The Septuagint of 2 Samuel

Objectives

The Academy of Finland Research Fellow project The Septuagint of 2 Samuel (PI: Tuukka Kauhanen) aims at producing a critical edition of the Greek Septuagint version of 2 Samuel for the series Septuaginta: Vetus Testamentum Graecum of the Göttingen Academy of Sciences. The project involves text-critical and text-historical study.

The main objective of the project is to produce a critical edition of the Septuagint of 2 Samuel for the Göttingen Septuagint series. The secondary objective is to carry out research on the textual history of 2 Samuel—the Hebrew and Greek texts and the secondary versions—and the impact of the Septuagint on a broader field of studies on the Books of Samuel-Kings.

In 2019–2023, there will be an Academy of Finland funded project The Septuagint and Its Ancient Versions that complements the research.

See a sample of the edition: II Regnorum Sample 2018.pdf

The edition includes:

  1. A critically established Greek text that presents the closest possible approximation to the text that the translator produced.
  2. Apparatus I, that lists all the meaningful variant readings in the Greek manuscript traditions, the significant readings from Latin and other secondary versions, as well as a selection of readings from indirect sources such as quotations from 2 Samuel by Greek and Latin patristic authors. The apparatus follows the well-established format of the Göttingen editions.
  3. Apparatus II, that lists extant Hexaplaric readings derived from later Jewish Greek versions (Aquila, Symmachus, Theodotion). These readings can be found in the margins of the manuscripts and noteworthy indirect witnesses such as the Syrohexapla and ancient commentaries.
  4. An Introduction that explains the format of the edition and outlines the textual history.
  5. Appendices providing information of orthography and other details not noted in the apparatus.

For the text-historical research, the key questions are:

  1. What are the underlying causes for the textual variation in the witnesses?
  2. How are the witnesses divided? How should they be arranged into groups and families?
  3. What is the contribution of the secondary versions and quotations to the textual history?
  4. How the remains of the Hexapla help to recognize early Hebraizing corrections in the Septuagint witnesses?

It is expected that the new critical text of 2 Samuel will differ from the current Rahlfs text by 1% in chapters 1–9 (so-called non-kaige section) and by up to 4% in chs. 10–24 (kaige-section). The new critical text will allow for a more refined reconstruction of the underlying Hebrew Vorlage to be used in the textual criticism of the Hebrew 2 Samuel. In addition, it can be expected that the research included in the editing process will refine the view on the textual history of the Greek 2 Samuel and, by extension, of all the Books of Samuel-Kings. The expected results can be formulated as the following hypotheses:

  • There are occasional kaige-type Hebraizing corrections in codex Vaticanus (B) even in the non-kaige section, perhaps once per 100 words. By contrast, the amount of the kaige readings in the kaige section may be somewhat lower than generally supposed (3–4 per 100 words).
  • In the witnesses of 2 Samuel there are less Hexaplaric readings than in 1 Samuel and for the most part they are confined to three manuscripts (codex Alexandrinus [A] and group O). The Lucianic manuscript group (L) especially has fewer Hexaplaric readings in 2 Samuel than in 1 Samuel.
  • The nature of the L group does not change between the non-kaige and kaige The stylistic revision is clearly visible throughout the book and these Lucianic stylistic variants can be found 10–12 times per 100 words.
  • Clear copying errors can be expected to be found roughly as often as Lucianic variants. These are found in almost every witness.
  • There is a distinct category of secondary readings that can be termed free copying. This phenomenon has not been treated much in textual criticism of the Septuagint, but it is recognized in the manuscripts of the Greek New Testament. Unlike unconscious errors, the amount of free copying varies greatly between different lines of transmission. Counting all the details, a variant that results from free copying can be found as often as for every three words of text.

Three scientific breakthroughs can be expected:

  • Reconstruction of the closest-to-original (or “oldest attainable”) text in the kaige section of 2 Samuel (2 Sam 10/11–24). Such a reconstruction of a kaige section has not been previously attempted. The reconstruction requires choosing between the B and L texts in every variation unit. In addition, it can be expected that a conjectural emendation has to be made 10–20 times in cases where the original reading has been lost under two recensions.
  • The application of a diagnostic approach in the text-critical analysis. The approach is applicable to the textual criticism of most texts and it is highly adaptable.
  • An electronic tool S2S WebApp, including a database and web applications, that stores and handles the text-critical data and allows its use both for printing and online. The database structure is general enough to work with any kind of text that can be subjected to textual criticism.

The scientific impact includes:

  • A critical edition of the Greek 2 Samuel is a prerequisite for text-critical, translation-technical, as well as lexicographical study of the book.
  • The ongoing project for producing a critical edition of the Hebrew Bible has to take into account the reconstructed Hebrew Vorlage of the Septuagint. That, in turn, requires a critically established Greek text.
  • By spearheading in the use of databases in philological research the project has the potential of spreading the best data management practices within the field of Humanities.
  • The diagnostic approach utilized in the evaluation of the secondary readings can be adapted to the study of other ancient texts. The approach can be expanded to categorize all types of changes in texts and traditions.
  • The results of the research on the secondary versions and patristic quotations from 2 Samuel can be used in the study of the languages of the versions and in patristic studies.

Diagnostic Approach

Establishing the critical text requires an analysis of the cause for the secondariness of every reading that is rejected from the critical text and placed in the apparatus. The diagnostic approach that is being developed by the PI aims at defining tangible criteria for recognizing the reasons for second­a­riness or, to employ diagnostic terms, conditions. The variant reading—the “patient”—should exhi­bit certain symptoms that are brought about by an underlying condition. Thus the focus is on the se­cond­ary readings, but it should be emphasized that secondariness is always secondariness in respect to the original reading. Criteria for originality, not only for secondariness, have to be taken into account.

The conditions can be tentatively divided into seven categories.

1. Kaige readings are few in the non-kaige section (ch. 1–9) and not very numerous even in the kaige section (10–24). They often concern lexical choices and offer stereotypical renderings of the Hebrew words. Kaige readings are often attested by B along with some or most other manuscripts.

2. The Greek 2 Samuel is expected to contain only a small number of Hexaplaric readings. These readings are typically quantitative differences that agree with the MT and they are mostly found in A and O.

3. The Lucianic recensional readings are most often stylistically or logically better readings. They are attested by L, now and then joined by an handful of other manuscripts. In 2 Samuel L is not likely to preserve a significant number of Hexaplaric readings. It can be estimated that about one in six secondary readings can be attributed to the Lucianic reviser.

Errors can be divided into those that (4.) are not likely to happen independently in several witnesses, i.e., monogenetic, and those that (5.) can happen independently (polygenetic). This distinction is important for text-historical research: while a close relationship between the witnesses is easiest to recognize by the common errors they share, not all errors are equally important from that point of view. Monogenetic errors are such in which there is not a clear transcriptional or other reason for the error, but the variant is not likely to be a deliberate change either. Polygenetic errors are often of the type that results in a loss of a part of the text, esp. homoioteleuton errors. They can happen any time in any witness. Together the two categories of errors explain about every fifth secondary reading.

The remaining reasons for secondariness go under free copying and, like errors, they can be divided into (6.) monogenetic and (7.) polygenetic. These two are the most usual causes for secondariness and together they explain almost 60% of all variation. Free copying means changing the text consciously or, at least, half-consciously: they are changes that do not fit the recensional patterns (kaige, Hexaplaric, Lucianic) but cannot be labelled as unconscious errors either. Free copying appears to have taken place in all lines of manuscript transmission, but their large number can be misleading: often a free variant is found only in a single witness. Accordingly, only a few lines of transmission, such as O, b, and 107´, have a considerable number of these; and some, such as M and L, are very resistant to free copying. Monogenetic free readings tend to be additions while polygenetic free readings tend to be omissions: adding requires more effort than omitting, and thus small clarifiying additions are less likely to happen independently in several lines of transmission.

The following table provides the estimates for the relative occurrence of the seven types of secondariness along with examples and brief notes.

Secondariness Estimation, % (by section) Example Notes
non-kaige kaige
kaige 2 5 17:25 τὴν  στρατιάν L 554mg] της δυναμεως rel = Ra צבא-δύναμις is a typical kaige rendering
Hexaplaric 2 5 1:4 πεπτώκασιν πολλοι] tr A O corresponds to MT
Lucianic 17 15 1:2 καὶ ἐγένετο] εγενετο δε L L favors the better Gk expression
error, monogen. 12 12 1:3 ἐγώ] λεγων 121 68´ transcr. error, but for no obvious reason
error, polygen. 9 8 1:2 om λαοῦ 44 245 transcriptional error (τοῦ λαοῦ)
free copying, monogenetic 27 26 1:2 αὐτόν] + προθυμως 707 free addition which is not really needed
free copying, polygenetic 31 29 1:6 τὸ ἀπάγγελλον αὐτῷ] > O 509 106 381 71 not transcriptional, but the omitted notion is not needed

Combining Textual History and Textual Criticism

The methodology of diagnosing the conditions is tightly related to the over-all view of the textual history: these two points of view affect each other. This relationship can be illustrated with a circle. (The items are numbered in a logical order; the working order is not necessarily the same.)

  1. The view on textual history defines what kind of conditions there are in the first place. It also includes a crude eva­luation of the witnesses: e.g., kaige readings are expected to be found mostly in B and Lucianic readings in L.
  2. Therefore, recognizing symptoms does not happen purely without any pre­con­ceptions. For example, any type of Hebraizing correction is ex­pect­ed to correspond more or less to the MT and an error is ex­pect­ed to manifest itself in an odd read­ing. More­over, any reading in the L group is more easily recognized as a Lucianic reading while a similar reading in the b group, largely unaffected by Luci­anic read­ings, will probably be judged as free copying.
  3. The evaluation of the witnesses is refined on the basis of recognizing the symptoms that actually are there. The witnesses that have a large amount of symptoms carry less weight when evaluating the attestation to a putative Old Greek reading. By contrast, witnesses with a small number of symptoms have a greater value when establishing the critical text. The nature of the symptoms, on the other hand, allows to define certain witnesses as, e.g., more Lucianic or less erroneous than the others. This evaluation is in a key role in dividing the manuscripts and other witnesses (versions and patristic sources) into groups and families.
  4. The process continues with a better definition for the symptoms. A certain symptom, e.g., an agreement in a minus with the Hebrew text, can be symptom of an error or of free copying, but it most often goes together with Hexaplaric revision. Defining symptoms by attestation means that if a witness appears to contain a large amount of certain type of symptoms, these are more easily defined as belonging to the condition that the said witness is mostly affected by. E.g., it is probable that new kaige readings will be recognized in B and thus more kaige symptoms can be defined. This, in turn, refines the view on the nature of the kaige revison, leading to a better understanding of what kind of symptoms it produces.

While the sectors 1 and 3 in the circle have more to do with textual history, and sectors 2 and 4 are more about the methodology, none of them can be done in isolation: the process is a flux or a continuum rather than a step-by-step procedure. The circle illustration is chosen to emphasize that neglecting any of the sectors distorts the whole.

See also The Septuagint and Its Ancient Versions.