Ancient Burials of Metal Documents in Stone Boxes

This paper is an expanded version of a paper presented earlier at the Library History Seminar VI in March 1980. It deals with the persistence, for something like three thousand years, of a strange documentary custom of the Mesopotamian kings, which was distinct and separate from the scribal tradition of clay-tablet writing associated with Assurbanipal. This custom led to numerous regal burials of metallic documents (often encased in stone boxes or other special containers), which were concealed in the foundations or other inaccessible recesses of temples and palaces. The discovery of metal documents beneath the foundations of the Serapis Temple, which housed the Serapeum Library at Alexandria, has also established an archaeological connection between the building practices of the Ptolemies and the Mesopotamian kings.


A farmer in the western Peloponnesus was digging a well. Twenty feet down he came upon a stone box. He smashed in its lid. Inside there was a big object “like a bundle,” dark in color and crumbly in texture. He thought he saw letters written on it. He informed the police, who informed the local director of antiquities; but for some time they could not get out to the farm.

It was 1944-45, and Communist squads were trying to control the roads. When at last the director was able to reach the farm, the object was gone. The farmer had thrown it on the dunghill “because it was not a treasure: it looked like dung and it fell to pieces quite soon.” Others, however, had seen “many letters” on it and said that, although fragile, it held together on the dunghill for some days. Clearly it was a book roll . . . ; clearly it was precious to the man who buried it in a stone casket; certainly it would have been precious to us. But it was of no use to the farmer, and it is gone.1

On the Ancient Preservation of Writing

Throughout antiquity, records of all kinds were intentionally buried for one reason or another. The Qumran literature, for instance, was not driven underground by the ravages of war. It was deliberately laid to rest in the “solemn communal interment” of a documentary funeral,2 which served as the “final concealment” of a whole community library.3

This could only have taken place when the community was on the point of dying out. When that happened, however, we do not know. . . . But we know for certain that . . . when Josephus wrote his Antiquities . . . , the religious order [of the Essenes] was in a vigorous condition and could have had no reason to store its books carefully in a hidden and inaccessible place.4

The Qumran documents were apparently “embalmed” before they were buried. “The careful way in which the MSS were deposited” suggests, more than anything else, “the intention of preserving them as long as possible.”5 There are some intriguing instructions for preserving library materials in the Assumption of Moses, where the aging prophet says to Joshua:

Receive thou this writing [about the preservation of documents] that thou mayest know how to preserve the books [of the Pentateuch] which I shall deliver unto thee: and thou shalt set these [books] in order and anoint them with oil of cedar and put them away in earthen vessels.6

These instructions, or something similar, were also behind the creation and preservation of written legal deeds for the transfer of real estate in Jeremiah 32:6-15. The documents, which were duly certified by witnesses, had been drawn up in duplicate (with both a sealed and an open copy) by Jeremiah, who then directed his scribe to “put them in an earthen vessel, that they may continue many days” (Jeremiah 32:14).7 The documentary methods of Moses and Jeremiah, furthermore, have been attested all over the ancient world. They occur in the Talmud, to be sure, but they are also “fully described in Greek sources” and found in the literatures of both Mesopotamia and Rome.8 Their presence in the West is implicit in a persistent legend about the books of King Numa, the traditional founder of Roman legal and religious institutions. Refusing cremation, he ordered his followers to make two “stone coffins” (lithinas sorous) in order to “bury his books along with his body.” When he died, therefore, they sealed the coffins with lead, “the one holding his remains, the other containing the holy books he had written with his own hand,” and buried them as directed at the foot of Janus Hill on the west bank of the Tiber.9 Four or five centuries later,10 the coffins were accidentally discovered intact.11 When the lids were removed by breaking their leaden seals, Numa’s body had wasted away to nothing,12 whereas all of his books had been preserved, not merely well, but “in mint condition.”13 The contrast was impressive: the books, written on papyrus scrolls, had been buried with their regal author in a hole in the ground,14 but they outlasted him hands down because the West, which learned to preserve its documents by procedures derived from the embalming and entombment of corpses, never deigned to mummify its dead.15 Pliny, following Hemina (who deviates somewhat from other accounts of the burial and retrieval of Numa’s books), describes the process in part:

How these books were able to last so long was amazing to many. But the man who found them had this explanation: a stone cube placed in the center of the coffin had been bound up with waxed cords running in every which direction. On [or in] the top of this stone [or stone box?] three books had been placed [or inserted]; and that probably explains why they had not decayed. Besides, the books themselves had been treated with citrus oil; and that doubtless explains why the moths [or gnawing worms] had not touched them.16

Numa’s books (three, twelve, or fourteen) survived for half a millennium, if only to be burned by the Romans who found them,17 because deliberate measures were taken to ensure their survival. They were chemically treated for protection against moth and rust, sealed in a special stone container, and buried deep in the bowels of the earth. Citrus oil, waxed swaddling cords, hewn stone containers, leaden seals—all of this smacks of the cedar oil, waxed linen wrappings, unique earthenware jars, and tightly sealed lids used for preserving the Dead Sea Scrolls.18 The parallel is too close to be accidental. If using these things at Qumran “proves that the scrolls were hidden in the cave for safe preservation,”19 if “everything was done to preserve the scrolls as long as possible,”20 can we say anything less of Numa’s books? The Dead Sea Scrolls survived for more than 2000 years to be read in our own day.21 Why, then, couldn’t the scrolls of King Numa survive in good condition for less than one-fourth as long?

Other buried libraries have survived for many centuries in both the Far and Middle East. About A.D. 1035, for example, the Buddhist monks of Chinese Turkestan, who were “under the threat of invasion,” walled up their entire collection of books in the cave of Tun-Huang. In A.D. 1900, almost nine centuries later, “the hiding-place was accidentally discovered by a Tibetan monk.” Orientalists subsequently explored the cave, “where they found 20,000 scrolls preserved, dating from the sixth and seventh centuries, in Chinese, Tibetan, Sanskrit, and other languages.”22 A second Buddhist library, discovered in the ruins of a tower at Gilgit, “also contained a great number of manuscripts, some dating perhaps from the fourth century.”23 The Nag Hammadi library, a Gnostic “Qumran” of Christian documents, was retrieved through an “earthenware ‘time capsule’ discovered in the sands of Egypt” when peasants, hunting for fertilizer in 1945, dug up “a large jar filled with leaves of papyrus, bound together like books.”24 The library, “well buried in a tomb very far away from all the monasteries,” was virtually intact after more than 15 centuries.25 It has been described as “the most remarkable ancient library we possess.”26 Its early codex-volumes, whose beautiful leather bindings “are among the oldest ever to survive,”27 were preserved by the same techniques employed at Qumran.28 Eusebius even mentions Ksisouthros, better known as Noah, who was commanded before the Deluge “to bury his books (which discussed the beginnings, middles, and endings of all things) in the sunlit city of Sippar.” When the flood subsided, therefore, Noah took his family “back to Babylon as commanded, in order to retrieve the buried documents from Sippar and transmit them unto men.” Accordingly, they “dug up the documents and began founding cities, setting up temples, and rebuilding Babylon.”29 These records were preserved temporarily, through extremely hazardous circumstances, by special techniques unknown to us. The clay tablet libraries have also survived through documentary techniques differing in significant ways from those which preserved their papyrus cousins.30

There are no Qumrans or Nag Hammadis in the West, for classical literature “is like a city which has been bombed and partially burned”; most of its streets and buildings are in ruins, although many have remained partially (and some wholly) intact.31 The literature we have is largely from the discard. The tablets from Crete and Mycenae, for example, “were not even fired: they became permanent only when the palaces were burned down.”32 Virtually all of the Greek and Latin papyri, furthermore, “were found quite literally in rubbish dumps or in the ruins of abandoned houses.”33 A few manuscripts have nevertheless survived “because they were deliberately buried.” These include two retrieved from coffins, one from a stone box found twenty feet below ground, and several from the wrappings of “cheap mummy cases”;34 some have even come from the “mouths ‘and other cavities’ ” of embalmed sacred crocodiles!35 But many of the writings buried in the West, as in the East, have been metallic documents clearly meant “to survive as long as possible.”36 Lillian Jeffery mentions the use of various metals for writing in the ancient Near East and among the Greeks, who “apparently passed on the practice to the Latin and Etruscan people,”37 as the Roman use of bronze is firmly established.

The bronze plaque (pinax or deltos) was widely used. . . . The Greeks themselves appear to have had a tradition that texts of really pre-historic antiquity were (or should be) inscribed on bronze. Thus Agesilaos of Sparta, on opening a tomb at Haliartos . . . , found there . . . a pinax chalkous [bronze tablet] covered with barbaric characters which resembled Egyptian. . . . Akousilaos the Argive historian was said to have compiled his genealogies from deltoi chalkai [bronze tablets] which his father found while digging on his premises. . . . When Lucian’s Alexandros went to Kalchedon to stage an elaborate piece of deception, he . . . arranged to excavate deltoi chalkai of incredible age from the old temple of Apollo there, containing alleged statements by Asklepios and Apollo his father.38

We have no gold tablets from archaic Greece, although “a fifth-century inscription at Selinous appears to mention one.”39 The nine golden plates of Orphism, however, had been carefully interred in coffins as guidebooks for the dead; they have helped explain the strange Near Eastern overtones of platonism because “Plato and the buried plates were drawing on the same eschatological literature.”40 And the metal tablets from Pyrgi, found “some thirty miles north of Rome” in 1964, were “buried by pious hands” after the smaller of two temples, the sanctuary of Thefarie Velianas, had been reduced to ruins.41 Rubble from the sanctuary was found “in a rectangular niche between the two temples, carefully and piously disposed” to protect its most valuable records. “There, between large blocks of tufa” salvaged from its walls “and three slabs of its terminal tiles,” lying beneath “a heap of terracotta fragments, three sheets of gold leaf, with inscriptions on the outer face, had been hidden.”

Together with these gold leaves, there was a mysterious fourth inscribed sheet of bronze, in very poor condition. . . . The inscription on the bronze sheet with the three others on gold sheets suggests that the niche between the two temples had been made to preserve . . . a part of its archives, which contained different documents established on various occasions.42

The Pyrgi tablets recall many ancient burials of metal documents, which include: (1) the legal agreements of a town in Spain with both its guests and its Roman overlords—two bronze tablets, “one placed exactly over the other with their written sides down,” discovered beneath “two roofing tiles carefully laid against each other and covered with debris”;43 and (2) the golden “Torah” of Pali Buddhism found “in the brick chamber of an old mound”44 at Hmawza—”a manuscript in every way similar to the palmleaf manuscript so common in India and Burma but with [twenty] leaves of gold” and two gold covers,45 which contains “the Law or Dharma Preached by the Buddha.”46

There is, finally, an interesting burial from the Bertiz Valley near the Turkish province of Maras, where some small silver plates “completely covered with Semitic characters” were discovered in the late 1940s. They had apparently been “unearthed in a badly dilapidated Bronzekugel,” a brazen sphere “disregarded by the farmers who emptied it because of its beat-up condition.”47 Unusual burials like this are often dismissed as one of a kind. But there is nothing unique in this account: it resembles the Assyrian reburial, probably by Shalmaneser III (858-824 B.C.), of a small silver plate and two small gold plates from the reigns of Shalmaneser I (1274-1245 B.C.) and Tukulti-Ninurta I (1244-1208 B.C.).

The three tablets had been imbedded in sand in a small bowl. A second, similar bowl was inverted over the top and the two were apparently laced together through holes in their rims. This little “capsule” was half-sunk into the ground, a larger bowl was inverted over it, and the whole thing was buried.48

These remarkable burials—of special documents carefully placed in peculiar containers designed specifically to preserve them—may actually be related to the long history of incantation bowl inscriptions which were interred well into the Christian era.49 They introduce quite naturally the ultimate attempt of the ancients to immortalize their records—the gold and silver plates from Persepolis.50

Before and after Persepolis

Old Persian studies got a new lease on life in 1926, “when an inscription of Darius was found at Hamadan, in duplicate on gold and silver tablets.”51 The inscription, wrongly thought to be “wholly novel as to its form and content,” was discovered in an old foundation “between two square hewn stones that had been carefully prepared to receive it.”52 The find, which established the exact location of ancient Ecbatana, also elicited Herzfeld’s prediction that “we may expect with certainty the discovery of similar documents in the excavations at . . . Persepolis” and elsewhere.53 This prophecy was fulfilled in September 1933, when Herzfeld discovered that “two shallow, neatly made stone boxes with [sealed] lids, each containing two square plates of gold and silver, had been sunk into the bedrock beneath the walls at the corners of . . . the apadana”54 (the multicolumned audience hall of the Palace at Persepolis). (See figure 1.) The plates, which bore the same inscription as their counterparts from Hamadan, “were laid down, probably in the presence of Darius, in 516-515 B.C.”; they were retrieved 2,500 years later in perfect condition, “the metal shining as the day it was incised.”55 There were now six metallic copies of the same inscription, three complete sets of duplicates proclaiming the majesty of Darius and the vast extent of his kingdom.

All these tablets—one gold and one silver from Hamadan, two gold and two silver from Persepolis—were discovered in situ. . . . The texts of the gold tablets from Hamadan and Persepolis vary only in the line arrangements imposed by different formats. The Persepolis tablets underlie the issuance of this “edition,” whose unconventional writing [of a particular word] . . . shows that all of its copies were created from one and the same Urtext in a central office. Darius had undertaken simultaneous building projects in Persepolis, Susa, and Ecbatana, and the administration of these buildings was a unified thing.56

Four more gold tablets found at Hamadan bear inscriptions issued by Ariaramnes, Arsames, Artaxerxes III, and Darius II.57 Of the six inscriptions from Hamadan, a full two-thirds—the silver tablet and three of the five gold tablets—were rescued from looters who had cut them into pieces for the purpose of melting them down.58 One shudders to think of the many similar documents which have not escaped the cutters and melters. The Persepolis plates constitute the high point in a long tradition of concealed metallic documents which extend from Sumer to Alexandria. The stone boxes found in holes cut into rock foundations prove conclusively that the plates were building deposits. The Darius inscription on gold and silver tablets is therefore “of the same type as the foundation inscriptions on metal tablets of Warad Sin of Larsa [1843-1823 B.C.], of . . . the wife of Rim Sin [1822-1763 B.C.] . . . , of Tukulti-Ninurta I [1244-1208 B.C.], and of Sargon II [721-705 B.C.].”59 Metallic foundation texts are older than that, however, possibly reaching as far back as Early Dynastic II (ca. 2700-2500 B.C.).60 The stone chest may be older still, if an object dated ca. 2900 B.C. or earlier, which was found in a temple at Tell Brak, is actually an “early dynastic foundation box.”61 The metallic foundation tradition, though frequently interrupted,62 lived on until the crash of the Late Assyrian Empire (ca. 626-609 B.C.), when it perished because the Neo-Babylonians instituted other documentary procedures. It was briefly resurrected from the Late Assyrian period by the Achaemenid dynasty of Persia (539-331 B.C.),63 only to die once more, at least to all appearances, when Alexander the Great fired the palace at Persepolis. But the metallic foundation inscription surfaced yet again at Alexandria in the excavations of (1) a granite box for holding the writings of a late Greek author,64 and (2) dozens of small metallic plates from the foundations of the Serapis Temple, which housed the Serapeum Library.65

The “flames of Persepolis” symbolize in every way the significance of Persia as a major “turning-point in history.”66 She was the mystic counter of Greek naturalism, who created a comprehensive “synthesis of Near Eastern cultures” by combining all of the influences from the Fertile Crescent, “including those of Persia itself, Mesopotamia, Asia Minor, the Syria-Palestine coast, and Egypt.”67 Her material wealth in gold alone was staggering. Antiochus I minted more than $7,250,000 in coins from the golden roofing tiles of one Ecbatana palace;68 and Alexander the Great systematically looted the palace at Persepolis for “a treasure estimated . . . at over $150,000,000” before putting it to the torch, plus virtually all of the valuable objects “which Persian art had made or Persian conquest gathered.”69 The figures are revealing, even without correction for inflation. The culture of ancient Persia, which “reached one of the high peaks of human experience,” also produced the carefully hewn stone boxes of Darius with their magnificent cargo of gold and silver plates. The Darius inscriptions thus mark the “culmination of a metal art which had been at least 2000 years maturing, gathering inspiration from a variety of cultures.”70

It remains, then, only to review the history of metallic foundation inscriptions before and after the Darius plates, and to summarize its significance for library history. Before doing that, however, we must ask an intriguing question. Only two stone boxes were discovered by Herzfeld, who retrieved them from the northeast and southeast corners of the apadana. But “the cavity meant to hold a third such box was [also] found at the destroyed northwest corner.”71 Who destroyed the northwest corner before the excavators got to it? Could it be that Alexander the Great and his men actually found the missing limestone box with its fabulous treasure of gold and silver plates?

Before Persepolis

The history of metallic foundation inscriptions provides too many boxes and documents to discuss each one separately. This paper therefore reviews that history only in relation to (1) three Neo-Sumerian kings, whose peg deposits probably led to the later burials of metal documents in stone boxes; and (2) nine subsequent rulers, including one Kassite, one Chaldean, two Amorite, and five Assyrian kings, who ruled from the nineteenth through the seventh centuries B.C. The paper thus ignores a mass of material, which includes the numerous metal tablets from Early Dynastic peg deposits,72 the Akkadian bronze tablet from Samarra,73 four deposits with uninscribed bronze plates from the Isin Larsa period,74 the mysterious stone and metal tablets from Old and Middle Assyro-Babylonian times,75 the vague references to metals deposited in foundations by Shamshi-Adad I (1813-1781 B.C.) and Esarhaddon (699-680 B.C.),76 the built-up brick boxes from Lagash,77 the many brick boxes from the Neo-Sumerian and later periods,78 the door pivot boxes,79 and the trinkets (beads, amulets, etc.) found embedded in bricks.80 Hundreds of documents like the Elamitic inscription on a bronze plate (ca. 600 B.C.) found in the treasury of the Persepolis palace, are also ignored because they are not associated with building deposits.81

The stone box loaded with metal documents is probably derived from the peg deposits of the Neo-Sumerian Renaissance at Mari in the Ur III period (ca. 2100-2000 B.C.).82 Parrot uncovered “six foundation deposits” of Niwar-Mer, which had been embedded in the materials used to construct an ancient building. Four of these deposits, “placed very precisely at its corners, identified the building as the Ninhursag Temple, thanks to the inscribed bronze plates,”83 which they included.

In each case a bronze plate, about 15 cm. square, was placed directly on the mud bricks. Each plate had a short inscription in one corner. In the center of each was a round hole through which was thrust vertically a bronze peg 12 to 14 cm. long. A slab of wood about the same size as the metal plate was put on top, and a miscellaneous collection of small objects—a spindle whorl, beads, small plaques, a pendant—was placed beside it.84

Three of the corners in the temple of Dagan have also produced the foundation deposits of Ishtup-Ilum. More complex than the previous deposits, they definitely suggest a development toward the stone box of Darius. They were found “inside the wall a little above the footing at the base of the temple in a rectangular space”85 that had been carefully prepared to receive them.

In one corner of this rectangle was placed a box made of two square stone slabs. The lower slab had a square depression in which a bronze plaque about 13 cm. square was placed. A bronze spike about 27.5 cm. long was thrust through holes in the bronze plaque and the stone slab, and into the mud brickwork beneath. A second stone slab, of the same size as the first but without the depression or hole, was placed over the first. The rest of the . . . rectangle reserved in the brickwork was covered with a layer of round pebbles, among which were numerous small objects. . . . Next to the stone box, buried among the pebbles, were a tablet of white limestone and one of schist. The tablets and the bronze plaque bore identical inscriptions.86

The several deposits of Apil-kin, one of Mari’s early governors, were concealed in the boxlike cavities of false bricks built directly into or beneath the foundations themselves. The governor had found “a real hiding place” beneath the inner doors of the sahuru, a small entrance hall leading to the “Lions’ Temple,” which he had built behind the Temple of Ninhursag. This cachette was “arranged with much more care” than his predecessors had bestowed on theirs. He had actually “made a box by hollowing out one of the rough bricks in the footings beneath the foundation.”

In this box a bronze plate had been deposited without being nailed down. It was encased in wood, as the cavity was larger than the metal plate. A [wooden] plank, cut to the exact dimensions of the cachette, covered both the plate and its framework. A mat was then placed over the whole thing, the hiding place with its hollow brick was concealed, the brick foundation was laid atop all this as though nothing had happened and construction continued.87

The foundation deposits of Niwar-Mer, Ishtup-Ilum and Apil-kin are also related to the elaborate boxes made up of baked bricks “laid flat in bitumen, in courses measuring 3 x 2 1/2 bricks.”88 All of these deposits with their various containers point to the long development which culminates in the rock holes, stone boxes, and metal documents of Darius.

Of more than a dozen rulers listed by Oppenheim, Warad-Sin (1834-1823 B.C.) and Rim-Sin (1822-1763 B.C.) are “the only Larsa kings who used peg deposits”;89 but both of these rulers were involved with either the boxes or the documents of the metallic foundation deposit. While clearing a small temple site in southeastern Ur of its superimposed ruins from the Ur III and Isin-Larsa periods, Woolley dug into the remains of an old wall. He quickly found, in the rubble beside the wall, some “clay foundation cones . . . from its destroyed upper courses.” Then about six inches below the wall’s highest remaining surface, he uncovered “a box of burnt brick contrived in the mud-brick core of the wall.” The box contained “an intact foundation-deposit consisting of the copper figure of the king” and a “brick-shaped inscribed steatite tablet.” The cones, the statuette, and the tablet all bore the same inscription, which stated that “the temple was dedicated to En-ki, the water god of Eridu, . . . by Rim Sin king of Larsa,” in the ninth year of his reign. The building and its deposit “can therefore be accurately dated to the year 1990 B.C.”90 The excavation disclosed no metal tablets, however, and none are known from Rim-Sin; but Simat-Inanna, “one of the wives of Rim-Sin,” did deposit inscribed limestone and copper tablets in the foundations of a Larsa temple, which she dedicated to the goddess Belit-ekallim “during part of the reign of Hammurabi at Babylon [ca. 1792-1750 B.C.].”91 No deposits actually made by Warad-Sin have ever been recovered, and the same is true of Kurigalzu II (1345-1324 B.C.). But excavation of the later Ningal Temple, built by “the Assyrian governor of Ur in about 650 B.C.,”92 has produced a pair of steatite and copper tablets from each of those rulers. “The temple had been restored by Nabonidus [555-539 B.C.],”93 the last Neo-Babylonian king, who also restored its foundation deposits. This reburial of tablets from the Amorite and Kassite dynasties not only proves that Warad-Sin and Kurigalzu II deposited foundation inscriptions in their buildings, but also demonstrates the astonishing antiquity and vitality of this vigorous metallic tradition.

Under the floor [of room three] there was found loose in the soil a [white] limestone foundation-tablet of Kuri-Galzu and close to this two copper tablets and one of black steatite; one copper tablet was a duplicate of that in limestone and recorded the restoration of an ancient temple . . . , the other two also formed a pair and recorded the building by Warad-sin of “a great wall which like a tall mountain cannot be undermined” . . . ; neither of the two texts can have any reference to the site in which they were found; they must have been unearthed in the Neo-Babylonian period and given pious reburial under the new temple that was in course of construction.94

After Kurigalzu II, the Assyrian kings more or less monopolized the metallic foundation deposit until the breakup of their empire (ca. 600 B.C.) by the Neo-Babylonians. The elaborate reburial by Shalmaneser III of a Schalenkapsel containing gold and silver plates from Shalmaneser I and Tukulti-Ninurta I has already been discussed.95 The only other building documents from Shalmanesers I and III are an inscription of the former stating that he “placed stones, silver, gold, iron, copper, tin, and aromatic plants” in foundations,96 and a lone gold tablet of unknown provenance from the latter.97 It is nevertheless known that “small tablets of precious metal were used from the time of Shalmaneser I onwards.”98 The most complicated foundation deposits of Mesopotamia, on the other hand, come from the later Ishtar Temple of Tukulti-Ninurta I (1244-1208 B.C.), who dedicated its twin shrines to Ishtar Asshuritu and to Dinitu. The deposits from this temple to Assur constitute “a very elaborate combination of [inscribed] slabs and tablets, large and small, of various materials,” installed with “a lavish use of beads and nondescript fragments of stone.”99 The slabs, which include seven made of lead (averaging about 5″ x 15″ x 30″ in size and 880 pounds in weight) and two of limestone (one almost 9′ x 5′ x 16″, the other about 4′ x 6′ x 12″), constitute “the most massive [deposits] so far discovered in Mesopotamia.”100 The tablets include thirteen made of gold or silver and seven each of lead and alabaster.101 The complex arrangements of these twenty-seven documents defy verbal description, but they were partially disposed as follows:

First three lead blocks were placed upon the mud brick sub-foundation; two small inscribed tablets of gold and silver and a tiny square of sheet copper were placed on the middle block. A few baked bricks were laid along the wall face to make a level bed for the stone slab. Glass beads, fragments of stones, and . . . twigs or bits of wood were strewn over these objects, and the limestone slab was placed over them. . . . Mats were laid over the block, and . . . [near] its rear edge were placed more valuable trifles, including beads and . . . bits of ivory. On this “cushion” of beads and mortar went two more gold and silver tablets, and a square of sheet gold. Then the fourth lead block was laid over the lot and the construction of the wall continued in mud brick.102

Additional gold and silver tablets were positioned, “together with beads and stone chips, on the cella pavement beneath the dais.” Another complex deposit of similar foundation inscriptions was also discovered “beneath and behind the Dinitu shrine.”103

An important pair of gold and silver plaquettes has survived from Assurnasirpal II (883-859 B.C.).104 “The actual provenance of these two inscriptions is unknown,”105 but they were very probably found at Nineveh in the Temple of Nabu, the god of learning, writing, scribes, and secretaries.106 The possible linkage of Nabu with the tablets is interesting for they present Assurnasirpal II as saying explicitly: “I laid the foundation of the palace at the city of X, the foundations of my royal residence, on tablets of silver and gold.”107 The actual wording of the tablets, as a matter of fact, means “to establish the foundation on documents.”108 In all of cuneiform literature, Bottéro knows specifically of “only one other formula somewhat like this one.” It occurs “in the Prism [text] describing the 30th year of Assurbanipal,” the librarian-king from Nineveh. In this inscription, which deals with the Temple of Nergal at Kutha, Assurbanipal says: “In a favorable month, on a propitious day, I established its subfoundation on GULA oil, that fine oil, and upon tables of silver and gold.” This statement, Bottéro notes, incorporates “the same verb (addi), the same preposition (ina), and the same mention of gold and silver tablets as in our text.”109 It suggests that foundation documents are not merely inscriptions discovered in foundations. They are basic documents bearing witness to the founding of important royal and religious buildings on writing, which was known anciently as “the King’s Secret”—a mysterious something giving him both the right and the power to rule.110 The regal habit of building upon inscriptions, furthermore, probably symbolizes the original founding of the temple, the palace, and the city-state upon the written document,111 and possibly upon the metallic document. At any rate, the practice was firmly established in ancient Mesopotamia.

Archaeological digs have amply documented this custom, observed by the Mesopotamian kings, of burying among the substructures of the temples or palaces they built or restored such things as clay nails; cones, barrel cylinders, and stone or metal tablets, on which they inscribed a permanent record of their labors.112

The utter seriousness of the kings who made these foundation deposits is exemplified by the solemn curse of Assurnasirpal: “If anyone should efface my name which I have written here, or misuse this document for his own pleasures or purposes, may Assur, the Great Lord, destroy his army, ravage his throne, and cut off from the land his name and all of his descendants!”113

The inscribed stone box “appeared for the first time in the reign of Assurnasirpal II [883-859 B.C.],”114 the last of the Middle Assyrian kings. All previous examples of boxes, including the possible instance from Tell Brak and the boxlike cachette of Apil-kin,115 were either uninscribed or directly incorporated into the structure of some building. In 1929, however, “a damaged stone box bearing an inscription” by Assurnasirpal II showed up in Philadelphia.116 The box came from the ancient city of Apqu, also known as Bumariyah or Tell Abu-Maria, “some twenty miles west of Mosul, near Telefar,” in Iraq.117 It was pieced together by E. A. Speiser, who “identified it as a foundation box, and deciphered the [long] cuneiform inscription” on its sides and lid.118 It was probably taken from a foundation hole, although “there is no means of knowing the [actual] conditions under which it was found.”119 Moreover, since the gold and silver tablets of Assurnasirpal II may also have come from Apqu, “it is possible that they were [originally] enclosed in the foundation box.”120

Another inscribed stone box inscribed by Assurnasirpal II was retrieved from “a mound called Balawat,” supposedly the ancient Imgur-Bel near Nineveh, “about fifteen miles to the east of Mossul.”121 It was found while Rassam was in Mossul by the local foreman of the dig, who described it as “a stone coffer with a lid, containing two tablets of stone covered with inscriptions.”122 The foreman, who may or may not have removed the box from its find-spot, did rebury it for protection until Rassam returned to the site. It was apparently taken from the entrance to a burnt-out temple chamber, where Rassam also found, lying on a marble altar, “an inscribed marble tablet of the same size and shape as the other two.”123 Because the stone box had exactly enough room for this third tablet, he concluded that it “belonged to the same set” of documents, that it had been removed from the box and placed on the altar for reading, and “that before the priests had time to deposit it back in the coffer, the temple was burnt down, either by accident or by an enemy.”124 The cavity of the stone box was something like 8″ x 9″, large enough to hold three tablets “twelve-and-a-half inches long, eight wide and two-and-a-half thick.”125 As that is less than half the length and width of the box and perhaps three-fourths its depth, the box itself probably measured about 12″ x 18″ x 28″. It was a massive marble chest, whose great weight, though unspecified, was sufficient to tax Rassam’s ingenuity in transporting it to Mossul.126 There is yet another ninth-century example of this kind from the son of Assurnasirpal II, the first Neo-Assyrian king. What little is known of the box, which is engraved on three sides, has been stated by Ellis.

A similar stone box of Shalmaneser III [858-824 B.C.] was found on the ruins of the west gate of the outer wall of Assur. Unfortunately it was empty, and it had evidently rolled down from some other position to its find-spot. In spite of its evidently secondary position, the box lay on some agate beads, which may have been inside when it rolled to its final position.127

These boxes seem to break with the conventional understanding of foundation inscriptions as documents about buildings. The box from Tell Bumariyah, for example, “does not include a building text” of any kind, and was probably “used for some other purpose.”128 The gold and silver tablets it may have housed also make it clear that Assurnasirpal II was founding buildings upon documents, not depositing documents about buildings.129 The Balawat box, on the other hand, mentions the building or rebuilding of both a city and a temple, but “did not appear to have been buried,” and “does not seem to have been a building deposit.”130 There is not much to say about the stone box of Shalmaneser III, as its find conditions are unknown: the king mentions rebuilding the city wall at Assur and urges its future rebuilder to “restore its ruins” and “to return my inscription to its place.”131 But where was its place? It is possible, certainly, that foundation documents served a double purpose, and that at least some copies “of building inscriptions were kept in the temples, for safekeeping or in order to keep the record . . . permanently before the god,”132 or even for reading. Marinatos thought a similar marble chest from Mesenia “could have been a library-box.”133 If such a box “was considered a container suitable for stone tablets” or other documents, as at Balawat, “it may be that the stone boxes of Assurnasirpal II and Shalmaneser III served a similar purpose, and were not meant to be deposited in structures.”134 Their inscriptions, which deal mostly with the Great King and his domains, would seem to bear this out. Excavations at Nimrud and Arslan Tash in northern Syria have also disclosed six or seven inscribed “Assyrian statues of deities holding square boxes” in their arms.135 Their inscriptions state explicitly that “they were set up for . . . Nabu,” the learned god of the written word who was also known as “the perfect scribe.”136 All these statues, and especially those from Nimrud, “are close chronologically to the boxes of Assurnasirpal II and Shalmaneser III,” and it is difficult to deny a connection between them. Mallowan, at any rate, has suggested that the statue boxes “might have been meant to hold tablets, in view of Nabu’s association with writing and scholarship.137

Sargon II (721-705 B.C.) indicates that he deposited inscribed materials of four to nine different kinds in foundations.138 The fact is that building deposits from the late Assyro-Babylonian kings (858-539 B.C.) often include such inscriptions, a documentary custom actually “mentioned in texts from Sargon’s time down to Nabonidus’s reign [555-539 B.C.].”139 The metallic foundation inscription flourished under the Neo-Assyrian kings, and it is therefore no surprise that “the depositing of inscribed documents was greatly elaborated in Sargon II’s palace at Khorsabad.”140 The excavator of this palace, Victor Place, “was intrigued by the unusual thickness (nearly 26 feet), of one of its dividing walls.” On digging into the wall he found “two inscribed barrel cylinders” and “an alabaster block which he carefully unearthed.” The block turned out to be “a stone box (whose lid had been broken by the weight of the wall), which measured about 11 x 15 x 17 inches; and in it he discovered five foundation tablets” on which Sargon II had “described the building of Khorsabad” from scratch.141 “These epigraphical documents have a high value for their texts themselves”; but in addition to that, “the material on which they were engraved increases, if possible, their extreme rarity,” because “one of the tablets was made of gold, another of silver, the third of bronze, a fourth of lead, and the last” of a mysterious “white material,” perhaps alabaster or magnesite, which has proven harder to identify.142 Of the three metallic inscriptions, the bronze tablet is the largest, the gold tablet the smallest, and the silver tablet somewhere in between.143 The lead tablet and the inscribed stone box,144 which completed this series of foundation documents from Khorsabad, disappeared in the infamous naufrage des collections of 23 May 1855, “in which so many of the archaeological materials gathered by the French were lost.”145 Here again, “the box with its tablets was not actually [discovered] in the foundations,” but in a wall “above the level of the floor.”146 This proves that foundation inscriptions were not deposited solely in foundations. It does not prove that the tablets of Sargon II were something other than foundation inscriptions, for they state repeatedly that he founded the city of Dur-Sharrukin (Khorsabad) and built its wall, the various shrines for its gods, and its several palaces; and they also say—again repeatedly—that he inscribed his name on those same tablets and deposited them in the “foundation walls” of the palaces.147 For what they are worth, there are also some Urartean deposits from the Haldis Temple at Toprakkale near Lake Van in Asia Minor, which are probably contemporary with Sargon II.148

At each corner of the square shrine a square depression, about 20 cm. on a side and 3-4 cm. deep, had been sunk into the bedrock. In two of these depressions were found deposits, each consisting of a square bronze plate and two tiny scraps, one of sheet gold, the other of sheet silver. None of these objects was inscribed.149

The metallic foundation inscription came to an end with the fall of the Neo-Assyrian Empire ca. 626-609 B.C. “It was not adopted by the Neo-Babylonian rulers,” who preferred “clay cylinders, the only type of inscribed building deposit used in their time.”150 A clay box151 and a brick box152 are associated with the first and the last Chaldean kings, and there may be others; but there were few if any stone or metal inscriptions. The years between 626-609 B.C. thus mark a chronological datum before which foundation documents were inscribed on metals but not after. “The custom was briefly revived by the Achaemenids,” who intentionally resurrected it from the Neo-Assyrian or Urartean past.153 It died for the second time in 331 B.C. when the Persian Empire was toppled by Alexander the Great, but it also underwent a second resurrection, this time in the great city of Alexandria.

The Alexandrian Echo of Persepolis

Archeology is problematic at Alexandria, where “excavation has yielded, and can yield, but little material for its reconstruction at any period.”154 There are many reasons for this, but the major causes are two:

The first is a general subsidence, probably of about four meters, which has taken much of the coastal region of the ancient city beneath sea level. . . . This subsidence is complicated by a second, man-made difficulty. . . . Intense building activity [since ca. 1850] has created a new and wholly artificial coastline, to a depth of some three hundred meters [900 feet] at its widest extent, in the area . . . where the Corniche was completed in 1906.155

The stratigraphy and ceramic sequences of Alexandria have thus been largely disrupted, as most of the “fill” for the modern city was taken from the ancient city, sherds and all.156 These artificial conditions of her coastline unfortunately “exclude any possibility of accurate determination of the contours of the most important part of the city.”157 Excavators have therefore been forced to concentrate on the east and west sides of Alexandria, the former containing her ancient cemeteries and the latter her famous Temple of Serapis.158 “The Serapeum,” as a matter of fact, “is the only excavated temple” in the city; and its foundation deposits “may reasonably be described as the most important archaeological find of the Ptolemaic period [ever] made in Alexandria.”159 It is very disconcerting, therefore, to learn that “not only Parsons, The Alexandrian Library . . . , but also serious works like the Handbuch der Bibliothekswissenchaft . . . or the Geschichte der Textüberlieferung . . . [have] failed to take notice of the excavations.”160 The failure is understandable, however, as the archaeological and literary evidence for this temple is so confusing that virtually nobody can make sense of it.161

The Serapeum has been unfortunate in its principal excavators, Botti and Rowe. In the reports of the former it is frequently not clear what structures he is discussing, while the latter had little understanding of the historical problems connected with the site, and was unable to interpret satisfactorily his discoveries, important though some of these were. . . . Detailed interpretation of their plans and descriptions is [therefore] a task of considerable uncertainty.162

On 23 August 1943, Alan Rowe discovered “a set of ten foundation plaques bearing bilingual inscriptions in hieroglyphs and Greek stating that Ptolemy III had built the Temple and the Sacred Enclosure for Serapis.” They were found in a hole sunk into a rock foundation beneath the southeast corner of the Serapeum at Alexandria. The set included (1) three metal plates of gold, silver, and bronze; (2) five opaque glass plates; (3) a tablet made of faience; and (4) a mud tablet, apparently uninscribed.163 The find was repeated on 31 December 1944, when a “similar set of ten plaques of Ptolemy III” were taken from another deposit hole in the foundation trench under the southwest corner of the same temple.164 The inscriptions, materials, and arrangements of the plaques were essentially the same as before, as was the actual find-spot.165 “The holes themselves were filled with sand after