The Western Non-Interpolations

Contents: Introduction *The Major Western Non-Interpolations *Other Possible Western Non-Interpolations *Outside the Gospels


The textual theory of Westcott and Hort recognized four text-types --the Neutral, the Alexandrian (these two really being different phasesof the same type, and now generally called "Alexandrian"),the Syrian (what we call the Byzantine), and the Western.

Of these types, in their view, the Alexandrian is restrained, the"Western" is marked by extensive paraphrase and expansion,and the Byzantine is a smooth combination of the two.

It is a good rule of criticism that, when manuscripts go against theirtendencies, the significance of this reading is increased. So, for instance,when the "Western" text preserves a short reading, thatreading is more likely to be original than when it preserves a longerreading. This is the basis on which Hort isolated the "WesternNon-interpolations."

If Hort's theory is to be believed, the "Western Non-interpolations"are in fact places in which readings have been interpolated into the Neutraltext (and usually the Byzantine text as well). Although Hort usually rejects"Western" readings, in this case he regards them as original,placing the common reading of the Neutral text in double brackets, [[ ]].The non-interpolations are described in §240-242of Hort's Introduction [and] Appendix.

The "Western Non-interpolations" actually fall into two classes.The first are the full-fledged non-interpolations, of which there are nine(all placed in double brackets by Hort). All of these are supported byDea (Codex Bezae)and the Old Latins, and inall cases Hort regards the words as "superfluous, and in some casesintrinsically suspicious" (§240). The second class consists ofreadings which, due either to shifts in the manuscript evidence or todifferences in the way he assesses them, Hort regards as doubtful enoughto place in brackets but not to reject as clearly spurious.

The force of Hort's argument was so strong that for three-quartersof a century most editions and translations (including the RevisedStandard Version and the New English Bible) omitted these nine passages.Then P75was found (which included all the "non-interpolations"for which it was extant). Such was the respect for this manuscript thatthe passages began to re-assert their place in the editions -- notably inUBS/GNT and its follower the New Revised Standard Version.E. C. Colwell, however, in "Hort Redivivus: A Plea and a Program,"offers this assessment of the case:

[Aland] reverses Westcott and Hort on the Westernnon-interpolations because P75 disagrees with them in agreeingwith Codex Vaticanus. But there is nothing in that agreementthat is novel to Hort's theory. Hort did not possess P75, buthe imagined it. He insisted that there was a very early ancestorof his Neutral text, that the common ancestor of Vaticanus andSinaiticus was a remote ancestor, not a close ancestor. P75validates Hort's reconstruction of the history, but P75 doesnot add a new argument for or against that theory.

To put it another way, P75 -- despite its age -- is just anotherAlexandrian witness. Its existence does not alter the case thatthe "Western Non-interpolations" are just that. They arestill present in the Alexandrian text and missing in the "Western."The student may well feel that they belong in the text, but the existenceof P75 should not sway this decision.

The list below gives the nine full-fledged Non-interpolations; this isfollowed by a list of some of the more questionable interpolations. In eachcase the support for the shorter reading is listed. It is noteworthy thateight of the nine Non-interpolations are in Luke (and the remaining oneis not a true example of the form). If the Non-interpolations are notaccepted as original, their presence should offer strong evidence forthe theory that D is an edited text -- at least in Luke.

The Major Western Non-Interpolations

Other Possible Non-Interpolations

The following readings are omitted in certain authorities (especially the Latins)which may be considered "Western," and are placed in single brackets byWestcott & Hort as possible "Western Non-interpolations." As above,the support for the shorter reading is listed, as are lacunae in certainof the major "Western" witnesses (D, the Old Syriac, a b e k and sometimesothers of the Latins; recall that k contains Matthew and Mark only, so it is notmentioned for Luke or John).

Outside the Gospels

Westcott and Hort did not extend the concept of the "Non-interpolations"outside the Gospels. Such caution was probably justified in the case of Acts, wherethe text of Codex Bezae is extraordinarily wild. But the "Western" textof Paul (as represented by D F G Old Latin with some support from 629 Vulgate) ismuch more restrained. The possibility of such "non-interpolations" mustbe conceded. A few candidates are listed below (this list is not comprehensive,and includes weak as well as strong candidates. Most of these deserve to berejected, although at least two have very strong cases. The others I leave forthe reader to judge). I have listed only readings whichare at least two Greek words long and which do not have support from the majoruncial witnesses P46ℵ A B Cor from the Byzantine text. If B is omitted fromthis list, we find a few other candidates, e.g. Rom. 5:2, Eph. 6:1).