Prolegomena to the Masoretic te'amim

Prolegomena to the Masoretic te‘amim David C. Mitchell Good morning. It is a great honour to address you today. My specialism is the Hebrew Bible, particularly the book of Psalms. I’m also a Director of Music within the Anglican tradition. Both as musician and biblical scholar, I have followed recent developments in archaeomusicology for some years now, and am amazed by the ingenious breakthroughs in this field in recent decades. I think it is a stroke of genius on the part of Professor Dumbrill to hold this conference on ‘The Masoretic te‘amim and other musical notations of the Ancient Near East’. So please allow me to offer a general introduction to the Masoretic te‘amim. FROM EGYPT TO JERUSALEM The Book of Exodus relates that the Israelite nation came out of Middle Kingdom Egypt in the mid-second millennium BC, a claim endorsed by later Israelite literature and some Egyptian sources.1 They were led, we are told, by a man called Mosheh who had been an Egyptian prince. Mosheh means, in Egyptian, ‘a son’, but in Hebrew ‘drawing out’. And like their leader, the Israelites were a son drawn out of Egypt. They entered the Levant, surviving until, in the eleventh century, a series of kings unified their clans into an effective military force. The second king, David ben Yishai, took advantage of the hiatus of Egyptian power in the Third Intermediate Period to seize the city of Uru-shalem, Egypt’s power-base in the Levant. Renaming it Yerushalayim, he established it as the cult centre for the worship of Israel’s God. In a sweeping innovation for Israel, he made music and song central to the new cult, establishing a guild of 288 élite singers, in twenty-four wards of twelve men, offering songs of praise morning and evening to render the deity propitious to Israel. And propitious he indeed appeared. For, undefeated in every conflict, Israel expanded from Egypt to Damascus. Ben Yishai and his son Shlomoh built for their city a great and magnificent temple. And within this temple, for a thousand years, with only the brief interruption of the sixth-century Babylonian exile, the guild of temple singers, dedicated only to the ministry of song, maintained their art from generation to generation. Now a nation born in Egypt, whose lawgiver was an Egyptian prince, whose capital was a former Egyptian colony, whose kings married Egyptian princesses, who had Pharaohs for fathers-in-law, and maintained military, diplomatic, and commercial links with Egypt, might be indebted also to Egypt’s musical culture. They certainly shared Egypt’s instruments—harps, flutes, trumpets, drums. And they surely knew the diatonic scales of New Kingdom Egypt, the harmonic minor and Persian scales, and probably the Mesopotamian modes as well. But, more importantly, they knew music as something led by cheironomers. Cheironomy was a central feature of Egyptian musical practice. Here are some depictions of it in bas reliefs of the Old Kingdom period. And, as you can see from the pictures, the cheironomers employ both hands to lead their team of musicians. 1 For a short but scholarly defence of the Exodus story, see M. Warker (ed.), Ancient Israel in Egypt and the Exodus (Washington, DC: Biblical Archaeology Society, 2012). The 3rd century BC Egyptian scribe Manetho tells how the Hyksos were led out of Egypt into the Levant by a man called Osarsiph and Mosheh. Josephus identifies the Hyksos, who ruled Egypt from the 18th–16th centuries BC, with the early Israelites. Figure 1. From a 3rd-millennium-BC mastaba. Three cheironomers signal to two harpists and a flutist. Figure 2. Egyptian cheironomers and musicians (2700 BC). Bas-relief from the tomb of Nencheftkai. Figure 3. Egyptian cheironomer (3rd from left) covering left ear. Figure 4. Top row: cheironomers and musicians (Tomb of Khnumhotep, c.2400 BC). Such two-handed cheironomy was the central feature of Egyptian music from at least the early third millennium BC. Hans Hickmann, of course, spent his life studying it, defining its symbols and identifying them with different steps of the scale. And, while views diverge on his claim that they show ancient Egyptian heterophony, there is a general consensus that different cheironomical signs did represent different musical notes. It appears that the Israelites, like the Egyptians, employed cheironomy in musical performance. They left no painted friezes, for they had strict attitudes to images. But they left writings which preserve details of their cultus. The Books of Chronicles describe its origins as follows. These are they whom David appointed over the hands of song (‘al y’dei shir) of the house of the LORD after the resting of the ark there. They ministered before the tabernacle of the tent of meeting with song, until Shlomoh built the house of the LORD in Jerusalem. (1 Chr. 6.31–32) This phrase, ‘hands of song’, is quite unique in the Bible. And while the phrase ‘al y’dei (by/over the hands) may refer to the transmission of a tradition, I would suggest that, as used here, it refers to the cheironomy which the Israelites saw and learned in Egypt, as will become clear later on. The Israelite musical director—in Hebrew the mnatseaḥ—would have been the one who made the cheironomic gestures. By these gestures he led his team of singer-musicians within the temple. A century and a half later, when the infant Joash was proclaimed king (c. 828 BC), we read of temple singing ‘al y’dei david, that is, ‘according to the hands of David’. Yehoyada appointed…the Levite kohanim to offer the burnt offerings of YHVH, as written in the Torah of Moshe, with rejoicing and singing according to the hands of David (‘al y’dei david). (2 Chr. 23.18) Three hundred years later, after the Babylonians had destroyed the temple, taken the people captive, and sacrifice and worship had ceased for seventy years, the Israelites returned to Jerusalem, rebuilt the temple, and resumed the cult. Then the Levites offered praise ‘according to the hands of David’. And when the builders laid the foundation of the temple of YHVH, they set the priests in their apparel with trumpets, and the Levites the sons of Asaf with cymbals, to praise YHVH, according to the hands of David (‘al y’dei david) king of Israel. And they sang antiphonally in praising and giving thanks to YHVH, for he is good, for his loyal love endures for ever toward Israel. (Ezra 3.10–11) Now, the fact that these ‘hands of David’ survived the cataclysm of the temple’s destruction, the dissolution of the Levite chorus, and the Babylonian exile, makes me suppose that we are dealing with a written cheironomic tradition attached to the sung text. Israel was a highly literate civilization. They kept records of everything. In the temple they kept their sacred scriptures, their genealogical records, the temple financial records. They kept records of their kings, of their history, and of their treaties with other nations. There are even records of how much cheap wine was issued to the builders labouring in tenth-century Jerusalem. They knew that a written record was a thousand times better than any oral tradition. And, if the Levite singers prized their musical tradition—which I cannot doubt—then they would certainly have found a way to record their cheironomic gestures, so as to prevent the loss or corruption of their tradition. In fact, I imagine they would have done this sooner rather than later. But, for their songs to survive the Babylonian exile uncorrupted, some such system must have been in place before the sixth century BC. Such a view is confirmed by rabbinic traditions which speak of cheironomic notation being added to biblical texts when they were written, in the time of Moses and Solomon. 2 Indeed, when Deuteronomy says that Moses was instructed to ‘write down this song’ (31.19–22), it probably implies more than just the song-text. So I suggest it was some such form of notation that enabled the Israelites to return from the Babylonian exile with their temple songs intact to resume worship in the second temple. A few hundred years later, in the Alexandrian period, forms of musical notation were appearing all round the Mediterranean. Here is a fragment of Euripides’ Iphigenie in Aulis, dating from the mid-third century BC, with signs above the text to direct the singers in their performance. Figure 5. 3rd century BC fragment of Euripides with cantillational symbols3 If the Greeks had such notation at this period, then surely any notion that the Israelites did not have a similar system by the same time is unlikely. Just as the kohanim kept master copies of the scriptures in specially-designated safe rooms in the temple, so the Levites, generation after generation, must have kept cheironomically-annotated scrolls, probably in the singers’ chambers, beneath the upper court of the temple. Then came the Roman siege of Jerusalem in 70 AD. Foreseeing the impending cataclysm, the temple’s scroll-keepers hid their precious treasures. Some were hidden in the desert of Judah. Some of these became, last century, our Dead Sea Scrolls. 4 But most of the Zadokite kohanim and the Levite singers perished by fire and sword in the conflagration that engulfed the city and temple. And, to all appearances, the singers’ cheironomical traditions perished too. Yet the songs of the temple were not forgotten. They were orally transmitted throughout the first millennium. The church of the east, as we shall see, took over some of the psalm-singing traditions of the temple. Judaism too preserved some memory of the temple song, especially in the grand and solemn worship of the Babylonian Jews, who, even in medieval times, prided 2 B. Erub. 21b. The picture and transcription of Leiden Papyrus Inv. 510 are from D. Jourdan-Hemmerdinger, ‘Un nouveau papyrus musical d’Euripide’, in Comptes rendus des séances de l’Académie des Inscriptions et Belles-lettres (Paris: Klinsksieck, 1973): 292–302. 4 In contrast to the older Qumran-Essene theory of Dead Sea Scrolls origin, there is a growing consensus that these documents came from the temple library in Jerusalem and perhaps elsewhere (K.H. Rengstorf; N. Golb; Y. Hirschfeld; Y. Magen; Y. Peleg). The variety of thought and handwriting among the scrolls makes it unlikely that they are from a single source, while the Qumran ‘Essene’ establishment was actually a Hasmonean fort. 3 themselves on preserving the temple song.5 Some form of cheironomy also survived. Within two generations of the temple’s destruction, Rabbi Akiva speaks of one-handed cheironomy being practised in the Holy Land when reading Torah scrolls. 6 Medieval records note the same thing.7 But, more importantly, from the sixth to ninth centuries, Bible manuscripts containing simple cantillational systems with a handful of symbols begin to appear. There are two such systems, called the Babylonian and Palestinian. Both feature supralinear symbols, like the Euripides manuscript, to guide the chanting. Both systems are fragmentary, not only in survival, but in conception. The symbols are rudimentary and the texts are inconsistent in their use of them. Perhaps, having lost the ancient cheironomical notation, the rabbis were devising their own systems. Suddenly, in the mid-ninth century, something new appeared: Bible manuscripts whose Hebrew texts were surrounded by an astonishingly sophisticated system of symbols indicating how the text should be pronounced and cantillated. The first of these was the Codex now called British Museum Orientales 4445. And it was followed by a series of other codices using the same system: Moshe ben Asher’s Codex of the Prophets in the late ninth century; the Codex Babylonicus Petropolitanus in the early 10th century, the great Aleppo Codex a few years later, and others. Unlike the probative attempts of the Babylonian and Palestinian accent systems, and unlike the 1,000 year evolution of western musical notation from neumes, this new Masoretic system appeared fully-formed and was employed with remarkable consistency throughout the Masoretic manuscripts. These manuscripts had a common origin. Rather than being products of Rabbanite Judaism, they came from the Karaite Jewish community. The Karaites—their name means Scripturalists—were a Jewish group who rejected the rabbinic authority of the Talmud and were guided by the Bible alone. They had little in common with the Rabbanites. The The Ben Asher family of Masoretes two communities lived separate lives in Rabbi Asher, ‘the great Shaikh’ (fl. c. 770) separate lands: the Rabbanites mostly in His son, Nehemia ben Asher (fl. c. 800) Babylon, the Karaites mostly in the Holy His son, Moshe ben Asher (fl. c. 835) Land. And most of these manuscripts were His son, Asher ben Asher (fl. c. 865) written or supervised by a prominent Karaite His son, Moshe ben Asher (fl. 895), family, the Ben Asher family, who lived in produced the Codex Cairensis Tveriya or Tiberias in Galilee.8 They came to His son, Aharon ben Asher (fl. 930), be known as the Ba‘alei Massorah, or produced the Aleppo Codex ‘Masters of the Transmission’. In English, we call them Masoretes. 5 Nathan ben Isaac ha-Bavli, Aḥbar Baghdad (10th century; Neubauer 1895: 83); Petaḥiah of Regensburg (11th century; Grünhut 1905: 24–5); Rashi (11th century) on B. Ber. 62a; Benjamin of Tudela (12th century; Adler 1907: 39 [ET], §60 [Hebrew]). 6 B. Ber. 62a. 7 Rashi on B. Ber. 62a; the Parma Psalms ms shows a cantor leading his singers with one-handed cheironomy. 8 Tiberias was the centre of Jewish life in the Holy Land from the time of the Bar Kokhba revolt, when Jews were expelled from Jerusalem, to the time of the Islamic conquest, when they were allowed to return to the Holy City. However, even after Jews resettled in Jerusalem, Tiberias remained important. THE ORIGINS OF THE MASORETIC TE‘AMIM The Masoretes, as I said, began to publish their te‘amim or cantillation marks in the ninth century. But where did their system come from? This is a vital question, for any interest that they may have for us is inextricably linked to their origin. Let us examine it closely. For some centuries now, the prevailing view in biblical scholarship has been that the Masoretes invented these symbols. Such a view covers a spectrum of attitudes. At one end, some think the Masoretes just invented these symbols to record their view of how the Bible should be read and chanted. Others, on the other hand, think they invented the symbols in order to record ancient oral traditions which they had received. But, one way or another, it is widely held that the Masoretes invented this system of notation. Yet closer inquiry reveals difficulties with this theory. Here are twelve of them. First, the Masoretes testify in their own words that they did not invent the te‘amim but received them. The late-ninth-century Masorete, Moshe ben Asher, wrote the ‘Song of the Vine’, which was kept with his Codex of the Prophets in the Ben Ezra synagogue in Cairo. In it, he relates how the Masoretes received the accents from a group called the Elders of Bathyra: The vine of God [is] the tribes of Jacob…The branches of the vine are the prophets…The perfect ones of the vine are the elders of Bathyra, the heirs of the prophets, who possess knowledge of understanding. Deep waters that utter mysteries; their heart brings forth wisdom like a flowing brook. As delights they established the te‘amim of the Scripture [hitqinu ta‘amei miqra], which give sense [to the Scripture] and interpret its word… Afflictions surrounded them from the kings of the Greeks, and they exiled them and dispersed them to No [Egypt] and its provinces.9 Now the elders of Bathyra are a group known to us from the Talmud as a prominent Israelite priestly clan of temple times. They lived in their own towns, called Bathyra, both in Babylon and the Holy Land. Ben Asher’s comment that they suffered under Greek rulers suggests that they suffered under the rule of Antiochus IV in 165 BC, who made Israel’s kohanim a particular target. Later, in the first century BC, the men of Bathyra held the presidency of the Sanhedrin, until they were ousted by Hillel and the Pharisaic party. Some of them survived the destruction of Jerusalem in 70 AD, and the family were influential into mishnaic times. These are the elders of Bathyra to whom Ben Asher credits the te‘amim. Moshe ben Asher’s son, Aharon, the author of the Aleppo Codex, says something similar in his Diqduqei ha-te‘amim, where he writes: The te‘amim [were received]…from scholars and scribes (soferim) established by the elders of the semi-circle [the Sanhedrin], the prophets and princes of the Exile…received from the prophets and sages who were sealed with the seal of the prophets of God. Here the younger Ben Asher says that the te‘amim were established by the Sanhedrin and passed on by a continuous chain of tradition. The Masoretes received them from scholars and scribes, who received from the Sanhedrin who established them, who in turn received them from earlier prophets and sages, sealed with the prophetic seal of God. The statement that the te‘amim were received from scribes suggests that the te‘amim were passed on as a written tradition. This accords with the statement of Ben Asher père about the Elders of Bathyra, the presidents of the Sanhedrin in the 2nd century BC. 9 Kahle, Cairo Geniza (1959), p. 84–85. Now these are surely remarkable testimonies. Here are the two foremost Masoretes, the patriarchs of their foremost clan, testifying that they received the te‘amim as a scribal tradition from a source predating the Sanhedrin. And their testimony is widely ignored. Second, those who think the Masoretes invented this system are at a complete loss to explain its sudden appearance. As one of them says, ‘The Tiberian system is a highly perfected system, but no previous stages of its graphemic development are known to us.’10 Now I think it is fair to say that highly-perfected systems do not appear overnight. It takes centuries for any system of writing or notation to achieve precision. The neumes of 8th-century Byzantium took a millennium to evolve into modern music notation. But the sudden appearance of the Masoretic system is explained by Ben Asher’s claim that they received the te‘amim from an ancient source? Third, if the Masoretes had invented this entirely new gamut of symbols and added it to the Bible text, they would have done so in direct defiance of their own epithet, Ba‘alei massorah, ‘The Masters of the Transmission’. They would have done so in defiance of their own guiding philosophy of ‘hiding nothing and adding nothing to what was received’. Paul Kahle, who led the study of the Masoretic texts of the Cairo Geniza, says, ‘What we know of their activity shows an obstinate clinging to the smallest detail of what was transmitted to them’.11 Or as Sanders says, their goal ‘was not to innovate, but to preserve the finest textual traditions that they knew.’12 Their own codices pronounce terrible curses against anyone who should tamper with or add to the text within. 13 Surely, then, the Masoretes, when adding the te‘amim to the Bible text, must have believed that these symbols possessed a supreme authority to merit a place within the scriptural text. Fourth, the Masoretes’ claim to have received the te‘amim is supported by their apparent inability to explain their own work. Two Masoretic works describing the te‘amim have come to us. One is the 10th-century Diqduqei ha-te‘amim of Aharon ben Asher and the other is the Arabic Hidāyat al-qāri (Guide to the Reader) by the 11th-century Karaite, Abu al-Faraj Harun of Jerusalem. 14 If the Masoretes had invented the system, one would expect that these books would provide a systematic explanation of how the signs should be sung. But no such thing is found. They sometimes speak of hand-signs for certain symbols, or of a rising or falling voice. But their principle concern is the notation and pronunciation of the vowels. Elsewhere their work is, as Graetz says, ‘for the most part incomprehensible’. 15 Fifth, the detail, sophistication, and consistency of the Masoretic system shows that it was the work of experts in this system. Such expertise must have arisen from a group who were concerned with this musical notation on a regular basis. But these could not have been the Masoretes themselves. They were scribes, not singers. Nor is there any evidence of a highlydeveloped musical culture in 10th-century Tiberias. Sixth, the work of the Karaite Masoretes was not unknown to their Rabbanite contemporaries. On the contrary, the Rabbanites knew it and opposed it. When the first 10 A. Dotan, ‘The Relative Chronology of Hebrew Vocalization and Accentuation’, Proceedings of the American Academy for Jewish Research 48 (1981) 87–99 (p. 87). 11 P. Kahle, Schweich Lectures (1947), p. 110. 12 P. Sanders, ‘The Ashkar-Gilson Manuscript: Remnant of a Proto-Masoretic Model Scroll of the Torah’, Journal of Hebrew Studies 14 (2014), article 7 <DOI:10.5508/jhs.2014.v14.a7>: pp. 1–22 (22). 13 P. Kahle, Schweich Lectures (1947). 14 Eldar, 1992: 67–73; 1994: 40-43. 15 H. Graetz, History of the Jews. 6 vols. (trans. P. Bloch) (Philadelphia: The Jewish Publication Society of America, 1902) from Geschichte der Juden (1853–70): III.207. anonymous Masoretic Codex, Codex Orientales 4445, appeared in the mid-ninth century, it elicited a rebuke from the religious head of the large Babylonian Rabbanite community, Mar Natronai ben Ḥilai, Gaon of Sura academy from 853 to 861. According to Mar Natronai, the te‘amim were an ancient tradition which must not be divulged. We are forbidden deliberately to add [to the text] anything at all for fear of transgressing the Law, You shall not add anything to it [Deut. 12.32]. This is why we do not vocalize the scrolls of Torah. Neither may we add the te‘amim, even though these latter signs were revealed at Sinai. And when Aharon ben Asher produced his great codex in the following century, he, and indeed the whole Karaite movement, was fiercely opposed by Natronai’s successor, Sa‘adya Gaon. Now these Rabbanites did not object to the Masoretes’ work because they thought the te‘amim a novelty. On the contrary, they thought them ancient but sealed. They evidently held this view sincerely, as a tradition of their own. For, if they had just wanted to discredit the Karaites, they would simply have denied the antiquity of the te‘amim and made the Masoretes out to be falsifiers of the Bible text. But this they never did. Seventh, their rabbinic predecessors, the sages of mishnaic times, some seven hundred years before, also speak of the te‘amim as ancient and authoritative. The Talmud records their views that the te‘amim originated with the writers of the Bible books, such as Moses and Solomon, and that they were systematized by Ezra.16 They speak of cheironomic gestures and cantillational symbols associated with the Torah. They affirm that these things sprang from an ancient and authoritative source.17 Yet, like Natronai and Sa‘adya, they apparently did not have direct access to these symbols. 18 Eighth, the te‘amim influence the very meaning of the sacred scripture. Within Judaism, such an intervention in the text would only be tolerated on adequate authority. And yet, after the initial wave of opposition from the Babylonian Rabbanites, the Masoretic texts were soon accepted throughout Judaism. The great Rabbi Moshe ben Maimon (Maimonides) pronounced Ben Asher’s Aleppo Codex to be the most perfect of all Bible texts, and it was venerated throughout the Jewish world.19 Ninth, cantillation marks per se are of great antiquity. They are found in Mesopotamian texts of the 2nd millennium BC, in Greek poetic texts of the 3rd-century BC, and in Greek hymns from the early Christian period. There can therefore be no objection to an ancient origin for the te‘amim on the basis of the history of musical notation. One must ask, if the Greeks had such a system hundreds of years before Christian times, why the Jews had not? Were they backward or uneducated? Did they not think their scriptures were ancient and sacred? Could they not see the benefits of such a notation, or of the risks posed by the lack of it? In fact, one is struck by just how closely the Masoretic te‘amim resemble the symbols of 1,000 or more years before their own time in the Greek texts and the Dead Sea Scrolls.20 If the Masoretes were introducing 16 B. Meg. 3a; B. Ned. 37b; Y. Meg. 4.1; Gen. R. 36.8, ascribe the te‘amim to Ezra, after Neh. 8.8. B. Ned. 37b; B. Erub. 21b ascribe to Moses and Solomon the cantillation of their own books. 17 Yesh em l’miqra (B. Sanh. 4a): There is [an ancient and authoritative] source for the cantillation [of scripture]. 18 They speak of the fifteen puncta extraordinaria, which appeared in every synagogue scroll both then and now, as something seen, but they speak of the te‘amim as something only known about. 19 Mishneh Torah, Hilkhot Sefer Torah 8.4. In 1973, M.D. Cassuto wrote that Maimonides did not in fact work from Ben Asher’s codex. But Ofer shows that Cassuto’s view was based on a distorted text of Mishneh Torah (Y. Ofer, ‘The History and Authority of the Aleppo Codex’, in M. Glatzer, Jerusalem Crown Bible Companion Volume. Jerusalem: Ben-Zvi, 2002: 25–50; cf. 37–38). 20 The Euripides’ Iphigénie in Aulis fragment (1973: 292–302) has cheironomical symbols resembling Masoretic munaḥ and silluq. Liebermann (1950: 38-46) and Talmon (1969 i-xxvii) also note the similarity of Masoretic their own cantillational notation, why did they not use neumes, which were current in the Levant in their own time, and which Jewish communities as far away as Rome were using soon after.21 To employ the system of a millennium earlier was as anachronistic as for us to annotate a modern Bible in runes. Tenth, there is evidence of a history of musical development within the biblical te‘amim themselves. For instance, the cantillation of Psalm 136 is different from that of its neighbour, Psalm 137. Psalm 136 is almost devoid of supralinear symbols; Psalm 137 is replete with them. The cadences of Psalm 136 are invariably etnaḥ and silluq in each line. But Psalm 137 cadences not only on etnaḥ and silluq, but also on revia and tifḥa and zinnor and oleh ve-yored. And Psalm 137 features sign-pairs—revia mugrash and oleh ve-yored—totally absent from Psalm 136. In fact, a glance at the cantillation shows that they are as different as night and day: Psalm 136 is stark, repetitive, monolithic; while Psalm 137 is fluid, rhapsodic, and unconstrained. Why should this be? The answer surely lies in their dates of origin. Psalm 136 sings of the Exodus, the crossing of the sea, and the entry to the Promised Land. It is silent on themes like Jerusalem, the temple, or the house of David. It appears to be pre-monarchic, from the second millennium BC. Psalm 137, by contrast, speaks of the Exile in Babylon, and must date from after the mid-sixth century BC. Since the two psalms date from 500 hundred or more years apart, we may suspect that their different cantillation reflects the singing of two different periods of history. Other psalms confirm the same thing.22 Such evidence of historical musical development within the te‘amim themselves is hardly consistent with the view that the Masoretes invented this system and applied it to the ancient psalms. Eleventh, there is evidence that the other set of accents, the nekudot or vowel-points, were also received, not invented. For instance, the vowel points indicate different pronunciations of the same word in different periods. The Hebrew letters for a carbuncle stone—Bet Resh Kaf Tav—are given the vowels bareket in the book of Exodus, but, in the book of Ezekiel, written hundreds of years later, they are vocalized as barkat. It would be an exceptional oral tradition indeed that would record such historical differences in pronunciation. Twelfth, the Masoretes frequently remark upon inconsistencies in the accentuation of the text.23 But if they had invented them themselves, they would simply have put the correct accents to the words and left it at that. It defies all reason to think that they invented the vowel signs, then deliberately put the wrong signs in the text, then provided a marginal corrigendum to correct their own intentional error. All in all then, the notion that the Masoretes invented the te‘amim is hard to maintain. Even if they did receive ancient oral traditions of how the Scriptures were chanted and only invented symbols to represent it, that would hardly accord with their claim that they received a scribal tradition originating from before the time of the Sanhedrin, nor would it explain many of the other dissonances outlined above. It seems to me that many of these facts can only be explained accents to Greek accents of the Alexandrian period. E. Werner, ‘Musical Aspects of the Dead Sea Scrolls’ in The Musical Quarterly (1957), No. 1, p. 24. 21 The Parma MS shows use of neumes in the 12th century Jewish community in Rome. 22 See, for instance, Psalm 114, which refers to the Exodus, but also to the existence of a sanctuary in Judah, and is perhaps from the early monarchic period. It shares the simple etnaḥ-silluq cadence-scheme of Ps. 136, but also features the sign-pair revia mugrash, which is quite absent from Ps. 136. See too, Ps. 132, whose early section, apparently a song from King David’s time, is marked by the etnaḥ-silluq cadence-scheme, while the elaborate Solomonic-period flourishes in vv. 11–12 feature complex tetrapartite verses, with revia mugrash. 23 In 1 Sam. 5.9; 6.4, the shva of the qere ‘techorim’ inserted into the ketiv, creates the impossible form efolim, resulting in ayin vocalized with a vocal shva. In Jer. 42.6, the word aleph-nun-tav vocalized with the vowels of the qere ‘anachnu’ results in the impossible combination of a shuruk and a vocal shva in a single tav. by the fact that the Masoretes were reproducing a written tradition which they had received from antiquity. WHERE DID THE TE‘AMIM COME FROM? So, if the Masoretes did not invent the te‘amim, where did they get them? Well there are, I imagine, two options. Either they always had them, or else they obtained them. If we favour the first opinion, then the explanation may hang on the person of the first real Karaite leader, the prince Anan ben David. A descendant of the royal house of King David, Anan was a candidate for the Babylonian Exilarchy in 761 AD. But, since he rejected the Mishnah and Talmud, his appointment was opposed by the geonim of Sura and Pumbedita, the heads of Babylonian Judaism. And so, gathering others of anti-rabbinic sentiment, he founded the Karaite movement. Now it seems likely from his writings that ‘Anan was in possession of scrolls dating from temple times. 24 And these may have included cantillated scrolls which his successors published the following century. The second option, that the Karaites found a cache of ancient biblical scrolls, was suggested by the great German scholar of the Cairo Geniza, Paul Kahle. Speaking of Moshe ben Asher’s claim that the te‘amim are of ancient origin, Kahle drew attention to the letter of the Nestorian Patriarch Timotheus of Jerusalem which tells of the discovery of a great cache of ancient biblical scrolls in a cave near Jericho around the year 800. Kahle says, ‘We hear that the Jews from Jerusalem who had been informed collected the manuscripts and studied them. The Jews chiefly interested in these manuscripts seem to have been Karaites, and we have to assume that the Karaites were considerably influenced by these manuscripts.’ (1953: 82; 1959: 16.) Either way, I suggest that, whether transmitted or discovered, the Masoretes possessed ancient written scrolls of the temple cantillation. They recognized these accents as both ancient and authoritative, and combining them with the best scribal practices of their own day, began to reproduce them in codex form from the ninth century on. But, if the musical notation of the te‘amim existed before the Babylonian exile, it is worth pausing for a moment to ask why they were so long unseen. Why are the te‘amim not more fully represented in, say, the Dead Sea Scrolls? Why do the talmudic rabbis speak of them as something known of but unseen? How can we account for this? The answer, perhaps, is that religious establishments protect their interests by laying claim to powers and privileges unattainable by the average man. They surround these claims with ceremonies of initiation that demand the acolyte’s subservience to the hierarchy. Since this still exists in our egalitarian age, how much more so in ancient times? Curt Sachs tells how a form of musical notation was kept in great secrecy among the initiated in Babylon; and likewise, later, among the singers of the Abyssinian Church.25 One imagines that Israel’s temple singers would have been equally secretive about their craft. It was not for public viewing. It would have been kept on a strictly need-to-know basis among the temple’s Levite chiefs. Yet such precious scrolls could have come to a prince like Ana ben David. And the arrival of such secret scrolls in Karaite hands 24 His Sefer ha-Mitzvot (Book of Precepts) of 770 reflects the opinions of other older forms of anti-rabbinic Judaism. These views may have come from old Sadducean and Essenic writings in his possession. 25 C. Sachs, The Rise of Music in the Ancient World: East and West, 1st ed. (W.W. Norton & Co. Inc., 1943) pp. 85-86). would fully explain the sudden publication of what was hitherto known of, but unseen, just as it would fully explain the Rabbanite reaction to the Masoretic codices. CHARACTERISTICS OF THE TE‘AMIM Now, turning from the origins of the te‘amim, let us look more closely at the Masoretic punctillation. First, the punctillation includes ten signs, including a mobile dot, indicating the vowels and inflections of the Hebrew consonants. These are called the nekudot or pointings. Here they are. And here is the psalm-heading ‘A Song of Ascents’ written with the vowel signs. In addition, there are some twenty cantillational symbols or te‘amim indicating how the text should be cantillated or sung. Here they are. And here is the same psalm-heading with its cantillational signs or te‘amim. And in the Bible texts both sets of signs are combined to produce an accented text of great complexity and precision, as seen again in the same psalm-heading. SUBLINEAR & SUPRALINEAR SYMBOLS Now the te‘amim symbols occur in two separate rows, one above and one below the text. Figure 6. Supralinear symbols Figure 7. Sublinear symbols Each symbol usually appears in only one position, either above or below the text. But four of them may appear either above or below, in which cases they have different names. Figure 8. Sub- and supralinear symbols And some appear in combination, either as doubles, or with other symbols, these compound signs again having their own names. Figure 9. Compound symbols or sign-pairs Now two important observations emerge from these two lines of symbols above and below the text. First, the Masoretic system differs completely from its predecessors in this matter. The Babylonian and Palestinian systems had supralinear symbols only. And what is interesting in this regard is that the ancient form of Egyptian and Israelite cheironomy was two-handed, but, as we saw, one-handed cheironomy became the norm in rabbinic times. And it seems likely that the bilinear Masoretic system reflects the ancient two-handed cheironomy, whereas the Babylonian and Palestinian systems reflect rabbinic one-handed cheironomy. Second, the sublinear symbols are more important than the supralinear ones. For some texts continue for many verses with sublinear symbols alone. Psalm 136, for instance, features only two supralinear symbols in 26 verses of text. But not one verse in the Bible has only supralinear symbols. From this, one may conclude that it was possible to notate some form of musical performance with sublinear symbols alone, and that the sublinear symbols are therefore the prime determinants of the notation. TWO DISTINCT SYSTEMS: POETIC & PROSODIC The next thing we need to say about the te‘amim is that they are not one system but two. There is one system for poetic texts another for prose texts. The two systems share some symbols, but each has unique symbols of its own. The poetic system is found in the three books which consist principally of poetry; that is, Psalms, Proverbs, and Job. While the prosodic system is found in the other twenty-one books of the Hebrew Bible which consist principally of prose. The Book of Job is a unique hybrid: its prologue and epilogue are written in the prosodic system while its poetic core is in the poetic system. The two systems, poetic and prosodic, share twelve symbols in common. Six are sublinear and six are supralinear. Sublinear te‘amim common to all Bible books ‫ֽא‬ ‫֥א‬ ‫֖א‬ ‫֑א‬ ‫֣א‬ ‫֤א‬ Silluq Merkha Ṭifḥa Etnaḥta Munaḥ Mahpakh Supralinear te‘amim common to all Bible books ֙‫א‬ ‫֜א‬ ‫֘א‬ ‫֡א‬ ‫֗א‬ ‫֓א‬ Pashta Geresh Zarqa Pazer Revia Shalshelet Meanwhile some fifteen symbols and symbol-combinations are unique to each system. Five of them are unique to the poetic books and ten to the prosodic books. Te‘amim only in the poetic books ‫̧א‬ ‫֪א‬ ‫֬א‬ ‫ד֫ ֥בר‬ ‫֝ד ֗בר‬ Galgal Etnaḥ hafukh Illuy Oleh ve-yored Revia mugrash Te‘amim only in the prosodic books ‫֧א‬ ‫֛א‬ Darga Tevir Ladder (Scale) Turn ‫֔א‬ ‫֕א‬ Zaqef qatan Little upright Zaqef gadol Great upright ‫֒א‬ ‫֞א‬ Segolta Gershayim Cluster Expulsions ‫֠א‬ ‫֩א‬ ‫֟א‬ ‫֦א‬ Telisha gedolah Great pulling away Telisha qetanah Little pulling away Qarney parah Horns of cow Merkha kefula Double merkha So this is a sophisticated and detailed system. And it is applied with rigorous consistency. There are standard patterns and, dare one say, laws, that one meets over and over again. HOW MANY TE‘AMIM ARE THERE Finally, I have referred several times to “twenty or so” te‘amim because opinions vary on their number. In particular, there are three symbols—galgal, yeraḥ ben yomo, and etnaḥ hafukh— which were confused in later Masoretic codices, those from after AD 1,000, and even more so in subsequent printed editions. As a result, one sign, the first sign here, was called galgal in the poetic books and yeraḥ ben yomo in the prosodic books, while the third circular sign disappeared. It was not until the last few decades, when first-millennium Masoretic texts, like the Aleppo Codex, became more accessible, that it was realized that there were actually three different symbols. There was the half-moon symbol, yeraḥ ben yomo, in the prosodic books. Meanwhile in the poetic books, there had been two different symbols which had both merged to a V-shaped symbol called galgal. This was first noted by Professor Yeivin, who identified the three-quarter circle here as the true galgal, and gave the name etnaḥ hafukh to the V-shaped symbol, which resembles an upside-down etnaḥ.26 26 The Unicode system recognizes two out of these three symbols, distinguishing between etnaḥ hafukh and yeraḥ ben yomo, but does not yet have the three-quarter circle galgal symbol. RECREATING THE MUSIC OF THE TE‘AMIM If we want to recreate the music of the te‘amim, then there are several avenues of approach. (1) First, there is the inductive approach, proceeding from the symbols themselves and their likely musical meaning in context, just as we would crack a linguistic code from knowing the basic principles of language. (2) Second, there are surviving fragments of temple psalmody in church and synagogue chant that allow us to make comparisons. As we shall see, both these traditions have preserved at least one temple melody through two millennia in the tonus peregrinus. Yet both traditions also exhibit much diversity and none possesses a pure record of the temple song. 27 Yet, there is gold in the rivers, for those who will seek it. (3) Third, the phenomenon of parallel passages in the poetic and prosodic books may hold promise of understanding both better. (4) Fourth, we may gather insights from secondary sources, such as the Masoretic commentaries on the te‘amim, and from related notational systems, like the systems of the Dead Sea Scrolls, the Babylonian and Palestinian traditions, the Byzantine kontakia and other ancient Christian traditions. For now, I would like to show you an example of how we may proceed by combining the first two avenues; that is, the inductive approach and synagogue tradition. Let’s begin with the inductive approach. I noted earlier that some texts feature no supralinear cantillation marks, but only sublinear ones. From this, we recognized that some form of music can be notated with the sublinear signs alone. And so we concluded that they are determinative. Looking further at the sublinear signs, we find that there are eight in each system. Eight, as we all know, is a significant number in music: it suggests the notes of a scale over one octave. And since the idea of musical scales, and symbols for each step of the scale, was well established long before Israel’s monarchy period, we may surmise that these lower symbols represent notes of the scale. Our next step requires us to look at the notation. Here are the first lines of two of the oldest psalms, Psalms 114 and 136. The English reads from left to right; the Hebrew, of course, from right to left. ‫ממצרים‬ ֑ ‫בצאת י֭ שראל‬ ֣ Psalm 114.1 In-going-forth of-Israel from-Egypt ‫לﬠז‬ ֽ ‫מﬠם‬ ֥ ‫בי֥ ת יעקוב‬ House-of-Jacob from-a-people stammering. 27 During the first millennium after the temple’s destruction, some Jewish communities claimed to preserve the authentic temple song. But, after a millennium of improvisations by successive generations of cantors, it had already diverged into many streams in the lands of the diaspora. When, by the thirteenth century, hand-written copies of the Masoretic texts began to circulate throughout Judaism, each community interpreted the te‘amim in terms of its own existing cantoral traditions, inventing divergent interpretations of each sign which they then reimposed upon their divergent melodies. ‫הודו ליהו֣ה כי־ט֑וב‬ ֣ Psalm 136.1 Give-thanks to-the-LORD for-he-is-good; ‫חסדו‬ ֽ ‫֭כי לעול֣ם‬ for until-forever his-loyal-love. What we can see is that the perpendicular bar silluq occurs at the end of every verse. In fact, it comes at the end of every verse in the Bible. Now, the fundamental psycho-acoustic fact of music is the home note or tonic. Music revolves around the relation between the tonic and the departures from it. And we find that, in all surviving forms of ancient church and synagogue chant, every verse closes on the tonic. 28 So we can, with some confidence, identify silluq as the tonic. Can we perhaps proceed to identify another note of the scale in the same way? The second most important note of the scale, I think you will agree, is the fifth step, the dominant. If the tonic is the home note, the dominant is the bold departure. If the tonic is the grey-bearded patriarch sitting on his throne, then the dominant is the handsome prince going forth to win his kingdom. The dominant is the sound of the trumpet. Blow a horn and your first notes are tonic and dominant. Yes, the dominant may not venture as far as the sixth or seventh or octave. But they are so high that they lie in the realms beyond. For confident boldness, for heroic strength, the dominant is our man. So is there a sign here that might represent the dominant? Look at the beginning of Psalm 114. The words are strong words. B’tset yisroel mi-mitsrayim – At the going-out of Israel from Egypt. That looks like a good candidate for the dominant. B’tset – going out, the departure. Now Psalm 136. Hodu l-adonai ki tov. Praise the Lord, for he is good. That seems like a strong declaration, and another good candidate for the dominant. Now in each case the first stressed syllable has the backward-L-shaped sign munaḥ. Could it be that munaḥ represents the fifth? What else might it represent? The minor second sounds sad. The major second sounds apologetic. The minor third sounds sad again. The major third and fourth sound a bit better, but not as bold as the dominant. The major sixth sounds too exceptional, and the minor sixth too sad and emotional. The seventh sounds weird. No note suits these words and this sign half as well as our trumpeting friend, the dominant. Having identified likely candidates for the tonic and dominant, the next thing to note is that the musical system of the te‘amim is structured around cadences. Cadences, as you know, are the pauses, half-pauses, and stops that mark the structure of a musical phrase. And, when we look at the words of the Psalms, we find that semantic pauses, the pauses in the sense of the text, where musical pauses would also occur, are always associated with the same sublinear signs. The most common by far is this sign etnaḥ or atnaḥ. We notice it in the two psalms we cited at the pause in the middle of the verse. Ps. 114.1. 28 In the going-forth of Israel from Egypt (etnaḥ), the house of Jacob from stammering folk (silluq). Some forms of synagogue recitation psalmody finish above or below the tonic, looking forward to the tonic resolution at the beginning of the next verse. But these local and relatively modern forms are absent from the ancient psalmodic traditions. Ps. 136.1. Give thanks to the Lord for he is good (etnaḥ); for until forever is his loyal love (silluq). What note might this cadential sign etnaḥ represent? We have already identified the tonic and the dominant. That leaves us the second, third, fourth, sixth, seventh, and octave. But the sixth, seventh and octave would be sensational and over-reaching, useless for a cadence point. The second would be sad and weary. The only real candidates are the third and fourth. Let’s try it with Psalm 114. If we cadence on the third then the harmony remains within the tonic triad. But the fourth is more adventurous. It is a cadence, a resting point, in new harmonic territory. When the handsome prince, the fifth, has won his kingdom, then he sits down in new territory, on the fourth. So we now have three notes, the first, fifth, and fourth steps of the scale. Let’s look a little further. The other sign in the first half of the verse from Psalm 114 is tifḥa. The options that remain to us are now the second, third, sixth, seventh, and octave. The seventh is impossible, the octave is unlikely. The second and sixth are possible, but less so than the third. For the third keeps the opening phrase in familiar tonic-harmony territory until the cadence on etnaḥ, whereas the second and sixth depart too soon into new harmonic territory. And when we try the same options with the same tifḥa sign in Psalm 136, then the same impression is confirmed. I don’t have time here to go through any more steps. I invite you to experiment with all the options yourselves. But this is how we proceed with an inductive approach to the cantillation. And, using the same method, I would that the other sign in Psalm 114, merkha, on ‘house of Jacob’ is the second note of the scale. But now I’m going to leave the inductive approach, and confirm what we have found from the other source I spoke of, that is from ancient chant traditions. Some years ago, Erich Werner showed that the Gregorian chant melody called the tonus peregrinus, uniquely reserved for Psalm 114, is the same melody as exists in Jewish tradition for the same psalm. Yet, Werner noted, the Jewish melody cannot derive from the Gregorian rite, for it is found not only in the west, but also in the east, far beyond the influence of the Roman church. Werner concluded that the Jewish and Christian forms of the melody must both derive from a single source predating both Church and Judaism, namely, the Jerusalem temple. The tonus peregrinus is therefore our best remaining fragment of temple psalmody. Now, the Sephardi version of the tonus, preserved in the east beyond the influence of the Roman rite, is likely to be closest to the temple original. Here it is. 29 Example 1. Sephardi form of the tonus peregrinus to Psalm 114 What you should notice, if we now compare the Sephardi melody with the inductively-deduced version of the sublinear te‘amim, is that they are extremely close indeed. Bear in mind that while the synagogue version has the same melodic line for every verse, the Masoretic version has little nuances in every line to paint up the words. So, for instance, the first line of the cantillated version has a variant second half. But look at the beginning of the first and third lines. Point by point, the synagogue version and the cantillated version are the same. They both open with the dominant, drop to the third, and cadence on the fourth. And see how the second half of verses two and three open with the fourth and cadence by way of the second on to the tonic. The first five steps of the scale that appeared from the inductive process are confirmed exactly by the ancient chant: silluq represents the tonic, munaḥ the fifth, etnaḥ the fourth, tifḥa the third, and merkha the second steps of the scale. With this insight, I suggest, we have the key to begin to unlock the other signs. 30 29 This version of the tonus is from Morocco. Idelsohn (1929.63) recorded it with Ps. 29 and Werner 1959: 419 has it for the same psalm. The te‘amim of the first verse of Ps. 29 (without heading) are similar to Ps. 114 (as Werner 1959: 420 shows), and so the same tune serves the words of Ps. 114. But while the opening melodies of Pss. 29 and 114 were alike, the tonus does not fit Ps. 29 throughout. The bipartite verses of Ps. 114 fit the bipartite phrases of the tonus like a glove, but Ps. 29’s tripartite and unipartite vv. 3 & 7 fit the tonus only under duress. 30 I confess my debt to the insights of the French musicologist Suzanne Haïk-Vantoura. Her observation that the sublinear te‘amim take precedence over the supralinear clears the way to recognizing the sublinear signs as steps of the scale. And I, by an inductive approach, have reached the same conclusion as she as to which signs represent the first five notes of the scale. Yet a few caveats are in order. First, she spoke of seven sublinear signs for the poetic books, whereas there are now seen to be eight. This was not her fault. She worked from inferior printed editions of the Masoretic text, with no access to the Ben Asher codices, now visible online and in excellent printed editions. But we, now aware of these eight signs, must make sense of them. Second, Haïk-Vantoura was, I believe, CONCLUSION I commend to you the Masoretic te‘amim. I propose that they are not of medieval origin, but that they are the musical notation of the Jerusalem temple singers of the first millennium BC. As such, they are the largest and most perfect system of ancient musical notation which has been preserved to us. Since they represent the cheironomy of ancient Israel, one may suppose that their bilinear notation reflects the two-handedcheironomy that the Israelites learned in Middle Kingdom Egypt. I believe that, with further research, the te‘amim may reveal not only the music of ancient Israel, but also offer insights into the music of ancient Egypt and other surviving systems of ancient musical notation. too confident about her own interpretations of some of the supralinear signs. They may well be ornaments to the sublinear notes, as she suggests. But some of her suggestions seem unlikely, and I have suggested other possibilities on the basis of synagogue traditions.