The Discovery in Chamber 25
The two smallest and most famous pieces of scriptural antiquity in the world are rolled-up sheets of silver thinner than a kitchen foil. Each, when unrolled, fits inside the palm of an adult hand. Yet in the forty-odd years since a thirteen-year-old volunteer brushed dust out of a collapsing tomb ceiling in Jerusalem, these scraps of beaten silver have rewritten the early history of the biblical text. They are the Ketef Hinnom scrolls, often called the Priestly Benediction amulets, and they are the oldest surviving witnesses to verses from the Hebrew Bible.
The story of how they came out of the ground begins, by all accounts, with a chain of small accidents. The Ketef Hinnom, Hebrew for “shoulder of Hinnom,” is a low ridge that overlooks the Hinnom Valley just southwest of the Old City of Jerusalem. The site lies near St. Andrew’s Church, on ground that is today part of the Menachem Begin Heritage Center, close to the spot where the Valley of Rephaim meets the Valley of Hinnom — historically the road south from Jerusalem toward Bethlehem. Across the centuries the area was used as a burial ground, and the soft limestone was cut into long horizontal chambers with small rectangular openings that could be resealed with slab stones.
In the summer of 1979, the Israeli archaeologist Gabriel Barkay, then a young professor at Tel Aviv University, began a small excavation here. Funding was limited. Volunteers were a dozen or so schoolboys and schoolgirls, twelve and thirteen years old, members of an archaeology club for youth organised by the Society for the Protection of Nature in Israel. The site looked, by all conventional signs, exhausted: the chambers had last been used to store rifles during the Ottoman period. On the second day, while clearing debris from a rock-cut burial cave, one of the children — accounts vary, sometimes a boy, sometimes a girl — dislodged a slab. Part of the rock ceiling gave way. Behind the rubble a sealed chamber opened up. It turned out to be chamber 25 of what eventually became known as tomb 24, an undisturbed Iron Age family burial cave that had been closed since the early sixth century BCE.
What came out of that chamber in the weeks that followed was a small museum’s worth of late First Temple material: roughly a thousand objects, including more than a hundred items in silver, three dozen iron arrowheads, gold jewellery, pieces of ivory, glass beads, bone implements, and about a hundred and fifty semi-precious stones. Most striking for the excavation team were skeletal remains that clearly reflected disturbed burials, indicating that the tomb had been reopened for later interments — a regular practice in Iron Age Judah before the cemetery filled up. Several deposits went straight into labelled plastic bags on site.
Among the silver pieces were two small cylinders. From a distance they looked like little cigarette butts. Up close they had a slightly rougher surface and a faint seam where the sheet had been rolled. Both were sealed closed by being simply squeezed shut: there were no caps, no solder, no cords. They had been compressed, in life, so that they could be worn against the body.
The boys and girls on the dig noticed them first. Barkay himself was not present at the moment of discovery. One volunteer later remembered picking the object up and handing it across the trench; the photographs taken that morning show the heavier of the two still partially caked in chalky dust. By lunchtime both cylinders had been placed in separate bags and labelled “KH1” and “KH2” — the Ketef Hinnom amulet one and two — together with their find context.
The excavators had stumbled on a sealed context of the late First Temple period, and inside it lay two personalised ritual objects that, within five years, would be recognised for what they contained.
The Tomb and Its Date
The Ketef Hinnom site sits on the western shoulder of a geological valley that was, in the Iron Age, a cemetery serving the south-western quarter of biblical Jerusalem. Barkay’s team excavated four burial caves in two seasons, between 1979 and 1980. Tomb 24, the largest, contained seven chambers along its length. Chamber 25, where the silver was found, lay in the deepest recess of the complex, which is one reason its contents were preserved.
The dating of the tomb is now reasonably settled, though it took two generations of scholarship to do so. Three independent indicators point in the same direction: the ceramics, the stratigraphy, and the paleography of the inscribed script. The ceramic types found in the chamber are characteristic of the late seventh and very early sixth centuries BCE — late Iron II Judah, the horizon of the prophet Jeremiah and of King Josiah’s reforms. The stratigraphy is undisturbed: the chamber was sealed by a roof collapse, and the deposits that built up inside it can be matched one-for-one with the pottery sequence used elsewhere in the southern Levant. The paleography — the forms of the individual letter shapes incised on the silver — is now read as unambiguously late First Temple, the late seventh or very early sixth century BCE.
Barkay’s own original 1989 publication dated the inscriptions to the late seventh or early sixth century BCE, on the basis of comparative letter forms and the immediate ceramic context. Some scholars in the 1990s pushed the date down a little, toward the late eighth or seventh century, citing the formality of certain letter shapes. Others thought the placement was mid-seventh. A wave of technical re-examinations in the late 1990s and the early 2000s — most importantly a long collaboration with Bruce Zuckerman and Marilyn Lundberg of the University of Southern California’s West Semitic Research Project — used high-resolution digital photography and computer enhancement to take better readings of the silver, which had previously been almost opaque to normal light. The consensus now is that the letters are consistent with a date within a generation or two of 600 BCE, and most scholars place them in the last decades of the kingdom of Judah before the Babylonian destruction of Jerusalem in 586/7 BCE.
This dating matters enormously because it puts the scrolls right at the boundary between the First Temple period and the Babylonian exile. Whoever wrote the words on those silver sheets did so in the years leading up to the destruction of Jerusalem.
What did the silver sheets themselves contain? In 1983 — four years after the discovery — conservators at the Israel Museum’s laboratories, working under Yael Dvorsah, were finally able to unroll them mechanically. They had to use tweezers and a fine tool because the metal was extremely thin, heavily corroded, and far more fragile than expected. When amulet 1 was unrolled it measured about 97 by 29 millimetres, almost four inches by just over an inch. Amulet 2 was smaller, roughly 39 by 11 millimetres when flattened out. Both were around ninety-nine per cent pure silver, beaten to a thinness comparable to three sheets of paper. The script, in both cases, was incised rather than engraved: a stylus had been used to scratch the letters into the metal.
The faintness of the incisions is part of why the Ketef Hinnom scrolls became a long-running technical project. Initial readings were uncertain. The first readings of amulet 1 in the 1980s were almost over the shoulder of a single scholar using oblique light. Published photographs under normal lighting showed almost nothing. It was only with the USC team’s digital enhancements — photographing the scrolls in raking light at very high resolution, then electronically cleaning up the negatives — that the readings stabilised. The results of that work, with Andrey Lebedev and other collaborators, were published in the early 2000s.
The Texts That Were Hidden in the Silver
Both amulets contain, in the words of the excavator, “early Hebrew versions of biblical verses,” but they are not identical and they do not preserve exactly the same verses. The fuller of the two is KH1. When the readings are set against the standard Hebrew Bible text of Numbers 6:24-26, the Priestly Benediction, the parallels are striking:
– KH1 begins with what appears to be a divine epithet — something like “the Great One who keeps the covenant and is gracious to those who love him and keep his commandments” — followed by a passage about redemption, a rock, and then a version of the opening lines of the Priestly Benediction: “May YHWH bless you and keep you; may YHWH make his face shine …”
– KH2 is more compact. Its preserved text reads almost exactly: “May YHWH bless you and keep you; may YHWH cause his face to shine upon you, and grant you peace.”
The correspondences with Numbers are not coincidental. Numbers 6:24-26 in the standard Hebrew Bible reads:
24 — May YHWH bless you and keep you.
25 — May YHWH make his face shine upon you, and be gracious to you.
26 — May YHWH lift up his countenance upon you, and give you peace.
KH2 is essentially a four-line condensation of this. KH1 is more elaborate, layering extra theological language around the same core. Both preserve the tetragrammaton — the four-letter name of God written as YHWH — in its standard Hebrew form, which is itself significant: the Ketef Hinnom amulets are the earliest known extra-biblical witnesses to the Tetragrammaton written in this style anywhere outside the biblical manuscripts.
The texts are also interesting for what they are not. They are not exact copies of any one manuscript version of Numbers. The wording of Numbers 6:24-26 in the Masoretic tradition is fixed to a particular phraseology, but the Ketef Hinnom formulae diverge in places — using words like “keep” instead of “guard,” and rearranging the structure of the lines. The differences are small enough that the texts clearly count as versions of the same tradition, but they are different enough that scholars have proposed that the verses were fluidly remembered and remixed, rather than transcribed mechanically from a written copy. In other words, the Ketef Hinnom texts are evidence that biblical passages were being memorised, taught, and remixed well before the Hellenistic period.
Reading the amulets has also opened a wider interpretive question. Both texts are written not on a scroll but on a small wearable sheet meant to be tied or sewn onto the body. The standard interpretation, dating back to Rabbi Hanina bar Yitzhak in the Babylonian Talmud and the oral-mystical tradition behind it, was that the Priestly Benediction was used as an incantation text in the late Second Temple period at the latest. The Ketef Hinnom scrolls push that practice back four hundred years, into the First Temple period, and make the protective use of Numbers 6:24-26 a Judahite practice long before Ben Sira or any known Second Temple literature.
What was being protected, in real life, was almost certainly a living person. As Barkay has argued from the start, the two silver sheets must have been amulets. Their location inside a burial cave is not itself evidence of an afterlife function — the corpus of broken silver objects packed into a tomb is more typical of Iron Age Judah than of any other period — but the size, shape, and purity of the scrolls is consistent with the wear conditions of a personal protective object. A silver sheet of this size and thinness is the sort of thing a family would have worn around the neck or sewn into a garment; this is exactly how the Mishnah and later rabbinic literature describe amulets of this period.
The most likely interpretation today is that these were protective amulets worn by two separate individuals buried in the cave, inscribed with the text that ritual experts in late First Temple Judah considered most efficacious for the purpose.
The Cave, the Corpus, and the Survival Problem
Outside the silver scrolls, what makes the Ketef Hinnom discovery so important is the way it lets us see the rest of an Iron Age Judahite burial cave as a closed context. Burial caves in Judah from the late First Temple period were routinely reopened for new interments as family members died. By the time the cave was finally sealed — usually when the rock-cut chambers had filled up — its contents often resembled the inside of a packed deposit, with displaced bowls, broken storage jars, corroded metalwork, and personal ornaments all jumbled together. This is also why the Ketef Hinnom material is so hard to interpret at the object level: the found associations between items may be largely coincidental.
For the purposes of biblical scholarship, however, that does not matter as much as the date. The crucial point is that the silver scrolls, despite the chaos of the surrounding deposit, can be tied by their immediate context to the late seventh or early sixth century BCE. The next older Hebrew inscriptions that preserve biblical text by hundreds of years: the earliest substantial biblical manuscripts known before 1979 came from the Dead Sea caves of the first century BCE and the Wadi Daliyeh papyri of the fourth. Ketef Hinnom took the earliest known biblical text back four centuries.
The find also created a small conservation problem that has shaped Ketef Hinnom research ever since. Silver corrodes in soil. The Ketef Hinnom sheets are made of nearly pure silver that has been immersed for two and a half millennia, which means that the metal surface has been replaced by a fine layer of silver chloride and silver sulfide corrosion. The text is preserved not in the original silver but in the indentations left by the original incisions in the underlying, less-corroded metal. That is one reason the readings are so hard to photograph in normal lighting. It is also why the Israel Museum chose to limit conservation to leaving the amulets in their rolled state after the initial readings. Unrolling them too often, or treating the corrosion too aggressively, would damage the record itself.
The Ketef Hinnom project has gone on to inform a broader set of questions about religious life in late First Temple Judah. The two amulets are part of a wider little-discussed category: protective objects inscribed with divine names and short prayers, of which the Ketef Hinnom amulets are simply the earliest and most thoroughly studied. There are now comparable inscribed amulets from sites like Tell el-Hesi and Khirbet el-Rai, most of them considerably later. The Ketef Hinnom amulets did not create a genre; they confirmed an older pattern of practice that we now know extends well into the Persian and Hellenistic periods.
For textual criticism of the Hebrew Bible, the Ketef Hinnom scrolls are also a benchmark. They predate every Masoretic witness by more than a thousand years. They predate the Dead Sea Scrolls. They are older than the Samaritan Pentateuch in its currently surviving form. They are older than the Septuagint. The earliest non-biblical Hebrew text that names YHWH — outside the Ketef Hinnom amulets themselves — comes from a half-century later. The texts of amulets 1 and 2 corroborate the antiquity of the Priestly Benediction: the words that rabbis in late antiquity associated with Aaron were already being worn on the body in the last years of the First Temple.
That was the surprise in 1979. The students who lifted those cylinders out of chamber 25 could not have known that they were picking up verses that, in their first written form, were already two hundred years old. They were, in any case, picking up the oldest known sentences from the Hebrew Bible ever recovered by an archaeologist. They are now on display in the Israel Museum, Jerusalem, in a small case built to keep light to a minimum. The case label is short, and so is the story. But it is one of the few moments in biblical archaeology where the gap between text and object has become almost nothing.
Why the Ketef Hinnom Scrolls Still Matter
Five decades on, the Ketef Hinnom scrolls sit in a peculiar place in the archaeology of religion. They are physical objects, but they carry a Bible verse. They are tiny, but they have forced us to revise the chronology of texts that we had only previously seen inside the larger manuscripts of a thousand years later. They are mundane — small pieces of beaten silver of the kind that high-status families in Judah could afford — but they are unique: nothing of comparable date has come down to us with such a clearly preserved scriptural text on it.
The Ketef Hinnom amulets matter for several distinct reasons. They are evidence that biblical texts were circulating in late First Temple Judah, well before the exile, in forms recognisably close to the Masoretic text. They show that those texts were being remixed, condensed, and rephrased in the process of writing them down — the verse we know as the Priestly Benediction was already a flexible liturgical text. They show that the tetragrammaton was part of the working vocabulary of late seventh-century Judah: people were wearing the divine name on their bodies, not hidden away in a temple archive. They show that the protective use of biblical verses, which late Second Temple and rabbinic literature treats as a given, was already a folk practice at the moment the Babylonian armies first came within sight of the city walls.
What changed in 1979 was not that the words of Numbers 6:24-26 became any older than they already were. It was that the version of them we have from the Masoretic text a thousand years later turned out, suddenly, to have an archaeological location attached to it. Numbers 6:24-26 is older than it was; its words were being worn on the body in the time of Jeremiah. The little silver cylinder in the case at the Israel Museum is, simply, the closest we have got to a biblical verse in the hand of a person who lived before the exile.
Most biblical archaeology finds are either great monuments (broad walls, royal inscriptions, water systems) or else inscriptions that sit outside the canonical text (Mesha Stele, Lachish letters). The Ketef Hinnom amulets are unusual in that they fuse the two categories. They are small, fragmentary, and a thousand years apart from the Masoretic tradition — and yet the inscription on them is one of the verses that is still read aloud in synagogues around the world every week, in the same kind of voice, in the same kind of cadence, as it would have been sung in the First Temple.
What we see, when the amulets are unrolled, is something close to the heart of the religion that produced the texts of the Hebrew Bible. Not the theology, not the doctrines, not the court scribes: just a small piece of metal, with a few short lines scratched into it, intended to be worn close to the body of someone who wanted a blessing. That is the kind of thing that survives, by accident, for two and a half thousand years. And it is the kind of thing that tells us, more clearly than any monumental inscription could, what the religion of pre-exilic Judah actually looked like at the level where it met the bodies of its practitioners.
The Ketef Hinnom scrolls are small, but in the archaeology of the Bible they sit at the centre.
The Cave Inventory and Barkay’s Later Work
What the Ketef Hinnom amulets came out of, very literally, was a typical Iron Age Judahite burial cave. The list of objects recovered from chamber 25, after consolidation in the Israel Museum storerooms, gives a snapshot of what an upper-status Jerusalem household of the late seventh century BCE was burying with its dead: a hundred and twenty-five silver objects, mostly small personal ornaments; several dozen iron arrowheads of the kind associated in Iron Age contexts with mounted archers or with household weapons stocks; three dozen gold beads and beads of ivory and bone; about a hundred and fifty semi-precious stones, including carnelian, agate, and a few imported faience; iron tools and bronze fibulae; glass beads of the kind imported from the Phoenician coast; multiple small ceramic vessels; and the skeletal remains of at least a dozen individuals.
This kind of assemblage is no longer remarkable in late First Temple archaeology. By the late 1970s there were already several comparable tombs published from the Jewish Quarter excavations and from sites like Tell el-Ful, Tell en-Nasbeh, and the tombs of Ketef Hinnom’s neighbour at the side of the Hinnom valley. What made chamber 25 different was the silver.
In a burial context, the iron arrowheads reflect either a military family or — and this is more likely for a Jerusalem suburb — household weapons stored at a time of increasing Assyrian and Babylonian pressure. The glass beads show trade contacts with the Mediterranean coast. The pottery shapes are predominantly local Judahite types with a few Assyrian and Phoenician imports. None of this individually dates the deposit; together, the patterns tie it firmly to the late seventh or very early sixth century BCE.
Gabriel Barkay, the excavator, went on to excavate or to direct further work at Ketef Hinnom in a second season in the spring of 1980, with the same youthful volunteer group. Two more seasons were conducted in 1984 and 1985, producing additional tombs but no comparable silver inscriptions. He later became the senior archaeological adviser to the Israel Exploration Society and a long-time professor at Tel Aviv University. He died on 11 January 2026 at the age of eighty-one, just a few days after a brief illness. Obituaries highlighted the Ketef Hinnom amulets as his most famous find, but he had been working on Iron Age Jerusalem, on the Temple Mount’s southern wall, on the second temple period cemetery at Mount of Olives, and on the survey of the Rephaim Valley for several decades.
The early-2026 obituary in Christianity Today noted that one of the Ketef Hinnom volunteers — then a fifteen-year-old — was an American Christian student named John Monson, who went on to become a senior archaeologist at Wheaton College and a scholar of the southern Levant. Monson’s account of digging at Ketef Hinnom as a teenager has been one of the most reprinted first-hand pieces about the amulets. The presence of Christian-volunteered archaeologists on a Jewish Israeli excavation site in 1979-1980 is itself a small illustration of how the find acquired its unusually wide religious audience.
The Wider Inscription Landscape
It is important to remember, when reading about the Ketef Hinnom amulets, that they belong to a much larger category of late First Temple Judahite inscriptions. Iron Age Judah was far more literate than used to be assumed. The major corpora of late First Temple inscriptions — the Lachish letters, the Arad ostraca, the Gezer calendar, the Siloam tunnel inscription, several bullae from the City of David, the personal name lists from the Ketef Hinnom caves and other sites, and a long sequence of stamped jar handles — all illustrate a network of writing in official, military, priestly, and private contexts. The Ketef Hinnom amulets, however, stand out in this corpus because of what they preserve: not administrative chatter, not a measurement or a date or a receipt for goods, but recognisable verses from a later book of the Hebrew Bible.
The closest Iron Age parallels to the Ketef Hinnom silver are not literary inscriptions at all. They are small inscribed objects of the same kind of size and shape: the Tell el-Hesi pendant, an inscribed silver pendant from a burial context at the site of Tell el-Hesi, probably seventh-century, with a non-biblical inscription; the inscribed ostraca from Horvat ‘Uza in the Negev, with very short texts including personal names; the inscribed arrowheads of the late Iron Age, with names of soldiers and units. None of these preserve biblical material.
This is what makes the Ketef Hinnom amulets exceptional. Late First Temple inscriptions are mostly administrative: receipts, receipts for goods, sealed jars, sealed documents, instructions for soldiers. The Ketef Hinnom texts preserve Hebrew Bible verses in a form already established enough to be worn on the body for protection.
Two corollaries follow. First, the scriptural importance of the Ketef Hinnom inscriptions is not so much that they exist at all (in a literary culture, people write down many things), but that they preserve biblical text in a place and a period when biblical text was thought to be confined to oral or formal temple contexts. Second, the amulets suggest that by the late seventh century BCE, the Priestly Benediction was already established enough in the liturgy of households and individuals to be considered the protective verse by choice. Whether the formulation of Numbers 6:24-26 was yet canonically fixed is a separate question, but at the very least the verse was a recognised liturgical item.
The textual relationship between the Ketef Hinnom amulets and the later Masoretic text of Numbers 6:24-26 is interesting. The divergences in wording are small, but they show the kind of variation that argues for a fluid oral and scribal tradition rather than a single fixed master text. The smaller of the two amulets, KH2, in particular, looks like a lay scribe’s condensation. The forms of the common Hebrew words used suggest that whoever inscribed the silver did not have the verse before them in a fixed written form; they were reproducing it from memory.
This is one of the more unexpected conclusions the discovery supports: a few weeks of digging in a Jerusalem backyard produced evidence for the practice of biblical memorisation in a lived, household, lay context, in the late First Temple period.
The Texts in Their Function
Reading the amulets requires a small step back into the religious world that produced them. Numbers 6:24-26 is, in the Hebrew Bible, a passage taught by God to Moses for Aaron and his sons to use in blessing the children of Israel. In synagogue liturgy it remains a widely recited benediction, used as the closing blessing of the priestly course and adopted into a number of other ritual contexts. In household and apotropaic practice, the same text has been written, worn, or recited as a protective verse across much of late antiquity and the medieval period.
By the time of the Ketef Hinnom amulets, several features of the verse had become sufficiently fixed that lay scribes could be expected to know them by heart, although individual scribes might condense or expand the verse according to family practice. The very structure of the verse — a tripartite form in which each verse is bounded by the name YHWH — would have made memorisation relatively straightforward, and the repetition of the word “may” (in Hebrew “yevarekha,” “yaser,” “yisah”) would have given each verse a familiar cadence.
The fact that the Ketef Hinnom amulets were worn as protective objects is consistent with a long tradition of biblical apotropaic practice. By the time of the Mishnah and the Talmud, scribes were producing protective scrolls and amulets in great numbers, and the Talmudic discussion of which verses were permissible for apotropaic use (and which were not, on grounds that some verses were too sacred for everyday wear) is detailed and well-studied. The Ketef Hinnom amulets push this tradition back four hundred years and confirm that, in late First Temple Judah, the Priestly Blessing was already established as the preferred protective verse for lay use.
What the Ketef Hinnom amulets do not tell us is whether other verses were also written on silver for the same purpose. The two amulets found in chamber 25 were just two of about a hundred and twenty-five silver objects in the deposit, and they are not the only candidates for amulet status in the assemblage. Most of the silver in the chamber has not been studied to the same level of detail as the two inscribed pieces, and small inscribed pendants, finger rings, or other personal ornaments may yet yield additional scripture. The next phase of Ketef Hinnom research is likely to be a re-examination of the broader silver assemblage for similar small inscriptions.
For textual scholars, this is a research horizon of considerable value. Even if the surviving evidence does not yield a second biblical verse from the late First Temple period, it has the potential to confirm the practice of biblical apotropaic writing in pre-exilic Judah in a way that single finds cannot.
Two things are nonetheless clear. The first is that the texts on the silver are recognisable as biblical, and are recognisable enough that we can match them up verse by verse with Numbers 6:24-26 in the Masoretic text. The second is that the texts were being worn as protective verses on the body, and were thus living scripture at the household level, in a period before any of the established post-exilic scribal schools had begun their work.
This is a connection between the modern synagogue recitation and the household piety of late First Temple Judah that no other surviving artefact provides. The simplest and most accurate way to describe the Ketef Hinnom amulets is that they are the oldest verses of the Hebrew Bible ever recovered by an archaeologist, and they were worn by people whose piety we can read off the metal.
References
- Ketef Hinnom scrolls — Grokipedia (overview article with citations to primary publications).
- About the Ketef Hinnom Silver Scrolls — Gabriel Barkay, project page (archaeologist’s own account of the discovery).
- The Blessing Of The Silver Scrolls — Biblical Archaeology Society, “Bible and Spade,” 2017.
- The Ketef Hinnom Scrolls: Earliest Biblical Text Ever Discovered — Armstrong Institute of Biblical Archaeology.
- The Ketef Hinnom Amulets: Wearing the Priestly Blessing for Protection — TheTorah.com (Hadley, Skinner, et al., academic summary).
- Gabriel Barkay — Grokipedia biographical entry on the excavator (with bibliography).
- The Jewish Archaeologist Who Inspired a Generation of American Christians — Christianity Today, February 2026 (obituary and account of the dig).
- The Silver Scrolls — Jewish Press, account of the discovery.
- Image: Gabriel Barkay, “The Priestly Benediction on Silver Plaques from Ketef Hinnom in Jerusalem,” Tel Aviv 19.2 (1992) — Biblical Archaeology Society Library (image and citation to the original excavation report).
- Ketef Hinnom — Grokipedia overview of the site and its context.
- Priestly Blessing — Grokipedia overview of Numbers 6:24-26 (the biblical verses inscribed on the amulets).
- Numbers Chapter 6 / Mechon-Mamre Hebrew-English Bible — Hebrew text of Numbers 6:24-26 with English translation (for the modern comparator readings).