A Helper and Protector

One of the many tasks that we will engage in with this Blog is an investigation into the nature of the text of the Bible in the original languages and the nature of the Greek translation of the Old Testament known commonly as the Septuagint.  What follows is a serious philological examination of three Septuagint translations of an archaic Hebrew refrain, two of which end up in our liturgical texts. The content of this post may be particularly heavy in regard to Greek and Hebrew content, but it is hoped that the reader will be able to follow along even without knowledge of these languages.

Contrary to the dominant views held by many Orthodox apologists for the Septuagint (LXX), the Greek translation of the Hebrew Old Testament, the relationship between the LXX and the Hebrew Bible is extremely complex from the point of view both of the transmission history of the Septuagint and the translation itself. What we know today as “The Septuagint” is actually an amalgamation of several sources each with its own history of transmission and redaction. Understanding the relationship of the LXX and the Hebrew Bible requires diligent philological work in the original languages, an understanding of the transmission of each text, and the knowledge of the discipline of textual criticism. Generalizing statements about the LXX or the Hebrew Bible are almost always proven false upon a close examination of the texts, and as a rule, a case-by-case examination should be made before trends can be established in any of the books that comprise the Old Testament. As an illustration of the complexity of the LXX as a translation, we will examine a particular hymnic refrain found three times in the Hebrew Old Testament and three different translations of it found in the LXX.


The Hebrew Text

In the Song of Moses found in Exodus 15, we find in verse 2 a rather peculiar line of Hebrew poetry:

Exodus 15:2

עזי וזמרת יה ויהי לי לישוע
ˤozzī wǝzimrāṯ yāh wayyǝhī lī līšūaˤ

As far as scholars can tell, the text as found in the Masoretic Hebrew Bible and the Dead Sea Scrolls manuscript 1QIsA presents several difficulties, making direct translation very difficult. The KJV translates it thusly:

“The LORD is my strength and song, And He has become my salvation.”

This phrase, found prominently in the Song of Moses, must have become a rather popular refrain in Israelite hymnography, for it occurs twice more in the Hebrew Bible, Isaiah 12:2 and Psalm 118:14 (117:14 LXX) with only minor variation or none at all.

Isaiah 12:2

עזי וזמרת יה  יהוה ויהי לי לישוע
ˤozzī wǝzimrāṯ yāh YHWH wayyǝhī lī līšūaˤ
KJV: “YAH, the LORD, is my strength and song; He also has become my salvation.”

Psalm 118:14

עזי וזמרת יה ויהי לי לישוע
ˤozzī wǝzimrāṯ yāh wayyǝhī lī līšūaˤ
KJV: “The LORD is my strength and song, And He has become my salvation.”

There are two chief problems in this refrain, the grammatically anomalous זמרת zimrāṯ and the particle יה yāh, which we will now examine in detail.

The word זמרת zimrāṯ is a noun used in series with עזי ˤozzī, normally translated “my strength.” However, unlike עזי ˤozzī, זמרת zimrāṯ does not have the pronominal suffix –ī meaning “my.” Rather, it stands in the construct case as if it did have a pronominal suffix at one time or was joined to another word. By itself, the word should be זמרה zimrāh without the t at the end. As it stands, it could be literally translated “My strength and the song of …” which does not make sense.

The particle יה yāh is normally taken to be a shortened form of the Tetragrammaton יהוה YHWH, the name of God normally pronounced as Yahweh. A shortened form of it, יה yāh, was commonly used in the expression הללו-יה hallǝlū yāh “Praise Yah(weh)” or in many Hebrew names that included it as a theophoric element, e.g. זכריה zǝḵaryāh “Zechariah (Yah has remembered).” This reading renders the translation “My strength and a song of Yah(weh),” which does not make much sense, or “Yah(weh) is my strength and my song,” ignoring the ungrammatical זמרת zimrāṯ as explained above. The other option is to see it as the remnant of an original, archaic pronominal suffix זמרתיה zimratuya “my song.” As we will see below, this will turn out to be the preferable reading. If it is taken to be such, it is possible that the Tetragrammaton in full or in short was omitted through haplography, a scribal error of writing a sequence of letters or words once though they occur twice or more in the exemplar text.


The Greek Text

Now let’s look at how the Greek translators understood this phrase. In Exodus 15:2, the original LXX translator gives us a very unique reading:

Exodus 15:2

βοηθὸς καὶ σκεπαστὴς ἐγένετό μοι εἰς σωτηρίαν
“A helper and protector has become for me salvation.”

This might be familiar to many, as this rendering is used as the irmos of Ode 1 of the Great Canon of St. Andrew of Crete chanted during the first week of Great Lent, usually in the English translation “A helper and a protector – He has become my salvation.”

We should ask then, how did our Greek translator get “helper” and “protector” from “strength” and “song” as we found in the Hebrew? Keep in mind that Hebrew is written only with consonants and no vowels with the exception of some long vowels, which are represented by certain consonants that are able to do “double duty.” עזי ˤozzī could have been read in a number of ways, and one of those ways could have been עזר ˤōzēr, which is an active participle from the root עזר ˤZR meaning “to help” giving us “helper.” The last consonant, י yod, could have been easily mistaken in a hand-written manuscript for ר resh, and it appears that this is exactly what happened with our Greek translator – he read עזי ˤozzī “my strength” as עזר ˤōzēr “a helper.”

Moving now to זמרת zimrāṯ, we need to turn to any standard lexicon of the Hebrew Bible where we may find that the root זמר ZMR actually has two definitions, (1) the most common usage meaning “to sing,” and (2) a rarer usage meaning “to protect.” The Hebrew consonant ז zayin is historically derived from two different sounds that existed in archaic Hebrew, /z/ and /ḏ/ pronounced like the th in this. As the Hebrew language developed, these two sounds fell together into /z/. So, the two separate roots, ZMR meaning “to sing” and ḎMR meaning “to protect,” fell together into homonyms. This being the case, our Greek translator very adroitly chose the correct meaning of the word זמרת zimrāṯ contrary to virtually every other translation ever made! “Song” is not likely the correct meaning, as the word is placed in series with עזי ˤozzī as a hendiadys, two words put together that mean virtually the same thing such as “cloud and smoke” in Isaiah 4:5. The Greek translator subsequently may have ignored the particle יה yāh and read זמרת zimrāṯ as zōmére, a feminine participle, which could be used in an abstract sense. Nevertheless. the Greek approximates what we could posit for a reconstruction of the original Hebrew reading, “(Yahweh is) my strength and my protection, And he has become my salvation.”

Isaiah 12:2

ἡ δόξα μου καὶ ἡ αἴνεσίς μου Κύριος καὶ ἐγένετό μοι εἰς σωτηρίαν.
“My glory and my praise  is the Lord, and he has become for me salvation.”

The Greek translator of LXX Isaiah reads זמרת zimrāṯ in the sense of “song,” ἡ αἴνεσίς. The other Greek versions follow suit with Aquila reading ἐγκώμιον “panegyric” and Symmachus and Theodotion reading ὕμνησις “praising.” Deriving ἡ δόξα from עזי ˤozzī is more difficult, though it is more likely that the translator took a freer approach to his translation and chose a word more in keeping with ἡ αἴνεσίς. In other words, the translator of LXX Isaiah was more interested in creating a translation that read well in Greek rather than being slavishly literal to the Hebrew original. This is an element of what LXX scholars refer to as translation technique.

Psalm 118:14 (117:14 LXX)

ἰσχύς μου καὶ ὕμνησίς μου ὁ Κύριος καὶ ἐγένετό μοι εἰς σωτηρίαν.
“My strength and my song is the Lord, and he has become for me salvation.”

The Greek translator of LXX Psalms translates the Hebrew as we normally find it in English translations. Many will be familiar with this rendering from such translations or may recognize it from the Sunday prokeimenon of tone 2 often rendered as, “The Lord is my strength and my song. He has become my salvation.” The LXX translator took עזי as ˤozzī “my strength”,  יה yāh as “Yahweh,” i.e. “The Lord,” and understood זמרת zimrāṯ as being the absolute זמרה zimrāh.


What to take away

As illustrated from the preceding discussion of three different Greek translations of the same Hebrew phrase, establishing which text is “right” or “wrong” is not an easy thing to do and may, in fact, be a red herring altogether. For example, we might look at the LXX translation of Exodus 15:2 and say that this is “right” and the English translations of the Hebrew are “wrong” – until we find the very same “wrong” translation in Isaiah 12:2 and Psalm 117:14 (118:14 Heb). When the Hebrew text is difficult to translate due to textual corruption or poorly understood archaisms, the various translators of the LXX often arrive at different conclusions, and it is beyond us to pronounce any one of them as being more “inspired” and thus more “correct” than another. Indeed, if we are to understand the LXX as being inspired by God in any way, we cannot place strictures on what that meant for the translators, who were often dealing with extreme difficulties in the Hebrew text, difficulties that only modern philology has been able to understand. It is established beyond doubt that these inspired translators often translated the Hebrew text in ways that modern philology would consider erroneous. This does not mean that the LXX translators were knowingly in error or would have had any way of avoiding those errors without the accumulated knowledge of centuries of philological and linguistic investigation. Nevertheless, whatever we may understand to be inspiration must not prohibit us from understanding the human realities of the translation enterprise itself. Inspiration of scripture in any form does not mean that the text will be free of scribal or translational errors.

The variety of the LXX translations, whether they agree or disagree with the Hebrew text as we have it today, should not be placed in opposition to the Hebrew text as if one were inspired and correct and the other neither inspired nor correct. Instead, both the LXX and Hebrew texts of the Old Testament – along with the Syriac Peshitta, the Latin Vulgate, and many other ancient translations – exist together as a veritable “cloud of witnesses” that have all played a role in bringing God’s written revelation to his people. As the LXX itself demonstrates in this case where the same hebrew phrase is rendered in three different ways, as the saying goes, “there is more than one way to skin a cat.”

Click the image below for a flow chart explaining the textual changes and translations as they occurred.

Ozzi wezimrat-ya - New Page


  1. Another great article! Thanks very much. How would you consider the “great cloud of witnesses” of texts? Is there a translation of the Bible that is favored for study for anyone who is not linguistically adept in the ancient languages? I think I heard somewhere that the new Orthodox Study Bible is basically the NKJV, although I’m not certain of that.

  2. Thank you for such an interesting and thorough discussion. I especially appreciate what you write under What to Take Away. Moreover you raise a significant point about how liturgical use of texts should be considered when discussing text critical issues. It is with fear and trembling I comment on something written by a doctoral student at University of Chicago. I am a relative novice when it comes to textual criticism. I do not disagree with your analysis but do want to share a few thoughts and questions.

    1) One thing that really bothers me is the relationship between Exod 15:2 and Isa 12:2 and Ps 118:14. Are Isa 12:2 and Ps 118:14 based on Exod 15:2? How and more importantly when? Because if there are problems with text of Exod 15:2 we have two (later?) verses that are identical. (With the addition of יהוה in Isa 12:2 as a gloss?)

    1b) This makes your excellent analysis of Exod 15:2 LXX, Isa 12:2 LXX, and Ps 117:14 LXX all the more interesting. Same Hebrew words. Three very different translations.

    2) You mention Exodus 15 is archaic. As you know there is some debate how many examples of Archaic Biblical Hebrew (ABH) we have in the Hebrew Bible. The consensus seems to be Exodus 15, Judges 5 and… beyond those two texts much less agreement. Could that play a role in Exod 15:2? Such as the apparently ungrammatical וְזִמְרַת wezimrat? Could this be an archaic feminine singular absolute ending? Especially since fsa ־ָה –ā(h) developed from proto-Hebrew –at(v)? See also GKC §80f.

    3) I like the suggestion וזמרת could be based on ḏmr. Thus “protection” and be in parallel with עז “strength”. (Has this been published? Worth a short note?) Could a scribe wanted this to be read as both zmr “song” and ḏmr “protection”? Compare the work on polysemy (or multivalence) and Janus Parallelism by Cyrus Gordon, Scott Noegel, Gary Rendsburg, and others; also C-L Seow’s recent article on orthography in the book of Job. I am looking at this issue in 2 Sam 11-12. When we have Hebrew words that are (deliberately?) ambiguous translators have to make a choice.

    4) At the risk of contradicting myself in (3) what about theological context? There are parts of the Hebrew Bible that seem to emphasize “victory/strength through superior liturgical practice (rather than military-political power)”. “Worship rightly… and the walls/enemy shall fall”. Perhaps עז “strength” and זמרה “song” can be parallel.

    5) I have trouble accepting reconstruction ומזרתיה with final ־יה as archaic 1cs possessive suffix. Especially final short –a with mater ה. Are there any other examples of this?

    6) At the risk of contradicting myself in (2) I like your proposal that we have an example of haplography *וזרמתי יה “and my song (protector?) is Yah(weh)” > וזרמתיה. “The reading that explains all the others is to be preferred”. And that seems to explain the others best.

    7) How did you produce the beautiful graphic? Mac? Windows? Which application?

    8) I have to challenge a little bit the way that graphics mentions “Masoretic Corruption” (when it already addressed this as haplography) but “Greek translation”. So the Hebrew Bible represents “corruption” (negative) of the original text but the Septuagint is a “translation” (neutral) even though it is based on the same reconstructed original of Exod 15:2 yes?

    Over the last few years have become increasingly aware of the importance of the Septuagint for studying the Hebrew Bible/Old Testament. Protestant Christians often fail to appreciate this. Thank you for this stimulating and interesting post.

    1. Thanks for the thoughtful engagement with this post. I’ll try to answer your questions the best I can.

      (1) The relationship between the texts is difficult to ascertain. Likely, Ex 15:2 is the oldest, and the others are referencing it. I would suspect that the Song of Moses was well known, and that refrain would have been easily picked up and put elsewhere.
      (2) The notion of what constitutes ABH is difficult, because one has to determine that a text is actually archaic and not just archaizing. What has been considered ABH has recently come under question for that reason. I take at least 15:2 to be archaic because of the grammatical irregularities and the use of zmr as ḏmr.
      (3) I assume it has been published, but if not, it is at least noted in the standard lexica. I doubt polysemy is possible for the original author… whoever that was.
      (4) Your suggestion seems to be a bit of a stretch. In keeping with basic conventions of poetry, I would prefer to keep the two nouns as hendiadys.
      (5) There would not likely be simply zimrat without a case vowel, i.e. zimratu. So, if the /t/ is there, then I need a case vowel… which just made me realize that I have committed an error in vocalization. It should be ḏimratuya, with the nominative case, not ḏimratiya.
      (7) The graphic was done with an online tool, the name of which I have forgotten. A simple Google search should bring it up.
      (8) I say “corruption” because of the vocalization only. But you’re right to some extent, the corruption had occurred much earlier, and the Masoretes were giving it there best shot.

      There will be a lot more content like this on the blog, so I hope you will follow it and keep engaging in the comments.

  3. Thanks for the wonderful Post. I have only studied Greek and never Hebrew so I was a little at sea, but still able to hang on. I too would like further illumination about the Masoretic corruption. That might be worthy of its own post.

    1. Well, I use it as a technical term to describe the process of transmission and editing of the text. I don’t mean to say at all that the Masoretes intentionally screwed with the text or did anything underhanded. As a whole, the Masoretes did a fantastic job in preserving and vocalizing the text. I will do a post on that soon. Thanks for the suggestion!

  4. One possibility here is that the original text in Hebrew did not match the text in Isaiah 12:2 and Psalm 118:14, but had some similarities, and some scribe in the chain of transcription of the Masoretic or pre-masoretic text corrected the text in Exodus in the light of those texts. I realize that later Masoretic transcriptions were very meticulous, and that even texts that they knew to be in error were preserved, and the error was noted in the margin, but we do not know that this level of meticulous transcription was always the standard between the time of the Septuagint and the time the Masoretic text took the essential form that we have it today.

    1. The three texts are substantially the same in Hebrew. They all preserve the same grammatical anomalies. But, I’m curious what textual evidence you have for the time “between the time of the Septuagint and the time of the Masoretic text” “as wee have it today”?

      1. The three are substantially the same in the masoretic text that we have now, but whether or not that was true of the Hebrew text at the time of the translation of the Septuagint is another question. Aside from the dead sea scrolls, we don’t generally have pre-masoretic texts to compare.

        My point here is that we don’t know how stringent the scribal standards were prior to the masoretic rescension. We don’t have anything like the textual evidence we have in the case of the Greek New Testament.

        Now in this case, if the Peshitta (which is of equal authority to the Septuagint in the Church) has it as the Hebrew text does, then that presents a good argument that the reading in the Septuagint is not the best reading in this instance.

        1. Fr. John, I’ve already stated in the post that the Hebrew as contained in the MT is corrupt, and that the LXX preserved the more proper reading in Ex 15:2. That it didn’t in the other cases is of little importance other than to say that the LXX translators were able to come up with more than one translation of a given Hebrew phrase. This does mean, however, that the corruption in the proto-MT came about at least by the time that the LXX of Isaiah and Psalms was translated.

  5. I’ve read that the primary focus of the disagreement between Jews and Greek Christians in Antiquity boiled down to which version of the scriptures were used (Septuagint vs. Masoretic). I’ve also read that Justin Martyr believed that b/c verses in the Hebrew version seemed less Christological, that therefore they had been corrupted.

    This reality seems difficult if not impossible to square with the vision of inspiration you’ve presented… just sayin. I think your explanation seems like the only reasonable one, but I’m not sure how to accept it without jettisoning the Patristic view of scripture, which I thought included inerrancy. Please correct me if I am wrong.

    1. Jacob, it’s important to remember that the Greek and Latin Fathers by and large did not have access to the Hebrew Bible and were thus ignorant of it. Those that did bother to learn Hebrew, such as St. Jerome and Origen, knew its utility and revered it as scripture. This is not to speak ill of the Fathers, but only to emphasize that the accessibility of the Hebrew Bible and the ability to learn the Hebrew language is much easier than it was in antiquity. The Hebrew Bible is not by any means less Christological than the LXX if understood properly. In fact, there are some cases when the LXX actually may obscure messianic prophecies found in the Hebrew text (e.g. the “branch” prophecies in Jeremiah). Suffice to say, most of the patristic polemics against the Hebrew Bible are unfounded (especially those of St. Justin). Again, this is not to speak ill of them, but only to contextualize their place within intellectual history and the accessibility of information and knowledge.

  6. Eric,
    First, I am excited to have found your blog. Like an earlier commentator, I am not proficient in Hebrew yet, I also believe I can benefit greatly from your insights.

    My question has to do with the LXX as an edition. Peter Williams contends that the Septuagint as an actual bible may have never existed instead, what we possess are Greek translations of Hebrew texts made by numerous people over a period of time that never circulated together as an alternative to the Hebrew text. Obviously, he means in Second Temple Period or before since we have codices from the fourth century, A.D. that include what we call LXX. Any thoughts?


    1. This is true, though the rhetoric is tricky. We do actually have the LXX as a “Bible” because the Bible is not nor ever can be any one particular set of texts. It has always existed as a tradition of texts accepted by a religious community, which is more or less fluid. The LXX was developed over the course of some 500 years to become what is used normatively by the Byzantine churches (and her daughter churches). The LXX as a “Bible” can and does exist, because the Orthodox churches that use it define what it is, the same as the Roman Catholic Church or any Protestant communion has the ability to define a Bible for their own purposes. In other words, what is “the Bible” is particular to any religious community that defines its sacred literature. The mass of texts that exists out there from which various “Bibles” are determined by individual religious communities are what they are. Ancient Jewish and Christian literature is what it is, i.e. these texts exist independently of any religious community. Their significance, however, i.e. whether or not they are sacred and what versions of them are sacred, is determined by each individual religious community.

      1. I see an evolution of bibles, the first being the Pentateuch followed by an informal acceptance by religious leaders of additional Hebrew manuscripts. The NT was similar but the path was much shorter. Initially a core of individual manuscripts were accepted as inspired text. Over a few hundred years additional texts were accepted at the same time that the Christian church was finding its feet. By about 300 CE the church leaders were pretty much in agreement on those texts it considered inspired, with lingering doubts on a few. But by 400 the Church accepted, as canon, its bible. This was the first official Christian bible and it soon became official for both the Eastern and Western churches. The Protestant churches, a millennial later, rose. And their bibles became edited versions of the Catholic bible. It appears that divine inspiration of the Protestant bibles is an inheritance passed on from the Catholic hierarchy. Of note: I see divine inspiration occurring at two levels. First, the authors were inspired, and secondly the church leaders were inspired when compiling the bible.

        Richard Swarthout

Leave a Reply

Your email address will not be published. Required fields are marked *