|qarney para||֟||telisha gedola||֠|
|telisha qetana||֩||yerah ben yomo||֪|
Hebrew cantillation, trope, trop, or te'amim is the manner of chanting ritual readings from the Hebrew Bible in synagogue services. The chants are written and notated in accordance with the special signs or marks printed in the Masoretic Text of the Bible, to complement the letters and vowel points.
These marks are known in English as 'accents' (diacritics), 'notes' or trope symbols, and in Hebrew as taʿamei ha-mikra (טעמי המקרא) or just teʿamim (טעמים). Some of these signs were also sometimes used in medieval manuscripts of the Mishnah. The musical motifs associated with the signs are known in Hebrew as niggun or neginot (not to be confused with Hasidic nigun) and in Yiddish as trop (טראָפ): the word trope is sometimes used in Jewish English with the same meaning.
There are multiple traditions of cantillation. Within each tradition, there are multiple tropes, typically for different books of the Bible and often for different occasions. For example, different chants may be used for Torah readings on Rosh Hashana and Yom Kippur than for the same text on a normal Shabbat.
Three systems of Hebrew punctuation (including vowels and cantillation symbols) have been used: the Babylonian, the Palestinian and the Tiberian, only the last of which is used today.
Babylonian Biblical manuscripts from the Geonic period contain no cantillation marks in the current sense, but small Hebrew letters are used to mark significant divisions within a verse. Up to eight different letters are found, depending on the importance of the break and where it occurs in the verse: these correspond roughly to the disjunctives of the Tiberian system. For example, in some manuscripts the letter tav, for tevir (break), does duty for both Tiberian tevir and zaqef. In general there are no symbols for the conjunctives, though some late manuscripts use the Tiberian symbols for these. There is also no equivalent for low-grade disjunctives such as telisha gedolah: these are generally replaced by the equivalent of zaqef or revia.
Nothing is known of the musical realization of these marks, but it seems likely that they represent breaks or variations in a set melody applied to each verse. (A somewhat similar system is used in manuscripts of the Qur'an to guide the reader in fitting the chant to the verse: see Qur'an reading.)
This system is reflected in the cantillation practices of the Yemenite Jews, who now use the Tiberian symbols, but tend to have musical motifs only for the disjunctives and render the conjunctives in a monotone. It is notable that the Yemenite Jews have only eight disjunctive motifs, thus clearly reflecting the Babylonian notation. The same is true of the Karaite mode for the haftarah; while in the Sephardi haftarah modes different disjunctives often have the same or closely similar motifs, reducing the total number of effective motifs to something like the same number.
The Babylonian system, as mentioned above, is mainly concerned with showing breaks in the verse. Early Palestinian manuscripts, by contrast, are mainly concerned with showing phrases: for example the tifcha-etnachta, zarqa-segolta and pashta-zaqef sequences, with or without intervening unaccented words. These sequences are generally linked by a series of dots, beginning or ending with a dash or a dot in a different place to show which sequence is meant. Unaccented words (which in the Tiberian system carry conjunctives) are generally shown by a dot following the word, as if to link it to the following word. There are separate symbols for more elaborate tropes like pazer and telisha gedolah.
The manuscripts are extremely fragmentary, no two of them following quite the same conventions, and these marks may represent the individual reader's aide-memoire rather than a formal system of punctuation (for example, vowel signs are often used only where the word would otherwise be ambiguous). In one manuscript, presumably of somewhat later date than the others, there are separate marks for different conjunctives, actually outnumbering those in the Tiberian system (for example, munach before etnachta has a different sign from munach before zaqef), and the overall system approaches the Tiberian in comprehensiveness. In some other manuscripts, in particular those containing Targumim rather than original text, the Tiberian symbols have been added by a later hand. In general, it may be observed that the Palestinian and Tiberian systems are far more closely related to each other than either is to the Babylonian.
This system of phrasing is reflected in the Sephardic cantillation modes, in which the conjunctives (and to some extent the "near companions" such as tifcha, pashta and zarqa) are rendered as flourishes leading into the motif of the following disjunctive rather than as motifs in their own right.
The somewhat inconsistent use of dots above and below the words as disjunctives is closely similar to that found in Syriac texts. Kahle also notes some similarity with the punctuation of Samaritan Hebrew.
By the tenth century CE, the chant in use in medieval Palestine had clearly become more complex, both because of the existence of pazer, geresh and telisha motifs in longer verses and because the realization of a phrase ending with a given type of break varied according to the number of words and syllables in the phrase. The Tiberian Masoretes therefore developed a comprehensive notation with a symbol on each word, to replace the fragmentary systems previously in use. In particular, it was necessary to invent a range of different conjunctive accents to show how to introduce and elaborate the main motif in longer phrases. (For example, tevir is preceded by mercha, a short flourish, in shorter phrases but by darga, a more elaborate run of notes, in longer phrases.) The system they devised is the one in use today, and is found in Biblical manuscripts such as the Aleppo Codex. A Masoretic treatise called Diqduqe ha-teʿamim (precise rules of the accents) by Aaron ben Moses ben Asher survives, though both the names and the classification of the accents differ somewhat from those of the present day.
As the accents were (and are) not shown on a Torah scroll, it was found necessary to have a person making hand signals to the reader to show the tune, as in the Byzantine system of neumes. This system of cheironomy survives in some communities to the present day, notably in Italy. It is speculated that both the shapes and the names of some of the accents (e.g. tifcha, literally "hand-breadth") may refer to the hand signals rather than to the syntactical functions or melodies denoted by them. Today in most communities there is no system of hand signals and the reader learns the melody of each reading in advance.
The Tiberian system spread quickly and was accepted in all communities by the 13th century. Each community re-interpreted its reading tradition so as to allocate one short musical motif to each symbol: this process has gone furthest in the Western Ashkenazi and Ottoman (Jerusalem-Sephardi, Syrian etc.) traditions. Learning the accents and their musical rendition is now an important part of the preparations for a bar mitzvah, as this is the first occasion on which a person reads from the Torah in public.
In the early period of the Reform movement there was a move to abandon the system of cantillation and give Scriptural readings in normal speech (in Hebrew or in the vernacular). In recent decades, however, traditional cantillation has been restored in many communities.
There are two systems of cantillation marks in the Tanakh. One is used in the twenty-one prose books, while the other appears in the three poetical books of Psalms, Proverbs and Job. Except where otherwise stated, this article describes the "prose" system.
The current system of cantillation notes has its historical roots in the Tiberian masorah. The cantillation signs are included in Unicode as characters U+0591 through U+05AF in the Hebrew block.
The names of some of the cantillation signs differ in the Ashkenazi, Sephardi, Italian and Yemenite traditions; for example Sephardim use qadma to mean what Ashkenazim call pashta, and azla to mean what Ashkenazim call qadma. In this article, as in almost all Hebrew grammars, the Ashkenazi terminology is used. The names in other traditions are shown in the table below.
A primary purpose of the cantillation signs is to guide the chanting of the sacred texts during public worship. Very roughly speaking, each word of text has a cantillation mark at its primary accent and associated with that mark is a musical phrase that tells how to sing that word. The reality is more complex, with some words having two or no marks and the musical meaning of some marks dependent upon context. There are different sets of musical phrases associated with different sections of the Bible. The music varies with different Jewish traditions and individual cantorial styles.
The cantillation signs also provide information on the syntactical structure of the text and some say they are a commentary on the text itself, highlighting important ideas musically. The tropes are not random strings but follow a set and describable grammar. The very word ta'am, used in Hebrew to refer to the cantillation marks, literally means "taste" or "sense", the point being that the pauses and intonation denoted by the accents (with or without formal musical rendition) bring out the sense of the passage.
The cantillation signs serve three functions:
|Syntax||They divide biblical verses into smaller units of meaning, a function which also gives them a limited but sometimes important role as a source for exegesis. This function is accomplished through the use of various conjunctive signs (which indicate that words should be connected in a single phrase) and especially a hierarchy of dividing signs of various strength which divide each verse into smaller phrases. The function of the disjunctive cantillation signs may be roughly compared to modern punctuation signs such as periods, commas, semicolons, etc.|
|Phonetics||Most of the cantillation signs indicate the specific syllable where the stress (accent) falls in the pronunciation of a word.|
|Music||The cantillation signs have musical value: reading the Hebrew Bible with cantillation becomes a musical chant, where the music itself serves as a tool to emphasise the proper accentuation and syntax (as mentioned previously).|
In general, each word in the Tanakh has one cantillation sign. This may be either a disjunctive, showing a division between that and the following word, or a conjunctive, joining the two words (like a slur in music). Thus, disjunctives divide a verse into phrases, and within each phrase all the words except the last carry conjunctives. (There are two types of exception to the rule about words having only one sign. A group of words joined by hyphens is regarded as one word so they only have one accent between them. Conversely, a long word may have two—e.g., a disjunctive on the stressed syllable and the related conjunctive two syllables before in place of meteg.)
The disjunctives are traditionally divided into four levels, with lower level disjunctives marking less important breaks.
The general conjunctive is munach. Depending on which disjunctive follows, this may be replaced by mercha, mahpach, darga, qadma, telisha qetannah or yerach ben yomo.
One other symbol is mercha kefulah, double mercha. There is some argument about whether this is another conjunctive or an occasional replacement for tevir.
Disjunctives have a function somewhat similar to punctuation in Western languages. Sof pasuk could be thought of as a full stop, atnach as a semi-colon, second level disjunctives as commas and third level disjunctives as commas or unmarked. Where two words are written in the construct state (for example, pene ha-mayim, "the face of the waters"), the first noun (nomen regens) invariably carries a conjunctive.
The cantillation signs are often an important aid in the interpretation of a passage. For example, the words qol qore bamidbar panu derekh YHWH (Isaiah 40:3) is translated in the Authorised Version as "The voice of him that crieth in the wilderness, Prepare ye the way of the LORD". As the word qore takes the high-level disjunctive zakef katon this meaning is discouraged by the cantillation marks. Accordingly, the New Revised Standard Version translates "A voice cries out: 'In the wilderness prepare the way of the LORD, ...'" while the New Jewish Publication Society Version has "A voice rings out: 'Clear in the desert a road for the LORD'."
Most cantillation signs are written on the consonant of the stressed syllable of a word. This also shows where the most important note of the musical motif should go.
A few signs always go on the first or last consonant of a word. This may have been for musical reasons, or it may be to distinguish them from other accents of similar shape. For example, pashta, which goes on the last consonant, otherwise looks like kadma, which goes on the stressed syllable.
Some signs are written (and sung) differently when the word is not stressed on its last syllable. Pashta on a word of this kind is doubled, one going on the stressed syllable and the other on the last consonant. Geresh is doubled unless it occurs on a non-finally-stressed word or follows kadma (to form the kadma ve-azla phrase).
Cantillation signs guide the reader in applying a chant to Biblical readings. This chant is technically regarded as a ritualized form of speech intonation rather than as a musical exercise like the singing of metrical hymns: for this reason Jews always speak of saying or reading a passage rather than of singing it. (In Yiddish the word is leynen 'read', derived from Latin legere, giving rise to the Jewish English verb "to leyn".)
The musical value of the cantillation signs serves the same function for Jews worldwide, but the specific tunes vary between different communities. The most common tunes today are as follows.
There has been an attempted reconstruction of the original melody by Suzanne Haïk-Vantoura, on the basis of the shapes and positions of the marks and without any reference to existing melodies, as described in her book La musique de la Bible révélée and her records. That reconstruction assumes the signs represent the degrees of various musical scales, that is individual notes, which puts it at odds with all existing traditions where the signs invariably represent melodic motives; it also takes no account of the existence of older systems of notation, such as the Babylonian and Palestinian systems. Musicologists have rejected her results as dubious and her methodology as flawed. A similar reconstructive proposal was developed by American composer and pianist Jeffrey Burnsand posthumously published in 2011.
In the Ashkenazic musical tradition for cantillation, each of the local geographical customs includes a total of six major and numerous minor separate melodies for cantillation:
The Ashkenazic tradition preserves no melody for the special cantillation notes of Psalms, Proverbs, and Job, which were not publicly read in the synagogue by European Jews. However, the Ashkenazic yeshiva known as Aderet Eliyahu, or (more informally) Zilberman's, in the Old City of Jerusalem, uses an adaptation of the Syrian cantillation-melody for these books, and this is becoming more popular among other Ashkenazim as well.
At the beginning of the twentieth century there was a single Ottoman-Sephardic tradition (no doubt with local variations) covering Turkey, Syria, Palestine and Egypt. Today the Jerusalem-Sephardic, Syrian, Egyptian and Baghdadi melodies recognisably belong to a single family. For example, in these traditions the Torah reading is always or almost always in Maqam Sigah. There are some variations, among individual readers as well as among communities: for example the Egyptian melody is related to the more elaborate and cantorial form of the Syrian melody and was transitioning toward Maqam Huzzam before the mass expulsion in 1950. The Karaite tradition, being based on the Egyptian, also forms part of this group.
Another recognisable family consists of the Iraqi (Mosul and Iraqi diaspora), Spanish-Moroccan and Spanish and Portuguese melodies. The probable reason for the occurrence of similar melodies at opposite ends of the Arab world is that they represent the remains of an old Arab-Jewish tradition not overlaid by the later Ottoman-Sephardic tradition that spread to the countries in between. There may also have been some convergence between the London Spanish and Portuguese and Iraqi melodies during British rule in India and the British Mandate of Mesopotamia.
The Jews of North Africa, the Middle East, Central Asia and Yemen all had local musical traditions for cantillation. When these Jewish communities emigrated (mostly to Israel) during the twentieth century, they brought their musical traditions with them. But as the immigrants themselves grew older, many particular national melodies began to be forgotten, or to become assimilated into the "Jerusalem Sephardic" melting-pot.
As with the Ashkenazim, there is one tune for Torah readings and a different tune for haftarot. Spanish and Portuguese Jews have a special tune for the Ten Commandments when read according to the ta'am elyon, known as "High Na'um", which is also used for some other words and passages which it is desired to emphasize. Other communities, such as the Syrian Jews, observe the differences between the two sets of cantillation marks for the Ten Commandments but have no special melody for ta'am 'elyon. There is no special tune for Rosh Hashanah and Yom Kippur in any Sephardic tradition. As with Ashkenazim, the normal musical value of cantillation signs is replaced by a "coda" motif at the end of each Torah reading and of each haftarah verse (though there is no special coda for the end of the haftarah), suggesting a common origin for the Sephardi and Ashkenazi chants.
Eastern Jewish communities have no liturgical tradition of reading Ecclesiastes, and there is no public liturgical reading of Song of Songs on Passover, though brief extracts may be read after the morning service during the first half of Nisan. (Individuals may read it after the Passover Seder, and many communities recite it every Friday night.) There are specialized tunes for Song of Songs, Ruth, Esther and Lamentations. The prose passages at the beginning and end of the book of Job may be read either to the tune of Song of Songs or to that of Ruth, depending on the community. The Ruth tune is generally the "default" tune for any book of the Ketuvim (Hagiographa) that does not have a tune of its own.
Unlike the Ashkenazic tradition, the eastern traditions, in particular that of the Syrian Jews, include melodies for the special cantillation of Psalms, Proverbs and the poetic parts of Job. In many eastern communities, Proverbs is read on the six Sabbaths between Passover and Shavuot, Job on the Ninth of Av, and Psalms are read on a great many occasions. The cantillation melody for Psalms can also vary depending on the occasion. The Spanish and Portuguese Jews have no tradition for the rendering of the Psalms according to the cantillation marks, but the melody used for several psalms in the evening service is noticeably similar to that of Syrian psalm cantillation, and may represent the remnants of such a tradition.
Yemenite cantillation has a total of eight distinctive motifs, falling within four main patterns:
This is true equally of the system used for the Torah and the systems used for the other books. It appears to be a relic of the Babylonian system, which also recognised only eight types of disjunctive and no conjunctives.
Some communities had a simplified melody for the Torah, used in teaching it to children, as distinct from the mode used in synagogue. (This should not be confused with the lernen steiger used for studying the Mishnah and Talmud.) For example, the Yemenite community teaches a simplified melody for children, to be used both in school and when they are called to read the sixth aliyah. The simplified melody is also used for the reading of the Targum, which is generally performed by a young boy.
Conversely, the Syrian community knows two types of Torah cantillation, a simpler one for general use and a more elaborate one used by professional hazzanim. It is probable that the simpler melody was originally a teaching mode. Today however it is the mode in general use, and is also an ancestor of the "Jerusalem-Sephardic" melody.
Some communities, such as the Portuguese community of Amsterdam, have a simplified melody for the Prophets for study purposes, distinct from that used in reading the Haftarah: the distinction is mentioned in one medieval Sephardic source.
The following table shows the names of the te'amim in the Ashkenazi, Sephardi, and Italian traditions together with their Unicode symbols.
|Sof passuk||ב ׃||U+05C3||סוֹף פָּסֽוּק
Sof pasuk / silluk
|Zakef katan||ב֔||U+0594||זָקֵף קָטָ֔ן
|Zakef gadol||ב֕||U+0595||זָקֵף גָּד֕וֹל
|Shene pashtin||ב֨ב֙||U+0599, U+05A8||שְׁנֵ֨י פַּשְׁטִין֙
|Qarne farah||ב֟||U+059F||קַרְנֵי פָרָ֟ה
Qarne farah / pazer gadol
|Telisha gedola||ב֠||U+05A0||תְּ֠לִישָא גְדוֹלָה
Geresh / Azla
|אַזְלָ֨א or קַדְמָ֨א
Azla or Qadma
|Telisha ketana||ב֩||U+05A9||תְּלִישָא קְטַנָּה֩
|Mercha kefula||ב֦||U+05A6||מֵרְכָא כְּפוּלָ֦ה
|Yerach ben yomo||ב֪||U+05AA||יֵרֶח בֶּן יוֹמ֪וֹ
Yerach ben yomo/ galgal
|יֵרֶח בֶּן יוֹמ֪וֹ
Yeraḥ ben yomo
|יֵרֶח בֶּן יוֹמ֪וֹ|
Yerach ben yomo
The following additional symbols are found in the three poetical books: their names do not differ among the various traditions.
|Name||Symbol||Unicode||Hebrew name in Unicode|
|Geresh muqdam||ב֝||U+059D||גֵּרֵשׁ מֻקְדָּם֝|
|Atnach hafukh||ב֢||U+05A2||אתנח הפוך֢|
For learning purposes, the t'amim are arranged in a traditional order of recitation called a "zarqa table", showing both the names and the symbols themselves. These tables are often printed at the end of a Chumash (Hebrew Pentateuch).
The order of recitation bears some relation to the groups in which the signs are likely to occur in a typical Biblical verse, but differs in detail between different communities. Below are traditional Ashkenazi and Sephardi orders, though variations are found in both communities.
Numbers 35:5 (in Parshat Mas'ei) has two notes found nowhere else in the Torah:
The rules governing the sequence of cantillation marks are as follows.
|Nearest preceding lower level
disjunctive ("near companion")
|Other lower level disjunctives
|First level disjunctives|
|Sof pasuk||Mercha||Tifcha||Zaqef qaton|
|Second level disjunctives|
|Zaqef qaton||Munach||Pashta||Revia||Zaqef gadol|
|Third level disjunctives|
|Munach legarmeh||Geresh, Telisha gedolah, Pazer|
|Geresh/Azla/Gershayim||Telisha gedolah, Pazer|
|Geresh/Azla/Gershayim||Telisha gedolah, Pazer||Yetiv|
|Tevir||Mercha or Darga;
Qadma Mercha or
|Geresh/Azla/Gershayim||Telisha gedolah, Pazer|
|Fourth level disjunctives|
Telisha qetannah Qadma
|Qarne farah||Yerach ben yomo|
The following sequences are commonly found:
The system of cantillation signs used throughout the Tanakh is replaced by a very different system for these three poetic books. Many of the signs may appear the same or similar at first glance, but most of them serve entirely different functions in these three books. (Only a few signs have functions similar to what they do in the rest of the Tanakh.) The short narratives at the beginning and end of Job use the "regular" system, but the bulk of the book (the poetry) uses the special system. For this reason, these three books are referred to as sifrei emet (Books of Truth), the word emet meaning "truth", but also being an acronym (אמ״ת) for the first letters of the three books (Iyov, Mishle, Tehillim).
A verse may be divided into one, two or three stichs. In a two-stich verse, the first stich ends with atnach. In a three-stich verse, the first stich ends with oleh ve-yored, which looks like mahpach (above the word) followed by tifcha, on either the same word or two consecutive words, and the second stich ends with atnach.
Major disjunctives within a stich are revia qaton (immediately before oleh ve-yored), revia gadol (elsewhere) and tzinnor (which looks like zarqa). The first (or only) stich in a verse may be divided by dechi, which looks like tifcha but goes under the first letter of the word to the right of the vowel sign. The last stich in a two- or three-stich verse may be divided by revia megurash, which looks like geresh combined with revia.
Minor disjunctives are pazer gadol, shalshelet gedolah, azla legarmeh (looking like qadma) and mehuppach legarmeh (looking like mahpach): all of these except pazer are followed by a pesiq. Mehuppach without a pesiq sometimes occurs at the beginning of a stich.
All other accents are conjunctives.
Some manuscripts of early Rabbinic literature contain marks for partial or systematic cantillation. This is true of the Sifra, and especially of Genizah fragments of the Mishnah.
Today, many communities have a special tune for the Mishnaic passage "Bammeh madliqin" in the Friday night service. Otherwise, there is often a customary intonation used in the study of Mishnah or Talmud, somewhat similar to an Arabic mawwal, but this is not reduced to a precise system like that for the Biblical books. Recordings have been made for Israeli national archives, and Frank Alvarez-Pereyre has published a book-length study of the Syrian tradition of Mishnah reading on the basis of these recordings.
On the relationship between the cantillation marks found in some manuscripts and the intonation used in Ashkenazi Talmud study, see Zelda Kahan Newman, The Jewish Sound of Speech: Talmudic Chant, Yiddish Intonation and the Origins of Early Ashkenaz.
The Jewish born Christian convert Ezekiel Margoliouth translated the New Testament to Hebrew in 1865 with cantillation marks added. It is the only completely cantillated translation of the New Testament. The translation was published by the London Jews' Society.
Khazdan E. (2021) From Masoretic Signs to Cantillation Marks: Initial Steps (On the Virtual Dialogue between Alfonso de Zamora and Johannes Reuchlin). Lietuvos muzikologija. T. 22.
The recordings held at the Commons are organized by the Vayavinu Bamikra Project at Wikisource in the following languages: