The Western Text
This is not meant as an exhaustive study of either the Western Text-type or the mss that contain it, but rather as an introduction to the issues surrounding it. For more details, e.g. specific differences between the Western and other text-types, please see the links at the bottom of this page.
What is the Western Text type?
The Western text-type is one of several text-types used in textual criticism to describe and group the textual character of Greek New Testament manuscripts. It is the term given to the predominant form of the New Testament text witnessed in the Old Latin and Peshitta translations from the Greek; and also in quotations from certain 2nd and 3rd-century Christian writers, including Cyprian, Tertullian and Irenaeus. The Western text had a large number of characteristic features, which appeared in text of the Gospels, Book of Acts, and in Pauline epistles. The Catholic epistles and the Book of Revelation probably did not have a Western form of text. (Wikipedia)
The name ‘Western’ suggests that this text was predominately found in Italy, France (Gaul), and North Africa. However, although this may have been the region in which it originated, the name ‘Peripheral’ may perhaps better describe this text-type, as not only was it also used in the East (e.g. Syria and Armenia), but today we find it mostly on the outskirts (the periphery) of what was the Roman Empire, away from the greatest controlling influence of the Catholic Church. Despite the fact that today the Western text is found only in a small minority of mss, there is strong evidence that the reverse was once the case, with it having been used at least as early as the mid-2nd century, and found in the works of the church fathers Cyprian, Hippolytus, Irenaeus, Justin, Marcion, Origen, Tatian, and Tertullian, as, for example, indicated by Robert Waltz in The Encyclopedia of New Testament Textual Criticism:
The Latin text of the Adversus Hæreses [Irenaeus] gives its quotations in a distinctly "Western" form, perhaps most closely resembling the European Latin. Irenæus is one of the chief supports for the belief in the antiquity of the "Western" text…
… But all evidence seems to indicate that his [Marcion’s] text was highly interesting and very early (e.g. it clearly omitted the reference to Ephesus in Eph. 1:1). Readings associated with him seem to have been transmitted in the "Western," P46/B, and 1739 texts; they are rarer in the Alexandrian text (Compare Souter, who writes -- based on what we should note is incomplete evidence -- that "We find him in company with the Latin witnesses, especially the European Old-Latin MSS., but not infrequently also with the Old Syriac. He is never on the side of the great Greek uncials against both these versions.").
Waltz does then note some problems with the text of the fathers, but concludes:
Why, then, do we bother with such difficult sources of information? Because the Fathers, unlike manuscripts or versions, can be so precisely located. In most instances, we know with fair precision both where and when a particular author wrote. Thus, a judicious use of their testimony can allow us to localize particular readings and text-types.
In addition, many of the Fathers are early, and their texts predate all but our earliest continuous-text witnesses. They thus give us insight into a period where the history of the text would otherwise be completely dark. The earliest Greek witnesses to the "Western" text, for instance, date from the fifth century and after. The earliest Latin witnesses come from about the fourth. But in the quotations of Irenaeus, Tertullian, Cyprian, and others, we have fragments of "Western" texts going as far back as the second century.
The pattern of usage of the ‘Western’ text that we see fits a scenario in which the ‘Western’ text was in use very early, but as it spread outwards it was supplanted from the center by other forms of text (notably the Caesarean, Byzantine, and Alexandrian). According to Westcott and Hort:
On all accounts the Western text claims our attention first. The earliest readings which can be fixed chronologically belong to it. As far as we can judge from extant evidence, it was the widely spread text of Anti-Nicene times; and sooner or later every version directly or indirectly felt its influence… To what extent the earliest MSS of the distinctively Western ancestry already contained distinctive Western readings, cannot now be known. However they may have differed from the apostolic autographs, there was at all events no little subsequent and homogeneously progressive change.
What were the ‘distinctive Western readings’ mentioned above? Westcott and Hort described the characteristics of the Western text in this (much mis-quoted) passage:
The chief and most constant characteristic of the Western readings is a love of paraphrase. Words, clauses, and even whole sentences were changed, omitted and inserted with astonishing freedom, wherever it seemed that the meaning could be brought out with greater force and definiteness. They often exhibit a certain rapid vigour and fluency which can hardly be called a rebellion against the calm and reticent strength of the apostolic speech, for it is deeply influenced by it, but which, not less than a tamer spirit of textual correction, is apt to ignore pregnancy and balance of sense, and especially those meanings which are conveyed by exceptional choice or collocation of words.
In a page on the Western Text, biblicaltraining.org states that Westcott and Hort “also found a tendency to assimilate words and phrases found close to each other and, more seriously, through a process of harmonization to obliterate differences in similar or parallel passages,” and then continues:
This process of harmonization is of course most readily found in the gospels, where it was particularly easy through carelessness or particularly tempting in the interests of consistency to make passages conform to each other. It was in the writings of Luke, particularly in Acts, that the Western Text was found to diverge most from the other types.
In Western form Acts is nearly ten percent longer than what is commonly regarded to be the original text of that book. Given the connection between Luke and Acts, it might therefore be expected that the Western form of Luke would show a similar increase in length. Indeed, there are some places in which the Western form contains unique additional readings, but nevertheless the most well known characteristic of the Western form of the Gospels (and Luke in particular) is that it is shorter, and there are 27 places where there are strong indications that the shorter Western text could be original, usually referred to as the Western Non-Interpolations.
The Western Non-Interpolations
Towards the end of the 19th century Hort identified nine omissions in the Western text of the NT (one at Mt 27:49, one in Lk 22, and the remainder in Lk 24) that he considered to represent the original form of the text. They are referred to as ‘Non-Interpolations’ to indicate that he considered the majority of mss to have been interpolated at these points. Hort made his decision that these few shorter pieces of text were original on the basis that it was the opposite of the expansive tendencies of the mss in which they were found, and in the introduction to ‘The New Testament in the Original Greek’ he wrote:
They are all omissions, or, to speak more correctly, non-interpolations, of various length: that is to say, the original record has here, to the best of our belief, suffered interpolation in all the extant Non-Western texts. The almost universal tendency of transcribers to make their text as full as possible, and to eschew omissions, is amply exemplified in the New Testament. Omissions of genuine words and clauses in the Alexandrian and Syrian [Byzantine] texts are very rare, and always easy to explain. In the Western text, with which we are here concerned, they are bolder and more numerous, but still almost always capable of being traced to a desire of giving a clearer and more vigorous presentation of the sense. But hardly any of the omissions now in question can be so explained, none in a satisfactory manner. On the other hand the doubtful words are superfluous, and in some cases intrinsically suspicious, to say the least; while the motive for their insertion is usually obvious. With a single peculiar exception (Matt. xxvii 49), in which the extraneous words are omitted by the Syrian as well as by the Western text, the Western noninterpolations are confined to the last three chapters of St Luke.
Although this view remained popular for many decades, in more recent years Hort’s theory of the Western Non-Interpolations (now expanded to cover 27 passages) has increasingly fallen out of favor, not least due to the discovery of P75, dating back to the 2nd century, that contained these passages. Their fate was seemingly sealed by the United Bible Societies (UBS) 1971 revision of the Greek text of the NT, in which of the original nine omissions noted by Hort, only that at Mt 27:49 was omitted in their text. Bruce Metzger (one of the UBS committee members) wrote the following note explaining this decision regarding Hort’s theory:
… In recent decades this theory has been coming under more and more criticism. With the acquisition of the Bodmer Papyri, testimony for the Alexandrian type of text has been carried back from the fourth to the second century, and one can now observe how faithfully that text was copied and recopied between the stage represented by Papyrus 75 and the stage represented by codex Vaticanus. Furthermore, scholars have been critical of the apparently arbitrary way in which Westcott and Hort isolated nine passages for special treatment (enclosing them within double square brackets), whereas they did not give similar treatment to other readings that also are absent from Western witnesses [A reference to the 27 passages mentioned above].
… During the discussions a sharp difference of opinion emerged. According to the view of a minority of the Committee, apart from other arguments there is discernible in these passages a Christological-theological motivation that accounts for their having been added, while there is no clear reason that accounts for their having been omitted. Accordingly, if the passages are retained in the text at all, it was held that they should be enclosed within square brackets. On the other hand, the majority of the Committee, having evaluated the weight of the evidence differently, regarded the longer readings as part of the original text.
It is undeniable that the mss evidence for the omissions is thin. Not only do the great majority of mss contain the passages, but the majority of the support for the omissions is found in non-Greek mss. Against this evidence the ‘Christological-theological motivation’ referred to above simply did not weigh heavily enough. However, the discovery of P75 should not have had the effect that it did, since it is still undeniable that the Western text predominated in the 2nd century. In essence, on its own P75 just isn’t old enough to have made such a difference to the prevailing opinion.
As indicated above, the Western text (at least in the Greek) basically does not exist in the Catholic Epistles and Revelation, and even in the other NT books its character varies significantly, to the point where it is hard to define exactly what ‘Western text’ really means. For example, it is not even certain that the ‘Western’ text of Paul should even be considered as such:
The Western text of the Epistles of Paul - as witnessed in the Codex Claromontanus and uncials F and G - does not share the periphrastic tendencies of the Western text in the Gospels and Acts, and it is not clear whether they should be considered to share a single text-type. (Wikipedia)
There is also the distinction between the Western form of Acts, which is 10% longer than it is in canonical form, and the Gospels, which although in Western form include some additional text, also contain the Western Non-Interpolations, as follows (With Hort’s original nine highlighted – for more detail see The Encyclopedia of New Testament Textual Criticism):
Mk 2:22; 10:2; 14:39
Mt 6:15, 25; 9:34; 13:33; 21:44; 23:26; 27:49
Lk 5:39; 10:41-42; 12:19, 21, 39, 22:19b-20, 62; 24:3, 6, 9, 12, 36, 40, 51, 52
Jo 3:31; 4:9
The shorter readings in Luke 22, 24 are as follows:
vv. 22:19b-20 – “this do in remembrance of me. Likewise also the cup after supper, saying, This cup is the new testament in my blood, which is shed for you.” is not present;
v. 24:3 – “of the Lord Jesus” is not present;
v. 24:6 – “He is not here, but has been raised!” is not present;
v. 24:9 – “from the sepulchre” is not present;
v. 24:12 – The whole verse regarding Peter seeing the clothes but not the body is not present;
v. 24:36 – “and said to them, ‘Peace be with you.’” is not present;
v. 24:40 – “When he had said this, he showed them his hands and his feet.” is not present;
v. 24:51 – “and was taken up into heaven.” is not present;
v. 24:52 – “worshiped him and” is not present.
Metzger (above) refers to there being: “a Christological-theological motivation that accounts for [the Western Non-Interpolations] having been added,” and this can be most clearly discerned in the passages shown above at the end of Lk. In The Orthodox Corruption of Scripture Bart Ehrman remarks as follows:
Contrary to widespread opinion, there is no reason to take all the so-called Western non-interpolations en masse, as if they all stand or fall together… All the same, it is striking that a pattern appears to be emerging among the so-called non-interpolations considered to this point (Matt 27:49; Luke 22:19-20; 24:12): the corruption in each case represents an early interpolation (outside of the Western tradition) that works against a docetic form of Christology… Remarkably, the trend appears to continue in the variants I will now consider… [vv. 24:3, 6, 36]
What we have seen in the preceding deliberations is that a number of the readings that Hort isolated as Western non-interpolations evidence a theological Tendenz. Or to put the matter more accurately, the non-Western interpolations evidence this Tendenz; for in these cases, the “Western” text evidences no scribal tendency at all, but simply attests the original text that came to be corrupted in another stream of the tradition early on in the history of its transmission. Moreover, these secondary corruptions of which the Western tradition is innocent all work in the same direction: each functions to counter the docetic Christologies that can be dated to the time of their creation, the early to mid-second century.”
The Western text does not exist in the abstract. Instead, it is represented by a number of (predominately non-Greek) mss. However, this is not to say that all these texts agree, and the above statement from Westcott and Hort that: “Words, clauses, and even whole sentences were changed, omitted and inserted with astonishing freedom” is indicative of that. Nevertheless, at least in some areas there is significant consistency across the various witnesses. Ehrman again:
… Even today, however, with all the advances since Hort, no one can yet deny that the Greek witness codex Bezae, the Old Latin manuscripts, and (often) the Old Syriac tradition evidence wide-ranging agreements with one another and that these points of agreement can scarcely be explained except on the theory of relative antiquity…
The so-called Western witnesses occasionally attest readings found neither among the witnesses of the Alexandrian text nor among any other witnesses. In almost every instance these variants appear clearly secondary: they are harmonizations, explanatory additions, or paraphrases. But what about instances in which a Western witness is shorter and more difficult, where in fact the [longer] text attested elsewhere does not fit in its broader literary context and can be explained as a harmonization or an explanatory addition? Such Western readings cannot be discounted without further ado. They are, after all, evidenced in witnesses of the fourth and fifth century in Greek, Latin, and Syriac, so that if they did not originate in the autographs, they must have been generated in quite early times, at least by the end of the second century. Moreover, it should not be too quickly forgotten, as Hort’s detractors appear to do, that several of the papyri discovered subsequent to Hort’s investigations, despite their Egyptian provenance, do derive from just such a Western stream of tradition.
Just as the group of mss classified as Alexandrian do not agree in all instances, and may actually have significantly different readings in many places, so too the Western mss. Despite their differences, there exists (at least in the Gospels) a core set of distinctive, shorter, readings that not only support a docetic Christology, but appear to be earlier than what appear to be anti-docetic Alexandrian expansions. Consequently, perhaps the ‘Western Non-Interpolations’ should more accurately be termed the ‘Non-Western Interpolations.’
What are the Origins of the Western Text Type?
Apart from the characteristics of the textual differences between the Western Text-type and other (e.g. Alexandrian) Text-types, what makes the Western Text-type unusual is that it exists in very few Greek mss. It is found in the fragmentary papyri P37 (Mt), P69 (Lk), P29 (possibly), P38, and P48 (Ac), uncial 0171 (fragments of Mt and Lk), in codex Sinaiticus in Jn 1:1–8:38, in codex Washingtonensis (W/032) in Mk 1:1-5:30, and a small number of later Greek mss. In addition, it is found in the Greek/Latin diglot codices Bezae (Gospels and Acts), and Claromontanus, Augiensis, and Boernerianus (Paul).
However, the main mss support for the Western text-type comes from the Old Latin, which was widespread in Italy, Gaul (France), North Africa, and Egypt. Because the Western text-type has support in both Greek and Latin, we need to determine which came first. As the diglots contain both languages, an obvious place to start would be to compare the Greek with the Latin in each ms, to see which appears earlier. The problem is that none of the diglots contain a Greek text that is an exact translation of the Latin, or vice versa. Instead, in all cases the texts show signs that either one language was edited after having been translated from the other, or that they were both simply copied from earlier mss.
For example, it is likely that Augiensis (F/010, 9th century) was copied from a ms containing parallel Greek and Latin columns in which the Latin did not match the Greek (which has various missing passages), but instead was closer to the Latin text of the Vulgate, written at the end of the 4th century. In contrast, Boernerianus (G/012, also 9th century) was probably copied from a ms containing interlinear (alternate lines) Greek and Latin, in which the Greek is almost identical to that of Augiensis (even to having the same missing passages), and the Latin was nearly, but not quite, a strict translation of the Greek. In Claromontanus (D/06, 6th century) the Greek and Latin texts (on facing pages) are similar, but again neither is a direct translation of the other. It is sometimes assumed that the Latin pages were translated from an earlier Greek text that was similar to the Greek pages, but it could have happened the other way round – the Greek could have been translated from a Latin original.
Then there is Bezae (05), in which the Greek (D) and Latin (d) are close, but again neither of which is a direct translation of the other. While it appears that in some places the Latin has been altered to conform to the Greek, in other places the Greek has been altered to conform to the Latin. However, according to Frederic Henry Chase, the Greek of Bezae shows strong signs of being influenced by the Syriac. For example, he writes of Acts 1:2:
It seems clear that the Bezan Latin ('usque in eum diem quem susceptus est quo praecepit apostolis') is an awkward translation, and not the original, of the Bezan Greek. On the other hand the Syriac seems to offer an explanation of the variation in order from the common Greek text.
He then comments on the relationship between the Bezan Greek and Latin in Acts:
But a study of such passages as iv. 32, xi. 26, 27, xv. 29 shews that the Latin scribe had no clue to the meaning of an enigmatical Greek rendering of a Syriac gloss, and that therefore the formation of the Bezan Latin must be independent of, and later in time than, the formation of the Bezan Greek, and further, that we have no ground for thinking that the birthplace of the one is the birthplace of the other.
Waltz notes the following about the mss:
When discussing the Old Latin, of course, the great question regards the so-called "Western" text. The standard witnesses to this type are the great bilingual uncials (D/05 D/06 F/010 G/012; E/07 is bilingual but is not particularly "Western" and 629 has some "Western" readings but its Latin side is Vulgate). That there is kinship between the Latins and the "Western" witnesses is undeniable -- but it is also noteworthy that many of the most extravagant readings of Codex Bezae (e.g. its use of Matthew's genealogy of Jesus in Luke 3:23f; its insertion of Mark 1:45f after Luke 5:14) have no Latin support except d. Even the "Western Non-interpolations" at the end of Luke rarely command more than a bare majority of the Old Latins (usually a b e r1; occasionally ff2; rarely aur c f q).
He then suggests that Bezae is too late for it to represent the early Western text:
It is the author's opinion that the Old Latins, not Codex Bezae, should be treated as the basis of the "Western" text, as they are more numerous and show fewer signs of editorial action.
Not everyone agrees with this suggestion, and part of the issue is that, in comparison with the other text-types, the number of extant Western mss is small, with most being quite late and/or fragmentary. In addition to the differences noted above by Waltz, there are other passages found in Luke in Bezae that are not in other Western mss, the most well known of which is the following text between Lk 6:4 and 6:5:
On that day, seeing a certain man working on the Sabbath, he said to him, ‘Man, if you know what you are doing, blessed are you. But if not, cursed are you and a transgressor of the law’.
Another addition is the text “and having placed him there he positioned before the tomb a stone that scarcely twenty people could roll” in Lk 23:5. Also uniquely in Bezae we find that in the genealogy of Jesus in Lk 3:23-31 the first part follows Mt’s order, which suggests that these verses were missing in an ancestor of Bezae, and that a scribe used a ms of Mt to supply the missing text. However, despite these unique additions, the predominant characteristic of the Western text in the Old Latins is omissions or shorter readings, about which Willker makes the following observations:
The main problem is this: Internally the readings are very difficult to evaluate. In this commentary they have almost all been rated "indecisive". Overall, purely on internal considerations, there is often a slight tendency in favor of the shorter readings.
On the other hand, the shorter text is supported (for most of the readings) by D plus the old Latin only, without any other Greek support whatsoever. This is very strange. If original, this would point to a very early change of the text. It would mean that the archetype of the Western text departed earlier than all other texts from the main root, already quite heavily edited, but containing genuine readings not preserved anywhere else. This is of course basically possible and was the opinion of Hort.
The analysis of all variants in this commentary has shown that D has more secondary readings in Lk than any other Greek codex. D has thus to be considered very unreliable in general. To accept that exactly this codex preserved more than 10 important original readings alone in Lk is therefore rather improbable from the outset. It must be admitted that from this consideration it appears more probable that the shorter readings in D have been taken over from the Latin, as it happened elsewhere in the codex. They then never existed in Greek, but only in the Latin version.
J. Rendel Harris agrees with Willker that many readings in D originate in the Latin, ending Chapter 11 of his study of Bezae as follows:
And so we might continue our examination, but the results are sufficiently patent: we may say that the hypothesis of Latinization is shewn conclusively to be the right one for the explanation of the text, since so many readings of D are unsupported in Greek, while almost all are followed by the Latin. Next we see that occasionally whole battalions of later uncials take up the Latinized reading, while a small company remains faithful, usually including B…
The majority of the Latin texts (perhaps all of them) are derivable from a common source, their concurrence in singular errors being inexplicable on any other hypothesis, but whether this source be European or African, Gallican or Roman, remains as yet uncertain. And this being the case, and the authority of D having, for the greater part, been reduced to that of d, the practical problem is, to restore the lost Western text in its primitive Vulgar Latin form, and to reason from the single form thus reached, as being the equivalent of a very early Greek MS.
So extensively has the Greek text of Codex Bezae been modified by the process of Latinization that we can no longer regard D as a distinct authority apart from d. In the first instance it may have been such; or, on the other hand, it may have been the original from which the first Latin translation was made. But it is probably safest to regard D + d as representing a single bilingual tradition. The process of Latinization is not a late one consequent on the rapprochement in a bilingual codex of two texts, an old Western Greek and an old Western Latin respectively; for this bilingual tradition goes back to the earliest times. It can be traced in Irenaeus, in the ancestry of אCL, and in the parentage of the Egyptian versions. Any residual divergences between D and d are due to unequal criticism of correcting hands.
Harris here concludes that Bezae originates in one or more early Greek mss that were translated into Latin and began the family of old Latin documents that we see today, one branch of which led to d (the Latin side of Bezae), and by translation back to Greek, D. The importance of the old Latin, the Western Text-type, and in particular Bezae, is that as well as containing many unique readings due to problems in translation, D contains differences (mainly omissions) in Luke that are not matched by similar changes in other mss, but in many cases do match differences in Marcion's gospel.
If the Latin, not the Greek, is the key to the Western text-type, then it should be possible to trace the Latin back at least as far as the Greek, and Alexander Souter suggests that writings from Tertullian and Cyprian indicate that Latin translations of the New Testament date back to the mid-2nd century:
It is perfectly clear from references in Tertullian, who wrote at Carthage (mainly in Latin, but also in Greek) between A.D. 195 and 218, that Latin translations of at least some parts of Scripture existed in his time. Tertullian's regular practice was to use the Greek original and to translate for himself. But, in addition to his actual mention of existing Latin translations, it is clear that he sometimes used them himself. A study of his quotations by Monceaux has shown that he must have possessed translations of Luke, John, Galatians, First Corinthians, Romans and Ephesians. The existence of a (relatively) complete New Testament in Africa first comes into clear view in the writings of Cyprian (+ 258), who quotes a Latin Bible abundantly and accurately. The fact that on close study the translation used by him shows secondary characteristics confirms the conclusion that in Tertullian's time a Latin New Testament already existed in Africa, and suggests that it is the result of a long period of translation commenced not later than 150.
It seems clear that the Old Latin form of the Western text is old, but how old, and how did it originate? For any hypothesis regarding the origin of the Western text to be viable, it must account for (or at least allow for) the following:
Some passages in the Western text of Luke are longer than the non-Western texts, while others (e.g. in Lk 24) are significantly shorter;
In codices Bezae and Washingtonianus (and non-Western codices Monacensis and Tischendorfianus IV, and 0234) the gospels are in what is known as the ‘Western’ order: Mt, Jn, Lk, Mk.
Codex Washingtonianus is Western in Mk 1:1-5:30 only, with the rest of Mk and other gospels written using other text-types.
Codex Sinaiticus is Western in Jn 1:1–8:38 only, being close to Codex Bezae in these verses.
The Western text of Acts is approximately 10% longer than other forms;
Although an apparently Western text of the Pauline epistles exists, it is far less ‘wild’ than in the gospels or Acts, and it has been suggested that it is in fact a different text-type.
The text in 1 Corinthians usually found as vv. 14:34-35 is placed after v. 40 in the Western Greek, but not in the Western Latin.
Although some traces of Western text of the Catholic Epistles exist in the Old Latin, there is no evidence that it ever existed in Greek;
There is no evidence of the Western text of Revelation in Greek, although it is present in a few Latin, Syriac, Coptic, and Armenian mss.
These differences between the Western text in the Greek and Old Latin in the various New Testament books essentially rule out any overall ‘common process’ that applied to them all. Instead, they suggest that the Gospels, Acts, Paulines (possibly not the Catholic Epistles and Revelation) were translated/edited individually several times by different people, leading to the existence of multiple different translations, as noted by Jerome in his ‘Preface to the Vulgate Version of the New Testament’ at the end of the 4th century:
For if we are to pin our faith to the Latin texts, it is for our opponents to tell us which; for there are almost as many forms of texts as there are copies. If, on the other hand, we are to glean the truth from a comparison of many, why not go back to the original Greek and correct the mistakes introduced by inaccurate translators, and the blundering alterations of confident but ignorant critics, and, further, all that has been inserted or changed by copyists more asleep than awake?
The evidence indicates that there never was an early Old Latin NT. Instead, individual books were translated into Old Latin by different people in different places, with different levels of skill, and with some (the Catholic Epistles and Revelation) only apparently translated later in order to complete an Old Latin NT, perhaps as a reaction to Jerome’s creation of the Vulgate.
Conclusion
The usual supposition is that the Western text (as found in the Old Latin mss) was created in order to meet the needs of early Latin speaking Christians, and a reasonable assumption is that the Old Latin mss contain translations of the text found in non-Western Greek mss. The question that then arises is which Old Latin ms(s) are most likely to contain the original text.
As there is essentially no evidence for the existence of a non-western Old Latin text, it appears that the Western text-type itself was created as part of the translation process, rather than as changes to a previous faithful translation of the Greek. In addition, as there is no evidence of an early Western Greek form of the NT, it appears that this is largely a translation from a existing Western text, although not always the Old Latin. In the Encyclopedia of New Testament Textual Criticism Waltz comments as follows on codex Bezae:
So the final question is, is Bezae a proper witness to this text which underlies the Latin versions? Here it seems to me the correct answer is probably no. To this extent, the Alands are right. Bezae has too many singular readings, too many variants which are not found in a plurality of the Latin witnesses. It probably has been edited (at least in Luke and Acts; this is where the most extreme readings occur). If this is true (and it must be admitted that the question is still open), then it has important logical consequences: It means that the Greek text of Bezae (with all its assimilations to the Latin) is not reliable as a source of readings. If D has a reading not supported by another Greek witness, the possibility cannot be excluded that it is an assimilation to the Latin, or the result of editorial work.
The previous discussion supports this comment, and it most likely also applies to the other diglots. In none of these mss was one of the languages translated directly from the text of the other, but instead the two texts appear to have been copied independently from separate exemplars, although either the Latin or the Greek may also have been at least to some degree conformed to the other.
This makes it likely that it is the text of the early Old Latin mss that most closely represents the original Old Latin form of the Western text, and that the diglots are in that respect ‘second best.’ Unfortunately, as the early Old Latin mss are so fragmentary then we still have to rely on the diglots for much of the text, even though their texts (both Latin and Greek) appear to have complicated histories that are unlikely to ever be fully understood.
References
Aland, Kurt and Aland, Barbara: The Text of the New Testament: An Introduction to the Critical Editions and to the Theory and Practice of Modern Textual Criticism, 1987
Chase, Frederic Henry: The Old Syriac Element in the Text of Codex Bezae
Ehrman, Bart D: The Orthodox Corruption of Scripture : The Effect of Early Christological Controversies on the Text of the New Testament ...
Head, Peter M: Acts and the Problem of its Texts, 1993
Klijn A. J: A Survey of The Researches Into the Western Text of the Gospels and Acts, part 2, 1949-1969
Marlowe, Michael: Hort's Theory of 'Western Non-Interpolations' and its Influence on English Versions of the New Testament
Metzger, Bruce M: A Textual Commentary on the Greek New Testament, 1975, pp. 259-72
Richards, W.L: The Present Status of Text Critical Studies in the Catholic Epistles, 1975
Smith, Ben C: Significant Textual Variants
Snapp, James, Jr: Equitable Eclecticism: The Future of New Testament Textual Criticism
Souter, Alexander: The Text and Canon of the New Testament, 1913
Waltz, Robert B: The Encyclopedia of New Testament Textual Criticism, 2013, pp. 875, 1274
Westcott, B.F. and Hort, F.J.A: The New Testament in the original Greek, the text revised by B.F. Westcott and F.J.A. Hort, 1881
Willker, Wieland: A Textual Commentary on the Greek Gospels - Western non-interpolations
Wilson, J.M: The Western Text of the Acts of the Apostles, Introduction. pp. 1-37, 1923
Witherington, Ben, III: Revelation, 2003