Dating the Great Stemma


To discover the date when any ancient diagram or text was originally written, one starts by analysing the content of the existing manuscript copies. This often allows one to construct a hypothesis about the history of the copying process. This hypothesis can be tested, and sometimes be enhanced, by seeing if it is consistent with external historical data.

At each step along the way, one tries to push back the dating by proposing a new terminus ante quem, the last possible point in time which the item of evidence under consideration would allow as a date of origin. This recent-first, earliest-last order will be employed in the summary that follows. Many of the points on this page are discussed in more detail elsewhere in this website.

Physical: The most recent terminus ante quem is the date at which our oldest extant manuscript was penned. The Morgan Beatus, now kept in New York, is an artefact that was certainly made before the year 945, according to specialists including art historians who have written about that codex as a whole. It contains a 14-page version of the Great Stemma which resembles the other 17 extant copies in most key points.

Incorporation: Wilhelm Neuss and Henry Sanders were the first modern scholars who recognized the genealogy-of-Christ content as a separate work from the text by Beatus of Liebana in the codices, and attempted to date it.[*] Jerome and Isidore are named in the decorative dedication to the Beatus Commentary, since they were among the patristic sources read by Beatus. But they are certainly not sources for the Great Stemma, which was almost certainly drawn up before either flourished. Mention of them in the Commentary front matter cannot be used to argue that the Great Stemma post-dates them. Although his overview was limited, Neuss was presciently certain that the stemma could never have been the creation of Beatus the monk. He concluded that it was an external resource that was incorporated into the Apocalypse Commentary under the editorial authority of Beatus himself in 786. More detail.[*]Neuss.

The Neuss argument that Beatus himself was the person responsible for the incorporation was later cogently attacked by John Williams, who thought a later copyist took the decision. But the Neuss view cannot entirely be ruled out. More detail.

A named scribe: Justo, a notary and scribe who died in 772, unwittingly copied a fragment of the Great Stemma when he commenced a copy of the Four Gospels which was later to be bound into the Codex Ovetense. Teófilo Ayuso Marazuela noticed this clue.[*]Ayuso Marazuela, 1943, 161: Justo's Gospel text began, Sicut Luchas evangelista..., a text found on the last page of the Great Stemma. The Gospels manuscript can be dated because Justo entered his name in a closing prayer at the end, and a later hand added the precise date of Justo's death. Unfortunately these pages of the codex are now lost and we must trust intermediary scholars on this matter.. He suggests that Justo's prototype may have been a badly damaged old bible which had had a Great Stemma frontispiece where everything except the 14th and last page had been torn out and lost by the time it came into Justo's hands. More detail.

Anthropology: Some manuscripts of the Great Stemma contain anthropological characterizations that are clearly related to the conflict that began with the Umayyad Islamic invasion of Visigothic Iberia in 711. These characterizations vary so much among the different recensions that it would be hard to argue they are not additions by later editors. In the Roda Codex, for example, Magog is described as the ancestor of the Moors (a quo leviunde est mauritunie filiibus futh dicitur / de isto natus mauri) while Gomer is described as ancestor of the Goths (de isto natus gothi). It is plausible to suppose that these glosses reflected contemporary beliefs when the document was edited, rather than deriving from its original authorship. This is thus reinforcement for the hypothesis that the original material in the Great Stemma predates the period of the Moorish invasion.

The overburden: The manuscripts we have today contain various accretions which have been layered onto the Great Stemma in the course of its transmission, including the anthropological observations noted above. The dates of such overburden are also indicators of the age of what lies underneath. A notable accretion is a recension during the reign of King Wamba of the Ordo Annorum Mundi, a text setting out the number of years from the Creation of the World up to the "present moment" at the time of writing, which was shortly after the year 672 in Visigothic Spain.[*]Ayuso Marazuela, 1943 contends that the Ordo Annorum is by an author other than that of the Great Stemma. See my discussion for a reply.

Since elements of the ORR show up with the Epsilon, Delta and Beta recensions, it is likely that the entire range of recensions are reproductions, at one or two removes, of an intermediate and now-lost Iberian Great Stemma manuscript dating from about 672 when this information was current. More detail.

Scriptural sourcing: Our principal dating evidence relies on those elements in the unamended versions of the Great Stemma which are drawn from the Latin biblical and apocryphal texts in use before the Latin Vulgate Old Testament was translated in the period 390-405 by Jerome. Simplifying, we can take 400 as an approximation of Jerome's step-by-step "publication date" of the various texts and assume that the Vulgate came into general use over the following 150 years.[*]On the speed at which the new pushed out the old, see Fischer's article, ?Zur Überlieferung altlateinischer Bibeltexte im Mittelalter?, Nederlands archief voor kerkgeschiedenis, 56 (1975), 19. This would make it highly unlikely that the Great Stemma could have been created after 550.

There are many features in the Great Stemma that draw either on Vetus Latina texts or on early Christian or Jewish legends. These elements include:

Peregrinus: Ayuso Marazuela promoted in his writings a hypothesis that a variety of additions to Spanish bibles, including the Great Stemma, might have been drafted by Peregrinus, a shadowy Iberian scholar and theologian who was active about the year 450. Very little is known about Peregrinus other than the fact that he edited a single Priscillianist text, effectively so as to reconcile it with mainstream Catholic belief, and that he is described in the author credit for that text as a bishop. Ayuso's argument that Peregrinus had a hand in various enhancements to Spanish Bibles, including drawing up the Great Stemma, is essentially one by elimination: Ayuso names the known bible scholars of Late Antique Iberia, and by ruling each one out arrives at Peregrinus as the only possible author remaining. Although his scholarly account of the issue is illuminating, the argument is not a very convincing one and lacks any real evidence. In terms that seem usually vituperative for an exchange among two eminent Catholic priests, but are perhaps understandable, given the rivalry between the Vetus Latina enterprise at Beuron in Germany and Ayuso's Vetus Latina Hispana project in Spain, Bonifatius Fischer (1915-1997) assailed Ayuso's attribution of the bible edits to Peregrinus as "absurd.".[*]Zaluska, Feuillets, 241 summarizes the debate. Ayuso asserted that Peregrinus worked from a copy of the Liber Genealogus (see below) in the 5th century, drawing up what was essentially a new recension of it in graphic form..

Another sighting: There is compelling evidence that the G recension of the Liber Genealogus, drawn up in 427 by an anonymous author, is a textual reworking of a pre-existing Great Stemma. It adopts the same peculiar ordering of data as the Great Stemma on the macro level, proceeding (1) from Adam to David, then (2) via Matthew's genealogy to Christ, doubling back to show (3) Luke's genealogy of Christ, then setting out (4) the ancestry of Moses and Aaron. On the micro level, G contains several depth-first traverses that are clearly based on a diagram source, not on scripture. Its Vetus Latina name forms are broadly the same and it reproduces the key explanatory passage almost verbatim from the Great Stemma. More detail.

Proceeding from the opposite direction, we need to consider a succession of terminus post quem dates, that is to say, points in history which necessarily precede the document's creation.

The Vetus Latina: The obvious starting point in time for this sequence is that period when the gospels of Matthew and Luke, the main basis of the Great Stemma, were first translated from Greek into Latin, perhaps late in the second century. Keith Elliott points out that it is not entirely certain when this happened. Such translations probably began in North Africa rather than in western Europe, where Greek speakers dominated the Christian community: at Lyons, Irenaeus wrote in Greek, and in second-century Rome, Ignatius, Justin Martyr and Hermas all wrote in Greek. Even the Roman bishop Cornelius in 250 CE was writing in Greek to his churches. [*]Elliott, J. K. 'The Translations of the New Testament into Latin: The Old Latin and the Vulgate.' Aufstieg und Niedergang der römischen Welt 2.26.1. Berlin: de Gruyter, 1992, 198-245

Paola Marone says that, by contrast, the Scillitan Christians of North Africa possessed at least the Epistles of Paul in the Latin version by 180 CE, and by the time of Cyprian (who died in 258), a Latin version of the whole Bible certainly existed. The adoption of Latin as a language of theological discourse probably did not happen until significant parts of scripture were also available in Latin, though this is not a point on which one can be certain.[*] Marone, Paola. Optatus and the African Old Latin. Journal of Biblical Textual Criticism, 2006 (online).

Whose Old Testament? The Vetus Latina version of the Old Testament used by the Great Stemma author is also a potentially useful indicator of the document's age. We have reviewed above the various uncanonical and Septugiunt names that spill over into the Great Stemma text. We have also looked elsewhere at the appearance of extreme textual variants, such as the relocation of the passage dealing with the "judge" Shamgar.

The Growth of a Marian Cult: The central proposition of the Great Stemma, that Luke's genealogy attaches to Mary, the mother of Christ, and that she was the child of the uncanonical couple Joachim and Anna, also necessitates a terminus post quem late in the second century. There is no attested mention of the source of this tradition, the uncanonical Protevangelium of James, before this time. The first patristic author who alludes to it, Clement of Alexandria, died in about 215, and indeed, an explicit citation from the Protevangelium is not found until the third century, in a work by Peter of Alexandria who died in 311. It is not until the fourth century that we find a datable Latin work with content from the Protevanglium: the hymn that is preserved in a papyrus codex which is now in Barcelona.

The motivation to write the Great Stemma can probably be seen in the beginnings of the Christian cult that venerated Mary. At the very least, one might argue that the Great Stemma's placement of the ancestors of Christ in a kind of roll of honour is consistent with growing fourth century interest in sainthood and the growing emphasis on saints in religious worship. More detail.

Dependency: The Great Stemma clearly draws on earlier chronographic work by early Christian writers. It has a marked affinity, in both form and content, with the tabulation of biblical history contained in the pages of the Chronological Canons, drawn up in Greek in about 300 by Eusebius of Caesarea, and translated into Latin by Jerome (Eusebius Sophronius Hieronymus) in about 380 (Jerome's text concludes in 378). Jerome's Latin is nowhere quoted in the Great Stemma, so it is possible that the Great Stemma was drawn up after consultation of the original Greek of Eusebius. More detail.

The final area to consider is that less substantial evidence which does not yield any date, but allows us to test the plausibility of any dating for negative evidence.

The roll form: The layout of the stemma itself strongly suggests that the original— and probably any direct copies from the author's manuscript— was not a document cut into sections and spread over several pages of a book, as we see it today, but was drafted as a single large chart. If so, that chart is less likely to have been drawn on parchment (which must be stitched together to make big sheets and tends to be impracticable as a roll material) and is more likely to have been penned on papryus (which is easier to manufacture directly in large sheets).

The choice by the author of a wide, horizontal canvas for the Great Stemma would be consistent with a date of creation in the period when papyrus rolls, designed to be read left to right, provided the low-cost, standard substrate for written works. The transition from the roll, which can easily be made 20 or 30 times wider than it is high, to the codex— a bound book of many pages laid over one another— began in the 1st century AD and was more or less complete by the end of the 4th century.[*]See for example Netz, 70. Roger Pearse has pointed out a recently rediscovered text by Galen in which that Greek author mentions losing his codices in library fires. For Eusebius of Caesarea, drafting his multi-column Chronological Canons in the 4th century, the page-by-page presentation had become the natural and easily transported form of presentation.[*]See for example Rosenberg, 26.

These considerations would be consistent with the Great Stemma, which in any case is less sophisticated than the Chronicle in its use of synchronisms between the different timelines, being in fact the older of these two works. The evident disappearance of the early models by the time that the 10th-century copies were made is also consistent with the fact that scrolls of papyrus are physically vulnerable and do not age well.

Metrics: The number of reproduction cycles required to achieve our earliest existing witnesses late in the first millennium of the Christian Era should also be considered. We have seen above one instance in which a copy had seemingly been destroyed by heavy use, neglect or fire when the notary Justo had the remnants in his hands in about 760. Whether they are bound into bibles, Beatuses or chronicles, all the witnesses derive from some single earlier archetype,which itself was clearly defective and was not the document drawn up by the author.

One supposes that apart from the author's version, several accurate copies could have been created in a first cycle of reproduction. In a second cycle, at least one faulty reproduction was made. By the time a third cycle of reproduction began, the original, the first-cycle versions and all but one of the second-cycle versions must have been unavailable for some reason, since all 19 versions in existence today draw on the single faulty reproduction. The high degree of corruption in the surviving 9th and 10th century orthography also implies transcription from very aged models containing damaged, faded or unfamiliar script.

The key fault in this lost intermediate archetype, the muddling of the Southern Kingdom royal wives, is discussed on a separate page of this website.

The overall degree of corruption suggests a timespan in the order of centuries between the original authorship and the scribal copying that took place in 943 or 945. The loss of author data for Great Stemma would also be consistent with an early origin: the document may have enjoyed the repute in early medieval Spain of being a document dating back to the patristic years of Christianity, even though editors lacked any record of its age and could no longer name the document's author.

Silence: The Great Stemma's silence about other religious intellectual currents cannot be used to establish any date: at most, such silence should be studied to test our dates for plausibility. The Stemma seems to take no stance on the great trinitarian controversies of the fourth century. It contains no evident references to the body of theological writing composed between 391 and 430 by Augustine of Hippo.

Back to History Table of Contents

Back to Macro-Typography on the Web