My research, based on all the collations of the Greek Psalter that have been available in Göttingen and have now been completely digitised as part of the Academy project “Die Editio critica maior des griechischen Psalters”, has led to an extremely important insight, the consequences of which are certainly of inestimable value for the constitution of the Greek Psalter text: A careful analysis of the collation results has shown that, contrary to previous assumptions, an Origenic recension (O-group) is detectable in the primary tradition of the Greek Psalter.

Until now, it was assumed that the Origenic recension or O-text,[1] apart from a few Greek fragments of the hexaplaric tradition, had only been preserved indirectly in the Latin tradition of the Psalterium Gallicanum. However, I have now discovered that two important witnesses of the Catena Palestinensis seem to have preserved the Origenic recension more or less directly. I first presented my research results in March 2023 at a conference in Vienna organised by Uta Heil and Sebastiano Panteghini.[2] The publication is currently being prepared. At this point I would only like to summarise my most important insights, because they are in my opinion of essential importance for further editorial work on the Greek Psalter.

The current state of research indicates that the ancient Palestinian catena — which is primarily based on the commentaries of Origen, Eusebius, Didymus, Apollinaris and Theodoret — has been transmitted in two editions, one in tripartite and one in bipartite form. An overview of the tradition, which is divided into a direct and an indirect tradition, can be found in the Clavis Patrum Graecorum under the siglum C12.[3] According to this overview, which is based on Karo-Lietzmann, the direct tradition comprises two classes or types:

• Psalms 1–50 of this type are classified as Catena Typus VI = CPG C19.[4]
• Psalms 51–150 of this type are classified as Catena Typus XI = CPG C24.[5]

The tripartite edition of the Catena Palestinensis has survived only for Ps 1–100 (i.e. the first two volumes) and the bipartite edition only for Ps 77–150 (i.e. the last volume). The following Rahlfs Mss. belong to this catena form: [6] 1021-1113-1121-1270, 1209-1675-1756-1906.

The table indicates which psalms are contained in each of the manuscripts:

The most important witness to the tripartite edition for part one is manuscript Ra 1121 (Cod. Oxoniensis, Bodleian Library, Barocci 235, late 9th century). The most important textual witness of the tripartite edition for part two is manuscript Ra 1209 (Codex Taurinensis, Biblioteca Nazionale Universitaria di Torino, C. II. 6, 10th century). The most important textual witnesses of the bipartite edition, in the case of which only the second part has survived, are Ra 1675 (Cod. Ambrosianus F 126 sup., 12th century) and Ra 1756 (Cod. Patmiacus, Ἰωάννου τοῦ Θεολόγου, 215, 13th century).

In my study, I selected sample passages from the intersection of the bipartite and tripartite tradition, i.e. from Psalms 77–100. In doing so, I was guided by two questions: First, whether the textual witnesses of the Catena Palestinensis form a separate group within the catena manuscripts. Secondly, whether the biblical text of the Catena Palestinensis has its own textual character.

First, I investigated the question of whether the group 1209-1675-1756-(1906)[7] could be isolated within the catena manuscripts. The result was clear. Ra 1209 testifies to a non-Lucianic text. This was evident from the passages where Ra 1209 stood against L. Often this manuscript was accompanied by the two witnesses of the bipartite Catena Palestinensis, namely Ra 1675 and Ra 1756, but these partly deviated and went with the Lucianic text, thus showing Lucianic influence. Other manuscripts joined the attestation sporadically.[8]

These observations were followed by a follow-up question: Does the biblical text of the Catena Palestinensis have a textual character of its own? The close relationship between the biblical text of the Catena Palestinensis and the Origenian tradition was methodically demonstrated by comparing the biblical text of the Catena Palestinensis and the Septuagint column of the Origenian tradition in Rahlfs-Ms. 1098. Indeed, characteristic readings could be found that testified to a convergence of the Catena Palestinensis with Ra 1098(ο′). The fact that the representatives of the Palestinian catena stood out supported the idea that the Catena Palestinensis represented its own type of biblical text. Could this initial assumption be substantiated in more detail?

In a further step, I therefore examined the textual character of the Catena Palestinensis in the case of Psalms 1–50. Four manuscripts have preserved the first volume of the tripartite Palestinian catena of the Psalter: Ra 1021, 1113, 1121, 1270. Manuscript Ra 1121 is the oldest (late 9th century) and most important witness that offers Psalm 1–50 in its entirety. Thus, the relationship of Ra 1121 to the other three manuscripts, i.e. to Ra 1021 (s. X), Ra 1113 (s. XI) and Ra 1270 (s. X) remained to be determined. Ra 1270 proved to be entirely Lucianic, while Ra 1113 was largely and Ra 1021 partly shown to be under Lucianic influence. An extended comparison with Ra 1098(ο′) made it clear that Ra 1121 always goes together with Ra 1098(ο′), and in parts even with the Psalterium Gallicanum.

To sum up: The tripartite edition of the Catena Palestinensis transmits a biblical text that is of high text-critical value. For the area of Psalms 1–50, this evidence is placed on a somewhat broader basis with a total of four witnesses (Ra 1021, 1113, 1121, 1270), whereby Ra 1121 stands out from the others. For the area of Psalms 51–100, only Ra 1209 is to be mentioned as a witness. However, in comparison with the representatives of the bipartite Catena Palestinensis, which has survived only for the second part, i.e. Psalms 77–150, this shows that the tripartite edition bears witness to a special text which, unlike all the other Catena manuscripts of the Psalter, is not influenced by the Lucianic tradition. A further comparison with the Septuagint column of the hexaplaric fragments of Ra 1098 has proved extremely revealing. It illustrates that both Ra 1121 and Ra 1209 have preserved a biblical text that is very close to the text of the Origenic recension.

I therefore conclude that an Origenic recension resp. O-text is traceable in the primary tradition of the Greek Psalter, which I call the O-group and whose main witnesses are the Mercati Psalter (Ra 1098, ο′-column), which admittedly preserved only parts of the Psalter in the Origenic recension, alongside the Psalterium Gallicanum (and the Vulgate) as well as the sources already mentioned by Rahlfs (Ra 2005, Jerome), and above all the two outstanding witnesses of the Catena Palestinensis, i.e. Ra 1121 and Ra 1209. I believe that the text-critical significance of these two catena manuscripts for the constitution of the Greek Psalter text can hardly be overestimated. Thus, the following notation has emerged, which will be used as a working hypothesis in the further editorial work: O = 1098(ο′)-1121-1209-Ga.