Nov 14, 2010
Aland: TC Rules
Review: K. Aland, Text of the NT, 2nd ed. xlat. E.F.Rhodes, (Eerd., 1989)
Introduction: - Holmes on Eclecticism
Introduction to Contemporary Practice
Brief Summary of Rules by http://www.earlham.edu/
The Rule Against Conjectural Emendation
The textbook by Kurt Aland and Barbara Aland, The Text of the New Testament (trans. Erroll F. Rhodes; 2d ed.; Grand Rapids, Mich.: Eerdmans, 1989), 280-281, presents 12 basic rules for textual criticism. This post takes a look at the first one:
1. Only one reading can be original, however many variant readings there may be. Only in very rare instances does the tenacity of the New Testament tradition present an insoluble tie between two or more alternative readings.
Textual difficulties should not be solved by conjecture, or by positing glosses or interpolations, etc., where the textual tradition itself shows no break; such attempts amount to capitulation before the difficulties and are themselves violations of the text.
Variants caused by Author, Secondary Copies
"Basically, this rule asserts that the praxis of New Testament textual criticism lies in selecting one of the surviving variants as the original. I have no doubt that this is appropriate for the vast majority of variation units, but let us explore the situations in which this basic rule may require some nuance.
The first sentence, “[o]nly one reading can be original,” raises the issue of authorial variants (if “original” means that it owes it origin to the author). It is possible that the author himself could have generated some textual variants: there could be (marginal or in-line) corrections in the autograph, for example. If we had to pick one reading, which one should be chosen?
Often, in other fields of textual criticism, it is supposed that what the text critic ought to be interested in is the last authorial version of a text. But if this is the case, we should acknowledge that our usual canons for selecting readings may not work. The external evidence is irrelevant for authorial/autographic variant readings, and, since we expect authorial corrections to improve or supplement the author’s earlier draft, the canons of preferring the harder reading or the shorter reading do not seem appropriate.
A related issue is the question of multiple originals / autographs, each somewhat different. For example, what if Paul retained a copy of his letter to the Galatians? Which copy is considered the original? The one that was sent to the Galatians or the one that Paul retained? If both are “original,” how do we adjudicate between the inevitable differences in copying? These questions do not have easy answers, and the answers may well take us beyond the “basic” rules for textual criticism.
The flip-side of the denial that there are multiple original readings at a variation unit is the denial that none of the surviving readings at a variation unit are original.
This raises the issue of conjectural emendation, the positing of a reading that, although no longer attested among the surviving witnesses, it nonetheless provides the best explanation for both the internal evidence and the extant external evidence.
What makes the Alands’ rejection of conjectural emendation for the New Testament very interesting is that it completely the opposite attitude than that of the classical textual critics, who, for classical works, feel that conjectural emendation is not only necessary but the most important part of editing literary works from antiquity.
They assert on the contrary that conjectural emendation is necessary whenever the archetype (the most recent common ancestor of all the surviving witnesses) differs from the autograph, which they assume is true for nearly every ancient work of literature. For example, Paul Maas, Textual Criticism (Oxford, 1958), 11, holds:
If the archetype of a complete work proves to be entirely free of corruptions, it may be the original, i.e. the split in the tradition may have started with the original. I know no major classical work where this possibility is to be reckoned with, and in shorter works it would not bring us any further.
When Archetype NOT = Original: Paul
But is it reasonable to equate the archetype and the autograph for the New Testament? For Paul’s letters, for example, I’m not so sure. As far as I can tell, all of our surviving witnesses to Paul’s letters stem from a collection of Paul’s letters, perhaps compiled at the end of the first century. Since Paul did not compose all his letters at once in a single, integrated collection (the letters are occasional, after all), this archetypal collection cannot simply be equated to the autographs of the individual letters.
Indeed, for there to be no conjectural emendation in theory it would be necessary to assume that there are no differences in the text of the letters between those in the archetypal letter collection and those of the autograph. I don’t how to justify this a priori and appeals to the “tenacity of the New Testament tradition” seem to be too late to be pertinent.
Moreover, the problem of multiple originals for Paul’s letters also affects our assessment of the difference between the autograph and archetype. For example, suppose Paul sent a letter to the Galatians but retained his own copy. If there are differences, we would ideally prefer the form of the text that the Galatians themselves received. But, if the archetypal letter collection is derived from Paul’s retained copies, then any differences between the retained copy and the sent autograph would be reflected in the archetype of Galatians.
To assert a priori that conjectural emendation is completely unnecessary for Galatians is equivalent to assuming that there is no difference in the text of the letter Paul sent to the Galatians, in the text of his retained copy of that letter, and in the text of the letter as copied into the archetypal letter collection. I don’t know how to justify that assumption.
Other Conjectural Emendations
As for whether conjectural emendations are “themselves violations of the text,” I think we need to acknowledge that every (imperfect) copy violates the text of its exemplar and so all surviving witnesses in some way represent violations of the text. The issue with conjectural emendation is whether to privilege some ancient violations of the text over modern attempts to address the difficulties.
Indeed, some attempts to preserve ancient readings might necessitate violating the proper understanding of the Greek language. Some attempts to justify the reading in 1st Cor. 6:5, ἀνὰ μέσον τοῦ ἀδελφοῦ αὐτοῦ ( "between his brother") as good Greek require positing a unique construction in Greek for which there is no other corroborating evidence, as Jeffrey Kloha, “1 Corinthians 6:5: A Proposal,” NovT 46 (2004): 132-142, pointed out. (See my post, Emendation by Translation in 1 Cor 6:5 [July 12, 2010].)
What’s the greater violation: emending the text of 1 Cor 6:5 or emending the Greek language to permit the construction?? I might suggest that, since the evidentiary basis for our knowledge of the Greek language is so much greater than the textual tradition of 1 Corinthians, the greater violation lies in accepting the [probably] corrupted reading.
Despite these provisos, Alands’ basic rule here should work for the vast majority of cases. For some, hopefully very few, cases, however, the text critic is forced in deciding between multiple bad options. I think that conjectural emendation should be a bad option that, where appropriate, the textual critic must consider."
Faulty Premises (msg #3614)
"mr. Scrivener" posted the following additional critique on TC-Alternate-List:
I however would like to consider other problems with A&A's Rule 1:
Its presuppositions: This rule presumes that the original text is always preserved somewhere in the (Greek?) textual stream, and will be reasonably easily found and identified with certainty by applying their own rules.
The first premise (that the original reading still exists in extant documents), whether strictly true or not, is a good and probably unavoidable working basis, and will be likely true in any case for most accidental variants. But how could that apply to deliberate variants, which might have been expunged from a transmission stream?
The second premise (that A&A's own rules are adequate to the task of determining the text) is much more precarious a claim.
It is not even that the rule list not comprehensive (though it is quite incomplete).
Nor is the problem in the articulation (although they are wordy and sloppy).
No Hierarchical Order for Rules
The main problem, as with all previous "wishlists" of Canons, is that there is no hierarchical structure of priority or precedence, when the rules come in conflict. The solution to what rules or combination of same have the power to outweigh others, is left completely hanging.
It is just presumed that either the rules will be mutually exclusive in scope (i.e., always applying to different variation units), or else that it will be obvious when the time comes which rule will take precedence in case of conflict or confusion.
Particularly disturbing, is A&A's feeble attempt to address the obvious potential conflict between "internal" and "external" evidence or arguments. Here they suggest (in rule 3) that TC "must always begin" from the textual evidence, and only after this should "internal criteria" be considered.
What is disturbing is not just the idea that textual evidence always trumps internal evidence, but that A&A seem oblivious to observations that were made as far back as Michaelis!
(a) All textual criticism is "conjecture" of a reasoned kind and involves interpretation of textual evidence. The textual evidence can't interpret itself.
(b) A&A give no system or criteria for determining what arguments exactly are "external" and what are "internal". This dichotomy is extremely suspicious because of its lack of any definition.
(c) In practice, "external" evidence is always interpreted by "internal" evidence. Internal criteria (history, linguistics, style, content) are always used to evaluate textual evidence. Thus the two categories always have reversed precedence in actual practice.
(d) The naive suggestion of using the temporal order of operations (external first, internal second) is a fantasy. Once analysis is done (and mountains of TC work already exists), it can be looked at in any order. Scientists often present arguments and even historical discoveries and explanations in a different order than they happened, to make them appear more scientific and/or convincing.
Order of operations here has not been shown by A&A to make any difference in outcome, but if it does (and this is likely), this is even worse for TC practice. Who is then to say what the order of operations should be? This requires a theory, with an actual scientific basis. In other words, no scientific procedure has yet been proposed, and this is a necessity if one would claim to have a scientific method.
On this I have to agree with many of Nazaroo's complaints. What we are seeing is not yet science. Its still just a 'wishlist'.
2. Only the readings which best satisfies the requirements of both external and internal criteria can be original.
3. Criticism of the text must always begin from the evidence of the manuscript tradition and only afterward turn to a consideration of internal criteria.
Stephen Carlson comments:
"This rule shows the Alands’ emphasis on external evidence over internal, an emphasis which can be contrasted with that of Hort who sometimes turned to internal criteria to overturn “Neutral” readings (e.g., the so-called “Western Non-Interpolations”). One difference between Aland and Hort is that Hort did not have the benefit of the famous 20th-century papyrus discovery, though it must recognized that Hort’s theory had effectively predicted the existence of texts like P75.
Rule 3’s calling for examination of the external evidence first before considering internal critical is how I go about evaluating individual places of textual variants, but there is more going on underneath the surface. In particular, the rule is formulated in terms of the temporal order of consideration (“must begin … only afterward”), but there must be more to it than that.
Like a mathematical operation, the order of examination should be only significant only if the results would actually differ. For example, addition is commutative, so it makes no difference if you add 2 to 3 or 3 to 2: with either order the result. Yet, because addition is commutative it would make little pragmatic sense to insist on a particular order of addition. On the other hand, subtraction is not commutative, so 2-3 is not equal to 3-2. Because the Alands do specify an order, it must be the case that the external evidence somehow affects the outcomes, perhaps by constraining which variants are considered or how much weight a particular variant is to be given. The Alands go in more detail about the role of external evidence in Rule 6, so I will defer further discussion of that until then.
Another wrinkle is that determining which external evidence is good and weighty depends in part on internal evidence. The good manuscripts are those which have good internal readings. This means that the external and internal criteria are mutually interacting: one drives the other and vice versa.
I don’t think that the process must be or has actually been logically circular, but avoiding circular reasoning is something that careful textual critics always have to be mindful of."
Once again, a rule or Canon of criticism exposes itself as other than what it purports to be:
This is actually not a TC at base-level, but rather its a meta-rule about how to apply base-level rules: In this case, the meta-rule governs not just say the precedence or weight of two competing rules, but attempts to govern the whole set, with a presumption that they have already been previously sorted into "textual" and "internal".
Of course, to some degree the "canons" of NT TC have been previously sorted, into rules for each of the above categories, and even via Hort (but actually preceding him) a sub-grouping of Internal into "transcriptional" (scribal habits) and "intrinsic" (author-based).
While these rule categories appeared adequate in Hort's day (1882), they hardly do now. For instance, a whole new category of literary rules has sprung up around Source, Form, and Literary criticism. Many of these new categories of criteria have yet to be fully assimilated and integrated into NT TC.
As well as the inflood of new guidelines and observations regarding the NT text(s), the original problem remains:
The lack of any kind of organized hierarchy for existing canons and rules, for when they conflict. But before this problem even got off the ground in terms of solutions for rule-conflicts, a whole new realization has crept into the mix: The realization that any hierarchical system for sorting powers, precedence and priority is indeed affected by two dynamical aspects:
(1) The one exposed by Stephen Carlson above, namely that temporal order of operations affects outcomes; we'll return to this shortly, and
(2) That any hierarchy must actually be dynamic, not static, and have the flexibility to cope with a wide variety of phenomena and types of Variation Unit; this means that the "meta-rule" system cannot be 'fixed', but must adjust and accommodate changing circumstances like textual and linguistic phenomena, as well as changing scribal practices over extended time-periods.
All very well to see what is needed, but NT TC still drags its feet and is very reluctant to admit that the "emperor has no clothes" yet.
As well as basic canons, and meta-rules, we also need a General Procedure. This was recognized way back in the 1880s, as various early Textual Critics tried to articulate exactly what the procedure should be. Grandiose schemes were proposed, such as that of Milligan, as outlined here:
Milligan on TC - - Click here.
This issue, that of the overriding general methodology has never been resolved, in spite of great advances in many categories of research and sub-tasks. Thus we find four different methodologies in current practice, (see D. A. Black's overview here):
D.A. Black: 4 Methods -- Click here.
But also we have fresh minds attempting to offer yet more variants, for example Mr. Snapp's offerings here:
James Snapp Jr. on Eclecticism -- Click here.
But the real elephant in the room must be that after some 200 years, NT TC is still badly split over both the methodology to be used, and the resultant text (WH vs. Maj.). What is painfully obvious to all is that starting premises, and axioms, and especially one's view of the history of the textual transmission drastically affects the NT text.
One can say then that the need for a methodology that is truly independant of ideology and starting premises, that is deterministic, and scientific, is paramount. It is one of the big three problems facing TC today.
Why is it taking so long? Why has such little progress been made? The answer is found in the fact that the field has been ideologically dominated since its inception, even though specific ideologies and agendas have come and gone, like fads.
This has created an inertia, a stubbornness and resistance to challenge and change unprecedented in other historical fields. No surprise: its inextricably entrenched in popular religion and philosophy.
4. Internal criteria (the context of the passage, its style and vocabulary, the theological environment of the author, etc.) can never be the sole basis for a critical decision, especially when they stand in opposition to the external evidence.
Secondary Status of Versions and Patristic Evidence
..the 5th of the 12 basic rules of textual criticism in, K. Aland & B. Aland (Trans. E. F. Rhodes), The Text Of The New Testament: An Introduction To The Critical Editions And To The Theory And Practice Of Modern Textual Criticism, 1995 (2nd Revised Edition), William B. Eerdmans Publishing Company: Grand Rapids (Michigan), p. 280:
5. The primary authority for a critical textual decision lies with the Greek manuscript tradition, with the versions and Fathers serving no more than a supplementary and corroborative function, particularly in passages where their underlying Greek text cannot be reconstructed with absolute certainty.
6. Furthermore, manuscripts should be weighed, not counted, and the peculiar traits of each manuscript should be duly considered. However important the early papyri, or a particular uncial, or a minuscule may be, there is no single manuscript or group or manuscripts that can be followed mechanically, even though certain combinations of witnesses may deserve a greater degree of confidence than others. Rather, decisions in textual criticism must be worked out afresh, passage by passage (the local principle).
7. The principle that the original reading may be found in any single manuscript or version when it stands alone or nearly alone is only a theoretical possibility. Any form of eclecticism which accepts this principle will hardly succeed in establishing the original text of the New Testament; it will only confirm the view of the text which it presupposes.
8. The reconstruction of a stemma of readings for each variant (the genealogical principle) is an extremely important device, because the reading which can most easily explain the derivation of the other forms is itself most likely the original.
9. Variants must never be treated in isolation, but always considered in the context of the tradition. Otherwise there is too great a danger of reconstructing a "test tube text" which never existed at any time or place.
10. There is truth in the maxim: lectio difficilior lectio potior ("the more difficult reading is the more probable reading"). But this principle must not be taken too mechanically, with the most difficult reading (lectio difficilima) adopted as original simply because of its degree of difficulty.
11. The venerable maxim lectio brevior lectio potior ("the shorter reading is the more probable reading") is certainly right in many instances. But here again the principle cannot be applied mechanically.
12. A constantly maintained familiarity with New Testament manuscripts themselves is the best training for textual criticism. In textual criticism the pure theoretician has often done more harm than good.