POINTING TO THE ACCENTS IN THE SCROLL : FUNCTIONAL DEVELOPMENT OF THE MASORETIC ACCENTS

IN THE HEBREW BIBLE

Sung Jin Park Midwestern Baptist Theological Seminary

It is widely recognized that the accents developed by the Masoretes have three functions: they mark (1) stress, (2) musical notations, and (3) punctuation. However, the order in which these functions developed remains contestable. The present study demonstrates that indication of stressed syllables was a secondary rather than primary function that became dominant at a relatively late stage. It is most likely that the punctuational function of the accents for recitation gave rise to other relational divisions of a unit. Then the accents came to be used for cantillation in addition to punctuation, although this never fully developed into a system of musical notation. The masoretic accents that we know are thus reflective of an unfinished transition to such notation, intended for singing, from punctuation designed for recitation.

1. THREE FUNCTIONS OF THE MASORETIC ACCENTS

There are two systems of accentuation in the Masoretic manuscripts of

the Hebrew Bible: one used in the Three Books Psalms, Proverbs, and Job (except for the narrative portion of Job, 1:1 3:1 and 42:7 17) and the other in all of the remaining material. The present study examines only the latter, simpler system, although both of them are functionally the same in terms of the issues to be discussed here. Concerning the functions of the masoretic accentuation system, nearly all scholars agree that the accents have three distinctive functions.1 The first is to indicate stress, with the marker usually placed above or below the main stressed syllable of each word. As in many other languages, stress in the masoretic accentuation

1 W. Wickes, A Treatise on the Accentuation of the Twenty-One So-Called Prose Books of the Old Testament (London: Clarendon, 1887), pp. 1 2; M. Breuer, " " (The accents of the Bible in the twenty-one books and the three books; Jerusalem: Mikhlalah, 1982), pp. 3 9; I. Yeivin, Introduction to the Tiberian Masorah (SBLMS 5; Missoula: Scholars Press, 1980), p. 178; M. Aronoff, Orthography and Linguistic Theory: The Syntactic Basis of Masoretic Hebrew Punctuation, Language 61 (1985): 32 36; D. M. Weil, The Masoretic Chant of the Bible (Jerusalem: Rubin Mass, 1985), p. 4; E. J. Revell, Hebrew Accents and Greek Ekphonetic Neumes, in Studies in Eastern Chant (ed. M. Velimirovi ; London: Oxford University Press, 1979), 4:140 170; D. B. Weisberg, The Rare Accents of the Twenty-One Books, JQR 56 (1966): 317; N. Janis, A Grammar of the Biblical Accents (Ph.D. diss., Harvard University, 1987), pp. 3 6; A. Dotan, prolegomenon to W. Wickes, Two Treatises on the Accentuation of the Old Testament (Jerusalem: Ktav, 1970), pp. vii viii.

Hebrew Studies 55 (2014) 74 Park: Pointing to the Accents”

system is phonemic, that is, the meaning of a word can be changed depend-ing on which syllable is stressed.2 For example,

Then his brothers said to him, Are you actually going to reign over us? Or are you really going to rule over us? (Gen 37:8).

the tower that men had built (Gen 11:5).

When the sun had set (Gen 15:17).

Behold, Rachel his daughter is coming (Gen 29:6).

Since the position of the stress is phonologically significant in Biblical Hebrew, it is reasonable to suppose that marking its position would be an important role for any diacritical system. Marking stress, however, should not be considered the primary function of the masoretic accentuation system since, if this were the only function, the Masoretes would not have needed so many different accents. Furthermore, accents for indicating stress over each word cannot be found at the early stage of the Babylonian Masorah as well as in the Proto-Palestinian Masorah.3 In fact, even in the Tiberian system, some postpositive or prepositive accents, such as segolta, dechi, and telisha are not marked over or below the stressed syllables.

For this reason, Werner suggests that this function of accents is a later development.4 It is worth noting, however, that some Palestinian manu-scripts show more detailed features of conjunctives than the Tiberian system

2 J. Blau, A Grammar of Biblical Hebrew (PLO 12; Wiesbaden: Harrassowitz, 1976), p. 19. Revell states, The claim that stress position is phonemic in Hebrew cannot be maintained unless it can be shown that its

retraction occurs under particular conditions (E. J. Revell, Nesiga (Retraction of Word Stress) in Tiberian Hebrew [Textos y Estudios Cardenal Cisneros 39; Madrid: Instituto de Filología, 1987], p. 9). 3 A. Dotan, Masorah, EncJud 13:637; S. Morag,

(Some aspects of the methodology and terminology of the early Massoretes), Leshonenu 38 (1973 1974): 49 77. 4 E. Werner, The Sacred Bridge: The Interdependence of Liturgy and Music in Synagogue and Church during the First Millennium (New York: Columbia University Press, 1959), p. 413.

Hebrew Studies 55 (2014) 75 Park: Pointing to the Accents”

but do not place their accents upon stressed syllables.5 Revell thus concludes that marking stress positions was not the primary concern of the Palestinian accentual signs.6 Moreover, Shoshany recently observed that accents over stressed syllables are found even in a few early Babylonian manuscripts.7 Therefore, the function of marking stress should not be considered a later development. Although we have no documental evidence regarding the existence of stress in Biblical Hebrew, various phenomena reflecting stress positioning can be reasonably postulated in it.8 It is evident that stress existed in the biblical period since its existence on a word is more ancient than any other phonological phenomena in Biblical Hebrew; stress is a suprasegmental quality innate to a word in a language.9 If so, why does this function show up mostly in the later Masoretic tradition? The reason might be that the Masoretes did not feel an urgent need to mark the stressed sylla-bles since the text s readers already knew where to put the stress. In any case, it is almost certain that marking it was not considered a major purpose of the masoretic accentuation system.

The second function of the masoretic accents is musical and the third is punctuational. Although scholars debate which of these two is primary, due to several reasons, most tend to believe that it is the former. First, cantilla-tion of biblical texts guided by the masoretic accentuation system is prac-ticed in the synagogue to this day.10 Wickes comments,

5 E. J. Revell, Hebrew Accents, p. 148. 6 E. J. Revell, Hebrew Accents, p. 148. 7 R. Shoshany, (A new aspect of correlation in the development of the biblical vocalization and accentuation in the Babylonian tradition), in (Israel: Linguistic studies in the memory of Israel Yeivin; ed. R. I. Zer and Y. Ofer; Jerusalem: Magnes, 2011), pp. 251 267. 8 A. Goetze, Accent and Vocalism in Hebrew, JAOS 59 (1939): 431 459; J. C. L. Gibson, Stress and Vocalic Change in Hebrew: A Diachronic Study, Journal of Linguistics 2 (1966): 35 56; Z. S. Harris, Development of the Canaanite Dialects: An Investigation in Linguistic History (New Haven: American Oriental Society, 1939); A. Poebel, The Antepenult Stressing of Old Hebrew and Its Influence on the Shaping of the Vowels, AJSL 56 (1939): 225 230; A. Poebel, Penult Stressing Replacing Ultimate Stressing in Pre-Exilic Hebrew, AJSL 56 (1939): 384 387; G. Bergsträsser, Hebräische Grammatik (Leipzig: F. C. W. Vogel, 1918 1929). In the article listed above, Goetze comments, The rearrangement of vocalic quantities in Hebrew took place under an Early Hebrew accent which was different from that of Masoretic Hebrew. Hence, the interpretation of Hebrew vocalism has largely become a matter of accentuation (A. Goetze, Accent and Vocalism p. 453). 9 It is beyond the shadow of doubt that vowels and consonants of a word are the most fundamental phonological entities: S. J. Park, A Study on Poetic Typology of Biblical Hebrew Poetry: Generative Metrical Approach (Ph.D. diss., Hebrew Union College-Jewish Institute of Religion, 2012), pp. 22 24. 10 I. Yeivin, Introduction to the Tiberian Masorah, p. 178; M. Breuer, The Accents of the Bible, pp. 3 9; D. M. Weil, The Masoretic Chant ABD 4:594 596; E. J. Revell,

142.

Hebrew Studies 55 (2014) 76 Park: Pointing to the Accents”

The Hebrew accentuation is essentially a musical system. The accents are mu-sical signs originally designed to represent and preserve a particular mode of cantillation or musical declamation, which was in use for the public reading of the Old Testament text at the time of their introduction, and which had been handed down by tradition from much earlier times.11

Second, the accents musical function may be also supported by early rabbinical references. According to b. Meg. 32a,

' '

' R. Shefatiah further said in the name of R. Johanan, As for one who reads (the Scripture) without melody and studies (the Mishnah) without a tune, the Scripture says about him, Wherefore I gave them also statutes that were not good etc.

Likewise in b. Ber. 62a:

'

'

' Why should one not wipe with the right hand but with the left? Raba said, Because the Torah was given with the right hand, as it is written, At His

right hand was a fiery law unto them [Deut 33:2]. Rabbah b. Bar Hanah said, Because it is brought to the mouth. R. Simeon b. Lakish said, Because one binds the tefillin [on the left arm] with it. R. Nahman b. Isaac

said, Because one points to the accents in the scroll with it. Similarly, R. Eliezer says, Because one eats with it. R. Joshua says, Because one writes with it. R. Akiba says, Because one points with it to the accents in the scroll.

Rashi thought that the term the accents of the Torah referred to the cantillation of Scripture, with corresponding movements made with

the right hand according to the musical accents, since it was prohibited to use an accentuated scroll for public reading. Furthermore, he mentioned that the Palestinian readers of his day (late eleventh century) practiced cheironomy when learning the Scripture.12

11 W. Wickes, A Treatise on the Accentuation, pp. 1 2. 12 As mentioned by Werner, cheironomy was a very important ancient practice, especially in the cultic settings of Egypt, Mesopotamia, and ancient Greece (E. Werner, The Sacred Bridge, pp. 107 109). On

Hebrew Studies 55 (2014) 77 Park: Pointing to the Accents”

Compelling as they may seem, these two considerations, based on Jewish tradition, are not entirely adequate because some other Talmudic passages support a punctuational function for the accents. According to b. Ned. 37b,

"

" " < >"

"

As for one who maintains that payment is for the teaching of accentuation, why does he reject the view that it is for acting as guardian? He argues, Do daughters then need guarding? Then, as for one who maintains that the fee is for guardianship, why does he reject the view that it is for teaching accents? He holds that accents are also biblical. For R. Ika b. Abin said in the name of R. Hananel in Rab s name, What is the meaning of, And they read in the book, in the law of God, distinctly, and they gave the sense, so that they understood the reading [Neh 8:8]? They read in the book, the law of God refers to Scripture; distinctly , to Targum; and they gave the sense , to the division of sentences; so that they understood the reading , to the accen-tuation; others say, to the masoroth.

Likewise in b. Hag. 6b (= b. Yoma 52b),

'

R. Hisda asked, How is this verse to be understood, And he sent the young men of the children of Israel, who offered burnt-offerings lambs, and sacri-ficed peace-offerings of oxen unto the Lord ? What if both were oxen? What difference does it make? Mar Zutra said, In regard to the punctuation . R. Aha, the son of Raba, said, In regard to one who says, I vow [to offer] a burnt-offering like the burnt-offering which Israel offered in the wilderness , what [must he offer]? Were they oxen or lambs? It remains [undecided].

R. Hisda s question is, apparently, whether Exod 24:5 should be under-stood as saying that both burnt offerings and peace offerings ordered by Moses consisted of oxen. In response, Mar Zutra points to the atnakh on

EncJud 11:656 664.

Hebrew Studies 55 (2014) 78 Park: Pointing to the Accents”

, interpreting it as a punctuation mark indicating that a sentence ends here and that consequently the reference to oxen after this sign does not necessarily apply to burnt offerings mentioned before it. R. Hisda s addition of the word sheep to the first half of the verse (whose Masoretic text reads ) is therefore justified.13

This brings us to the third consideration, based not on extra-biblical tra-dition but on internal evidence. If punctuation was the primary function of accents, it is unclear why the Masoretes needed so many of them.14 The existence of numerous disjunctives would still be understandable due to their hierarchical and distributive characteristics. However, since there is no hierarchical relationship among conjunctives, musical value is the only pos-sible explanation of their existence.15 In particular, according to the rules of dichotomy, some disjunctives appear in combinations only with certain conjunctives.16 The conjunctive munakh is the first servus of revia, always being followed by it.17 Thus, Wickes comments, This formal dichotomy necessarily supplied (as far as it went) the basis for the musical, and from its constant recurrence seems to have suggested to the originators of the accen-tual system a guiding principle for the musical divisions in general. 18 Simi-larly, Revell contends that the different combinations among the accents may relate to the music of several different forms of chant, calling groups of accents motifs or tropes. 19

At the same time, recent scholarship has recognized that the division pattern in the masoretic accentuation system is quite close to the perfor-mance structures based on the prosodic structure of sentences.20 The main

13 E. Tov, Textual Criticism of the Hebrew Bible (2nd ed.; Minneapolis: Fortress, 2001), p. 69. 14 M. Breuer, The Accents of the Bible, p. 20. 15 Price offers an ordered ranking of conjunctives, but this ranking refers to the order in which they are arrayed before their governing disjunctive accents rather than to the conjunctives own hierarchy (J. D. Price, The Syntax of Masoretic Accents in the Hebrew Bible [SBEC 27; Lewiston: Edwin Mellen, 1990], p. 34). 16 J. D. Price, The Syntax of Masoretic Accents, pp. 32 35; I. Yeivin, Introduction to the Tiberian Masorah, pp. 176 218; W. Wickes, A Treatise on the Accentuation, pp. 29 129. 17 I. Yeivin, Introduction to the Tiberian Masorah, pp. 192 193; W. Wickes, A Treatise on the Accentua-tion, pp. 97 98. Munakh also regularly plays the role of the first servus of zarqa. 18 W. Wickes, A Treatise on the Accentuation, pp. 29 30 (italics his). The law of substitution in the accentuation system also supports the musical purpose of the accents. See J. D. Price, Syntax of Masoretic Accents, pp. 31 32, 154 155. 19 E. J. Revell, Masoretic Accents, p. 594. 20 Although the performance structures had been researched by psycholinguists, it was Janis and Dresher who first connected the pattern of these structures with the division pattern of the masoretic accentuation

Language 70 [1994]: 1 52).

Hebrew Studies 55 (2014) 79 Park: Pointing to the Accents”

purpose of a performance structure is to determine the prominence of boundaries in the case of a real utterance by obtaining empirical data such as pausal duration, transitional error probabilities, and parsing values.21 About 70% of all pauses occur at major constituent breaks.22 The division pattern in the masoretic accentuation system is similar to the performance structures in several respects. First, both are characterized by hierarchy. Second, the lay-out in both cases is more or less symmetrical. This means that the most prominent break occurs at the midpoint of a sentence while those next in prominence take place in the midpoint of the half-sentence circumscribed by the most prominent break.23 For example, performance structures can be schematically depicted as follows:

She discussed the pros and cons to get over her suprisingly apprehensive feelings. 4 15 1 9 1 25 1 5 8 6 16 10

In addition to his files the lawyer brought the office s best adding machine 2 10 3 3 33 1 8 17 0 10 13

21 Cognitive Psychology 15 (1983): 411

Cognitive Psychology 11 (1979): 58 81; A. Christophe, MThe Case of the Head- Developmental Science 6 (2003): 211 220. 22

Phonetica 31 (1975): 144184. 23 81.

Hebrew Studies 55 (2014) 80 Park: Pointing to the Accents”

Here the number between the words indicates the pause duration per-centage. In the first sentence, the number 25 between cons and to refers to 25% of the pausal breaks that take place when the whole sentence is spo-ken. It is by far the longest break or the most prominent pause, and it is located approximately in the middle of the sentence.24 The second-longest breaks, marked by the numbers 15 and 16, take place approximately halfway between the most prominent pause and the boundaries of the sentence, re-sulting in a symmetric structure. In the second sentence, the longest break is skewed toward the beginning but the second-longest ones remain centered.

Of course, we cannot prove with certainty that the masoretic accentuation system divisions and the performance structure are identical, if only because we do not know how the ancient Hebrew speakers read their biblical texts. At the very least, however, we do know that the division patterns based on the rule of dichotomy in the masoretic accentuation system and the pause duration percentage in performance structures share the same linguistic con-cept regarding the prosodic structures of sentence (i.e., pause in real speech).25 The masoretic accentuation system divisions are thus interpretable as suggesting that the major purpose of the masoretic accents was to mark punctuational pauses for proper recitation.

2. MUSICAL OR PUNCTUATIONAL?

How then do we make the choice concerning the main function of accent

in the masoretic system? Was it punctuational for recitation or musical for song? Admittedly, it is by no means a simple task because the masoretic ac-centuation system clearly indicates its musical elements by means of various kinds of conjunctives and patterned accent combinations while also desig-nating its punctuational elements by means of the pausal duration pattern. Did it serve these two purposes concurrently or consecutively?

In this respect, two studies are worth noting. Revell has observed that a considerable proportion of pausal forms can be found in Biblical Hebrew with minor disjunctive accents and even with conjunctive accents.26 Based

24 416. 25 Park demonstrates how the prosodic boundaries (phonological/intonational phrases) divided by the masoretic accents resemble the metrical boundaries of Biblical Hebrew poetry. He comments, The divisions by the Tiberian accentuation system are the prosodic representation, and they are legitimate colometric divisions for biblical Hebrew verse since they are phonologically connected with the metrical hierarchy Park, A Study on Poetic Typology of Biblical Hebrew Poetry, p. 129). 26 E. J. Revell, Pausal Forms in Biblical Hebrew: Their Function, Origin and Significance, JSS 25 (1980):

iblical Hebrew the language of the Tiberian Masorah as represented by the received text of the Hebrew Bible, including the vowel and accent signs.

Hebrew Studies 55 (2014) 81 Park: Pointing to the Accents”

on this observation, he offers some suggestions: (1) the accent system cannot be the conditioning factor of the pausal forms because the pausal forms and the accents reflect different ways of grouping the words of the text into semantic units; (2) pausal forms originated from the intonation patterns of normal speech because the division of speech into sense units was delimited by intonation patterns; (3) these intonation patterns were later replaced by the musical patterns of chanting controlled by the accents.27 Revell suggests that the pausal forms reflect an early stage of the reading tradition prior to the period of the Masoretes and the masoretic accents, which reflect the later musical patterns.28 In other words, the punctuational function reflected in pausal forms of the reading tradition developed earlier than the musical function reflected in the masoretic accents. The latter replaced pausal forms; thus, we can see only their rare traces in pausal forms with minor disjunctive accents and conjunctives.

To prove his hypothesis, Revell must first figure out the interrelationship between pausal forms and intonation patterns based on normal speech. Yet, he does not provide any evidence showing how these pausal forms arose from intonation patterns or how the divisions delimited by intonation pat-terns differ from those delimited by the masoretic accentuation system. In his recent article, Revell proposes that the occurrence of pausal forms is de-termined by the marking of the main divisions denoted from the significant semantic or syntactic unit, not by the accentuation system.29 His suggestion, however, is not without problems. The foremost one is Revell s lack of at-tention to the fact that pausal forms with minor disjunctives and with con-junctives occur very rarely.30 According to Price s data on the distribution of pausal forms in the prose books of the Hebrew Bible, 97.6% of pausal forms (35923 out of 36798) occur with atnakh or silluq.31 Even if one assumes for the sake of argument that Revell s hypothesis is correct, such a small per-centage of pausal forms occurring with minor accents (including conjunc-tives) signifies that the word groupings indicated by recitation and by

27 E. J. Revell, Pausal Forms 179. 28 E. J. Revell, Pausal Forms, p. 178. He insists that there were two separate orthographic systems: one accentual and another based on pausal phonology. 29 E. J. Revell, The Occurrence of Pausal Forms, JSS 57 (2012): 213 230. 30 Revell found about 465 pausal forms in the book of Deuteronomy. Almost 76% of these occur with atnakh or silluq and 19% with segolta, zaqef, and tifkha. Only 21 forms (5%) appear with the minor disjunctive accents: nine with revia, five with pashta, three with telisha, and one each with zarqa, tevir, geresh, and the conjunctive munakh l Forms, pp. 165 166). 31 J. D. Price, Exegesis and Pausal Forms with Non-Pausal Accents in the Hebrew Bible (paper presented at the Southeastern regional meeting of the ETS, March 18, 2006), p. 5. Online: http://www.jamesdprice .com/masoreticstudies.html.

Hebrew Studies 55 (2014) 82 Park: Pointing to the Accents”

cantillation are nearly identical. Rather than reflecting another kind of word grouping, as Revell has suggested, pausal forms with minor accents may ap-pear for other purposes. Based on an exhaustive computer analysis, Price maintains that pausal forms with the so-called non-pausal accents (any ac-cents other than atnakh and silluq or ole weyored) mark exegetically sig-nificant words or phrases in the text with rhetorical emphasis.32 For example, in Isa 49:18,

Lift up your eyes around and see. All of them gather together, they come to you. As I live, says the Lord, you shall surely put on all of them as jewels, and bind them on as a bride.

As in some other verses, the conjunctive munakh marks the pausal phrase

with a rhetorical emphasis, which is exegetically important. In the second study, Shoshany presents a peculiar methodology regarding

the relationship between recitation for reading and cantillation for singing.33 She builds upon the prosodic structural marker theory developed by Stein and Gil.34 In a seminal 1980 article, they isolated five consistent and univer-sal prosodic markers stress, number of syllables, sonority, syntactic com-plexity, and semantic import.35 Later, Gil further developed this theory by examining 146 languages and suggested that

iambic languages are characterized by SOV [Subject-Object-Verb] basic word order, simple syllable structures, high consonant-vowel ratios, and the absence of phonemic tones, whereas trochaic languages are typified by SVO [Subject-Verb-Object] basic word order, complex syllable structures, low consonant-vowel ratios, the presence of phonemic tonal distinctions.36

32 J. D. Price, Exegesis and Pausal Forms, pp. 1 21. 33 R. Shoshany, (The original purpose of biblical accentuation), in (Mas at Aharon: Linguistic studies presented to Aron Dotan; ed. M.

Bar-Asher and C. E. Cohen; Jerusalem: The Bialik Institute, 2009), pp. 469 486. This article is a revised version of a sub-chapter in her doctoral dissertation: Babylonian Accentuation System: Rules of Division and Accentuation, Stages of Development, and Relationship to the Tiberian System (Ph.D. diss., Tel-Aviv University, 2003). 34 Theoretical Linguistics 7 (1980): 173

Folia Linguistica 20 (1986): 165 231. 35 D. Stein and D. Gil 240. 36 D. Gil, Prosodic Typology, p. 167.

Hebrew Studies 55 (2014) 83 Park: Pointing to the Accents”

Gil s main claim is that poem for recitation tends to be predominantly iambic, while song for singing tends to display trochaic structures.

Applying Gil s hypothesis to the Babylonian and the Tiberian systems, Shoshany divides the former into two stages: (1) early and relatively simple Babylonian accentuation; (2) late, more complex Babylonian accentuation combined with Tiberian conjunctive accents.37 She then analyzes the pro-sodic structures of both phases and compares the results with the Tiberian system. One of her examples is Isa 49:22:

... - ...

The first line represents the early Babylonian system with the text di-

vided according to the rules of its accentuation; the second line, the Tiberian system. According to the former, the verse is divided into two parts: the first consists of one word and the second includes four. Thus the word ratio is 1:4: it is iambic, according to Gil s theory, since a weak component (i.e., a lesser number of syllables) is followed by a strong one (i.e., a greater num-ber of syllables). In the Tiberian system, by contrast, the word ratio is 3:1. It is trochaic since a strong component is followed by a weak component. Shoshany concludes, therefore, that the early Babylonian system is iambic whereas the Tiberian system is trochaic.38 In other words, the early Babylonian system more closely resembles poetry in recitation, but the Tiberian system more closely resembles song in singing.

Among the 1948 verses examined by Shoshany, 1665 (85.47%) are di-vided identically in the Babylonian and Tiberian systems, 251 (12.89%) display iambic division in the early Babylonian system and trochaic division in the Tiberian system, and 32 (1.64%) reverse the latter pattern.39 Based on these data, she concludes that, since the Tiberian accentuation system de-veloped from the early Babylonian one, accentuation was originally intended for recitation and only later developed for music.40

37 R. Shoshany, Babylonian Accentuation System, pp. 57 378. This grouping was originally suggested by A. Spanier, Die massoretischen Akzente. Eine Darlegung des Systems nebst Beiträgen zum Verstandnis ihrer Entwicklung (Berlin: Akademie-Verlag, 1927), but Shoshany further subdivides it into four sub-groups: early a, late a, b, and c. 38 R. Shoshany, The Original Purpose of Biblical Accentuation, p. 477. 39 R. Shoshany, The Original Purpose of Biblical Accentuation, p. 480. 40 R. Shoshany, The Original Purpose of Biblical Accentuation, p. 484.

Hebrew Studies 55 (2014) 84 Park: Pointing to the Accents”

Shoshany s hypothesis is thought-provoking. It does, however, have some vulnerabilities. Most importantly, Gil s theory, which serves as the foundation of Shoshany s work, has been challenged. Schweiger, in particu-lar, critiques his treatment of the statistical data concerning the languages that he examined. First, Schweiger proposes that Gil s sampling is not repre-sentative and offers representative data that demonstrate a similar average consonant-vowel ratio among both Subject-Object-Verb and Subject-Verb-Object languages. This means that it is impossible to judge which of them are more iambic or trochaic in their linguistic structures. Second, Schweiger points out that there is no homogeneity in Gil s data: for example, the range of the average consonant-vowel ratio in Subject-Object-Verb languages is too broad, from 3.79 to 4.49.41 Yet another problem is that Gil s theory is too abstract with regard to the iambic or trochaic qualities. In prosodic theory, the baseline for distinguishing between an iamb or a trochee is estab-lished by foot-leveled examination.42 Gil, however, begins with the level of the sentence, which is divided into two parts by the accentuation rules of dichotomy; therefore, his analysis depends on the number of words in each part. Consider, for example, Ps 23:4:

<^> ...

In the Tiberian system, the word ratio is 2:2. Is the verse iambic or trochaic? Or is it treated as relatively iambic in the Tiberian system when compared with the early Babylonian system? In sum, Gil s suggestion that whether the prosody is iambic or trochaic is determined by the word ratio is not compel-ling. It is overly simplistic and mechanical in determining whether the given text was intended for reading or singing.

Based on the discussion so far, neither Revell s studies nor that of Shoshany is helpful for determining the masoretic accentuation system s functional development. Here, it is worth its while to turn our attention to the empirical insight of modern cognitive theorists that there are no apparent dif-

41 Folia Linguistica 24 (1990): 123 125. 42 B. Hayes, The Prosodic Hierarchy in Meter, in Phonetics and Phonology, Volume 1: Rhythm and Meter (San Diego: Academic Press, 1989), pp. 201 202; E. O. Selkirk, Phonology and Syntax: The Relation between Sound and Structure (Cambridge: MIT Press, 1984).

Hebrew Studies 55 (2014) 85 Park: Pointing to the Accents”

ferences between intonation patterns of normal speech and musical melodies except for some minor factors, such as duration.43 According to this re-search, it would not be surprising if by their nature the masoretic accents were both punctuational (in their distribution of pauses for normal speech) and musical (in their various combinations to produce music motifs). The two functions likely coexisted in the masoretic accentuation system.

3. FUNCTIONAL DEVELOPMENT OF THE MASORETIC ACCENTUATION

SYSTEM With the above discussion in mind, is it possible to determine which of

this system s two functions was the first to develop? Werner suggests that as a norm, punctuation emerges earlier than musical notation.44 Can his sugges-tion be applicable to the masoretic accentuation system? Several considera-tions are relevant in answering this question. To begin with, there are three reasons to believe that development from major disjunctive accents to conjunctives within the masoretic accentuation system took place in several stages.

The first reason is Revell s observation that the distribution of spacings in Rylands Greek Papyrus 458, one of the oldest Septuagint manuscripts, is identical to that of some major disjunctive accents in the Masoretic Bible.45 Since Papyrus 458 is dated to the second century B.C.E., the colometric tradition concerning major disjunctives must be relatively old. In Revell s words,

For the Hebrew Bible, [the manuscript] shows clearly that the basis of the system of cantillation represented by the later accents was already firmly established in the second century B.C.E., and was so much a part of the formal reading of the Torah, that it was also used for the Septuagint.46

43 A. Patel, Music, Language, and the Brain (Oxford: Oxford University Press, 2008), esp. pp. 96 299; D. Proceedings of the National Academy of

Sciences of the United States of America 104 (2007): 9852 9857; D. L. Bowling, K. Gill, J. Choi, J. Prinz, Journal of the

Acoustical Society of America 127 (2010): 491Toronto Working Papers in Linguistics 27 (2008):

43 58. 44 E. Werner, Sacred Bridge, p. 104. 45 E. J. Revell, The Oldest Evidence for the Hebrew Accent System, BJRL 54 (1971/1972): 214 222; E. J. Revell, Biblical Punctuation and Chant in the Second Temple Period, JSJ 7 (1976): 181 198; E. J. Revell, Pausal Forms in Biblical Hebrew, JSS 25 (1980): 165 179; E. J. Revell, Pausal Forms and the Structure of Biblical Poetry, VT 31 (1981): 186 199. 46 E. J. Revell, Oldest Evidence, p. 222.

Hebrew Studies 55 (2014) 86 Park: Pointing to the Accents”

Second, originally conjunctives were not counted in ancient tradition among the since they appear with words in phrases as connectors to the following disjunctives.47 Disjunctive accents indicate stresses as well as phrasal divisions ranging from full stop to various shades of shorter pauses. 48 Hence, disjunctives function as separators. Stress, however, can-not be recognized with disjunctives alone because by its nature, it should be distributive to show its conspicuous emphasis.49 For this reason, it became necessary to develop an additional accent type conjunctives. They were not initially categorized as accents but later they came to be treated as such because they bear relatively weak stresses as connectors between domi-nant disjunctive accents.50

Third, there were no conjunctives in the earlier Babylonian tradition, whose accentuation system bears better witness than the Tiberian one to the initial stages of the accents development. Of course, lack of conjunctives in relatively early sources does not definitively prove that these signs did not exist at that time. However, taken together the three arguments that have just been outlined do indicate the functional primacy of major disjunctives over conjunctives, probably signifying that the former were the first to emerge. Here it is important to mention that this applies only to major disjunctives because some rare accents like shalshelet and patser gadol were certainly added much later by medieval Jewish communities.51 It is apparent, there-fore, that the masoretic accentuation system did not emerge in its final shape but rather developed in stages over time. It is a mixture of practices, some of which reflect ancient traditions while others were developed by the Masoretes for their own exegetical purposes.52 Although it is difficult to separate the two elements, the punctuational function of the disjunctive ac-cents should be considered ancient, most likely of earlier provenance than conjunctives.

47 M. Breuer, (The punctuation of the accents in the Bible; Jerusalem: Hahistadrut Hatsionit, 1958), p. 58. 48 B. K. An Introduction to Biblical Hebrew Syntax (Winona Lake: Eisenbrauns, 1990), p. 29; M. B. Cohen, The System of Accentuation in the Hebrew Bible (Minneapolis: Milco, 1969), pp. 4 5, 59. 49 B. Hayes, Metrical Stress Theory: Principles and Case Studies (Chicago: University of Chicago Press, 1995), pp. 24 Linguistic Inquiry 8 (1977): 334; E. O. Selkirk, Phonology and Syntax. 50 M. Breuer, The Punctuation of the Accents, p. 58. 51 D. B. Weisberg, The Rare Accents, JQR 56 (1966): 315 336; D. B. Weisberg, The Rare Accents, JQR 57 (1967): 57 70, 227 238. 52 E. J. Revell, The Oldest Accent List in the Diqduqe Hat.e amim, Textus 8 (1973): 138 159; D. B. Weisberg, The Rare Accents, pp. 326 333.

Hebrew Studies 55 (2014) 87 Park: Pointing to the Accents”

Further, the masoretic accents are not musical notes but rather musical neumes. Werner well explains the difference as follows:

The basic distinction between a neume and a note is that: the former usually stands not for a single note, but for a whole phrase, whereas modern notation has one sign for each individual note. Not all early neumes and none of the ecphonetic systems can be transcribed in modern notation without the help of oral tradition.53

In most cases, synagogue and church practices display a trend of gradual transition from ecphonetic accents to a more exact musical notation.54 The standard masoretic accentuation system, with its various combinations of the accents (i.e., neumes), made it possible for ancient readers to chant the bibli-cal texts with ambiguous directions of upward or downward voice move-ment in accordance with an orally transmitted tradition. However, that is not the same as singing these texts because chanting is in fact a form of speech marked by cadential formulas.55

Finally, since the masoretic accents are not musical notes there is no way with them to write down rhythmic flow or temporal periodicity of music in the text. This implies that the masoretic accentuation system could be in the process of developing from the punctuational purpose for recitation to the musical purpose for singing. To be sure, the above considerations do not preclude the possibility that musical realization existed from the outset, sup-ported by an orally transmitted tradition. However, there is no documental evidence of such tradition, and in the masoretic accentuation system we can trace only two major functional elements: punctuation for recitation and cantillation for chanting with melodic formulas (originated by neumes), not yet for singing. Therefore, it appears likely that the masoretic accentuation system functionally developed from punctuation to cantillation. That is why we are able to observe the two functional elements coexisting in this system.

4. CONCLUSION

The above discussion makes it evident that marking stressed syllables

was a secondary rather than primary function of the masoretic accentuation

53 E. Werner, Sacred Bridge, p. 105. 54 E. Werner, Sacred BridgeMusic- Early Music History 4: Studies in Medieval and Early Modern Music (ed. I. Fenlon; Cambridge: Cambridge University Press, 1984), pp. 135 208. 55 L. Treitler, Reading and Singing p. 179.

Hebrew Studies 55 (2014) 88 Park: Pointing to the Accents”

system that developed relatively late because this was not an urgent matter. Then, the punctuational function of the accents for recitation probably gave rise to other relational divisions of a unit. Finally, the accents came to be used for cantillation in addition to punctuation, although this development did not bring about a full-fledged system of musical notation.