Additional Janus Parallels in the Book of Mormon

  • Article Formats:
  • MP3 audio
  • PDF
  • MOBI
  • ePub
  • Kindle store
  • NOOK store
  • Order Print Copy

Abstract: A little more than 40 years ago, Cyrus Gordon discovered and described for the first time an ancient literary technique which he had found in the Hebrew Bible, and he gave it a name — a Janus parallel. That is why no one, more than 40 years ago, could have faked a Hebrew Janus parallel in an English translation of an ancient document. But, as I reasoned, if Janus parallels were a Hebrew literary device at the time Lehi left Jerusalem (for an analog see chiasmus), then such parallels probably can be found in the Book of Mormon. In this article I describe the technical methodology for discovering Janus parallels in an English translation, and I provide two new examples.

Cyrus Gordon, in his 1978 publication “New Directions,” drew attention to a heretofore unrecognized ancient literary technique in the Bible1 and called this ancient technique “Janus Parallelism.”2 In doing so, he was giving a nod to the Roman god Janus who has one face on the front of his head looking forward and another on the back of his head looking back.3

In its simplest and strictest form,4 a Janus parallel consists of a tristich, three lines or stichs, often of poetry. Hebrew poetry in general consists of two parallel stichs, or colons, and Hebrew prose also often employs parallel lines. But a Janus parallel must consist of at least three stichs to allow for the double parallel, as will be explained. The middle line of such a tristich usually contains a polysemous word or phrase. One meaning of the second line anticipates a synonymous or antithetical parallel with the third line, while the other meaning of the second line looks retrospectively at a synonymous or antithetical parallel in the first line. In a well-constructed Janus parallel, the first line and third line will [Page 82]not usually parallel each other closely, while the middle stich looks both forward and back.

By way of example, I offer the analysis of Gary A. Rendsburg, a student of Cyrus Gordon, published in 1980, on a verse of interest to Latter-day Saints.5 In the King James translation of Genesis 49:26, Jacob blessed his son Joseph saying, “The blessings of thy father have prevailed above the blessings of my progenitors unto the utmost bound of the everlasting hills.” This tristich can be structured as follows:

The blessings of thy father have prevailed
above the blessings of my progenitors
unto the utmost bound of the everlasting hills.
ברכת אביך גברו
על ברכת הורי עד
תאות גבעת עולם

The Janus parallel turns on the two Hebrew words at the end of the second stich, הורי עד. Depending on which vowels are supplied to the unpointed text, it can be read as “my progenitors of old,” which parallels with “your father” in the first stich. Or with different vowels, it can be read “mountains of old” which proleptically points to “everlasting hills” in the third stich.6

The King James translators followed the first interpretation which produced a translation similar to the following (my own) rendering:

The blessings of your father surpass
The blessings of my progenitors of old,
The delight7 of the eternal hills.

The Septuagint (the ancient Greek translation of the Old Testament) and a few modern translations preserve the second interpretation, which I render from the Hebrew as follows:

The blessings of your father surpass
The blessings of the mountains of old,
The delight of the eternal hills.

It can clearly be seen in this tristich that “the blessings of your father surpass” in line one and “the delight of the eternal hills” in line three are neither synonymously nor antithetically parallel with each other. Yet the polyvalence of the middle stich produces parallels both with “the blessings of your father surpass” and with “the delight of the eternal hills.”

If, as Gordon, Rendsburg, and others have demonstrated, ancient Hebrew literary techniques included Janus parallels, they could appear in the Book of Mormon.8 However, just as with the King James translation [Page 83]of Genesis 49:26, there is a major issue with recognizing Janus parallels in an English translation of an ancient text. Janus parallels depend most often on the polysemous nature of a word or phrase in the middle line of a three-line parallel set, as can be seen in the above biblical example. Even a practiced and adept translator will find it nearly impossible to find a word or phrase in the target language that replicates the polysemy of the middle line of the original text and allows the middle line to still parallel with one of its meanings the first line and with its second meaning the third line. (As a thought experiment, the reader should try to find an English rendering of the second line of this biblical tristich that at once parallels the first and third stich.)

Succinctly put, the English translation of the Book of Mormon will probably not replicate any polysemous parallels of a Nephite vorlage. If polysemous Janus parallels exist on the gold plates the only way to ferret them out is to make a conjectured but well-reasoned backtranslation from the English text to a supposed Hebrew or Egyptian vorlage, an obvious highly speculative exercise rife with subjective traps and semantic pitfalls.9

Despite the difficulties, to my delight I eventually found a plausible example in the Book of Mormon of a Janus parallel in 1 Nephi 18:16 and published my discovery in 2017:10

I did look unto my God,
and I did praise him all the day long;
and I did not murmur against the Lord because of mine afflictions.

The polysemy centers on the verb in the second stich, praise. The best candidate in this verse for a back translation of “praise” into Hebrew would be the piel verb zammēr, which can mean both “to praise” and “to sing.” With these two meanings of zammēr, 1 Nephi 18:16 can be read two ways:

I did look unto my God,
and I did sing unto him all the day long;
and I did not murmur against the Lord because of mine afflictions.


I did look unto my God
and I did praise him all the day long
and I did not murmur against the Lord because of mine afflictions

[Page 84]Since publishing this proposed Janus parallel, I have found two more possible examples of Janus parallels in the Book of Mormon. The first example I discuss, 2 Nephi 4:20, is fairly straightforward (I have italicized the possibly polysemous word):

My God hath been my support;
he hath led me through mine afflictions in the wilderness;
and he hath preserved me upon the waters of the great deep.11

The Janus feature of this verse depends on the polysemous nature of the posited Hebrew vorlage of “led” (lead in the present tense). The verb led in the middle stich could be the English rendering of the Hebrew verb nh that can signify two (slightly?) different homographic lexemes: √1, nh, “lead” or “conduct” as in Isaiah 11:2; or √2, nh, “join with, stand by (someone)” as in Isaiah 7:2.12 Thus, if the two meanings are separated and rendered individually, the text in my rendering can be read either with the √1 meaning,

My God hath been my support;
he hath conducted me through mine afflictions in the wilderness;
and he hath preserved me upon the waters of the great deep;

or, with the √2 meaning,

My God hath been my support;
he hath stood by me through mine afflictions in the wilderness;
and he hath preserved me upon the waters of the great deep.

In the first reading, “hath conducted me” in the second stich anticipates the parallel “hath preserved me” in the third stich. In the second reading, “hath stood by me” is parallel retrospectively with “hath been my support” in the first stich. Though not a strict requirement for a simple Janus parallel, the semantic separation between the first stich and the third stich is not strong. Nevertheless, the subtle nuances of the parallels in 2 Nephi 4:20 are quite poetic and beautiful.

My second example, found in 1 Nephi 10:13, is quite different in that it does not depend on a possible backtranslation into Hebrew. In fact, unlike nearly all explanations of Janus parallels, it does not depend on polysemous words or phrases at all.13 The parallels in 1 Nephi 10:13 work thematically, with the theme of the middle stich pointing retrospectively to the first stich and proleptically to the third. Because it is not dependent [Page 85]on polyvalence or polysemy it survived intact the translation into English. Structured as a tristich, 1 Nephi 10:13 reads,

that we should be led with one accord into the land of promise,

unto the fulfilling of the word of the Lord,
that we should be scattered upon all the face of the earth.

The Janus parallel in this verse can be overly simplified as being led ← fulfilling the word of the Lord → being scattered. The author has deftly placed before the middle stich the subordinate clause, “that we should be led with one accord into the land of promise,” which is then explained in the second line as “the fulfilling of the word of the Lord.” The third line, like the first line, is also couched as a subordinate clause, “that we should be scattered upon all the face of the earth.” This third line is also explained by the middle (second) line as “the fulfilling of the word of the Lord.” In other words, the second line explains the first and third line. Yet the first and third lines are not close parallels, even though when viewed through the historical lenses of the Restoration, the phrases “led with one accord into the land of promise” and “scattered upon all the face of the earth” can be viewed as somewhat parallel, if not synonymously then antithetically.

The three examples of Janus parallelism in the Book of Mormon that I have discussed so far, two here and one in a previously published paper, are surely only the tip of the iceberg of an ancient literary technique on display in this sacred volume. Their presence in the Book of Mormon makes it increasingly more evident that the door John W. Welch opened when he discovered chiasmus in the Book of Mormon more than 40 years ago was only the incipient notification to students of the Book of Mormon, prescient as it was, to pay attention to ancient literary techniques that bring out the meticulous construction and exquisite literary and doctrinal beauty evident in its pages.14

[Page 86]

1. “New Directions,” Bulletin of the American Society of Papyrologists 15, no. 1/2 (1978): 59–66.
2. For a first-hand account of Gordon’s discovery from the point of view of an observer, see Gary A. Rendsburg, How the Bible Is Written (Peabody MA: Hendrickson, 2019), 362n7.
3. For those familiar with the Harry Potter series, the character Quirinus Quirrell appears to be modeled after the Roman god Janus for whom our month of January is named.
4. In the ancient, pre-Hellenistic world, as far as we know, there were no handbooks nor instructions on how to construct a Janus parallel, nor for that matter on how to write a chiasm, or how to construct any literary form. The forms were known by all educated individuals, and certainly some attempts were more successful than others. But other than a general public opinion or private judgments, no one oversaw the correctness of individual attempts. Thus, there were exquisite compositions created by gifted literati on the one hand, and on the other hand, rather clumsy attempts.
5. Gary A. Rendsburg, “Janus Parallelism in Genesis 49:26,” Journal of Biblical Literature 99, no. 2 (1980): 291–93. I gladly acknowledge my discussion here of Genesis 49:26 as heavily dependent on Rendsburg. Because his article consists of slightly more than two pages, I refrain from referencing page numbers. Numerous other examples exist, but as will be explained, English translations do not reveal the Janus parallels that are in the Hebrew text (e.g., Genesis 15:1, Proverbs 31:21, Song of Solomon 2:12, etc.). For LDS interest in Genesis 49:26, see Wilford Woodruff’s 1878 statement that the Lord would “gather together the house of Israel to the valleys of these everlasting hills [in Utah], according to his decree to old father Jacob” (Journal of Discourses, 19:359). See also Orson Pratt, in Journal of Discourses, 2:292 (1855).
[Page 87]6. The form of the Masoretic lexeme, הורי, would require the translation “progenitors, forefathers,” as in the King James translation. Rendsburg, however, provides ample justification in his article allowing for the translation “mountains,” which does not have a waw (הרי) which was probably a mater lectionis anyway (Rendsburg, “Janus Parallelism in Genesis 49:26”). Rendsburg offers evidence that, in fact, it was read as “mountains” and/or “progenitors” anciently. The most compelling piece of evidence is that the LXX translates it as ὀρέων, “mountains.” Additionally, according to Rendsburg, Rashbam, the medieval Jewish commentator, translated “the blessings of your father … surpass the blessings of the mountains.” Rendsburg also suggests that both meanings, “mountains” and “forefathers,” were current when Targum Yerushalmi (also called Targum Pseudo-Jonathan; its date is disputed, though most date it to the 8th century with a range between 4th to 14th century) was written because it appears to reproduce both meanings in its translation:

May the blessings of your father be added to the blessings,
Wherewith Abraham and Isaac who are like mountains blessed you,
And to the blessings of the four mothers who are like hills,
Sara and Rebekah, Rachel and Leah.

7. In my rendering of this verse, I have chosen to follow the definitions in The Hebrew & Aramaic Lexicon of the Old Testament, eds. Ludwig Koehler, Walter Baumgartner, and Johann Jacob Stamm, trans. M. E. J. Richardson, [CD-ROM] (Leiden: Brill, 1994–2000), (hereafter, “HALOT”), for תאוה, namely, תאוה from תאוה(< אוה), meaning “longing, wish, yearning.” HALOT, citing Genesis 49:26, also suggests “abundance” and “splendour.” Thus, the New English Bible reads “bounty.” In a separate entry, HALOT also notes that the homograph תאוה(< אוה) means “marking; mark a boundary,” which explains where the “utmost bound” of the King James translation in Genesis 49:26 comes from. However, HALOT does not cite Genesis 49:26 as an example of this second homograph.
8. See, for example, Jeff Lindsay, “Janus Parallelism in the Book of Job: A Review of Scott B. Noegel’s Work,” Interpreter: A Journal of Mormon Scripture 27 (2017): 213–20,, and Jeff [Page 88]Lindsay, “The Possibility of Janus Parallelism in the Book of Mormon,” Interpreter: A Journal of Mormon Scripture 28 (2018): 1–20,
9. The first of many potential perils with this method is to rely on post-Iron Age Hebrew. This is not to say that post-Iron Age Hebrew cannot provide evidence, but it does mean that the further removed the backtranslation is from Iron Age Hebrew, the less likely it will reflect a Book of Mormon vorlage.
10. The discussion of 1 Nephi 18:16 was published as Paul Y. Hoskisson, “Janus Parallelism: Speculation on a Possible Poetic Wordplay in the Book of Mormon,” in “To Seek the Law of the Lord:” Essays in Honor of John W. Welch, eds. Paul Y. Hoskisson and Daniel C. Peterson (Orem, UT: The Interpreter Foundation, 2017), 151–59. I can now add to that discussion the fact that 1 Nephi 18:16, in addition to being an example of a Janus parallel, is one of the better examples in the Book of Mormon of what some scholars call “the propensity for languages to place shorter items before longer ones in a series.” Rendsburg, How the Bible Is Written, 424. That is, notice how the first stich of the Janus parallel in 1 Nephi 18:16 is the shortest line, while the third stich is longer than either of the previous two lines. For a general discussion of this phenomenon in the Bible (and other languages, including English) see Rendsburg, “Shorter Before Longer — and Divergences Therefrom,” in How the Bible Is Written, 424–42.
11. Royal Skousen, The Book of Mormon: The Earliest Text (New Haven, CT: Yale University Press, 2009) also parses the structure of this verse as I have done.
12. For the Hebrew see HALOT, s.vv. נחה 1 and נחה 2. It would make little difference if other possible roots with the requisite consonantal structure were involved, such as נוח, because the 3rd masculine singular qal perfect forms with a pronominal suffix for either root would be identical in Iron Age Israel, (י)נחנ.
13. Because most Janus parallels depend on polysemy, the temptation to create a backtranslation of 1 Nephi 10:13 containing a contrived polyvalence, such as a qal versus piel form of a verb, is rather enticing. But there is no need, and therefore I forbear.
[Page 89]14. For a recent publication about literary beauty see Donald W. Parry, Preserved in Translation: Hebrew and Other Ancient Literary Forms in the Book of Mormon (Provo, UT: Religious Studies Center, Brigham Young University, 2020).
Posted in Article and tagged , on . Bookmark the permalink.
Referenced scriptures: ,

About Paul Y. Hoskisson

Paul Y. Hoskisson, emeritas professor of Religious Education at BYU, received his PhD from Brandeis University in ancient Near East languages and history. In addition to teaching in Religious Education at Brigham Young University beginning in 1981, he served as institutional representative on the Board of Trustees of the American Schools of Oriental Research, Director of the Laura F. Willes Center for Book of Mormon Studies, Richard L. Evans Professor of Religious Understanding, Associate Dean of Religious Education at BYU, Coordinator of Near Eastern Studies at BYU, and epigrapher for the 1983 ASOR excavation at Qarqur, Syria. Prior to coming to BYU, he taught ancient Near East languages at the Universität Zürich.

Go here to see the 4 thoughts on ““Additional Janus Parallels in the Book of Mormon”” or to comment on it.