Of the languages of the ancient Near East, one has the odd distinction of having been spoken almost everywhere and written in almost nowhere, until suddenly it was written everywhere. Aramaic began as the speech of a scatter of Semitic peoples in the Syrian uplands and the Euphrates valley, the tongue of some minor city states that the Assyrians swallowed and deported. Within two centuries it was the working language of an empire, the written vehicle of its diplomacy and its accounts, the script of its clerks, and the medium in which scripture was read and taught long after the empire that had spread it had gone. It then did something rarer still: it outlived every political circumstance that had made it important, and when those circumstances had all ended, it was still the language in which a man in Baghdad said his prayers, in which a boy in a Galilean village said the prayer that his mother had taught him, and in which a congregation in Kerala in the twentieth century heard the psalms read aloud in a tongue that had not been spoken there since the seventh century.
This essay is an attempt to tell the biography of the language itself. That is a different undertaking from the study of any of its textual corpora, and this journal has already published a good many of those: the Aramaic Pentateuch and the targumim, the Peshitta and its Hebrew Vorlage, the Syriac New Testament, the Armenian Bible and the alphabet that Mesrop Mashtots built from it, the Elephantine papyri, the Genesis Apocryphon, the Nippur tablets of the Murashu house. Those are histories of books. What follows is a history of a tongue: where it came from, how it was lifted into an empire, what happened to it when the empire stopped existing, how its letters escaped to become the letters of half the world, what it did for scripture, and how it is still alive in the villages of the Zagros and the Jabal al-Druze in a form that no longer resembles the Aramaic of Daniel by very much at all.
It is worth saying at the outset what “Aramaic” does and does not mean, because the word covers a great deal of ground and the distinctions matter. It can mean the old Aramaic of the Aramean states; it can mean the official Aramaic of the Achaemenid, Parthian and Sasanian empires, which was a bureaucratic koine written in a standardised hand; it can mean the Jewish Aramaic of Babylonia and Palestine, in which the targumim were made; it can mean Syriac, which is the most literary of the Aramaic dialects and the only one with a substantial Christian literature; it can mean Mandaic, the language of the Mandaeans of southern Iraq; it can mean the small, archaic literary dialects of the Jews of Galilee in the early Roman period; and it can mean the Neo-Aramaic languages of the present, a continuum of village dialects stretching from Sanandaj in western Iran to the mountains of Kurdistan. These are not one language with a family tree but rather one language with a long history of branching, and the branches have had very different fates. The imperial dialect went out of use as an official language within a few centuries of the Arab conquests. Syriac became the liturgical language of a church. Mandaic survives as the ritual speech of a community of perhaps a few thousand baptised adherents in Iraq. And the Neo-Aramaic dialects, which nobody ever used for an imperial purpose at all, are the ones that have quietly carried on.
The Arameans and the Language They Brought
The Arameans were not a nation in the sense that the Assyrians or the Babylonians were. They were a set of related Semitic-speaking peoples who occupied the arc of country running from the upper Euphrates down through the Balikh and Khabur valleys and eastwards towards the desert, and who appear in the Assyrian records as a dozen separate polities rather than one. Aram-Damascus was the largest of them and the one with the longest reach, its kings ruling from the middle of the ninth century BCE and resisting Assyria for a century and a half. There was a state called Aram-Soba, named after its capital, and one called Bit Adini, and there was Hamath. All of them were absorbed, and absorbed early, and the manner of their absorption is the single most important fact in the history of the language.
The Assyrian conquest of the Aramean states was not a single event but a campaign conducted across nearly a century, and its most important mechanism was not the army but the deportation. From the reign of Tiglath-Pileser III in the middle of the eighth century BCE, Assyria developed the mass transfer of conquered populations as a deliberate instrument of imperial policy, and the transfers were carried out on a scale that is difficult to overstate. The population of the northern kingdom of Israel was removed after the fall of Samaria in 722 BCE and resettled across the Assyrian provinces, mixed in among other deported groups. The Aramean states were picked off in turn. The policy had an immediate linguistic purpose, whether or not it was conceived in those terms: if you move a people away from their land and mix them with others, their language stops being the vehicle of their particular political loyalties, and the imperial language becomes the only language with any future. Assyria was, in effect, the first state to run a deliberate language policy, and it ran it in Aramaic.
The mechanism by which Aramaic won is more specific than that, and it has to do with the peculiar situation of Akkadian. Akkadian, the cuneiform language of Sumerian and Akkadian scribal tradition, had been the written vehicle of Mesopotamian administration for well over a millennium. It possessed what no Aramaic scribe could match: an enormous, exquisitely trained, self-perpetuating scribal establishment with its own apprenticeships, its own connoisseurship, its own scholarly canon, and a written literature of such standing that it could not simply be displaced. When Aramaic began to displace Akkadian in the administration of the Assyrian heartland in the eighth and seventh centuries BCE, it did so despite Akkadian, and it did so by a route that can be reconstructed from the documents themselves.
Aramaic had two advantages. The first was that it was not the language of the old scribatic establishment, so there was no vested guild defending its monopoly of the tablet; the Aramaic-speaking deportees had no scribal caste to defend and no canon to uphold. The second was that Aramaic was the language of the road. It was the speech of the merchant and the caravan and the clerk, and it was already spreading as a trade language across a wide region that Assyrian power had never reached. Assyria did not create Aramaic; it enrolled it. The empire adopted the language because the language had already done the harder work of becoming useful, and Assyria then gave it the one thing a trade language cannot get for itself, which is an administration.
There is a small and quite devastating document from Nineveh that captures the moment. It is a letter written in Aramaic by a scribe to his own pupil, or rather it is a text that presents itself as such, preserved in the British Museum, in which the tutor complains that the boy’s native Akkadian is being crowded out. The Aramaic is the language of instruction, the language of the letter, and the language the tutor is worried is swallowing the boy’s own tongue. The document has traditionally been read as a sort of lament for the displacement of Akkadian by Aramaic in the Assyrian administration, and it is almost exactly what you would want: an insider’s view of a language change happening from the inside of the bureaucracy that was making it happen. The tutor’s complaint tells you that by the time the letter was written, probably in the seventh century BCE, the change was far advanced. Aramaic was no longer a foreign intruder being used against Akkadian. It was simply the language, and the older one was what needed defending.
The format of the letter is worth pausing on, because it is the format that would carry the language everywhere. The Aramaic letter of the Assyrian and Persian periods has a recognisable architecture: a greeting in the third person, the name of the addressee, the self-description of the writer as servant, the body, and a formula of greeting at the close. The writer’s designation of himself as the addressee’s servant is not a rhetorical flourish but an institutional statement, and it tells you that the relationship being asserted is one of subordinate official to superior official, with the sender claiming as his merit that he has written to say nothing more important than to send greetings and to ask for the recipient’s favour. The same architecture survives in the letter from Elephantine to the authorities of Babylon and Samaria, in the Aramaic of the Bar Kokhba documents, and, in a later register, in the epistolary Syriac of the eastern monasteries. A script and a set of formulae together are a portable form of administration.
Aramaic and Hebrew are sister languages rather than one a relative of the other. Both are Northwest Semitic; Hebrew is usually described as a Canaanite language, descended from a dialect of the early West Semitic continuum that the Assyrian inscriptions call “Amorite”, and Aramaic is a close cousin of that same continuum which moved off in a different direction. The result is that the two languages are mutually intelligible in a limited way, in the way that Danish and Norwegian or Serbian and Croatian are, and that they shared a long period of contact that left a mark on both. Hebrew took a great many Aramaic words into its vocabulary, among them the everyday terms for the domestic animals, the trades and the body, and Aramaic in turn absorbed a smaller number from Hebrew. The two languages then had a long, tangled and affectionate relationship in which each supplied the other with literature. Some of the most famous lines of Hebrew wisdom literature are spoken in the idiom of a reader who grew up hearing Aramaic.
The oldest Hebrew scripture itself knows the Arameans, and knows them at the level of the household rather than the battlefield. Abraham’s servant is described as an Aramean of Aram-Naharaim, the Aram of the two rivers; Laban, the brother of Rebekah and the employer of Jacob for two decades, is “Laban the Aramean” and complains about Jacob in Aramaic. The phrase Paddan-Aram, the Aram of the fields, is another name for the same country. These are not references to a distant empire but to the speech of the household in which the patriarchs lived. Whatever language the books of Genesis were written in, the men who spoke in them most likely spoke Aramaic, or something very like it, and the audible traces of that remain in the text. When the Assyrians deported the northern kingdom and scattered the Israelites among Aramean-speaking populations, they did not in fact introduce a new language to them. They put them among their own.
The peoples of Aram had, as their name suggests, a name that probably means highland, and the ethnonym itself is uncertain in a way that is typical of the early history of the language. What is certain is that the label was applied by outsiders. The Assyrians, who had their own designations for everything, called the language and its speakers Chaldean, and that word became the standard term for Aramaic in a great many contexts for the next two thousand years. When Daniel’s magicians are called Chaldeans, the Hebrew word used is chaldi, which is not a gentilic for the people of Chaldea but the Aramaic loanword for the class of Babylonian scholars who read the heavens. The Septuagint translators rendered the same word the same way. The name stuck, and eventually it came to mean Aramaic-speaking Jew in a way that was both accurate and hopeless, since from the Persian period onwards the Jewish communities of Babylonia were as Aramaic-speaking as they were Jewish, and their neighbours looked for the one that was not Aramaic.
The Aramaisation of Assyria
It is worth being precise about the pace of the change, because the assumption that a language spreads because an empire orders it to is too simple. Aramaic became the language of Assyrian administration gradually, through a series of small institutional decisions rather than one decree. There was no moment at which Akkadian was abolished. What there was, in the seventh century BCE, was a state in which the scribal curriculum itself was changing, in which the official correspondence of the empire was conducted in Aramaic, and in which an Aramaic letter was the ordinary way to communicate with a provincial official. The letter from Nineveh is evidence of this process at the level of the schoolroom, which is the level at which language change is actually decided.
The scribal consequence was enormous. Writing in Aramaic meant writing in an alphabetic script rather than in cuneiform, and that single change did more than any other to make the language portable. Cuneiform was powerful and beautiful and it was the product of a specific technical tradition: a wedge-shaped stylus pressed into wet clay, requiring training, a supply of clay, and the physical presence of the clay tablet. The alphabet required only a reed pen, a reed cut to a flat edge, and a surface that would take ink: papyrus, parchment, leather, or wood. An alphabetic script in Aramaic could be written on anything, by anyone, and carried anywhere. This is not a small point. It is the technical precondition for everything that follows, from the Aramaic documents of the Jewish colony at Elephantine to the Aramaic letters of the Bar Kokhba revolt in the caves of the Judean desert, and it is the reason the language travelled at all.
Aramaic scribes were not rare. They were, in fact, an industry, and the industry had a geography. The concentration of Aramaic scribal activity in Babylonia, and the survival of the tablets of the Murashu family from Nippur in the form that this journal has described elsewhere, show a working commercial world in which writing a document in Aramaic was an ordinary professional skill practised across generations in the same house. The scribal families of Nippur are the clearest surviving window onto this world, and the literary Aramaic that came out of the same excavations, the incantation texts that preserve older Sumerian and Akkadian magical formulations in Aramaic dress, shows what the scribes did for pleasure as well as for pay. An Aramaic-speaking scribal class had its own professional pride, its own connoisseurship and its own literary ambitions, and it supplied the empires that employed it for three consecutive dynasties.
The Aramaic of the administrative sphere was a koine in the strict sense, a composite language built for clarity rather than for beauty. It had a small closed vocabulary for the things that mattered, a fixed phraseology for formulae of address and for the opening and closing of a document, and a grammar stripped of the complications that make literary languages difficult to use at speed. It was, in effect, a language designed for filing. That it was also capable of carrying the subtleties of the Book of Daniel, of the Psalms of David and of the theological prose of the targumim is one of the more remarkable facts about it, and it is the reason that the study of Aramaic has never been only a study of paperwork.
The administrative Aramaic of Assyria was, however, about to be inherited rather than celebrated. When the Assyrian empire fell in a single catastrophic year of 626 BCE, when Nineveh was sacked and the empire dissolved within a generation, the Aramaic scribal class did not vanish with it. It dispersed, and it was recruited. The Babylonians, who had been subjects and then conquerors, had no Aramaic scribal tradition of their own of comparable depth and they needed one, so they employed the men who had been trained in Assyria. The Persian kings who followed conquered Babylonia in the 540s BCE and found the same arrangement waiting for them: an Aramaic-speaking administrative class, with the Aramaic language of the great empires running its bureaucracy. The language that had been adopted as a tool of Assyrian control outlived Assyria by three hundred years, and was adopted by everyone who conquered Assyria’s successors.
It was in Babylon, in fact, that the Aramaic of the administration and the Aramaic of the synagogue first came into sustained contact, and the result was the tradition of the targumim that this journal has described at length in its accounts of Targum Onkelos and of Targum Jonathan ben Uzziel. In the courtyards of the Babylonian academies a Hebrew scripture was expounded through an Aramaic translation, and the translation was not a classroom exercise but a liturgical act with a defined form, a conventional style, a name for its practitioner, and eventually a print tradition of its own. The Jewish communities of the east had taken the language of their rulers and made it the vehicle of their scripture, and the result was that the Hebrew Bible was read for two thousand years in a language that was not the one in which it was written. That is the second great irony of Aramaic’s history, and it belongs with the first: the language spread by conquest and administration was then used to make the conquered people’s own scriptures accessible without ceasing to be the conquerors’ tongue.
Imperial Aramaic and the Fall of the Empire
The most striking thing about the transfer of power from Assyria to Persia in the sixth century BCE is how unremarkable it was. When Cyrus the Great entered Babylon in October 539 BCE, he did not, so far as we know, order any change of administrative language. He found an empire that already ran on Aramaic, and he inherited it. The Cyrus Cylinder, the document in which Cyrus announced that he had taken the throne and granted his subjects their rights, was itself written in Akkadian, in the cuneiform script, in the old style of Babylonian royal inscription. The last great Akkadian monument was produced at the moment that Akkadian had already ceased to be the language anyone used for anything practical. It is a document written in the past tense by a court that had stopped believing in its own language, and it is one of the most moving pieces of administrative writing in the ancient Near East.
What the Achaemenid kings did with Aramaic was not to reform it but to industrialise it. The Persian empire was larger than any that had preceded it, stretching from the Aegean to the Indus and from Central Asia to the Horn of Africa, and it was administered by satrapies whose governors needed to communicate with each other, with the royal court at Susa or Persepolis, and with a mixed subject population. The language they settled on was Aramaic, and the reason was simply that it was the only language in the empire with a trained scribal establishment, a portable script and a working convention for official documents. Aramaic was the administrative Latin of the ancient Near East, and like Latin it held the post because there was no serious competitor, not because anybody loved it.
The evidence for this is startling in its ordinariness. In the 1930s the German archaeologist Ernst Herzfeld excavated the great terrace complex at Persepolis and, in the fortification wall behind the treasury, found thousands of clay tablets written in Aramaic. They had been sealed inside the wall as a foundation deposit, a scribal archive sealed up and forgotten, and when the masons came to the surface in the 1930s they found inside it the paperwork of the Achaemenid administration. The Fortification Tablets, as they are now called, and the related Treasury Tablets, record the flow of food, animals, silver, wine and workers through the imperial centre. Richard T. Hallock devoted much of his career to editing them, and they remain the single best window onto how the Achaemenid state actually functioned day to day.
What the tablets reveal is a state that ran on Aramaic in every direction at once. The language of the accounts was Aramaic. The names of the people in them are a spectrum: Babylonian, Elamite, Egyptian, Persian and Greek names all appear, and the scribes are identified by their names and their ethnic affiliations. Workmen who built the terrace came from across the empire. The interesting detail, for the historian of language, is the one that the tablets were written in a single script, a single language, and in a form with almost no variation from one end of the empire to the other. A clerk at Persepolis and a clerk at Susa were writing in the same bureaucratic register, and a Persian official could draft an instruction in Aramaic and be understood in Egypt, in Bactria and in Babylon without translation.
This is the point at which Aramaic stops being a language of one region and becomes an imperial instrument. It also begins to be a slightly different language from the one the Murashu scribes had used in Nippur, though the difference is a matter of register rather than kind. Imperial Aramaic was a chancery koine with a closed and carefully policed vocabulary, a stock of formulae for opening a letter and closing it, and an orthography that was conservative in a way that literary Aramaic never was. It was, in effect, engineered for reliability. Two clerks in different satrapies, trained in different traditions, could be expected to write the same word the same way, because the whole point of the orthographic rules was that a document drafted in one place would be read without ambiguity in another.
One of the most telling details of Imperial Aramaic is what it did not do. Aramaic writes consonantal text, with no vowel points, and Imperial Aramaic was more radical than most about this. There were occasional devices to mark a final yod and the occasional attempt to disambiguate, and there were a very small number of foreign words marked with a special sign, but the system was austere. The result is a script that is easy to write and comparatively hard to read, which is exactly the profile of a document language. When the Jews of the Elephantine colony received official replies from the Persian authorities in Aramaic, they could read them, because Aramaic was what they spoke. When a Greek-speaking administrator in Egypt received an Aramaic draft of a contract, he did not read it; he had it read to him, and he relied on the formula and on the scribe.
Monuments tell a different story from the paperwork, and the difference is instructive. The great Achaemenid inscription, the Behistun relief and text of Darius I, is trilingual: Old Persian, Elamite and Babylonian, in three scripts. There is no Aramaic version of the main text. Aramaic was the language of the administration and not the language of the monument, because the monumental form was aimed at subjects whose reading was tied to particular traditions and whose prestige required the use of their own ancestral languages, and the Persians were careful to say so in their own words. The royal inscriptions of the Achaemenids are a declaration that the empire was multi-national by design. Aramaic was the instrument of that multi-nationalism, not its ornament.
Aramaic nevertheless appears in monumental context in Babylon itself, and its appearance there is significant. A series of inscriptions cut for Nebuchadnezzar II in the early sixth century BCE carries essentially the same text in three languages: Babylonian, Aramaic and Greek. The presence of Greek is the surprise, and it marks the point at which Aramaic ceased to be merely the lingua franca of a Semitic world and became one of the standard languages of a wider Mediterranean system, alongside Greek and Babylonian, on a footing that a conquering language was willing to recognise. The Aramaic of these Babylonian inscriptions is a little different in register from the Aramaic of the Persepolis tablets, being more literary and more closely tied to the Aramaic of the Babylonian scribal and temple establishments, and it is worth noticing that the Babylonians did not adopt it cold. It was there already, because the Chaldean temple administration had used Aramaic for a long time.
Alexander’s conquest in 331 BCE changed the empire’s ruling dynasty but barely touched the language of its paperwork. The Aramaic chancery passed, like everything else, to the Seleucids, and the documents that survive from their dominions are the most copious in the whole imperial series. The Zenon papyri from Ptolemaic Egypt, the correspondence of the Judean community at Elephantine with its officials in Egypt, the Aramaic documents from the cave of letters in the Judean desert, the contracts and letters from Bactria: these are the working papers of an administrative world in which an Aramaic-speaking scribe in Bactria and an Aramaic-speaking scribe in Egypt used the same orthography, the same formulae and the same calendar conventions. The language had become infrastructure.
It remained infrastructure through three more successive Persian empires. The Parthians, an Iranian dynasty that ruled from the first century BCE, adopted Aramaic as their chancery language and wrote their own Iranian language, Parthian, in letters derived from Aramaic. The Sasanians, who replaced them in the third century CE, did the same, and their Middle Persian was written in an alphabet derived from Aramaic that is still, in its broad strokes, recognisable as an Aramaic script. This is one of the strangest and most durable facts about the Near East: for some eight hundred years, from Cyrus to the Arab conquest, Aramaic was the documentary language of Persian empires whose own languages were Iranian and whose own scripts were invented by their own scribes. It was not a case of a conquered language keeping its vitality. It was a case of an administrative instrument being valued for what it was, and defended by the state because a state that changed its document language would have to retrain its entire civil service.
Side by side with this imperial Aramaic ran another Aramaic, the Aramaic of the eastern churches. The Sasanian state was Zoroastrian, and the churches of the East were its rivals in matters of theology and, increasingly, in matters of political allegiance. Those churches, whose adherents spoke and wrote Syriac, a literary Aramaic, maintained their own institutions, their own bishops, their own schools and their own canonical literature. The juxtaposition is worth dwelling on: within the same Sasanian empire, the state used Aramaic for its administration and its coinage legends, and the dissident Christian communities used Aramaic for their theology, their liturgy and their letter. The language was the common ground of the apparatus of the state and of its opposition to it, and neither party could have operated without it.
Imperial Aramaic ended, as imperial things in the Near East tend to end, not with a single decree but with a long administrative transfer. In the seventh century CE the Arab armies overran the Sasanian provinces, and the new rulers inherited an Aramaic-speaking administrative class because they had no alternative. Aramaic chancery documents survive from the early decades of the eighth century, some of them written by officials who had served the previous regime. But the Arab state was building a different administrative language on a different script, Arabic, written in letters that were themselves descended from Aramaic, and the writing of the two languages diverged as their administrations did. By the end of the century the administrative Aramaic of Iran was over. What remained was the script, which was about to prove far more durable than the language, and the dialects, which went on living in the villages, in the monasteries and in the mouths of Christians, Jews and Mandaeans.
It is worth pausing on the scale of what was lost. The administrative Aramaic of the Achaemenid, Seleucid, Parthian and Sasanian states was not merely a language; it was a whole apparatus of writing, spelling, dating, accounting and authority, and it went out of use in a single generation across a territory the size of France. No one voted for it to stop. The decision was made in the offices of the caliphate, in the ordinary way that new administrations are built, and the old language was retired the way a superseded ledger is retired: quietly, completely, and without much ceremony. The people who had spoken it went on speaking it. But the state stopped writing in it, and a language that a state stops writing in has lost the principal thing that kept it alive for a thousand years.
A Script and Its Descendants
Aramaic’s alphabet is the part of it that outlived everything else, and the first thing to say about that alphabet is how small it is. The classical Aramaic alphabet has twenty-two letters. It is an abjad, a writing system in which the letters represent consonants and the vowels are left to the reader to supply, and it is written from right to left. The letters have names, and the names are inherited from the older Semitic world: aleph, bet, gimel, daleth, he, waw, zayin, heth, teth, yod, kaph, lamed, mem, nun, samekh, ayin, pe, tsade, qoph, resh, shin and taw. The shapes have changed over the centuries and vary between hands and regions, but the basic skeleton of the forms is recognisable across the whole family of scripts descended from them, and the order of the letters has never been in doubt. This is an alphabet that is nearly three thousand years old and has never been out of use.
The forms that won were the ones that suited the Aramaic chancery hand of the Persian period, and they won because they were the forms that the scribes of the Jewish community in the Persian province of Yehud chose to adopt for Hebrew. The coins struck in Yehud in the fourth and third centuries BCE carry legends in a script that is Aramaic in every respect but its language, and they mark the moment when Hebrew ceased to be written in the old Phoenician-derived forms and began to be written in what is now called square Hebrew script. That script is, in its essential structure, an Aramaic script. Its letter shapes are Aramaic letter shapes with the angles squared off and the proportions regularised. It was a domestic adoption of an imperial instrument, and it happened because the Aramaic letters were what the coinage of the province looked like, and the coinage was what everyone could read.
From the square Hebrew script the steps to the other great scripts of the region are short and traceable. The Syriac scripts, of which there are several, descend from Aramaic letterforms reshaped for a hand of very different sort, the monumental Estrangela of the earliest manuscripts, the rounder and more cursive Serto of the Rabbula Gospels and the Peshitta tradition, and the Eastern Syriac script that carries the liturgical writing of the Church of the East. The Mandaic script of the Mandaeans of southern Iraq descends from the same source and keeps its Aramaic name for God, the Great Living One. The Nabataean script, used for the Aramaic of the Nabataean kingdom and its heirs, gave the Arabic script its shape, and the modern Arabic script is, in essence, Nabataean Aramaic with a handful of added letters, a handful of dropped ones, and a very large number of new uses for the dots that Aramaic’s single letter shape had never needed.
The dispersion of the script is one of the great unintended consequences of the Aramaicisation of Assyria, and it has no single cause. The Syriac letters travelled with the Syriac Christians of the East, who carried them from Mesopotamia into Persia, into Central Asia, and as far as China, where a stone of the year 781 CE records the arrival of a Christian community in the capital of the Tang empire. The Arabic letters travelled with the Arab conquests and with the imperial administration that followed them, and then with the religion that came out of them, and then with the languages that the religion’s scholarship required to be written down, so that the script that the Muslims took from the Nabataeans ended up carrying Persian, Turkish, Kurdish, Urdu, Malay and Javanese as well as Arabic. The Hebrew letters travelled with the synagogue and with the Christian Hebraists, and with the printing of the Hebrew Bible. What the script carried was not the language. It was the technique, and the technique was the durable part.
Two of these lines deserve more attention. The first is Syriac, which is the line that carried the script furthest east and converted it into a liturgical alphabet. The Syriac letters of the Church of the East were used to write a liturgy in Syriac, and the liturgy in turn carried the alphabet into new languages, among them Sogdian, a language of the great Silk Road cities of Central Asia written in an alphabet of Syriac descent, and the Old Turkic inscriptions of the eighth and ninth centuries, which are among the earliest written records of a Turkic language and use a script derived from Syriac. Whether those Turkic letters were borrowed directly from Syriac or through an intermediary is debated, but the descent is not in doubt. The Church of the East carried its letters two thousand miles east, and they were still being used to record a Turkic state in the Yenisei basin generations after the empire that had taught Syriac Christianity to the Mongols had gone.
The second line is the one this journal has told in its own account of the Armenian Bible, since Mesrop Mashtots built his alphabet in the early fifth century CE from Greek and from the Iranian scripts of the Sasanian world, and those Iranian scripts were themselves Aramaic. The alphabet of Caucasian Albanian, the language of the ancient kingdom centred on the Caspian, is in the same family of derivations. The Ge’ez script of the Aksumite and Ethiopian churches is a separate creation of the fourth century, generally traced to Greek uncials, though some of its letter forms show Syriac influence and its use of a vocalised script for the transcription of the Ge’ez language has no precedent in the Aramaic world. Ge’ez is written from left to right, unlike the Aramaic family it resembles. Every one of these scripts is a reminder of a point that is easy to miss: the Aramaic alphabet did not spread because Aramaic speakers carried it abroad. It spread because alphabets travel with institutions, and the institutions that carried it were churches, schools, bureaucracies and banks, each of which needed a way to write and each of which found the Aramaic letters ready to hand.
What finally distinguishes the Aramaic script from everything around it is the length of its unbroken life and the completeness of its family. The scripts of Mesopotamia, the cuneiform tradition, went out of use entirely. The Egyptian hieroglyphs and demotic went out of use and were revived only as a religious curiosity, and even that revival was the work of one man. The Phoenician script survived, but only as a substrate, in Greek and then Latin and then in the alphabets of Europe, which changed so thoroughly that their descent is invisible on the page. The Aramaic script, by contrast, is still in daily use in the Middle East and the Caucasus, still read by Mandaeans and Syriac Christians and Jews, and still the direct ancestor of the letters used every day by something over two hundred million people who write Arabic. In the standard used by every computer in the world, the Aramaic block sits immediately next to the Hebrew block, the two alphabets that share a letterform family numbered in their thousands, side by side in the list.
It is a strange thing to reflect on, a century and a half after the Assyrian deportations that set the whole process in motion, that the most durable legacy of the Arameans is not a language at all. The Aramaic language is a set of living dialects in the Zagros, in the Jabal al-Druze, in Maaloula and in the villages of the Jewish hills of Galilee, and its literature is a library of targumim, of the Syriac scriptures and of the Mandaean rituals, and it is living and precious. But the script is the thing that spread. It went from the offices of Nineveh to the temples of the Magi, from the citadel of Hatra to the monasteries of Tur Abdin, from the chancery of Susa to the steppe of Mongolia, and in the end it became the vehicle of a poetry and a philosophy in a language that has nothing to do with it. That is not a metaphor. It is the ordinary mechanism of history: a people lose their country, and their alphabet becomes the most successful export they ever had.
Aramaic and the Scriptures
The single most surprising place to find Aramaic is inside the Hebrew Bible. Between two and three per cent of the Old Testament is in Aramaic, and it is concentrated in exactly the places where a reader would expect to find the language of an imperial administration. The book of Daniel, the book of Ezra and a single verse in the book of Jeremiah are the Aramaic books, and the concentration is not accidental. Aramaic is in the Bible because the Bible preserves, in its very texture, the administrative language of the empire in which it was written.
Daniel is the case that has to be handled carefully, because the book is in two languages and the division is meaningful. The opening of Daniel is in Hebrew, and the narrative sections that follow are also in Hebrew: the young Daniel at the court of Nebuchadnezzar, his interpretation of the dreams, the writing on the wall, the lions. The visions, the court disputes, the decrees and the state conversations are in Aramaic. The effect is of a book that changes language when it enters the world of the court and the empire. Daniel speaks Hebrew to God and Aramaic to the king. The Aramaic of Daniel is not a translation; it is the language of the story world, and the Hebrew is the language of the interior life, and the alternation is one of the most accomplished pieces of literary construction in the ancient Near East.
The world of Daniel is the world of the exile and the return, the world in which the Temple had been destroyed and the imperial administration of Babylon and Persia had become the framework of ordinary life. It is exactly the world in which Aramaic was the working language of the court. The book does not translate the Aramaic passages into Hebrew for its readers, which implies that a substantial part of its original audience did not need it translated. The Septuagint translators, working a century or more later, rendered the Aramaic sections into Greek; that they thought it necessary to do so is itself evidence of how the balance of readership had shifted by the third century BCE.
Ezra is even more administrative, and for the historian of language it is the most instructive book in the Bible. The second half of Ezra, from chapter four onwards, is largely a series of memoranda, letters and official replies between the Jewish community in the province of Yehud and the Persian authorities. The letters are in Aramaic and they are in the idiom of the imperial chancellery. They begin with a greeting, they name the sender and addressee, they refer to previous correspondence, and they appeal to the memory of an earlier document, and they do all of this in the same formulae as a contract from Bactria or a receipt from Persepolis. The community at Elephantine in Egypt wrote letters to Persia in exactly this style. The men of Ezra are writing to the same offices, using the same conventions, and they are unusually good at it, because the community they represent had learned the paperwork from the Persians who had governed them for a century.
The letter of chapter seven of Ezra, in which the prophet is given permission by Artaxerxes to teach and to establish the law in the province, is the moment at which Aramaic stops being the language of empire and becomes the vehicle of a minority’s claim to be the legitimate religious authority of a territory. The genre has changed from memorial to petition to something closer to a decree. The language has not changed at all, which is precisely the point: the community has decided that the imperial language is the right vehicle for its own purposes, and it intends to use the empire’s own paperwork to establish its own law.
Jeremiah’s single Aramaic verse, chapter ten verse eleven, is an outlier and a curiosity. It is a taunt against Babylon, dismissing its gods as vapours and thin smoke, and it is written in Aramaic in the middle of a Hebrew poem. It is very likely an Aramaic scribal gloss or interpolation, or a piece of Aramaic popular verse that was taken up into the Hebrew book, and its presence is a reminder that in the Persian period Aramaic was not merely the language of administration and of the synagogue. It was a language with its own jokes, its own popular forms, and its own presence in the scribal tradition of Judah.
Those last two points lead to the targumim, which this journal has described in detail in its accounts of Targum Onkelos and of Targum Jonathan ben Uzziel, and it is not necessary to repeat that story here. What is worth adding from the point of view of the language’s own biography is the scale of what happened. Before the targumim, the Hebrew Bible was a book that could be read, at least in principle, by anyone who knew Hebrew. After the targumim, for a very large part of the Jewish world, it became a book that was heard, and the hearing required an interpreter. Aramaic became the gateway to the Hebrew text rather than its rival. A Jew who could not follow the Hebrew could follow the Aramaic, and a synagogue service consisted of the Hebrew being read and the Aramaic being said aloud in the gaps. The language that Assyria had used to break up a people had become the thing that kept their scripture accessible to them.
The same dynamic operated in Greek, and it is one of the underappreciated results of the Septuagint that its Aramaic Vorlagen are now a principal source of textual criticism of the Hebrew Bible. Where the Greek translator appears to have misunderstood the Hebrew, the reason is often that the translator was working from an Aramaic version and was carrying the Aramaic reading into the Greek. The reverse is also demonstrable in a few passages, where the Hebrew appears to have been adjusted to match a Greek that had been translated from Aramaic. The Greek Septuagint is therefore not simply a translation of the Hebrew. It is a translation, sometimes at several removes, of a mixed textual tradition in which Aramaic stood between the Hebrew and the Greek, and this is why the story of the text has become more complicated, not simpler, as the documents have multiplied. The same phenomenon appears in the Syriac Peshitta, which this journal has treated at length in its account of the Peshitta and its Hebrew Vorlage, and which carries an Aramaic layer between the Hebrew and the Syriac.
Jerome’s Latin Vulgate, completed in the early fifth century, is the point at which this layering becomes visible in a single document. Jerome worked from the Hebrew, with the help of Jewish and Christian tutors, and for the Hebrew Bible proper he produced a genuinely new translation rather than a revision of the Greek. But when he came to the Aramaic books, Daniel and Ezra, he did not work from the Hebrew alone. He used the Greek Septuagint of Daniel, which was itself translated from an Aramaic Vorlage, and he reproduced a number of its readings. The result is a Latin text which, at a handful of points, preserves a Hebrew reading that has been filtered through Aramaic and Greek and back again, a chain of transmission that would have satisfied a media scholar. The passage of the Aramaic through the Latin tradition was also unusually faithful in one particular respect, and it is the point at which the New Testament comes into the story.
The New Testament was written in Greek, but the Greek rests on an Aramaic foundation, and in a dozen places the Aramaic shows through the Greek like a subsiding tide leaving stones behind. The most famous is the cry from the cross, which the Gospels record as the last word of Jesus: Eloi, Eloi, lema sabachthani, my God, my God, why have you forsaken me. The underlying form, quoted in Aramaic and then translated by Matthew and Mark, is the opening of Psalm 22 in a popular Aramaic paraphrase rather than the Hebrew of the psalm itself. It is the one place in the New Testament where a quotation of scripture is given in a form that is not the form in the Hebrew Bible, and it is a reminder that Jesus’s own scriptural practice included a popular Aramaic text of the psalms.
The other traces are scattered but instructive. The word Jesus uses in the Gospel of Mark when he tells the deaf man to be healed is ephphatha, and the Gospel translates it, be opened. The word he uses to the girl in the synagogue at Capernaum is talitha kumi, and the Gospel translates that too, little girl, get up. The Aramaic talitha is the emphatic feminine singular of the verb meaning to rise. The word for the coin in the story of the widow’s mite is mammonas, Aramaic for the small denomination of copper currency, and the Greek calls it a lepton, which is not the same coin. The Aramaic word for the-skilled, used in the Gospel of Matthew to describe the servants in the parable who are beaten, is a Greek transcription of an Aramaic term. The names of the two disciples James and John, Boanerges, sons of thunder, are an Aramaic phrase preserved in Greek. The Aramaic word Golgotha is the Greek rendering of an Aramaic place-name, and in Matthew it appears with its Aramaic meaning given in Greek as the skull. And the Aramaic word Abba, which Jesus used in the Lord’s Prayer, is preserved in the Greek with the comment that it means father, an explanatory gloss that tells us at once that the Greek-speaking church did not expect its readers to know the word.
The last of these is the most consequential, because it means that the Aramaic prayer that the earliest Christians used was, by the end of the first century, already a foreign object to a substantial part of the church. The Greek text of the Lord’s Prayer that most Christians have prayed for two thousand years is a translation of an Aramaic original, and the Aramaic original is lost. The Peshitta preserves the Lord’s Prayer in Aramaic, in the form that the eastern churches used, and it is a different text from the Greek in several places, including the opening: in the Peshitta the prayer begins not with the address to the Father but with the petition for the kingdom to come and for the will to be done on earth as it is in heaven.
That is the New Testament’s Aramaic in a nutshell. Where the Greek text is silent, it preserves something; where it translates, it admits that translation is needed; where it quotes scripture, it sometimes preserves a form of scripture that is not the Hebrew form. A language that was spoken by a rural Galilean Jewish carpenter and his followers has left traces in the most influential book in the Western tradition, and those traces are not ornamental. They are load-bearing.
Outside the scriptures, Aramaic continued to be used for the purposes it had always served. The letters from the Bar Kokhba revolt, found in the caves of the Judean desert in the 1950s, were written in Aramaic and are among the most vivid documents of the revolt’s administration, with the tone of some of them remarkably familiar. They are, as this journal has described in its account of the Elephantine papyri, part of a continuous tradition of Jewish communal writing in Aramaic that runs from the fifth century BCE in Egypt, through the Persian and Hellenistic periods, to the second century CE in the desert. The letter from the Jewish community at Elephantine to the Persian authorities, sending its greeting to the governor of Babylon and the authorities of Samaria, was written in the same language and the same formulae as the Aramaic sections of Daniel, and the continuity between them is not a matter of resemblance but of unbroken practice.
The Long Afterlife
Aramaic outlived every political circumstance that had made it important, but it did not do so in a single form. What survived was a set of religious communities, each with its own dialect, its own script, its own calendar and its own relationship to the empires that surrounded it, and the history of the language after the seventh century CE is largely the history of those communities and of the pressures that slowly closed over them.
The most consequential of them was the Syriac Christians. The Church of the East traced its origin to the missions of the first century and its institutional form to the metropolitan see at Seleucia-Ctesiphon in Mesopotamia, and it grew to a communion stretching from Egypt to eastern Persia, into Central Asia and, as the stele of 781 CE records, into the Tang empire in China. Syriac was its liturgy, its literature and, increasingly, its administrative language. Its scriptures were the Peshitta, and its doctrinal disputes with the Roman empire were conducted in a language that Roman Europe had never heard. The long-running quarrel that produced the Church of the East and the West Syriac or Melkite church was a quarrel about the nature of the single nature of Christ, and it was fought for the most part in Syriac, and the schism that resulted in 431 or just after it is, in linguistic terms, the point at which Aramaic became a confessional boundary.
The Syriac Christians did not merely speak a language; they made one. The fifth and sixth centuries produced a literary culture in Aramaic of extraordinary range. Its central monument is the Rabbula Gospels of 586 CE, a masterly Syriac manuscript whose text stands at the head of the Peshitta tradition and whose script, the rounded serto hand, became the standard book hand of the East Syriac churches for the next thousand years. Around it grew a translation movement that carried Greek philosophy, Greek medicine and Greek science into Aramaic, and a school of interpretation that produced the only surviving manuscript of the Aramaic text of the Book of Mani, the scripture of the Manichaeans, copied in the ninth century and now held at Bonn. The East Syriac world produced its own scholastic tradition, its own calendar, its own martyrologies and its own cycle of feasts, all of them in Aramaic, and it produced them with the apparatus of a literate empire, including a school at Edessa, traditionally founded in the fifth century, that taught Syriac, Greek and medicine together and was moved east to Nisibis when the Persians took Edessa in 489.
The Arab conquest changed the legal and political conditions of the eastern churches without ending them. Christians in the new Islamic empire became dhimmi, protected but subordinate, and their bishops acquired new obligations to Muslim government. The great irony of the early Islamic centuries is that the Syriac-speaking Christians of the Abbasid caliphate became the translators of the classical and scientific literature of Greece and Persia into Arabic, in the celebrated translation movement associated with the House of Wisdom in Baghdad, and that the medicine, mathematics and astronomy of the medieval Islamic world passed through Aramaic and Syriac on their way. Syriac-speaking Christian physicians and scholars produced Arabic versions of Galen, Hippocrates and Dioscorides, and those versions were then translated into Latin in the twelfth and thirteenth centuries and became part of the European scientific tradition. A language that had been the vehicle of the Assyrian and Persian state had become a bridge between the ancient world and the medieval one, and it had done so while remaining, in its own view, the language of a minority faith.
The Mandaeans of southern Iraq were the other ancient Aramaic community to survive into the modern period, and they survived in a more surprising way. The Mandaeans, whose ritual language is a form of Aramaic close to the older literary dialects, have preserved a set of beliefs that incorporate elements of ancient Mesopotamia, a Gnostic dualism, and a baptismal rite performed in running water. They are a small community now, in the low thousands, concentrated in the marshlands of southern Iraq and in the diaspora, and they have never had the institutional weight of the churches. But their survival is a reminder of a point that is easy to overlook: the Aramaic communities of the ancient Near East were not exhausted by the two great religious traditions. There was a third, and possibly more, and the reasons why each survived or declined are not reducible to anything so simple as the importance of the institutions that carried them.
The Jewish communities of Babylonia are the clearest case of a long Aramaic life, and the clearest case of its ending. The Jews of Iraq traced their presence there to the Exile, and they remained Aramaic-speaking for two and a half millennia. The targumim were made in their synagogues, the Babylonian Masoretic tradition and the Babylonian supralinear vocalisation system were developed for the study of the Hebrew text in an Aramaic environment, and the great codexes of the Babylonian tradition, including the Aleppo Codex in its later Iraqi provenance, are the work of Aramaic-speaking scribes. Then, in the middle of the twentieth century, the community was destroyed as a community. Between 1950 and 1951 the Jews of Iraq were subjected to a systematic campaign of persecution, and the great majority left, a mass exodus to Israel that removed the population almost entirely and with it the last generation of native speakers of Baghdadi Judeo-Aramaic. The language is now spoken by a handful of elderly people in Israel, and by a diaspora that has kept it alive in the United States and in a small number of other places.
The Christian communities of the Ottoman and modern Middle East were more durable, and then catastrophically diminished. The Chaldean Catholic Church, whose head was the patriarch of the Church of the East until the sixteenth century, and the various Syriac Catholic and West Syriac churches, maintained their languages, their rites and their canonical literature into the modern period, and the villages of Tur Abdin in the Mardin mountains remained, until the twentieth century, a heartland of East Syriac monasticism. The twentieth century changed that decisively. The Iraqi wars of the late twentieth and early twenty-first centuries, the position of the Mandaeans after 2003, the displacement of the Assyrian communities from northern Iraq after 2014, and the flight of millions of Syriac Christians from Syria after 2011 have reduced the population of the East Syriac churches in their historic homeland to a fraction of what it was. The language is now, overwhelmingly, a diaspora language, spoken in Sweden, Germany, the United States and Australia.
And then there is the Neo-Aramaic of the villages, which is the strangest survival of all, because it is the one nobody planned. Neo-Aramaic, in the strict sense, is a group of closely related dialects spoken by communities who have no institutional connection to one another and no shared literary tradition, most of whom never had a canonical language in the sense that the targumim or the Peshitta are canonical. The main groups are the Assyrian Neo-Aramaic of northern Iraq, north-eastern Syria, south-eastern Turkey and western Iran; the Chaldean Neo-Aramaic of Iraq and Iran; the Western Neo-Aramaic, spoken in and around Sanandaj in western Iran and in the Sinjar region; and the Jewish Palestinian Neo-Aramaic, spoken by a small number of families in northern Israel whose speech preserves features of the Aramaic of the ancient Near East that have no counterpart anywhere else. These are the descendants, in the most literal sense, of the Arameans whose deportation by Tiglath-Pileser III set the whole story in motion, and they have been doing their own thing in their own villages ever since.
The best known of the Neo-Aramaic communities is the smallest. Two villages in the Jabal al-Druze, an hour or two south of Damascus, Maaloula and Jdeidet Maaloula, speak a form of West Syriac Neo-Aramaic, and they have kept it as a distinctive community identity in a way that few language minorities anywhere have managed. The villages are majority Christian, of the Melkite rite, and the language is used in the liturgy and in daily life in a way that no larger Neo-Aramaic community manages. It has not been an unmixed blessing. The war in Syria reached Maaloula, and in November 2013 the Islamic State attacked the village, murdering residents and destroying homes and forcing the population to flee. The village has been partly rebuilt and partly repopulated, but the episode is a reminder that the survival of a language community in the modern Middle East has never been guaranteed by anything other than luck.
The prognosis for Neo-Aramaic is not good, and it is worth being plain about the reasons. The speakers are small in number, dispersed by war and by economics, and living in a region where the dominant languages are Arabic, Kurdish, Turkish and Persian, none of which a Neo-Aramaic speaker needs to learn for any economic purpose that matters to their children. Modern schooling is conducted in those dominant languages. Marriage pools have widened. And the process is the familiar one: a child who speaks the home language at home and the national language at school ends up, at seventeen, fluent in the national language and increasingly rusty in the other. The Assyrian Neo-Aramaic of the villages of northern Iraq is in this position now, and the Western Neo-Aramaic of Sanandaj is in a worse one, since the villages there have been steadily emptied of their young.
What happens next is not entirely in anyone’s hands, and it is not entirely unprecedented either. Languages in this position have been revived before, usually by a combination of determined community effort, a willingness to write and publish in the language, and the accidents of political change that give a minority more room than it had. The Aramaic of the Assyrians has a written standard, a small body of modern literature, a church that still uses a closely related liturgical form, and a diaspora with the resources and the incentive to do something. The Mandaeans have a ritual vocabulary that is genuinely literary and a body of priestly practice that has been kept alive deliberately for more than a thousand years. None of this guarantees survival, and it is better to be honest about that than to produce a reassuring paragraph. But it is worth noticing that the language which won an empire and then lost one, and then lost every administrative function it ever had, is still spoken, is still written, and is still being taught to children.
The last thought is the simplest one. The Arameans were scattered by the Assyrians, and their language was made the empire’s instrument and then its gravestone. It carried the correspondence of the Achaemenid state, the scriptures of the synagogue, the gospels of the Church of the East and the liturgy of the Mandaean priests. And in the end it did the one thing that languages rarely manage: it outlived the purposes for which it was learned. A tongue that began as the speech of a dozen small city states in the Syrian uplands is today the liturgical speech of a handful of villages in the Jabal al-Druze and the village talk of perhaps a few hundred thousand people, and its letters are read every morning by a quarter of the world’s population in a script that descends from it and carries no memory of where they came from.
References
- Aramaic language – Grokipedia
- Imperial Aramaic – Grokipedia
- Aramaic alphabet – Grokipedia
- Arameans – Grokipedia
- Old Aramaic – Grokipedia
- Assyria – Encyclopaedia Iranica
- Assyrian – Encyclopaedia Iranica
- Neo-Assyrian Empire – Grokipedia
- Assyria – Grokipedia
- Assyrians – Grokipedia
- Cuneiform – Grokipedia
- Akkadian language – Grokipedia
- Sumerian language – Grokipedia
- Achaemenid Empire – Grokipedia
- Behistun Inscription – Grokipedia
- Fortification Tablets – Grokipedia
- The Persepolis Treasury Tablets, by Richard T. Hallock – University of Pennsylvania Press, 1968, WorldCat
- Parthian Empire – Grokipedia
- Sasanian – Encyclopaedia Iranica
- Aramaic – Encyclopaedia Iranica
- Syriac – Encyclopaedia Iranica
- Mandaeans – Encyclopaedia Iranica
- Hatra – Encyclopaedia Iranica
- Church of the East – Grokipedia
- Syriac Christianity – Grokipedia
- Syriac Church – Grokipedia
- East Syriac Rite – Grokipedia
- West Syriac Rite – Grokipedia
- Syriac language – Grokipedia
- Old Syriac – Grokipedia
- Eastern Orthodox Church – Grokipedia
- Rabbula Gospels – Grokipedia
- Manichaeism – Grokipedia
- Edessa – Grokipedia
- Nisibis – Grokipedia
- Syriaca.org, the Syriac Encyclopedia – Syriaca.org
- The Syriac Bible – British Library
- Collection of Syriac manuscripts – British Library
- Targum – Grokipedia
- Jewish Aramaic – Grokipedia
- Peshitta – Grokipedia
- Book of Daniel – Grokipedia
- Sefaria, a library of Jewish texts – Sefaria
- Elephantine – Grokipedia
- Language of Jesus – Grokipedia
- Lord’s Prayer – Grokipedia
- Talitha kumi – Grokipedia
- Christian Palestinian Aramaic – Grokipedia
- Samaritan Aramaic – Grokipedia
- Samaritan script – Grokipedia
- Hebrew alphabet – Grokipedia
- Nabataean Aramaic – Grokipedia
- Nabataeans – Grokipedia
- Palmyrene Aramaic – Grokipedia
- Palmyra – Grokipedia
- Ge’ez script – Grokipedia
- Armenian language – Grokipedia
- Semitic languages – Grokipedia
- Neo-Aramaic – Grokipedia
- Assyrian Neo-Aramaic – Grokipedia
- Chaldean Neo-Aramaic – Grokipedia
- Sanandaj – Grokipedia
- Maaloula – Grokipedia
- Chaldean Catholic Church – Grokipedia
- Imperial Aramaic chancery texts, monographic and bibliographic search – WorldCat
- Aramaic language, monographic and bibliographic search – OpenLibrary