Zechariah 12:10 Corruption in the NWT

by Sea Breeze 84 Replies latest watchtower bible

  • Earnest
    Earnest
    @ aqwsed12345

    Let me see if I understand you correctly.

    When John writes in John 19:37 "And again, a different scripture says [Zechariah 12:10]: “They will look to the one whom [ν] they pierced”, he is not altering the Hebrew of Zechariah 12:10 "because John presupposes that ν refers back to the first-person speaker".

    But when the NWT translates Zechariah 12:10 in the same way "it corrupts the Hebrew".

    I agree with you that John is not altering the Hebrew, and neither is the NWT. Why do I say this? You suggested that the evangelist had before him 8 ḤevXIIgr itself, another scroll of the same recension, or a liturgical extract derived from it when he quoted from Zechariah 12:10. If John wrote his gospel in Ephesus it seems very unlikely he would have had in front of him a scroll that ended up in a cave in the Judean desert. So there must have been at least two copies of this translation of Zechariah. And Origen must have had a copy when he created the Hexapla in Caesarea. In fact, it looks like this was the copy of the Greek scriptures in common use in John's time, doesn't it?

    You say about it that it is "the Jewish revision that already matched the Hebrew more closely than the older LXX". I agree. Bynum argues convincingly in his article on John 19:37 "that John’s citation from Zechariah is both accurate and historically reliable. Carefully considering the biblical textual milieu of the era brings to light John’s concern for fidelity to the prominent Hebrew text of his day, and for the correctness of the Septuagint form of the citation." Whether this is a new translation or a revision of an existing translation, it shows that it was used used by Jews at least some of whom could compare them with proto-MT. In other words, these were semi-bilingual Jews who made use of both Greek and Hebrew texts, and required, like those who used Aquila later on, a translation which accorded with the Hebrew text before them. To put it more plainly, the Hebrew text that the translators had available to them reflected the Greek of John 19:37 and not that of the Masoretic Text.

    Just one other thing to add while we are talking about 8 ḤevXIIgr. This is a translation into Greek which contains the tetragrammaton multiple times (Jon 4:2; Mic 1:1, 3; Mic 4:4, 5, 7; Mic 5:4, 4; Hab 2:14, 16, 20; Hab 3:9; Zep 1:3, 14; Zep 2:10; Zec 1:3, 3, 4; Zec 3:5, 6, 7). John was using just such a translation when he wrote his gospel. So apart from providing support for the Zechariah 12:10 translation, it also supports the use of God's name in the Greek mss used by the gospel writers.

  • aqwsed12345
    aqwsed12345
    @Earnest

    In the exchange you have reduced a complex, three-stage trajectory—Hebrew oracle → Jewish Greek revision → Johannine citation—to a single step and then concluded that the NWT is exonerated because its English wording resembles John’s Greek. The similarity is superficial, and it does not touch the fundamental point: the NWT alters the Hebrew it claims to translate, whereas John does not.

    1. What the Hebrew says.
    All extant Hebrew witnesses, whether medieval codices (e.g., Aleppo, Leningrad) or the fragmentary proto-Masoretic texts from the Judean Desert, read the same consonantal sequence in Zech 12:10:

    וְהִבִּ֣יטוּ אֵלַי אֵ֥ת אֲשֶׁר־דָּקָ֖רוּ
    “they shall look unto me, ‹’ēt› whom they pierced.”

    The first-person pronominal suffix -ay (“unto me”) is not disputed; no Qumran fragment or medieval margin offers אֵלָיו (“unto him”). The difficulty is rhetorical, not textual: YHWH, the speaker, declares himself to be the one pierced.

    2. How the late Second-Temple translators handled that line.
    The so-called kaige–Theodotionic revisers, whose work is preserved in 8 evXII gr (1 st c. BCE/CE) and, centuries later, in Codex L (8 th c.), did not suppress the first person. They rendered literally—ες μ ν ξεκέντησαν—while adding the relative pronoun ν required by Greek syntax. The result mirrors the Hebrew meaning (“unto me whom they pierced”) yet is stylistically smoother than the Old Greek of Vaticanus, which had diverted the thought by changing the verb (“they insulted me”). Nothing in the kaige revision implies a different Hebrew Vorlage; it is a conscientious accommodation of Greek to the same consonantal text we still read.

    3. What John does with that Greek.
    John 19:37 cites Zechariah in virtually that kaige form, with one synonymic substitution (ψονται for πιβλέψονται) and the omission of μ—an omission that neither erases nor contradicts the first-person implication, because the antecedent can be supplied from context. John has just narrated the lance thrust (19:34); he then affirms that this fulfils the Scripture that says “they will look ες ν ξεκέντησαν.” Because the pierced one in Zechariah is YHWH, John’s christological inference is that the crucified Jesus shares YHWH’s identity. The citation is interpretive, but it is not a textual alteration of Zechariah: it rests transparently upon the Jewish Greek then in circulation.

    4. What the NWT does instead.
    When the NWT comes to translate Zech 12:10 from Hebrew, it replaces אֵלַי with a third-person construction: “they will look to the one whom they pierced.” That English is indeed congruent with John’s Greek surface form—but only because the translators have first excised the inconvenient first-person pronoun from the Hebrew clause. No Hebrew manuscript supports their move; no critical apparatus lists such a variant; it is an emendation introduced solely so that the Hebrew verse will no longer equate the speaker (YHWH) with the pierced figure whom John identifies as Jesus. In other words, the NWT has harmonised the source text to its own reading of the citation in order to avoid the divinity-claim that the citation, in its original literary setting, makes inevitable.

    5. Why the analogy with John fails.

    You begin by suggesting that I claim John does not alter the Hebrew of Zechariah 12:10 because he presupposes that ὃν (“whom”) refers back to the first-person speaker, while the NWT’s similar rendering somehow corrupts the Hebrew. This mischaracterizes my position slightly, so let me clarify. John is acting within the normal bounds of Second-Temple Jewish exegesis: he quotes an accepted Greek revision and draws a theological conclusion from the conjunction of that text with the events he has witnessed. The NWT, conversely, presents itself as a translation of the Hebrew Bible but silently rewrites the clause to protect an Arian christology. The former is legitimate interpretation; the latter is textual manipulation.

    My argument is not that John’s use of ν inherently refers back to the first-person speaker in a grammatical sense divorced from context, but that John cites a Greek tradition—aligned with kaige-Theodotion as seen in 8evXIIgr—that interprets the Hebrew text faithfully, and he applies it theologically to identify Jesus, the pierced one, with the divine speaker, YHWH. The NWT, by contrast, does not merely adopt this Greek rendering; it rewrites the Hebrew text itself, changing the Masoretic Text’s (MT) וְהִבִּיטוּ אֵלַי אֵת אֲשֶׁר־דָּקָרוּ (“they shall look unto me, [namely] the one whom they pierced”) into “they will look to the one whom they pierced,” thereby eliminating the first-person suffix אֵלַי (“unto me”) and breaking the link between the speaker and the pierced one. This is not a matter of following an alternative textual tradition, as no Hebrew manuscript supports such a reading, but a deliberate emendation driven by the NWT’s theological presupposition that Jesus cannot be YHWH.

    You assert that John is not altering the Hebrew, and neither is the NWT, because John likely had access to a Greek text like 8evXIIgr or a similar recension, which you argue was widely used in his time. You further suggest that this Greek text reflects a Hebrew Vorlage different from the MT, one that aligns with John 19:37’s ψονται ες ν ξεκέντησαν (“they will look to the one whom they pierced”). Let’s examine this claim closely. The Hebrew text of Zechariah 12:10 in the MT is unambiguous in its consonants: אֵלַי is a first-person pronominal suffix, meaning “unto me,” and the relative clause אֵת אֲשֶׁר־דָּקָרוּ identifies this “me” as the one pierced. No extant Hebrew manuscript, including those predating the medieval codices like Aleppo and Leningrad, reads אֵלָיו (“unto him”) instead. The Dead Sea Scrolls, while fragmentary for Zechariah, offer no evidence of such a variant, and the proto-MT tradition, which 8evXIIgr’s translators worked from, consistently preserves אֵלַי. Your suggestion that the Hebrew text available to the kaige translators reflected John 19:37 rather than the MT lacks textual support. The kaige-Theodotion tradition, as evidenced by 8evXIIgr, renders the verse as ες μ ν ξεκέντησαν (“unto me whom they pierced”), retaining the first-person reference of the Hebrew and introducing a relative pronoun ν to clarify the syntax for Greek readers. This is not a departure from the MT but an interpretative rendering of it, wrestling with the awkward Hebrew construction while preserving its meaning.

    6. On the alleged “different Hebrew” behind the kaige revision.
    No evidence whatever points to a Hebrew reading with a third-person object. The kaige revisers are famous precisely because they laboured to make the Greek line up more closely with proto-Masoretic Hebrew; their work forms the bridge between the Old Greek and the later strictly literal versions (Aquila, Symmachus, Theodotion). That bridge presupposes, and everywhere witnesses to, the same first-person אדני/אלי forms. The fragmentary state of 8 evXII gr at this verse does not weaken the point: enough letters are extant to establish ες — ν ξε[κέντησαν, and the pattern of the revision elsewhere requires ες μ before the lacuna.

    You cite Bynum’s argument that John’s citation is “accurate and historically reliable” and reflects “the prominent Hebrew text of his day,” implying that this Hebrew text differed from the MT. However, Bynum’s point, as articulated in his article (The Fourth Gospel and the Scriptures, Brill, 2012), is that John’s fidelity lies in his use of a Greek text from the kaige recension, which sought to align the Septuagint more closely with the proto-MT, not that the Hebrew itself varied. The kaige translators were not working from a Hebrew text that read אֵלָיו; they were translating the same אֵלַי found in the MT, as confirmed by scholars like Dominique Barthélemy (Les Devanciers d’Aquila, 1963) and Emanuel Tov (The Greek Minor Prophets Scroll from Naal ever, 1990). The fragmentary state of 8evXIIgr at Zechariah 12:10 prevents a complete reading, but reconstructions based on spacing and context support ες μ, consistent with the MT. John’s adaptation, replacing ες μ ν ξεκέντησαν with ες ν ξεκέντησαν, simplifies the phrasing but does not negate the underlying Hebrew; it reflects a Greek interpretive choice that still allows his theological identification of the pierced one with YHWH, especially given the context of John 19:34–37 and his broader Christology (e.g., John 1:1, 20:28). The NWT, however, does not translate this Greek tradition back into Hebrew accurately; it alters the Hebrew אֵלַי to a third-person construction without manuscript evidence, a move that cannot be justified as fidelity to either the MT or the Greek.

    Your logistical point about John’s access to 8evXIIgr—suggesting it’s unlikely he had a scroll that ended up in a Judean cave while writing in Ephesus—misses the mark. I’m not claiming John used that specific scroll; rather, he likely drew from a Greek text of the same kaige-Theodotion type, which was part of a broader Jewish revisionist movement in the late Second Temple period. These texts circulated widely enough to influence Origen’s Hexapla centuries later, and their use by semi-bilingual Jews, as you note, supports their prominence. But prominence does not mean they reflect a different Hebrew text; it means they were Greek translations of the proto-MT, designed to hew closely to the Hebrew while making it accessible in Greek. John’s citation aligns with this tradition, but his theological application goes beyond mere translation, identifying Jesus as the divine figure pierced, a conclusion reinforced by Revelation 1:7’s apocalyptic echo of the same verse.

    7. On the tetragrammaton in Greek scrolls.
    The appearance of יהוה written in palaeo-Hebrew letters inside some Greek biblical manuscripts is a Jewish scribal convention; it left no trace in the NT textual tradition. Every papyrus and uncial of John reads κύριος or θεός in OT quotations, never the divine Name. Consequently the NWT’s retrojection of “Jehovah” into the NT is another conjectural emendation, unsupported by documentary evidence and unrelated to the question of Zech 12:10’s pronominal switch.

    The presence of the Tetragrammaton in some Greek manuscripts of the Hebrew Bible reflects a Jewish scribal practice to preserve YHWH’s sanctity, seen in texts like 8evXIIgr and certain Qumran fragments (e.g., 4Q120). However, NT manuscripts, including those of John (e.g., 𝔓66, 𝔓75), uniformly use κύριος (“Lord”) or θεός (“God”) when quoting the OT, following the Septuagintal convention of substituting κύριος for YHWH. There is no evidence that John’s autograph included the Tetragrammaton; the earliest Christian witnesses, such as Papias or Justin Martyr, show no trace of it in NT citations. The NWT’s restoration of “Jehovah” in John 19:37 and elsewhere is a modern conjecture, unsupported by Greek manuscript evidence, as even Jason BeDuhn notes in Truth in Translation (2003), critiquing it as a “conjectural emendation” lacking textual basis.

    8. Conclusion.
    John and the kaige translators transmit the same underlying message: Israel’s God speaks in the first person and declares himself to be the One who will be pierced. John identifies that divine speaker with the crucified and risen Christ. The NWT, finding such a claim incompatible with its doctrine, edits the Hebrew sentence so that “me” becomes “the one,” thereby severing the clause that John exploits. That, and not John’s legitimate use of Jewish Greek, is the corruption that textual scholarship detects and that a Trinitarian reading exposes. the core issue remains: the NWT’s rendering of Zechariah 12:10 as “they will look to the one whom they pierced” is not a translation of the Hebrew אֵלַי but a substitution drawn from the Greek tradition, adjusted to avoid John’s Christological implication. John’s ν presupposes a connection to the divine speaker when read in context, whereas the NWT severs that link by reworking the Hebrew into a third-person construction, unsupported by any textual witness. This is corruption—not because it differs from John’s Greek, but because it claims to translate the Hebrew while inventing a reading to fit an Arian Christology. John, by contrast, uses a legitimate Greek rendering to affirm that the pierced one is God incarnate, a truth the NWT explicitly rejects.

    In short, your defense of the NWT conflates John’s use of an interpretative Greek tradition with the NWT’s alteration of the Hebrew text itself. The evidence—from the MT, 8evXIIgr, and John’s Gospel—shows that the Hebrew says “me,” John interprets it as fulfilled in Jesus as YHWH, and the NWT changes it to protect a sub-divine Christology. The charge of corruption stands firm.

  • Sea Breeze
    Sea Breeze
    Let me see if I understand you correctly.

    Indeed.

    I must admit that I like shortcuts. I pretty much just see how the first few generations of Christian leaders either quoted or paraphrased a particular verse that heretics say is wrong. If I find two or more supporting the traditional reading, (as I do 98% of the time) that's good enough for me.

    Helps me skip a lot of reading about grammar and foreign languages so I can do more important things. But, I do appreciate all the posts on these "corruption" threads. Very enlightening.

    Did I mention my three sons and I caught 8 large red snapper (the limit) in the Gulf of America last weekend? Yes, I do other things than hang out here. Hope you all do too!

  • aqwsed12345
    aqwsed12345

    It is best to begin where all text–critical discussion must begin: with the surviving Hebrew. Every complete Masoretic witness and every extant Judean-Desert fragment that preserves Zechariah 12:10 gives exactly the same consonantal sequence,

    וְהִבִּיטוּ אֵלַי אֵת אֲשֶׁר־דָּקָרוּ
    “they shall look unto me, ‹’ēt› whom they pierced.”

    Nothing in the Hebrew tradition, early or late, replaces אֵלַי (“unto me”) with אֵלָיו/עָלָיו (“unto him”). The so-called waw/yōd interchange raised by Burney is well known; yet all presently collated DSS or medieval manuscripts continue to read the first-person suffix. Even the two medieval codices that register a keri-note of עָלָיו retain אֵלַי in the ketiv; scribes flagged the verse as theologically disquieting; they did not supply a demonstrably ancient alternative reading. Consequently, modern critical editions (BHS, BHQ) reproduce אֵלַי without siglummed variants.

    Because the MT yields the striking claim that YHWH himself is “pierced,” most scholars—whether or not they are persuaded by Christian dogma—take the line as a textbook illustration of the lectio difficilior. Scribes amend what shocks them; they do not invent it. That is why the locus classicus in Gesenius–Kautzsch–Cowley is headed, not “probable text,” but “we should probably read,” i.e., an editorial emendation, not a reading with external support. To treat that conjecture as though it were an attested Hebrew variant is to confuse scribal evidence with modern editorial suggestion. Men who translate for the pew cannot legitimately substitute an un-attested conjecture for the universally attested text.

    Turning from Hebrew to Greek, the Old Alexandrian translator of the Minor Prophets did transform the clause: “they shall look to me because they have mocked” (πιβλέψονται πρς με νθ’ ν κατωρχήσαντο). Why? Probably to defuse the anthropopathic shock. In the late second-temple period another cadre of Jewish scholars—the circle today called kaige–Theodotion—produced a counter-revision meant to restore formal fidelity to the Hebrew. Their work is preserved (i) in the Naal ever scroll 8evXII gr (late 1st BCE/early 1st CE) and (ii) in the eighth-century Codex L (Rahlfs 309). At Zech 12:10 the kaige text reads, as one would expect, “κα πιβλέψονται ες μ ν ξεκέντησαν”—a literal rendering of אֵלַי plus a relative pronoun demanded by Greek syntax. This is the Jewish-Greek form which the Fourth Evangelist knew. His citation in John 19:37 is verbally identical except that he replaces πιβλέψονται with the synonymous ψονται and, in a typical compression, omits the explicit μ. Nothing in that omission effaces the first-person sense, because the reader who knows Zechariah will supply it; indeed Revelation 1:7 makes the identification between the pierced One and the eschatological Lord explicit.

    Greg Stafford’s defense of the NWT ignores that textual trail. He insists that because John 19:37 no longer contains μ, modern translators are free to remove אֵלַי from the Hebrew and to write “they will look to the one.” That is a non sequitur. John is not translating Hebrew; he is quoting a current Jewish Greek recension whose own fidelity to אֵלַי is beyond dispute. To erase the first person from the Hebrew base text is therefore not to “follow John,” it is to alter the only Hebrew that John and every other ancient witness had in front of them.

    Nor does the appeal to other modern versions rescue the move. Most modern English Bibles that read “him/the one” put “me” in the margin and explain in a foot-note that the Hebrew is first person; their editorial committees judged the stylistic roughness intolerable but made no claim of manuscript authority. The NWT alone presents the conjecture as the Hebrew original, without alerting the reader that no Hebrew witness supports it—a procedure the United Bible Societies’ Principles and Procedures for Translation (§1.4) labels “unacceptable.”

    Finally, the theological tail is unmistakably wagging the textual dog. A strict Arian Christology cannot permit a verse in which YHWH speaks and yet is the One whom Israel has pierced, for that is precisely the identification John exploits to say that the crucified Jesus shares YHWH’s identity. The NWT therefore suppresses the first-person suffix, harmonizes Zechariah to its own reading of John, and then turns round to claim that John vindicates the alteration. Circularity could scarcely be clearer.

    The objection from 1 John 4:12 (“No one has ever seen God”) rests on a category mistake. John’s Gospel does not claim that the crowd saw the divine nature; it claims they pierced the enfleshed Logos, the One in whom “all the fullness of deity dwells bodily” (Col 2:9). Classical Christology calls that the communicatio idiomatum: what is done to the incarnate Son is predicated of the person, and therefore—because the person is divine—of God. That is why John can say in the same breath that “no one has seen God” (1:18) and still insist, quoting Zechariah, that the soldiers looked upon him whom they pierced.

    In sum, Stafford’s defense collapses at every textual checkpoint. The Hebrew says “unto me”; the earliest and best Jewish Greek says “unto me”; John writes in conscientious dependence on that Greek; the NWT alone rewrites the Hebrew, unbacked by any manuscript, in order to silence a verse that otherwise discloses what orthodox Christianity has always confessed: the crucified Jesus is none other than Israel’s God in human flesh, the Lord who was pierced and who yet promises, “they shall look upon me.”

  • Sea Breeze
    Sea Breeze
    The objection from 1 John 4:12 (“No one has ever seen God”)...

    God was veiled in human flesh that same way that Moses had to wear a veil because his face showed the brilliance of the Lord after meeting with him on Sinai.

    Having therefore, brethren, boldness to enter into the holiest by the blood of Jesus, By a new and living way, which he hath consecrated for us, through the veil, that is to say, his flesh. - Hebrews 10: 19&20

    • Chrysostom quoted Hebrews 10:20 in his homilies on Hebrews and interpreted the verse, stating that "Christ's flesh shows us the Godhead as if it were under a veil, For otherwise we could not stand the brightness of it."

Share this

Google+
Pinterest
Reddit