Distributional Analysis of Cantillation Marks. Author: Scott Alexander Gabriel Reiss. Developers:

Transcription

1 Cantillizer Distributional Analysis of Cantillation Marks Author: Scott Alexander Gabriel Reiss Developers: Music Editors: Susan Owen, John Wheeler, John McMurtery What imagination the scrupulous originators manifested in these creative pictographical representations. It is paramount to translate with precision this message transmitted with so much love. 1 Suzanne Haïk-Vantoura Cover Page Cantillizer is copyright protected under the LGPL (GNU Lesser General Public License). This text is copyright protected under the FDL (GNU Free Documentation License). Cover Page 1

2 Table of Contents Cover Page... 1 Table of Contents... 2 Table of Figures Cantillation Vocalization Semiotics Versification Hermeneutics Architecture System Database Server Client Database Specifications, Requirements & Constraints Raw Data Input Data Processing Data Structure Application Graphical User Interface Menus Configuration Database Query Syntactic Analysis Syntactic Pattern Recognition Musical Analysis Freeform Analysis Graphics Appendix A Table of Contents 2

3 List of Emendations Appendix B Tools & Libraries Appendix C Abbreviations Endnotes Table of Figures Figure 1-1 Cantillation Chironomy... 7 Figure 1-2 Hierarchy of Disjunctive Signs... 9 Figure 1-3 Hierarchy of Prosodic Signs Figure 1-4 Suzanne Haïk-Vantoura Figure 1-5 Syntactic Structure of Cantillation Marks (A) Figure 1-6 Syntactic Structure of Cantillation Marks (B) Figure 2-1 System Diagram Figure 3-1 Aleppo Codex, Isaiah 9: Figure 4-1 Cantillizer Main Screen (next page) Figure 4-2 Configuration Dialog Box Figure 4-3 Search for Signs Dialog Box Figure 4-4 Phrygian Mode Scale in C Major Figure 4-5 Other Prosodic Sublinear Signs Figure 4-6 Prosodic Appoggiature Figure 4-7 Prosodic Melismata Figure 4-8 Harmonic Mode Scale in E Minor Figure 4-9 Other Psalmodic Sublinear Signs Figure 4-10 Psalmodic Appoggiature Figure 4-11 Psalmodic Melismata Figure 4-12 Blowing the Shofar on Rosh Hashanah Figure 4-13 Salomon Helperin Blowing the Yemenite Shofar (2006) Figure 4-14 Marc Chagall, The Capture of Jerusalem by Nebuchadnezzar (1956) Figure 4-15 Psalm 137 By the Rivers of Babylon in E Minor Table of Figures 3

4 Figure 4-16 Synagogue at Gaza, Mosaic of King David Strumming the Harp (6 th century CE) Figure 4-17 Cadence of Psalm 137 with Ornament Resolution Figure 4-18 Psalm 137, Verse 1, Syllabified Figure 4-19 Structure of Psalms 120, 124, 129, Figure 4-20 Tree Diagram Table of Figures 4

5 1 Cantillation The goal of the Cantillizer project is to extract and process cantillation data from the Hebrew Bible for the purpose of studying the order in which cantillation marks occur. The database holds cantillation information and provides display and statistics showing the patterns or structure of cantillation marks. 1.1 Vocalization Hebrew (like other Semitic languages) was originally and is still written without most vowels. 2 Sometime between Jerome (c ), the Dalmatian theologian and author of the Latin translation of the Bible, and Saadia ben Joseph ( ), aka Gaon, the Jewish Egyptian philosopher and author of the Arabic translation, who testify respectively to the absence and presence of vowels, three rival schools of vocalization arose, the Babylonian, the Palestinian, and the Tiberian, with the last (and latest) eventually prevailing. 3 As the Jerusalem Talmud (written in Tiberias, 4 th century CE) and the Babylonian Talmud (written in Sura, 5 th century CE) collected and organized different oral traditions of biblical commentary, the three pointing systems synthesized diverse local phonetic and musical phenomena. Moreover, beginning in the second half of the eighth century amid political turmoil in the caliphate of Baghdad, the Karaites, a schismatic Jewish sect, posed a grave threat to rabbinical authority by opposing traditional biblical commentary in a back-to-the-text movement. The besieged Tiberian rabbis fought back by creating a textual standard that they called,מסורה the Masorah or tradition. The Palestinian school under ben Naphtali (flourished c , given name either Jacob or Moses), the Jewish scribe and philologist, produced its own standard, but it has not survived, although many of its readings are known through secondary sources. Tradition attributes the vowels either to Sinaic origin or to Ezra, a priest and legal scribe in the Great Synagogue (established under his jurisdiction c. 444 BCE). None however is attested until the High Middle Ages, more than five hundred years after Hebrew had ceased to be a native language (gradually replaced by Aramaic and other vernacular tongues as the spoken languages of the Jews). The authors of the Masorah have exerted more influence on the history of biblical scholarship than all of the Talmudists and exegetes put together, for they in large part determined what following generations of readers and philologists would understand as the words of the Bible. In adding vowels to the text, eliminating polysemy by suppressing homonymy, they essentially rewrote it. Since the Septuagint (translated from lost sources c BCE) preceded the vocalized Masorah by a thousand years, its readings must be regarded as at least as authoritative. Philo (c. 20 BCE-50 CE), the Jewish Egyptian diplomat and philosopher, describes the circumstances of the Greek translation: He [Ptolemy II (c BCE), aka Philadelphus, king of Egypt], then, being a sovereign of this character, and having conceived a great admiration for and love of the legislation of Moses, conceived the idea of having our laws translated into the Greek language; and immediately he sent out ambassadors to the high-priest and king of Judea, for they were the same person. And having explained his wishes, and having requested Cantillation 5

6 him to pick him out a number of men, of perfect fitness for the task, who should translate the law, the high-priest, as was natural, being greatly pleased, and thinking that the king had only felt the inclination to undertake a work of such a character from having been influenced by the providence of God, considered, and with great care selected the most respectable of the Hebrews whom he had about him, who in addition to their knowledge of their national scriptures, had also been well instructed in Grecian literature, and cheerfully sent them. 4 Much of this tale may be deemed apocryphal, but the authors of the Septuagint spoke and wrote a language fairly close to Ancient Hebrew. The authors of the Masorah spoke medieval Aramaic, and learned to read Mishnaic (c CE) and biblical Hebrew. The vocalization of the Bible ignited a controversy that burned for more than five hundred years, until the advent of moveable type 5 allowed the advocates of the Masorah to impose its readings definitively. The tradition of printing the Bible with vowels, while almost all other Hebrew texts (including books and newspapers) lack them, is not a quaint usage benevolently conceived on behalf of Diaspora Jewish readers less skilled in the Hebrew language, but an ideological tactic to shrink the plethora of biblical variants down to one unique vision. Indeed this seemingly innocuous practice amounts to censorship. The actual Torah is manuscripted without vowels precisely because it is kept in the synagogue between the hands of the rabbis, and shown to the layman only under supervision. Modern Jewish women have fought for the right to touch the scrolls not merely as a symbolic gesture, but in a legitimate demand to read the unadulterated text of the Bible. 1.2 Semiotics לעב rotcel stnahc קריאה Cantillation refers to Jewish liturgical chant. In synagogue the from the unvocalized Torah and Hagiographa with the help of a prompter following in a cantillated text. He reads the entire Pentateuch and the five scrolls (Ruth, Song of Songs, Ecclesiastes, Lamentations, and Esther) in an annual rotation of weekly passages on the Sabbath and holidays. The הליפת לעב precentor psalmodizes from a vocalized סדור psalter. On the Sabbath and holidays he sings psalms and songs from the Bible, as well as other prayers and poems both ancient and modern. In the Middle Ages, and as late as the twentieth century in certain Jewish communities, such as those of Rome, Cairo, and Yemen, a signer cued the congregation to cantillation marks by means of chironomy or hand signals. The חזן cantor may double as lector or precentor in addition to his role as soloist and/or choirmaster, depending on the size, wealth, individual talents, and cultural traditions of the temple. Untold historical and geographical variations in synagogal custom and organization are attested. Cantillation 6

7 Figure 1-1 Cantillation Chironomy Mehupakh Revia Pashta Merekha Tifkha Little Zakef Atnakh Great Zakef Sof Pasuk Cantillation marks belong to a complex system of punctuation or textual annotation (the dots, lines, and curves written above, below, within, and between Hebrew letters) that convey a enormous amount of information with breathtaking economy as pertains to the following: 1. Melody, modulation, and rhythm (cantillation). 2. Tonic accents, intonation and pauses (cantillation). 3. Metrical units (logical). 4. Syntactic relationships (logical). 5. Vowels (diacritical). 6. Homographs and phonetic shifts (diacritical). Cantillation marks (referred to collectively as טעמים taste ) perform the first four functions, but are only attested in conjunction with pointing, the diacritical marks (referred to collectively as נקודות dots ) that perform the last two. All arose at the same time (evolving in three rival schools over a period of five hundred years, c ), and clear distinctions are seldom drawn among the characters (including digraphs and homographs), their names (including synonyms), and their intertwined roles. Since a thousand years of unrecorded harmonic and phonetic transformations separated the punctuators from the authors of the Bible, cantillation and diacritical marks do not accurately reflect historical phenomena of the biblical period, notes and vowels the rabbis had never heard. They may however reflect the Jewish culture, music, and dialects of the time and places in which they were written, medieval Sura (in Babylonia, modern-day Iraq) and Tiberias (on the shores of the Sea of Galilee). They may also accurately represent logical relationships (grammatical and metrical) actually present in the text. In his notes to the publication and translation into French of the anonymous Yemenite Hebrew grammar compilation Manuel du lecteur (1870), Joseph Derenbourg, the Jewish French Orientalist and philologist, gives this colorful account of cantillation marks: Accentuation is like the first stuttering of an unconscious grammar, and would perhaps never have undergone this development had it not been destined to compensate for science, which had not yet been formulated. This incomparable punctuation may only be understood as the expression of a tradition that had to materialize, for want of the ability to call to its aid the exact observation of the organism of language. 6 Cantillation 7

8 Whatever their original role may have been, musical, syntactic, prosodic, or phonological, cantillation marks constitute a series of data punctuating a linguistic text. The signs are non-random in order, and their sequential patterns are easily discerned, if not so easily interpreted. 1.3 Versification Two systems of cantillation marks occur in the Masorah, referred to as psalmody (Psalms, Proverbs, body of the book of Job) and prosody (prologue/epilogue of Job, and the rest of the books). Moreover Sephardic, Ashkenazi, and Yemenite cantors interpret cantillation marks differently in their trope (musical phrasing conventions). About ninety percent of the words in the Bible carry a cantillation mark, taking into account that some polysyllables carry two signs, and considering makef (similar to a hyphen) as a word separator. Hebrew is an oxytonic language, whose accent regularly falls on the last syllable of an utterance. Almost all cantillation marks occur in the stressed syllable, although half a dozen pre- or postpositive signs may, depending on the manuscript, be reiterated in the accented syllable when they precede or follow it Hermeneutics Traditional Parsing Samuel Bohl ( ), the German Orientalist and philologist, divides cantillation marks into five organizational levels 8 : רסיק ro Imperator Emperor ךלמ xer ro King הנשמ xud ro Duke שילש semoc ro Count תרשמ suvres ro Servant Prose Psalm Prose Psalm Prose Psalm Prose Psalm Prose Psalm Silluk Silluk Segolta Ole Veyored Revia Revia Geresh Little Pazer Munakh Merekha Atnakh Great Shalshelet Atnakh Zarka Revia Mugrash Double Geresh Great Shalshelet Mehupakh Tarkha Little Zakef Pashta Tsinor Little Pazer Azla Legarmeh Merekha Azla Great Zakef Yetiv Dekhi Great Pazer Mehupakh Legarmeh Double Merekha Munakh Tifkha Tevir Great Telisha Munakh Legarmeh Darga Azla Little Telisha Galgal Iluy Mehupakh Galgal Little Shalshelet Tsinorit The first four groups (violet, blue, red, and green) are disjunctive (indicating syntactic breaks) or pausal, with members listed in descending order of hierarchical or structural power, while the last (yellow) is conjunctive (indicating syntactic links) or non-pausal. The following diagram illustrates the disjunctive system as it applies to half a verse. Cantillation 8

9 Figure 1-2 Hierarchy of Disjunctive Signs In modern symbolic notation the above diagram might be represented as follows: {Emperor (King1 [Duke1 <Count1> <Count2>] [Duke2 <Count3> <Count4>]) (King2 [Duke3 <Count5> <Count6>] [Duke4 <Count7> <Count8>])} In practice, however, and contrary to modern nesting practices, cantillation marks hold cumulative mandates, i.e. the emperor supersedes king 2, duke 4, and count 8, king 1 supersedes duke 2 and count 4, dukes 1 and 3 supersede counts 2 and 6 respectively. This economy allows for eight signs instead of fifteen. The following table represents their role as implemented in linear order: Level 4 Level 3 Level 4 Level 2 Level 4 Level 3 Level 4 Level 1 Count Duke Duke Duke King King King Emperor Since the data are arranged sequentially, four members of an inferior rank are promoted to preserve the ascending hierarchy, without respect to level and contrary to modern nesting practices. Thus, level 3 king duke, and level 4 king/dukes count. Priority between two signs of equal value (with no intervening sign of greater value) goes to the first sign, except that priority goes to the last emperor, who governs the whole verse. 9 In his Cantillation of the Bible (1957), Solomon Rosowsky, the Jewish Latvian cantor and composer, elaborates on the relationships among prosodic disjunctive and conjunctive cantillation marks in the Pentateuch. Cantillation 9

10 Figure 1-3 Hierarchy of Prosodic Signs Dotted lines indicate the fealty of vassal lords to emperors and kings. Horizontal arrows indicate the proxy of regents for kings and dukes in absentia. Solid lines indicate the government of servants by lords. Conjunctive signs are seldom deemed to play a structural role in the syntax of the sentence or the meter of the verse. Attempts to reconstruct the music of the Bible tend to hold that disjunctive signs beat the rhythm, while conjunctive signs carry the melody. In his Treatise on the Accentuation of the Twenty-One So-Called Prose Books of the Old Testament (1887), William Wickes, the British Orientalist and philologist, calls the elaborated classification fanciful and misleading, and defends the rabbinical bipartition of disjunctive מלך or dominus master and conjunctive משרת or servus servant. He nevertheless lists the former in an almost identical hierarchical order (rather than, for example, alphabetically), while quibbling about a few signs that he deems over- or underrated. No one seems to ignore the subordinations. Scholars disagree in their analysis of those subordinations. Derenbourg once again waxes poetic: But the worried and restless spirit of these doctors [rabbis], endlessly bent over the sacred text, divided and subdivided the words of each verse; the slightest nuances were spotted, not only breaks were noted, but also links, and despite the rule, that a prince should not be demoted to the level of a servant, nor should the latter be promoted to the level of a ruler, [quotation from translated source text] a veritable hierarchy was established, a rather burlesque feudal system of accents, which entertained a few subtle savants of the fifteenth, sixteenth and seventeenth centuries. In this scale the lower nobility was confused with the lackeys, and accents such as telisha maintained their rank of master with difficulty. Throughout the ongoing creation of new dignitaries, the small stroke, straight or curved, placed above or below the line, tilted to the right or to the left, became the insignia of new ranks. Finally the denominations overflowed and overran, whether still more distinctions were made, or the nakdanim [punctuators] invented new names for the same accents and afterwards new uses were sought for these innovations until then unknown. 10 Cantillation 10

11 Perhaps cantillation marks once held meaning, musical, syntactic, prosodic, or phonological. However, as Derenbourg shrewdly recognizes, by the Renaissance this meaning had already broken down and been crushed under the burden of ever-increasing expectations and ever-more-minute analysis. Cantillation marks had become, and remain to this day, a semiotic system, with all the complex rules, exceptions, homonymy, synonymy, polysemy, and ambiguity common to such systems, but emptied of all semantic content. Today they constitute half a sign, signifier bereft signified Suzanne Haïk-Vantoura Heedless of the syntactic and semantic distinction between disjunctive and conjunctive signs, Suzanne Haïk-Vantoura ( ), the Jewish French organist and musicologist, in her Musique de la Bible révélée (1976), interprets cantillation marks as neumata, splitting them into groups of sublinear and superlinear signs. The former determine the notes of the diatonic scales, while the latter perform the ornamental roles known as appoggiatura and melisma. Figure 1-4 Suzanne Haïk-Vantoura Cantillizer supports Haïk-Vantoura s music theory, defined in 4.7 au-dessous Constituent Structure Analysis Generative grammar, while also repudiating Bohl s system, sees in cantillation marks evidence of constituent structure analysis. 11 Tree diagrams represent the syntactic relationships of sentences, as in the following figure illustrating Genesis 6:22. Figure 1-5 Syntactic Structure of Cantillation Marks (A) The cantillation marks are then interpreted and manipulated in much the same way as linguistic concepts such as sentence (S), noun phrase (NP), and verb phrase (VP), as follows: Atnakh governs the verse. Tifkha 1 governs the subject and verb of the first independent clause. Tifkha 2 governs the second independent clause and the subordinate clause. Tevir governs the subject and predicate of the subordinate clause. Cantillation 11

12 Great Telisha governs the object of the first independent clause (antecedent of the subordinate clause) and the verb phrase of the subordinate clause. The system seems to break down in the tevir and great telisha nodes, as the following analysis of the very similar verse Genesis 7:5 suggests. Figure 1-6 Syntactic Structure of Cantillation Marks (B) Atnakh governs the verse. Tifkha 1 governs the subject and verb of the independent clause. Tifkha 2 governs the subject and predicate of the subordinate clause. This interpretation, moreover, gives no structural role to silluk, which does not even appear in the tree diagrams, thus reducing the strongest of the cantillation marks to little more than a pilcrow sign Distributional Analysis Cantillizer retains none of these theories a priori, preferring to perform distributional analysis. Instead of appealing to such diverse fields as music, grammar, prosody, or phonetics, cantillation marks are seen as a coherent semiotic system worthy of study independent of other disciplines. Distributional analysis determines the role of an element in a system by studying its environment, i.e. its position relative to (preceding and following) the other elements of the system. Repetitive patterns tend to appear in nonrandom sets of sequential data, such as an alphanumeric code to be deciphered. This structure, once discerned, defines the relationships of the discrete elements to one another in the working of the system. For a full explanation and examples of how Cantillizer applies distributional analysis to cantillation marks, see 4 au-dessous. Cantillizer is just one small possible use for distributional analysis and structural linguistics in the study of the Bible. From the time of Saadia to that of David Kimhi (c ), aka Radak, the Jewish Provençal philologist whose Book of Completion, edited and translated in 1952 by William Chomsky, was the standard Hebrew grammar for six hundred years, Hebrew and Semitic philology led the world in the fields of phonetics and morphology. Since 1812 the standard Hebrew Grammar is the work of the German Orientalist Wilhelm Gesenius. A complete, Unicode text-and-data display-and-analysis database tool containing the whole Bible would perform distributional analysis of all characters, so-called letters, vowels, dagesh and other diacritical marks, and cantillation marks, as well as grammatical analysis of morphemes. (Syntactic and semantic analysis are also possible, but would enlarge the scope and complexity of the project considerably.) A world-wide commercial market for such a piece of software probably exists. It would be an invaluable tool to Orientalists, musicologists, university and yeshiva students and professors, seminarians, priests, cantors, and rabbis. Cantillation 12

13 Modern linguists use distributional analysis to determine the phonemes of a given language by means of minimal pairs, such as bow and vow in English. No such minimal pair is found in modern Castilian Spanish, where /b/ and /v/ are not distinct phonemes. The same method is used to determine the syntactic categories of given language. Assuming a written language in which the syntagmatic (linear language) axis runs horizontally, lexical items that are interchangeable in the paradigmatic (or vertical) axis play the same role, and therefore belong to the same part of speech. There is no other valid method of determining the phonemes or parts of speech of a given language. Cantillation 13

14 2 Architecture The system is designed as a web application. 2.1 System Figure 2-1 System Diagram Database Client End-User Server 2.2 Database Database Query 2.3 Server Client-Server Query 2.4 Client End-User Query Architecture 14

15 3 Database Cantillizer holds that the interest of any semiotic system that consists of a fixed number of discrete signs, such as natural human languages, mathematics, or musical notation, lies not in the signs themselves, which are arbitrary, unmotivated, and conventional, but in the complex relationships among the signs, which are logical and rule-based. Cantillizer discovers the logical rules of cantillation marks. Cantillizer makes the following minimalist assumptions (observations) about cantillation marks: 1. Cantillation marks are discrete signs, signifiers whose signified has been lost or forgotten. 2. Cantillation marks occur in conjunction (above, below, between) with the graphemes of a written and oral text (the Bible) in a natural human language (Hebrew). 3. Cantillation marks occur in the same linear order as the text with which they coincide. 4. This sequential order, far from being random, is the key to the interpretation of cantillation marks. The goal of Cantillizer is to perform systematic statistical and distributional analysis of the order in which cantillation marks occur in the Bible. 3.1 Specifications, Requirements & Constraints Specification Requirement Constraint Sites Operating System Programming Environment Programming Language Online web application built primarily or exclusively by means of open-source tools in a LAMP (Linux, Apache, MySQL, PHP/Python/Perl) environment.. LGPL license GPL (and compatible tools only) 100mb DB & application on Sourceforge. Works well in Windows, Linux, Mac OS. Exports to spreadsheet and/or to tab-delimited text files. Should be standard and portable. Should be standard and portable. Database Some version of SQL is proposed, but not required. Should be standard and portable. Storage Raw data must be stored and accessed independently. If better data becomes available, system must be able to accept a complete set of new raw data with minimal changes. Memory DB and processing should be small enough to work well on a laptop computer in a standard RAM configuration. Database 15

16 Specification Requirement Constraint Language English (left-to-right). Should not require Hebrew fonts or right-toleft text direction capabilities. Interface Plug-In Database is composed of alphanumeric characters and structural rules in machine code, so it is language-neutral. Interface could easily be translated into Hebrew for integration and localization purposes. Could be created and should be optimized for porting as a plug-in to an existing Hebrew Bible reader or editor. 3.2 Raw Data Input Aleppo Codex The Aleppo Codex (c. 930), written in or around Tiberias (on the shores of the Sea of Galilee) by Shlomo ben Buya a under the direction of Aaron ben Asher (flourished first half of tenth century), the Jewish scribe and philologist, was the earliest extant complete vocalized Bible until 1947, when Syrian rioters burnt down the synagogue where it had been housed and diligently copied for five hundred years, since its removal from Jerusalem via Cairo. Jews managed to rescue about sixty percent of the manuscript and smuggle it back to Israel. The source text for virtually all subsequent editions, the Aleppo Codex is the single most important document in the three-thousand-year history of the Hebrew Bible. Database 16

17 Figure 3-1 Aleppo Codex, Isaiah 9:6 In his guidelines for biblical scribes Moses ben Maimon ( ), aka Maimonides, Rambam, the Jewish Spanish physician and theologian, writes of this text: The scroll on which I relied on for (clarification of) these matters was a scroll renowned in Egypt, which includes all the 24 books (of the Bible). It was kept in Jerusalem for many years so that scrolls could be checked from it. Everyone relies upon it because it was corrected by ben Asher, who spent many years writing it precisely, and (afterward) checked it many times. 12 The spiritual leader may be prescribing more than describing, but his words carried enormous weight. 3.3 Data Processing The raw data input is a file of the entire vocalized, punctuated, and cantillated Bible, some ten million characters. Processing will reduce this to cantillation data only, some two million characters. Cantillizer obtained the best Unicode Aleppo Codex text currently published online, available from: where it may be freely and openly downloaded: without any special permission, although a small donation is requested. We have not and will not publish any of their data. Under the doctrine of fair usage we have extracted a small portion (about 20%) of their data (cantillation marks only) emended the order of some signs that are misencoded (despite being correctly rendered in graphical browser representation), processed it, analyzed it, and are publishing that emended and processed data, as well as the results of that analysis, in a format utterly incompatible with that of the original source text. 13 Database 17

18 3.3.1 Data Conversion Algorithm To convert input data: 1. Select Unicode notation from the following: Unicode decimal (sorting advantages) Unicode hexadecimal Other 2. Select display font, such as Ezra SIL SR. 3. Create character map of Unicode (all signs in source text) to display font, and verify visually. To create the character map, a list of every unique character in the text is created by means of a One of Many function. 4. Convert data to selected Unicode notation Data Processing Algorithm A good resource for checking signs, especially if we use some version of the BHS, is available at the following site: sq=1&lvl=99 To process input data: A batch file performs the following Find/Replace commands and keeps a numerical log of substitutions/deletions. 1. Convert four consecutive spaces to stich dummy sign: space character + space character + space character + space character > space character Convert verse ends to return character: <BR> > return character. Delete all remaining line breaks: <BR>. 3. Convert book number: Use the first two decimal digits of the file name (26 [= Psalms] in the example below) to extract the data. <SCRIPT TYPE="text/javascript"> ('c26f0')</SCRIPT> > </BNum><BNum ID="##"> This will create an error, which will have to be corrected semi-manually, in the first and last books of the Bible. 4. Convert tome number: Use the letter following the first two decimal digits of the file name (a [= 1] in the example below) to extract the data. <SCRIPT TYPE="text/javascript"> ('08a01.htm')</SCRIPT> > </TNum><TNum ID="##"> This will create an error, which will have to be corrected semi-manually, in the first and last tomes of each book. 5. Convert chapter number: Use the last two hexadecimal digits of the file name (f0 [= 150] in the example below) to extract the data. <SCRIPT TYPE="text/javascript"> ('c26f0')</SCRIPT> > </CNum><CNum ID="###"> This will create an error, which will have to be corrected semi-manually, in the first and last chapters of each tome/book. 6. Convert verse number anchors: <A NAME="###"> </A> > </VNum><VNum ID="###">. This will create an error, which will have to be corrected semi-manually, in the first and last verses of each chapter. Database 18

19 7. Delete all characters (letters, vowels, punctuation), ampersand (&), and pound sign (#), but not the semicolon separator (;), except cantillation marks, makef (1470), pasek (1472), sof pasuk (1475), and space character, i.e. retain the following numbers: 1425, 1426, 1427, 1428, 1429, 1430, 1431, 1432, 1433, 1434, 1435, 1436, 1437, 1438, 1439, 1440, 1441, 1443, 1444, 1445, 1446, 1447, 1448, 1449, 1450, 1451, 1452, 1453, 1454, 1469, 1470, 1472, Delete all variants and annotations: (text), [text], and {text}. 9. Convert meteg + sof pasuk to silluk: 1469;1475 > 38. Convert meteg[space character] to silluk. (There should be only a dozen or fewer, including Genesis 35:22, Exodus 20:2, Deuteronomy 5:6. If there are many more, then there is a problem.) According to another source, verses with two silluks are Gen 35:1, Exo 20:2, Deu 5:3, and Isa 44:4. 14 We will correct this manually, if necessary. Convert remaining metegs: 1469 > 37 Check for additional sof pasuks, and delete. There should be very few. 10. Convert makef (1470) to space character. 11. Delete [space character][space character]pasek (1472). This assumes that (as in my text of Leningrad) that a space character always precedes pasek. Space character prevents misidentification of digraphs. For example, munakh[space character][space character]pasek indicates munakh followed by pasek punctuation (not cantillation) after the next word, while munakh[space character]pasek indicates munakh legarmeh. Makef needs to be checked for its effect on legarmeh signs (check munakh legarmeh visually), but I doubt the first element of a legarmeh digraph can precede makef. If I am wrong about that, then makef need not be retained in step Convert the following digraphs: Great shalshelet (shalshelet + pasek): 1427;[space character];1472 > 33 Munakh legarmeh (munakh + pasek): 1443;[space character];1472 > 15 Azla legarmeh (azla + pasek): 1448;[space character];1472 > 14 Mehupakh legarmeh (mehupakh + pasek): 1444;[space character];1472 > 13 Check for remaining pasek characters and delete, if necessary. There shouldn t be any. 13. Convert reduplicated pashta: 1433;1433 > 23 Check to see if reduplicated pashta is not encoded as azla: 1448;1433 or even 1448;1448. Space character prevents misidentification of digraphs. Pashta[space character]pashta indicates consecutive rather than reduplicated signs. Also for following step. 14. Repeat the preceding step for each of the following (Results will probably be null.): Reduplicated segolta: 1426;1426 > ## (any unused number, i.e. >39). Reduplicated zarka: 1454;1454 > ## (any unused number, i.e. >39). Reduplicated great telisha: 1440;1440 > ## (any unused number, i.e. >39). Reduplicated little telisha: 1449;1449 > ## (any unused number, i.e. >39). Reduplicated dekhi: 1453;1453 > ## (any unused number, i.e. >39). 15. Delete remaining space characters. 16. Convert ole: 1451 > 36 and mugrash: 1437 > Convert remaining signs (including tifkha, single zarkha, single pashta etc., but excluding tsinor and tarkha): #### > ## 18. Check for remaining ####: to ensure that all signs have been converted. Troubleshoot. 19. Convert semicolon separator (;) to position numbers: ; > </PNum><PNum ID="##">. The ## has to be a serial number per verse, 1-nn. This will create an error, which will have to be corrected semi-manually, in the first and last positions of each verse. 20. Convert zarka (and reduplicated zarkha, if attested) to tsinor (and reduplicated tsinor) and tifkha to tarkha in psalmodic books. Database 19

20 3.3.3 Notes on Aleppo Codex There are a few differences between this text and the Leningrad codex: Book and chapter data are given in Hebrew in the headings, not verse by verse. We will need to translate and incorporate it. This might have to be done semi-manually. This operation might best be performed on the intermediary 24 book files. It will be especially difficult in the Minor Prophets book, which contains eight booklets. Verse data is given in Hebrew. We will have to input Arabic verse numbers. This might have to be done semi-manually. This operation might best be performed on the intermediary 24 book files. It will be especially difficult in the Minor Prophets book, which contains eight booklets. Word seems to interpret the space character preceding pasek as a non-breaking space character. I doubt that BabelPad recognizes the difference, so this probably doesn't matter. Word and FontPage seem to interpret line breaks as manual line breaks. I m not sure what BabelPad will do with this. We need regular line breaks at the end of verses, and nowhere else. We may have to specifically define verse ends (especially in prose) as sof pasuk, which is OK. There should be no exceptions. Step 1 or 2. The text indicates stich breaks by four consecutive spaces (breaking and non-breaking depending on the software used to read the file). In one of the earlier steps of the Data Processing Algorithm, we will convert [space character][space character][space character][space character] to an unused number (>38) that defines stich. This will have to be dealt with in the PNum property later. Step 1 or 2. Delete {} and text between. To obtain PNum property, convert data + semicolon to, for example, <PNum ID="#">data</PNum> for each verse. That might be a little hard to write, but it should work. Another option, of course, is to run the processed data through a spreadsheet converting semicolon to column division. The Data Conversion Algorithm is slightly different: 1. Open HTML file in browser. 2. Save as text. 3. Open text file in BabelPad. 4. Select All and Convert\Unicode to NCR (Decimal) Data Definition To define input data: Many more musical properties need to be defined. 1. Define books as either prosodic or psalmodic (separating Job into three booklets). 2. Define books as Pentateuch, Prophets, or Hagiographa. 3. Define PNum property per sign per verse as PNum 01-nn Define stichs by atnakh/ole veyored/silluk. 5. Define disjunctive and conjunctive signs. 6. Define prosodic and psalmodic syntactic value of signs. 7. Define sublinear and superlinear signs. 8. Define Degr property of sublinear and superlinear signs. 9. Define Rthm property of sublinear and superlinear signs. 3.4 Data Structure The raw data input is a file of the entire vocalized, punctuated, and cantillated Bible, some ten million characters. Processing will reduce this to cantillation data only, some Database 20

21 two million characters. The database retains an individual record of each of the two million instances of the signs. The following properties are associated with each record Items & Properties Cantillizer database contains the following data: Datum Property Expression Description Remarks Book BNum Conventional cannon of 39 books. Class 1-3 Pentateuch, Prophets, or Hagiographa. Type 1-2 Prosodic or psalmodic. Mode 1-2 Prosodic = Phrygian. Psalmodic = harmonic. Tonic 0-6 Defined per Mode property. Chapter CNum 001-nnn Conventional Hebrew chapter number. Book properties are defined in audessous. The book of Job is divided into three filees, the prologue and epilogue (prosodic) and the body (psalmodic). The body begins at 3:2, and the epilogue begins at 42:7. Conventional classification. Psalmodic: Psalms, body of Job, Proverbs. All other prosodic. Defined per Type property, it determines the display note value of the Degr property of the tonic, and of all other signs. (The Mode property is defined in 4.7 au-dessous.) For the moment tonic is always E? What happens when it s C/C' in prosodic, if possible? Not sure if this needs to be a separately defined property or embedded in Mode. If more than one file is used to store data, file names observe the following five-character convention: bbccc, where: bb = BNum ccc = CNum, if necessary. Verse VNum 001-nnn Conventional Hebrew verse number. Stich a-n Defined by atnakh/ole veyored/silluk. A few other signs, revia, segolta, also define the stich. Algorithm must be reverseengineered Book, chapter, verse, and stich are displayed as follows: Xyz ###:###a (lead zero suppressed). In psalmodic, revia and other signs mark the stich. This algorithm will be reverseengineered. Database 21

22 Datum Property Expression Description Remarks PNum 01-nn Position within the verse (1 st sign, 2 nd sign, 3 rd sign, etc.) ActvSubl 01-nn Rules define active sublinear sign per PNum property. The sign meteg is counted. The signs mugrash + revia and ole + merekha are counted as two signs each. Position will be adjusted in configuration selection (see au-dessous). In the display, position is not an intrinsic property of sign, as position of signs changes according to View/Hide Conjunctive Signs and Verse/Stich Display. Necessary for superlinear sign algorithms. Examples: Preceding sign Following sign Active sublinear rules are defined in au-dessous. Sign SNum 01-nn Unique identifier. Sign properties are defined in au-dessous. Syn 1-2 Disjunctive or conjunctive syntax. ProsSyn 01-nn Syntactic value in prosodic books. PslmSyn 01-nn Syntactic value in psalmodic books. Case 1-2 Sublinear or superlinear. Degr Value of diatonic scale degree. SNum is essentially a random serial number, although for the moment it reflects hierarchy as sorted by ProsSyn and PslmSyn below. Some overlap of prosodic and psalmodic is attested, e.g. pashta (Psalms 52:7, 103:1, 17), tevir (Psalms 103:46). A null value (00) could be assigned in such exceptional cases. Values of signs are expressed as a formula relative to the tonic, which is alone defined as an absolute value. The tonic needs to be defined both in the scale (0) and as a sign (silluk), here or in the Mode/Tonic property: TonicValu = 0 TonicValu x (0-6) ActvSublValu y (0-6) A few signs (such as silluk, munakh, and galgal) may need to be assigned an override (absolute) DispNote value. Database 22

23 Datum Property Expression Description Remarks Rthm 96, 48, 24, 12, 08, 06, 03 DispNote 0-11 or MIDI value per Rthm. Whole, half, quarter, eighth, triplet, sixteenth, thirtysecond notes. Integer/MIDI notation of the 12 notes of the chromatic scale per rhythm. For display options see 4.7 audessous Determined by Mode and Tonic properties of the book and Degr property of the tonic and sign in conjunction with the Rthm property of the sign. May occasionally be determined by an override value (such as silluk, munakh, and galgal). MIDI notation has the advantage of identifying the octave Cantillation Mark Properties The metrical division of cantillation marks into prosodic and psalmodic signs (based on the books of the Bible in which they occur), the syntactic distinction between disjunctive and conjunctive signs, and the hierarchy of signs within each of these groups, are a priori assumptions necessary to the creation of the database. The goal of Cantillizer is to test, confirm, refute, and refine these and other assumptions about cantillation marks. As the system is implemented and tested, a loopback of information (or reverse engineering) will take place. For example, the hierarchy indicated by the ProsSyn and PslmSyn columns in the following table is based on traditional (non-distributional) analysis of cantillation marks, methods by and large untested in statistical models enabled by computer technology. Once more rigorous data is available through use of Cantillizer, some signs may well be promoted or demoted depending on the results of distributional analysis. More radical changes, including the rejection of the metrical (prosodic/- psalmodic) and/or the syntactic (disjunctive/conjunctive) distinctions, are not to be ruled out in advance. Cantillizer will most likely provide data that necessitates the modification of the system. The examples in the following table of cantillation marks display GIF images of the Hebrew letter mem ( ) in 24 point Ezra SIL SR font. The names and even the forms of the signs vary considerably in the literature, including but not limited to differences in the Ashkenazi and Sephardic traditions. Allographs are attested within the same font family: MenuName Ezra SIL SR Ezra SIL Zarka, Tsinor Little Pazer Darga Transliteration is intended only to help readers of English recognize and pronounce the names of the cantillation marks, and does not mean to imply anything about Hebrew phonetics or orthography. The spelling kh represents the phoneme [x] as -ch in Bach. 17. Database 23

24 The abbreviations in the DispName column make some effort to account for unvocalized Hebrew spelling, so vowels are most often suppressed. Database 24

25 Syn ProsSyn PslmSyn SNum (a-z) DispName MenuName UTF8 Name UTF8 Dec (14-) UTF8 Hex (05-) Sign Alin Case Degr Rthm Issues Disj Atnk Atnakh Etnahta Posv Subl TonicValu Disj 24 NA Sgol Segolta Segol Post Supl ActvSublValu - 1 ActvSublValu ActvSublValu Haïk-Vantoura claims that segolta is always preceded by zarka. Conj NA 02 LtSl Slsl Little Shalshelet Shalshelet Shalshelet Posv Supl ActvSublValu - 2 ActvSublValu - 2 ActvSublValu In musical configuration display and process shalshelet Second and third values represent chromatic intervals (not diatonic degrees). Disj GtSl Slsl Great Shalshelet Shalshelet Shalshelet 27;72 93;C0 Posv Supl ActvSublValu - 2 ActvSublValu - 2 ActvSublValu Lower syntactic value (less than revia, tsinor, little pazer) in psalmodic books. In musical configuration display and process shalshelet. Second and third values represent chromatic intervals (not diatonic degrees). Disj 22 NA LtZk Little Zakef Zaqef Qatan Posv Supl ActvSublValu Disj 21 NA GtZk Great Zakef Zaqef Gadol Posv Supl ActvSublValu - 1 ActvSublValu Conj NA 08 Trka Tarkha Tipeha Posv Subl TonicValu Homograph of prosodic tifkha. Disj 20 NA Tfka Tifkha Tipeha Posv Subl TonicValu Homograph of psalmodic tarkha. Disj Rvia GtRv LtRv Revia 18 Revia Posv Supl ActvSublValu ActvSublValu - 1 ActvSublValu In musical configuration display and process great revia in prosodic and little revia in psalmodic. Database 25

26 Syn ProsSyn PslmSyn SNum (a-z) DispName MenuName UTF8 Name UTF8 Dec (14-) UTF8 Hex (05-) Sign Alin Case Degr Rthm Issues Conj NA 01 Tnri Tsinorit Zarqa Prep Supl ActvSublValu - 1 ActvSublValu Disj 17 NA Psta Pashta Pashta Post Supl ActvSublValu Disj 17 NA Psta RdPs Pashta Reduplicated Pashta Pashta 33;33 48;33 99;99 A8;99 Post Supl ActvSublValu + 1 ActvSublValu In syntactic configuration display and process pashta. In musical configuration display and process reduplicated pashta. Disj 16 NA Ytiv Yetiv Yetiv 34 9A Prep Subl TonicValu Disj 15 NA Tvir Tevir Tevir 35 9B Posv Subl TonicValu Disj 14 NA Grsh Geresh Geresh C Posv Supl ActvSublValu Disj NA 16 RvMg GrMk Revia Mugrash Geresh Mukdam Geresh Muqdam D Prep Supl ActvSublValu In syntactic configuration display and process revia mugrash. Ignore following revia. In musical configuration display and process geresh mukdam. Display and process following revia as distinct sign. Disj 13 NA DbGr Double Geresh Gershayim 38 9E Posv Supl ActvSublValu + 2 ActvSublValu ActvSublValu Database 26

27 Syn ProsSyn PslmSyn SNum (a-z) DispName MenuName UTF8 Name UTF8 Dec (14-) UTF8 Hex (05-) Sign Alin Case Degr Rthm Issues Disj 11 NA GtPz Great Pazer Qarney Para 39 9F Posv Supl ActvSublValu + 3 ActvSublValu + 2 ActvSublValu + 1 ActvSublValu ActvSublValu ActvSublValu + 1 ActvSublValu + 2 ActvSublValu In musical configuration fourth and fifth values are combined in one sixteenth note (06), if possible. Disj 10 NA GtTl Great Telisha Telisha Gedola 40 A0 Prep Supl ActvSublValu + 3 ActvSublValu + 2 ActvSublValu Disj LtPz Little Pazer Pazer 41 A1 Posv Supl ActvSublValu + 2 ActvSublValu Conj Mnkh Munakh Munah 43 A3 Posv Subl TonicValu May be determined by an override value (B) in DispNote property. Probably not necessary. Disj 09 NA MnLg Mnkh Munakh Legarmeh Munakh Munah 43;72 A3;C0 Posv Subl TonicValu In musical configuration legarmeh constructions display and process without pasek. May be determined by an override value (B) in DispNote property. Probably not necessary. Conj Mhpk Mehupakh Mahapakh 44 A4 Posv Subl TonicValu Disj NA 10 MhLg Mhpk Mehupakh Legarmeh Mehupakh Mahapakh 44;72 A4;C0 Posv Subl TonicValu In musical configuration legarmeh constructions display and process without pasek. Conj Mrka Merekha Merkha 45 A5 Posv Subl TonicValu Higher syntactic value (greater than mehupakh, munakh) in psalmodic books. Database 27

28 Syn ProsSyn PslmSyn SNum (a-z) DispName MenuName UTF8 Name UTF8 Dec (14-) UTF8 Hex (05-) Sign Alin Case Degr Rthm Issues Conj 05 NA DbMr Double Merekha Merkha Kefula 46 A6 Posv Subl TonicValu - 1 TonicValu Conj 04 NA Drga Darga Darga 47 A7 Posv Subl TonicValu Conj Azla Azla Qadma 48 A8 Posv Supl ActvSublValu Higher syntactic value (greater than mehupakh, munakh) in psalmodic books. Disj NA 11 AzLg Azla Azla Legarmeh Azla Qadma 48;72 A8;C0 Posv Supl ActvSublValu In musical configuration legarmeh constructions display and process without pasek. Conj 02 NA LtTl Little Telisha Telisha Qetana 49 A9 Post Supl ActvSublValu + 1 ActvSublValu + 2 ActvSublValu Conj Glgl Galgal Yerah Ben Yomo 50 AA Posv Subl TonicValu May be determined by an override value (D#) in DispNote property. Probably necessary only in prosodic mode. Disj NA 19 OlVy Ole Veyored Ole 51;45 AB;A5 Prep Supl ActvSublValu + 3 ActvSublValu Conj NA 05 Iluy Iluy Iluy 52 AC Posv Supl ActvSublValu + 4 ActvSublValu In syntactic configuration display and process ole veyored. Ignore following merekha. In musical configuration display and process ole veyored. Display and process following merekha as distinct sign. 24 Iluy rises four degrees if the active sublinear sign is tarkha (G or lower), falls three degrees if the active sublinear sign is atnakh (A or higher). Disj NA 14 Dkhi Dekhi Dehi 53 AD Prep Subl TonicValu Database 28

29 Syn ProsSyn PslmSyn SNum (a-z) DispName MenuName UTF8 Name UTF8 Dec (14-) UTF8 Hex (05-) Sign Alin Case Degr Rthm Issues Disj NA 15 Tnor Tsinor Tipeha 54 AE Post Supl ActvSublValu - 1 ActvSublValu Homograph of prosodic zarka. Disj 18 NA Zrka Zarka Tipeha AE Post Supl ActvSublValu - 1 ActvSublValu Homograph of psalmodic tsinor. Haïk-Vantoura claims that zarka is almost always followed by segolta. NA NA NA NA Gaya NA Gaya Meteg 69 BD Posv Subl TonicValu 24 In syntactic configuration meteg is ignored. In musical configuration display and process gaya, equivalent in value to silluk (tonic). Disj Sluk Silluk 22 Meteg 69 BD Posv Subl TonicValu 24 Tonic may be determined by an override value (E) in DispNote property. Probably not necessary. NA NA NA NA NA Maqaf 70 BE NA NA NA NA NA NA NA NA NA NA Paseq 72 C0 NA NA NA NA NA NA NA NA NA NA Word Break?? NA NA NA NA NA NA NA NA NA NA NBSP A0 NA NA NA NA NA NA NA NA NA NA Stich Break?? NA NA NA NA 24 Dummy sign to indicate stich breaks. Should be ignored in Posn property. In syntactic configuration display (2 horizontal bars). In musical configuration display // (2 slashes on the top line of the staff). Database 29

30 MenuName BNum DispName Class Type Remarks Active Sublinear Sign Arguments At each position in the verse, one sublinear sign is considered active. This argument is necessary for the calculation of superlinear sign arguments (see au-dessus). The ActvSubl property defines the active sublinear sign per PNum property. A sublinear sign is always active in the position it occupies. The following rules apply to positions occupied by superlinear signs: The preceding sublinear sign is considered active by default. In position 1 of the verse (if it is not occupied by a sublinear sign), the preceding (and active) sublinear sign is considered to be the last sublinear sign (almost always silluk) of the preceding verse. The following sublinear sign (usually gaya) is considered active. (This rule will not work, as Aleppo does not always have gaya as the first sign in the verse. Following a sublinear sign corresponding to more than 1 note values (such as double merekha), the last note value is considered active Books of the Bible Properties Genesis 01 Gen Pent Pros Exodus 02 Exo Pent Pros Leviticus 03 Lev Pent Pros Numbers 04 Num Pent Pros Deuteronomy 05 Deu Pent Pros Joshua 06 Jos Prph Pros Judges 07 Jud Prph Pros 1 Samuel 08 1Sa Prph Pros 2 Samuel 09 2Sa Prph Pros 1 Kings 10 1Ki Prph Pros 2 Kings 11 2Ki Prph Pros Isaiah 12 Isa Prph Pros Jeremiah 13 Jer Prph Pros Ezekiel 14 Eze Prph Pros Hosea 15 Hos Prph Pros Joel 16 Joe Prph Pros Amos 17 Amo Prph Pros Obadiah 18 Oba Prph Pros Jonah 19 Jon Prph Pros Micah 20 Mic Prph Pros Nahum 21 Nah Prph Pros Database 30

31 MenuName BNum DispName Class Type Remarks Habakkuk 22 Hab Prph Pros Zephaniah 23 Zep Prph Pros Haggai 24 Hag Prph Pros Zechariah 25 Zec Prph Pros Malachi 26 Mal Prph Pros 1 Chronicles 27 1Ch Hgph Pros 2 Chronicles 28 2Ch Hgph Pros Psalms 29 Psa Hgph Pslm Job (prologue) 30 Job Hgph Pros Chapters 1-3:1 are accessible from the Books of the Bible drop-down menu only through the All 8 Prosodic Hagiographa item. Job (body) 31 Job Hgph Pslm Chapters 3:2-42:6 Job (epilogue) 32 Job Hgph Pros Chapter 42:7-17 is accessible from the Books of the Bible drop-down menu only through the All 8 Prosodic Hagiographa item. Proverbs 33 Pro Hgph Pslm Ruth 34 Rut Hgph Pros Song of Songs 35 SoS Hgph Pros Ecclesiastes 36 Ecc Hgph Pros Lamentations 37 Lam Hgph Pros Esther 38 Est Hgph Pros Daniel 39 Dan Hgph Pros Ezra 40 Ezr Hgph Pros Nehemiah 41 Neh Hgph Pros Known Problems The Decalogue (Exodus 5:2-17 and Deuteronomy 5:6-21) attests two systems of cantillation marks, often referred to as upper (or Babylonian) trope and lower (or Palestinian) trope. Either/both are retained. (I haven t looked at this at all.) The following signs may be reduplicated: segolta, zarka (Isaiah 2:15, 26:5), great telisha (2 Kings 17:24), little telisha (Genesis 48:7, Deuteronomy 21:13, Jeremiah 11:10, Esther 6:13), tsinor, and dekhi. If so, proceed as with pashta in au-dessus. Prepositive signs (such as geresh and dekhi) tend to get reversed with the following sign in the conversion to Unicode decimal notation. The Internal Silluk There should be only a dozen or fewer occurrences of the internal silluk, variously listed as: Genesis 35:1 (Aleppo) Database 31

32 Genesis 35:22 (BHS) Exodus 20:2 ff. (Aleppo & BHS) Deuteronomy 5:3 (Aleppo) Deuteronomy 5:6 ff. (BHS) Isaiah 44:4 (Aleppo) Most or all of these passages seem to be cases of double punctuation. David Robinson and Elisabeth Levy comment: Meteg is used to mark the secondary tone, reminding the reader to give its vowel full pronunciation. In BHS meteg is also used sometimes to help the reader distinguish a short o from a long a. An identical mark is used for the punctuation character Silluq. In fact several of the accents do double duty in this way, but there is always a clear algorithm to distinguish their use [ ] Silluq is the strongest disjunctive accent, the equivalent of a modern full stop. It is written as a vertical bar under the tone syllable of the last word in a sentence. In appearance it is exactly the same as meteg. In the vast majority of cases, silluq is written under the word immediately before sof passuq (:) so it is usually redundant as a punctuation mark. But the Masoretes made good use of it in a few cases where they disagreed with the sentence divisions they had inherited from earlier rabbis. In Gen 35.22, for example, the end of the verse is doubly accented. The earlier rabbis had not placed a sof passuq between and Israel heard it and the sons of Jacob were twelve, although the structure of the narrative clearly requires one--it seems likely that this was a rather delicate means of passing over an unpleasant subject by minimising its emphasis. The Masoretes were not free to insert a sof passuq, and they obediently pointed the text in the form they had received it, but also inserted silluq at the end of and Israel heard it to indicate that there should have been a verse division at that point. Similar emendations of the traditional verse structure are to be found in Ex 20.2ff and Deut 5.6ff. With these exceptions silluq is always the last accent on a word. Any mark which appears before it is to be ignored for the purposes of punctuation. This beautiful commentary seems cogent, but unfortunately the authors do not specify their clear algorithm. One such rule might obtain, according to Helmut Richter: Meteg vs. Silluq. Silluq (as part of Sof Pasuq) occurs only as last mark before the colon-shaped Sof Pasuq whereas Meteg is never the last cantillation mark in a word. Database 32

33 4 Application The best approach is to create a system of Configurations that in turn determine the display, find, search, sort, analysis, pattern recognition, and graphics parameters. Users select a configuration (or use the default syntactic configuration) and then make queries without having to use too many different radio and toolbar buttons. The configuration, which may be changed at any time, determines which menus and buttons appear in the toolbars and dialogs or which objects and items are enabled. (For details on configuration see 4.3 au-dessous). In all three configurations Cantillizer integrates the metrical distinction between prosodic and psalmodic books of the Bible (see au-dessous). Most users (traditional Bible students) require the default configuration, observing the syntactic (disjunctive/- conjunctive) distinction, while other users, musical (sublinear/superlinear or case distinction) and freeform (no distinction between signs beyond prosodic/psalmodic), will seldom change configuration, once they have made their selection. Development schedule: 1. Create database and interface to accommodate all three configurations. 2. Create capabilities for syntactic configuration, the default configuration. 3. Test thoroughly and rigorously. Great shalshelet, merekha, and azla should be slated for extensive testing to confirm hierarchy. 4. Adjust database according to data returned in syntactic configuration, if necessary. 5. Add capabilities to enable musical configuration. 6. Test thoroughly and rigorously. 7. Adjust database according to data returned in musical configuration, if necessary. Given the complexity of musical configuration, it may be prudent to postpone implementation of freeform configuration indefinitely. 8. Add capabilities to enable freeform configuration. 9. Test thoroughly and rigorously. 10. Adjust database according to data returned in freeform configuration, if necessary. 4.1 Graphical User Interface Main Screen Use the Cantillizer menus and toolbars to define data sources, select display settings, query database, sort results, perform analyses (syntactic, musical, or freeform), recognize patterns, and create graphics. Figure 4-1 Cantillizer Main Screen (next page) Application 33

34 Configuration Find Search Sort by Verse/ Hierarchy Verse/Stich Display Show/Hide Conjunctive Signs Show/Hide Superlinear Signs Analyze Display Statistics as Integer/ Percentage Pattern Recognition Pattern/ Verse Display Charts & Graphs Export to Excel Cantillizer File Edit View Format Tools Window Help Books of the Bible All Prosodic Books All Psalmodic Books All Pentateuch All Prophets All Prosodic Hagiographa Genesis (Pentateuch) Exodus Leviticus Numbers Deuteronomy Joshua (Prophets) Judges 1 Samuel 2 Samuel 1 Kings 2 Kings Isaiah Jeremiah Ezekiel Hosea (Minor Prophets) Joel Amos Obadiah Jonah Micah Nahum Habakkuk Zephaniah Haggai Zechariah Malachi 1 Chronicles (Hagiographa) 2 Chronicles Psalms Job (psalmodic) Proverbs Ruth Song of Songs Ecclesiastes Lamentations Esther Daniel Ezra Nehemiah Chapter Disjunctive Signs Conjunctive Signs Conjunctive Sign Position Sublinear Signs Superlinear Signs Superlinear Sign Position Atnakh Azla Legarmeh Dekhi Double Geresh Geresh Great Pazer Great Shalshelet Great Telisha Great Zakef Little Pazer Little Zakef Mehupakh Legarmeh Munakh Legarmeh Ole Veyored Pashta Revia Revia Mugrash Segolta Silluk Tevir Tifkha Tsinor Yetiv Zarka Azla Darga Double Merekha Galgal Illuy Little Shalshelet Little Telisha Mehupakh Merekha Munakh Tarkha Tsinorit Ultimate Penultimate Antepenultimate Preantepenultimate Preterpreantepenultimate Atnakh Darga Dekhi Double Merekha Galgal Gaya Mehupakh Merekha Munakh Silluk Tarkha Tevir Tifkha Yetiv Azla Double Geresh Geresh Geresh Mukdam Great Pazer Great Revia Great Telisha Great Zakef Illuy Little Pazer Little Revia Little Telisha Little Zakef Ole Veyored Pashta Reduplicated Pashta Segolta Shalshelet Tsinor Tsinorit Zarka 1 st postpositive 2 nd postpositive 3 rd postpositive 4 th postpositive 5 th postpositive No verses displayed. No disjunctive sign displayed. No conjunctive sign displayed. No conjunctive sign position displayed. No patterns displayed. Application 34

35 The above image was created in such a way as to show the full menu contents. The actual GUI should contain three toolbars, the Cantillizer toolbar, the Syntactic Analysis toolbar, and the Musical Analysis toolbar. Configuration selection determines which of the last two toolbars is enabled. The two analysis toolbars may be two halves of the same physical toolbar, as space allows Cantillizer Toolbar Books of the Bible drop-down menu Chapter drop-down menu Configuration button (opens Configuration dialog box) Find Verses button Search for Signs button (opens Search for Verses dialog box) Sort Data by Verse/Hierarchy toggle button Previous button Next button Analyze button Display Statistics as Integer/Percentage toggle button Pattern Recognition button Pattern/Verse Display toggle button Charts & Graphs button (possibly disabled in this version) Print button Export to Tab-Delineated Text File button (necessary for export to music editor) Syntactic Analysis Toolbar Disjunctive Signs drop-down menu Conjunctive Signs drop-down menu Conjunctive Sign Position drop-down menu Verse/Stich Display toggle button (not relevant to and confusing in musical configuration) Show/Hide Conjunctive Signs toggle button Musical Analysis Toolbar Sublinear Signs drop-down menu Superlinear Signs drop-down menu Superlinear Sign Position drop-down menu Show/Hide Superlinear Signs toggle button Display Letter Notes/Fixed-Do/Signs radio buttons 4.2 Menus A fairly simple, console-style GUI is probably sufficient, making use (primarily or exclusively) of toolbar buttons. The menu bar menus may be informed as necessary. Conventions for commands, tools, utilities, and their shortcuts/hotkeys should be observed to the extent possible. Application 35

36 4.2.1 File Edit View Format Tools Window Help Documentation will probably be the final version of this document in PDF format. 4.3 Configuration Cantillizer fully supports two different interpretations of cantillation marks, the traditional rabbinical syntactic and semantic theory based on the historical distinction between disjunctive and conjunctive signs, and the music theory of Suszanne Haïk- Vantoura based on the empirical case distinction between sublinear and superlinear signs. In addition Cantillizer supports freeform analysis, which makes no such distinction between signs. Each of these configurations observes the metrical distinction, treating signs in the prosodic and psalmodic books differently. Configuration may be changed at any time. In the toolbar, press the Configuration button to open the Configuration dialog box and set your preference for analysis. Figure 4-2 Configuration Dialog Box Configuration Perform: Syntactic analysis (disjunctive/conjunctive) Musical analysis (sublinear/superlinear) Freeform analysis OK Syntactic Configuration (default) The Disjunctive Signs and Conjunctive Signs drop-down menus are enabled upon selection of books. Only signs occurring in selected books are enabled. The Show/Hide Conjunctive Signs button is enabled. The Conjunctive Sign Position drop-down menu is enabled upon selection of a conjunctive sign. Only those positions in which the selected sign occurs are enabled. Application 36

37 The Analyze and Pattern Recognition buttons take into account the syntactic (disjunctive/- conjunctive) distinction. The following information appears in the status bar: No verses displayed (toggle stichs), No disjunctive sign displayed, No conjunctive sign displayed, No conjunctive sign position displayed, No patterns displayed. The Position property is adjusted to exclude meteg and to count geresh mukdam + following revia and ole veyored + following merekha as one sign each Musical configuration The Sublinear Signs and Superlinear Signs drop-down menus are enabled upon selection of books. Only signs occurring in selected books are enabled. The Show/Hide Superlinear Signs button is enabled. The Superlinear Sign Position drop-down menu is enabled upon selection of a superlinear sign. Only those positions in which the selected sign occurs are enabled. The Analyze and Pattern Recognition buttons take into account the musical (sublinear/- superlinear) or case distinction. The following information appears in the status bar: No verses displayed (toggle stichs), No sublinear sign displayed, No superlinear sign displayed, No superlinear sign position displayed, No patterns displayed Freeform Configuration Syntactic configuration parameters (sign names and values) may also apply to freeform configuration. The All Signs drop-down menu is enabled upon selection of books. Only signs occurring in selected books are enabled. The Analyze and Pattern Recognition buttons take into account all prosodic or psalmodic signs without distinction. The following information appears in the status bar: No verses displayed (toggle stichs), No sign displayed, No patterns displayed. The Position property is adjusted to exclude meteg and to count geresh mukdam + revia and ole + merekha as one sign each. 4.4 Database Query au-dessous assume that the syntactic (disjunctive/conjunctive) distinction obtains. For more information on musical and freeform analysis, see 4.7 and 4.8 au-dessous Find Verses To search database & display data: 1. In the Books of the Bible drop-down menu, click to select a biblical book or group of books in which to search and display data. Press Ctrl to multi-select, Shift to select range. You may select either prosodic or psalmodic books only. 2. Press the Show/Hide Conjunctive Signs toggle button to show conjunctive signs. 3. Press the Verse/Stich Display toggle button to view stich display. Verses are divided into stichs (a, b, c, d) by ole veyored or atnakh. 4. Press the Search button to display data. The number of verses and stichs displayed appears in the status bar. Application 37

38 Assuming the options Show Conjunctive Signs and Stich Display, Cantillizer returns data as follows: X Stichs Position 1 Position 2 Position 3 Position 4 Position 5 Position 6 SoS 6:1a 27 Yetiv 17 Munakh 33 Little Zakef 31 Tifkha 36 Atnakh SoS 6:1b 27 Yetiv 17 Munakh 33 Little Zakef 31 Tifkha 37 Silluk SoS 6:2a 28 Pashta 17 Munakh 33 Little Zakef 31 Tifkha 36 Atnakh SoS 6:2b 28 Pashta 33 Little Zakef 31 Tifkha 37 Silluk SoS 6:3a 16 Mehupakh 28 Pashta 17 Munakh 33 Little Zakef 31 Tifkha 37 Silluk SoS 6:4a 12 Azla 16 Mehupakh Pashta 33 Little Zakef 31 Tifkha 36 Atnakh SoS 6:4b 31 Tifkha 37 Silluk SoS 6:5a 16 Mehupakh 28 Pashta 33 Little Zakef 15 Merekha 31 Tifkha 36 Atnakh SoS 6:5b 28 Pashta 17 Munakh 33 Little Zakef 31 Tifkha 37 Silluk SoS 6:6a 28 Pashta 17 Munakh 33 Little Zakef 31 Tifkha 36 Atnakh SoS 6:6b 28 Pashta 33 Little Zakef 31 Tifkha 37 Silluk By default data is sorted in ascending order by book (as in the Books of the Bible menu), chapter, verse, and stich. [Color scheme is wrong. Workspace and data should appear in some standard DB format. A color distinction. between conjunctive (superlinear) (gray?) and disjunctive (sublinear) signs only is necessary. ProsSyn and PslmSyn (defined in au-dessus) are necessary in display so that user understands sorting. ProsSyn, PslmSyn, and DispName should be non-proportional font so that columns are fixed width] Search for Signs TBD Application 38

39 Figure 4-3 Search for Signs Dialog Box Search for Signs Books of the Bible All 21 Prosodic Books All 3 Psalmodic Books All 5 Pentateuch (prosodic) All 8 Prophets (prosodic) All 8 Prosodic Hagiographa Genesis (Pentateuch) Exodus Leviticus Numbers Deuteronomy Joshua (Prophets) Judges 1 Samuel 2 Samuel 1 Kings 2 Kings Isaiah Jeremiah Ezekiel Hosea (Minor Prophets) Joel Amos Obadiah Jonah Micah Nahum Habakkuk Zephaniah Haggai Zechariah Malachi 1 Chronicles (Hagiographa) 2 Chronicles Psalms (psalmodic) Job (psalmodic) Proverbs (psalmodic) Ruth Song of Songs Ecclesiastes Lamentations Esther Daniel Ezra Nehemiah Exact sequence And Or Without And Or And Without And Find Verses Find Patterns Disjunctive Signs Disjunctive Signs Disjunctive Signs Disjunctive Signs Disjunctive Signs Disjunctive Signs Atnakh Azla Legarmeh Dekhi Double Geresh Geresh Great Pazer Great Shalshelet Great Telisha Great Zakef Little Pazer Little Zakef Mehupakh Legarmeh Munakh Legarmeh Ole Veyored Pashta Revia Revia Mugrash Segolta Silluk Tevir Tifkha Tsinor Yetiv Zarka In the Books of the Bible drop-down menu press Ctrl to multi-select, Shift to select range. You may select either prosodic or psalmodic books only. The Disjunctive Signs menu is enabled upon selection of books. Selection of a sign enables the Exact sequence drop-down menu and opens/enables another Disjunctive Signs drop-down menu. Selection of another sign opens/enables the corresponding And drop-down menu. User should be able to select at least five or six signs. In syntactic configuration (default) the Disjunctive Signs drop-down menu appears. Only signs occurring in selected books are enabled. In musical configuration the Sublinear Signs drop-down menu appears. Only signs occurring in selected books are enabled. In freeform configuration the All Signs drop-down menu appears. Only signs occurring in selected books are enabled. Application 39

40 The Find Verses and Find Patterns buttons are enabled upon selection of signs. Press either the Find Verses or the Find Patterns button. For best results select at least 2 or 3 signs. Select either the Exact sequence or the And item. To find multiple occurrences of the same sign per verse, select the same sign more than once. Select either the Exact sequence (for consecutive occurrences) or the And (for all multiple occurrences) item. The Exact sequence drop-down menu occurs only once. If you select this (default) item, no And drop-down menus will appear below. If you unselect it, they will appear Sort Data Use the Cantillizer toolbar to define data sorting options. In the text that follows, the term hierarchy refers to ProsSyn and PslmSyn defined in au-dessus. To sort data by hierarchy of signs: 1. Perform data search, as in au-dessus. 2. In the toolbar, press the Sort by Verse/Hierarchy toggle button to sort data in descending order by hierarchy of signs per column from right to left. (Tiebreaker is ascending order by book, chapter, verse, and stich.) Cantillizer sorts data as follows: X Stichs Position 1 Position 2 Position 3 Position 4 Position 5 Position 6 SoS 6:3a 16 Mehupakh 28 Pashta 17 Munakh 33 Little Zakef 31 Tifkha 37 Silluk SoS 6:4a 12 Azla 16 Mehupakh 28 Pashta 33 Little Zakef 31 Tifkha 36 Atnakh SoS 6:5a 16 Mehupakh 28 Pashta 33 Little Zakef 15 Merekha 31 Tifkha 36 Atnakh SoS 6:5b 28 Pashta 17 Munakh 33 Little Zakef 31 Tifkha 37 Silluk SoS 6:1b 27 Yetiv 17 Munakh 33 Little Zakef 31 Tifkha 37 Silluk SoS 6:2a 28 Pashta 17 Munakh 33 Little Zakef 31 Tifkha 36 Atnakh SoS 6:6a 28 Pashta 17 Munakh 33 Little Zakef 31 Tifkha 36 Atnakh SoS 6:1a 27 Yetiv 17 Munakh 33 Little Zakef 31 Tifkha 36 Atnakh SoS 6:2b 28 Pashta 33 Little Zakef 31 Tifkha 37 Silluk SoS 6:6b 28 Pashta 33 Little Zakef 31 Tifkha 37 Silluk SoS 6:4b 31 Tifkha 37 Silluk 3. Click any column header to toggle between ascending and descending (default) order by hierarchy of signs in that column. 4. Repeat as desired for additional columns. Sorting is nested, i.e. previous sorting is retained. 5. Press the Sort by Verse/Hierarchy toggle button to clear sorting and revert to ascending order by book, chapter, verse, and stich. Application 40

41 4.4.4 Compare Signs add feature, compare distribution of 2 signs. only syntactic config only disjunctive signs (?) only following sign of greater value (recursive!?!) 1,7 x 500 = 50% 1,8 x 250 = 25% 1,9 x 250 = 25% 2,8 x 1 = 10% (10% same) 2,9 x 9 = 90% (25% same) = 35% similar distribution of 1 and 2 This algorithm must eventually be used to reverse engineer sign hierarchy, moving backwards from verse break to stich break (caesura)... Also move forward from verse beginning, 1st sign is value 1, second sign is value 2, third sign is value 3, until recurrence of 1st sign, then recommence... Final sign hierarchy algorithm: value = average value of following sign of higher value -1. Typical example, from the Song of Songs: pashta, little zakef, tifkha, atnakh, pashta, little zakef, tifkha, silluk Values: Silluk = 5 Atnakh = 4 Tifkha = 3.5 ([5+4]/2-1 Little zakef = 2.5 Pashta = Syntactic Analysis Statistics are provided per book and per user-defined group of books. Use the Cantillizer toolbar to define data processing and statistics options Disjunctive Statistics To obtain disjunctive statistics: 1. Perform data search, as in au-dessus, selecting the Hide Conjunctive Signs option. 2. In the toolbar, press the Analyze button to display statistics. 3. Press the Integer/Percentage Display toggle button to view percentage display. Occurrences of sign x following or preceding sign y are displayed as a percentage of total occurrences of sign x.. Application 41

42 Statistical analysis provides the following data for the selected book or books: Number of occurrences of each disjunctive sign following and preceding each disjunctive sign. The following table illustrates such a display of statistics for five common signs: Sign 28 Pashta 31 Tifkha 33 Little Zakef 36 Atnakh 37 Silluk Occurrences # # # # # Environment Follows Precedes Follows Precedes Follows Precedes Follows Precedes Follows Precedes 28 Pashta # # # # # # # # # # 31 Tifkha # # # # # # # # # # 33 Little Zakef # # # # # # # # # # 36 Atnakh # # # # # # # # # # 37 Silluk # # # # # # # # # # Verse break # # # # # # # # # # Default sorting is in ascending order of hierarchy. Data may be sorted in ascending or descending order by the frequency of any (or all cumulatively) of the Follows/Precedes columns by clicking the desired column header Conjunctive Statistics by Sign To obtain conjunctive statistics by sign: 1. Perform data search, as in au-dessus, selecting the Show Conjunctive Signs option. 2. In the Conjunctive Sign menu, click to select the sign whose statistics you wish to obtain. 3. Press the Analyze button to display statistics. The name of the selected conjunctive sign appears in the status bar. Statistical analysis provides the following data for the selected book or books: Number of occurrences of the selected conjunctive sign in each position preceding each disjunctive sign. The following table illustrates such a display of statistics for munakh: Preterpreantepenultimate Preantepenultimate Antepenultimate Penultimate Ultimate Disjunctive # # # # # 37 Silluk # # # # # 36 Atnakh # # # # # 35 Segolta # # # # # 34 Great Shalshelet # # # # # 33 Little Zakef # # # # # 32 Great Zakef # # # # # 31 Tifkha # # # # # 30 Revia # # # # # 29 Zarka # # # # # 28 Pashta Application 42

43 Preterpreantepenultimate Preantepenultimate Antepenultimate Penultimate Ultimate Disjunctive # # # # # 27 Yetiv # # # # # 26 Tevir # # # # # 25 Geresh # # # # # 24 Double Geresh # # # # # 23 Little Pazer # # # # # 22 Great Pazer # # # # # 21 Great Telisha # # # # # 20 Munakh Legarmeh Default sorting is descending order by column from right to left. Data may be sorted in ascending or descending order by the frequency of any (or all by nested sorting) of the columns by clicking the desired column header Conjunctive Statistics per Disjunctive Sign To obtain conjunctive statistics per disjunctive sign: 1. Perform data search, as in au-dessus, selecting the Show Conjunctive Signs option. 2. In the Disjunctive Sign menu, click to select the sign whose statistics you wish to obtain. 3. Press the Analyze button to display statistics. The name of the selected disjunctive sign appears in the status bar. Statistical analysis provides the following data for the selected book or books: Number of occurrences of each conjunctive sign in each position preceding the selected disjunctive sign. The following table illustrates such a display of statistics for atnakh: Conjunctive Preterpreantepenultimate Preantepenultimate Antepenultimate Penultimate Ultimate Disjunctive 17 Munakh # # # # # 36 Atnakh 16 Mehupakh # # # # # 36 Atnakh 15 Merekha # # # # # 36 Atnakh 14 Double Merekha # # # # # 36 Atnakh 13 Darga # # # # # 36 Atnakh 12 Azla # # # # # 36 Atnakh 11 Little Telisha # # # # # 36 Atnakh 10 Galgal # # # # # 36 Atnakh Default sorting is descending order by column from right to left. Data may be sorted in ascending or descending order by Conjunctive or by the frequency of any (or all by nested sorting) of the columns by clicking the desired column header. Application 43

44 4.5.4 Conjunctive Statistics by Sign per Position & per Disjunctive Sign To obtain conjunctive statistics by sign per position & per disjunctive sign: 1. Perform data search, as in au-dessus, selecting the Show Conjunctive Signs option. 2. In the Disjunctive Sign menu, click to select the sign whose statistics you wish to obtain. 3. In the Conjunctive Sign menu, click to select the sign whose statistics you wish to obtain. 4. In the Conjunctive Sign Position menu, click to select the desired position: Ultimate (last before disjunctive sign) Penultimate (last but one before disjunctive sign) Antepenultimate (last but two before disjunctive sign) Preantepenultimate (last but three before disjunctive sign) Preterpreantepenultimate (last but four before disjunctive sign) Only those positions in which the selected sign occurs in the selected books are enabled. I don t think the occurrence of six consecutive conjunctive signs is attested. 5. Press the Analyze button to display statistics. The name of the selected disjunctive sign, and the name and position of the selected conjunctive sign appear in the status bar. Statistical analysis provides the following data for the selected book or books: Number of occurrences of each conjunctive sign in each position surrounding the selected conjunctive sign in the selected position and preceding the selected disjunctive sign. The following table illustrates such a display of statistics for munakh in the ultimate position preceding atnakh: Conjunctive Preterpreantepenultimate Preantepenultimate Antepenultimate Penultimate Ultimate Disjunctive 17 Munakh # # # # 17 Munakh 36 Atnakh 16 Mehupakh # # # # 17 Munakh 36 Atnakh 15 Merekha # # # # 17 Munakh 36 Atnakh 14 Double Merekha # # # # 17 Munakh 36 Atnakh 13 Darga # # # # 17 Munakh 12 Azla # # # # 17 Munakh 36 Atnakh 36 Atnakh 11 Little Telisha # # # # 17 Munakh 36 Atnakh 10 Galgal # # # # 17 Munakh 36 Atnakh Default sorting is descending order by column from right to left, excluding the disjunctive sign and the selected conjunctive position. Data may be sorted in ascending or descending order by the frequency of any (or all by nested sorting) of the columns by clicking the desired column header. Application 44

45 4.6 Syntactic Pattern Recognition Pattern recognition is provided per book and per user-defined group of books. Use the Cantillizer toolbar to define pattern recognition settings Disjunctive Patterns To perform disjunctive pattern recognition: 1. Perform data analysis to obtain disjunctive statistics, as in au-dessus. 2. For best results, in the toolbar, select the Stich Display option. 3. Press the Pattern Recognition button to display patterns. The number of patterns displayed appears in the status bar. 4. Click to select a row, and press the Pattern/Verse Display toggle button to view the verses corresponding to the selected pattern. Pattern recognition provides the following data for the selected book or books: 25 most frequently attested patterns of disjunctive signs. To view more patterns, press the Next button. The following table illustrates such a display of patterns: Occurrences Position 1 Position 2 Position 3 Position 4 # 28 Pashta 33 Little Zakef 31 Tifkha 37 Silluk # 28 Pashta 33 Little Zakef 31 Tifkha 36 Atnakh Default sorting is descending order by occurrences of pattern. Data may be sorted in ascending or descending order by the frequency of any (or all by nested sorting) of the position columns by clicking the desired column header Disjunctive Patterns by Sign TBD. This command filters results so that only patterns including pashta, little zakef, and/or tifkha, for example, are displayed. This is performed using the Search for Signs dialog box in Conjunctive Patterns by Sign per Position To perform conjunctive pattern recognition by sign per position: 1. Perform data analysis to obtain conjunctive statistics by sign, as in au-dessus. 2. In the Conjunctive Sign Position menu, click to select the desired position: Ultimate (last before disjunctive sign) Penultimate (last but one before disjunctive sign) Antepenultimate (last but two before disjunctive sign) Preantepenultimate (last but three before disjunctive sign) Preterpreantepenultimate (last but four before disjunctive sign) Only those positions in which the selected sign occurs in the selected books are enabled. 3. In the toolbar, press the Pattern Recognition button to display patterns. The name of the selected conjunctive sign and the number of patterns displayed appear in the status bar. Application 45

46 4. Click to select a row, and press the Pattern/Verse Display toggle button to view the verses corresponding to the selected pattern. Pattern recognition provides the following data for the selected book or books: 25 most frequently attested patterns of surrounding conjunctive signs and the disjunctive sign following the selected conjunctive sign in the selected position. To view more patterns, press the Next button. The following table illustrates such a display of patterns for munakh in the ultimate position, i.e. immediately preceding a disjunctive sign: Occurrences Preterpreantepenultimate Preantepenultimate Antepenultimate Penultimate Ultimate Disjunctive # 17 Munakh # 17 Munakh 36 Atnakh 33 Little Zakef Default sorting is descending order by occurrences of pattern. Data may be sorted in ascending or descending order by the frequency of any (or all by nested sorting) of the position columns by clicking the desired column header Conjunctive Patterns per Disjunctive Sign To perform conjunctive pattern recognition per disjunctive sign: 1. Perform data analysis to obtain conjunctive statistics per disjunctive sign, as in au-dessus. 2. In the toolbar, press the Pattern Recognition button to display patterns. The name of the selected disjunctive sign and the number of patterns displayed appear in the status bar. 3. Click to select a row, and press the Pattern/Verse Display toggle button to view the verses corresponding to the selected pattern. Pattern recognition provides the following data for the selected book or books: 25 most frequently attested patterns of conjunctive signs preceding the selected disjunctive sign. To view more patterns, press the Next button. The following table illustrates such a display of patterns for atnakh: Occurrences Preterpreantepenultimate Preantepenultimate Antepenultimate Penultimate Ultimate Disjunctive # 17 Munakh # 15 Merekha 36 Atnakh 36 Atnakh Default sorting is descending order by occurrences of pattern. Data may be sorted in ascending or descending order by the frequency of any (or all by nested sorting) of the position columns by clicking the desired column header. Application 46

47 4.6.5 Conjunctive Patterns by Sign per Position & per Disjunctive Sign To perform conjunctive pattern recognition by sign per position & per disjunctive sign: 1. Perform data analysis to obtain conjunctive statistics by sign per position and per disjunctive sign, as in au-dessus. 2. In the toolbar, press the Pattern Recognition button to display patterns. The name of the selected disjunctive sign and the number of patterns displayed appear in the status bar. 3. Click to select a row, and press the Pattern/Verse Display toggle button to view the verses corresponding to the selected pattern. Pattern recognition provides the following data for the selected book or books: 25 most frequently attested patterns of conjunctive signs surrounding the selected conjunctive sign in the selected position and preceding the selected disjunctive sign. To view more patterns, press the Next button. The following table illustrates such a display of statistics for munakh in the ultimate position preceding atnakh: Occurrences Preterpreantepenultimate Preantepenultimate Antepenultimate Penultimate Ultimate Disjunctive # # # # # 17 Munakh # # # # # 17 Munakh 36 Atnakh 36 Atnakh 4.7 Musical Analysis The Sublinear Signs, Superlinear Signs, and Superlinear Sign Position dropdown menus allow the user to perform analysis according to the biblical music theory of Suzanne Haïk-Vantoura. To perform musical analysis, in toolbar, press the Configuration button. In the Configuration dialog box select the Musical analysis radio button. Find, search, sort, statistical analysis, and pattern recognition all proceed as in au-dessus with sublinear signs replacing disjunctive signs and superlinear signs replacing conjunctive signs. Keep in mind that the conjunctive sign depends on the following disjunctive sign, while the superlinear sign depends on the preceding (active) sublinear sign. Musical configuration sorts sublinear signs in descending order as follows: tonic, supertonic, mediant, subdominant, dominant, superdominant, subtonic. Superlinear sorting? Musical configuration output may be displayed in the following three notation systems, by selecting a radio button from the music toolbar: Letter notation (for export to music editors) Notation of accidentals? Fixed-do (for European users) Notation of accidentals? DispName property (sign names) Display of Rthm property? (Note problem with Great Pazer.) musical config use display feature (color?) to distinguish notes derived from sub/superlinear signs and to distinguish consecutive ornaments, if applicable Application 47

48 4.7.1 Prosodic Mode & Scale Cantillizer calculates note values by means of algorithms operating on the degrees of the diatonic scales (corresponding to the sublinear signs) in the prosodic and psalmodic modes (Figure 4-4 and Figure 4-8 below). The monody (both modes considered) spans eleven degrees from middle C to F', a typical soprano vocal range of an octave and a half. Tenor and bass cantors will sing one octave lower than indicated by the treble clefs below, beginning on 'C. Figure 4-4 Phrygian Mode Scale in C Major Drga Tvir Sluk Mrka Tfka Atnk Mnkh 23 Mhpk 24 In her Music of the Bible Revealed Haïk-Vantoura analyzes the structure of this In addition to these characteristics, the melody actually terminates (as the medieval treatises on the te amim [cantillation marks] state) by progressing from high to low, rather than low to high. * Nearly every verse begins with a rising melody, starting from * The text above in parentheses is misleading and may be deemed anti-semitic. A more faithful rendering reads: as our treatises have taught us. One can infer that the translator correctly understands the firstperson plural anaphors to refer to the author and her fellow Jews by his interpolation of the adjective medieval, which does not occur in her text. The medieval treatises on cantillation marks were written by Jews, as opposed to later treatises that were written by Jews and Christians alike. The English translation silently suppresses Haïk-Vantoura s reference to her own religious education and cultural background, which context alone explains her early exposure to and lifelong interest in the cantillation of the Hebrew Bible. The American edition mentions nowhere that the author was Jewish, although the biographical note relates without comment that her studies were interrupted by World War II. Haïk-Vantoura and her family fled Vichy France for their lives. More than thirty years later she identifies herself in the disputed phrase as a Jewess writing in the context of Jewish tradition. On the other hand, the editor s preface tells of his own childhood memories of Christian Sunday School. With regard to the translation John Wheeler states: I have taken pains to examine closely both the classical French text (with its figurative and often highly elliptical expressions) and Dennis Weber s English translation (which was originally made for study purposes). I have attempted to retain the literal sense of the French text and of Mr. Weber s translation as much as possible without sacrificing clarity. What editorial alterations I have made of the translation were made with the aim of presenting clearly and faithfully the message of the French book. The qualification of Haïk-Vantoura s writing style ( classical, figurative, and elliptical ) does not ring true. Francophone readers may form an esthetic judgment of the source text from the quotations in the endnotes below. (For a concrete example of her wordy prose, see endnote 1 and comment.) Her longwinded style bears little resemblance to classical (late-seventeenth-century) French language and literature, and would have no reason to do so. Application 48

49 the tonic. All concluding cadences, without exception, depict a descending movement as they rejoin the same tonic note. Furthermore, this tonic note s attraction is not to a lower leading tone but an upper one. * And finally, as the recent deductions of musicologists would lead one to presume, the finale or tonic has an invariable median position in the tonal scale from whence it governs the relationships. 25 Haïk-Vantoura writes fixed-do musical notation, as note lettering is not used in France (nor in many other Romance- and Slavic-language countries). Her musical thought may bear the influence of her education. In tying the syllable do to the note C, this method frees the tonic to wander upwards from C to E (with respect to the modern, western, classical convention that defines the tonic as the first note in the scale) more easily than does movable-do notation, which fits the syllables of solfège to the degrees of the scale. (To compare fixed- and movable-do representations of the scales, see Prosodic Pitch Chart and Psalmodic Pitch Chart below.) Figure 4-5 Other Prosodic Sublinear Signs DbMr 26 Gaya 27 Glgl Ytiv 28 A sublinear sign remains active until the occurrence of the next sublinear sign. Every unmarked syllable is sung on its tone, which recurs after an intervening superlinear sign, whose duration is limited to the syllable it marks. Since Cantillizer does not support Hebrew-language text, the sublinear sign s note is represented only once. The cantor will realize this written notation in the melody and reiterate it, since every sung syllable must by definition receive a pitch value. Prosodic Pitch Chart Integer MIDI Letter Degree Sign Function Fixed-Do Movable-Do C 6 Darga Superdominant Do Mi C# Do# D 7 Tevir Subtonic Re Fa Apparently born in Paris, Haïk-Vantoura seems to have been the child of a Yiddish- and/or Alsatianspeaking Ashkenazi mother and a Ladino-speaking Sephardic father. At the time of her birth, historically germanophone Alsace, home to a large Jewish community, had been German territory for more than thirty years, and would resist gallicization for another fifty years. Jews expulsed from Spain by the Inquisition in 1492 continued to speak Judeo-Spanish in Turkey for more than four centuries. If French was the primary language spoken in the Vantoura home, it was almost certainly a foreign tongue to both parents. Translator Dennis Weber and editor John Wheeler have both been given the opportunity to explain in this space. Neither has answered yet. [Note and italics by SAGReiss.] * This characteristic upper leading tone in the Dorian [Haïk-Vantoura refers to the Phrygian mode as Dorian in an older acceptation of the latter term. Note by SAGReiss.] mode is called, however, the 2 nd degree. The melody rises so naturally before returning towards its starting point. [Author s note and italics. Translation by Dennis Weber.] Application 49

50 Integer MIDI Letter Degree Sign Function Fixed-Do Movable-Do D# Re# E 1 Silluk Tonic Mi So F 2 Merekha Supertonic Fa La F# Fa# G 3 Tifkha Mediant Sol 30 Ti G# Sol# A 4 Atnakh Subdominant La Do A# La# B 5 Munakh Dominant Si 31 Re C' 6 Mehupakh Superdominant Do Mi Prosodic Rhythm & Melody The syllable count determines the rhythm of cantillation. Not enough is currently known about Ancient Hebrew versification to characterize the rhythm of the prosodic books of the Bible. Suffice to say that it feels quite irregular (or prosaic) to the modern reader, except for a few passages (Song of Deborah, Song of Hannah, Song of Moses, Song of Songs, Song of the Sea, Song of the Well) that feel more poetic, in content if not in form. Cantillizer hopes to contribute to the research in this field. The esthetics of prosody vary wildly according to the style of the text, from mythological narrative (Genesis) to liturgy and legislation (Deuteronomy), from epic history (Samuel) to ideological harangue (Jeremiah), from divine hallucination (Ezekiel) to philosophical poem in prose (Ecclesiastes). The sublinear signs determine the pitch value of the subordinate degrees (appoggiature) and ornamental figures (melismata), corresponding to the superlinear signs. Nevertheless, these melodic embellishments are mandatory, contrary to the optional or improvised grace notes and flourishes of baroque music. Haïk-Vantoura enumerates the functions of prosodic ornaments: Let us bear in mind that the cadential significance of the 5 th degree, compared with that of the 4 th degree (when it is part of the ancient prosody), is more suspensive (deprived as it is of harmonic support); from whence comes the feminine ending which affects it so naturally. This ending generally takes one of two characteristic aspects: one being a simple flection [little zakef], the other a more shaded flection [great revia]. This ending s reserved contours not only give a particular twist to the word concerned, but at the same time to the phrase it terminates. It is a sort of cautious commentary, justified by its special location at the caesura in relationship to the context. Now we can see the functional meaning of the different melismas as instruments of the punctuation, although we must not generalize nor lose sight of the primacy of the tonal functions in this office. Let us summarize them here. This melisma [great telisha] often introduces the first word of a phrase, whereas this one [little telisha] produces a prolongation, as does moreover this next one [great zakef]. This sign [little pazer] sets off the word like a curl in the middle of the discourse. This next melisma [segolta], which never appears in the middle of the phrase, distinguishes itself from the others by a recognized characteristic: it does not mark a consequential rest, but a simple caesura, and it is always preceded by this sign [zarka] which delicately shades the middle of the same incidental clause. 32 Application 50

51 (For more information on the function of these melismata, see au-dessous.) In Figure 4-6 and Figure 4-7 below, it is assumed that silluk, the tonic E (default value), is the active (preceding) sublinear sign or constituent degree, based on whose pitch the note values of superlinear signs are calculated. The tonal relationships remain constant as the sublinear sign changes. The rhythm of the ornament always equals the value of a quarter note. (All formulæ appear in au-dessus.) For example, the formula for great pazer is the following. Great Pazer (three 32 nd notes, one 16 th note, three 32 nd notes): 1. Constituent degree + 3 diatonic degrees 2. Constituent degree + 2 diatonic degrees 3. Constituent degree + 1 diatonic degree 4. Constituent degree 5. Constituent degree + 1 diatonic degree 6. Constituent degree + 2 diatonic degrees 7. Constituent degree + 3 diatonic degrees Figure 4-6 Prosodic Appoggiature Azla 33 Grsh LtZk Psta Figure 4-7 Prosodic Melismata DbGr GtPz GtRv GtTl GtZk LtPz LtTl RdPs 34 Sgol Slsl 35 Zrka Application 51

52 Haïk-Vantoura puts great emphasis on the precise syllabic location of superlinear signs, although it is not clear how she translates two-dimensional graphical space into linear phonetic time in order to define the beginning, middle, and end of a syllable. 36 Since it does not support Hebrew-language text, Cantillizer lacks this information. Hebraists may wish to create these linguistic data for themselves. If the syllable following a superlinear sign (except ole veyored, which already returns to the constituent degree) is marked with a sign other than the active (preceding) sublinear sign, it is suggested to resolve the ornament s note sequence to the constituent degree, and adjust the rhythm so that the total duration remains equal to the value of a quarter note Psalmodic Mode & Scale The psalmodic mode scale lacks middle C, the superdominant in the lower octave. 38 Figure 4-8 Harmonic Mode Scale in E Minor Glgl Sluk Mrka Trka 39 Atnk Mnkh 40 Mhpk 41 Haïk-Vantoura explains the specificity of this scale: The mode actually takes shape at the same time as the scale, being a chromatic minor similar to our own, with one slight difference: the connection between the 6 th and 7 th degrees (in their fundamental nature as represented by the lower signs) is avoided. The scale limits itself to the upper 6th degree at the top and the lower 7th degree at the bottom. The characteristic interval of the augmented 2 nd between them therefore does not fundamentally exist. Only the presence of subordinate degrees makes it occur. 42 Figure 4-9 Other Psalmodic Sublinear Signs Dkhi 43 Gaya 44 Psalmodic Pitch Chart Integer MIDI Letter Degree Sign Function Fixed-Do Movable-Do C Do Fa C# Do# D Re Application 52

53 Integer MIDI Letter Degree Sign Function Fixed-Do Movable-Do D# 7 Galgal Subtonic Re# Si E 1 Silluk Tonic Mi La F Fa F# 2 Merekha Supertonic Fa# Ti G 3 Tarkha Mediant Sol 47 Do G# Sol# A 4 Atnakh Subdominant La Re A# La# B 5 Munakh Dominant Si 48 Mi C' 6 Mehupakh Superdominant Do Fa Psalmodic Rhythm & Melody The rhythm of the psalmodic books of the Bible feels more regular, more poetic, than the rhythm of prosody. The typographical convention of leaving a blank space to mark the cæsura reinforces this impression. The esthetics of psalmody embrace song and prayer (Psalms), theological thought (Job), and aphoristic wisdom (Proverbs). Haïk-Vantoura characterizes the style of psalmody: The following melismas in the prosodic system are absent in the poetic books: [great zakef, segolta, great telisha, little telisha, double geresh]. Upon reflection, this is not surprising. They figured in the narratives and the exhortations, either for directing the believer s attention to key words [great telisha], or for marking with retrocession certain occasions of an incidental clause [great zakef], or, on the contrary, terminating a word with the flourish of a panache [little telisha]. These means, these artifices we are tempted to say (even though legitimate everywhere they are found), have no place in the Psalter. The psalmodic cantillation is a harmony which conjugates with that of the words. Both blend their respective purity in order to support the special effusion of prayer. It is useless to employ devices in order to attract the attention of Him who knows everything! When all is said and done, the presence of these melismas, authentic oratorical structures that they are, would be most unusual in the Psalms. 49 In Figure 4-10 and Figure 4-11 below, it is assumed that silluk, the tonic E (default value), is the active (preceding) sublinear sign or constituent degree, based on whose pitch the note values of superlinear signs are calculated. The tonal relationships remain constant as the sublinear sign changes. The rhythm of the ornament always equals the value of a quarter note. (All formulæ appear in au-dessus.) For example, the formula for shalshelet is the following. Shalshelet (a triplet of eighth notes): 1. Constituent degree 2 diatonic degrees 2. Constituent degree 2 chromatic intervals 3. Constituent degree 1 chromatic interval Application 53

54 Figure 4-10 Psalmodic Appoggiature Azla 50 Iluy 51 LtRv 52 GrMk 53 Figure 4-11 Psalmodic Melismata LtPz OlVy 54 Slsl 55 Tnor Tnri 56 Haïk-Vantoura describes the distinctive leap and fall of the superlinear sign ole veyored when following the constituent degree of galgal (D#): Melodically speaking, the subtonic note the only basic scale degree below the tonic note in psalmody exercises a curious influence, highly different from that of our present-day leading tone [ ] Even though it is just a half-step from the tonic, it almost never cedes to its attraction nor resolves directly to it (which is not the case at all for our classical leading tone). In the concept which emerges, this sub-tonic creates the effect of a set of balances. * As soon as this 7 th note of the scale is attained by the melodic line, as if in recoil, it nearly always sets in motion a highly characteristic rebound to the upper 4 th (most often a diminished 4 th ) and, after a melodic oscillation, arrived at again, it proceeds to finally conclude on the 2 nd degree of the mode [see 4.7.6, v. 6, below] The melody thus determines a suspending cadence through a figure charged with expression Psalmodic Instrumental Accompaniment A score of Psalms call for the accompaniment of specific musical instruments in their ascription (v. 1): Psalm Section Transliteration Instrument 22 Hapax legomenon ayelet hashakhar 58 8, 84 Strings gitit 9 Hapax legomenon mut laben 53, 88 Dis legomenon makhalat leanot 4, 6, 54, 55, 61, 67, 76 Strings neginot (al-hasheminit) - 5 Woodwinds nekhilot 46 Dis legomenon alamot * An observation curiously supported by its etymology: literally, a wheel. [Author s note. Translation by Dennis Weber.] Application 54

55 Psalm Section Transliteration Instrument 12, 45, 60, 69, 80 Horns shushan (edut) Psalm 150 (v. 3-5) seems to evoke a street orchestra and a parade of dancers: Section Transliteration Instrument Text Strings Strings Woodwinds Percussion Horns Percussion kinor nebel ugab tselatsali shofar tof Unfortunately the interpretation of these Ancient Hebrew technical terms (harp, lute, pipe, horn, timbrel, cymbal, etc.) remains highly speculative, so much so that certain translations of the Bible simply transliterate many of the words. Some of the above may not even refer to the names of musical instruments. The shofar, or ram s horn, is the only musical instrument that survives in an unbroken tradition from biblical antiquity to modern times. 59 The word informs the Sephardic name of two cantillation signs, shofar holekh (munakh) and shofar mehupakh (mehupakh). Figure 4-12 Blowing the Shofar on Rosh Hashanah Application 55

56 Figure 4-13 Salomon Helperin Blowing the Yemenite Shofar (2006) In the Bible the shofar announces battles, parades, festivals, the new moon, coronations, and religious ceremonies. Today the shofar still sounds in synagogues around the world on Rosh Hashanah (Jewish New Year) and Yom Kippur (Day of Atonement). 60 Haïk-Vantoura gives the following argument in favor of instrumentation, absent from the tradition of synagogues, where the cantor generally sings a cappella: [ ] the psalmody rather than the prosody, seems to have been created with an instrumental accompaniment in mind. * Its configuration (with its often disjointed melodic movements, notably in thirds, and particularly the V-I cadence found frequently at the end of verses) testifies to this. Its structure implies a harmonization. Again, these given facts reveal a music created for a specific purpose. 61 Composers may wish to partition and/or orchestrate the choral scores. While Haïk- Vantoura transcribed the music in several modes and keys, Cantillizer supports two, Phrygian in C major for prosody and harmonic in E minor for psalmody. Musicians may need to transpose scores in order to accommodate their instruments and arrangements. * But apart from the Song of Songs and other canticles, of course, it is quite possible that the prose texts were accompanied by the lyre or harp in common use [ ] [Author s note and italics. Translation by Dennis Weber.] Application 56

57 4.7.6 Psalm 137 By the Rivers of Babylon 62, 63 Psalm 137 celebrates, laments, and threatens brutally to avenge the fallen city of Jerusalem. The poem appears to have been written between 586 BCE (date of the destruction of the Temple of Solomon and of the enslavement of the Jews by King Nebuchadnezzar II of Babylonia) and 538 BCE (date of the liberation of the Jews by King Cyrus II of Persia), or in any case before 516 BCE (date of the destruction of Babylon by King Darius I of Persia, and of the construction by the Jews of the second Temple). Authorship has been diversely attributed to King David (died c. 970 BCE) and the prophet Jeremiah (died a refugee in Egypt after 562 BCE), however the narrator s claim to be a Levite singer seems compelling. Figure 4-14 Marc Chagall, The Capture of Jerusalem by Nebuchadnezzar (1956) Application 57

58 Figure 4-15 Psalm 137 By the Rivers of Babylon in E Minor Application 58

59 Figure 4-16 Synagogue at Gaza, Mosaic of King David Strumming the Harp (6 th century CE) Application 59