saurocr from the rncknt mt east

vobune 2, frsckk 2

a hurrian musical score from ugrrit:

 sources from the ancient near east

volume 2, fascicle 2

a hurrian musical score from ugarit:

the discovery of niesopotamian niusic

bY

marcelle duchesne-guillemin

undena publications

malibu 1984
 SOURCES AND MONOGRAPHS ON THE ANCIENT NEAR EAST
Editon: Ciorgio Buccellati, Marilyn KeUy-BucoeIlrti
AdstantEditor: P I t r i c t ~

These two series make available original documents in English translation (Sources) and important studies by modern scholars
(Monographs) as a contribution to the study of the history, religion, literature, art, and archaeology of the Ancient Near East.
Inexpensive and flexible in format, they are meant to serve the specialist by bringing within easy reach basic publications,
often in updated versions, to provide imaginative education outlets for undergraduate and graduate courses, and tc reach
interested segments of the educated lay audience.

SANE 212
a hurrian musical score from ugarit:
F
the discovery of mesopotamian music
by m. duchesne-guillemin

The discovery of a Babylonian musical theory, published by the author in

1963, has repeatedly been confirmed by further discoveries,notably by that of a
Hurrian tablet containing a musical notation. Of this notation, three interpreta-
tions have been offered so far. This is a fourth one.
Nothing was known about Babylonian music, apart from instruments, until a
Babylonian tablet (in The Museum of the University of Pennsylvania) was pub-
lished by Ann Kilmer in 1960 and interpreted by the present author in 1963.
It revealed the existence of a theory of the scale. This discovery created quite
a stir among scholars and made it possible for an asyriologist, 0. Gurney, and a
musicologist, D. Wulstan, to interpret an unknown fragment in the British
Museum which gives a method for passing from one mode to another, thus
proving the existence of seven modes as far back as the 18th century B.C - The
next step was the publication, in 1970, of a Hurrian tablet of the 14th century
B.C. found at Ugarit (Syria) containing a musical score the interpretation of
which is very difficult. Three attempts have been made so far. A fourth one is
offered here with a recording made at Lihge in 1975 by a talented group of
amateur singers specialized in ancient music. This theory is based on the assump
tion that polyphony never existed in the Middle East, and confirmed by a
comparison with traditional Jewish psalm-songs and ancient Syro-Chaldean
Christian liturgical melodies.

@ 1984 bj IJndena Publicetions

ISSN: 0732-6424 ISBN:0-89003-158-4 i

All rights reserved. No part of this publication may be reproduced or transmitted in any 'formor by any means, electronic or
mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing
from the publisher.
 TABLE OF CONTENTS

Preliminary Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1. Thescales ............................ 8
2. Thestrings ............................ 9
3. TheModes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4. The Name of the Instrument . . . . . . . . . . . . . . . . . . . . 12
5. Tuning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
6. TheNotation . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
6.1. Philological Background . . . . . . . . . . . . . . . . . . . . . 14
6.2. First Interpretations . . . . . . . . . . . . . . . . . . . . . . 15
6.3. Criticism . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
6.4. A New Interpretation . . . . . . . . . . . . . . . . . . . . . . 19
7. Comparisons with Jewish and Syrdhaldean Music . . . . . . . . . . . . 21
8. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
A Recording of the Hurrian Song from Ugarit . . . . . . . . . . . . . . .Tape
Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Plates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . after 16

[SANE 2. 651
 LIST OF FIGURES

1. U. 3011, column 1. Sumerian-Akkadian lexical text ( = Nabnitu Tablet 32) . . . . . 26

2. Graphic interpretation of Tablet CBS 10996 . . . . . . . . . . . . . . . . 26
3. Tablet KAR 158 VIII VAT 10101 . . . . . . . . . . . . . . . . . . . . 27
4. Fragment U 7/80 (UET VII 74) . . . . . . . . . . . . . . . . . . . . 27
5. Explanation of Metaboles by M. Duschene-Guillemin . . . . . . . . . . . . . 28
6. Hurrian Tablet, E. Laroche’s transcription . . . . . . . . . . . . . . . . . 29
7. A. Kilmer’s interpretaion .. . . . . . . . . . . . . . . . . . . . . . . 29
8. Megiddo Ivory . . . . . . . . . . . . . . . . . . . . . . . . . . 30
9. Interpretation of musical terms, with digits left out . . . . . . . . . . . . . . 30
10. Concordance of Human text and musical themes on the Ras Shamra tablet . . . . . 31
11. Transcription by M. Duchesne-Guillemin . . . . . . . . . . . . . . . . . 31
12. Thematic analysis of a Jewish Psalm . . . . . . . . . . . . . . . . . . . 32
13. Syro-Chaldean chant . . . . . . . . . . . . . . . . . . . . . . . . 32
14. Final part of a Syro-Chaldean charit . . . . . . . . . . . . . . . . . . . 32

LIST OF PLATES

Rate Illustration

I 1. Lyre-Kithara from Ur
2. Oblique silver lyre-kithara
I1 3. Tell0 relief
111 4. Nippur Tablet CBS 10996
Iv 5. Ur standard
6. Shell plaque from Ur
v 7. Giant lyre-kithara from Ischali
8. Funerary pillar, Xanthos
VI 9. Hurrian Tablet, copy by E. Laroche
10. Hurrian Tablet, photographed by A. Kilmer in Damascus Museum
11. Hurrian Tablet, copy by A. Kilmer
I

Twenty years ago, very little was known about Mesopotamian music. Musical instruments
were depicted on monuments or mentioned in texts; the famous excavations at Ur had
brought to light a hoard of harps and lyres adorned with gold, silver, lapis lazuli and
other semi-precious stones dating from about 2,600 B.C. (see P1. I,l)! However, no music
had been discovered and the existence of a music theory was not even envisaged.

Some attempt had been made to decipher a Babylonian ‘‘musical score” in 1924 by the
ethnomusicologist Curt Sachs? Working with a bilingual hymn on a tablet in the Berlin
Museum, Sachs tried to reconstruct a scale from the tablet’s first column. He did this
through a statistical estimation of the frequency of syllables. The resultant melody,
however, was astonishing and unacceptable. In my doctoral dissertation, “The Musical

‘Beside the many articles scattered in encyclopedias or journals, dealing with the numerous musical
instruments recovered from excavations or represented on reliefs, mosaics 02 paintings, the following
fundamental works will give a fair idea of the subject;
Curt Sachs, Geist und Werden der Musikinsmmente, Berlin, 1929 (chiefly ethnological)
Francis W. Galpin, The Music of the Sumerians and their immediate succebsors the Babylonians and
Assyrians, London, 1937 (beautifully illustrated but philologically unreliable, and faulty in its restoration
of the British Museum harp and of the pseudemusical notation.)
Marcelle Guillemin and Jacques Duchesne, “Sur I’Origine asiatique de la Cithare grecque”, Antiquite
classique 4, 1935, pp. 117-124.
M. Duchwne-Guillemin, “La harpe en Asie occidentale ancienne”, Revue d’Assyriologie 34, 1937, pp.
2 9 4 1.
Idem., “La harpe i plectre irauienne: son origine et sa diffusion”, Journal of Near Eastern Studies 28,
1969, pp. 109-115; “Note complementaire sur l’instrument AlgaI”, Journal of Near Eastern Studies 29,
1970, pp. 200-201.
Idem,“Mesopotamie”, DictionMire de la Musique, Bor&s, 11, Paris, 1976, pp. 597-601.
Idem., “Music m ancient Mesopotamia and Egypt”, World Archaeolob 12, 1981, pp. 287-297 and
plates.
Joan Rimmer, Ancient Musical Instruments of Western Asia in the British Museum, London, 1969.
From the philological point of view: Henrike Hartmann, “Die Musik der sumerischen Kultur”
(Dissertation), Frankfurt-am-Main, 1960 (a serious study, though little conclusive.)
The mast important find was in Ur, where the royal tombs yielded the richest instruments: Leonard
Wodley, Ur Excavations 11, The Royal Cemetery, 2 volumes, London and New York, 1934.
Barnett in his article “New facts about musical Instruments from Ur”, Iraq 31, 1969, pp. 96-103,
insists (p. 101) on the authenticity of the lyre with rampant stag kept in the University Museum of
Philadelphia (U. 123555) (see Pl. 1,2), which had been questioned by Wilhelm Stauder in his book Die
Harfen und Leiern &r Sumerer, Frankfurt-am-Main, 1957, p. 48, and by his disciple H. Hartmann, op.
cit., pp. 22-23 (Figs. 16a-16b). But R Barnett, after a “careful study of the photographs of the
objects in the ground” and of a pair of stags in copper lying near by, is convinced that the instrument
is quite distinct from the copper pair. The silver lyre is unique with its oblique yoke and its boat-
shaped sounding box, but the stag on it can be compared with the animals on the Failaka seal and cia&
relief from Tell0 kept in the Louvre (see PI. 11,3). Of the harp of Queen Pu-Abi in the British Museum,
W. Stauder proposed a good reconstruction, which has been accepted by the specialists and carried 0 s
in the Museum.
’Curt Sachs, “Die Entzifferung einer babylonischen Notenschrift”, Sitzungsberichte d Preuss. Akad
d Wissenschaften 18, 1924, pp. 120 ff.

[SANE 2, 671
6 M. Duchesne-Guillemin [SANE 212

Instruments of Ancient Western Asia” (Lii?ge, 1932, unpublished3), I showed Sachs’

position to be untenable. This judgment was later upheld by Benno Landsberger, who
rejected Sachs’ argument on philological grounds and proved that the syllables in ques-
tion had nothing at all to do with music.4 Landsberger’s rebuttal did not prevent an
English canon, Francis Galpin, proposing another interpretation of these syllables in his
1937 book on Sumerian music5. Errors die hard: Galpin’s attempt was still accepted
thirty years after in the EncyclopCdie des Musiques sacr4es.6

In 1960 Anne Kilmer published a mathematical tablet7 part of which concerned musical
strings. This tablet came from the ancient town of Nippur and is kept at the Museum of
the University of Pennsylvania in Philadelphia.8 Although, according to the curator Samuel
Noah Kramer, the late Legrain had judged it interesting, it remained unpublished for seventy-
five years. It is written in a peculiar, rather difficult type of cuneiform script, seemingly be-
longing to the Kassite period, i.e. about 1500 B.C. (see P1. III,4 and interpretation in Fig. 2).

In her commentary, Kilmer wrote (p. 274):

Column I in particular is unique in that what is preserved of this column deals with the strings
of one or more musical instruments. The numbers that begin each line may or may not be
coefficients; in any case their function is obscure. The first five preserved lines are arranged
m the usual number-object manner of the coefficient lists, while the lines that follow appear
to be elaborated, in that the numbers are “defmed” before they are given: e.g. numbers 1, 5
are preceded by the names of the strings to which they apply, fore string and fifth string.
What is being given, therefore, seems t o be the string names together with .their numbers, and
their relationship t o other string names, or, possibly, to certain stringed instruments. That
these are string names is made clear from an unpublished tablet from Ur (U. 3011), of which
the writer has been able to utilize (through Prof. Landsberger)-a hand copy available by
courtesy of Prof. 0. R. Gurney; the obverse deals with a certain nine-stringed instrument. The
consecutive numbering of strings 1 t o 9 is: 1, 2, 3, 4, 5 , 4, 3, 2, 1: the last four are said to
be “behind.” The instrument must have, therefore, either two rows of strings,‘-one placed
behind the other, or a two-part arrangement in a single row, one set of which is numbered
in one direction, and the other from “behind ” (see Fig. 1).

My preference, based on the non-existence of attested instruments with two rows of strings and
on the fact that the animal’s head represented the forepart of’the instrument, as can be seen on
the Ur standard (see P1. IV,5) and on the scene with musical animals on the shell plaque from Ur
(see PI. IV,6), was to adopt the second of Kilmer’s suggestions: a single row of strings,

3Except for articles extracted from it:

M. Guillemin and J, Duchesne, “Sur l’origine asiatique . . .”, see above, note 1.
M. Duchesne-Guillemin, “Note sur la provenance asiatique d’un tambour egyptien”, Archuologirche
Mitteiktngen aus Iran 8, 1936, pp. 54-55.
Idem., “La harpe en Asie . . .”, see above, note 1.
Idem., “La Musique en Egypte et en Mekopotamie ancie~es”,Encyclopidie de la Ple‘iade. Hktoire de
la Musique, I, Paris, 1960, pp. 352-362.
4Benno Landsberger, “Die angebliche babylonische Notenschrift. Archiv f i r Orientforschunn 1.1933, pp. 17(1178.
’Francis Galpin, The Music . . ., see above, note 1.
6Robert Statlender, “Sumer et Babylone”, Encyclopedie des Musiques sncre‘es, Paris, 1968, pp. 303-309.
‘A Draffkorn-Kilmer, “TWO new lists of key-numbers for mathematical operations”, Orientaliu 29,
1960, pp. 273-308.
*C. B. S. 10996.

[SANE 2, 681
 19841 The Discovery of Mesopotamian Music 7

the last four being counted backwards. In fact, this was a succession of nine string.

The progression of numbers, from one line to another, suggested the notation of a scale,
and the fact that the progression neverexceedsseven (although there were nine strings)
argued for a heptatonic scale. Another fact then occurred to me that had escaped the
editor’s attention: the first five lines on the one hand, with the strings merely numbered,
not “defmed”, and, on the other hand, the following ones in which the strings were
both numbered and “defmed” constituted two versions-one abridged the other written
in full-of one and the same text. Since the tablet was damaged at both ends, only the
latter part of the first version and the fmt part of the second version were extant. SO
that it was possible, by combining the two versions, t o reconstruct the whole text.
However, in Kilmer’s edition, some of the numbers did not agree: in line 2 she read
“4,3” where a comparison with line 21 led one to expect “6,3”; and in line 3 she read “3,6”
instead of the “3,5” corresponding to line 22. This prompted me to ask Samuel Noah Kramer
to send a photograph. Instead, he kindly brought the tablet itself to Chicago, where my
husband was a visiting professor, and we examined it together with Hans Giiterbock and
Anne Kilmer. Following this examination, Giiterbock and Kilmer provided a new edition
of the tablet in Studies in honor of B. Landsberger. In a later article lo Kilmer declared
herself “happy to say that as a result of Dr. Duchesne-Guillemin’s analysis, not only
were many readings improved, but we were able to restore the preceding broken section
to such an extent that the progression from one to seven and again to one was firmly
established”

The entire sequence of numbers on the tablet, in the unabridged version, is now as
follows:

line 11: 1-5 line 18: 1-3

line 12: 7-5 line 19: 5-2
line 13: 2-6 line 20: 24
line 14: 1-6 line 21: 6-3
line 15: 3-7 line 22: 3-5
line 16: 2-7 line 23: 7-4
line 17‘: 4-1 line 24: 4-6

Fig. 2 shows the jumps from one string to another. The terms at the end of each line
are given according to the latest readings, obtained by comparison with the Hurrian
tablet (see below, 6.) The translations remain tentative. It will be noted that only
seven of the nine strings are taken into account, and that there are three groups of
fwe strings, two of six, four of four, and fwe of three. If we apply this to the
oblique lyre, we may surmise that the forestrings were lower in pitch and that this
1 scale is therefore an ascending one, unlike the scale of the Greek theoreticians.

9A. Draffkorn-Kilmer, “The s h g s of musical instruments: their names, numbers and significance”,
Studies m honor of B. Landsberger, Chicago, 1965, pp. 261-268, with an Appendix by M. Duchesna
Guillemin, “Note complkmentaire suf la dkcouverte de la gamme babylonienne”, pp. 268-272.
lo A. Draffkorn-Kilmer, “The discovery of ancient Mesopotamian Theory of Music”, Proceedinn of
the American Philosophical Society 115, 1971, pp. 131-149; especially p. 134.

[SANE 2, 691
8 M. Duchesne-Guillemin [SANE 212

1. THE SCALES

What was the tonal system employed? C. Sachs had taught that all oriental music in
antiquity was governed by the pentatonic system, i.e., one based on a division of the oc-
tave into fwe notes with no half-tone, as we find, for example, on the black keys of a
piano. If we adopt this formula, however, the three jumps of five strings do not have
the same amplitude, nor are they consonant.

When I tried the enharmonic system, according to some Greek traditions the most ancient,
I encountered the same difficulties regarding the jumps of four strings. Thus this h y p e
thesis, too, was discarded.

There remained a third possible solution, none other than o x diatonic heptatonic system.
This was already suggested by the fact that only seven out of the nine strings of the
instrument occurred in the pairings of strings. This hypothesis gave a division of the
octave comparable to that of the white keys of the piano. The five-string jumps were
equal and consonant fifths. As a working hypothesis, I assumed that the designation of
the third string as “thin” could signify “higher in pitch” and therefore sounded nearer
to the fourth string, thus indicating the place of the semi-tone. There is of course a
second semi-tone in the diatonic scale, that between the seventh and eighth strings,
but this was left out by the theory, since it did not go beyond the seventh string.

The “tkin” string was followed by the fourth, which was named after the god Ea:
“Ea-made-(it)”. Ea was supposed to have been the creator of the arts. The designation
therefore seemed to indicate the importance of that string, just as the fourth note or
string in the Greek scale, mese, was prominent as the basis of the tuning of the lyre.
We shall return to this analogy below (see 2).

The presence of jumps of a third alternating with jumps of fourths and fifths in the
Nippur-Philadelphia tablet made me think of a method of tuning, in which the sixths
(1-6, 2-7) could be inversions of the thirds which corresponded to 8-6 and 9-7 and
were excluded from a theory extending only to seven strings. While my first article on
the theory was being printed,”I had a friendly exchange of letters with H. S. Powers of
the University of Pennsylvania, who convinced me that the tablet was not a tuning
method. My second article12 admitted as much, but the real tuning method was to be
discovered later on (see 5). 5).

In short, this first tablet gave the names of each of the three fifths, two sixths, four
fourths, and f i e thirds included in the seven note scale and differing according to their
respective positions in it. It should be noted that, as indicated by the arrows in Fig. 2,
the fifths and the sixths are ascending, while the fourths are descending: as for the thirds,

M. Duchesne-Giullemin, ‘‘Wcouverte d‘une gamme babylonienne”, Revue de Musicologie 49, 1963,

pp. 3-17.
121dem., “A l’aube de la the‘orie musicale: concordance de trois tablettes babyloniennes”, Revue de
Musicologie 52, 1966, pp. 147-162.

[SANE 2, 701
 19841 The Discovery of Mesopotamian Music 9

four of them are ascending while the highest goes down. This cannot be explained
except as a survival of ancient gestures.

My interpretation of the Nippur-Philadelphia tablet as representing the heptatonic system

was received with skepticism. Landsberger, especially, was cautious, after having refuted
Sachs. However, my thesis served to draw the attention of scholars to the problem, for
it incontrovertibly pointed to a musical theory which the Babylonians had worked out
many centuries before the Greeks. The earliest preserved Greek theory dates to the fourth
century B. C., one thousand years after the Kassite period.

% 2. THE STRINGS

In 1965, Anne Kilmer, while examining material for inclusion in the Chicago Assyrian
Dictionary, discovered on a tablet from Assur, now in Berlin (VAT 10101-see Fig. 3)
seven of the names designating pairs of strings. This text was a catalogue of Assyrian
hymns classified according to seven of the terms in the Philadelphia tablet, namely,
those of the fourths and fifths (underlined in Fig. 2). Kilmer published this find, together
with the first column of a lexical text from Ur, now in the British Museum (U. 3011-
Nabnitu XXXII see Fig. 1) transcribed by Gurney, and a new edition of the Philadelphia
tablet, referred to above.14 I also contributed to the same volume with a brief commen-
tary," and I showed the convergence of the three documents in a longer article in the
Revue de Musicologie.16

The Ur tablet had provided Kilmer with a clue to the names of the strings. Now, the peculiar
way of numbering the strings made me suspect that the Greeks designated the strings of the
lyre in a'similar way. In fact, they turned out to have counted the strings as did the Sumerians,
namely from both ends.I7 Their list of strings is: hypate, parhypatt?,lichanos, mest?,paramest?,
tritt?, paranzti?, nett?. Since parhypati? follows hypatt?, and paramese' follows mest?, parantte'
must similarly have followed nett?. Hence the counting of the last three strings must have begun
with nEtZ and proceeded backwards: nt?te',parane'tt?,triti?. Tritt?thus comes third, which
clinches the argument. The system can be represented as foll'ows:

hypati?, parhypatt?, lichanos, mest?, paramesz, triti?, paranZt8, nett?

> <
This is parallel to the Sumerian counting, except for one small point: while the Sumerians,
on their instrument, operated with nine strings, the Greek theory was limited to the eight
1 strings of the octave. Consequently, there was no string in the Greek system corresponding

"In fact, as will be seen below, p. 12, the theory dated back at least to the 21st century B.C., and
the Philadelphia tablet prwed to be Neo-babylonian.
14See abwe, note 9.
Idem.
l6 See above, note 12.
17M. DuchesneGuiUemin, "Survivance orientale dans la d6signation des cordes de la lyre en Gre'ce",
S y k 44, 1967, pp. 233-246.

[SANE 2, 711
10 M. Duchesne-Guillemin [SANE 212

to the “fourth behind” of the Sumerians. It is notable also that the notation of “middle”
in the Greek theory includes the paramesi? (5th note), just as in the Mesopotamian names
of the intervals or portions of scales qablitu extends from the second to the fifth string.

3. THE MODES

1968 witnessed the most important contribution to our knowledge of Babylonian musical
theory in the form of a fragment also from Ur (UET VII 74-U. 7/80 see Fig. 4), found
in the British Museum by its curator E. Sollberger. It was published by 0. Gurney, with
the help of the musicologist D. Wulstan’* and confirmed the heptatonic principles sur-
mised on the basis of the Philadelphia tablet. The same seven terms found on both the
Philadelphia and Berlin texts here designated seven different diatonic scales and the method
for passing from one to another, i.e., changing pitch, on a nine-stringed instrument. The
style of writing indicated that this text dated from the 18th century B.C. The Babylon-
ians therefore already at this early date knew seven diatonic scales, each formed of fwe
tones and two half-tones and capable of constituting a mode, i.e. a fmed succession of
notes as a basis for a melody.

Changing the mode was brought about by displacing the half-tones in the octave. To
make this clear, we may use the white keys of the piano, starting from a C. The first
half-tone is between E and F, the second one between B and C. In other words, the
first half-tone will be between the 3rd and 4th notes, the second between the 7th and
the 8th. If we start from D, the first half-tone will be between the 2nd and the 3rd
notes, the second half-tone between the 6th and 7th. If we start from E, the first half-
tone will be between the 1st and 2nd notes, the second between the 5th and 6th notes,
and so on. This alters the aspect of the scale and consequently-an essential point recog-
nized by the ancients-the ethos of the melody. The relations between the principal
notes varied according to the mode chosen, and as in Greece, there were seven modes in
Babylonia. This is why songs could be classified according to their modes, as attested
in the Berlin tablet found by Anne Kilmer.

Theoretically, modulation could also be achieved by shifting the scale, either down or up,
along the white key-board, so that from

c D E-F G A B-c ( - indicating the half-tone)

you get either

B-c D E-F GA B and so on,

I8Oliver R. Gurney, “An old Babylonian treatise on the tuning of the Harp”, - . ~ q30, I 68, pp. 229-
223. In the same issue of this journal, David Wulstan, who had brilliantiy collaborated with Gurney,
tried to prove m another article, “The tuning of the Babylonian Harp”, pp. 215-228, that the basic
scale of the Nippur-Philadelphia tablet was established on the note D. But later on, after reading Kummel’s
article (see below, note 24), he retracted (see below, n. 27).

[SANE 2, 721
 19841 The Discovery of Mesopotamian Music 11

or D E-F G A B-c D and so on.

The latter is the process described by the Greek theoretician Ptolemy.

The Babylonians used another method, encompassing only seven strings. They had noticed
that the tritone interval is dissonant. They called it impure. This was the interval to be
altered. It is either an augmented fourth, made up of three whole tones, or a diminished
fifth, made up of a half-tone, two tones and a half-tone. In order to make the fourth
and f i t h consonant, the dissonant fourth must be diminished into the true fourth (two
tones and a half-tone), and the diminished fifth must be augmented to three tones and
a half-tone.

The Babylonians distinguished two processes: tuning down and tuning up (see Fig. 5,
showing’the successive alterations), and treated them on the tablet in two sections separ-
ated by two signs isolated on a line: N U SU, a Sumerian expression (a hapax) which
Gurney left untranslated and which Dr. Kilmer interpreted as “no more”, probably mean-
ing “end of this matter, now for something else”.

As an example of the first p m s s : to diminish the augmented fourth F-B,the B string

(the highest string of the group) is tuned down to B flat, producing the true fourth F-B
flat. To augment the diminished fifth E-B flat, the E string (the lowest of the group) is
tuned down, producing the consonant fifth E flat-B flat. In the second process the
augmented fourth F-B is made consonant by tuning up the lowest string of the group: F
becomes F sharp. The diminished fifth F sharp€ is purified by tuning up the highest
string of the group, C, to C sharp, producing the consonant fifth F sharp€ sharp.

It is clear that each process produces a d e f i t e succession of modes, one in reverse order
of the other. The alterations occur in the same order as in our modern theory:
B E A D G C F for the sequence of flats, and F C G D A E B fq_r the sequence of
harps

The instrument is originally tuned in C, although its lowest note is E. The order of
strings is therefore (since there are nine strings):

EFGABCDEF
The eighth string is always altered together with the first, or the ninth with the second,
a proof that the division of the octave is heptatonic. AU this was seen quite correctly
by Gurney and Wulstan, but the latter wondered what the relationship was between - the
names chosen to designate we octave species. In my third article (1969)19 I explained
a constant in the choice of the terms designating the mode: the fourth or fifth after
which the octave species was named always had its half-tone at the upper end of the
group. From this observation we can deduce that the terms designated not only inter-
vals and modes but also portions of scales, obeying a strict order: tone+tone+half-tone
for the typical fourth; and tone+tone+tone+half-tone for the typical fifth.

l9 M. Duchemffiuillemin, “La thdorie babylonienne des mdtaboles musicales”, Revue & MuSicoZogie
55, 1969, pp. 3-11.
12 M. Duchesnffiflemin [SANE 212

Aaron Shaffer’s interesting article, “A new musical term in ancient Mesopotamian Music”,
Iraq, 43, 1981, p. 79 ff., suggests two alternative meanings for the term, Akkadian
is-su,-ha-up, Sumerian Su,-Su, , namely “overturning” or “throwing down”. Musicologically,
however, only the latter makes sense. And this can only support Gurney and Wulstan’s
excellent interpretation of the British Museum tablet On the other hand, it is most im-
portant to note that the evidence adduced by Shaffer from the self-laudation of King
Sulgi proves that the tuning method was already in use with the Sumerians as far back
as the 21st century.

4. THE NAME OF THE INSTRUMENT

The British Museum fragtl?_nt gives a further important fact: the name of the instrument.
It is called in Sumerian @Z-A.Mi, a name occuiring $I several texts but unfortunately
without any description. (@%AM1 corresponds to W Z A C S A L in the earlier readings).
This instrument is commonly thought to be a stringed instrument made of wood (this
is confirmed by our fragment, since gis is the determinative for wood), but there is
hesitation between lyre and harp. The lexicographic lists published by Landsberger2’
give the Babylonian equivalent: mum-mu-u. This is mentioned on several occasions and
may refer to several instruments which are closely related but which differ in some
particulars. It is also said to be associated with the goddess Inanna This may help us to
recognize its nature, for the instrument of the goddess is elsewhere2’ called zasn-nu-ru,
which Laroche compares with Hittite z i w . There is notab& a hun-zinm “great zinar”,
which might recall the large Sumerian lyre-kithara which rested on the ground and
continued in use after Sumerian times (cf. the Ischali terracotta in the Chicago Oriental
Institute A 9361, P1. V,7). A specimen of such an imposing instrument was found by
N. b g i i q at Inandyk. It is shown on a beautiful vase in the Ankara Museum: two,
musicians, their backs toward us, are seen playing together on one instrument taller
than they are22. This great lyre, dating from the 16th century B.C., exactly resembles
the great Sumerian upright lyre played in the scenes of animal-musicians found in Ur
(see PI. IV,6).

On the other hand, the lexicographical lists referred to above render gi“zA.Mi s i - d by
iiurtu. This may be an instrument, or the name of its tuning; we know that, in the
theory, it designates a certain group of strings (2-6). However, the translation of most of
the technical terms is still questionable.

To sum._up this philological digression, the instrument named in the British Museum frag-
ment, @%A.Mi, is not a harp but a lyrekithara. This interpretation, given in my 1969
article, 23 is now commonly accepted.

2o Materialien zum sumerischen Lexikon VI, p. 122, line 44.

“lbid p. 119, line 43.
22The instrument is still unpublished; but N. Ozguq showed me a good photo.
23 See above, note 19.
 19841 The Discovery of Mesopotamian Music 13

5. TUNING

In 1970, the German assyriologist H. K t i ~ n m e lexplained

~~ satisfactorily how the instru-
ment was tuned by alternating’d&cending fourths and ascending fdths, a tuning later
called Pythagorean. It appears from an examination of the process described in the
British Museum fragment that the tuning is governed by four rules:

1) an ascending fifth and a descending fourth are used;

2) the heptachord is a limit not to be exceeded m the alternating process. Hence
the alternation is interrupted on the 4th string (the Ea-string) and gives way to a
succession of two descending fourths;
3) the first tuning gesture starts on the group of string after which the mode is
named and which is characterized by hiwing the half-tone between its highest two notes;
4) the tuning ends on the tritone. In order to change to the next scale, the tri-
tone is inflected to reach consonance.

This is the only detailed, exhaustive description known in history of the secalled
Pythagorean tuning.

An important conclusion for the definition of the Pythagorean tuning is that contrary to
the common opinion, m the course of the alternating process the octave is not mentioned.
It is only implied when changing the pitch of the first string for the 8th, or of the
second for the 9th.

Only one scale is completely governed by the alternating process; it is the scale of nid-qabli,
or C scale, which appears to be the basic scale underlying the theory reflected in the CBS
tablet.

i
6. THENOTATION *

When the Hurrian tablets from Ras-Shamra were published by E. L,aroche,Z’ Guterbock
immediately recognized in one of them (see P1. VI,9 and 10) a slightly Humanized form of the
musical terms used in the Philadelphia tablet (see Fig. 2). The terms, in Akkadian, were written
underneath a Human hymn and from this Gilterbock inferred a musical score.26 A year later,
a fnst attempt at interpretation was made by t&e musicologist W u l ~ t a n . 2 In
~ 1973, this

24 Hans M. Kiimmel, “Zur Stimmung der babylonischen Harfe”, OrientuZiu 39, 1970, pp. 252-263.
25 Emmanuel Laroche, ‘Documents en langue hourrite provenant de Ras Shamra, II, Textes hourrites
en cuniiformes syllabiques”, Ugaritica V, 1968, pp. 462-496. The year before, I had written to
Rofessor Jacques Chailley, of the Sorbonne, Park: “Si les Babyloniens avaient voulu noter une melodie,
ils n’e’taient pas loin d’en trouver le moyea Et je ne de’sespe’re pas qu’on --)Lume, un jour, un texte
ou tous ces mots-cle‘s seront mis p&-m&le. Alors, on aura enfm de la musique!”
26 Hans Giiterbock, “Musical Notation in Ugarit”, R e w e d’AssyrioZogie 64, 1970, pp. 45-52.
27 David Wulstan, “The earliest musical notation”, Music w d Letters 52, 1971, pp. 365-382.

[SANE 2, 751
14 M. Duchesnffiuillemin [SANE 212

was questioned by Kilmer, who produced a very different, polyphonic rendering.*’ I later
refuted these two analyses and proposed instead a monodic melody, with parallels in
the traditional Jewish and SyrdJhaldean Christian music?’ The history of these various
evaluations is a complex one and will, therefore, be detailed below.

6.1 Philological Background

The tablet, currently in the Damascus Museum, is made up of three fragments (R. S.
15.30 + 15.49 + 17.387), assembled by Laroche in his work under number h.6, pp. 463 .
and 487 (see P1. VI,9 and Fig. 6). The writing on the tablet is divided into two parts:

1) The four lines on the upper part of the tablet are a Hurrian hymn, the meaning
of which escapes us almost entirely because of our imperfect knowledge of the language.
We can recognize only a few words: the goddess Nikkal, wife of the.moon-god; a gift;
a heart; fathers . . .. The text is written in a manner unattested in Mesopotamia: each
line, starting on the obverse side is continued around the right edge to the reverse.
And the last few syllables on the first three lines on the reverse are repeated at the
beginning of the following line on the obverse.

2) Below the four lines, on the obverse, two dividing lines run across the tablet
with two pairs of wedges inscribed between them. Below this are the six lines of nota-
tion consisting of the Babylonian terms, each followed by one of the numbers 1, 2, 3,
4, 5(?), and 10.

Finally, on the reverse there is a colophon which indicates-the mode, nfd qibli, in which
the song is composed, as well as the name of the composer and the scribe.30 This mode,
in Babylonian nT. qabli (Sumerian NIM or SUB-MURUB), is none other than that of the
scale which I had reconstructed on the basis of the Philadelphia tablet, without then
imagining that there were seven possible modes. As it is simply the scale of C, the pitch
of each of the notes represented by the terms on the Hurrian score can be deduced from
the pattern of the Philadelphia tablet (see Fig. 2).

However, three words presented problems. One, in the middle of line 5 , was poorly
preserved. A second word in line 7 had been read tuppunu by Laroche, which seemed
to make some sort of sense, namely “our tablet”, but musically was worthless. I, there-
fore, suggested to Laroche that, by slightly altering the first and third syllabic signs-the
second standing equally well for pu or bu-one could recognize, under its Hmianized
form, one of the terms defined in the Nippur-Philadelphia tablet, enzbiibe. Laroche
accepted the suggestion and, in fact, as appears on Kilmer’s very good photo on the

28 A. Kilmer, “The cult song with music from ancient Ugarit: another interpretation”, Revue
d’Assyriologie 68, 1974, pp. 69-82.
29 M. Ducheme-Guillemin, “Les probldmes de la notation hourrite”, Revue d’AssyrioZogie 69, 1975,
pp. 159-173, and “De’chiffrement de la musique babylonienne”, Accademia dei Lincei Roma, 1977,
Quaderno 126, p p . 3-24.
30 In the lot of fragments published by Laroche in Ugmiticu V (see above, note 25) there are traces
of sixteen colophons, each belonging by definition to a different tablet. On five of those colophons
Laroche was able to decipher the mode: it is always, curiously enough, the nid/Mt RibZi (or, as written
by Kilmer, qubli), viz. the C-mode.
[SANE 2, 761
19841 The Discovery of Mesopotamian Music 15

cover of Sounds of Silence, the correct reading, given by her, is umbube, a plausible
Hurrian adaptation of ernbiibe?I About the third word, udtamari, see below, p. 20.

6.2 First Interpretations

As outlined above, three attempts at interpretation were made previous to my own.

The English musicologist D. Wulstan, who as we know had brilliantly explained the
British’Museum tablet with Professor Gurney, presented his analysis, along with a record-
ing of the music, at the 1971 Rencontre Assyriologique in Paris32 The digits following
the musical terms indicated, in his opinion, the number of the note to be taken from the
portion of the scale defined by each term. But since the number 5 appeared after a
term for a four-note portion, he inverted the fourths into fifths in order to find a
fifth note. This, however, did not account for the occurrence of the number 10 after
a group of three notes. After this disappointing result, anyone might feel justified in
taking up the problem again.

The second try was made in 1972 in Guterbock’s lectures on the subject in Chicago and
Germany. These unfortunately have not been published. Professor Giiterbock counted the
syllables of the hymn but did not succeed in making them tally with the notes. On the
other hand, he believed that the syllables repeated at the beginning of three of the lines
were refrains.

Anne Kilmer presented her interpretation at the International Congress of Orientalists in

Paris in 1973. She divided the religious text according to the written lines, taking as
refrains the last words of a line on the reverse which were repeated at the beginning of
the next line on the obverse, as had Professor Giiterbock The musical terms were inter-
preted as chords of two notes, “dyads” or “dichords”, in which the upper note is the
melody ‘while the lower one is the accompaniment. Kilmer claimed that these dichords
are to be repeated the number of times indicated by the digit following each term.
Since the total number of dichords thus obtained did not match the number of syllables
in the hymn, she repeated the whole song, but in an inverted musical order. Moreover,
she used some of the musical terms in line 5 for the music of the “refrains”. Unfortun-
ately, this does not tally. Kilmer then added a coda with music taken from line 5 , sing-
ing it twice to account for the digit 10. Thus line 5, combining refrains and coda, is
isolated from the song proper, which begins on line 6. This is complicated and arbitrary.

6.3 Criticism

Anne Kilmer’s attempP3 can be refuted on several grounds. For one thing, a grammatical
feature pointed out by Laroche should be taken into account The first word of a

31E. Laroche, “Etudes hourntes”, Revue d’Assyriologie 67, 1973, pp. 119-130.
A. Kilmer, Revue d‘Assyriologie 68, 1974, p. 73; Sounds from Silence. p. 12.
”!See above, note 27.
3 3 ~ above,
e note 2s.

[ W E 2, 771
16 M. Duchesnffiuillemin [SANE 212

Hurrian sentence is provided with an enclitic, al. We can therefore divide the religious
text into its sentences and determine whether the refrains occur between sentences, as we
would expect This is not the case, for we find seven sentences arranged as follows:

1) xxxhanuta niyaia ziwe Binute zuturiya ubugara xxhuburni

2 ) TaSal killa zili Sipri bumarubat uwm’ humambat uwari wandanita ukuri kurkurta
3) fialla ulali kabgi allibgi Sirit murnuju.
wegal tatib tiSia-unuga kapiili unugat akli Sam-amme xlil.
5) Uklal tununitax(xxkka kalitanil nikala kalitanil nikala. i
6 ) Ni&ur&l hana &muteti.
7 ) A ttayaStal atarri hueti hanuka (xxxxxxxhziiati wewe bnuku.

The repeated words occur once at the end of a sentence (sentence 5)’ once in the middle
(sentence 2 ) and once at the beginning (sentence 4). (Sentences 1, 3, 6 and 7 have no
refrain at all.)

The theory that these are refrains must therefore be dropped, and we must return to
Laroche’s suggestion to take the repeated words as ~atchlines.3~ Kilmer discarded Laroche’s
suggestion on the ground that a catchline never occurs in Babylonia between different
parts of a tablet, but only between successive tablets of a series. However, the hypothesis
that these are catchlines is justifEd by the fact that the writing goes recteverso-an arrange-
ment unknown in Mesopotamia. Matching the transcription (Fig 6) with the cuneiform
(PI. VI,3), the reader may see that the repetitions could serve the function catchlines
as the musician turned the tablet over during a performance, o~ when rehearsing.

In addition, if we examine the musical terms, we cannot help but notice sequences of
terms repeated in the same order, which must constitute musical themes. (see Figs. 6
and 7). The first of these, which we may call Theme A (line 6), is titim Sarte.
zirfe, Sahri, xxte, irbute, which recurs with the last word of line 7 and the fist half of
line 8. Theme B appears in lines 8 and 9; it is made up of a triple repetition of Sahri-
iaSate. Finally, Theme C, also made up of two expressions, kitme-qablite, repeated
three times, terminates the song. Such a structure, which in.my opinion is of fundamen-
tal importance, escaped Kilmer’s notice; for example, she interrupts Theme B after two
dichords in order to introduce her so-called refrain.

An examination of h n e Kilmer’s copy (Pl. VI,ll) reveals three details which confirm my
interpretation.

1) At the beginning of the hymn two or three syllables are missing. Kilmer only
counts one.
2 ) In the first line of the musical notation there appear to be only two vertical
wedges after the third word rather than three, and according to the spacing of the
existing signs there is no room for a third. However, the third wedge is necessary to
support Kilmer’s hypothesis that the number of repetitions (3+1+?) is equal to seven,
to account for the seven syllables of the refrain.

34 Laroche, Ugaritica V, 1968, p.464.

[SANE 2, 781
19841 The Discovery of Mesopotamian Music 17

3) The restored Su iini “of two, double” in the middle of line 5 is essential to
Kilmer’s interpretation, since it suggests to her the idea of repeating the melody, the re-
frains and the coda in order to make the musical notation tally with the words of the
hymn. However, there hardly seems to be enough room for the three syllables required
by the restoration proposed. Moreover, it is difficult to accept so fundamental a recon-
struction on the basis of a mutilated passage.

Kilmer’s theory is also questionable from a musicological point of view. I expressed this
to her in Paris and Wulstan also demurred in a review article.35 It must be stressed that
music has remained monodic in the Near East up to the present day. If it had been only
a question of heterophony, which may have been known in Greek music, one would ex-
pect to find first of all among the dichords the octave, which is the most natural heter-
ophony. However, there are no octaves. In addition, there is no example of noting together
both a song and its accompaniment.

Dr. Kilmer published her interpretation in the Revue d’A~syriologie~~ and issued, with the
assistance of Richard L. Crocker and Robert R. Brown, a record and a booklet,w Sounds
from Silence, in 1976. Of the latter work, the philological and historical sections, which
deal with the opening of this new area of Assyriology, are almost entirely correct: the
reproduction of the three fragments which Laroche so fortunately joined is clear and the
photographs of the tablet, taken at the Damascus Museum and slightly enlarged, are
excellent. However, the musicological discussion is questionable, not only in the matter
of polyphony but on several other grounds. For one thing, it accepts a useless hypothesis
~ ~ reconstructed the theory of the Philadelphia tablet as if
put forward by S t a ~ d e rwho
it were meant to be demonstrated on a nine- or seven-stringed lute, an instrument which
never existed in ancient Mesopotamia. Mesopotamian theory was based on a

When Kilmer and Crocker attempted to restore the British Museum tablet, they stripped
it of the stage df the process which started from the iiartu, and began with the qablitu-
tuning. This left only six modes in the first column, instead of seven. One wonders why
the instrument should not begin with its normal tuning, that of the iiartu in the C-scale
(beginning with E). It is clear, that Kilmer and Crocker did not understand that the two
methods of modulation consisted in flattening and sharpening, not in de-sharpening and
sharpening. This is certain, since the method, especially in its fmt stage, must always
start from the usual “accordatura” (Fig. 5 ) . The two processes are described successively
in the British Museum tablet.

Finally, the reconstruction of the Megiddo lyre suggested by R. Brown needs to be qual-
ified (Fig. 8). On D. -20 of the booklet. we read:

=’D. Wulstan, “Music from ancient Ugarit”, Revue d’Assyriologie 68, 1974, pp. 125-128.
36 See above, note 28.
37 A. Draffkorn-Kilmer, Richard L. Crocker, Robert R Brown, Sounds from Silence, Berkeley, 1976,
in which Dr. Kilmer’s hypothesis is essentially unchanged.
38 W. Stauder, “Ein Musiktractat aus dem zweiten vorchristlichen Jahrtausend”, Festschrift W. Wiora,
Kassel, 1967, pp- 157-163.
39See above, n. 19.
18 M. Duchesne-Guillemin [SANE 212

It is impossible to determine on the basis of the Megiddo ivory whether the strings were ter-
minated by being inserted in the top of the sounding box (in the manner of a harp), or
whether they passed over a bridge (on the side toward the musician’s body) and were attached
to the side or underside of the sounding box. Since the latter method might have interfered
with the playing of the instrument, as depicted, Brown4’ decided to insert the strings into the
top of the sound box, i.e. the sound board, by means of small wooden pegs which secure the
knotted strings.

1 cannot see why the other method should have “interfered with the playing of the instru-
ment” when, in fact, it is the only method attested by all the lyres which have been pre-
served or unambiguously represented in the ancient Near East. It is essential to the defini-
tion of the lyre that the strings run parallel to the sounding ~ G X ,passing over a bridge
which transmits their vibrations to the sound board, and that they be attached to the
bottom of the box. This was also noted by Professor Th. J. H. Krispijn, of Leiden,
Netherlands, who made a correct reconstruction of the Megiddo lyre which he demonstrated
at the University of Lewen, Belgium, in 1979.

Another attempt must be mentioned which, as far as the music is concerned, entirely relies
on Anne Kilmer’s theory. This is the study which was submitted at the Rencontre Assyrb
logique in Paris in 1977 by Hans Jochen ThieL4’ He does not offer an original musical
reconstruction, but presents an investigation of the rhythmic structure of the hymn. He
obtains rhythmical patterns and symmetries which seem at fmt sight very impressive. On
closer examination, however, several objections arise.

1) He postulates the existence of two hymns assembled on a iingle tablet, in

spite of which he cannot make use of all the Babylonian terms: at certain points he is
short of syllables, while at others, for instance between the two songs, a note without
words must be repeated as often as ten times.

2) To obtain his rhythmic symmetries he must alter the text. For example, he
must read the four syllables ni-&i-ru-SaZ as three, and the three of u-kur-ri as one!

3) The stanzas of his second hymn begin with the repeated words at the end of
the reverse-lines, which is a very strange place indeed for a hymn or stanza to begin.

4) Does the reconstructed pattern, 7-7-5-9-9, have any known parallel?

5 ) The music is strangely distributed. The second song begins with the repetition
of theme A, already used in the first song. This is unlikely, even according to Thiel’s
own theory, since the two songs differ in their rhythmic pattern. How, then, can they
use the same tune?

6 ) The author seems to have been fascinated by the digits to the extent that he
neglected the differential meaning of the musical terms.

40 Since the booklet was a collective publication, it W a s necessary to specify who was responsible for
the reconstruction of the lyre.
41 Hans J. Thiel, ”Der Text und die Notenfolgen des Musiktextes aus Ugarit”, Studi Micenei ed Egeo-
Anatolici 18, 1977, pp. 109-136.

[SANE 2, 801
 19841 The Discovery of Mesopotamian Music 19

7) In the latter he sees only dichords, as does Kilmer, instead of groups of notes,
the only solution compatible with the absence of polyphony in the East.

6.4 A New Interpretation

My own interpretation now follows. Professor Laroche’s remark on the division of the
sentences by means of the enclitic suddenly shed light on the problem. Starting from his
i
observation, I realized that the beginning of the musical themes should coincide with that
of the sentences.

It was necessary to start all over again from reliable elements. The mode given by the
colophon indicated the Cscale and, consequently, the place of each of the groups. They
could only be portions of the scale, not dichords (polyphony) or intervals, for a series
of jumps without intermediary notes would have resulted in an impossible melody. We
also knew from the British Museum tablet U7/80 that the names of fourths and fifths
designated portions of a scale which always had the semitone as their upper part. We were
aware of the directions (see arrows in Fig. 2) in which the portions of the scale ran. The
next step consisted of putting all these portions in a row without taking the digits into
account (see Fig. 9). In two instances I noticed awkward intervals between them: one
between irbute and nuat kubli, in line 8, was an augmented fourth, a tritone which was
considered impure by the Babylonian theoreticians; the other was a jump of a seventh,
upper B to lower C (line 6 between irbute and k % z t e ) , an interval which would be rather
difficult for the singem to perform. I concluded that the digits following each term must
serve to manage a transition between the successive portions of the scale and that they
probably represented notes added to avoid dissonance. Moreover, these digits also pro-
vided more notes to match the syllables of the hymn (cf. Fig. 6).

Two problems now remained:

1) Which additional notes were selected by the musician? Logically, they had to
belong to the portion of the scale designated by the preceding term, since otherwise
another term would have been used.

2) What was their place? This depended, on the one hand, upon their number and,
on the other hand, upon the necessity of avoiding dissonance. For example, number 1
b could not merely be a repetition of the previous note, for this would not have changed
anything to the dissonance. The solution had to be the simplest and easiest for the singers
to remember; it could very well be the last-but-one note of the group. Moreover, for the
digits 2, 3, 4, 5 and 10 the interval in which they had to move should not exceed the
interval of a second (i-e., two contiguous notes), because if the additional notes had
reached the interval of a third or more they could have been designated by one of the
expressions of the Philadelphia tablet. Above all, the method had to be the same for
all cases.

The interval of a second was the only one recommended by logic, simplicity and ease
of remembrance. Such an interval had to be designated by a digit, since there was no
term for a second in the Philadelphia tablet. In addition the digit 10, placed after the
20 M. Duchesne-Guillemin [SANE 212

third titim-iiurfe (line 5 ) could now be easily interpreted as a triZZ, which the ancient Greek
musicians called teretismos. The presence of the trill allowed us to recognize the music
as melismatic, that is, ornamented, not syllabic; each syllable could be spread over several
notes.

This opened up new perspectives. I approached the problem of structure in trying to

make the beginnings of sentences coincide with those of the themes (Fig. 10). The f i t
sentence, along with the f i t line of the notation (uncertain because of one unknown and
one erased word) I temporarily left aside. The second sentence could begin on theme A,
while the third one coincided with the intermediary development and the fourth began
with the reentry of theme A. I was greatly encouraged when I saw that theme B coin-
cided with sentences 5 and 6. Since the length of sentence 5 is approximately double the
length of qntence 6, and since the theme is a triple repetition, it follows that sentence
6 has to tally with the final repetition of Su&i (B, A, G) and SasSate (C, D, E, F, G, A).
This was the equivalent of nine notes, while the text ‘ni&msaZ &na &znuteti’ has 10
syllables-a proof that the digits designated additional notes.

The trill is not an isolated ornament; the groups of sixths and their additional notes p r e
vide melismatic passages in a rhythm alternating between ternary and binary, according to
the distribution of syllables. In fact, the rhythmic element is highly conjectural, since
we have no idea how Hurrian was accented. However, it seems safe and fairly plausible
to conclude that the beginning of each word coincided, as far as possible, with the
beginning of a musical term. Sometimes two short words naturally cojncide with one term
and its additional notes. When a syllable is sung on a single note I give it the value of
one beat; if it is extended over several notes I break it,down into smaller values. The
measure is free, which is usual in oriental music. The tempb is rather slow.

We are left with the fmt and third of the three difficult words mentioned aboveP2 At
the erased spot in line 5 there is room for two syllables, but not three. In searching
through the list of Human terms published by Laroche43 (pp. 484-485), I found that
the only disyllabic one was eSgi or iigi. Musically this hypothesis is plausible, for this
term designates the C D E third, which is included in the modal fourth nid qibZi (F E
D C).

As to the unknown word uitumuri, we have nothing to guide us except aesthetics. It

seemed to me that after a trill it was quite possible for a musical phrase to end on a long
note, whether it was repeated or not. Did the Hurrians have words for single notes?
They had quite a few terms of which the meaning is still unknown to us. One of these is
etumuieuni. which Laroche interprets as the Ugaritic equivalent of the string called “Ea
made it” in Babylonia. Since the latter term is used in the notation it may have desig- t
nated the note of the fourth string. Could we therefore say that uitamari designates the
note of the third string? This is merely a conjecture, and I have put it between brackets
r
in my musical transcription.

This transcription (Fig. 11) has four distinctive features:

42See above, p. 14.

43 See above, note 25. About etamajeani see note 31, P. 128.

[SANE 2, 821
 19841 The Discovery of Mesopotamian Music 21

1) It offers a thematic structure corresponding to the sentences of the hymn.

2) It uses melismata, and because of the groups formed by the sixths and their
added notes, the trill does not stand isolated.

3) It offers word endings on contiguous notes, or endings of the melody on

seconds, which is highly characteristic.

4) The rhythm fluctuates between binary and ternary passages.

i 7. COMPARISONS WITH JEWISH AND SYRO-CHALDEAN MUSIC

The overall result compares interestingly with the traditional Jewish music which has been
carefully preserved in the liturgy. The imposing corpus of Jewish s o n s collected by
Idelsohn between 1914 and 1932 in Jewish communities of the Near East and Europe
has helped a great deal in establishing this comparison.44 There is, among other examples,
a psalm of the Babylonian Jews (Fig. 12) in which:

1) the ambitus never exceeds seven notes, just as in the Hurrian hymn. It there-
fore seems to go back to a rather primitive period.

2) The structure offers three themes on a continuous text. The latter cannot
therefore be the cause of the musical repe4itions.

3) The grouping of the syllables produces passages which are slightly adorned on
alternate binary and ternary rhythms, with groups of six notes or more, as in our hymn.

4) There are endings on seconds comparable to the added notes in our tablet.

It seems reasonable to infer that the Babylonian tradition.survived not only in Mesope
tamia but in the whole Diaspora, as attested by other examples. The Hurrians, who had
become neighbors of Israel, may have acted as intermediaries long before the captivity
of the Jews in Babylon.

In the rendering of the reconstructed melody I have repeated the first phrase, on the
analogy of the Jewish songs; in our case the two pairs of small angle wedges on the
double horizontal dividing line may well signify this repetition. On the other hand, this
fmt phrase was probably sung by a soloist rather than by a choir, because of the trill.45

44 Abraham Z. Idelsohn, Tiresmrnrs of Oriental Hebrew Melodies, Paris-Berlin-Jerusalem, 1914-1932.

The example cited is in vol. 11, Cesinge der habylonixhen h&n,1922, n 95.
45A suggestion made to me by M. Marc Honegger, prkident de la SocK'te' de Musicologie de France,
to whom I wish to express my gratitude.
22 M. Duchesne-Guillemin [SANE 212

Further points of comparison are afforded by the old Christian music of Syria. This
was collected by Dom P a r i ~ o at
t ~ the
~ end of the 19th century and more recently by
Dom Jeannh4’ Although this collection was made only fifty years ago, it represents the
art of a small minority which clung to its tradition. It is diatonic music, often built on
three themes (Fig. 13); there are also ornamented alleluias on fluctuating rhythms, with
endings on second. Finally, I have found in the last part of a long composition a structure
(Fig. 14) remarkably similar to theme C in the Hurrian hymn; after a short introduction,
a double motif recalling the kitme-kablite pair is repeated three times and ends on an
abridged form of the same pattern. It appears that musical patterns go down the centuries.
The Gregorian chant, heir to the oriental tradition, still carries the echo of endings in
seconds. I am thinking of the “Veni Creator.”

The rendering recorded on the cassette was made by the male choir of the group of
Maurice Triaille, LiQge (Belgium). This music does not sound particularly strange to our
ears. We seem instinctively to recognize it, and I think it may be considered as part of
our ancestral heritage.

8. CONCLUSION

To quote Professor R. P. Winnington-Ingram, “It is really fascinating the way these

documents have turned up in succession and provided a progressive illumination. And it
is astonishing to find such a highly developed theory at this early date.” 48 The most
remarkable point is probably the fact that the Babylonian theory was based on a hepta-
tonic system similar to ours. Babylonian influence had reached the Mediterranean coast
and seems to have extended, through the Jewish and Syrdhaldean heritage, to early
medieval Europe, where it remained fundamentally unaltered until the invention of
polyphony.

How did Babylonian influence affect the Greek theory? This problem must be approached
with caution. With regard to musical instruments, it seems probable that the use of the
lyre spread to Greece and even beyond, for it is attested in the Halstatt culture of the
Iron Age. The Greeks called this instrument ‘kithara.’ 49 They did not doubt the oriental
origin of the kithara, but they did not look further east than Asia Minor. When, in my

46Dom J. Parisot, Rapport sur une mission scientifiue en lbrquie d’Asie, Paris, 1899, p. 234, n. 334.
i
4’Dom J. Jeannin, Me‘lodies liturgiques syriennes et chaldiennes, Paris, 1924, 2 vol. Examples cited
from voL 11, p. 10, n. 13, and pp. 46-47, n 64.
48 In a letter dated January 24, 1969, one of many exchanged with the British specialist in Greek
music.
49 The kitham was the concert instrument, the one used in competitions. The Greeks had also a more
simple instrument, made from a tortoise shell and the horns of an animaL They called it Zym. They
believed it to be the more primitive type, and that they had invented it. After the Sumerian discoveries,
many musicologists now believe the lyre to be a degenerate form of the kithara. Prehistoric instruments
are often taller than those attested later. In Ethiopia, both types (the big bagannu and the small kissar)
survive.
 19841 The Discovery of Mesopotamian Music 23

first article,50 I proposed to view the richly adorned SumereBabylonian instruments as

ancestors of the kithara, most specialists in Greek music (though not C. Sachs and 0.
Gombosi) were unconvinced or hostile. The chief objection was that the animal’s head
present on the Sumerian lyre was absent in the Greek kitham However, the Xanthos
excavations have yielded a relief depicting Apollo playing a kithara which was adorned
with a small animal on both arms (Pl. V,8). This proved a real connection with the
Philadelphia Lyre PI. 1,2). The use of a small stag in the decoration of the instrument
survived in Lycia as late as the 6th century B. C?, and, as recently put forward at the XXXth
Rencontre Assyriologique, Leiden 1983, on numerous Attic vases, see M. DuchesneGuillemin,
“L‘animal sur la cithare, nouvelle lumihe sur l’origine sumtrienne de la cithare grecque”, Orientalia
J. Duchesne-Guillemin emerito obhta (=Acta Iranica 23), 1984, pp. 129-142.

It now seems plausible that the Babylonian influence also included the method of playing
the instrument, its tuning, the arrangement and numbering of the string and the number
of the diatonic scales. However, the Greeh were not aware of this heritage; they had their
own myth about the invention of the lyre and they considered the Dorian mode to ‘x
their own national mode, just as the Babylonians thought that the iWtu mode was the
property of the land of Akkad: akkadi ki.52 There is a striking similarity between the
Dorian and the isartu modes in the arrangement of the string, beginning with the low-
pitched diatonic E. The name mesE and the numbering of strings from both ends argues
in favor of a practical technique adopted along with the instmment. Indeed, how could an
instrument have been borrowed without knowledge of the technique?

However, the Greek theoreticians based their speculations on the tetrachord and the octave,
rather than on the heptachord. Although they started from the diatonic system, the
Greeks appear to have reached stages which, as far as we know, were unknown to the
Babylonian theoreticians. They added the chromatic and enharmonic systems to the an-
cient diatonic one, and distinguished not only the seven modes but also the tonoi (i.e.,
positions in absolute pitch or transposed scales). Finally, the Greek system of instrumental
notation, which used letters of the alphabet in three different positions to designate each
note, was probably more practical than the Babylonian system, as far as we can judge from
the single preserved instance, our Hurrian hymn. We do not see, for example, why all the
groups of fourths are descending and those of fifths ascending (is this perhaps a reminis-
cence of tuning gestures?) The Babylonian system was abandoned, as was cuneiform writing
in general, perhaps for similar reasons.

L
My musical transcription of this hymn, in which I have tried to be logical and honest, has
character and even a certain beauty. However, it remains hypothetical. Perhaps our quest
is not yet at an end Our Human material contains many words which are still untrans-
latable and which are probably musically relevant. We must wait for further disc0veries.5~

“Sur l’origine asiatique . . .”, see above, note 1.

Pierre Demargne, Fouilles de Xanrhos, tome €, Les Piliers funiraires, Paris 1958, Plate XIII, relief n. 706.
While this study was going through the press I found the same animal, only with its head turned back-a
typical Sumerian motif-in numerous representations of kitharas on 6th century blackfigure Attic vases (cf.
my paper to the Rencontre Assyriologique, Leiden 1983, to be published in Acra Iranica 23, 1984).
52 See figure 3, line 45.
53 I am very grateful t o Professor Irkne Simon, of Likge University, for correcting the English of a h t
draft of this work.

[SANE 2, 851
24 M. Duchesne-Guillemin [SANE 212

A novel interpretation of the musical cuneiform texts is offered by Raoul Vitale, “La musique
sumCro-accadienne, Gamme et notation musicale”, Ugarit-Forschungen. 14, 1983, pp. 241-263.
It is based on the assumption that the front part of the instrument is the one facing the musician.
But this is contradicted by thlr presence of an animal’s head, obviously marking the front and held,
held, as can be seen on the Ur standard, away from the musician.

GLOSSARY

(Adapted, with due modifications, from C. Sachs, History ofMusical Instruments, New York, 1940, and from
Webster’s Dictionary.1
J
Accordatura: the tuning scheme of a stringed instrument (g d’a’e” is the usual accordatura of
a violin).
ambitus: the compass of a melody.
catch line: a line containing a catch word.
catch word: a word standing under the right hand side of the last line on a book page that
anticipates the first word of text on the following page.
chromatic: giving all the tones of the chromatic scale, consisting of twelve notes separated
by semi-tones.
coda: a final or concluding musical section that is formally distinct from the main
structure of a composition or movement.
colophon: an inscription usually placed at the end of a book, manuscript or tablet and
usually containing facts relative to its prodtiction, such as the scribe’s name.
consonant: agreeable in sound; specifically, harmonically satisfying, as contrasted with
dissonant.
diatonic: relating to a standard scale of eight sounds to the octave without
chromatic deviation. The succession of intervals is, for instance, in the C scale,
as follows: tone, tone, semi-tone, tone, tone, tone, semi-tone.
diaspora: the movement of Jews to areas outside Palestine.
dichord: a combination of two tones sounded together.
dissonant: disagreeable in sound, as contrasted with consohant. The second, the augmented
fourth and the diminished fifth are the dissonant chords; the last two are called
tritones.
enclitic: a syllable without independent accent, attached in pronunciation to a preceding
word.
enharmonic: in ancient Greek music: relating to that genus of scale employing quarter tones.
fifth: the musical interval embracing fme diatonic degrees.
fourth: the musical interval embracing four diatonic degrees. i
harp: a musical instrument in which the plane of the strings is perpendicular, not
parallel (as it is in the lyre) t o the sound board. The strings are attached to
the sound board, but run vertically away from it, and not along it. I
L

hep tatonic: composed of seven musical notes.

heterophony : a singing or sounding of the same melody by two or more voices or instruments,
usually with some modification (as in rhythm or ornamentation) by one or both
of the performers. The most natural heterophonic interval is the octave, owing

[SANE 2,861
 19841 The Discovery of Mesopotamian Music 25

to the difference in pitch between male and fern& voices. The Greeks admitted
also the fifth and the fourth, but not the third or the sixth, which occur some
times in modern popular oriental music.
lute: an instrument composed of a body and a neck which serves both as a handle and
as a means of stretching the strings beyond the body and of stopping them at -
different lengths, to vary the pitch.
&re: a stringed musical instrument made with a hollow body and two upright arms
that are joined at the top by a yoke; the strings run parallel to the body or
sound box, to which their vibrations are transmitted by means of a bridge.
With the Greeks, the lyra was a subspecies of the kithara, with the body made
from a tortoise shell.
melismatic: relating to or having melisma.
melisma: a group of notes or tones sung to one syllable in plainsong; a melodic embellish-
ment or ornamentation.
mesP: Greek for “median”. It was the basic note, the first to be tuned.
metabole: Greek for change.
mode: a musical arrangement of the diatonic notes or tones of an octave according to
one of various fmed schemes of their intervals.
modulation : the act or process of changing from one tonality to anotker.
monody: a melody sung by one voice or sung by several voices in unison.
pentatonic: consisting of five musical tones, without any semi-tones.
pentatonic scale: a musical scale of frve notes without any semi-tone, in which the octave is
reached at the sixth note.
polyphony: musical composition in simultaneous and harmonizing parts or vaices.
second: a musical interval embracing two diatonic degrees.
sixth: a musical interval embracing six diatonic degrees.
tetrachord: the basic unit of analysis in ancient Greek music consisting of a diatonic or
disjunct series of four notes or tones with an interval of a perfect fourth
between the first and last and distinguished by the relative position of the
semi-tones or quarter-tones in the series.
third: a musical interval embracing three diatonic degrees.
tritE: the Greek ordinal number“third”; the third string or note, counting backwards.
tritone: a musical interval either of three whole tones (an augmented fourth) or of a
semi-tone, two tones and a semi-tone (a diminished fifth). It is a dissonant
interval: the Babylonians considered it impure.

.
The recording that accompanies this text was made at Lihge
in 1975 by The Ensemble Vocal Maurice Triaille.

[SANE 2,871
 26 M. Duchesae-GuiUemin [SANE 212

Kabnitu XXSII column i

sa.di fore (string)
sa.u5 lies t (string)
sa.3 sa.sig third, thin (string)
sg.4.tur Sum. : fourth, sinall
{Akh.: r4 a-crest01
5. s a . d i . * 5 (test 4) ha-am-[h] fifth (string)
sa.4.a.ga.gul ri-bi ul~-ri-i[m] fourth of the behind (string)
s 8 . 3 . 8 . g n . g 11 1 Xal-Si tih-ri-im third of the behirid (st.ring)
1s a . 2 . a . g a’. g u 1 J’i-ni Bh-ri-im second of the behirid (string)
[s a . 11 . f a 1 . g a . g u I u‘h-ru-um thc behind-one (string)
10. [9]. s 3. a !J pi-it-nu nine strings
[ 1 s (Y> pi-is-mu ...
[ I i-fci r-ti ...
[ I [Cji-‘!iil-zir i-h--/uv) hritlge, . . .
[ 1 Ik i-i]t-mu cover
15. [ I [x y(z) k]i-il-mu . . . covcr
[ 1 [syj-urn ...
i 1 [sy-u]m ...
(remainder of col. i broken)

Fig. 1. U. 3011, cdumn 1. Sumerian-Akkadian lexical text ( = Nabnitu Tablet 32)

(Reprinted from Anne Draffkorn-Elmer, ‘“The strings of musical instruments:
their names, numbers and significance,”Studies in Honor of B. Landsberger. 1965, p. 264)

Fore “Thin” Ea Behind

c--------l
1st 2u 3 4 5 4. 3 2 1
BABYLONIAN HURRlAN TE’NTATIVE TRANSLATION

11 ni3 gaburf ? ?
12 S h i!aa&i song.. .
13 &rturn ismte straight (lyre), normal
14 salsatu Whte third
15 embhbu embube Pipe
16 rebiitu irbute fourth
17 nid qabli nit kablite .. .of the middle
18 SW esgi ?
19 qablitu kablite middle (mese)
20 titw qabli titar kabli bridge of the middle
21 kitmu kitme . . .covered
22 titur ihrtum titimiiizrte bridge of iSartum
23 pitu ? .. .opened?
24 serd (last reading) zirte ?

c d

Fig. 2. Graphic interpretation of Tablet CBS 10996

(See also Pl. III,4)
[SANE 2, 881
19841 The Discovery of Mesopotamian Music 27

45 23 inZtu k e-&-re Akkaca KI ( e h e = Isarturn) 23 love songs of the iSartu (type)

46 17 imru k ki-it-me 17 love songs of the kitmu (type)
47 24 iniruiueb-bu-be 24 love songs of the.embtibu (type)
48 4 i ~ t &pi-i-re
u 4 love songs of the pitum (type)
49 [ . . .lircStu k ?&it,&MURUB (MURUB = qablim) -.. . love songs of the nid qablitu (type)
50 [. - .]irCrru kai-*i3 GABN . . . love songs of the nis gabari (type)
51 [..-]ir&sa~m~~-re . . .love songs of the qablitu (type)

Fig. 3. Tablet KAR 158 VlIt VAT 10101

U.7180

Fig. 4. Fragment U 7/80 (UET W 74)

(Courtesy of the Trustees of the British Museum. Reprinted from Gurney, Iruq 30,1968, pp. 229-233)

[SANE 2,891
28 M.Duchesne-Guillemin [SANE 212

NID YURU.
uItum

embubu
kitmu

NU SU

I W

:
I I 0

.
I I]
pltum
enlbobu

NI8 GABPK
WID Y V R U .
-

Fig. 5. Explanation of Metaboles by M. DucheneGuPemm.

(From M. Duchesne-Gudlemin,“La theorie babylonienne des metaboles
musicales,’’ Revue de MusicoIogie LV, 1%9, facing p. 8.)

[SANE 2,901
19841 The Discovery of Mesopotamian Music 29

HuS Pub-li-re! 3 ii-bu-re 1 tub-li-re 2 ? X - X - x [ti~-ti-mi-.fur-te 10 uZ-ta-mua-ri

6 ti-ti-mi-Jur-te2 zi-ir-te 1 Ju-lalb-ri 2 x-x-& 2 ir-bu-re 2
7 cm-)ri-bo 1 311-uJ-h-re 2 ir-lu-te x [3]u-[uEJja-t[r] x ti-tur-kab-li 1 1i-timi-hr-te4
8 a - i r - r c 1 lu-ub-ri L’ Ju-uJ-Jo-t[r]4 ir-hi-tc 1 nu-at-ku&-li 1 h u b - r i 111
9 Ju-cr.-.in-re 4 Iu-ub-ri 1 iu-a&I[u-t]e2 h-ub-ri 1 3u-crI--h-te L’ ir-[/+-[r.] 2
10 ki-it-ntr L’ kuli-li-te 3 ki-it-[me] 1 Puldi-re Pi-it-me 1 PuLli-te .i?
________ - - -
11 ni-it-Pib-!i zu-[lu-zi
lun-nri]-ri ,-u-rcni-ni~r-cr.t-h-.~~~ 1x SU mAnr-niu-ru-bi

Fig. 6. Human Tablet

E. Laroche’s transcription, showing text above,
musical notation underneath, with em-bu-beemended by M. DuchesneCuillemin
(From E. Laroche, “Documents hourrites provenant de Ras Shamra,” Uguriticu V, 1968, p. 487)

Ir . r 1 r .I r r r l - b - n -..c b - * t . t u

.-WR.r~wtlm..lUI B- ~k1-4

Fig. 7. A. Kilmer’s interpretation, with musical themes A, B, C added by M. Duchesne-Guillemh

(From E. Laroche, “Documents hourrites provenant de Ras Shamra,” Ugmiticu V, 1968, p. 463)

[SANE 2, 911
30 M.Duchesne-Guillemin [SANE 212

Fig. 8. Megiddo Ivory

(From Anne Kilmer, Sounds from Silence, 1976, fig. 28; Bit Enki drawing based on photo.
The Rockefellar Museum, Jerusalem)

Fig. 9. Interpretation of musical terms, with digits left out

[SANE 2,921
 19841 The Discovery of Mesopotamian Music 31

4 I l ’ r i a l t a l i ) t i J i a U W I ~ Ik a p
tili unitgat rkli ktnhmmr 1x1-lil
r

Fig. 10. Concordance of Hurrian text and musical themes on the Ras Shamra tablet.
(From M. DuchesneGdlemin, “Dkhiffrement de la musique babylonienne,”
Accademia dei Lincei, Quaderno 236, 1977, p. 15)

Fig. 11. Transcription by M. DuchesneGuilIemm

(From M. Duchesne-Guillemin, “D6chiffrement de la musique babylonienne,”
Accademia dei Lincei, Quaderno 236, 1977, p. 21)
32 M.Duchesn&uillemin [SANE 212

* F&. 12. .*matic rnalysi~of a ~cwishpsphn

(Reprinted from k Z. Idelsohn,Gem"nge der babylonicchen Juden.

-
Thesaurus of Oritntol Eebnw MeIodies, Vol. U, 1922,no. 95)

1. ? 1.
ILLUSTRATIONS
 Pi-
@
3
Y
E
I1 The Discovery of Mesopotamian Music [SANE 212

Ill. 3. Tell0 relief

(Courtesy of the Muse'e du Louvre)

[SANE 2, VI]
 19841 M. Duchesne-Guillemin

CBS 10996 Translation of CRS 10996

Obv. col. I (at least 5 lines broken away at beginning) Obv. col. I (five or mote lines broketl)
r21, r.ti TS.4 .T xl-tum r2,i r4i .. . .string
J. 3 SA kit-mu 1, 3 " covered " string

3.6 SA ti-x i-Sar-tum 3,6 . " . . .upright " string

7, .I sn- ti-lum 7, 4 " first " strinx

10'. J , 6 sa-mui-lwn 10'. 4, 6 * I' second(?)" string

SA qud-mu-ti ri SA 54ti 1.5 SA man-ga-ri fore string and fifth string 1, 5 " reed basket strinx
"

S.4 3 u k r i d SA 544 7,5 SA SAR.KfGIN .

third-behind string and fifth string 7, 5 . . . strinK
SA fa-CE, d SA 4 uh-ri 2, 6 SA i-Lr-tum second string and fourth-lwhind string 2, 6 upriRlit " string
"

SA qud-mu-ti d SA d uh-ri 1,6SA Sul-Sd-tum fore string and fourth-behind string 2, 6 third " string
"

15'. SA 3(!)-fd SIC li S.4 3 4 3 uk-ri 3 . 4 SA em-bu-bib 15'. third thin string and third-hchind string 3,4 " flute " string
';
3 SA Sd-GE, 4 SA 3-84 uh-ri 2, J SA 4-fu
SA 't-a-DfJ d SA qud-mu4 4 . 1 SA NIM.MfTRUB
second string and third-behind string 3, 4 fourth " string
" Ea-made(-it) " string ant1 fore string . .
1, 1 .. string
fore string and third thin string 1, 3 " Elamite " string
i SA qud-mu4 ri SA 3-Sd SIC I , 3 SA CIs.NIM.MA
SA MrN 3-53 d S.4 Sli-CE, 5, 2 SA MURUB-lu . . . .string and second string 5 , 2 " middle " string
20'. SA Jli-GE, h, SA di-a-L)o 2 , 4 SA fx1 MCRWB-tri 20'. second string and " Ea-made(-it) string 2,4 " ... .middle" string
"

S.4 4 nb-ri 4 SA 3-Sti SIG 6 , 3 S[A fourth-behind string and third thin string 6, 3 L 1I

SA 3 - ~ 3SIC ic SA 5 - s ~ r31, third thin string and fifth string rQ1. { 3

SA 3-hi uh-Ti ri [ tllirtl-behind string and [ 1
SA 't-a-DU'[ " Ea-made-it) " string I
(remainder of col. broken) (remainder broken)

111. 4. Nippur Tablet CBS 10996

(Courtesy of the University Museum, University of Pennsylvania.
Reprinted from Anne Kilmer, "Two Lists of Key Numbers", Orientuliu 29, 1960, pp. 278 and 281)

[SANE 2, MI]
IV The Discovery of Mesopotamian Music '
[SANE 212

Ill. 5. Ur standard
(Courtesy of the Trustees of the British Museum)

Ill. 6. Shell plaque from Ur

(Courtesy of the University Museum, University of Pennsylvania)

[SANE 2, VIII]
19841 M. Duchesne-Guillemin V

Ill. 7. Giant lyre-kithara from Ischali

(Courtesy of the Oriental Institute, University of Chicago)

Ill. 8. Funerary pillar, Xanthos

(Courtesy of P. Demargne)

[SANE 2; 1x1
 The Discovery of Mesopotamian Music . [SANE 212

111. 9. Human Tablet, copy by E. Laroche

(Reprinted from E. Laroche, "Documents houmtes . . .", Ugaritica V, 1968, p. 487)

,
.. -.
"-"--I .---
.. . ...- -~
-.*

IU. 10. Human Tablet, photographed by A. Kilmer in the Damascus Museum

(Courtesy of the Damascus Museum. Reprinted from Anne Kilmer, Sounds from Silence, 1976, cover)

111. 11. Human Tablet, copy by A. Kilmer

(Courtesy of Anne Kilmer. Reprinted from Sounds from Silence, 1976, p. 12)

[SANE 2, XI