or a wooden tablet: A manual skill that an illiterate could hardly have. A different matter is being able to draw single lines and circles or drawing simple crosses. Let us begin therefore with the most basic, though deeply significant, sign among the symbols meaning the name of Christ: the cross.
18 For example in E. Eisenlohr, Von ligierten zu symbolischen Invokations- und Rekognitionszeichen in frühmittelalterlichen Urkunden, in Graphische Symbole cit. (see note 8), p. 183. 19 O. Redlich, Die Privaturkunden des Mittelalters, München-Berlin, 1911, p. 5. 20 It is so, for example, in B.-M. Tock, Scribes, souscripteurs et témoins dans les actes privés en France (VIIe-débeut XIIe siècle), Turnhout, 2005, p. 147. 21 A reproduction also in Eisenlohr Von ligierten cit. (see note 15), p. 207, Fig. 1.
7
A person who was not able to write a signing line to a legal document – a so- called subscriptio22 – could however participate with his own hand to the documentary process, as author of the written contract or as witness, by drawing a cross in the empty space left for it in the line written by the scribe or by a third delegated literate person, in the following way, more or less: «Signum (sign) – empty space for the autograph cross – NN (name, title, role of the signer and sometimes with a comment that he is not able to write this line)». Consequently to what we have already mentioned above, the greek cross or the diagonal cross are the only two shapes attested as 'autograph' subscription from illiterate people in the sources from the fifth to the seventh centuries. The Tablettes Albertini, a set of estate documents of the 490s from Tebessa, an economically marginal zone of southern Byzacena, are the only documents we have from Africa under the Vandals: They comprise forty-five cedar-wood tablets, written in ink, containing thirty-four documents. Among twenty-eight signs of illiterates, eighteen are crosses (eleven are diagonal crosses, seven are greek crosses traced in very cursive manner). The remaining signs are little circles or a single short wavy line23. Although signs other than a cross are present in these sources, there is not doubt that the traced crosses were christian signs: particularly significant is the similarity of the shape of the traced diagonal cross the to the greek letter chi – χ –, initial of χριστός. Illiterate signers to greek private documents on papyrus from the early Byzantine Egypt seem to prefer drawing simple greek crosses.24 The same practice is attested in almost the same period as the Italian papyri, with their latin private documents from the Ostrogothic and then Byzantine Italy (from the fifth to the seventh
22 See the definition and related terminology in Vocabulaire international de la diplomatique, ed. M. Milagros Cárcel Ortí, Valencia 1997, p. 66, Nr. 254: «Les souscriptions (lat.: subscriptiones) sont les formules par lesquelles les partie, les témoins de l'acte juridique ou de l'acte écrit, le scribe, marquent la part qu'ils ont prise à cet acte et manifestent leur volonté personelle, leur consentement ou leur presence (...) All.: Subsktiptionen/Unterschriften; – angl. subscriptions (autograph ou non-autograph); – esp. suscripciones; – ital.; sottoscrizioni». For the structure of the 'subscriptions' in this period see L. Saupe, Die Unterfertigung der lateinischen Urkunden aus den Nachfolgestaaten des Weströmischen Reiches, Kallmünz, 1983 (Münchener Historische Studien. Abteilung Geschichtl. Hilfswissenschaften, 20); Id., Unterfertigung mit Handzeichen auf Urkunden der Nachfolgestaatendes Weströmischen Reiches bis zur Mitte des 8. Jahrhunderts, in Graphische Symbole cit., pp. 99-105. 23 See Tablettes Albertini cit. (see note 5), 1, p. 59. 24 J. A. M. Sonderkamp, Die byzantinische Privaturkunde, in Graphische Symbole cit. (see note 8), p. 112.
8
centuries)25. The evidences that we have from the poor relics of the documentary practice in Frankish Gaul (from sixth-seventh century) and Lombard Italy (from eighth century) show us again illiterates drawing a cross before their respective signing line (or within it) written by the notary or by the scribe of the document. The cross appears traced in one of the two typical forms inherited from the late antiquity: the greek cross or the diagonal one (the 'saltire' seems to be used only in the merovingian documents)26. At least in one Frankish charter the greek cross written by illiterate signers appears no more as such as, the ends of its two arms are bent at nearly right angle: Its form is that of a hooked cross27. In terms of writing practices by illiterates, the sources from fifth to eigth century do not show elements of discontinuity. Hence two considerations could be set out. For an illiterate drawing the cross was the only possibility of taking active part in the writing process of legal documents: given the use by literate of putting a christian symbol at the beginning of the autograph subscription – as we will see in the next chapter –, one could most easily conclude from it that the 'illiterate cross' – if we can say so – was in some way a sort of «"Unterschrift" in abgekürzter, rudimentärer Form»28. But the cross was a potent symbol and it never lost all the intensity of its simbolic meaning: abuses in that sacred name, traced by own hand, surely must have seemed inconceivable. In any case, the practice of drawing a cross by illiterates in the post-Roman period involves – it must be remembered – a fundamental assumption: that an illiterate Frank, Goth or Lombard could understand – still in the seventh century, at least – the written language of the document, which the notary used to read aloud again, at the end of the writing process, to all participants (author and witnesses), according to
25 See for example the six autograph crosses of the six illiterate sellers in ChLA, XX, n. 704 (Ravenna 551), lines 130-135. 26 See, for example ChLA XIII, n. 564, crosses at the lines 36 and 38 (Lamorlaye, 673 march 10th). 27 ChLA XIII, n. 558 (Clichy, 654 june 22th): cross of the vir inluster Auigulfo comes palatii and of the vir inluster Probato. 28 Sonderkamp, Die byzantinische cit. (see note 20), p. 112.
9
the expression "relegi ei chartam" contained in the document itself: that written language was latin29.
3. In the name of Christ: crosses and related signs as initial marks in autograph subscriptions
In the late roman antiquity the use of greek crosses and staurograms as symbolic invocations to Christ, drawn before writing the text and sometimes after its conclusion, is not attested exclusively as a feature of the practice of letter writing. The same use connected with a same original function –it would be reasonable to presume it – can be observed also in the subscriptions to official documents by the authorities in the name of which the document was issued. The late roman subscription traced by a high government bureaucrat at the end of an official document issued in his own name has the well known form of either a greeting, bene vale, or that of a statement, legi / legimus. The subscription was generally autograph, written in first person, in latin language and latin alphabet even though the document was written in greek language and greek script, and the author never expressed in it his own name. This genre of texts survives in a very few examples, all from the fifth century. In a short series of papyri containing orders of the praeses Thebaidos, the praeses has drawn before his legi or legimus-statement a staurogramm, sometimes the sign is traced also at the end of the line. Also the so called R barré – abbreviation meaning probably R(ecognovi), written by the scribe or by a secretary – has a greek cross traced over the letter R30. The well known Papyrus Butini, a fragment of charter from the chancery of a comes sacri stabuli, datable to the sixth century – famous, because written in that typical cursive of the roman provincial chanceries which palaeographers consider the
29 Important suggestions in M. Banniard, Niveaux de compétence langagière chez les élites carolingiennes: du latin quotidien au latin d'apparat, in La culture du haut Moyen Âge: une question d'élites?, sous la direction de F. Bougard, R. Le Jan, R. MckKtterick, Turnhout 2009, pp. 39-61. 30 ChLA, X, n. 464; ChLA, XLI, nn. 1192, 1195, 1196.
10
direct ancestor of the chancery scripts employed by Merovingian kings in Frankish Gaul – has the autograph subscription of the comes himself in the form of final greeting in latin, preceded by a small greek cross: + bene | uale31. At the end of the text there is also the subscription of the scribe himself, written in the form of bene vale. Althoug in that point the papyrus is damaged, the rest of something traced before the beginning of that subscription is still visibile: Jean Mallon considered it as the rest of the letter g and presumed the presence of the word signum 32; It seems to me most likely to be simply the remaining part of the long vertical stroke of a cross. In the case of subscription in form of a greeting in latin language – bene uale – the literal meaning of the sign of cross (the name of Christ) seems evidently to be headed towards the addressee, the person who shall have read it (and, of course, the document too): In nomine Christi bene uale ("In the name of Christ be well"). In the case of subscriptions in the form of legi / legimus the function of the christian sign (staurogram) written at the beginning of the phrase seems to be quite different. This time the symbolic message concerns evidently the statement legi itself, that is the action made by the author of reading the document and approving the text; the invocated name of Christ makes sure that the action of legere has been performed in full faith – In the name of Christ, the crucified, I have read and approved the text – and, at the same time, validates the truth of the statement itself. The same can be said also about the cross traced over the abbreviated word R(ecognovi) – "I have recognized – expressed by the secretary. An extraordinary case offers the P. Iand. Inv. 61; this small piece of papyrus can be dated around the year 500 and it was probably a card accompanying a gift33. The text is a statement with the name of the author expressed – Fl(auius) Symeonius cornicularius obtul(it) – written between two staurograms. Here too the function of those symbolic religious signs is to provide with truth the identity of Flavius Symeonius and what he declares, though in third person: obtulit, 'he offered a gift'. Remarkable is
31 See the facsimile in ChLA, I, n. 5. 32 J. Mallon, Le Papyrus Butini, «Bibliothèque d'Humanisme et Renaissance», 14 (1952), pp. 283-288 (reprint in De l'ecriture. Recueil d'études publiées de 1937 a 1981, Paris, 1982, pp. 216-219). 33 See J. O. Tjäder, Papyrus Iandana 68b. Eine paläographische Studie, Kurzberichte aus den Giessener Papyrussammlungen; 25 (1977). Facsmile in ChLA, XI, n. 490.
11
the shape of the two staurograms: The vertical stroke of the letter rho is thin long line and the horizontal one (that is the part of the monogramm which represents the letter tau) is not drawn as a straight line (as it should be) but as an elegant wavy line. It probably means that the practice of writing such personal declarations could make it possible to elaborate, and in some personal way, the consolidated and traditional structure of the staurogram. The archive of the archbishop of Ravenna furnishes us with a considerable large collection of private documents on papyrus dated to the Ostrogothic Kingdom – usually seen as Italy’s last period of Roman-style stability (490-554) – and to the later Byzantine period in Italy until the seventh century. Thus, the so called Italian Papyri may represent a sort of bridge between late antiquity and early middle ages, between the eastern Roman Empire and the Lombard Italy, of course in a relatively sense. All the autograph subscriptions attested in the Italian papyri – written by the authors of the contracts or by the witnesses or by the tabelliones, who were responsible in writing the text of the document – present an initial sign of invocation, sometimes in the form of the greek cross but in a larger number in the form of the staurogram. The function of these signs is the same of the staurogram written before the legi-subscription attested in the documents of the provincial chanceries of the late Roman state. But the most interesting feature that can be observed in these subscriptions of literates is the morphology of the sign of staurogram: Its structure shows in many cases variations, which interest the transverse stroke exactly analogous to that we have already observed above in P. Iand. Inv. 6134: The elaboration of that Flavius Symeonius could be seen then as a first phase – a sort of 'degree 0' – of a future successful development. The practice we found in the greek private documents on papyrus from Egypt of the early Bizantine period is substantial similar to that which is attested in Italian Papyri, with the a not irrelevant presence of the staurogram as opening sign in subscriptions35.
34 See, for example, ChLA, XXI, n. 713 (Ravenna 572), lines 10-13. 35 Particularly in Oxyrhynchus: M. Diethart – K. A. Worp, Notarsunterschriften im byzantinischen Ägypten, Wien, 1986, I, p. 13.
12
What emerges from the written documentation survived from the Merovingian Gaul, Lombard Italy and Visigothic Spain after the 600 is that literates used to begin their own subscriptions drawing a cross or, more frequently, a staurogram: it is the same we have seen in the latin private documents attested both in the Italian papyri of the sixth century and in the greek documents preserved in the Egyptian papyri. It is evidently a further, crucial element in the 'barbarian' inheritance from the writing practices of the late Roman world. The particular morphology of staurograms and its developments in the italian papyri seems to me to be a valid reason to reconsider in some way the origin and formation of the signs of invocation drawn in the subscriptions of visigothic documents and of those present in the Lombard charters as well, where scholars have seen them as indipendent developments within the early medieval tradition, without any connection with practices of the late Roman antiquity 36. An exception to this late Roman and early 'Barbarian' koinè seems to be the one attested for the Romans under the Vandals in Africa: the few signing lines of literates that have survived on the wooden tablets from Tebessa are conceived without any sign of invocation: Ego NN subscripsi ("I – name – have undersigned"). And finally, it is a proven fact that there is a lignée between the official document of the provincial chanceries of the late roman Empire and the document of the Merovingian kings37. Therefore there is absolute evidence of consonance between the subscription written by comes sacri stabuli – in the form of «+ bene uale» – in the P. Butini and the analogous «ben(e) ual(iat)» employed as «Siegelmarke» in the documents issued by the Merovingian kings 38 at the end of the charters of the Merovingians kings and it can be considered a further little piece of a well known puzzle.
4. Writing in a different way: shorthand and other graphic signs as ending marks in autograph subscriptions
36 So it is in Eisenlohr Von ligierten cit. (see note 15), pp. 201-206. 37 Classen, Kaiserreskript (see note 4). 38 See, for example: Die Urkunden der Merowinger, hg. v. Th. KÖLZER (MGH Diplomata regum Francorum e stirpe Merovingica), Hannover 2001, Bd. 1, Nr. 136 (692/693 November 1, Luzarches), p. 346.
13
The constant presence of the cross and staurogram as an initial mark in subscriptions (with the sole exception we have seen in Vandal Africa) can be considered one of the structuring elements of a wide koinè generated among the writing practices in the late Roman empire, that emerges still vital and alive in the post-Roman West. Not rare are the cases in which the text of an autograph subscription does not end with the last word the signer had written, whatever was the alphabet he had used: greek, latin or gothic.39 The signer had the possibility to continue writing his name (or very little else) for a second time, but this time in shorthand40. They could also certainly communicate something else, which however remains almost invariably obscure to us: This second message was, in fact, 'written' through tachigraphic notes, crosses, signs and a variety of lines drawn as interlocking pieces of a complex structure, which is impossible to 'read' and to understand. Although the ensemble ends up having a very characteristic shape, like a complex special sign. If the first 'block' (the initial sign of invocation, whether cross or staurogram) and the second one (the text of subscription itself) are always present in all subscriptions attested in our sources, not all the literate signers were able to realize that third 'block' consisting in a microtext written in shorthand (that is a system of codes completely different than that of greek and latin alphabet) and/or in a 'complex special sign'. But more significant than an occasional absence is the fact that the final block emerges, in the late Roman period, both in the greek private documents on papyrus of the Byzantine Egypt (in the notarial subscription) and in the latin italian papyri of the Ostrogothic and Byzantine Italy; and as well they are seen in the post-Roman period, both in the Merovingian documents and, more rarely, in the Lombard ones.
39 For subscriptions written in latin language but in greek alphabet, an example in ChLA, XXI, n. 714 (Ravenna, 575 febr. 25th); for subscriptions in gothic language and gothic alphabet, an example is ChLA XX, n. 704 (Ravenna 551). For subscriptions in greek language written partially in latin script, see Diethart–Worp Notarsunterschriften (see note 30), p. 14. 40 For greek shorthand: H. Boge, Griechische Tachygraphie und Tironische Noten, Hildesheim- New York, 1974. For latin shorthand: D. Ganz, On the History of Tironian Notes, in Tironische Noten, ed. P. Ganz, Wiesbaden 1990, pp. 35-51. The best survay on the Tironiana is M. Hellmann, Tironische Noten in der Karolingerzeit am Beispiel eines Persius-Kommentars aus der Schule von Tours, Hannover, 2000, pp. 1-98.
14
Absolutely remarkable is the structural similarity of the final complex sign in greek subscription from Egypt of the sixth century, in the latin subscriptions from Ravenna of the same period, in the latin ones from Frankish Gaul of the seventh; and with an isolated case, and therefore much more interesting, from Lombard Italy of the half of eighth century41. Two aspects then must be stressed here. First, as regards the graphic aspect, we see that substantially nothing differentiate the subscription of a notary from that of a literate signer involved in some way in the document: At least the Italian papyri show that clearly. The second aspect regards the use of tachygraphic notes, though I shall look here neither at the problem of their survival in Merovingian Gaul an Lombard Italy, nor at the huge problem of the survival of Tironian notes and their revival in Carolingian period: Writing in shorthand must be thought, at least in the fifth-seventh century, as a widespread ability (both in Ostrogothic and Byzantine Italy and Byzantine Egypt), not exclusively associated to notarii and exceptores. The Italian papyri attested that literate merchants and people of the middle class – like the viri honesti – had at least a basic knowledge of shorthand, so that they could employ it to mark their subscriptiones 42: It could be a trace to assume that a shorthand based on a simplified vocabulary was perhaps employed in business and commerce, as far as they could exist in Ravenna in the sixth and seventh century. Regarding the function of all that might be contained in this 'third' block of the ideal structure of the subscription, it can be guessed that it was added in order to make it easier to identify and recognise the subscription as autograph by the singner himself (and not by others) in case of a trial. 'Writing' the third 'block' was not compulsory: it is evident. Therefore after having finished the text of the subscription, there became, for the signer, a place available on the line, in which he could give his own view of himself,
41 For examples from Egypt, Ravenna and Gaul, see respectively: Diethart–Worp Notarsunterschriften (see note 30), 1, p. 50, n. 21.3.1, Taf. 15 (Arsinoites 663); ChLA XXI, n. 714 (Ravenna 575); ChLA, XIII, n. 558 (Clichy, 654 june 22th). The unique Lombard evidence is in the subscription of a bishop of Pisa (ChLA, XXVI, n. 803: Pisa 748), recently discussed in A. Ghignoli, Su due famosi documenti pisani dell'VIII secolo, «Bullettino dell’Istituto storico italiano per il Medio Evo», 106/2 (2004), pp. 25-27, pp. 25-27. 42 But Ganz, On the History (see note 40), p. 37 argued that only viri clarissimi and notaries of Ravenna copied their names in syllabic notes.
15
of course by writing even though in not ordinary way indeed. Identifying and understanding the graphic codes employed, therefore, would mean the possibility to link that not ordinary way of 'writing' to a social or cultural context.
5. The greek ΧΜΓ as origin of early medieval graphic swirls in latin documents?
The cryptogram ΧΜΓ – which could be considered an isopsephism too43 – has been the subject of many investigations since at least the late ninenteenth century. It appears widespread all over the Roman world from the fourth to the seventh centuries (in Egypt, in the arabic period), in a large variety of contexts, written almost all in greek language and greek script: christian letters, short texts as amulets, inscriptions, private documents, often in association with a cross or with a staurogram or christogram, or with tha isopsephism composed with the greek letters koppa and theta, whose numerical equivalent was 99 and stands for αµήν (amen). Scholars still debate over the origin and exact meaning of ΧΜΓ in certain contexts44. The three main interpretations argued by scholars are: Χ(ριστὸν) Μ(αρία) Γ(εννᾷ) – "Mary begat Christ" – ; Χριστὸς Μιχαὴλ Γαβριήλ – "Christ, Michael, Gabriel" – χ(ειρός) µ(ου) γ(ραφή) – "written with my own hand". Though there is general agreement that it was a Christian catchword or marker. Therefore it became not imperative in the investigation that only one rendering be preferred. That makes however quite weak the third possibility of rendering – "written with my own hand" –, proposed because of the presence of the group in some legal documents. Here indeed it stands often at the beginning of the contract, near the datation, where also an invocation formula, an expression of devotion and religiosity would be acceptable and appropriate.
43 Its numerical equivalent is 643 and could be interpreted as θεὸϲ βοηθόϲ ,"God is help". 44 J.-O. Tjäder, Christ, Our Lord, Born of the Virgin Mary (ΧΜΓ and VDN), «Eranos» 68 (1970), pp. 148-190; A. Blanchard, Sur quelques interpretations de ΧΜΓ, in Proceedings of th 14th Intern. Congress of Papyrologists, London 1975, pp. 19-24; A. Gostoli, Una nuova ipotesi interpretativa della sigla cristiana ΧΜΓ, «Studia Papyrologica» 22 (1983), pp. 9–14; T. Derda, Some Remarks on the Christian Symbol ΧΜΓ, «Journal of Juristic Papyrology», 22 (1992) 21– 27; A. Di Bitonto Kasser, Un nuova attestazione di χριστου µαρια γεννα, «Aegyptus» 78 (1998).
16
But the cryptogram is not an exclusively matter of greek papyrologists. In three latin legal documents on papyrus from Ravenna dating to the sixth-seventh centuries we find ΧΜΓ (more precisely χµγ, because it is written in late roman cursive, that is in a minuscule script), and therefore this criptogramm becomes interesting to us too. It seems likely that this evidence can be considered a matter of the importation of Byzantine scribal practices; but it is unlikely to me «that those who wrote, even though they were not aware of the phrase χ(ειρός) µ(ου) γ(ραφή), nevertheless believed they were expressing the idea 'written with my own hand"»45: This interpretation has not sufficient arguments to be preferred to the view that the group of greek letters was seen as a graphic sign in itself (not one attestation in the whole tradition shows the group marked by a sign of abbreviation). Given the meaning which the cryptogramm had in the contemporary sources of the greek part of the late Roman world, it could be seen as a 'sign' of blessing or an auspicious sign with apotropaic value. Let us see quickly these three latin documents. In P. Ital. 3046, a contract of sale dated to the year 539, χµγ stands, written by the notary, at the end of the text: after the datation – «[Ac]tum diae et quinquiaes p(ost) c(onsulatum) s(upra)s(cripti). Ravennae. χµγ» and before the beginning of the long series of subscriptions. In the same position it occurs in the much later P. Ital. 2547, which a fragment of a donation and long-term emphyteusis lease datable to seventh century. But here it is preceded by a staurogram: «Actum Ravenna, imperio, anno, die et ind(ictione) s(upra)s(crip)ta. (Staurogram) χµγ». It is noteworthy that in the chartae of the Lombard Italy (and Merovingian Gaul too) the expression feliciter – the so called apprecatio formula – will stand exactly in the same position occupied here by the χµγ- group. The group is not however traced by notary in the context of the document, in the third and last case, P. Ital. 648, which is a contemporary, we can say, of P. Ital. 30. It is a will dated to the year 575. Here the χµγ-group appears written by a witness – a certain
45 Tjäder, Christ (see note 38), p. 171. 46 ChLA, XX, n. 706. 47 ChLA, XXVIII, n. 843. 48 ChLA, XXI, n. 714.
17
Quiriacus – at the end of his subscription, of course in latin. The nature of a closing graphic group is evident. But what is really worth considering is the paleographical aspect: In the later document – P. Ital. 25, where the group stands at the end of the text, after the dating – the notary evidently traced the χµγ without being aware that they were three (greek) letters, and in any case he draw it very cursiv, without lifting his calamus. The result is an uninterrupted chain, a series of connected lines and loops. It is surprising the similarity of this graphic group to some kind of so-called row of "swirls", which often mark typically the end of a text (and after the datation in alternative with feliciter) and/or the end of a subscription in latin documents of the early medieval period, particularly in Italy49. These rows of loops and lines are usually considered by scholars a graphic nonsense, or merely decorative signs with the function to avoid illegitimate addictions of text in that place of the document. Even though the conjecture is based only on this kind of palaegraphical observations, the following hypothesis could be formulated: the 'original' cryptogram χµγ – with all the complex of its religious and apotropaic meanings, collected in its long life and widespread existence in the greek written practice of the late Roman world – has been changing its original shape as "signifier" (meant in the sense of Saussure's linguistic terminology), but it did not change its "signified" in the latin scribal practice of the Byzantine Italy, and from there it was transmitted to the scribal practice of the early medieval Italy as sign with the same – at least in the very first period – original apotropaic value. After all, in many aspects (first of all in that regarding writing practices) the process of transformation of the ancient Roman world in the early middle ages knew, more or less, analogous implications.
49 An example among the Lombard charters in ChLA, XXVIII, n. 855 (Milan, 777 march 8th).
18