Response to Boyd and Farrell Edwards’s Response to My “Critique of Alma 36 as an Extended Chiasm”

View PDF

Editor’s Note: This article has footnotes. To review them, please see the PDF below.

Extended chiasms were apparently an ancient Hebrew literary form. Since they were little known during Joseph Smith’s day, any extended chiasms in the Book of Mormon would be evidence of the book’s ancient Hebrew origins. John W. Welch has identified a number of extended chiasms in the Book of Mormon, one of his favorites being Alma 36. Among his many defenders are the father-and-son team of Farrell and Boyd Edwards, both physics professors. 

My analyses of Welch’s chiasms over the years leave me unimpressed. For example, in my recently published Dialogue article, I argued in part that, because of all the repetition in this chapter, Welch has been able to impose a chiasm on the text where none was intended. The Edwardses, however, using primarily not a literary but a statistical analysis, claim it is highly probable that Alma 36 as an extended chiasm was intentional and not inadvertent. While mentioning the Edwardses’ work in my critique, I did not address their statistical analysis directly, noting only that their approach is based on the order of words and ideas and that it explicitly disregards the literary merit of the chiasm. Reasoning that chiasmus is a literary device, I critiqued the literary merit of the chiasm they constructed from Alma 36, which differs from Welch’s construction. The Edwardses have responded to my critique and I have responded in turn at http://www.dialoguejournal.com/. Here I will address only the Edwardses’ statistical analysis. 

In their response, the Edwardses reassert the results of their statistical analysis, which assesses the likelihood that the elements in the chiasm would fall into a chiastic order by chance, that is, if they were drawn randomly from a hat. They recognize that their method supplements but does not replace literary analysis of a chiasm. Their method is illustrated in their paragraph 8, where they challenge my objection to their including multiple occurrences of key ideas within a chiastic section. They then prove mathematically that such multiple occurrences represent a higher degree of organization than a chiasm without multiple occurrences. 

Their mathematical proof is simple. Given two elements, a and b, for example, each repeated once, there are six ways to order them, only two of which are chiastic, viz., abba and baab, so that the likelihood of a random ordering of these two elements creating a chiasm is two chances in six, or one-third. Given an extra a, there are ten ways to order them, three of which are chiastic, viz., aabba, abbaa, and baaab, so that the likelihood of a random ordering creating a chiasm is three in ten, or 30 per cent. Since the likelihood of a random ordering creating a chiasm decreases with the extra a, the chance of its being random decreases, and the chance of its being intentional correspondingly increases. 

As applied to a paradigmatic, two-element chiasm such as “the first (a) shall be last (b) and the last (b) shall be first (a),” their method would seem to work. Repeating an a might even strengthen a chiastic element, e.g., “the first, yea, even the first shall be last and the last shall be first” equals aabba and might, as the Edwardses argue, represent an even higher degree of organization than the simple abba form. There is even an example of such a repeated idea in Alma 36. In verses 20 and 21, Alma rhapsodizes about his joy, referring to it three times. These two verses together might be a sound chiastic element even though joy is repeated. There is, however, no chiastic match for these two verses, and the Edwardses ignore them in their chiasm. 

Repeated elements in a more typical Alma 36 chiastic section, however, do not seem to represent a higher degree of organization. For example, the Edwardses’ section F’, the section I objected to and which objection they challenge in their response, contains three occurrences of born of God. (F’ comprises verses 23b-26a—120 words—which they pair chiastically with F, consisting of the first twenty-one words of verse 5 with one born of God.) Did Alma repeat born of God in F’ to strengthen this chiastic element? It seems doubtful. The first use, in verse 23, is about Alma’s being born of God and does indeed pair well with born of God in verse 5 (F) because it, too, is about Alma’s being born of God. (Verse 23’s born of God, however, is not the one Welch uses in his chiasm.) 

Then the account continues in verse 24 with Alma’s laboring to bring souls to repentance (which Welch uses in his chiasm but which the Edwardses ignore), so that others might taste of Alma’s joy (which is a nonchiastic match for the three jo;ys in verses 20 and 21 and which both Welch and the Edwardses ignore) and be born of God (which is also not the born of God used by Welch) and be filled with the Holy Ghost (which both Welch and the Edwardses ignore). In verse 25, Alma then expresses the joy he has received in the fruit of his labors (a fifth joy that both Welch and the Edwardses ignore). Finally, in verse 26, Alma proclaims that, be cause of the word he has received, many have been born of God. (Welch uses this born of God.) 

In short, because Alma is born of God, he goes to work so that others might be born of God; and indeed, at the time of his telling his story to Helaman, many had in fact been born of God. This is straightforward narrative that uses born of God three times. The Edwardses consider this narrative to be a chiastic section, although it begins with a dependent clause, is nearly six times longer than its chiastic counterpart, and contains extraneous language like Alma’s laboring to bring souls to repentance and nonchiastic pairings like Alma’s joy. They ignore all of these literary weaknesses.

The Edwardses’ statistical analysis permits this process and would apparently consider the three born of God’s to represent an even higher degree of organization than a single born of God. However, from a literary standpoint, their section F’ seems hardly to be an element of a chiasm. The Edwardses’ statistical analysis seems valid for truly random orderings of words, but the words an author uses are not put in a jar, shaken, and then withdrawn randomly. They appear in some order, but whether that order is chiastic must be determined by literary analysis, for which Welch’s fifteen criteria are helpful. The story of Alma’s conversion in verses 6-24 proceeds chiastically, from his rebellion against the church to his epiphany and his embrace of the church. It should be easy to find contrasting elements in such a story, and Welch and the Edwardses do, indeed, find some. What is surprising is that, given this splendid opportunity to create a real chiasm, Alma failed to do so. 

With respect to the literary merits of Alma 36 as a chiasm, the Edwardses do not challenge any of my data but only my misapplication of Welch’s proposed criteria. Welch’s criteria are useful in determining the presence of chiasmus but are explicitly neither finished nor authoritative and should not be made the issue. 

The details—and in critiquing a chiasm, the details can hardly be avoided—are found in Dialogue Online.