Responding to Thomas Holland's "Manuscript Evidences" regarding Revelation 16:5

Dr. Thomas Holland wrote the following regarding Revelation 16:5 (source)(cited by Khoo and from there by Moorman):

*** Start of Thomas Holland  *** 

Revelation 16:5:

And I heard the angel of the waters say, Thou art righteous, O Lord, which art, and wast, and shalt be, because thou hast judged thus.

The question arises concerning the Trinitarian phrase, "which art, and wast, and shalt be." Modern versions read, "which is, and was, the Holy One." Dr. Edward Hills has correctly cited passage as a conjectural emendation (Hills, 208). Bruce Metzger defines this term as,

The classical method of textual criticism . . . If the only reading, or each of several variant readings, which the documents of a text supply is impossible or incomprehensible, the editor's only remaining resource is to conjecture what the original reading must have been. A typical emendation involves the removal of an anomaly. It must not be overlooked, however, that though some anomalies are the result of corruption in the transmission of the text, other anomalies may have been either intended or tolerated by the author himself. Before resorting to conjectural emendation, therefore, the critic must be so thoroughly acquainted with the style and thought of his author that he cannot but judge a certain anomaly to be foreign to the author's intention. (Metzger, The Text Of The New Testament, 182.)

From this, we learn the following.

1). Conjectural emendation is a classical method of textual criticism often used in every translation or Greek text when there is question about the authenticity of a particular passage of scripture.

2). There should be more than one variant in the passage in question.

3). The variant in question contextually should fit and should be in agreement with the style of the writer.

Such is the case here. First of all, to change the Trinitarian phraseology (which is used in Revelation 1:4, 8; 4:3; and 11:17) does break the sense of the passage and is inconsistent with the phrase used elsewhere by John. Furthermore, the addition of "Holy One" is awkward and is repetitive of the use of the phrase "Thou art righteous, O Lord."

Secondly, there are some textual variances among the changes made. The Greek text of Beza reads, "o wn, kai o hn, kai o esomenoV" (who is, and was, and shall be). It should be pointed out that among the Greek manuscripts the reading is different. Most of them read, "o wn, kai o hn, o osioV" (who is, and was, the Holy one). The oldest Greek text of Revelation containing this passage, which is P47, has a textual variant. This Greek text reads, "o wn kai, o hn, kai osioV" (who is, and was, and Holy one). It is interesting to note that while the actual manuscript itself uses both kai and osioV, and that only the word osioV will fit, the text is rather worn here leaving the other words in the passage mostly unscathed.

Thirdly, P47 is not the only Greek text which is worn here. In fact, while P47 is slightly worn, the Greek text which Beza used was greatly worn. This is so noted by Beza himself in his footnote on Revelation 16:5 as he gives reason for his conjectural emendation:

"And shall be": The usual publication is "holy one," which shows a division, contrary to the whole phrase which is foolish, distorting what is put forth in scripture. The Vulgate, however, whether it is articulately correct or not, is not proper in making the change to "holy," since a section (of the text) has worn away the part after "and," which would be absolutely necessary in connecting "righteous" and "holy one." But with John there remains a completeness where the name of Jehovah (the Lord) is used, just as we have said before, 1:4; he always uses the three closely together, therefore it is certainly "and shall be," for why would he pass over it in this place? And so without doubting the genuine writing in this ancient manuscript, I faithfully restored in the good book what was certainly there, "shall be." So why not truthfully, with good reason, write "which is to come" as before in four other places, namely 1:4 and 8; likewise in 4:3 and 11:17, because the point is the just Christ shall come away from there and bring them into being: in this way he will in fact appear setting in judgment and exercising his just and eternal decrees. (Theodore Beza, Nouum Sive Nouum Foedus Iesu Christi, 1589. Translated into English from the Latin footnote.)

In addition to the Greek manuscript witnesses (which in this passage are few, as we have already noted), early translations should be considered. Again, the weight of the evidence falls on the side of "holy" and not "and shall be." Most translations, such as the Latin, omit the "and" using only "holy" (the Latin word is "sanctus"). Primasius, Bishop of Hadrumetum, wrote a commentary on Revelation around 552 AD and used the Latin word "pius" instead of "sanctus." They mean the same, but it does reveal yet another variance in the text. This, of course, brings us to yet another group of witnesses: Patristic citations.

Two things should be stated before continuing. One, as confirmed by Jerome, there were a number of various Latin editions of the New Testament which differed in both translation and content before and around 405 AD (when Jerome finished his Vulgate). Most of these we do not have. Two, as pointed out by Dr. John Wordsworth (who edited and footnoted a three volume critical edition of the New Testament in Latin) the like phrase in Revelation 1:4 "which is, and which was, and which is to come;" sometimes is rendered in Latin as "qui est et qui fuisti et futurus es" instead of the Vulgate "qui est et qui erat et qui uenturus est." (John Wordsworth, Nouum Testamentum Latine, vol.3, 422 and 424.)

Wordsworth also points out that in Revelation 16:5, Beatus of Liebana (who compiled a commentary on the book of Revelation) uses the Latin phrase "qui fuisti et futures es." This gives some additional evidence for the Greek reading by Beza (although he apparently drew his conclusion for other reasons). Beatus compiled his commentary in 786 AD. Furthermore, Beatus was not writing his own commentary. Instead he was making a compilation and thus preserving the work of Tyconius, who wrote his commentary on Revelation around 380 AD (Aland and Aland, 211 and 216. Altaner, 437. Wordsword, 533.). So, it would seem that as early as 786, and possibly even as early as 380, their was an Old Latin text which read as Beza's Greek text does.

It should be noted that none of the early English versions, nor the foreign translations, read as does the Authorized Version. However, they do not read as most modern versions do either. Instead they read somewhere in between using both the "and" with "holy." The New King James Version follows the reading of the Authorized Version.

New American Standard Version:

And I heard the angel of the waters saying, "Righteous art Thou, who art and who wast, O Holy One, because Thou didst judge these things.

The Great Bible:

And I herde an Angell saye: Lorde, whych arte and wast, thou arte ryghteous and holy, because thou hast geven soche judgementes.

Bishops' Bible:

And I heard the Angel of the waters say, Lorde, which art, and was, thou art righteous and that holy one, because thou hast given such judgements:

Luther's German Bible:

Und ich horte den Angel der Wasser sagen: herr, du bist gerecht, der da ist und der da war, und heilig, dab du solches geurteilt hast

Italian Bible:

Ed io udii L'angelo delle acque, che diceva: Tu sei giusto, O Signore, che sei, e che eri, che sei il Santo, d'aver fatti questi giudicii.

New King James Version:

And I heard the angel of the waters saying: 'You are righteous, O Lord, The One who is and who was and who is to be, Because You have judged these things.

*** End of Thomas Holland  ***  

This article (the same substance) was posted as early as March 6, 2001 (by Jody Adair, apparently on behalf of Dr. Thomas Holland, ThD) at the now-defunct (or perhaps re-homed to a different URL?) pure words website.  I assume that Dr. Holland is actually the contributor's name (not that it matters), but it is interesting to note that it is also the name of one of the King James translators (link to wiki bio).

1. Trinitarian Phrase

Holland's assertion that the phrase in Revelation 16:5 is a "Trinitarian phrase" is a mistake.  There is no argument to support the assertion, and treating the phrase as "Trinitarian" seems problematic.  If all that was meant is that "the Being" and the "Having Been" and the "Coming" are names or titles of God, who is Triune.   

2. Conjectural Emendation

Holland adopts and approves of Edward Hills identification of the reading "and shalt be" as a conjectural emendation (as discussed here).  Indeed, Holland offers a translation (apparently his own - at least I could not track it down to any earlier source than his of 2001) that suggests that Beza saw a manuscript with part of the text worn away and filled in the gap himself.

3. The Metzger quotation

The full quotation from p. 182 of Metzger's second edition of The Text of the New Testament, has:

The classical method of textual criticism regularly involves, as was mentioned at the beginning of this chapter, the exercise of conjectural emendation. If the only reading, or each of several variant readings, which the documents of a text supply is impossible or incomprehensible, the editor's only remaining resource is to conjecture what the original reading must have been.

A typical emendation involves the removal of an anomaly. It must not be overlooked, however, that though some anomalies are the result of corruption in the transmission of the text, other anomalies may have been either intended or tolerated by the author himself. (FN1) Before resorting to conjectural emendation, therefore, the critic must be so thoroughly acquainted with the style and thought of his author that he cannot but judge a certain anomaly to be foreign to the author's intention.

FN1: For a discussion of the paradoxical possibility of a textual critic's 'improving' on the original, see G. Zuntz's article on I Cor. vi. 5 entitled 'The Critic Correcting the Author', Philologus, xcix (1955), pp. 295-303.

I provide the full statement just to avoid any concern about the ellipses presented in the article, not to suggest a misquotation.  However, let's consider each of the lessons drawn therefrom, namely:

1). Conjectural emendation is a classical method of textual criticism often used in every translation or Greek text when there is question about the authenticity of a particular passage of scripture.

Textual criticism is not a translational technique and it is not limited to Greek (or other foreign language) texts.  It is not limited to Scripture.  In fact, on p. 183, Metzger provides an example of conjectural emendation with respect to an odd line in Shakespeare's play, Henry V.  

Moreover, the issue is less one of authenticity (as such) and more about originality.  Moreover, it is not applied in "every" case that such a question arises.  However, it can arise in many cases.  It's particularly common when a text is represented only by a single copy.  In such a case there are seldom variants (although there could be, via corrections of the text itself).  However, there may be passages that appear to be wrong, and in need of correction.

2). There should be more than one variant in the passage in question.

Actually, no.  The number of variants is not the standard.  Note that Metzger first mentions the case of "the only reading" before mentioning "each of the several variants."

3). The variant in question contextually should fit and should be in agreement with the style of the writer.

It's unclear what Holland means here.  Rather than "variant in question," Holland probably meant "conjecture in question."  If so, he would be correct.  Metzger is saying that conjectural emendation is the art of removing anomalies.  To do so, the reading (or all the variant readings if there is more than one reading) must be evaluated as to whether they represent an anomaly.

Holland continues:

Such is the case here. First of all, to change the Trinitarian phraseology (which is used in Revelation 1:4, 8; 4:3; and 11:17) does break the sense of the passage and is inconsistent with the phrase used elsewhere by John. Furthermore, the addition of "Holy One" is awkward and is repetitive of the use of the phrase "Thou art righteous, O Lord."

As mentioned above, the phraseology is not "Trinitarian." To this we must add that the phraseology at 4:8 (surely 4:3 is a typo), is different from that at Revelation 1:4, and 1:8 in that the first two elements are in the opposite positions.  The phraseology at Revelation 11:17, like that of Revelation 16:5 omits "the coming one" in the majority of Greek manuscripts.  Thus, the phrasing of Revelation 16:5 is not as inconsistent as Holland suggests.  Moreover, Beza's proposed fix is to use yet a different phrasing, thereby maintaining inconsistency with the previous instances, even if the TR reading of Revelation 11:17 were accepted against the majority of manuscripts.

Referring to Jesus as Holy (e.g. "O Holy One" or "the Holy One") is more harmonious than it is being credited - it is name given to him multiple times in the Old and New Testaments and it is a description uniquely applied to him in Revelation 15.  

Admittedly, it is unexpected in Greek to use an adjective as a title.  This, no doubt, accounts for the variants that we see, such as variants treating it as a coordinate adjective with "righteous."  However, to suggest that it is repetitive is a mistake.  Consider:

Psalm 145:17 The LORD is righteous in all his ways, and holy in all his works.

Is the word "holy" objectionably repetitive in that context? Certainly not.  So also, if it were the case that it were a coordinate adjective with "righteous" (as suggested by Beza), then it would simply be taken as a poetic reiteration for emphasis.

Similarly, we do not expect to hear Holland object to the King James Version's phrasing:

Revelation 22:11 He that is unjust, let him be unjust still: and he which is filthy, let him be filthy still: and he that is righteous, let him be righteous still: and he that is holy, let him be holy still.

Any anomaly suggested on this basis must have very little weight.

Holland continues:

Secondly, there are some textual variances among the changes made. The Greek text of Beza reads, "o wn, kai o hn, kai o esomenoV" (who is, and was, and shall be). It should be pointed out that among the Greek manuscripts the reading is different. Most of them read, "o wn, kai o hn, o osioV" (who is, and was, the Holy one). The oldest Greek text of Revelation containing this passage, which is P47, has a textual variant. This Greek text reads, "o wn kai, o hn, kai osioV" (who is, and was, and Holy one). It is interesting to note that while the actual manuscript itself uses both kai and osioV, and that only the word osioV will fit, the text is rather worn here leaving the other words in the passage mostly unscathed.

I assume that the "V" oddity arose from the use of a special font for Greek that treated a capital V as a terminal sigma.  That was a more popular way of writing Greek before Unicode Greek was in widespread use on the Internet.

The correct term is "textual variants," although I have heard folks even on "my side" say "variances."  

The note about the variants involves some mistranslation of the Greek corresponding to the variants.  Beza's text is more literally "the being [one], and the [one who] has been, and the shall being [one]." The majority text is more literally "the being [one] and the [one who] has been, O Holy [One]."  The P47 variant is more literally "You are righteous (the being [one] and the [one who] has been) and holy" - I've included more context for clarity.  This is the coordinate adjectives variant that I mentioned above.  Several letters of the word "osios" are worn in P47, as shown below, but it certainly is the word written in P47.

(source)

Holland continues:

Thirdly, P47 is not the only Greek text which is worn here. In fact, while P47 is slightly worn, the Greek text which Beza used was greatly worn. This is so noted by Beza himself in his footnote on Revelation 16:5 as he gives reason for his conjectural emendation:

This sentence is based on the wildly wrong translation Holland offers of Beza's annotation.  Contrary to Holland's mistranslation, the correct translation (see here for more context) is this:

5 And who [you] will be, καὶ Ό ἐσόμενος. It is commonly read, καὶ ὁ ὅσιος, the article indicating, against all manner of speaking, that the scripture has been corrupted. But whether the Vulgate reads the article or not, it translates ὅσιος no more correctly as "Sanctus" (Holy), wrongly omitting the particle καὶ, which is absolutely necessary to connect δίκαιος (righteous) & ὅσιος. But when John, in all the other places where he explains the name of Jehovah, as we said above, I.4, usually adds the third, namely καὶ Ό ἐρχόμενος, why would he have omitted that here? Therefore, I cannot doubt that the genuine scripture is what I have restored from an old bona fide manuscript (lit. old manuscript of good faith), namely Ό ἐσόμενος. The reason why Ό ἐρχόμενος is not written here, as in the four places above, namely I.4 & 8, likewise 4.8 & 11.17, is this: because there it deals with Christ as the judge who is to come; but in this vision, He is presented as already sitting on the tribunal, and exercising the decreed judgments, and indeed eternal ones.

You will notice that although Beza alleges corruption of the text, there is nothing about a line being worn away.  Beza says that the καὶ is necessary to join righteous and holy, and therefore the Vulgate is wrong.  Oddly enough, on this point the Vulgate happens to be right, as the majority of texts do not support Beza's coordinated adjective view.

Holland continues:

In addition to the Greek manuscript witnesses (which in this passage are few, as we have already noted), early translations should be considered. Again, the weight of the evidence falls on the side of "holy" and not "and shall be." Most translations, such as the Latin, omit the "and" using only "holy" (the Latin word is "sanctus"). Primasius, Bishop of Hadrumetum, wrote a commentary on Revelation around 552 AD and used the Latin word "pius" instead of "sanctus." They mean the same, but it does reveal yet another variance in the text. This, of course, brings us to yet another group of witnesses: Patristic citations.

The Greek manuscript witnesses here are not especially few, as compared to the remainder of Revelation.  Thankfully, Holland acknowledges the undeniable fact that nearly all the ancient translations stand with majority of Greek manuscripts.  

While Primasius' use of "pius" rather than the usual Latin translation "sanctus" is interesting, it does not suggest a different Greek Vorlage, as both words are legitimate translations of "osios."  To say that this reveals "yet another variance" (sic for "variant") in the text is really only the case with respect to the Latin tradition.

Two things should be stated before continuing. One, as confirmed by Jerome, there were a number of various Latin editions of the New Testament which differed in both translation and content before and around 405 AD (when Jerome finished his Vulgate). Most of these we do not have. Two, as pointed out by Dr. John Wordsworth (who edited and footnoted a three volume critical edition of the New Testament in Latin) the like phrase in Revelation 1:4 "which is, and which was, and which is to come;" sometimes is rendered in Latin as "qui est et qui fuisti et futurus es" instead of the Vulgate "qui est et qui erat et qui uenturus est." (John Wordsworth, Nouum Testamentum Latine, vol.3, 422 and 424.)

Once again, Holland's concession regarding Wordsworth's statement is significant.  There are a variety of "legitimate" ways to translate Greek to Latin, and pre-Jerome Latin translations varied.  Neither Jerome's nor the other is woodenly literal.

Holland continues:

Wordsworth also points out that in Revelation 16:5, Beatus of Liebana (who compiled a commentary on the book of Revelation) uses the Latin phrase "qui fuisti et futures es." This gives some additional evidence for the Greek reading by Beza (although he apparently drew his conclusion for other reasons). Beatus compiled his commentary in 786 AD. Furthermore, Beatus was not writing his own commentary. Instead he was making a compilation and thus preserving the work of Tyconius, who wrote his commentary on Revelation around 380 AD (Aland and Aland, 211 and 216. Altaner, 437. Wordsword, 533.). So, it would seem that as early as 786, and possibly even as early as 380, their was an Old Latin text which read as Beza's Greek text does.

Beatus of Liebana does present a different translation from Jerome at Revelation 16:5.  That said, Beatus' translation does not suggest a Greek Vorlage corresponding to Beza's Greek.  In particular, Beatus' translation has two Greek verbs with Holy.  The difference between Beatus and Jerome is the tense of the verbs.  This may be more easily accounted for by considering that the translator of Beatus' Latin may have been less literal and more paraphrastic in his translation.  The sense of "the being [one] and [who has] been" is, after all, an expression of God's eternal self-existence.  This same sense is conveyed by Beatus' Latin translation with different verb tenses.

Whether or not Beatus' and Tyconius' (as that does seem to be Beatus' source) Latin translation was paraphrastic, it only has two verbs and a word corresponding to osios, which emphatically is not Beza's proposal (three verbs and no osios).

Holland concludes (before quoting from various translations):

It should be noted that none of the early English versions, nor the foreign translations, read as does the Authorized Version. However, they do not read as most modern versions do either. Instead they read somewhere in between using both the "and" with "holy." The New King James Version follows the reading of the Authorized Version.

The significance of this is hard to evaluate.  The reason why other Reformation-era Bibles translated "and holy" is because Erasmus and Stephanus printed Revelation 16:5 with an extra καὶ, akin to P47.  While it is understandable that scribes would add such a καὶ to make "holy" coordinate with "righteous," the majority of manuscripts confirm the absence of such a καὶ.