Numărul 2 / 2010
REMARKS ON THE RELATION BETWEEN THE BREVES NOTITIAE AND THE NOTITIA ARNONIS*
Dr. habil. Tamás Nótári**
Summary. Acting perhaps also as prosecutor in the lawsuit against Tasilo, and thus considered the winner of the Frankish Machtergreifung, Arn had the Notitia Arnonis written by Deacon Benedictus. It is possible that the Notitia Arnonis had been recorded as early as in 788; however, a point of time somewhat later cannot be ruled out either; the text left to us by Rotulus gives the year 790 (anno dcc lxxxx) as title, and only insertions from the 15th century added the entry on 788 to it (anno dcc lxxxviii congestum), accordingly, the dating from the 12th century, the earliest known to us, allows to make the year 790 probable. Based on that, in summary, it can be declared that the statement demanded by Charlemagne at the end of 788 (eodem anno, quo ipse Baioariam regionem ad opus suum recepit), and readily presented by Salzburg, partly relying on and borrowing from earlier records and partly containing new information, was created during the occupation of Bavaria by the Franks, i.e., between 788 and-as it can be deduced from the text left to us-790. Heinrich Wanderwitz dated the creation of the Breves Notitiae to the years between 798 and 814; we can narrow this interval by presuming-as it is justified by historical circumstances taken into consideration-that the immunity granted to Salzburg was connected and in time coincided with making Salzburg Archbishopric. It can be declared that Charlemagne issued the immunities granted by him mostly between 787 and 800. Based on that, we can specify 798 as the terminus post quem and 800 as terminus ante quem of the immunity granted to Salzburg; thus, again the period between 798 and 800 can be given as the time of the creation of the Breves Notitiae. The Breves Notitiae can be by no means qualified as a later revision of the Notitia Arnonis, because their structural composition is independent, and linguistically-in terms of grammar and style-they show more proper solutions than the Notitia Arnonis, which manifests the linguistic deterioration of the age of the Merowings. The highly Carolingian supporting form of presentation used at certain points in the Breves Notitiae is acknowledged and accepted among researches, to various extents though. This tendency is indicated by the occurrence of Pippin and his mother, Hiltrud in the text, who are not mentioned by the Notitia Arnonis at all, and beside whom Duke Tasilo is reduced to a kind of supporting character by the narrative. The Breves Notitiae refer to Tasilo on five occasions in total compared to the nine loci in the Notitia Arnonis-the shorter source in volume; and grammatically he is included in the text as the subject on one occasion only. In relation to the Notitia Arnonis Herwig Wolfram uses the qualification "(eindeutig) prokarolingisch"; whereas regarding the Notitia Arnonis Heinrich Wanderwitz speaks about the unambiguous sympathy for the Agilolfings shown in the source, except for the introduction and the closing chapter. We can agree with Fritz Lošek on the point, that neither in relation to the Notitia Arnonis nor to the Breves Notitiae is it righteous to make such unambiguous and lapidary statements: each Bavarian Duke is adjudged differently, the absolutely positive image of Theodo and Theodbert is followed by portraits depicted of Odilo and Tasilo far from being flattering, which is supported by the conflict outlined in the Libellus Virgilii too.
Keywords: early medieval charters, legal history of Bavaria, Bavarian historiography.
As the reason for creating the Notitia Arnonis we could specify Charlemagne having seized power over Bavaria, which took place through the dethronement of Tasilo III, the last Duke of the Agilolfing dynasty. Up to the 780's, we cannot speak about resolute anti-Carolingian attitude engaged by the Agilolfings in the Frankish/Bavarian relations; these relations were determined by the current political constellation, in several cases the members of the Agilolfing dynasty, such as Odilo and Tasilo, were able to take the throne of Bavaria and strengthen their power only with the help of the Franks. The Carolingian/Agilolfing relation, as a matter of fact, did not lack rivalry but it became fatal for the Bavarian Duke only in 788, after Tasilo III had been dethroned by Charlemagne. At the end of our research, in spite of being incomplete and tendentious the sources clearly revealed what processes had led to this final result and the development of this picture. The Frank ruler's power politics did not necessarily require military clashes; after he had defeated his opponents and enemies in order to absorb also Bavaria-which had preserved its independence as the only entity on the territory under the former Merowing rule-in its empire, it was sufficient for him to isolate the Dukedom by cunning diplomatic tools and win over a group of the Bavarians who would support him in the future lawsuit. In the lawsuit not only his former sins, specifically harisliz alleged to have been committed in 763, i.e., arbitrarily leaving the king's armies, were cast in Tasilo's teeth because de iure this would not have been enough for announcing the death sentence, he was also charged with grave infidelity-breach of the oath of allegiance he made in 757 and 781, and of the vassal's oath he made in 787-and with entering into an alliance with the Avars, which was considered the most serious manifestation of infidelitas. The execution of the death sentence, however, would not have brought the result desired by Charlemagne because through Tasilo's execution he could not have annexed Bavaria eo ipso; closing Tasilo-and his family members-into a monastery enabled the Frankish ruler to dispose freely over the Dukedom now having no ruler. The creation of the Breves Notitiae is closely linked to raising Salzburg to Archbishopric in 798, and it was this register by which Charlemagne's fiduciary, Arn tried to prove his primacy to the rest of Bavarian bishops. Arn all the more earned Charlemagne's trust because he had willingly implemented his instructions in the critical situation evolved around Pope Leo III in Rome. In our view, this hostility was incited by Charlemagne's people, Paschalis and Campulus-and an attempt was made at throwing the Pope off his throne-primarily in order to enable the Frankish ruler to act subsequently as the Pope's savior, and to cause the Pope, seeing his reputation left in tatters, being politically defenceless, to crown him emperor. As an active player Arn could follow up these events-as it can be clearly established from the correspondence maintained with Alcuin.
I. The problems of dating of the both notitiae
After throwing Tasilo III off his throne in 788 in the lawsuit held at the assembly at Ingelheim, Charlemagne integrated the until then independent Bavarian Dukedom into his empire. Several of the Bavarian (secular and church) dignitaries, including Arn, Bishop of Salzburg and later a good friend of Alcuin, most probably took sides with the Frank ruler as early as during the reign of Tasilo, and helped him to legitimate the dethronement of the prince. In the years directly following the dethronement, between 788 and 790, it was at the instruction of Charlemagne that the great winner of the Carolingian takeover, Bishop Arn (785-821), who used to enjoy Tasilo's confidence for a long time, caused to write the notice on the donations to the Bishopric of Salzburg, the Notitia Arnonis, which was approved by the Frankish ruler, to ensure the benefices of his diocese. The Bishopric of Salzburg was raised to archbishopric in 798, and it was at that time that Arn received the pallium from Pope Leo III; however, the bishops now subjected to him apparently felt aversion to this decision adopted by Charlemagne and implemented by the Pope. So the Archbishop had to prove the origin and legality of the estates obtained, which resulted in the work entitled Breves Notitiae drafted between 798 and 800. It can be established that in spite of the Breves Notitiae and the Notitia Arnonis overlapping each other in certain points, none of the records served as the prefiguration of the other, so both documents add specific data to our knowledge on the economic and estate conditions of the age.
The text itself helps us to determine the date and aim of the creation of the Notitia Arnonis since the text contains a very important reference to the time of recording.Before looking at this sentence closely, it is worth surveying the sources related to the creation of the Notitia Arnonis. In the item numbered 168 from among Charlemagne's charters we can read about donations of kings and queens, dukes and other pious people. In addition to donations granted by dukes and other people, this charter dated to 791 by Herwig Wolfram mentions donations by kings and queens that the Notitia Arnonis does not refer to, and it can do so because-Pippin being Tasilo's guardian-Bavaria had established close relations with the Franks and Frankish kings qualified even retroactively benefactors of the Bishopric. Based on that this charter does not seem to be connected with the Notitia Arnonis, especially because the parallel point of text in the Breves Notitiae expressis verbis indicates the assistance of Pippin and his elder sister, Hiltrud.So if the charter number 168 referred expressly to the Notitia Arnonis, then quite peculiarly the document to be confirmed, that is, the Notitia Arnonis would not contain the facts enumerated in the charter. This charter, as a matter of fact, can be compared to several charters with similar content.
Charter number 162 from 788 also deals with current events in Bavaria, and this is confirmed also by the Traditio Frisingensis, according to which "anno secundo, quod domnus rex Carolus Baiuariam adquisivit ad Tassilonem clericavit", and "in secundo anno quo translatus est Tassilo dux de regno suo". Sources clearly date the aquisitio to 788, based on which we can with full certainty take the statement "eodem anno, quo ipse Baioariam regionem ad opus suum recepit" in the Notitia Arnonis refer to the year 788. More profound analysis of the quoted sentence reveals that the time adverbial complement (anno, quo ... recepit) belongs to the first predicate (exquisivi), and this should not be absolutely equal to what is denoted by the second predicate (conscribere ad memoriam feci), especially as the phrase standing behind and explaining the first predicate (a monachis et laicis) suggests the separation of the two predicates. On the other hand, this sentence does not necessarily have decisive significance in dating the Notitia Arnonis because Deacon Benedictus borrowed the idioms "viris valde senibus et veracibus diligentissime exquisivi" as well as "ad memoriam conscribere" from the Libellus Virgilii word for word.
So it is possible that the Notitia Arnonis was recorded as early as in 788; however, a somewhat later point of time cannot be ruled out either; the text left to us by Rotulus specifies the year 790 (anno dcc lxxxx) as title, and only insertions from the 15th century added the reference to 788 (anno dcc lxxxviii congestum); consequently, the earliest dating from the 12th century known to us allows to make the year 790 probable. Based on that, in summary it can be established that the report demanded by Charlemagne at the end of 788 (eodem anno, quo ipse Baioariam regionem ad opus suum recepit) and readily presented by Salzburg, which partly relies on and draws on older records, partly contains new supplementary data, was created sometime between the occupation of Bavaria by the Franks, i.e., 788 and-as it can be deduced from the texts left to us-790.
The survey of the structure of Notitia Arnonis does not cause any special difficulties; Fritz Lošek outlines it as follows:
A Donations made to the Ecclesia Sancti Petri of Salzburg (1-6.)
Donations by dukes (1-5.)
a Theodo (1.)
b Theodbert (2.)
c Hucbert (3.)
d Odilo (4.)
f Tasilo (5.)
Donations by nobles and semi-freemen (6, 1-25.)
a Donations (6, 1-21.)
b Cella Au (6, 22-23.)
c Cella Otting (6, 24-25.)
Ecclesiae parochiales (6, 26-28.)
B Nonnberg (7.)
C Cella Maximiliani (8, 1-7.)
Eschatolcollon (8, 8.)
Compared to the later record, the Breves Notitiae, the structure of Notitia Arnonis is more clearly arranged-although the topographical principle appears in addition to the chronological order naturally arising from donations by dukes, the author makes an effort to keep to the chronology within that.
The Breves Notitiae mention Arn as Archbishop on three occasions; however, it seems to be worth looking at "archiepiscopus" passages more closely. Regarding the first locus Wilhelm Levison has already called the attention to the point that presumably what we have here is an early glossa marginalis. The name "archiepiscopus" is fully righteous when it refers to Arn's acts after 798; however, the phrase "notum sit, quod" appears to be a later insertion, which seems to be confirmed by the plural of the term "pars" that occurs only here in the Breves Notitiae-based on that it is probable that this sentence was completely put in the text as a later interpolation. Regarding the second locus we can presume that it is not a later interpolation since it could not be logically explained why the copier would have wanted to stress Arn's dignity as Archbishop specifically in this case and would have referred to him merely as Bishop in the rest. We have no reason to doubt the originality of the third locus, and from chapter fourteen of the Breves Notitiae it can be observed and based on the chronological arrangement within the topographical grouping it can be declared that Arn was granted the donations mentioned in the quoted paragraph and the paragraphs following it as Archbishop, specifically after 798. The Breves Notitiae mention Arn as Bishop on several occasions, he is reliably referred to as Archbishop on two places, one of the loci refers clearly to a point of time after 798, and the other one to times around 798, from all the above we can draw the following two contradicting conclusions. On the one hand, we could presume that after 798-and, if we date the time of creation of the Breves Notitiae after 798 and before 816, this means eighteen years-so few legal transactions were implemented that the source was able to name Arn as Archbishop maximum on two places. On the other hand, we could set out from the point that the Breves Notitiae were recorded such a short time after 798, i.e., after Salzburg had become Archbishopric, that no other than transactions executed directly after 798 could be integrated in the text; that is, most of the donations, exchanges and purchases were entered into before 798.
Herbert Haupt took a stand on dating the Breves Notitiae to a later point of time, at certain points to the mid 9th century; he based his arguments mainly on linguistic and stylistic considerations. Within the source he distinguished between a Breves Notitiae I text (approximately chapters 1-14) and a Breves Notitiae II (chapters 15-24) text; he presumed that the Breves Notitiae I was created after 798 in view of the fact that those chapters mention Arn as Archbishop-yet, he was able to demonstrate this in two loci only, the first of which is certainly the result of a later interpolation. As a linguistic argument he pointed out the substantivum use of "fornax", stating that while the first part several times includes the phrase "fornacium loca", later on the formulation "fornacium I" is applied; however, this argument was refuted by Fritz Lošek by several quite apposite arguments. On the one hand, it is rather problematic to infer tendencies of linguistic development from a single locus; on the other hand, it is not absolutely necessary to see singularis nominativus in the structure "fornacium I", use as genitivus partitivus can be imagined just as well; on the third part, if "fornacium" were nevertheless singularis nominativus, then this would clearly fit in with the tendency of the shift from the more complicated third to the simpler second declinatio in the development of Middle Latin; on the forth part, differences in meaning can be also discovered between "fornax" and "fornacis locus"; and, on the fifth part, it is hard to explain why the form "fornacium" further from Classical Latinity would stand in the later part, which was recorded after emendation in the Carolingian Age. (The same applies to Haupt's argumentation regarding the word "fluvius".) It is an indisputable fact that the second part of the Breves Notitiae follows the somewhat schematic "David presbiter vir nobilis ad Salzpurch totum dedit, quod habuit ad Chirihheim" pattern, yet the reason for stylistic changes appearing there-which are really significant compared to the first part of a more narrative tone-should be looked for primarily in the differences in content.
These arguments, based on which we presume the year 798 as the date of the creation of the Breves Notitiae or accept a year a little later, are strengthened by the circumstances of the dispute over the church of (Michael)Beuern. "Ludowicus rex" referred to must have been Louis the Pious, and the dispute might have been concluded during the reign of Charlemagne, the latest at the end of August, 799 as the above-mentioned Gerold died on 1st September in the same year-i.e., what is contained in chapter thirteen of the Breves Notitiae is about events that took place the latest in 799. Dating to a point of time shortly after 798 is supported by the fact of Charlemagne providing Salzburg with immunity, which might have accounted for clarifying estate conditions; and although this charter of Charlemagne has been lost we can deduce its content from the confirming charters of Louis the Pious dated from 5th February, 816 and of Louis the German from 837. Based on the above, Heinrich Wanderwitz claims that Breves Notitiae were created between 798 and 814; however, this interval can be narrowed by presuming-and taking historical circumstances into consideration accounts for it-that the immunity granted to Salzburg is connected with, and for this reason in time almost coincides with making Salzburg an Archbishopric. It can be established that the immunities granted by Charlemagne were issued mostly between 787 and 800. Based on that, we can determine 798 as terminus post quem and 800 as terminus ante quem of the immunity granted to Salzburg; thus, again we can define the period between 798 and 800 as the period of the creation of the Breves Notitiae.
Following Fritz Lošek's division, the structure of the Breves Notitiae-which embraces both content and linguistic/stylistic aspects-can be outlined as follows:
A Theodo and Rupert (1-3, 7.)
a the Rupert legend and the first donations (1-2, 11.)
b Libellus Virgilii I/1. (3, 1-3, 7.)
B Rupert and Theodbert (3, 8-5, 5.)
a Libellus Virgilii I/ 2. (3, 8-3, 16.)
b Nonnberg (4.)
c Saint Peter Monastery (5.)
C Hucbert (6-7, 4.)
a Donations (6.)
b Interpolation: hunting law and use of forests (7, 1-7, 2.)
c Donations (7, 3-7, 4.)
D Odilo and Virgil (7, 5-10, 5.)
a Conditions of power and first donations (7, 5-7, 7.)
b Libellus Virgilii II. (8, 1-8, 15.)
c Odilo's donations to the Monastery of Saint Maximilian (9.)
d Nobles' donations to the Monastery of Saint Maximilian (10.)
E Virgil (and Tasilo) (11-13, 13.)
a Conditions of power and first donations (11, 1-11, 3a)
b Interpolation: supplements to donations (11, 3b-12, 3.)
c Libellus Virgilii III. (13, 1-13, 7.)
d Donations to Otting (13, 8-13, 13.)
Donations of nobles and the commons to Salzburg (14, 1-24, 4.)
The first part (1-13) is dominated by a chronological, the second part (14-24) by a topographical principle of arrangement; however, these principles are not used consistently. In the first part, the protagonists of the ecclesiastical and secular sides, so the Bishops of Salzburg from Rupert to Virgil and the Dukes of Bavaria from Theodo to Tasilo III, usually appear in pairs; as from Hucbert, at the end of passages, after dukes' donations, nobles' donation follow.
It is worth paying some attention to the relation between the Breves Notitiae and earlier texts from Salzburg, especially to the Gesta Hrodberti and the Notitia Arnonis. The hagiographic work on Rupert's life produced significant effect on the text of the Breves Notitiae, contrary to the text of the Notitia Arnonis, in which Rupert appears as "domnus Hrodbertus" rather than a saint, confessioner or bishop, on whose acts and the charters drafted thereon the writer of the memorandum can safely rely on. This difference between the Notitia Arnonis and the Breves Notitiae is made understandable by the fact that in the Notitia Arnonis it was not the legitimization of the primacy of Salzburg-which obtained primary importance later in the Breves Notitiae, and which could be highly advanced by reviving the Rupert tradition-that Bishop Arn and Deacon Benedictus kept in view. There are differences in the portrayal of Rupert's activity also in the Breves Notitiae emphasizing the apostolic function of the saint: chapter four of the Gesta Hrodberti asserts that the saint was working primarily on renewing the Christian faith; the Breves Notitiae claim that Rupert baptized Duke Theodo together with his dignitaries; chapter one of the Conversio extended this conversion to simple folks. The Breves Notitiae emphatically mention Rupert's activity covering the whole of Bavaria and its peoples-quite clearly, a tendency meant to support the primacy of Salzburg-yet, as a matter of fact, they are silent about his journey to Lorch referred to in the Gesta Hrodberti and his trip up to the frontiers of Pannonia described in the Conversio. The data regarding Seekirchen am Wallersee are contained in the Gesta Hrodberti and the Conversio as well as in the Breves Notitiae. The Gesta Hrodberti-and the Conversio-assert that having arrived to Salzburg Rupert found old buildings there from the Roman age at that time already in ruined state, and having received the permit asked from DukeTheodo he started construction works. Contrary to that, the Breves Notitiae claim that Rupert came to Salzburg with the Duke's consent, he found old buildings there, built a church, equipped the bishop's seat, and only after that did Duke Theodo appear and gave him the place as a seat, and made other donations.
From all the above it is apparent that Salzburg plays a more significant role and Rupert is more intensely tied to the town in the presentation of Breves Notitiae than in the narrative of either the Gesta Hrodberti or the Conversio: from among the places visited by Rupert before Salzburg only Seekirchen is given account of in view of the fact that, contrary to the town of Lorch which the Breves Notitiae are silent about, Salzburg had estates there; Rupert acted independently in Salzburg; only later did Theodo give his consent to the results of his acts, which emphasises the activity and autonomy of the Bishops of Salzburg. The prefiguration of the passage in the Breves Notitiae on founding the cella Maximiliani-a part of the Libellus Virgilii-must have been a relevant paragraph in the Gesta Hrodberti. The only sale and purchase, which is touched on also in the Gesta Hrodberti, the purchase of Piding can be found in the Breves Notitiae too. Regarding the nunnery founded by Rupert, the Gesta Hrodberti adds that the Bishop had brought his relative, Erintrudis from his own country, Worms and made her the head of the nunnery; the Breves Notitiae, as a matter of fact, do not refer to Worms since they do not discuss the issue of the saint's origin, On the one hand, and in accordance with the tendency of the source, the Bishop's own country could have been nothing else than Salzburg itself, on the other; similarly, the Breves Notitiae stress that Erintrudis was made Mother Superior in accordance with the Duke's permit and will, which clearly indicates the image of the Bishop and Duke proceeding in agreement, which is present throughout the Breves Notitiae.
The form of presentation of the Breves Notitiae, highly Caroling based at certain points, is acknowledged and accepted, albeit, to various extent, among researchers. This tendency is indicated by the occurrence of Pippin and his mother, Hiltrud in the text, who are not mentioned at all in the Notitia Arnonis, and beside whom Duke Tasilo is reduced by the narrative to a kind of supporting character. On no more than five occasions do the Breves Notitiae mention Tasilo, compared to nine loci in the Notitia Arnonis, which is the shorter source, and grammatically he functions as subject in the text only once. (One of the most striking examples for different handling of sources in the Notitia Arnonis and the Breves Notitae is the locus dealing with the donations granted to Nonnberg-both in terms of its content and its place in the text: whereas the Breves Notitiae rank them among the donations made in the age and through the assistance of Rupert and Theodbert, the Notitia Arnonis covers this donatio in a completely separate chapter and independent praefatio.) Herwig Wolfram uses the qualification "(eindeutig) prokarolingisch" with regard to the Notitia Arnonis; at variance, in relation to the Notitia Arnonis Heinrich Wanderwitz points out the unambiguous sympathy shown by the source towards the Agilolfings, except for the introduction and the closing chapter. Fritz Lošek declares that it is not righteous to make such an unambiguous and lapidary statement regarding the Breves Notitiae just as with respect to the Notitia Arnonis, and he has pointed out that specific Bavarian dukes are adjudged differently on an individual basis; the absolutely positive image of Theodo and Theodbert is followed by far from flattering portrayals of Odilo and Tasilo, which is supported by the conflict outlined in the Libellus Virgilii too.
Clearly, it can be established and declared that none of the two sources served as prefiguration for the other; that is, the Breves Notitiae cannot be considered an enlarged adaptation of the Notitia Arnonis; it was drafted completely independently of it. Both authors-Deacon Benedictus in the case of the Notitia Arnonis and the unknown author of the Breves Notitiae-had certain records at their disposal, which they could use as sources. A further essential difference is indicated by the title of the two sources itself: singular in the Notitia Arnonis shows that in its case what we have is a subsequently made uniform record based on collecting information; and plural in the Breves Notitiae indicates that most probably it merges several records made earlier. Furthermore, certain differences between the two texts can be demonstrated and proved by focusing on to what extent they updated, adjusted the sources they used-former notitiae-to their own age: such updating, which can be revealed by comparing loci with identical content in the Notitia Arnonis and the Breves Notitiae, is mostly indicated by the words "nunc" and "tantummodo". In modifications, quite often, Latin and Roman settlement names were replaced by German names; sometimes, efforts were made to make economic and topographical conditions correspond to the conditions of the period; the shorter text, i.e., Notitia Arnonis had suffered changes-on the other hand, updating tendencies can be specifically identified only in loci that were contained by both; yet, these modifications can never be found simultaneously in the loci with identical content of the two texts. From these results two conclusions can be drawn: first, as we have already established, neither the Notitia Arnonis, nor the Breves Notitiae served as prefiguration or source for the other text; secondly, it is apparent that the later text, the Breves Notitiae preserve the older text layer closer to the original sources, the notitiae.
As in terms of genres both the Notitia Arnonis and the Breves Notitiae are a peculiar mixture of charters and historical narrative (genus mixtum) regarding the early period of Salzburg, in the investigation of the linguistic/stylistic characteristics we have to take both usage of charters and the influence of Carolingian Latinity into account. In our work we have examined the following aspects-primarily in the text of the above-mentioned estate registers of Salzburg, and, as a matter of fact, by making an overview of the (Bavarian) sources of the period. Written confirmation by witnesses of donations made for the benefit of the Church. The relation between carta and notitia; similarities and differences in form and content. Key linguistic characteristics of estate registers of Salzburg; their relation to usage of charters. In-depth analysis of a few linguistic characteristics that occur in the estate registers.
To assert the fact of donation, the records use the verbs "dare", "donare" and "tradere", but the frequency of their occurrence is far from identical in the two lists: in the first part of the Breves Notitiae (chapters 1-13) the verb "dare" occurs approximately two hundred and forty times, while in the Notitia Arnonis only six times. The use of the verb tradere is almost identical in the two sources, which means that the fact of donation is indicated in ninety-nine percent with this verb in the Notitia Arnonis whereas the first part of the Breves Notitiae (the aforesaid chapters 1-13) express the execution of the transaction in approximately seventy-five percent with the verb "dare". Nevertheless, no difference can be discovered between the meaning of the two verbs in these sources when examining the character and legal fate of the estates indicated with various verbs that occur in loci identical in content in the two sources. Consequently, differences in the use of verbs can be attributed, instead of difference in content, to the simple fact that the Notitia Arnonis was addressed by the Bishop of Salzburg directly to the ruler, Charlemagne as it were as an official instrument, and in a quite critical period, and it was retaining the estates of Salzburg that was at stake. The verb "trado" better suited the official style of charters-in some sources (see the charters of donations from Mondee, Passau, Freising and Fulda) it was used in the pleonasm "trado et transfundo". Whereas, in the Breves Notitiae, which was to confirm the primacy of Salzburg within Bavaria, the outcome of the Carolingian language reform can be identified, and it is also due to the less formal nature of the record that the formal verb "tradere" was replaced by the compilor with the colloquial verb "dare".
Both records emphasized the necessity to identify donators unambiguously and clearly: to break the monotony of the text a little, instead of repeating the proper name, they often used the phrases known from charters "suprascriptus", "praenominatus", "supradicus" etc., which are typical items of the late antique and early medieval vocabulary of chancelleries and replaced the classical pronouns "idem" and "ipse". The terminology of the period worked out several forms of these phrases such as "iam dictus", "iam fatus", "iam scriptus", "memoratus", "praedictus", "praefatus", "praescriptus", "superscriptus", "supradictus", "suprascriptus". Striving to be absolutely precise, at certain points, the Notitia Arnonis elects to use rather pleonastic formulations: it links the anaphoric participium perfectum to the classical, anaphoric pronoun, thus creating, among others, the structure "ipseque dux iam scriptus" (quite frequent both in the vocabulary of the people and the Chancellary). Contrary to this, in two thirds of the relevant cases the Breves Notitiae use the pronoun "idem" instead of a paraphrase with participium, which again seems to show the impact of the Carolingian age, the felt need to get closer to literary language instead of the inveterate usage of the charters.
The language reform solicited by Charlemagne perfectly corresponds with the reform of writing also launched by him since these efforts were meant to advance a more efficient administration of a unified empire. As a matter of fact, reform efforts could not bring success unless built on the tradition of Latinity and the related culture of writing. The language reform, however, was in the first place to correct and eliminate phonetic/morphologic distortions (in this attempt the language of the Church Fathers was followed as a pattern); purification of word usage and syntax was attained only secondarily and accidentally. The retention of almost unchanged word usage and syntax can be attributed to a simple and logical reason: any amendment/correction in them would have led to perceiving uncertainty in law. Although the improper spelling of several phrases (habyre, pristetirunt, estipendiis) were corrected to ensure proper forms (habere, praestiterunt, stipendiis), syntax was left unchanged, except for correcting suffixes (e.g., "illut que" was replaced by "illud quod"). In the Notitia Arnonis and the Breves Notitiae this kind of tendency to improve the language can be also observed: dismissing the base of "villa nuncupante" in the Notitia Arnonis, in the relavant locus the Breves Notitiae use the phrases "villa nuncupata", "villa dicta", "villa quae dicitur", or "villa quae vocatur".
In the specification of the place of donation, the various forms of verbs "nuncupare" and "vocare" occur in a total of twenty-five times in the Notitia Arnonis, and far exceed the verb "dicere" preferred in the Breves Notitiae. On the other hand, the frequent use of the prepositions "in" and, more often, "ad" can be also noticed, which is linked to a more accurate specification of the place in a participle clause to shorten the sentence or in a relative caluse. In this respect the Notitia Arnonis uses participium imperfectum (mostly erroneously) more often, while the Breves Notitiae prefer participium perfectum. Comparison with former charters makes it possible to declare that the participium imperfectum "nuncupante" is an inveterate form quite recurrent in charters from the Merowing Age and the Formulae Marculfi. It is reasonable to assume that 8th century Bavarian charters the Notitia Arnonis was based on also used this form, which was borrowed by the record drafted on the orders of Arn since the impact of the Formulae Marculfi can be demonstrated in several Bavarian charters of the period, e.g., in charters from the Monastery of Mondsee and Sankt Gallen.
In the specifications of the donation the subject itself and the accessory that belongs to it are clearly separated. The actual subject, that is, the land is described in the Notitia Arnonis, on almost each occasion, more specifically, forty times with the formula "mansos inter vestitos et apsos", i.e., "partly cultivated, partly uncultivated lands". This phrase is never used in the Breves Notitiae; "mansus" is replaced by "manentes", clearly manifesting a paradigm shift both in language and content; and the adjectives "vestitus" and "apsus" are totally missing from later records. The adjectives "vestitus" and "apsus" are, accordingly, used as synonyms of "cultus" and "incultus"; the adjective "apsus" and its derivatives (absitas, absitus, absare) can be considered forms rooted in the vocabulary of Bavarian and Langobard charters. Also, Langobard impact is implied by the use of pronouns in Notitia Arnonis; especially in view of the rare occurrence of the pronouns "hic" and "iste" and the frequent occurrence of the pronouns "is" and "ipse", which manifests definite comparison with Leges Langobardorum. Similarly, it should be noted that "inter" is used as an adverb, in the sense "tam ... quam", which reveals connection with the usage of the Bible and the Formula Marculfi.
The system of listing accessories deserves special attention as the most representative loci in this respect in the Notitia Arnonis and the Breves Notitiae clearly demonstrate the differences and changes in the language of charters from the Agilolfing and Carolingian age. Beside corrections in morphology ("castrum superiorem" now is replaced by "castrum superius"), the lengthy listing of accessories, which is sometimes senseless and incomprehendable due to the rigid formula it follows, makes them more specific and comprehensible: so while the Notitia Arnonis applies "confinia, aquis aquarumque decursibus" and "adiacentiis", the Breves Notitiae use "finales loci, aquis circumquaque currentibus" and "adiacentibus". At the same time-just to ensure that the language of former charters, which both records drew on independently from one another, should not be totally changed in order to avoid uncertainty in law and endless disputes arising from that-the order of accessories is identical; subjection to the preposition "cum" and the sentence structure are the same in both records; and the various forms of the same words recur in both cases, which was knowingly meant to preserve close relation with original charters. In both cases, the list of accessories is introduced by the preposition "cum", which is confirmed by an adverbial "unā" in the Notitia Arnonis; "cum" stands, in each case, in ablativus, except for the aforesaid formal phrase "cum mansos", which implies Langobard roots.
The parts of the list of accessories are interrelated asyndetically; and wherever this interralation is polysyndetic, "vel" having a disjunctive sense in classical usage becomes the copulative element as a typical feature borrowed from folk language and adopted in the Latinity of charters. These features apparent in the earlier drafted, more archaic the Notitiae Arnonis influenced by folk language, manifesting both Bavarian and Langobard impacts-albeit, not more than ten years passed between the time of their drafting, and both of them were recorded on the orders of Arn-are no longer used in the Breves Notitiae, which demonstrates the traces of the Carolingian language reform. Comparing syntax, finally it should be added that the language of charters in the Breves Notitiae is quite often replaced by narrative style; especially in the sections where the author borrowed texts from the Libellus Virgilii that thematized the dispute evolving on cella Maximiliani and Otting: in these two loci the author makes use of the instrument oratio recta, which is absolutely not applied in the Notitia Arnonis.
What follows is an investigation of the use of the terms "manus" and "manentes", and "coloni" and "colonia" in the Notitia Arnonis and the Breves Notitiae-primarily taking into account the statements made by Fritz Lošek, the researcher who has the most extensive knowledge of these texts. In the Notitia Arnonis and the Breves Notitiae the terms "mansus", "manentes" and "colonium" occur quite differently: the noun "manus" and its derivatives occur eighty-five times in the Notitia Arnonis, and thirty-seven times in the Breves Notitiae; among them the occurrences in the Breves Notitiae can be identified in nineteen cases as the pluralis accusativus of "mansus"; consequently, they can be ranked among the fourth declination; the forms of "mansum" that can be identified as singularis accusativus allow ranking both among the second and the fourth declination; other occurrences (most frequently pluralis ablativus) usually allow ranking among the second declination. The forms "mansos", "manso" and "mansi" are absolutely not used in the Breves Notitiae; contrary to that, in the Notitia Arnonis "mansos" they are used forty-nine times, "manso" nine times, and "mansi" twice. The pluralis accusativus form "mansos" often stands with the preposition "cum"; "manso" stands in the number "I (uno)", on the one hand; and substitutes singularis accusativus, on the other; e.g., "tradidit ... in loco ... manso I vestito". The case "mansi" occurs twice right after one another. Furthermore, the singularis accusativus form (mansum) also appears, just as pluralis accusativus, on twenty-six occasions in total. The phrases "mansi tributales" and "tributarii aput mansos" usually refer to "Romani. "Mansi" might have been serviles, that is, servile souls too but it is impossible to ascertain whether in these cases this peculiarity is attached to the owner of the land or the land itself.
The phrase "mansus" so often used in the Notitia Arnonis is quite often replaced in the Breves Notitiae by the noun "manentes"; Wanderwitz has demonstrated that the reason that lies behind this linguistic change is related to content: this way the status of the persons to be donated could be more precisely determined. "Manentes" as the participle, or a form having become the participium substantivum of the verb "manere" occurs, except for a single case, solely in the Breves Notitiae-the only "manentes" passage in the Notitia Arnonis does not carry any legal content; there "manentes" means simply inhabitants. In the Breves Notitiae the transition from the participle usage to the substantive usage can be clearly followed up. In some of these cases, the form "manentes" is a simple participium coniunctum, which is meant to determine the position of the servi, or tributales (Romani); that is, to specify their capacity of having a house more accurately. Elsewhere the term "manentes" is used as a substantive, and so the relation of settlers and ploughmen (coloni) becomes closer with the land (mansus); that is, "mansus" and "colonia" most probably become synonyms. Between the two options of occurrence, in Lošek's view, the phrase "manentes in coloniis" constitutes transition, since "manentes" are handed over not only "in coloniis" but also "cum coloniis" in the text of the Breves Notitiae. The phrase "cum coloniis" defines donated persons more exactly as "colonia is also considered accessory. As a matter of fact, it might occur that the subject of donation is the villa itself; and mansi and manentes are only the accessories thereof. In the Breves Notitiae there are also cases where the subjects of donation are mansi, or manentes (servi); the participium in this case loses its substantive, and means the persons who reside on the mansus, or colonium.
The interpretation of "colonium" in the estate registers poses more difficulties than the interpretation of "mansus" and "manentes". Absolutely no traces of the phrase "colonium" can be found in the Notitia Arnonis, only in the Breves Notitiae does it appear; and here it is possible to witness a shift in the meaning from the thing towards the person. The form pluralis accusativus (colonia) appears on several places; but the author knows the word "colonia" too. At the same time, the concept of the person related to the colonium, the "colonus" is not unknown to the author either. The difficulty is caused by the pluralis ablativus forms but this can be solved-just like in the case of "mansus" and "manentes"-as follows: as the accessory of the thing that constitutes the subject of donation a person is indicated and as the accessory of the person a thing is indicated. However, a land can never be the accessory of a thing (land), and a person can never be the accessory of a person; that is, in this respect the author avoids tautology. The form "coloni" can be undoubtedly the singularis genitivus of either "colonium", or "colonus" but in the context of the three occurrences relevant in this respect persons constitute the subject of donation; therefore, it is reasonable to assume that in this case the specific item is the genitivus of "colonus".
From the comparable loci of the Notitia Arnonis and the Breves Notitiae-which sometimes contain the subject of donation and the accessories thereof in an identical structure, and sometimes exchange their interrelation-several conclusions can be drawn. The Breves Notitiae more often name persons as the subject of donation than the Notitia Arnonis; on the other hand, in several cases the Breves Notitiae mention only the land (mansus) as the subject of donation but says nothing of the persons that might belong to it. Following the tendency of separating persons from things, the Breves Notitiae more often use the phrases "colonium" and "colonus"; at variance with that, loci with identical content in the Notitia Arnonis expound on "mansus". On the grounds of the above, on the one hand, it can be assumed that the author's willful uniforming tendency lies behind the uniformity of the form of expression of the Notitia Arnonis, and contrary to that, the Breves Notitiae pass on a more varied formulation closer to the original charters; on the other hand, it cannot be excluded that it was the Notitia Arnonis that borrowed the uniform formulation from the original documents, and the Breves Notitiae adjusted it to the current conditions of the period. If, however, we assume that the donation procedure of Salzburg reflecting the several decades' long process was as colourful as the the documents on donations granted to Freising (Traditiones Frisingenses)-the original charters on such donations (contrary to those of Salzburg) have been left to us-then, in Lošek's opinion, it is the text of the Breves Notitiae that is closer to the original formulation. In this spirit, the condition of the lands specified cultivated or uncultivated in the Notitia Arnonis (mansos vestitos et apsos) reflects the general conditions typical of the period of the compilation of the record; furthermore, the term "colonia" goes back to a longer history in Bavaria than "mansus". All these facts serve as further proofs of the correctness of the statement that the Notitia Arnonis could not serve as a prefiguration of the Breves Notitiae.
* This paper was supported by the following research fund program: OTKA K78537.
 Cf. Diesenberger, M.-Wolfram, H.: Arn und Alcuin 790 bis 804: zwei Freunde und ihre Schriften. In: Erzbischof Arn von Salzburg. Hrsg. v. M. Niederkorn-Bruck- A. Scharer. Veröffentlichungen des Instituts für österreichische Geschichtsforschung, Ergänzungsband 40. Wien-München 2004. 86. sqq.
 See Demmelbauer, G.: Arno, der erste Erzbischof von Salzburg 798-821. (Dissertation) Wien 1950; Kerner, M.: Der Reinigungseid Leos III. vom Dezember 800. Die Frage seiner Echtheit und frühen kanonistischen Überlieferung. Zeitschrift des Aachener Gerichtvereins 84/85. 1977/78. 131. sqq.; Classen, P.: Karl der Große, das Papsttum und Byzanz. Die Begründung des karolingischen Kaisertums. Sigmaringen 1985. 42. sqq.; Wavra, B.: Salzburg und Hamburg. Erzbistumsgründung und Missionspolitik in karolingischer Zeit. Berlin 1991; Wolfram, H.: Arn von Salzburg und Karl der Große. In: 1200 Jahre Erzbistum Salzburg. Die älteste Metropole im deutschen Sprachraum. Hrsg. v. H. Dopsch- P. F. Kramml- A. S. Weiß. Salzburg 1999. 22. sqq.
 Notitia Arnonis 8, 8.
 MGH Dipl. Karol I. 226.
 Wolfram, H.: Die Geburt Mitteleuropas. Geschichte Österreichs vor seiner Entstehung. 378-907. Wien 1987. 189.
 Lošek, F.: Notitia Arnonis und Breves Notitiae. Die Salzburger Güterverzeichnisse um 800. Mitteilungen der Gesellschaft für Salzburger Landeskunde 130. 1990. 31.
 Lošek 1990. 31; Wolfram 1987. 190.
 Breves Notitiae 11, 1-2.
 Lošek 1990. 32.
 MGH D Karol. I. 219.
 Traditio Frisingensis Nr. 127a
 Traditio Frisingensis Nr. 127b
 Notitia Arnonis 8, 8.
 Lošek 1990. 32.
 Notitia Arnonis 8, 8.
 Lošek 1990. 33.
 Libellus Virgilii 2. Cf. Wolfram, H.: Libellus Virgilii. Ein Quellenkritisches Problem der ältesten Salzburger Güterverzeichnisse. Vorträge und Forschungen 20. Sigmaringen 1974. 177. sqq.
 Cf. Breves Notitiae 8, 12.
 Lošek 1990. 11. sq.; Keinz, F.: Indiculus Arnonis und Breves Notitiae Salzburgenses. München 1869. 3. sqq.
 Lošek 1990. 34.
 Lošek 1990. 34.
 Lošek 1990. 51.
 Breves Notitiae 14, 33.
 Levison, W.: Die älteste Lebensbeschreibung Ruperts von Salzburg. Neues Archiv der Gesellschaft für ältere deutsche Geschichtskunde 28. 1903. 216; Lošek 1990. 35.
 Breves Notitiae 14, 19. 29. portiones
 Breves Notitiae 14, 54.
 Lošek 1990. 36.
 Breves Notitiae 15, 4.
 Breves Notitiae 15, 4-7.
 Lošek 1990. 36.
 Breves Notitiae 14, 40; 16, 3; 18, 8; 19, 3. 4; 23, 1. 3. 5; 24, 2.
 Breves Notitiae 14, 54; 15, 4-7.
 Lošek 1990. 37.
 Haupt, H.: Zur Sprache frühmittelalterlicher Güterverzeichnisse. Mitteilungen des Instituts für österreichische Geschichtsforschung 83. 1975. 35.
 Breves Notitiae 14, 33. 54.
 Lošek 1990. 38.
 Breves Notitiae 2, 5; 4, 5; 9, 6.
 Breves Notitiae 14, 50.
 Haupt 1975. 47.
 Lošek 1990. 38. sq.
 Lošek 1990. 39.
 Breves Notitiae 16, 1.
 Haupt 1975. 46.
 Lošek 1990. 39.
 Breves Notitiae 13, 13.
 Wolfram 1987. 190.
 Lošek 1990. 40.
 SUB II. Nr. 5.
 Wanderwitz, H.: Quellenkritische Studien zu den bayerischen Besitzlisten des 8. Jahrhunderts. Deutsches Archiv für Erfoschung des Mittelalters 39. 1983. 58.
 Lošek 1990. 41.
 See Semmler, J.: Traditio und Königsschutz. Studien zur Geschichte der königlichen monasteria. Zeitschrift der Savigny-Stiftung für Rechtsgeschichte, Kanonistische Abteilung 45. 1959. 1-33.
 Lošek 1990. 42.
 Lošek 1990. 43.
 Lošek 1990. 44.
 Schmitt, F.: Zur Vita Ruperti. In: Frühes Mönchtum in Salzburg. Salzburger Diskussionen 4. Hrsg. v. E. Zwink. Salzburg 1983. 100.
 Wolfram, H.: Vier Fragen zur Geschichte des hl. Rupert. Eine Nachlese. In: Festschrift St. Peter zu Salzburg 582-1582. Studien und Mitteilungen zur Geschichte des Benediktinerordens 93. St. Ottilien 1982. 18.
 Lošek 1990. 26.
 Gesta Hrodberti 4.
 Breves Notitiae 1, 1.
 Conversio 1.
 Breves Notitiae 1, 2.
 Gesta Hrodberti 5.
 Conversio 1.
 Gesta Hrodberti 6; Conversio 1.
 Breves Notitiae 1, 3.
 Gesta Hrodberti 6.
 Gesta Hrodberti 7-8.
 Breves Notitiae 2, 1-3.
 Lošek 1990. 28.
 Cf. Koller, H.: Zur Frühgeschichte der ältesten Klöster in der Umgebung von Salzburg. Mitteilungen der Gesellschaft für Salzburger Landeskunde 117. 1977. 10.
 Breves Notitiae 3, 7-8.
 Gesta Hrodberti 7-8.
 Lošek 1990. 29.
 Gesta Hrodberti 8.
 Breves Notitiae 2, 4.
 Gesta Hrodberti 9.
 Cf. Gesta Hrodberti 1.
 Breves Notitiae 4, 1.
 Lošek 1990. 30.
 Wanderwitz 1983. 56; Wolfram 1974. 203.
 Breves Notitiae 11, 1-2.
 Lošek 1990. 47.
 Breves Notitiae 11, 1. 2; 13, 1. 10; 14, 4.
 Notitia Arnonis 5, 1. 7; 6, 1. 2. 13. 15. 22. 24.
 Breves Notitiae 11, 2.
 Notitia Arnonis 7, 9.- Breves Notitiae 4, 7; Notitia Arnonis 7, 3.- Breves Notitiae 4, 5; Notitia Arnonis 7, 12.- Breves Notitiae 4, 9.
 Breves Notitiae 4.
 Notitia Arnonis Praef.; 7.
 Wolfram 1974. 203.
 Wanderwitz 1983. 45.
 Lošek 1990. 21.
 Lošek 1990. 21.
 Breves Notitiae 18, 8; 16. 1; 16, 3; 18, 5.
 Lošek 1990. 22.
 Breves Notitiae 3, 1. = Notitia Arnonis 8, 1; Breves Notitiae 14, 1. = Notitia Arnonis 6, 2; Notitia Arnonis 1, 6. = Breves Notitiae 2, 7; Notitia Arnonis 5, 1. = Breves Notitiae 2, 10; Notitia Arnonis 7, 8. = Breves Notitiae 4, 10; Notitia Arnonis 7, 5. = Breves Notitiae 4, 4.
 Lošek 1990. 24. sq.
 Lošek 1990. 25.
 Haupt 1975. 36.
 Haupt 1975. 37.
 Cf. Fichtenau, H.: Das Urkundenwesen in Österreich vom achten bis in das dreizehnte Jahrhundert. Mitteilungen des Instituts für österreichische Geschichtsforschung, Ergänzungsband 23. Graz-Wien-Köln 1971. 19. sqq.
 Haupt 1975. 38.
 Haupt 1975. 3837.
 Notitia Arnonis 5, 7.
 Haupt 1975. 39.
 See Steinen, W. von den: Der Neubeginn. In: Karl der Große. Lebenswerk und Nachleben II. Hrsg. v. Braunfels, W. Düsseldorf 1965. 19. sqq.
 Haupt 1975. 40.
 E.g. Notitia Arnonis 1, 2. = Breves Notitiae 2, 4; Notitia Arnonis 7, 2. = Breves Notitiae 4, 2; Notitia Arnonis 7, 10. = Breves Notitiae 4, 7; Notitia Arnonis 4, 2. = Breves Notitiae 5, 3; Notitia Arnonis 2, 6. = Breves Notitiae 5, 4; Notitia Arnonis 2, 6. = Breves Notitiae 5, 4; Notitia Arnonis 3, 1. = Breves Notitiae 6, 1; Notitia Arnonis 4, 1. = Breves Notitiae 7, 6; Notitia Arnonis 5, 3. = Breves Notitiae 11, 2.
 See Diepolder, G.: Die Orts- und „in pago"-Nennungen im bayerischen Stammesherzogtum zur Zeit der Agilolfinger. Zeitschrift für bayerische Landesgeschichte 20. 1957. 364. sqq.
 John, W.: Formale Beziehungen der privaten Schenkungsurkunden Italiens und des Frankenreichs und die Wirksamkeit der Formulare. Archiv für Urkundenforschung 14. 1936. 79.
 Haupt 1975. 17. sq.
 Mittellateinisches Wörterbuch I. 1967. 50. 66. Cf. Haupt 1975. 43; Löfstedt, B,: Studien über die Sprache der langobardischen Gesetze. Beiträge zur frühmittelalterlichen Latinität. Acta Universitatis Uppsaliensis, Studia Latina Uppsaliensia 1. 1961. 254. sqq.
 Haupt 1975. 44.
 Notitia Arnonis 1, 1.
 Breves Notitiae 2, 3.
 Haupt 1975. 45.
 Fichtenau, H.: Arenga. Spätantike und Mittelalter im Spiegel von Urkundenformeln. Mitteilungen des Instituts für österreichische Geschichtsforschung, Ergänzungsband 18. Graz-Wien-Köln 1957. 2460; Fichtenau 1971. 24.
 Haupt 1975. 45.
 Breves Notitiae 8, 1. sqq.
 Breves Notitiae 13, 1. sqq.
 Breves Notitiae 8, 8; 13, 3.
 Cf. Lošek 1990. 58. sqq.
 Notitia Arnonis 6, 13.
 Notitia Arnonis 2, 4. 7.
 Notitia Arnonis 5, 4.
 Eg. Notitia Arnonis 6, 2.
 Cf. Notitia Arnonis 2, 7; 4, 8.
 Haupt 1975. 42.
 Wanderwitz 1983. 42.
 Notitia Arnonis 7, 6.
 Lošek 1990. 60.
 Breves Notitiae 2, 8; 4, 2; 7, 6; 13, 10; 18, 4.
 Breves Notitiae 1, 4; 2, 4; 21, 5.
 Lošek 1990. 61.
 Breves Notitiae 2, 9; 9, 4; 12, 3.
 Breves Notitiae 4, 7.
 Breves Notitiae 5, 2. 5; 9, 4. 5; 10, 4.
 Lošek 1990. 62.
 Breves Notitiae 14, 9. 23. 29. 45.
 Breves Notitiae 2, 7; 9, 4.
 Breves Notitiae 4, 3. 4; 14, 36. 51; 17, 1. 3; 18, 3.
 Eg. Breves Notitiae 4. 9. Cf. Lošek 1990. 62.
 Breves Notitiae 13, 12.
 Lošek 1990. 64. sq.
 Lošek 1990. 65.