I have decided to switch the subject of the Saturday Series study for this Holy Saturday, on the eve of Resurrection Sunday (Easter). I thought it worth reprising the content of an earlier article that dealt with the key textual variants (variant readings) in the Passion and Resurrection narratives. A study of such variant readings is central to the Textual Criticism of the New Testament—the primary goal of which is to determine, as far as possible, the original form of the text. For those who are familiar with the Gospel narrative, but perhaps have never studied these variant readings, the questions and issues surrounding the text may come as something of a surprise.
We will proceed through the key variation units (and the textual variants) one by one.
Luke 22:19-21
The first variant I will explore comes from the Lukan version of the Last Supper. To begin with, it might be useful to look at the three Synoptic accounts side-by-side, along with Paul’s traditional account from 1 Corinthians (notable add/omit variants are in square brackets):
Note: Unless otherwise indicated, the translation is my own, quite literal in style. Parentheses indicate helping English words; a slash indicates two alternates for rendering the same word, for the sake of clarity. Italicized words are left untranslated, as there is no single English word quite appropriate in context. The noun diath¢¡k¢ (diaqh/kh) literally means something that is “set/placed through(out)”, i.e. an “arrangement”. It is often used in the context of a “disposition” or “testament” (such as a last will). In the New Testament, the usage follows the Greek version(s) of the Old Testament, where diath¢¡k¢ translates the Hebrew b§rî¾ (tyr!B=), a word which fundamentally refers to a binding agreement—especially in the religious-theological sense of the agreement established by God with the people of Israel (Abraham and his descendants).
| Mark 14:22-25 | Matthew 26:26-29 | Luke 22:14-23 | 1 Corinthians 11:23-26 |
| 22And (at) their eating, taking bread (and) blessing, he broke (it) and gave to them and said: “Take, this is my body.” 23And taking (the) cup (and) expressing gratitude, he gave (it) to them, and all drank out of it. 24And he said to them, “This is my blood, of the [new] diatheke, which is (being) poured out for (the sake of) many. 25Amen, I say to you that no, no longer shall I drink out of the produce of the vine until that day when I drink it new in the kingdom of God.” | 26And (at) their eating, taking bread (and) blessing, Jesus broke (it), and, giving to the learners, said: “Take (and) eat, this is my body.” 27And taking [the] cup and expressing gratitude, he gave (it) to them, saying: “Drink out of it, all (of you), 28for this is my blood, of the [new] diatheke, which is (being) poured out around/concerning many unto the release/forgiveness of sins. 29And I say to you, I shall not drink again from the produce of the vine until that day when it drink it new, with you, in the kingdom of my Father.” | 14And when the hour came to be, he fell/sat down (to eat), and the apostles with him. 15And he said to(ward) them, “With longing, I have longed to eat this pascha with you, before by suffering. 16For I say to you that I shall not eat it until that (time) it is fulfilled in the kingdom of God. 17And receiving/taking (the) cup (and) expressing gratitude, he said: “Take this and divide (it) unto yourselves; 18for I say to you [that] no[, no longer] shall I drink from the produce of the vine from now on, until the (time) when the kingdom of God comes.” 19And taking bread (and) expressing gratitude, he broke (it) and gave to them, saying: “This is my body [that is given for you: do this unto my remembrance.” 20And like(wise) the cup, with/after the dining, saying: “This the cup is the new diatheke in my blood, that is poured out for you.] 21But more—see, the hand of the one giving me over (is) with me upon the table. 22That the Son of Man indeed travels according to that which was marked-out/determined, but more—woe to that man by whom he is given over!” 23And they began to question toward themselves (as to) the one of them who perhaps it might be, the (one) about to do this. | 23For I took over from the Lord that which I also have given over to you: that the Lord Jesus, in the night that he was given over, took bread 24and, expressing gratitude, broke (it) and said: “This is My body which is [broken/given] for you. Do this unto my remembrance.” 25Like(wise) the cup, with/after the dining, saying: “This the cup is the new diatheke in my blood: do this, how often if you drink, unto my remembrance.” 26For, how often if you eat this bread and drink this cup, you announce the death of the Lord until the (time) when he comes. |
Note especially the orange highlighted text above, to demonstrate how close 1 Cor. 11:24-25 is to the disputed portion (vv. 19b-20) of Luke 22.
The textual tradition of Luke 22:17-20 is somewhat confused, as is indicated by the fact that six major variants are attested for this passage. The Metzger/UBS Textual Commentary on the Greek New Testament (second edition, pp. 148-150) provides a nice table summary, which I include here captured out of Biblesoft’s electronic version:
Actually, these six variants really can be reduced down to two: a long version, which includes vv. 19b-20, and a short version, which does not have the verses. The text-critical question then is: which of these is most likely the original reading? Was vv. 19b-20 added (an interpolation) by scribes at some point in the process of transmission, or were they deleted?
Interestingly, the manuscript evidence is overwhelmingly in favor of the long version, as it is the reading found in every Greek MS except one. It is found in the oldest relevant papyrus (Bodmer, Ë75) and the major uncials (a A B C K L T W, etc.) as well as most miniscules and ancient Versions (translations). It is decidedly the Majority reading, including the entire early Alexandrian Tradition. On the other hand, the short version is only found in Codex Bezae (D) and in five Old Latin manuscripts (a d ff2 i l). Didache chap. 9 might also be a witness to an original cup-bread sequence (i.e., the short version).
The superior external (manuscript) evidence would seem to clinch the decision in favor of the long version, were it not for the fact that no one has been able to provide a good explanation as to how the shorter text ever could have happened. It does not appear to be the result of (any obvious) scribal accident. Moreover, a scribe, puzzled by Luke’s cup-bread-cup sequence, would more likely have remove the first mention of the cup, rather than the second, and thereby bring the sequence into harmony with the other Gospels (see above). Beyond this, it is a general rule of textual criticism that, in a choice between two readings, the shorter version is more likely to be original (lectio brevior potior)—though there are exceptions, of course. The long version has sometimes been called the more difficult reading (which generally is to be preferred); but I tend to regard both, in their own way, equally difficult. I must confess, it is a bit hard to imagine a pious scribe deleting vv. 19b-20, with their vital soteriological content. On the other hand, it is a bit easier to imagine a scribe adding these verses, given their obvious similarity to 1 Cor. 11:24-25—such familiar verses could have quickly taken root in the manuscript tradition, to be forever preserved in the Majority text.
One of the strongest modern advocates for the short text has been Bart Erhman, who devotes a lengthy discussion to the question in his book The Orthodox Corruption of Scripture (Oxford: 1993, pp. 197-209). While I disagree with much of his view of Lukan theology, he makes some excellent points regarding this passage. Here I cite a diagram (p. 206) which shows, from his point of view, the natural structure and continuity of the shorter text:
(A) And taking bread, giving thanks, he broke it and gave it to them, saying, This is my body (v. 19a)
(B) But (pl¢¡n) behold, the hand of the one who betrays (tou paradidóntos) me is on the table (v. 21)
(A1) For (hóti, continuation!) the Son of Man goes as it was ordained for him (v. 22a);
(B1) But (pl¢¡n) woe to that man through whom he is betrayed (paradídontai) (v. 22b)
In the late 19th and early 20th century, influenced by Westcott and Hort’s analysis (this passage was one of their “Western non-Interpolations”), more scholars were willing to regard the short text as original; today, very few are willing to do so. Joseph Fitzmyer’s discussion of the issue in his classic 2-volume commentary (Anchor Bible 28A pp. 1386-1392) is as good as any. Fitzmyer, among others, brings out how Luke’s unusual cup-bread-cup sequence may simply preserve more of an original (historical) Passover context. There would have been (at least) three cups in the ceremony: a first cup (qiddûš) to sanctify the feast day, a second (hagg¹d¹h) following the liturgy, and a third (“cup of blessing”, kôs šel b§r¹k¹h) following the meal proper. In this scenario, the cup of vv. 17-18 could be the first or second cup, with the cup of the diatheke (‘new covenant’) in vv. 19b-20 would be the third. While it does not at all explain the omission of vv 19b-20 in the short version, this is a most attractive interpretation.
Overall, it is impossible to ignore the external (manuscript) evidence for the long reading, and I would tend to accept it as original. However, I do not regard it nearly as certain as many do today.
Mark 14:24; Matthew 26:28
A much smaller, related variant was noted in the table above. Quite a few manuscripts, in both passages, read t¢s kain¢s diath¢k¢s (“the new covenant/testament”) instead of t¢s diath¢k¢s (“the covenant/testament”). As in the case of Luke 22:19-20 above, it is important to note that a high percentage of substantive textual variants involve the question of harmonization between passages (especially in the Gospels). Scribes were prone, intentionally or unintentionally, to modify the text of a Gospel to match that of another (also to modify an Old Testament quotation to match that of the Seputagint, and so forth). As a result, in choosing between variant readings, the one which more closely harmonizes with another passage, typically is less likely to be original. In this instance, kain¢¡s (“new”) is probably not original, and is most likely a harmonization, either from Luke 22:20 or 1 Cor. 11:25. It is also worth noting that scribes (orthodox ones, at least) were more apt to add a significant soteriological or Christological detail than to remove it.
Luke 22:43-44
The next variant involves a famous detail of the prayer scene in the Garden; the passage is as follows (with the disputed portion in double-square brackets, according to the Nestle-Aland critical text [27th ed.]):
40genomenos de epi tou topou eipen autois: proseuchesthe m¢ eiselthein eis peirasmon. 41kai autos apespasth¢ ap’ autœn hœsei lithou bol¢n kai theis ta gonata pros¢ucheto 42legœn: pater, ei boulei parenengke touto to pot¢rion a)p’ emou: pl¢n m¢ to thel¢ma mou alla to son ginesthœ. [[43œphth¢ de autœ angelos ap’ ouranou enischuœn auton. 44kai genomenos en agœnia ektenesteron pros¢ucheto: kai egeneto ho hidrœs autou hœsei thromboi haimatos katabainontos epi t¢n g¢n.]] 45kai anastas apo t¢s proseuch¢s elthœn pros tous math¢tas heuren koimœmenous autous apo t¢s lup¢s, 46kai eipen autois: ti katheudete? anastantes proseuchesthe, hina m¢ eiselth¢te eis peirasmon.
40And coming to be upon the place, he said to them: “Pray not to enter into testing.” 41And he drew out from them like a stone’s throw (away), and setting (down) the knees he prayed, 42saying: “Father, if you wish, carry away this cup from me, but more—(let) not my will but yours come to be.” [[43And a Messenger from heaven was seen (by/unto) him, strengthening him. 44And coming to be in agony, more fervently he prayed: and his sweat came to be like thick-drops of blood going down upon the earth.]] 45And rising from the prayer, coming to(ward) the learners he found them sleeping from sorrow, 46and he said to them: “What, you are asleep? Stand up (and) pray not to come into testing.”
Unlike the case of vv. 19-20 (see above), in this instance the external (manuscript) evidence is evenly divided:
-
- Manuscripts Ë69 (apparently), Ë75, aa, A, B, N, R, T, W, 579, family 13 mss, etc., as well as a number of key early translations (Syriac, Coptic, Armenian, etc.) and a number of Church Fathers (such as Origen and Clement of Alexandria), do not include vv. 43-44. A number of additional manuscripts include the verses but mark them with asterisks as suspect.
- Manuscripts a*, D, K, L, X, G, D, 565, family 1 mss, etc., along with key translations (Syriac, Coptic, Latin, etc.), and a number of Church fathers, do include the verses.
To judge by some of the best/earliest Alexandrian manuscripts, a slight edge would be given to the shorter text, as well as on the basis of the principle lectio brevior potior (“the shorter reading is [generally] to be preferred”). However, it is hard to say which is the more difficult reading. Did scribes add the verses, perhaps to help combat “docetic” Christologies by emphasizing the suffering of Jesus? Or, did scribes delete the verses, because they seemed to give too much emphasis on the human suffering of Christ? It is always easier to explain how such variants were preserved in the manuscripts, than to explain how they first came about.
In any event, the change, whichever direction it occurred (add or omit), must have taken place before the end of the second-century, since late-second- and early-third-century witnesses attest both forms of the text. Vv. 43-44 clearly represent an ancient tradition — early Church Fathers like Justin Martyr (see the Dialogue with Trypho c. 103) cite it, though not specifically as coming from the Gospel of Luke.
On the whole, the text-critical evidence appears to be slightly in favor of the shorter reading. So cherished and familiar are vv. 43-44, however—and such a powerful ancient tradition—that even scholars who reject them as original still feel compelled to include them (bracketed, as in the Nestle-Aland text above) and to comment upon them.
Mark 14:68, 72
This is an interesting instance of a small, but notable seeming discrepancy between the Gospels. Only Mark mentions the rooster “giving voice” (crowing) twice—both in the prophecy (14:30), and here in these verses. However, here the textual evidence is a bit confused, almost certainly due to attempts to harmonize the account—but in which direction? Was kai alektœr ephœn¢sen (“and [the] rooster gave voice [i.e., sounded/crowed]”) added to the end of v. 68 (it is missing from a number of manuscripts) in order to fulfill (literally) Jesus’ prophecy by recording two crowings; or, was it deleted in order to harmonize with the other Gospels. The manuscript evidence is divided. Again, in v. 72, a number of manuscripts do not have the words ek deuterou (“from/for a second [time]”), and the same question can be asked.
Luke 23:34
For students unfamiliar with these text-critical questions, it may come as a bit of a surprise that a good number of early manuscripts (Ë75, ac, B, D*, W, Q, 0124, 579, 1241, and some Syriac and Coptic translations) do not include Jesus’ prayer of forgiveness: ho de I¢sous elegen: Pater, aphes autois, ou gar oidasin ti poiousin. (“And Yeshua said, ‘Father, release/forgive them, for they know not what they are doing.'”) This is a wide range of good (and geographically diverse) textual witnesses, including the earliest relevant Papyrus (Bodmer, Ë75). On the other hand, the majority text, including both family 1 & 13 MSS, and the entire later Koine text tradition, along with key early manuscripts (a*, C, Dc, L, G, D, 0117) and many early translations, include the text. Once again, the manuscript evidence is fairly evenly divided, perhaps with a slight edge to the shorter reading, though it is hard to say for certain. Was this an ancient (authentic) saying of Jesus that was inserted in this location by early scribes? I disagree with scholars who claim that it is easier to explain its omission than its insertion. Orthodox scribes, on the whole, appear to have been reluctant to delete Christologically significant sayings or details, and were more likely to add or preserve them. It is quite possible that, once the saying became embedded in the textual tradition (however this exactly came about), it was really too wonderful ever to be removed.
Mark 15:25 and John 19:14
In order to harmonize the chronology between the Synoptic gospels and John, a few manuscripts and versions, read hekt¢ (“sixth [hour]”), instead of trit¢ (“third [hour]”) at Mark 15:25; correspondingly, the opposite variant occurs in a number of manuscripts (ac Dsupp L X D Y, etc.) at John 19:14. The apparent chronological discrepancies between the Passion accounts of John and the Synoptics are notorious, and represents a long-standing, and widely discussed area of New Testament interpretation. All three Synoptics appear to record the Last Supper as a Passover meal; and yet John (in 19:14 and 31) explicitly notes that Jesus was crucified on the day of Passover eve (when the lambs were slaughtered). There have been a number of attempts to reconcile these discrepancies, and so preserve strict historical accuracy in all four Gospels; but these solutions, while possible, are not entirely convincing. A theory, popularized by A. Jaubert (La date de la Cène [1957]; The Date of the Last Supper [1965]; and other articles) posited that Jesus and his followers, as recorded in the Synoptics made use of an older [364-day] solar calendar (also utilized by the Qumran sect, and in the Book of Jubilees, etc.), while John records Jesus’ crucifixion on Passover eve according to the [“official”] Jewish lunar calendar. This theory was once in vogue, but has since fallen somewhat out of favor.
In any case, this textual variant would seem to be more of a simple attempt at harmonization. A reminder that, as we have seen, this sort of variant occurs quite often in the manuscript tradition.
Mark 15:28
Quite a few of the early manuscripts do not have verse 28: kai epl¢rœth¢ h¢ graph¢ h¢ legousa, kai meta anomœn elogisth¢ (“And the Writing was fulfilled which says, ‘and he was counted with the lawless'”). If it is indeed an interpolation, it was most likely added from the parallel in Luke 22:37. As I have noted above, scribes were prone to adding Christological details (such as the fulfilment of messianic prophecy), or to harmonizing the text to that of the other Gospels.
Luke 23:42
Here the variant involves but a single preposition (eis versus en) with an accompanying change in case. The “good thief” on the cross asks of Jesus:
Mn¢sth¢ti mou hotan elth¢s (“Remember me when you come…”)
-
-
- eis t¢n basileian sou. (“…into your kingdom.”)
- en t¢ basileia sou. (“…in/with your kingdom.”)
-
The first variant, which seems to refer to Jesus coming into the presence of his Father in heaven (after death), is the reading of MSS Ë75 (the oldest relevant Papyrus), B, L, and the Latin versions. The second variant, would appear to have an eschatological meaning (i.e., when Jesus comes [again] in [or along with] his kingdom), and is attested by the majority of Greek manuscripts (a, A, C2, R, W, Y, 0124, 0135, family 1 & 13 mss, and the later Koine/Byzantine text tradition). It is hard to say for certain, based on the manuscript evidence, which reading is more likely to be original. Jesus’ response seems to imply the first variant, but he may also be “correcting” the second variant—that is, the thief asks Jesus to remember him when he comes to set up his kingdom, but Jesus responds that the thief will be with him in paradise today.
John 19:29
An interesting detail: a few manuscripts read hyssœ (“[putting round] a pole”), instead of hyssœpœ (“[putting round] a hyssop [branch]”). The latter is almost certainly the correct reading, the former arising perhaps as a scribal accident. However, it may have been preserved in these few manuscripts because it seemed to make more sense in context. John may be bringing out an explicit connection with Passover (see Exodus 12:20). It is important to recognize that textual changes, especially in the relatively rare instances they are made intentionally, are typically not made out of a malicious intent – rather, they generally are the result of a pious regard for clarifying the text when its meaning seemed to them ambiguous or obscure.
Mark 15:34
Another interesting detail: a few ‘Western’ manuscripts (including D) read ho theos mou ho theos mou, eis ti œneidisas me? (“My God, my God, unto what [i.e. why] have you reproached me?”), instead of ho theos mou ho theos mou, eis ti engkatelipes me? (“My God, my God, unto what have you abandoned me?”). oneidízœ has the basic sense of “insult, disgrace”, also “revile, reproach”. It would seem that a scribe, perhaps not understanding how God could, or would, “abandon / leave behind” Jesus, may have intentionally modified the text. This touches upon the sensitive question of intentional alterations (whether orthodox or heretical/heterodox) to the text of Scripture. As indicated above, these are relatively rare occurrences – indeed, it is often hard to tell for sure whether a scribal change was intentional or accidental. But it is a real phenomenon in the textual tradition.
Matthew 27:49-50
A number of important manuscripts (a, B, C, L, al) include the words allos de labœn longch¢n enyxen autou t¢n pleuran, kai ex¢lthen hydœr kai haima (“and another, taking a spear, pricked his side, and water and blood came out”) – a detail otherwise known from the Gospel of John (19:34) – at the end of the verse. Oddly, in Matthew these appear prior to Jesus’ cry and death. Are these words, then, original, having been deleted because of their strange location? Or was it introduced from John, perhaps accidentally, by way of a marginal comment. The evidence would rather seem to be against the words being original here. Again, however, the addition of Christological details, such as the water and blood from his side, would have been tempting to scribes, especially when they harmonize with other familiar passages in the Gospels.
Luke 24:6
A few manuscripts, primarily ‘Western’ (D with at least seven Latin mss), do not include the words ouk estin hœde, alla ¢gerth¢ (“He is not here, he has risen”). The shorter text, despite the slight manuscript support, has been accepted as original by a number of scholars (past and present), as a so-called “Western non-Interpolation” (for more on this term, see the concluding note below). The general argument in favor of these shorter readings (besides lectio brevior potior – “the shorter reading is [generally] to be preferred”), is that there is no good reason to account for the words being deleted or omitted, whereas they could easily have been added to harmonize with Matthew/Mark. However, I think that, in this instance, the superior manuscript evidence is decisive – I would tend to regard the longer text as original.
Luke 24:12 and 40
Two more examples of possible interpolations: a number of ‘Western’ manuscripts (D, with Latin and Old Syriac mss) do not include v. 12: ho de Petros anastas edramen epi to mn¢meion kai parakypsas blepei ta othonia mona, kai ap¢lthen pros heauton thaumazœn to gegonos (“But Peter, standing up, ran upon/to the memorial/tomb, and, stretching out [to look], saw the linen-strips alone, and he went from [there], wondering to himself [about] what had come to be”); as well as v. 40: kai touto eipœn edeixen autois tas cheiras kai tous podas (“And, having said this, he showed to them the hands and the feet”). The external manuscript support for these verses is overwhelming, including all the earliest/best manuscripts (except D): Ë75, a, A, B, K, L, W, X, D, etc. Notwithstanding, some scholars consider the shorter reading in each case to be original (‘Western non-Interpolations’ – see below), the verses having been added from John (20:3-7 and 20). Modern detailed studies of both Gospels, however, have led scholars rather to the general conclusion that, here in the Passion and Resurrection narratives, in particular, Luke and John are both drawing upon common tradition. Especially in the case of v. 12, where the account is greatly truncated compared with that in John, a scribal insertion is far less likely.
LUKE 24:51
Here again, Western manuscripts (D, Old Latin mss), along with the Siniaitic Syriac MS and the original hand of a have the shorter reading, without the words kai anephereto eis ton ouranon (“and he was carried up into the heaven”). The vast majority of MSS, including all the early/best Greek MSS (Ë75, a [corrected], A, B, C, K, L, W, X, D, etc.) contain the words. The manuscript evidence would seem to be decidedly in favor of the longer reading, but internal considerations make it a bit less certain. In which direction did the change occur? It is certainly possible that scribes, noticing the apparent discrepancy between v. 51 and Luke’s own account of the Ascension in Acts 1:1-11, deleted the words. In the Gospel, it would seem that the Ascension takes place on the same night as the Resurrection, whereas in Acts (v. 3) it occurs 40 days later. On the other hand, a scribe may have inserted the words, in order to have the Gospel end with an account of the Ascension. Of course, this may well have been Luke’s intention as the author. I suspect that the apparent discrepancy may be the result of Luke compressing/conflating the narrative, thereby giving the impression that it all happened on one night. This sort of handling of historical narrative was quite common with ancient writers, as unsatisfying as it might be to our modern sensibilities. The longer reading, I think, is more likely original.
Final Note on “Western Non-Interpolations”
A number of the most significant variants discussed above involve the so-called “Western non-Interpolations”. This awkward term comes from Westcott and Hort in their greatly-influential late-nineteenth century critical edition of the New Testament. I will examine this interesting topic a bit further, perhaps, in a subsequent article. All it really means is that there are a number of key instances where the “Western” group of manuscripts (of which the Beza Codex [D] is the most prominent) has the shorter reading. This is especially significant, because the Western text (in Luke-Acts) typically is more expansive and usually has the longer variant reading. In a number of such instances, Westcott and Hort, followed by later scholars, accepted the shorter reading as original, even when the vast majority of manuscripts agree with the longer reading. The originality of these shorter readings is being increasingly rejected by critical scholars today, largely due to the presence of the longer readings in the early (Bodmer) Papyri.