Authors, Books, and Readers in the Ancient World
1. Old Testament Times
Putting one's name to a written text may indicate presence as a witness, or agreement as a party to a deed, ownership of the text, responsibility for accuracy, or authorship. Each of these is attested in the Old Testament world. Authorship is necessarily admitted in letters and related to letters are prophetic communications. Numerous examples of prophecies from Mari in the eighteenth century BCE and from Assyria in the seventh century include or are accompanied by the names of those who spoke the oracles (Nissinen in Nissinen et al. 2003). Clearly there would be value in knowing which prophet's words proved to be true. Inscriptions celebrating a king's military prowess or building works were usually composed as if the monarch was speaking, ‘I am…’, but the authors were the royal scribes, all of whom remain unknown, with the exception of the writer of the formal letter to the god Ashur, ‘Sargon's Eighth Campaign’ (Luckenbill 1926–7: ii. 99, §178). Kings also declared their responsible rule by proclaiming laws, asserting that they did so at divine command. Earliest are the Sumerian ‘Reforms of Uru-inimgina’ (formerly read Urukagina), c.2350 BCE (Hallo 2000); then there are the ‘Laws of Ur-Namma’, c.2100 BCE, and of Lipit-Ishtar, c.1930 BCE (Roth 2000a, b). In Babylonia the name of the author of Laws of Eshnunna is uncertain, but Hammurabi proudly enacted his laws shortly after, c.1760 BCE (Roth 2000c, d). Later law collections do not retain their authors' names (Roth 2000e, f). Authorship is less commonly displayed in works of literature; indeed, the most extensive range of literary compositions from the Old Testament world, the Sumero-Babylonian, is largely anonymous. The authors of major compositions, such as the Epics of Creation, of Gilgamesh, of Atrahasis, and of the Descent of Ishtar to the Netherworld, are unnamed, as are those of the stories about Sargon of Akkad and other famous kings. No authors are known for the many Sumerian literary works preserved mainly in copies of the early second millennium BCE, with the exception of the hymns to Inanna (Ishtar) spoken by Enheduanna, the daughter of Sargon of Akkad, c.2300 BCE, whom he installed as high-priestess of the moon-god at Ur. Modern opinion accepts the attribution, making her ‘the first non-anonymous and non-legendary author in history’ (Hallo 1997: 519). An author who may be classed as ‘legendary’, is Shuruppak, who gives advice on life and behaviour to his son, Ziusudra, the Babylonian Noah, in ‘The Instructions of Shuruppak’. This ‘wisdom’ composition already existed among the oldest surviving Sumerian literary manuscripts copied about 2600 BCE at Abu Salabikh and lived on for another millennium or more, when there was a Babylonian translation (Alster 1974). Another sage whose advice to his son was copied in Hattusha and at Ugarit in the Late Bronze Age was the otherwise unknown Shube-Awilim (Foster 1996: 330–4). In some Babylonian works the authors reveal themselves obliquely through acrostics, where the first syllable of each line, read in sequence, spells out the name and profession, as is the case in the wisdom poem known as ‘The Babylonian Theodicy’ (Foster 1996: 790–8). In the ‘Poem of the Righteous Sufferer’, or ‘The Babylonian Job’, the speaker is named in the course of the text, although whether he was the author, or not, cannot be decided (Foster 1996: 306–23). Hymns and prayers to the gods from the third millennium BCE onwards were written in the names of kings who uttered them in rituals (e.g. Foster 1996: 68–71, 240–3, 724–44) and there are a few examples in the names of other individuals (Foster 1996: 486–94), with the names occasionally hidden in acrostics (Foster 1996: 610–11, 704–9). While those individuals may have composed the poems, it is impossible to tell whether the kings had or not; some indications may suggest that they could have done so (e.g. Foster 1996: 231, 697). The clearest statement of authorship in Babylonian texts is given at the end of the Epic of Erra: ‘The composer of its text was Kabti-ilani-Marduk, of the family of Dabibi. He (the god) revealed it to him at night, and when he spoke it while he was waking, he omitted nothing at all, not one line did he add. When Erra heard it he approved’, and uttered a blessing on everyone who recited or heard about his feats (Foster 1996: 759–89).
Beside the names occurring in the texts, Babylonian scribes in the first millennium BCE had traditions about authorship, associating gods and humans with particular works (Lambert 1962). Thus Ea, the god of wisdom, was credited with some incantations, rituals, and omen compendia, the early third millennium BCE king Enmerkar with a hymn about the date-palm, and a series of cultic personnel with a range of works including hymns and prayers, and the Theodicy whose author's name is given in an acrostic in the text itself (see above). Several identifiable poems, introduced as ‘The series of’ Gilgamesh, Etana, etc., are credited to specific individuals. The name given for the Gilgamesh Epic is Sin-leqe-unninni. That belongs to a name-type current in the second half of the second millennium BCE, so cannot be the name of the primary author, for copies of the epic survive from earlier in the millennium. The best conclusion is that this man edited older texts to produce the version that became widespread in the first millennium (George 2003: 28–33).
Analysis of some Babylonian works and the existence of manuscripts of different dates for some compositions disclose the activities of editors over the centuries. While some writings were handed down for generations unaltered, as the Laws of Hammurabi were for more than 1,000 years, changes ranging from single words through phrases to whole episodes occur in others (e.g. the Atrahasis Epic (Foster 1996: 160–203); Nergal and Ereshkigal (Foster 1996: 410–28)). There is a tendency for later copies to expand earlier texts, as in the Great Prayer to Ishtar, which ‘provides an excellent case study in how a Mesopotamian literary text could evolve or expand and still remain true in its essentials to the intentions of the original author’ (Foster 1996: 503–9, esp. 503). The changes that took place in the various manuscripts of the Gilgamesh Epic written between c.1800 and 180 BCE have been discussed extensively (Tigay 1982; George 2003: 39–47). However, a truly consistent pattern is hard to see. Having only an early second millennium copy, it would be difficult, if not impossible, to predict what changes might appear in a copy made eight or ten centuries later; nor could a second millennium version be reconstructed with any confidence at all on the basis of a first millennium copy alone. A short composition about the futility of life is preserved in Sumerian and in a Babylonian translation, in copies dating from c.1800 to 650 BCE found in Babylonia, Syria, and Nineveh which reveal ‘creative recensions’ (Lambert 1995: 37–42). Some compositions could be used in more than one of the extensive rituals, and older liturgical texts could be modified to be applied to new situations (e.g. a prayer to Shamash that had a place in a royal purificatory rite also served in the dedication of a statue (Seux 1976: 220–3); various Sumerian prayers (Cohen 1981: 36–9) or the same hymn may be addressed to one deity in one copy and to another in a second (e.g. Gula and Belet-ili (Foster 1996: 576–7)). That scribes might recognize how they were changing texts handed down to them is indicated in one case by the notation ‘Addition’ at the end of one Sumerian hymn in an extensive collection (Sjöberg and Bergmann 1969: 8, 24). The colophons of some copies of established works made in the first millennium indicate more than one exemplar, and there are commentaries which sometimes note variant readings, signs that scribes were aware of the vagaries of textual history.
The usual physical form of the book in Babylonia was the clay tablet, normally of a size convenient to hold in the left hand while the right impressed the cuneiform signs with a reed stylus. Larger tablets were laid on a flat surface for inscription, and were often ruled in vertical columns. The cuneiform signs appear to hang from horizontal lines, and those are physically ruled on some tablets. Scribes were concerned that their work should be intelligible to the readers, and usually avoided splitting a word between two lines, either crowding the signs into one line, or carrying them over two lines, usually leaving a space in the middle rather than at the end of a line. In letters scribes would include glosses on words they thought obscure—well-known examples occur in the El-Amarna Letters—but explanatory notes can be found in narrative texts also, and are clearly not added later to the compositions (Krecher 1969). From early in the second millennium onwards, Akkadian translations of Sumerian works were made, sometimes in interlinear form. Literary tablets which survive from the third and early second millennia BCE are mostly the products of pupil-scribes, copied as exercises in language, style, and accuracy. Attention to the accuracy is plain in the occasional notations ‘broken’ or ‘new break’ showing where the exemplars were defective. At the end of Sumerian compositions a title is given, ‘Hymn to X’, or an indication of the genre of song, prayer, or lament, and frequently those and other works end with a note ‘Praise the god(dess) X’ (e.g. Safati 1997; Cohen 1981). The copyists rarely added their own names to Sumerian works, but fuller information is given for three tablets of the Babylonian Epic of Atrahasis produced by a student c.1635 BCE, forming a classic colophon: name of the composition, number of the tablet if more than one, number of lines on the tablet with the total on the last tablet, scribe's name, date (Lambert and Millard 1969: 31–2). Similar colophons indicating who was responsible for the copy appear later in the second millennium and in the first, sometimes adding the name of a second scribe who collated the copy. Extra lines may state that the tablet was made for an individual or for a specific library, for reciting, singing, or reading (Pearce 1993). There may also be a note of ownership; Ashurbanipal's name and title were stamped on some tablets made for his library at Nineveh, and a curse might be added on anyone who stole the tablet (Hunger 1968).
Beside clay tablets, cuneiform was written on wax-covered wooden boards (le'um) which could be joined by hinges into pairs or longer sets. They were in use from the end of the third millennium BCE and were primarily intended for running accounts, registers, and similar administrative records. However, these boards could also contain compositions of a more literary type, and the colophons of some clay tablets state that they had been copied from writing boards. Actual examples of writing boards survive only from the first millennium: the ivory specimens from Nimrud which held the omen series Enuma Anu Enlil extending to some 5,000 lines, wooden fragments from Nimrud, and a wooden board from Assur (Wiseman 1955; Klengel-Brandt 1975).
The complexity of the cuneiform script meant that reading was a skill confined to those trained in scribal schools, some of whom may have progressed from the scribal profession to take other offices in temples and royal courts. Where colophons give the names of the owners of tablets, they were often professionals engaged in religious or semi-religious functions (exorcists, singers, diviners) as well as priests and scribes (Hunger 1968). Contrary to often repeated statements, books were not the prerogative of temples and palaces, a deduction made partly from the recovery of Ashurbanipal's ‘library’ at Nineveh in the earliest days of Near Eastern archaeology. Continuing excavations throughout Mesopotamia have recovered numerous collections of clay tablets from private houses of all periods, which, while mostly comprising business and legal documents, sometimes include literary works (Pedersén 1998). The tablets were stored in baskets, jars, or on shelves, and labels from such containers list literary compositions that were once in them (Hoffner 2002). Some of the householders belonged to the professions noted already; some were scribes who made their homes into schools, where they taught their craft to apprentices. Sumerian essays describe the experiences of such apprentices, their curriculum and stages of progress (Tinney 1999). They not only copied existing works, they also made up their own, for schooling involved the creation of didactic works, like the Sumerian ‘debates’ (Vanstiphout 1997).
An author's words, written by himself or by a scribe, would remain unknown unless they were copied or read by others. How a work reached an audience, or who that audience might have been, is not clear. Rituals required texts that would be recited regularly and precisely, and cultic changes could demand new texts, of which the Babylonian Creation Poem is a prime example, resulting from the pre-eminence given to Marduk late in the second millennium BCE (Foster 1996: 350–401). Individual scribes might compose hymns or prayers of personal piety for themselves or for clients; some prayers have a space left for an individual's name to be inserted, or the words ‘So and so’. How and why the major Babylonian literary works came to be composed, and then written, is obscure; there is no way to tell whether the written forms were the end result of lengthy oral traditions, or the compositions of individual scribes who put them into writing immediately (Vogelzang and Vanstiphout 1992). The praise of particular gods may have been one purpose of the Epic of Atrahasis, a story which explains how Babylonian society came to be (Atrahasis III. viii. 9–18; Foster 1996: 185). The Epic of Gilgamesh tells a good tale with a variety of characters and adventures, aiming to give the lesson that man is mortal (George 2003: 33–9). Questions about the cycle of life and life after death may lie behind the stories of deities descending to the underworld and either remaining there (Nergal and Ereshkigal), or supplying a substitute (The Descent of Ishtar). Unhappy historical circumstances in Babylon are apparently explained by the Erra Epic, and tales about kings of the past, like Sargon of Akkad (Foster 1996: 103–8, 251–7, 803–4), may have carried a sense of nostalgia. Some lines recur in two or more compositions (Reiner 1985: 32–3), and scribes sometimes quoted lines from standard works in letters; but who, apart from them, would recognize such allusions remains unknown. Most of the Sumerian compositions recovered are products of the schoolroom in the Old Babylonian period, but whether they had their origin there or were copied from tablets written in other contexts is unknown.
Most of ancient Egyptian writing is anonymous. Accounts of the pharaohs' achievements were composed at royal command by chosen, but unnamed, scribes for carving on temple walls and other monuments. To celebrate Ramesses II's famous battle with the Hittites at Qadesh (c.1275 BCE), a lengthy poem was created with the king as narrator, supplemented by a more prosaic, third-person narrative, both accompanying the pictorial presentation. The whole has been judged Ramesses' inspiration, carried out by his best scribes and artists to produce ‘a unique phenomenon in Egyptian literature’ (Gardiner 1960: 46, 47, 53; Kitchen 2000).
From early in the Sixth Dynasty, c.2300 BCE, come private tomb memorials relating notable deeds in the lives of the deceased, written in prose with some poetic passages, and these continue until the end of ancient Egyptian history (see Kitchen 1999). Undoubtedly the majority were composed by scribes on the basis of personal information provided. Most famous among these ‘autobiographies’ is the story of Sinuhe, the account of an exile's life in Canaan and eventual return to honour in Egypt in the twentieth century BCE. The oldest manuscripts of this text were copied about a century after the time of the setting of the story (Lichtheim 1997a). No tomb of Sinuhe has been found, so debates continue about the nature of the narrative, whether it is a straightforward tomb inscription, an adaptation of one to promote political interests, or a piece of fictional propaganda, the last appearing unlikely in view of the accuracy with which life in Canaan is portrayed (see Posener 1956; Parkinson 1997; Kitchen 1996b). Similarly, the report of Wenamun (Lichtheim 1997c) has been characterized as a factual, first-person account of a journey to Byblos to procure cedar wood c.1075 BCE (Čzerny 1952), but some Egyptologists treat it as a ‘novelistic’ composition (see Gardiner 1961: 306; Eyre 1996). The New Kingdom has provided a number of small funerary stelae set up by individuals which include ‘psalms’ praising the deity, pleading for favour and even suffering punishment from the god for sin and being rescued by him. In some cases there are clearly personal circumstances which imply either custom-made composition by a scribe or creation by the individual himself (examples in Kitchen 1999: 269–314).
The books for which authors are regularly named are the books of ‘wisdom’, for effective instruction requires authority, and ‘wisdom’ requires personal authority. Significantly, a thirteenth-century BCE scribe who lauds the immortality that authorship confers by saying that authors' names live on in the books they made, then cites eight names, four from the Old Kingdom and four from the Middle Kingdom (Wilson 1955). Two of the former are the authors of surviving ‘wisdom’ books, Ptahhotep and Hardedef (Wilson 1955), one, Kaires, may have been the father whose name is missing from the Instruction for Kagemni (Lichtheim 1973–80 : i. 59–61), and the fourth was the polymath Imhotep, to whom no extant books can be attributed. Two of the Middle Kingdom names are attached to ‘books of instruction’: The Prophecy of Neferti and The Complaints of Khakheperre-Sonb (Shupak 1997). One was said to be the author of the Instruction of Amenemhat (Lichtheim 1997b) and the fourth name has been associated with the Loyalist Instruction (Posener 1976). These authors' names were probably known because their works were copied in scribal schools (Lesko 1994: 138–43). The names of other authors are preserved in their works: Ipuwer, Amenemope, Amennakht, Dua-Khety, Any, Ankh-sheshonqy, mostly, again, known from the copies made by pupil-scribes, sometimes very corrupt. How far these names truly represent the authors of the works is debated. The Middle Kingdom language of Ptahhotep and Kagemni has led some to the conclusion that they are pseudepigrapha, although reflecting the Old Kingdom period (Lichtheim 1973–80: i. 6–7). Other scholars see no reason to doubt these attributions, the writers having held posts at various levels, some in the royal court, others as teachers (see Kitchen 1998: 346–9). When a king is said to have composed instructions for his son, as in the case of the Teaching for Merikare (the father's name is damaged; the date is a little before 2000 BCE), it may be assumed that the book was written at the king's behest; the extent of his own contribution cannot be known, but the fact that there is evidence for literate kings (see below) suggests they could have involved themselves in creating such works. Beside the ‘wisdom’ works, some Egyptologists identify a specific genre, the ‘Königsnovelle’, narratives in which the king is hero, solving an awkward problem, although its existence may be doubted (Loprieno 1996b; Leprohon 2001; contrast Redford 2001a; for a list of the Egyptian texts, see Redford 1992: 374–7).
Editorial activity can be seen or assumed within scribal circles. The copy of the Maxims of Ptahhotep, made about 2000 BCE, is written in Middle Kingdom Egyptian, not the language of the Old Kingdom, as Hardjedef is, and so, assuming the book is an Old Kingdom composition, the language has been modernized. By the thirteenth century BCE that version was so archaic that a thoroughly revised text was created (Lichtheim 1973–80: i. 61–80). In other cases modernization of language is evident when New Kingdom expressions replace Middle Kingdom ones or old ritual texts are rephrased in later language. Some of these substitutions may have been made unconsciously, but most will have been deliberate. Sufficient knowledge is now available about the history of the Egyptian language from a wide range of texts to enable such distinctions to be made.
The availability of stone in Egypt meant that multitudes of texts were carved in tombs, on rock surfaces, and on stelae. They are mainly royal inscriptions or funerary memorials, the latter often containing prayers and sometimes biographical narratives. Books were written on rolls of leather or, more commonly, papyrus; the oldest papyrus roll known is a blank one, buried in a tomb for the owner's use c.3000 BCE. The papyrus sheets were made from strips of pith peeled from the papyrus reed and laid side by side with a second layer on top at right angles to the first. The two layers were pressed or hammered together, the sap forming a glue, then the surfaces polished with a smooth stone. Writing was normally done along the horizontal fibres. The sheets were approximately 30 × 20 cms. (11 × 8 in.), glued side by side to make rolls on average 4 m. long, but extra sheets could be added at will, and one roll is 40.5 m. (133 ft.) long. The full height of the sheet was not often used for literary texts; rolls were bisected or cut even shorter, down to 10 cm. (4 in.). Excerpts from books were copied on ostraca, or wooden tablets, probably as writing exercises, so that some compositions are represented by many manuscripts of varying extent and quality. The texts were written with black ink made from soot and gum, with headings, opening phrases, and rubrics in red ink, made from red ochre. Exercise texts sometimes have red dots to show division of phrases or accentuation. Not all owners cared for their books. When papyrus was not readily available, the roll might be turned over and its back used for something else, or the ink washed off, or pieces might be cut from a roll and reused.
Besides the ‘wisdom’ works with named authors (see above), there is as great a variety of Egyptian literature as of Babylonian. The books range from technical treatises on topics such as mathematics, medicine, dream interpretation, and magic spells, through love poems and wisdom literature, to matter-of-fact narratives like Sinuhe and Wenamun, to fantastic tales like Cheops and the Magicians and the Shipwrecked Sailor up to myths about the gods and hymns to them. (For a catalogue of surviving Egyptian literary manuscripts, see Quirke 1996.) Absent from Egypt are collections of laws, perhaps because, in theory, the pharaoh's word could not be challenged, although records of lawsuits survive, and there were archives of past cases which could be consulted. Also absent are continuous chronicles of events covering more than one reign, distinct from individual kings' annals, the ‘demotic chronicle’ for the third century BCE being exceptional (Spiegelberg 1914).
In Egypt most scribes were occupied with administrative tasks; tomb models and tomb paintings show them listing deliveries, estimating yields, or ready to record their masters' orders. The able or well-connected might leave behind such mundane tasks as they rose through the hierarchy to positions where they employed secretaries of their own. The highest in the land might claim scribal skills: Ramesses II's monuments report that he ‘researched in the office of archives and unrolled the writings’, and his son Merneptah was described as ‘the Royal Scribe with skilled fingers’ (Kitchen 1996a: 183, 214). Tutankhamun's tomb was furnished with many imitation writing palettes, but a real one, gold-plated, from early in his reign and an ivory one from a later year bear his name and had been used, presumably by the boy-king (Carter 1933: 79, 80). Literary compositions present kings as able to read and write as a matter of course: e.g. the Middle Kingdom Prophecy of Neferti 15 ff., in which the king writes the words of the sage (Shupak 1997: 107).
Scribes were trained in ‘the house of life’ attached to temples or administrative offices in larger towns, and perhaps simply by following professional scribes in smaller places. When they graduated, they might copy books for patrons, for temples and for themselves. The activity of one scribe who composed five short poems and a brief ‘wisdom book’ in the twelfth century BCE has been traced, together with the work of some of his pupils (Bickel and Mathieu 1993). It seems to be the personal copies that have been recovered, and in a relatively few cases their provenance and treatment are known. Books might be placed in tombs for the delectation of the dead in the afterlife—copies of several classic literary works are thought to have been found in the Twelfth Dynasty tomb of their Theban owner (Parkinson 1997: 1). One letter tells of depositing papyri in an ancestor's tomb in the twelfth century BCE, although it is not clear whether that was for the benefit of the dead, or simply for safe-keeping (Pestman 1982: 156–7). Those papyri would have been kept in their owners' houses with the letters and papers of daily business, as found in the Twelfth Dynasty workmen's village at Lahun (Collier and Quirke 2002: pp. x–xiii) and in the Nineteenth and Twentieth Dynasty tomb-builders' village at Deir el-Medineh. The latter collection, accumulated within one family for over 100 years, included technical compositions (medical and magical prescriptions, a book of dream interpretations) and literary works, among them the Contendings of Horus and Seth, the Hymn to the Nile, and love poetry (Pestman 1982). In addition, hundreds of ostraca found in and around the village bear extracts from literary writings, the products of scribes amusing themselves, practising or quoting texts, or, most commonly, the exercises of their pupils. The papyri and ostraca together, the books from Deir el-Medineh, it has been said, ‘represent most of the literature surviving from ancient Egypt’ (Lesko 1994a: 133). Whether this collection would have been duplicated widely in other settlements in the Nile Valley, or only in major towns, is impossible to say, just as it is impossible to know how widely the literature was known outside scribal circles. While it has been said that although ‘99% of the documentation from the most important of Egypt's cities, towns, temples, and other work sites is missing—jar labels etc. indicate literacy extended to household servants’ (Lesko 1994a: 134), most ancient Egyptians, if they were familiar with the contents of any of their civilization's literature, had learnt it from oral presentations.
(c) The Levant
Scripts and Texts
Sumero-Babylonian texts were copied at Ebla in north Syria c.2300 BCE and at many sites during the Middle and Late Bronze Ages. There was probably a scribal centre at Aleppo in the eighteenth century BCE disseminating Babylonian traditions so that secretaries in several towns as far south as Hazor read and wrote cuneiform (Wiseman 1962). In the Late Bronze Age cuneiform was current across the region for diplomatic and local affairs. (For cuneiform texts found in Palestine, see Horowitz, Oshima, and Sanders 2002.) The El-Amarna archive from Egypt has letters from dozens of rulers of Levantine towns, and includes some copies of Babylonian books used for teaching Egyptian scribes to write in Babylonian. That was probably the purpose, too, of the fragment of the Epic of Gilgamesh found at Megiddo. Beside these imports, there is no native, ‘Canaanite’ literature written in cuneiform. One Egyptian legend seems to echo a Canaanite myth (Ritner 1997), but it is unlikely that Canaanite scribes used Egyptian for writing their own stories. If scribes were writing out literary compositions in Late Bronze Age Canaan, they were probably using the infant alphabet, developed during the previous centuries, but writing in ink on papyrus or leather rolls or on wooden boards, which have not survived. This supposition arises from the situation at Ugarit. There, scribes schooled to use Babylonian language and script learnt of the Canaanite alphabet and created a cuneiform alphabet on its pattern, for writing on clay tablets. Examples of their work include several books: the Epic of King Keret, the Story of Aqhat, the Ba‘al cycle of myths, and the myth of Dawn and Dusk (Pardee 1997). With the fall of the Late Bronze Age towns of the Levant, this script disappeared. The tablets from Ugarit demonstrate the existence of an indigenous literature in coastal north Syria beside the Mesopotamian cuneiform repertoire. These tablets date from the thirteenth century BCE, evidence for the existence of any local written literature at earlier dates has not been found (Millard 1999).
The Aramaean kingdoms of the first millennium BCE adopted the alphabet, but inevitably the only lengthy texts available today are the royal monuments on stone. The earliest is the Tell Fekheriyeh Statue, c.840 BCE, which is unusual in being largely a translation from an Assyrian original (Millard 2000a). The three Sefire stelae (c.750 BCE; Fitzmyer 2000) preserve the terms of a treaty with curses on the oath-breaker which evidently continue old and widespread forms. About 800 BCE someone at Tell Deir ‘Alla in the Jordan Valley copied a column of a book roll on to the plaster of a wall. The script is Aramaic, each word is separated from the next by a point, and certain phrases are written in red ink. A word omitted by haplography could be inserted above the line (Millard 1982: 148). The language is a local dialect related to Aramaic. The opening words serve as the title, ‘The Book of Balaam, son of Beor’. Here is the oldest example of a book in a West Semitic language written with the alphabet, and the oldest piece of Aramaic literature. It narrates visions of catastrophe given by the gods to Balaam (Levine 2000). The next piece of Aramaic literature, chronologically, is a badly damaged composition written on the walls of a tomb at Sheikh Fadl in Egypt in the fifth century BCE. The passages that can be read include references to rulers of the seventh century BCE, religious expressions, and declarations of love (Lemaire 1995). By this time the words are simply separated by spaces. From about the same time come the papyri from a colony of Aramaic speakers on Elephantine Island in the Nile, near Aswan. Apart from small pieces, there are two major compositions, one a version of the ‘Behistun Inscription’ of Darius I, the other the Wisdom of Ahiqar. Each was contained on a papyrus roll, the Ahiqar text being written on the back of an old account book which was about 13.5 m. long, although Ahiqar itself occupied only twenty-one columns or so, needing about 5.6 m. This is the oldest example of a West Semitic book roll (Porten and Yardeni 1993: C 1–2).
Some of the Ugaritic literary tablets concerning Ba'al, Keret, and Aqhat have an initial note lb‘l, krt, l'qht. While similar to the notes in the headings of some Hebrew psalms, these are not to be treated as notes of authorship; they simply designate the composition they belong to (Pardee 1997: 241 n. 4; 268 n. 241). Similarly, the Tell Deir ‘Alla text opens with ‘The book of Balaam’ (spr bl‘m), but there is no indication that he was reckoned its author, unlike Ahiqar, which starts, ‘These are the words of Ahiqar’ (Porten and Yardeni 1993: 26–7).
Among the literary tablets from Ugarit, a few end with colophons naming the scribe Ilumilku, a highly placed official in the royal court during the last days of Ugarit (Millard 1979: 613; Dalix 1999: 15). Whether he wrote them for reading to his master, or for some other purpose, is unknown. Who would read the Balaam text, or in what context, is not clear, for the nature of the building where it stood is uncertain, and the purpose of the Sheikh Fadl text is equally obscure. The Ahiqar and Behistun papyri were presumably copied by scribes or people able to write for their own education and delectation, perhaps for reading aloud to others.
(a) Extra-biblical The existence of Israelite literature much earlier than the seventh century BCE—indeed, of any extensive written texts in Israel—has been denied (Thompson 1992: 391; Blenkinsopp 2001: 41), but the epigraphic evidence from excavations there and the consistent testimony from other cultures contradicts that. A few fragments on potsherds and wall plaster imply the production of longer texts which would have been written on perishable papyrus or leather (Millard forthcoming).
(b) Biblical The Hebrew Bible is unrivalled as a collection of books from a single culture in the ancient Near East. No other culture is known to have created continuous historical narrative, a unified account of the nation's origins and establishment, such lengthy compilations of prophets' words, or so large an anthology of hymns. At the same time, the Hebrew books share forms and patterns which were current in other ancient Near Eastern societies. The books of the Hebrew Bible were produced over a long period and show great variety of content and differences of style (see Part IV in this volume).
The attachment of personal names to several biblical books has given the mistaken impression that those persons wrote the books bearing their names. In most cases nothing in the texts supports such a claim; the names indicate the, or a, major actor in the book: e.g. the books of Joshua, Samuel, Job, on the same pattern as the books of Kings. The ascription ‘Books of Moses’ is no different; Moses is the key figure in the Pentateuch. Just as the notations on Ugaritic tablets indicate their subject-matter, so the headings of many Psalms, ‘of David’ (ldwd) may celebrate him, although they may equally well signify authorship. Where individual authorship may be claimed is in books of wisdom and prophecy which are, significantly, the same types of composition for which authors are named in Babylonia and Egypt. In Wisdom literature the Proverbs of Solomon is a prime example, giving the king's name both in the initial sentence (1: 1) and at the head of chapter 25, where additions were made in Hezekiah's reign. As with Egyptian instructional works especially, the name need not denote the originator of every sentence, but rather that the collecting was done at his behest or in his honour. That does not mean that he had no part at all in the composition.
Prophetic collections indicate the speaker of the oracles: e.g. ‘The words of Amos’, without requiring him to be the author. Habakkuk was told to write his vision (2: 2), and Isaiah was ordered to write his unborn son's name and to seal a text (Isa. 8: 1, 16), and one of his visions is described as ‘words sealed in a scroll’ (29: 11), suggesting that prophets may have recorded their messages themselves. However, Jeremiah may not have been the only prophet who used a secretary (Jer. 36, etc.).
Nehemiah's first-person memoir, unique in the Hebrew Bible, may be compared with the Egyptian ‘autobiographies’ usually found on statues and in tombs (see above and, for the Persian period, the inscription of Udjahorresne, Lichtheim 1973–80: iii. 36–41).
Within the Hebrew Bible the first reference to writing in a book is Moses' record of the Amalekite war (Exod. 17: 14). Thereafter books are integral to Israel's culture, for religious and secular purposes (e.g. the Book of the Law, Josh. 1: 7–8; the Book of Jashar, 2 Sam. 1: 18; ‘the books of the chronicles of the kings of Israel’ and ‘of Judah’, 1 Kgs. 14: 19, 29, etc.). There is no reason to doubt the production of books throughout the history of Israel, written in the limited alphabet inherited from the Canaanites. Leather or papyrus rolls could contain books of any length, and the epigraphic evidence is sufficient to demonstrate the presence of writers from the days of the conquest onwards (Millard 1998a, forthcoming). The inscriptions show that the Old Hebrew script was normally written with a point separating each word from the next, but with no compunction about splitting a word between two lines (Millard 1970). When the Aramaic writing practices of the Persian Empire were adopted, resulting in the ‘square’ script, spaces were used to separate words, and words were not split across lines.
Intense analyses of the Hebrew books which attempt to discover sources and literary forms have been conducted within the texts, without external controls—that is, with relatively little attention to the abundant literature of the ancient Near East. Yet the Hebrew books conform in many ways to the customs known from elsewhere in the ancient Near East, although the books themselves are not attested in their present forms earlier than the second century BCE (the Dead Sea Scrolls). The lack of earlier copies renders any attempt to trace stages in a history of composition wholly hypothetical; that has proved possible for some Babylonian and Egyptian books only when manuscripts of different dates are available (see above). When the compositions in Egyptian and in cuneiform are taken into account, the Hebrew works can be objectively understood as arising from the same milieu, created in similar ways, and sharing similar features. Written accounts of political or military events, or prophetic utterances, could be made at the time they occurred and might be made simultaneously in prose and in poetry, as with Ramesses II's Qadesh narrations (see above) and for Tukulti-Ninurta I's conquest of Babylon (Grayson 1972: 108–34; Foster 1996: 211–29). Literary forms have analogies elsewhere (e.g. the Psalms; Gunkel and Begrich 1926; Gunkel 1967), and whole genres are clearly comparable to those in Egypt (Proverbs, Song of Songs; see also Chapter 5 above), while within the books, inclusio and chiasm, for example, mark discontinuity, and resumption of nouns etc., continuity (e.g. Baker 1983). At the same time, there are significant differences, notably in the continuous story of Israel from the exodus to the exile, with which the Babylonian Chronicle is only partly comparable, unlike the first-person records left by individual kings, and principally in the consistent theological perspective of history, hymnody, wisdom, and prophecy.
Discussion of the relationship between oral and written tradition is inconclusive. The existence of oral tradition throughout biblical history is certain, reaching to the ‘tradition of the elders’ in the Gospels (see Mark 7). There can be no conclusive identification in the written texts of passages that may have originated orally, for supposedly oral forms within the written text can do no more than point to an oral origin for those forms; a scribe could incorporate them in a new written composition without the necessity of any oral precursor. The weight given to written texts throughout the ancient Near East implies that oral versions carried less authority. Spoken words could be put into writing immediately in almost every circumstance, as is especially evident for prophecy (see Nissinen et al. 2003), and there would not be many places in Israel which lacked someone able to read or write. (See Niditch 1997 with Millard 1998b.)
According to the biblical texts, public reading of the Law took place at Sinai (Exod. 24: 3, 4, 7; 34: 32), was expected every seven years once Israel was settled in the Promised Land (Deut. 31: 9–13), was envisaged in the prescription for the monument on Mount Ebal, fulfilled after the conquest (Deut. 27: 1–8; Josh. 8: 32–5) and was reported on various occasions (2 Kgs. 23: 2; 2 Chr. 17: 9; Neh. 8). Jeremiah's prophecies were read to officials and to people in the temple (36: 10, 13–15). Two occasions are reported when books were read to a king and his court: the ‘book of the Law’ read to Josiah (2 Kgs. 22: 10) and Jeremiah's prophecies read to Jehoiakim (Jer. 36: 21–4). Xerxes, king of Persia, sleepless, had the records of his reign read to him, either to induce sleep or to occupy his mind (Esther 6: 1). The Israelite king was expected to provide himself with his own copy of the Law and to read it himself (Deut. 17: 18, 19). Other examples of individuals reading are rare, but Isaiah speaks of presenting a sealed scroll to one who can read and to one who cannot (29: 11–12). Whether or not credence is given to the biblical texts, they carry the assumption of their various authors that writing was accessible at the moments they describe. Although reading and writing were largely confined to the élite, the twenty-two-letter script was not hard to learn, and the epigraphic evidence, notably the graffiti, again suggests that there were people who could read scattered across the land, not confined to palaces and religious centres (Millard forthcoming).