Aland on Von Soden

With the support of Elise Konig, a wealthy patron who provided him with the necessary funds (and who deserves special recognition - may her tribe increase!), von Soden was able to enlist the efforts of about 40 colleagues in all for the task of collating MSS in the libraries of Europe and other lands. Under such favorable circumstances the 4-volume work appeared in 1902-1913 under the title Die Schriften des Neuen Testaments in ihrer altesten erreichbaren Textgestalt hergestellt auf Grund ihrer Textgeschichte (i.e., "The Writings of the NT, restored to their earliest attainable form on the basis of their textual history").

But this attempt must be adjudged "a failure, though a splendid one," to borrow the words from an outstanding American textual critic Kirsopp Lake with regard to t Westcott-Hort's edition in 1904. There are several reasons for this:

1. Its theoretical suppositions were false. Von Soden assumed that there were three major text-types: the K (Koine) text, the H (Hesychian, Egyptian) text, and the I (Jerusalem) text. According to von Soden these three recensions were derived from the archetype, the original text, which was still available to Origen. Therefore wherever all three text-types, or at least two of them, were in agreement, von Soden believed they preserved the original text of the NT. (cf. the brief summary of principles of reconstruction, 2:xxviii), provided it had not been distorted by the influence of Tatian (in the Gospels) or of Marcion (in the Pauline corpus). Apart from the fact that von Soden grossly exaggerates the influence of Tatian (for whose text his information was quite inadequate), the I-Text he described never existed, at least not as he describes it. Von Soden ascribed to I nearly all the MSS and traditions which did not conform to the K-Text or the H-Text.

Further, by placing the K-Text on a par with the other groups he ascribed to it a value it does not deserve. if the K-Text is given full vote in determining the original text, and if the agreement of two groups is decisive, the result must inevitably be a text that could never be called "original" (imagine the combination of K + I - a secondary and fictitious text-type!). It would necessarily be weighted in the direction of the Koine text [TR] - as actually happens in von Soden's text.

2. The system of symbols designed by von Soden makes his edition almost impossible to use. It could possibly have workd had he cited MSS individually instead of citing them usually by the various subgroups he established (or posited - nearly 30 for the K and I text-types), first identifying the subgroup and then listing individually only the members of the subgroup which do not support the reading! It is demanding enough to keep the varieties of subgroups clearly in mind, but it would tax all the powers of mnemonics expert to remember the members of all the various groups. Consequently the reader always needs to refer to the supplementary manuals, and these are usually either incomplete or awkward to use.

3. THe information in von Soden's apparatus is so unreliable that the reader soon comes to regard this remarkably full apparatus as little more than a collection of variant readings whose attestation needs verification elsewhere.

Von Soden's edition was distinctly a failure. ...But nonetheless, von Soden's edition is a necessary tool for textual critics. Part I is a vast quarry of information that is unavailable elsewhere. Von Soden's studies in the Koine text constitute pioneering research: the groups he distinguished, including even the congeries of groups he assembled under the lable of the I-Text, provide extenseive evidence for the establishment of MSS groups and families. Many of the spectacular discoveries made in recent decades may be found anticipated in von Soden's work, sometimes in a well-developed form. The text volume with its wealth of variant readings is also a useful source of information (despite its distribution into two, and frequently three, apparatuses).

- K. Aland,
Text of the NT
(Eerdmans, 1987,1995) p. 22 fwd

Von Soden's edition, its significance, and its limitations, has already been discussed (p.22). Anyone meeting it for the first time in the sample page (plate 12) will be somewhat surprised, first by the division of the apparatus into three parts, and (despite being forewarned) by its symbols and its system of designating MSS. The triple division of the apparatus has both advantages and disadvantages. The variant readings for every verse have to be looked for in three different places (and then reassembled in their proper sequence), although it is usually adequate to examine only the first two apparatuses - but it is this "usually" that raises the problem. When a pericope or a particular passage is under examination, all the variant readings in the tradition are of interest.

Von Soden's system can be explained only by its origins and its basic structural pattern, but for practical purposes even a regular user will always need some supplementary reference works (von Soden's notes on his format and apparatus in the preface to the text in volume 2 are quite inadequate, although anyone using the edition should certainly read them; they illustrate its limitations quite clearly).

To make things easier, Friedrich Kruger published in 1927 a Schlussel zu von Sodens Schriften des Neuen Testaments. Benedikt Kraft, Die Zeichen fur die wichtigeren Handschriften des griechischen Neuen Testaments, had already appeared a year earlier (1927, 3rd ed. 1955).

Both of these aids arrange the MSS according to von Soden's groupings and show the corresponding Gregory numbers, but both are incomplete and require the supplementary use of Sigla Index II (Correlating von Soden and Gregory) in the above-mentioned Kurzgefasste Liste (ANTF 1 1963:334-349). This index gives the symbols in numerical order (and also in von Soden's original order) to facilitate rapid reference. The intricacy of von Soden's system of symbols is due to his Germanic idealism in attempting to combine in each manuscript's symbol a description of its contents, its date, and its textual character.

Von Soden indicated the contents of a MS by prefixing a Greek letter to its number:
δ (διαθηκη "NT") for a MS containing the whole NT,
ε (ευαγγελιον "Evangelion") for the 4 gospels, and
α (αποστολος "Apostles") for the other NT books.

Thus there are three series in which von Soden arranged his known MSS. Unfortunately Revelation was often lacking in the MSS von Soden classified as δ (complete NTs), and the MSS of the 2nd part of the NT (α) were quite varied in their content, mostly representing either just the (Acts and Catholic letters) or the Pauline letters.

Von Soden also wished to include the date of a MS in its symbol: the third numeral from the end was assigned this function; e.g., δ150 for a MS of the 11th century, δ250 for the 12th cent, δ350 for the 13th century,...and δ50 to δ99 for a MS of the 10th century.

The plan was ingenious, but it had certain weaknesses and blind spots: the period before the 10th century was not accommodated (von Soden assigned to it the numbers δ1 to δ49), but even more awkward was the lack of a firm date for most of the miniscules. What would happen if a MS were assigned a new date? To preserve the system it would have to be assigned a new symbol (which is precisely what von Soden did on occasion).

Further, the system as observed for the δ-manuscripts could not be applied without adaptation to ε-manuscripts. There were too many ε-manuscripts from before the 10th century to be accommodated by the numbers 1 - 99, so that after that von Soden continued with ε01 to ε099. MSS of the 10th century began with ε1000 to ε1099, after which the 11th cent. MSS returned to the pattern with ε100.

Finally, it should also be mentioned that von Soden treated commentary MSS as a special group, indicating their authors by capital letters:

A for Antiochene commentaries (Chrysostom on Matt., Victor of Antioch on Mark, Titus of Bostra on Luke), and
A ^πρ for Andreas the Presbyter's commentary on the Apostolos,
Aν for Andreas of Caesarea's commentary on Revelation,
Aρ for Arethas' commentary on Revelation.

When the number of commentary MSS was sufficiently large, these were further distinguished by their century on the pattern of the δ, ε, α MSS, and if an author such as Theophylact commented on several books, these were distinguished, e.g., Θ^ε for his commentary on the gospels, Θ^π for his commentary on the Pauline letters, and then there is also Θδ for Theodoret's commentary on Paul.

In addition to all the complexities already posed by the δ, ε, α system - quite frequently with 4-figure numbers - the alphabetical prefixes require constant vigilance. The slightest slip could lead to utter confusion.

[Aland here gives two examples of deciphering the 1st part of the apparatus]

We have dealt with only two variants in the first apparatus from the sample page. There is no need for a similar treatment of the 18 variants in the 2nd apparatus, nto to mention the 54 variants in the 3rd, because these two examples should suffice as representative for the reader. The reader ...may undertake transcribing ...other variants...using the aids mentioned above. THe difficulties will be only too apparent, and even if they are successfully surmounted ...there is still no assurance that the results will be reliable because of ...inaccuracies...

Furthermore the miniscules cited are by no means all of real significance ...because so many are of secondary value."

- K. Aland,
Text of the NT
(Eerdmans, 1987,1995) p. 40-43