Criteria for Identifying and Evaluating the Presence of Chiasmus
John W. Welch
Abstract: This article defines fifteen criteria one can use to measure the strength or weakness of a proposed chiastic pattern in a given text. The need for rigor in such studies depends primarily on how the results of the proposed structural analyses will be used. Ultimately, analysts may not know with certainty whether an author created inverted parallel structures intentionally or not; but by examining a text from various angles, one may assess the likelihood that an author consciously employed chiasmus to achieve specific literary purposes.
In recent decades, numerous passages in the Bible and elsewhere have been analyzed by commentators who find those texts to be chiastic.1 Some of the suggested inverted structures are convincing and illuminating; others seem quite marginal. Some texts are strongly and precisely chiastic, while in other cases it may only be possible to speak of a general presence of balance or framing. From these studies it is apparent that all possible chiasms were not created equal and that in order to be clear in discussing chiasmus it is necessary for commentators to recognize that "degrees of chiasticity" exist from one text to the next.
Some attempts have been made in the past to define chiasmus.2 Lund, for example, proposes seven "laws" of chiastic structures: (1) the center is always the turning point, (2) a change in the trend of thought or antithetical idea is often introduced at the center, (3) identical ideas are often distributed in the extremes and at the center of the system and nowhere else in the system, (4) ideas will shift in many cases from the center of one system to the extremes of a corresponding system, (5) certain terms definitely tend to gravitate toward particular positions in a given system, (6) larger units are frequently introduced by frame-passages, and (7) chiastic and alternating lines frequently occur within a single unit. Obviously, Lund's "laws" are more descriptive than they are definitive; they describe features that are common to many chiastic passages, but they start from the assumption that the passages are recognizably chiastic. Furthermore, Lund's "laws" are riddled with subjective words like "often," "frequently," and "many," leaving unfinished the task of identifying the factors that are characteristic of strongly chiastic texts or that describe the point at which it is appropriate to denominate a passage as chiastic. Accordingly, this paper proposes a set of criteria a person may use to evaluate the degree to which chiasmus is present in a given passage.
The need for criteria to identify chiasms was addressed in the book Chiasmus in Antiquity. There I explained how the identification of chiastic passages involves both objective and subjective criteria. Objectively, the reader must be able to identify significant, balanced repetitions in an inverted parallel order with a focus or shift at the center.
If any aspect of chiastic analysis is to produce rigorous and verifiable results, the inverted parallel orders, which create the chiasms upon which that analysis is based, must be evidenced in the text itself and not imposed upon the text by Procrustean design or artifice of the reader.3
Defining literary units and determining their beginning and end, however, often involve subjective judgment, as does deciding which terms form significant matches in the two portions. Passages may vary in the strength of their chiastic structure, for chiasmus lends itself to a variety of applications and arrangements. Analysts must be cautious, for "where the inversion is less than perfect, . . . the analysis becomes much more complex and, depending to a large extent on what is to be proved thereby, may become controversial."4
It is hoped that further thinking about criteria for identifying the presence of chiasmus will assist scholarly analysis and exegesis of scripture in several ways. First, it should promote meaningful discussion about texts. A burden of persuasion rests on any person describing a passage as chiastic. It is not sufficient merely to affix the label "chiastic." Applying this term to a given passage must be justifiable; it should be possible for a listener to discern whether a commentator has used the term properly or improperly, aptly or inaptly. Factors discussed below give a framework upon which such a judgment may be made. Second, this effort to identify criteria should assist in evaluating the degree of a proposed chiasm. Whether one passage is more or less chiastic than another can be judged most securely on the basis of specific criteria. The following fifteen factors form a basis on which the degree of chiasticity in a given passage may be assessed. Third, known criteria should assist in appraising and appreciating the noteworthy characteristics of a text. Although evaluating any work of art is to some degree subjective, it is usually possible to describe, for example, what allows one to judge a Rembrandt better than a Van Holt. Likewise, whether a composition is chiastically praiseworthy, elegant, intricate, meaningful, or significant must not be merely a matter of one observer's predilection but something that should be communicable through careful observation and articulate description.
It is clear that texts can manifest varying degrees of chiasticity. Some passages are short, and their reverse parallel order is obvious and noncontroversial. For example, Isaiah 9:21 reads: "(a Manasseh, (b) Ephraim; and (b') Ephraim, (a') Manasseh; they together shall be against Judah." Genesis 1:27 reads: "(a) God created man (b) in his own image; (b') in the image of God (a') created he him." The order of the Hebrew words in Genesis 9:6 is: "(a) Whoso sheddeth (b) the blood (c) of a man, (c') by man (b') his blood (a') shall be shed." The inverted order of the words in these passages can be observed and verified objectively and concretely. In other instances, chiasmus can also be concretely created out of linguistic features, as in poetic strophes that have chiastically varying numbers of syllables, or where words appear in an order such that their cases (nominative-accusative accusative-nominative) or genders (masculine-feminine feminine-masculine) occur in an inverted sequence. In such cases, it may be debated what significance (if any) these arrangements may have, and whether they were intentionally created or are merely accidental, but it is usually not hard to agree that the arrangement exists.
The degree of certainty about the presence of chiasmus in a text usually varies in inverse proportion to the total length of the text. In other words, the more spread out the proposed chiasm, the less certain the fact of its chiasticity becomes, except in remarkable circumstances. Hence, the more extended the proposed chiasm, the greater will be the need for multiple corroborating factors before the passage can be meaningfully described as chiastic. Not every occurrence of repetition, balance, inclusion, or symmetry will amount to something that should be called chiastic; otherwise one might purport to find chiasmus in a telephone directory.
Several factors need to be addressed before one can establish the presence of chiasmus in a given text. The more of these criteria that are significantly present in a particular case, the higher will be its degree of chiasticity. For an illustration of the application of these criteria to a specific text, readers may wish to consult the paper entitled "Chiasmus in Alma 36" and a shortened version of that paper in Rediscovering the Book of Mormon.5
1. Objectivity. To what degree is the proposed pattern clearly evident in the text? If the process of identifying chiasmus is to produce verifiable results, the inverted parallel orders must be objectively evident. If a proposed chiasm consists of elements that are objectively observable in the text, rather than depending on distant parallels or clever linkages that require imaginative commentary to explain, it is more likely that the chiastic character of the text is strong and less likely that the reader has imposed an arrangement upon the text which he or she alone has brought to it. The more evident an arrangement, the greater the degree of chiasticity. Alma 36 manifests a very high degree of objectivity, for it features 26 key words or phrases that are identical or nearly identical in both the first and second halves. At one point in Alma 36:20 the comparison between two elements is explicitly drawn.
2. Purpose. Is there an identifiable literary reason why the author might have employed chiasmus in this text? Chiasmus is useful for several purposes, such as concentrating attention on the main point of a passage by placing it at the central turning point, drawing meaningful contrasts, aiding in memorization, or emphasizing the feeling of closure upon the conclusion of a lengthy repetition. Chiastic structures can enhance the sense of a passage in ways that extend beyond the molecular meaning of individual words and phrases. It is more plausible to assert that chiasmus exists in a passage when an author appears to have intentionally put it there for a stylistic purpose. The likelihood of such intent on the part of the author should be assessed as it relates to the ideas and characteristics of the text itself. Thus, in Leviticus 24:13—23, the style of chiasmus lends itself formally to the substantive content of talionic justice. Likewise, in Alma 36, no better literary device can be imagined to convey the sense of conversion—the complete reversal of spiritual attitudes and behavior—than does chiasmus; its turning point is purposefully focused on the sacred name of "Jesus Christ, a Son of God," which appears twice at the center of the structure.
3. Boundaries. A chiasm is stronger if it operates across a literary unit as a whole and not only upon fragments or sections which overlap or cut across significant organizational lines intrinsic to the text. These bounded units may be short,6 or they may comprise a full psalm or longer pericope.7 That is, in determining whether a passage in the Psalms is chiastic, one should consider the parts of the psalm as a whole. To the extent that the proposed structure crosses over natural barriers, unnaturally chops sentences in half, or falls short of discernible boundaries in the text as a whole, the more dubious the suggested chiasm becomes. A strong example of clear boundaries is found in Helaman 6:7—13, a remarkable chiasm that encompasses the entire report for the 64th year of the reign of judges.
4. Competition with Other Forms. Chiasmus is more dominant in a passage when it is the only structuring device employed there. Chiasmus becomes less significant to the extent that a competing literary device or explanation of the arrangement of the words or thoughts more readily accounts for an apparently chiastic placement of elements. For example, "Hickory, Dickory, Dock" cannot be considered strongly chiastic because it is primarily a limerick.
5. Length. The longer the proposed chiasm, the higher its degree of chiasticity. In other words, a chiasm composed of six words introduced in one order and then repeated in the opposite order is more extensively chiastic than a structure composed of three repeated words. Having a large number of proposed elements, however, is not alone very significant, for all the elements must bear their own weight. An extended chiasm is probably not much stronger than its weakest links.
6. Density. How many words are there between the dominant elements? The more compact the proposed structure, or the fewer irrelevancies between its elements, the higher the degree of chiasticity. Tightness in the text is indicative of greater craftsmanship, rigor, focus, intention, and clarity. In assessing the density of a passage, all significant words and phrases appearing in the system must be considered. What is disregarded or omitted is often just as important as what is included. Thus, if a proposed chiasm involves only a few terms spread out over a long text, it has a low density. Thus, for example, proposed chiasms covering the entire books of 1 Nephi, Mosiah, or certain biblical books are less dense than the tight pattern found in Helaman 6:10 or Genesis 7:21—23.
7. Dominance. A convincing analysis must account for and embrace the dominant nouns, verbs, and distinctive phrases in the text. Conversely, a weak construction relies upon relatively insubstantial or common words and ideas in the text. Accordingly, powerful chiastic structures revolve around major incidents, unique phrases, or focal words, as distinguished from insignificant or dispensable parts of speech. The more significant the elements in relation to the message of the text, the greater the degree of chiasticity. In the case of Alma 36, virtually all of the words that figure into the chiastic pattern are dominant words in the account; they completely convey the essence of Alma's story.
8. Mavericks. A chiasm loses potency when key elements in the system appear extraneously outside the proposed structure. The analyst is open to the charge of selectively picking and choosing among the occurrences of this element if some of its occurrences in the text are arbitrarily ignored. What is omitted from an analysis is often just as indicative as what is included when one turns to evaluating the creative success and conceptual value of a proposed chiasm.8 Again, to use Alma 36 by way of illustration, only three words appear in this chapter outside of their respective sections in the chiastic structure.
9. Reduplication. If the same word or element appears over and over within the system, the likelihood is greater that some other kind of repetition (including random repetition) is predominant in the passage instead of chiasmus. Reduplication is not a problem in Alma 36—this chapter contains 201 words that appear only once or twice; 58 words appear three, four, or five times; and only 42 words appear more than five times, usually in balanced sections or in close proximity to each other.
10. Centrality. The crux of a chiasm is generally its central turning point.9 Without a well-defined centerpiece or distinct crossing effect, there is little reason for seeing chiasmus. Inverting is the essence of chiasmus, so the clearer the reversal at the center point, the stronger the chiasticity of the passage. The talionic formula stands squarely at the physical and conceptual center of Leviticus 24:13—23. Similarly, nothing could be more central to the dramatic message of Alma 36 than its well-defined centerpiece in verses 17—19, whose key terms are
I rememberedHarrowed up no more.
Jesus Christ, a son of GodI remembered
Jesus Christ, thou son of God
11. Balance. How balanced is the proposed chiasm? Ideally, the elements on both sides of the proposed focal point should be nearly equal, in terms of number of words, lines, or elements. It reduces clarity and focus when the two halves of a purportedly chiastic passage are not balanced. Models of balance appear in King Benjamin's speech, where the chiasm in Mosiah 3:18—19 stands almost exactly at the center of the speech. In Alma 36, 52 percent of the words appear before the turning point, and 48 percent appear afterwards.
12. Climax. A strong chiasm will emphasize the central element of the passage as its focal climax. Where the concept at the center is not weighty enough to support the concentrated attention of the reader and to bear the author's paramount intention, the chiastic force of the passage is less than the case in which the idea at the center is an important one.
13. Return. A chiasm is more complete where its beginning and end combine to create a strong sense of return and completion. Second in importance to the central crossing effect in a lengthy chiasm is the way the chiasm begins and ends. The overall structure becomes more apparent when the boundaries are clearly defined and where the passage begins and ends similarly. Thus the journey into Alma's conversion explains how it is that he knows that one will be blessed and delivered by keeping the commandments of God, and accordingly one can see why Alma's words (Alma 36:1) can be equated with God's words (Alma 36:30) as the story returns in the end to its point of departure.
14. Compatibility. The chiasticity of a passage is greater when it works comfortably and consistently together with the overall style of the author. Chiasm is more likely to be meaningfully present if its author used chiasmus or related forms of parallelism on other occasions as well. If a proposed chiastic word order is an isolated phenomenon in the writings of an author, there is a greater chance that the occurrence in question was simply accidental. Accordingly, the fact that Alma makes remarkable use of chiasmus in Alma 41:13—15 enhances even further the degree of chiasticity in Alma 36.
15. Aesthetics. Finally, there is room for subjective appreciation. Computers alone cannot identify chiasmus. Since human readers must judge an author's artistic success, further factors become relevant in assessing a passage's degree of chiasticity, such as the author's fluency with the form; consistency in sustaining the structure, balance, and harmony; pliability at the turning point (which yet does not draw undue attention to itself); and meaningful applications of the form that do not resort to subtleties so obscure as to be esoteric or awkward.
How rigorous should one be in determining whether the foregoing criteria have been satisfied in a given text? The factors mentioned above indicate the types of questions that need to be asked in identifying and evaluating a proposed complex chiasm. In assessing the results one obtains by asking and answering these questions, a text critic will need to apply qualitative and quantitative standards.
How high a degree of chiasticity should be demanded before one can comfortably describe a passage as chiastic depends primarily on how the results of the analysis are to be used. If the interpreter only wishes to identify a general sense of orderliness or balance about the text, a fairly low level of chiasticity will support such an observation. If, however, the researcher intends to use the analysis for more specific purposes (for example, to interpret the precise meaning of a given word by contrasting it with a counterpart in the chiastic structure, or to compare the style of one author with that of another), the analysis must be more rigorous. The bolder the implications to be drawn, the greater the support the analysis needs.
The Intentional Fallacy
An issue closely related to the question of chiasticity is intentionality. Although one should not fall into the trap of the "intentional fallacy" (that just because a pattern is discernible in a text the author must have intentionally put it there), neither should one assume the opposite extreme, that no chiasms were intentionally created. What factors determine whether an author can be adjudged to have consciously (or subconsciously) created the asserted pattern? How intentional does chiasmus have to be? Can these complex patterns (occasionally proposed as embracing entire books of scripture) occur by accident?
It seems reasonable to believe that occurrences of simple chiasmus (like simple instances of alliteration, rhyme, or other obvious literary effects) are consciously created in many cases. But such inversions can also occur out of habit or convention, subliminally, subconsciously, and even inadvertently. Certainly many such simple effects occur in literature written by authors who do not know the technical term for the phenomenon.
When more complex chiastic arrangements are involved, it becomes increasingly difficult to know whether the author was aware he was creating or using the form. Nevertheless, the foregoing criteria can assist in establishing a presumption of intent. Moreover, we need not demand a showing of premeditation in order to conclude that the effect was "intentional." The following points seem relevant in discussing intentionality:
1. Degree of chiasticity. The higher the degree of chiasticity, the greater the likelihood that the chiastic structure was created intentionally. Factors such as length, purpose, and compatibility are particularly probative or persuasive when considering intentionality. Thus on some occasions the degree and precision of chiastic repetition will be high enough in the works of a given author or in a particular body of literature that it becomes highly likely that the author was aware of its creation. For example, it is plausible to believe that Homer and the Homeric bards were aware of the structure involved when Odysseus in the underworld asked the shade of his mother Anticleia seven things about how she died and how things were at his home in Ithaca, and then how she responded by addressing each of these seven in exactly the reverse order.10 The length of this inverted text is exceptional; the apparent purpose of the repetition is to aid in oral recitation; indeed, it is common in epic literature for commands to be given in one order and for them to be carried out in the opposite order. In other words, the discernible degree of intentionality may relate to the degree of chiasticity discussed above. The stronger the degree of chiasticity in a passage, the greater the likelihood that the author was aware of it and intended it.
By analyzing proposed chiasms thoroughly and from a number of angles, one can assess the likelihood that an author consciously employed chiasmus in a given case to achieve a specific purpose. Nevertheless, one can rarely speak with absolute certitude in this area, since few writers ever produce commentaries on their own works. Moreover, there will probably be some circularity in one's analysis here, for some of the factors used to determine the degree of chiasticity presume some degree of intentionality (e.g., purpose), yet those factors will then be relied upon in answering the question of whether the structure was intentionally created. Thus intentionality cannot be proved directly in terms of degree of chiasticity alone.
2. The idea of intentionality. In asking whether complex chiastic patterns can occur by "accident," one may be asking the wrong question or working with an erroneous model of "artistic consciousness," for the question of artistic intent is not an either/or proposition. An author may have intended a passage—more or less—to be chiastic. One must turn, therefore, to a broader model of the artistic process to discern in part how complex literary results can be created "intentionally," yet without blueprints, charts, or handbooks. Consider, for example, the improvisation of New Orleans Dixieland musicians. They ad lib, usually 8 or 16 measures at a time. While their music is spontaneous and "unconscious" (they are not reading music and have not sat down to figure out in advance what they are going to play), their complex rhythms, melodies, and chord progressions are nevertheless discernibly regular, structured, and organized. One would not call their music "accidental." Nor would one call it strictly "intentional." It fits, however, into a pattern, style, or convention that has become so natural with the jazz musician that the music just comes out that way. An old banjo player, when asked around 1850 if he could read music, innocently replied: "Can I read notes? . . . There are no notes to a banjo. You just play it."11
In a literary context, some poets and authors working within a literary tradition may likewise create complex artistic effects without being conscious of every facet of their compositions. Many people, notably children, regularly employ complex rules of syntax and language without "knowing" what they are doing. T. S. Elliot was once asked by a ladies' literary group in Oxford to explain what one of his poems meant. As many poets would have responded, he replied that he had no idea what the poem meant, explaining that it did not mean anything except in the minds of his listeners and readers. Yet his poetry is by no means meaningless, formless, or random.
In much the same way, while the degree of chiasticity in some chiasms may be strong enough to claim that they were expressly designed and implemented, one must expect that other chiastic patterns discovered by textual exegetes were not methodically crafted by the author to conform mechanically to some rigorously prescribed template. That does not, however, mean that the form was "unintentional" or "accidental"—the question of intentionality is not a black-and-white matter. Instead, it may be possible that some chiasms emerged out of the broader expressiveness of an author. Like the notes that sound good to the jazz musician who is deeply conditioned in certain cultural patterns, the words felt right coming out that way, given the conventions and structure of the author's language and literary culture. As rhythm and blues are at home with certain musicians, parallelism and chiasm were more a part of some languages and literatures than of others, as careful listening and reading will reveal.
3. The eye of the beholder. Finally, it is possible in some cases to conclude that a proposed chiasm exists exclusively in the eye of the beholder. If no claim of author intent can be mounted in a given text, this should be acknowledged, but should not preclude a careful observer from still appreciating the way in which an underlying orderliness or pattern happens to enhance the elegance and artistic achievement of the composition. If this is all that can be said of a particular textual phenomenon, however, it will change the way in which the literary analysis of the passage should be presented, what the analysis can claim for itself, and what conclusions or implications it will support.
In conclusion, I hope that the foregoing discussion will stimulate further thought and careful literary analysis of texts. Most aesthetic forms of literature and art do not lend themselves easily to formulaic definition or complete description, and the chiastic form is no exception. Thus it is not imagined that these proposed criteria will convert the study of chiasmus into a science. Nevertheless, it is surely possible for those who are interested in identifying and discussing scriptural instances of chiasmus to be clearer about their subject. Many proposed chiasms are impressive and interesting; others appear to be contrived or unremarkable. Textual analysts should be able to examine such instances of chiasmus and select among those that are better or worse, and then articulate reasons why they think some are better constructed or manifest a higher degree of intentionality than others. In my experience, chiastic passages manifest varying degrees of chiasticity. The criteria set forth in this paper are the main factors I consider in appraising the strength of one proposed example of chiasmus vis-Ã -vis another. Having advanced these criteria informally for comment in 1989, I now circulate these ideas more widely for further refinement and possible use.
1. A convenient listing of several hundred books and articles recognizing and employing chiasmus as a tool of literary criticism is found in John W. Welch, "Chiasmus Bibliography" (Provo, UT: FARMS, 1987).
2. See Nils Lund, Chiasmus in the New Testament (Chapel Hill: University of North Carolina, 1942), 40-41; see further the introduction in John W. Welch, Chiasmus in Antiquity (Hildesheim: Gerstenberg, 1981), and John W. Welch, "A Study Relating Chiasmus in the Book of Mormon to Chiasmus in the Old Testament, Ugaritic Epics, Homer, and Selected Greek and Latin Authors" (master's thesis, Brigham Young University, 1970), 6—17, proposing definitions that would allow critics to distinguish between simple, compound, and complex chiasmus; see also D. J. Clark, "Criteria for Identifying Chiasm," Linguistica Biblica 35 (1975): 63—72.
5. John W. Welch, "Chiasmus in Alma 36" (Provo, UT: FARMS, 1989), and John W. Welch, "A Masterpiece: Alma 36," in Rediscovering the Book of Mormon, ed. John L. Sorenson and Melvin J. Thorne (Salt Lake City: Deseret Book and FARMS, 1991), 129—30.
8. Along this same line, Yehuda Radday has developed statistical formulas for measuring the extent to which the order of words in a passage deviate from the ideal chiastic order. See his essay "Chiasmus in Hebrew Narrative," in Welch, Chiasmus in Antiquity, 50—115.
9. Nils Lund, Chiasmus in the New Testament, 41, asserts this as the first and foremost law of chiasmus: "The centre is always the turning point. At the centre there is often a change in the trend of thought and an antithetic idea is introduced." In meetings of the Hebrew Poetry Group of the Society of Biblical Literature in the 1980s, others have also suggested that the extremes may be as important as, if not more significant, than the center of a chiasm. I find Lund persuasive on this point, but even he ranked the extremes second only to the center in importance.