The Kingdom of Meroe straddled the Nile in what is now known as Nubia from as far north as Aswan in Egypt to the present–day location of Khartoum in Sudan (see Map 1). Its principal language, Meroitic, was not just spoken but, from the third century BC until the fourth century AD, written as well. The kings and queens of this kingdom once proclaimed themselves pharaohs of Higher and Lower Egypt and, from the end of the third millennium BC, became the last rulers in antiquity to reign on Sudanese soil.
Centuries earlier the Egyptian monarchs of the Middle Kingdom had already encountered a new political entity south of the second cataract and called it “Kush.” They mentioned the region and the names of its rulers in Egyptian texts. Although the precise location of Kush is not clear from the earliest attestations, the term itself quickly became associated with the first great state in black Africa, the Kingdom of Kerma, which developed between 2450 and 1500 BC around the third cataract. The Egyptian expansion by the Eighteenth Dynasty (1550–1295 BC) colonized this area, an occupation that lasted for more than five centuries, during which the Kushites lost their independence but gained contact with a civilization that would have a lasting influence on their culture.
During the first millennium BC, in the region of the fourth cataract and around the city of Napata, a new state developed that slowly took over the Egyptian administration, which was withdrawing in this age of decline. From 750 BC onward, all of Nubia would be Kushite again, and, in 732, the Kushite king, Piankhy, subjugated both Nubia and Egypt to Amun, the central deity of Napata and Thebes. His brother and successor, Shabaqo, is considered the first pharaoh of the Twenty-fifth Egyptian Dynasty, also called the “Ethiopian” dynasty after the Greek name for Sudan, which is today referred to as “Kushite.” This domination by Egypt lasted until the
| Image not available in HTML version |
A new era arrived without political or cultural upheaval during which “Kush” came to be ruled by the kings of Napata. It is there, not far from the holy mountain of Gebel Barkal, that the new monarchs were buried in the pyramids of the Nuri necropolis. The main events of this history can be understood with the help of the local inscriptions that were written in the Egyptian language and script.
About three hundred kilometers southwest of Napata, where the city of Meroe had already existed for nearly two centuries, a new development took place around 300 BC. It was not a clear cultural break with the previous period, and Napata remained the religious metropolis. Instead, the Kingdom of Meroe gradually distanced itself from Egyptian influences. There arose an official cult with long-known indigenous deities such as the lion god, Apedemak. They developed a local writing system for the Meroitic language, which, up until then, had not been written, apart from geographical and proper names transcribed in Egyptian script.
The end of this Meroitic kingdom came about in the fourth century AD; the circumstances of its demise are not clear to present-day historians. The power of the kingdom was lost, or perhaps divided, with the onslaught of the Noba, the ancestors of the Nubians of later days, and through incursions from the Ethiopians of Axum. By the time the Byzantine missionaries introduced Christianity in the sixth century AD, the old territory of Meroe had been divided into three kingdoms ruled by Nubian elites: Nobadia in the North, Makuria in the center, and Alodia in the South. Under Christianity, Meroitic civilization eventually disappeared from the historical record.
In 1819, the French-German architect Franz Gau, unknowingly, became the first to copy a Meroitic text when reproducing an inscription from the temple of Dakka. Frédéric Cailliaud from Nantes in France, passing by the great temple of Soleb in 1821, discovered an inscription he recognized as “Ethiopian,” that is, Sudanese. But most important of all, through his long expedition from 1842 to 1845 through Nubia, is the great Prussian Egyptologist Carl Richard Lepsius. He collected a number of texts, which he published in the Denkmäler aus Aegypten und Nubien. Although he was unable to decipher the script, he proposed that it was some form of ancient Nubian, a language to which he had dedicated considerable research with the publication of Nubische Grammatik in 1880. Between 1874 and 1881, there appeared studies of Nara (Barya), Beja, Nubian, and Kunama that became the first descriptions of African languages of their time. One can safely state that African linguistics is indebted to the research instigated by Lepsius in his attempt to reach a translation of Meroitic.
After the unfruitful attempts of Heinrich Brugsch (1887) and Archibald H. Sayce (1911), the decipherment of the Meroitic script became the work of the British Egyptologist Francis Llewellyn Griffith. After collecting at least a hundred texts, he made a systematic comparison. With the help of parallel versions of Egyptian and Meroitic, he found that Meroitic hieroglyphs were written in the opposite direction, a characteristic that had escaped his predecessors. The value of the signs was deciphered, but the vocabulary remained incomprehensible apart from names of people and deities, titles, and place-names. His philological method, based on a detailed examination of each text and cross-checking commonalities between one text and another, would have awarded him much progress if he had not been equally occupied with his immense and fruitful Egyptological work.
A few decades later, in 1930, the Austrian Egyptologist Ernst Zyhlarz published “Das meroïtische Sprachproblem,” in which he made many phonological, syntactic, and morphological comparisons. He was a student of Carl Meinhof, who suggested that certain African languages, one of them Meroitic (Meinhof 1921/22), were part of a group called “Hamitic” and not just “negroid” languages that were primitive. Zyhlarz confirmed the classification “Hamitic” for Meroitic and is thought to have found traits that they had in common. He analyzed a Meroitic inscription of Philae (REM 0101) and boldly presented an approximate translation. After the death of Griffith in 1934, it would not be until the 1950s that more serious attempts at understanding Meroitic would be undertaken.
It was after the Second World War that new research was launched. The main figure was the German Egyptologist Fritz Hintze, who would dominate Meroitic studies for the next thirty years. He sought to apply models inspired by generative linguistics through comparisons of funerary texts. Although it was admirable work, the results in terms of translations were rather limited.
During the sixties and seventies the construction of the Aswan Dam threatened Nubian monuments, and UNESCO launched a salvage campaign. Research internationalized, and specialized journals were set up to document the progress (Meroitica in Berlin, Meroitic Newsletters in Paris). The Canadian Bruce G. Trigger applied the methods devised by Joseph Greenberg to the study of Meroitic and classified the language as “Nilo-Saharan,” a newly created language group, rather than “Hamitic.” Hintze showed, however, that the translations of the Meroitic words that were used by Trigger had frequently been inaccurate and based on the unreliable work of Zyhlarz. Then the Frenchmen Jean Leclant and André Heyler launched a project for the long term. They entered the texts into computers under the name REM (Répertoire d’Épigraphie Méroïtique). The death of Heyler in 1971 stalled the project, which was not revived until recently (Leclant et al. 2000). A detailed catalogue has been the sole result, while the transliterations are still waiting for attention.
Only a Viennese group that convened around Inge Hofmann and the Groupe d’Études Méroitiques in Paris have continued to publish regularly on Meroitic. Elsewhere, progress has been slow, no new researchers have entered the stage, and no new inscriptions have been published in sufficient quantity to break new ground. From 1991 to 2007, there were no monographs on the Meroitic language. But since then, two volumes with the latest discoveries have been produced in French; they have reinvigorated the discipline and should inspire future research. This English-language volume is a synthesis of these two works and a handbook for those interested in the field.
The Meroitic language, despite its name, was not limited to the Kingdom of Meroe. It was previously thought that the language appeared in some form during the first Kingdom of Napata at the beginning of the first millennium BC and that it disappeared in the fourth century AD. Hofmann (1981a, 1981b) sets the birth of this language during the Eighteenth Dynasty. Instead, it has recently been demonstrated that Meroitic appeared in one form or another as early as the third millennium BC at the time of the Kingdom of Kerma.
When the Egyptians entered Nubia during the Middle Kingdom, circa 2000 BC, they encountered two types of languages. The rendering of the names of rulers that appear in the Egyptian texts (Posener 1940, Koenig 1990) indicates that one of these languages concerns the Medjay. Their language has a relatively high number of consonants, similar to what is found in Egyptian. The other is “Kush,” which has only a dozen consonants. The “Kush” names also do not have a single fricative except /s/. Although the corpus is limited, the two languages can be tentatively classified as Hamito-Semitic (Berber or Cushite), in the case of Medjay, and Nilo-Saharan (Eastern Sudanic), in the case of “Kush.” Unfortunately, linguists have used the term “Cushite” for a group of languages unrelated to “Kush.”
There is clearer evidence of early Meroitic in a document that was written in hieratic, the Golenischeff papyrus (cf. Erman 1911, Vernus 1984). It dates to the end of the era in which the Hyksos reigned in northern Egypt (around 1570 BC) and contains a list of fifty-seven foreign names carefully transcribed in syllabic script. The syllabic signs that were used made it possible for the scribes to enter the correct vowels in these non-Egyptian words. The resulting words are clearly proto-Meroitic, not only because they can be easily recognized through the original consonant system but also because there are entire words that contain names of two Meroitic deities: Mash (sun god) and, perhaps, Apedemak (creator god).
Contrary to the opinion of Erman in 1911, this list is not a record of slaves, since in such a case the names were not likely to be given in such phonetic detail. They were more plausibly ambassadors on a mission. The papyrus in question is from Crocodilopolis in the Fayum, a starting point on the string of oases that connected Lower Egypt and Kush. The ruling Hyksos had excellent relations with Kerma because both were opposing the Seventeenth Dynasty in Egypt. Since the Kingdom of Kerma was founded around 2450 BC and did not undergo dramatic cultural changes until the Hyksos period, the Meroitic language is likely to have existed since the third millennium.
From the Eighteenth Dynasty onward, the presence of the Meroitic language in Nubia quickly expands in the Egyptian texts. There is a servant’s name that reads, “beautiful is the Kushite” (Hofmann 1981b); there is an ostracon from the Rameses era (eleventh century BC) that states that “on the land of Kush” a goddess is given a name that contains the radical mk, meaning “divinity.” Ultimately, from the same period (see Zibelius-Chen 2005), there are supplementary Chapters 163–165 of the Book of the Dead with entire phrases in Meroitic that can, at least partially, be understood despite the Egyptian phonological rendering of the words.
During the Twenty-fifth Dynasty and the Kingdom of Napata, there remained some traces of Meroitic in proper names even though the administrative language was Egyptian. In the reign of Aspelta (around 600 BC), a rather interesting practice emerged. The transcriptions and hieroglyphs of certain Meroitic lexical elements appeared in proper names, for example, mlo “good,” mk “divinity,” and mte “child.” They were accompanied by determinatives or Egyptian ideograms that corresponded with their meaning: the heart and the trachea for mlo, the flag for mk, and the figure of a child sucking its thumb for mte.
From the second century BC onward, the development of the language can be traced in its entirety through the Meroitic writing system. Contrary to the Egyptian case, the cursive writing appeared first, probably to fill a commercial and administrative need. Inspired by Demotic writing, a script commonly used in Egypt during this time, Meroitic reads from right to left. The oldest examples are from Dukki Gel, near Kerma, and were used for pious graffiti on the walls of the temple of Amun. Not much later, toward the end of the second century, this script was enlarged with a hieroglyphic version reserved for religious and royal texts. The signs were taken from Egyptian, but the sound values in Egyptian were changed. For instance, Image not available in HTML version, which corresponds to Egyptian s3, is read as /ka/ in Meroitic. Also, the reading direction is opposite to that in Egyptian with the signs facing the end of the sentence rather than the beginning.
Table 1.1: The Meroitic scriptHieroglyphic
Cursive
Transliteration
Value
Image not available in HTML version
Image not available in HTML version
a
initial /a/ or /u/
Image not available in HTML version
Image not available in HTML version
b
/ba/
Image not available in HTML version
Image not available in HTML version
d
/da/
Image not available in HTML version
Image not available in HTML version
e
/e/, /ә/, no vowel
Image not available in HTML version
Image not available in HTML version
x
/Ãa/
Image not available in HTML version
Image not available in HTML version
h
/Ãwa/
Image not available in HTML version
Image not available in HTML version
i
modifier /i/
Image not available in HTML version
Image not available in HTML version
k
/ka/
Image not available in HTML version
Image not available in HTML version
l
/la/
Image not available in HTML version
Image not available in HTML version
m
/ma/
Image not available in HTML version
Image not available in HTML version
n
/na/
Image not available in HTML version
Image not available in HTML version
ne
/ne/, /nә/ or /n/
Image not available in HTML version
Image not available in HTML version
o
modifier /u/
Image not available in HTML version
Image not available in HTML version
p
/pa/ (?) or /ba/
Image not available in HTML version
Image not available in HTML version
q
/kwa/
Image not available in HTML version
Image not available in HTML version
r
/ra/
Image not available in HTML version
Image not available in HTML version
s
/sa/
Image not available in HTML version
Image not available in HTML version
se
/se/, /sә/ or /s/
Image not available in HTML version
Image not available in HTML version
t
/ta/
Image not available in HTML version
Image not available in HTML version
te
/te/, /tә/ or /t/
Image not available in HTML version
Image not available in HTML version
to
/tu/
Image not available in HTML version
Image not available in HTML version
w
/wa/
Image not available in HTML version
Image not available in HTML version
y
/ya/
Image not available in HTML version
Image not available in HTML version
:
separator
Whether cursive or hieroglyphic, the Meroitic script is not a system of consonants and ideographic signs as in Egyptian. Instead it is a syllabary with a default vowel /a/, also known for the Brahmi script in India, for instance. The following serves as a first introduction to the system of writing and an overview of its signs (see Table 1.1).
The system of writing is known as alphasyllabic, or abugida. Each consonant has an inherent vowel, in Meroitic read as /a/. Therefore Image not available in HTML version, normally transliterated as k, is read /ka/. For another vowel value, it is necessary to add a vowel modifier: Image not available in HTML version for /e/, Image not available in HTML version for /i/, Image not available in HTML version for /u/ (and perhaps /o/). This system is practical and efficient for sequences of the type CVCV (CV = consonant followed by vowel). But, as in certain Indian scripts, a problem presents itself for word-initial vowels and consonant clusters. Historically, the Meroites have chosen to use the modifier Image not available in HTML version e when representing a consonant without a vowel, but this may lead to obvious ambiguities.
There are four supplementary signs that do not have a vowel /a/. They have a fixed vowel value and cannot be followed by a vowel modifier: Image not available in HTML version ne, Image not available in HTML version se, Image not available in HTML version te, Image not available in HTML version to. These syllables correspond to common morphemes in the Meroitic language, a correspondence that explains their resistance to conforming to the norm.
Word-initial vowels are more complex in the system. The transcriptions have evolved strongly in the course of the centuries. In short, the sign Image not available in HTML version is transliterated a and serves to write both initial /a/ and /u/. The initial vowels /e/ and /i/ are written with the corresponding vowel modifiers Image not available in HTML version e and Image not available in HTML version i in combination with Image not available in HTML version y. So the group Image not available in HTML version ye and Image not available in HTML version yi with the Image not available in HTML version y in initial position are graphic rather than phonetic representations.
Finally, the Meroitic script has a separator sign transliterated “:” which allows different words and word groups to be separated, but this is seldom done systematically.
The Meroitic writing system is a local creation, probably inspired by the Egyptian “syllabic orthography.” Its influence is not Semitic, Greek, or Geez. In the first two cases, the principle of writing is completely different, and in the case of Geez, which resembles the system of Meroitic much more, the shape of the signs is unrelated. Also the date of its first appearance favors a Meroitic invention.
The Meroitic script continued to be used until the fifth century AD, that is, if the latest dating is taken to be correct. The inscription of King Kharamadoye (REM 0094), inscribed on a temple column of Kalabsha in Lower Nubia, is considered the latest attestation of Meroitic writing. It was thought to date to the fourth century, but a new analysis of the succession of kings (see FHN III: 1103–1107) by László Török has now placed the inscription around AD 420. The orthography is curious, with archaic forms and assimilations that characterize a script in decline. It is possible that an ancient stele served as a model but that only part of the texts was used.
The end of the Meroitic language and writing system is as little understood as the end of the Meroitic Kingdom. There are, however, two or perhaps three signs that appear in the Old Nubian alphabet: the nasal palatal Image not available in HTML version [ñ] that derives from Meroitic Image not available in HTML version ne; the Image not available in HTML version w that is from Meroitic Image not available in HTML version w; and Image not available in HTML version [Ñ], the velar nasal that is either the Meroitic Image not available in HTML version x or a modified Greek gamma.
The oldest documents using the Old Nubian script date to the end of the eighth century AD, indicating that the Meroitic signs continued to be read at least two centuries after the Kharamadoye inscription. Perhaps one day, in Sudanese archaeology, other evidence will fill in the gaps in our understanding of this history.
The phonetic values of the signs are, for the most part, reconstructed on the basis of two types of sources: the Egyptian transcriptions and Greek and Latin renderings of Meroitic words, mostly place-names and proper names, as well as the Meroitic transcriptions of Egyptian words, mostly loanwords, in addition to some rare Greek and Latin words.
For example, the name of the king Teqorideamani is written Teqoride-mani in Meroitic and transcribed Tqrrmn in Egyptian Demotic (Philae 416). The /d/ in intervocalic position in Meroitic may have had a retroflex character (see Rowan 2006:61–69) and resembles acoustically an /r/ for the speakers of a language such as Egyptian that does not have that type of consonant. This confusion is absent for the Meroites, who never confuse Image not available in HTML version d and Image not available in HTML version r in their transcriptions. There is, therefore, an opposition of /d/ and /r/ of which the first is spoken as [Ç], and the other as [r]. Conversely, the name of Caesar appears in Meroitic as Kisri, which would be rendered /kaisari/. The “general (strategos)” is pelmos and becomes /pәlamusa/ or /bәlamusa/, in Late Egyptian p-lmš‘. These two transcriptions show that the Meroites did not make a distinction between [s] in Caesar, transmitted no doubt via Greek, and the [ƒ] of Late Egyptian. They had only one phoneme /s/ in free variation with [ƒ]. These hypotheses are, of course, confirmed by other similar transcription examples.
On the basis of many such examples, a phonology of Meroitic consonants can be proposed (see Table 1.2). Nevertheless, some elements remain tentative and many points need clarification. Among others, the existence of the phoneme /p/, proposed by earlier scholars, is doubtful because of its absence in proto-Meroitic texts. The frequent permutations with /b/ in the texts and its occurrence mainly in word-initial position make it suspect. It may have been borrowed from Egyptian, where the Egyptian article p3 before nouns may have influenced its preference for initial positions.
The certainties about the vocalic system are less spectacular. The absence of a sign for /o/ is most surprising. The phonemes are limited to /u/, /ә/, /a/, /e/, and /i/. The central vowel /ә/ was perhaps not originally there. In the middle of the first century AD, an important phonetic change took place. In many cases the /a/ became centralized and often mute when it was in initial or final position.
Table 1.2: The Meroitic consonant systemLabials
Dentals
Palatals
Velars
Labialized velars
voiced stops
/b/ [b]
/d/ [d] ~ [Ç]
voiceless stops
/p/ [p] (?)
(< Egypt. ?)
/t/ [t]
/k/ [k]
/q/ [kw]
fricatives
/s/ [º]
/x/ [Ã]
/h/ [Ãw]
approximants
/l/ [l]
/y/ [j] (?)
trills
/r/ [r]
nasals
/m/ [m]
/n/ [n]
It is likely that there were two diphthongs, probably /auò/ and / aiò/, of which only the second part was written. Long vowels were transcribed in Meroitic with the use of y and w; hence, eyi is a group that is read as /iú/; owo or oyo then makes /uú/.
More than eleven hundred Meroitic texts have been uncovered up until today, and each year new discoveries appear from archaeological sites. They are classified with a number in the Répertoire d’Épigraphie Méroïtique, or REM (Leclant et al. 2000). The writing surfaces are of different kinds, mostly stone, but also ceramics (ostraca), papyrus, wood, skin, et cetera. The following paragraphs describe the different types of text, each with its own characteristics, as well as the present understanding of their composition and meaning.
These constitute more than half of the recovered texts and generally follow a stereotypical scheme that has been well documented and analyzed by Griffith (1911a).
In the vast majority of the funerary inscriptions, the invocation, addressed to the gods, is at the beginning of the text using a vocative-marker -i: Wos-i : “O Isis,” (A)sorey-i : “O Osiris.” In a few cases, this invocation is absent (3 percent of the cases in the texts from Karanog, Shablul, and Faras) or placed after the name of the deceased (REM 0331, 0525). This proves its independent syntax in relation to the rest of the text; therefore, Griffith identified this group of words as a vocative. The invocation is often repeated elsewhere in the funerary text, particularly at the end.