In 1979, excavating a tomb in a cave on the western slope of the City of David, the archaeologist Gabriel Barkay uncovered two small objects that were, at first, easy to overlook: discs of silver no bigger than a quarter of a coin, found rolled up as tightly as a miniature. When they were unrolled in the Israel Museum, tiny lines of ancient Hebrew script appeared in the metal. The text on one of them turned out to be a version of the priestly blessing from the book of Numbers – the blessing that begins “The Lord bless you and keep you”, the formula that stands at the heart of Jewish prayer and at the centre of Christian ordination – and the two objects are now generally known as the Silver Scrolls of Ketef Hinnom. They are among the oldest surviving texts of the Hebrew Bible, and among the smallest. They were also, in a sense that matters for everything else in this essay, an accident: they are the visible fraction of a much larger body of writing on metal that has mostly vanished, and that included curses. This essay is about that vanished body. It takes the Silver Scroll as its point of entry and then moves outward through the archaeology of Hebrew writing on metal – the lead curse tablets of the late Second Temple period, the copper scroll of treasure found at Qumran, the disputed lead tablet from Mount Ebal – asking what such objects tell us about the world they came from, and why the prayers and the curses they carry survive in silver rather than in parchment.<\/p>\n
The Ketef Hinnom scrolls are conventionally described by their findspot and by their material: two sheets of silver, inscribed in the paleo-Hebrew letterforms of the period, recovered in 1979 from a burial cave in the funerary complex at Ketef Hinnom, on the southwestern slope of the valley that runs up from the Kidron towards the city, and now held by the Israel Museum. Gabriel Barkay, who supervised the excavation, has described the finds as silver scrolls no bigger than the diameter of a quarter, originally worn as amulets around the neck. Their date, on paleographic grounds, is generally placed in the late seventh or the early sixth century BCE, in the closing decades of the First Temple period and the years immediately before the Babylonian destruction of Jerusalem. The edition of the texts that has become standard is the one prepared by Gabriel Barkay, A. G. Vaughn, M. J. Lundberg and B. Zuckerman, published in 2004 in the Bulletin of the American Schools of Oriental Research as “The Amulets from Ketef Hinnom: A New Edition and Evaluation”.<\/p>\n
Barkay’s team was cutting into the row of rock-cut burial chambers that face westward over the valley from the shoulder of Hinnom, and the chamber that produced the scrolls had been sealed, in antiquity, by a partial collapse of its ceiling, which is why the tomb robbers above it had missed it altogether. Beneath the burial bench they found a repository holding more than a thousand objects, and among them two coils of silver, the second recovered a week later when the soil across the chamber floor was sifted by hand. What broke the seal was not a trained eye but a boy of thirteen on the dig, who noticed a patch of disturbed fill in the corner of the chamber and struck it with a pick.<\/p>\n
Neither object could be read in the field, and it looked as though they never would be. The silver had been rolled so tightly that the two coils had fused into single masses of metal, and any attempt to open them would have destroyed the inscriptions on the inner face of the roll. Museums in England and Germany were offered the chance to try and both declined, judging the risk too great. Three years after the discovery the work was given to Joseph Shenhav, a conservator at the Israel Museum, who stripped away the corrosion and unrolled the coils under a microscope with surgical instruments, a fraction of a millimetre at a time over months that ran into years. Even then the letters were too faint to be read off the metal: it was Ada Yardeni of the museum, redrawing the scrolls for a display in 1986, who noticed the divine name written three times and found the prayer of Numbers in the gaps.<\/p>\n
What the two sheets carry is the blessing of Numbers 6:24-26, and the two handle it very differently. Unrolled, the larger, KH1, measures twenty-seven by ninety-seven millimetres and carries eighteen lines; the smaller, KH2, measures eleven by thirty-nine millimetres and carries twelve. On KH2 the blessing runs close to the received text, clause by clause, preceded by a prayer for protection from evil. On KH1 it sits inside a longer sequence, with epithets of God before and after it. Neither sheet reproduces a text; both are the work of someone who knew the blessing and needed it on metal. Rolled up again, they are objects the size of a thimble, and the smaller is not much larger than a coin, which is the phrase in the heading of this section and is not a figure of speech.<\/p>\n
What the scrolls contain is a variant of the priestly blessing of Numbers 6:24-26, the three verses that are recited at the end of every Jewish service and that stand behind the Christian blessing of the ordained ministry. Set out in the form in which it survives on the silver, the blessing runs through the divine name and the three clauses of blessing and the closing formula. It is recognisably the biblical text, and it is recognisably not identical with it: the wording of the Ketef Hinnom version departs from the received text in places, and it is written in a script and in a spelling that belong to the centuries before the Masorah fixed the biblical text in its present form. For this journal the significance is immediate. Here is a witness to the biblical text that is older than any of the great Masoretic codices, older than the Aleppo Codex and the Leningrad Codex, and independent of them, carried not on parchment but on a metal that will outlast any parchment by centuries.<\/p>\n
That last point is worth dwelling on, because it changes what a manuscript is. A codex of the Hebrew Bible is a large object: hundreds of leaves, quires of parchment sewn and bound, a collaborative effort of a scriptorium and a bindery, an object with a shelf life of centuries and a working life of generations. The Ketef Hinnom scrolls are the opposite of that. They are private, portable, precious and tiny. A sheet of silver was cut, inscribed with a stylus in letters too small to read comfortably, and rolled. What kind of writing is that? Not a book. Something closer to a talisman, carried against the body, whose virtue lay in the words and in the material together, and whose owner might never have expected to read it.<\/p>\n
The silver itself is a statement. Silver was expensive. It was the material of coin and of the vessels of the Temple, and a sheet of it in an Iron Age tomb is conspicuous. The biblical materials themselves associate precious metal with holiness and with the spoils of war that must be purified before they may be put to sacred use, and the practice of inscribing protective words on metal is not a late invention but a technique that the period knew well. The Ketef Hinnom objects also raise the question of reuse. Sheet silver was not cheap to discard, and the amulets were made from scrap. The letters were cut with a point into a metal surface that had already served some other purpose, or was destined for the furnace. The excavators had to work slowly, unrolling sheets that had been pressed tight for nearly twenty-five centuries, and the preservation is a consequence of the material’s value: the silver was not thrown away, because throwing away silver was not something one did.<\/p>\n
The findspot tells its own story as well. The scrolls came from a burial cave in the complex of chambers cut into the slope at Ketef Hinnom, and they were deposited with the dead – which places them in a category of their own, neither the household shrine nor the temple treasury nor the rubbish pit, but the tomb. Texts buried with the dead do one particular job: they accompany. They are put where the person they address will be, so that the words will be near them at the moment when the words can no longer be read by anyone with eyes. The Ketef Hinnom deposit is unusual not because the practice is rare but because the conditions preserved it: a sealed chamber, a floor of debris, and metal that does not rot. Most of what the ancient Levant wrote on metal was written to be read once, by one person, and then to be destroyed. What we have from Ketef Hinnom is what the practice left behind when it was interrupted rather than completed.<\/p>\n
It is worth pausing on the interpretive problem that the Ketef Hinnom scrolls pose for the study of amulets, since it bears directly on the question of who wrote and read the metal scrolls of the ancient Levant. The letters are minute. They were written for a wearer to carry rather than for a reader to study, and their smallness means that their meaning was not entirely in the text. It was in the text, and it was in the weight, and the shine, and the contact with the skin. As the amulet moved against the body, the words were replayed in the mind of the person who carried it, whether or not they were consciously read. That is a mechanism of devotion quite different from the one this journal usually describes, in which a text is copied by a scribe, collated by a reader, annotated by a commentator and printed in a city. The Ketef Hinnom scrolls sit at the far end of that spectrum, and they remind us that the spectrum exists.<\/p>\n
The blessing on the Ketef Hinnom silver was not the only thing the ancient Levant wrote in metal, and it was not the most aggressive. The other great class of metal texts is the curse tablet, and the oldest candidate for one is a small lead sheet found on Mount Ebal, the mountain across the valley from Jericho that the Bible associates with the blessings and the curses pronounced by the tribes of Israel. The tablet was reported in 2022 by a team associated with the Associates for Biblical Research, and it was announced with some excitement, because if the reading and the dating hold, the object would be a curse tablet in proto-alphabetic script and at least three thousand two hundred years old – which would push back the earliest evidence for literacy among the ancient Israelites by several centuries, the previous benchmark being the Khirbet Qeiyafa inscription of about the tenth century BCE.<\/p>\n
The claim has to be handled with the care that such claims now require. The tablet was not found in a clean archaeological context; according to the archaeologist Israel Finkelstein, who was not involved in the work, it came from a pile of debris left by an excavation carried out in the 1980s. Finkelstein questioned the proposed date, noting that comparison with other sites dated by radiocarbon analysis suggests the mountain’s occupation may belong to the eleventh century BCE rather than the twelfth, and he questioned whether the decipherment would bear another interpretation. No detailed peer-reviewed analysis was available at the time of the announcement, and Finkelstein’s summary of his position – irritation at “sensational claims of discoveries which ostensibly change everything that we know about the Bible and the history of ancient Israel” – is a fair description of how the field received it.<\/p>\n
What is not in dispute is the general point, which is older than the Mount Ebal tablet and better documented: curse tablets are a recognisable artefact class of the Second Temple period. The specialists who work on them describe a conventional form. A sheet of lead, folded, inscribed with a stylus in small letters, and deposited where it was meant to be found – in a tomb, under a threshold, in the foundations of a building, beside the body of the person it was meant to affect. The texts follow a recognisable pattern: a formula invoking the name of God, an addressee identified by name and usually by patronymic or place, the statement of the wrong, and the consequence – that the curse should fall on the person named, on his children, on his possessions, or on some part of his body. The language is blunt and repetitive, and it is often heartbreaking in its specificity, because the wrongs recorded are small and domestic. A curse on a neighbour who has withheld a deposit; on a man who has taken an oath and not kept it; on a woman who has wronged her husband’s household. The taboos are the ordinary taboos of a village society: property, reputation, boundaries, the family.<\/p>\n
And here the narrative that survives in the Hebrew Bible is the ancestor of the form. The blessings and curses of Deuteronomy 27 and 28, spoken from the twin hills of Gerizim and Ebal, are a repertoire of sanctions on the life of the community: on the man who does not bring the tithe, on the one who moves a boundary stone, on the one who takes a pledge from a poor neighbour and does not return it. The curse on one who “has cursed the Lord” is the earliest and the plainest: a word spoken against God cannot go unpunished, and the punishment is asked for in the imperative, “let it be as if he had not done it”. A late Second Temple curse tablet is a small, private, commercial application of that logic: the same grammar, the same retributive logic, applied to a neighbour instead of to a people.<\/p>\n
It is worth noticing how the scriptures themselves deliver a curse, because they almost always deliver it by voice. The blessings and curses of Deuteronomy 27 and 28 are formulae in the plural imperative, chanted by the assembled tribes, “cursed be the man who does this”, to be answered with “Amen”. Nothing is written, nothing is deposited, nothing is picked up by a wronged neighbour. The tablet inverts that arrangement: it is private, it is silent, and it is activated by whoever happens to find it.<\/p>\n
The Hebrew Bible itself knows a written curse, and it knows it in a sanctuary setting, which is worth noticing because the later tablets are personal versions of something that the scripture had made official. The fifth chapter of Numbers describes a ritual for a woman suspected of infidelity: the priest takes her into the tent, writes the names of the accused and the suspected on a scroll, and casts the scroll into water mixed with dust from the floor of the tent, with the words that invoke the woman’s own testimony against her. The written word there is not a record of a dispute but the instrument of the dispute, the means by which a case is put to the deity. The late antique curse tablets are the domestic counterpart of that rite: the same confidence that a word properly written and properly deposited will do what it says, applied to a neighbour rather than to a spouse, and applied in a domestic shrine – a tomb, a threshold, the foundations of a house – rather than in the tent of meeting.<\/p>\n
It is a striking thing that the medium for this is lead. Lead is soft, cheap, malleable, and not a metal anyone would choose for a valued object; it is the material of the disposable, of the expedient, of things made to be used once. It is also a metal that corrodes. A lead curse tablet was written to be buried, and the burial was part of the rite; the tablet’s purpose and its destruction were the same event. There is a neatness there that the Silver Scrolls invert. Silver was chosen because it was precious and because it would last; lead was chosen because it was cheap and because it was not meant to. Between them they bracket the range of what writing on metal could be, from the amulet of the individual to the instrument of the feud.<\/p>\n
The Qumran library holds the object most often mistaken for the greatest curse of all. The Copper Scroll, manuscript 3Q15, is not a curse: it is an inventory of deposits, with no addressee, no invocation of God or of any other power, no statement of a wrong and no consequence. It belongs to a wholly different genre, and the confusion has done real damage: a text that is technical, terse and coded will be taken for a weapon before it will be taken for what it is. The Israel Antiquities Authority, which catalogues it, describes it as listing what may be the sites of the Temple’s buried treasure. What is not open is that it is a list and not a malediction.<\/p>\n
The Dead Sea Scrolls were found in the caves above the Dead Sea between 1946 and 1956, and the library that came out of those caves is the largest body of Second Temple Jewish writing ever recovered. It includes its share of the genre of the curse. One of the manuscripts from Cave 5 is a collection of liturgical compositions whose contents are curses, and blessings and curses of this kind form part of the rulebooks of the community that lived at Qumran, where a member who crossed the boundary of the community was answered with a formula of malediction. The scrolls show what the Ketef Hinnom amulets can only hint at: that the writing of curses was a normal part of the religious technology of the period, carried out in the same scribal world that copied the scriptures and the rulebooks, by the same kind of literate people, with the same equipment.<\/p>\n
The Cave 5 compositions are worth looking at closely, because they are what a curse should look like: formulas of malediction, some of them arranged in a rhythm meant to be spoken aloud, and they belong securely to the same library and scriptorium as the psalms and rulebooks that surround them. They are worth setting against the private curse of the lead tablet, a different object and a different social act.<\/p>\n
One Dead Sea Scroll stands apart from the rest in exactly the way that a curse tablet does, and it has become famous. The Copper Scroll, manuscript 3Q15, was found in Cave 3 near Qhirbet Qumran in 1952, and it differs from every other scroll in the library in its material: it is written on copper rather than on parchment or papyrus. It is not a biblical text and not a rulebook. It is a list, written in Aramaic, of what appears to be buried treasure, with the places described in a code – how many cubits from a particular corner, how many stanzas from a particular column – and with instructions for what to do with the metal once it is recovered. It is the scroll that has been read as a treasure map, and it is the object that has done most to give the Dead Sea Scrolls their reputation for secrecy, because it reads like a coded instruction rather than a religious text. It is also the object that most obviously invites the question of how such a thing could be religious at all: the literature around it has oscillated between reading it as a pious deposit of sacred metal, as a practical manual for a hidden treasury, and as a cipher for something else.<\/p>\n
The lead tablets are a much larger and more varied body of writing, and the form in which the genre is best known. They begin in the Hellenistic period and run through the Roman centuries, and they are found across the eastern Mediterranean, not only in Judaea: the Egyptian and Near Eastern examples, scratched onto potsherds and flakes of stone as well as onto lead, are of the same character as those from Jerusalem. Two forms of text are common. One is prose, running through clauses: an invocation, the deity called down, the person targeted, the injury, the consequence. The other is verse, cast into a rhythm meant to be recited, because a curse that can be spoken aloud can be passed on. And they are put where a stranger will find them, in a tomb, under a threshold, in the foundations of a house, because the magic is in the finding: the tablet is a trap set for the ground, not a message sent to heaven. The practice is not a Jewish invention but an ancient Near Eastern technique that the communities of the Second Temple took up and wrote in their own language.<\/p>\n
The tablet from Mount Ebal belongs, on paper, to this class rather than to the Qumran one, and it is worth saying plainly what is argued and what is not. The object is a sheet of lead roughly two centimetres square, folded in half after it was inscribed, recovered in December 2019 when a team led by Scott Stripling wet-sifted spoil from excavations on the summit, and announced in March 2022 as the earliest Hebrew inscription yet known. The interior text was recovered by X-ray tomography, and the reading published in Heritage Science in 2023 gave a formulaic malediction in paleo-Hebrew letters, placed in the Late Bronze Age and compared with the covenant texts of Deuteronomy 27 and Joshua 24. What is not established is at least as important. The readings are partial, the letter forms are disputed, the exterior text has never been convincingly published, and the field has not settled whether the object is an inscription at all: one study in the same journal in 2024 concluded that the supposed letters are scratches and pitting of the kind that develop in buried lead, and others have proposed that it is a net sinker.<\/p>\n
The Copper Scroll is also the object that makes the provenance question unavoidable. A large part of the Dead Sea Scroll material did not come out of the caves under controlled conditions; it was bought from dealers and from those who had taken it from the caves, and the sequence in which fragments reached the caves, the dealers and the museums is often unrecoverable. By contrast the Ketef Hinnom scrolls have a known provenance in every particular: a cave, a chamber, a year, an excavation team and a published report. For the history of the Hebrew text that difference is not a technicality. It is the difference between a witness whose context is known and one whose context must be argued for.<\/p>\n
It is worth a word of caution about numbers, because the naming of these objects has been muddled in popular accounts for years. The catalogue of Cave 4 manuscripts assigns the number 4Q128 to a group of phylacteries, not to a silver scroll or a curse; the Silver Scrolls of Ketef Hinnom have no Qumran number at all, because they were not found at Qumran; and the numbers 4Q and 5Q are excavation designations that mean nothing outside the Dead Sea Scrolls. The first thing a careful reader of this literature learns is that the most vivid detail in a story about a small metal scroll is often the number attached to it, and that the numbers are usually the part that has been invented by repetition. The objects themselves are smaller and stranger than the numbers suggest.<\/p>\n
What can be learned from a scrap of metal inscribed with a few words is bounded, and the bounds are worth stating. The Ketef Hinnom scrolls tell us that in the last decades of the First Temple period there were people in Jerusalem who owned small silver amulets carrying the priestly blessing, and who buried them with the dead. They tell us that the script was still paleo-Hebrew, that the biblical text they carried differed slightly from the text that the Masorah would later fix, and that precious metal could be cut and written on with the same ease as a scribe ruled a line on a parchment. They do not tell us how many such amulets were made, how they were distributed, or who made them. Two discs cannot support a social history; they can support a point about what was possible, and they can raise a question about what is missing.<\/p>\n
The missing thing is the larger body of writing on metal that was never preserved, or was destroyed, or was simply lost. The Ketef Hinnom deposit was a sealed chamber with a floor of debris; what it gave us is two scrolls. The lead curse tablets of the Second Temple period survive in dozens, mostly because they were deposited in protected places, and they survive from the Hellenistic and Roman centuries, when the archaeology of the region has an unrivalled density. Between the seventh century BCE and the fourth century BCE lies a long stretch of the same kind of practice about which we know very little, and the Mount Ebal tablet, if its reading and its date hold, would belong to the beginning of that stretch. It is a reminder of how thin the record is, and how much of the history of writing on metal is a history of things we cannot see.<\/p>\n
The world the silver came from is nevertheless recoverable in outline. It is a world in which writing was not confined to professionals but reached into ordinary houses, worn against the body, and used for purposes that no library would have catalogued. It is a world in which the words of the Bible were not only transmitted but used – carried, worn, deployed, and then buried. And it is a world in which the contrast between the blessing and the curse is not as sharp as a modern reader might assume, because both are attempts to make a text act on the world: one to secure the presence of a blessing, the other to ensure the presence of a harm. Both are the work of people who believed that words, if written in the right form on the right material in the right place, would do what they said.<\/p>\n
There is a temptation, in a subject like this, to reach for the ancient story of the curse that does its work, and to read the moral of the genre as a lesson about the society that produced it. The temptation should be resisted in one direction and yielded to in another. Resisted, because a curse tablet is a legal instrument, not a sermon; the interest of the genre lies in its forms rather than in its morals, and in the fact that it tells us that in a world without courts worth the name, people still wanted the record of a wrong to be entered somewhere. Yielded, in one particular respect, because the wrongs the curses name are so small and so recognisable. A deposit withheld. An oath not kept. A neighbour’s name. The moral economy of the ancient village, in which a man’s character was demonstrated in the handling of what had been entrusted to him, was a real thing, and the curse tablets are where it surfaces in writing. The later Jewish tradition would develop a genre of blessings and curses of its own, and the moral core of it – the insistence that charity and honesty are not matters of private conscience but of public obligation – is already present in these small sheets of lead, addressed to no audience at all except the deity and the person named.<\/p>\n
The making of one of these objects was a small and skilled job. A sheet of lead was cut or torn from a larger piece, the letters were incised with a stylus point by point, and the sheet was rolled up, folded, or crushed into a shape small enough to hide in a crack, a wall or a floor. Nothing about the making was secret and nothing about it was expensive, which is precisely the point: lead was chosen because it is soft, malleable and cheap, because it takes an incised line cleanly, and because it was not worth keeping. The Copper Scroll was opened differently again: the corroded copper could not be unrolled at all, and was cut into twenty-three strips with a fine saw in Manchester.<\/p>\n
That cheapness also distorts the archaeological record in a way that is easy to miss. A parchment scroll is large, conspicuous and, in a dry cave, spectacularly durable; a rolled tablet of lead is small enough to be dropped, trodden into a floor, swept away or simply lost, and it corrodes into a brittle and unreadable lump unless it is found and treated in time. The curses we can study are therefore a tiny and unrepresentative fraction of what once existed, and they are the fraction that happened to be put somewhere protected enough to survive. The Ketef Hinnom sheets show how much that protection is worth: a scroll rolled tight for two and a half millennia could not be opened until a conservator had spent three years finding a way to do it, and without those years the oldest biblical text anyone has found in a controlled excavation would be two discs of corroded silver with no legible writing at all.<\/p>\n
And there is one more thing the silver knows, which is the thing this journal is ultimately about. The Ketef Hinnom scrolls carry the biblical text in a form that is not the Masoretic text, written in a script that the Masorah would have rejected, before the canon was closed and the text was fixed and the tradition of reading aloud with the Masorah and its notes was formed. They are a witness from the long run-in to the manuscript culture that this journal studies, and they are, in the most literal sense, scrolls: rolled up, tiny, portable, and hiding in a cave what the books in the libraries had not yet been asked to hold.<\/p>\n
In 1979, excavating a tomb in a cave on the western slope of the City of David, the archaeologist Gabriel Barkay uncovered two small objects that were, at first, easy to overlook:…<\/p>\n","protected":false},"author":1,"featured_media":858,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[9],"tags":[46,190,287,188,189,100,193],"class_list":["post-859","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-text-transmission","tag-dead-sea-scrolls","tag-gabriel-barkay","tag-judean-epigraphy","tag-ketef-hinnom","tag-priestly-benediction","tag-qumran","tag-silver-amulets"],"_links":{"self":[{"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/posts\/859","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/comments?post=859"}],"version-history":[{"count":1,"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/posts\/859\/revisions"}],"predecessor-version":[{"id":860,"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/posts\/859\/revisions\/860"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/media\/858"}],"wp:attachment":[{"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/media?parent=859"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/categories?post=859"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/masoretes.online\/wordpress\/wp-json\/wp\/v2\/tags?post=859"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}