This is a tribute to the splendid archaeological work done by Ali Hakemi in Shahdad. Hakemi, Ali, 1997, Shahdad, archaeological excavations of a bronze age center in Iran, Reports and Memoirs, Vol. XXVII, IsMEO, Rome. 766 pp.
I am thankful to Prof. Mehdi Mortazavi Assoc. Professor, University of Sistan and Baluchestan for the links and references provided. His insights and encouragement are gratefully acknowledged and have led me to this monograph. For the opinions expressed herein, I am responsible.
Pierre Amiet summarises Hakemi’s report with a brilliant exposition: “The discovery, long after that of the great Mesopotamian civilization, just after World War I, of an urban civilization which emulated that of Sumer in the Indus Valley, followed even more recently by the equally impressive civilization of Turkmenia, immediately raised the question of what presumably happened in the immense territory between th two, represented by the Iranian plateau…(Aurel Stein) had crossed Baluchistan and Kerman, ultimately reaching, on the westward side, the only historical entity of Iran predating the Persians – the ancient country of Elam – to all intents and purposes part of Mesopotamia, although essentially a country of mountaineers. In its geographic duality in which the mountain valleys of Fars were associated with the lowlying plains of Susiana, Elam, which was also an ethnic duality, was presumably linked with a hinterland that had remained in the wings of history and comprised the Kerman mountains dominating the salt pans of the Lut Desert. The province which was traditionally rich in stones and metals, and scantly explored by the pioneers, must have been a home to the major witnesses of what Gordon Childe as early as 1934 called the ‘mechanism of the spread’ of the conquests of civilization…in eastern Bactria, bounded the wide loop of Amu Darya, the site of Shortughai corresponds to a settlement of ‘colonists’ from Harappan India, with their characteristic pottery, who saw to the transit of copper and doubtless also of lapis lazuli. These observations seem to be indicative of what probably happened in western Bactria where fortresses housing stores, as at Dashly Tepe, may have been built by a merchant-colonist elite to guarantee trade with the workshops set up either at Shah-I Sokhta or at Shahda and Tepe Yahya and, through them, with Elam, as well as by sea, with Mesopotamia. Unlike Anatolia, where the intense metalworking activity does not seem to have produced any art specific to a given civilization or else highly customized before the 2nd millennium, Iran thus appears to hav been a huge community enlivened by a network of very long routes spreading out from the towns and villages of craftsmen who were creating a different art and using a wide range of techniques, perhaps simulated by Elam. These craftsmen worked copper and soft, colored stones, such as chlorite and alabaster, found locally, together with imported hard stones such as carnelian and lapis lazuli. They must have come into close contact with the transporters, presumably nomadic, according to the tradition of the bearers of the intercultural style. Shahdad lay at the crossroads of these routes, the one running north-south from Gorgan and Tepe Hissar and passing through Tepe Yahya on its way to the Persian Gulf, and those crossing the Lut desert or skirting it through Bampur, towards the north and south of the Hindu Kush and from there into India.” (Introduction, pp.8 - 10)
“The shaft is set on a 135 mm high pyramidal base. The thin metal plate is a square with curved sides set in a 21 mm wide frame. On the plate there is a figure of a goddess sitting on a chair and facing forward. The goddess has a long face, long hair and round eyes. Her left hand is extended as if to take a gift…a square garden divided into ten squares. In the center of each square there is a small circle. Beside this garden there is a row of two date palm trees…Under this scene the figure of a bull flanked by two lions is shown…The sun appears between the heads of the goddess and, one of the women and it is surrounded by a row of chain decorative motives.” (p.271, p.649). The inscriptional evidence discovered at this site which is on the crossroads of ancient bronze age civilizations attests to the possibility of Meluhha settlements in Shahdad, Tepe Yahya and other Elam/Susa region sites. The evolution of bronze age necessitated a writing system -- the answer was provided by Indus writing using hieroglyphs and rebus method of rendering Meluhha (mleccha) words of Indian sprachbund.
Shown are the glyphs of 1. zebu and 2. tigers which are also glyphs on Indus writing which I decode as related respectively to 1. blacksmithy on unsmelted metal (Adar Dhangar, zebu) 2. working with alloys (kol, tiger) !!! The tree is a smelter furnace (kuTi). The endless-knot motif is iron (meD, knot, iron).
I am thankful to Prof. Mehdi Mortazavi Assoc. Professor, University of Sistan and Baluchestan for the links and references provided. His insights and encouragement are gratefully acknowledged and have led me to this monograph. For the opinions expressed herein, I am responsible.
Pierre Amiet summarises Hakemi’s report with a brilliant exposition: “The discovery, long after that of the great Mesopotamian civilization, just after World War I, of an urban civilization which emulated that of Sumer in the Indus Valley, followed even more recently by the equally impressive civilization of Turkmenia, immediately raised the question of what presumably happened in the immense territory between th two, represented by the Iranian plateau…(Aurel Stein) had crossed Baluchistan and Kerman, ultimately reaching, on the westward side, the only historical entity of Iran predating the Persians – the ancient country of Elam – to all intents and purposes part of Mesopotamia, although essentially a country of mountaineers. In its geographic duality in which the mountain valleys of Fars were associated with the lowlying plains of Susiana, Elam, which was also an ethnic duality, was presumably linked with a hinterland that had remained in the wings of history and comprised the Kerman mountains dominating the salt pans of the Lut Desert. The province which was traditionally rich in stones and metals, and scantly explored by the pioneers, must have been a home to the major witnesses of what Gordon Childe as early as 1934 called the ‘mechanism of the spread’ of the conquests of civilization…in eastern Bactria, bounded the wide loop of Amu Darya, the site of Shortughai corresponds to a settlement of ‘colonists’ from Harappan India, with their characteristic pottery, who saw to the transit of copper and doubtless also of lapis lazuli. These observations seem to be indicative of what probably happened in western Bactria where fortresses housing stores, as at Dashly Tepe, may have been built by a merchant-colonist elite to guarantee trade with the workshops set up either at Shah-I Sokhta or at Shahda and Tepe Yahya and, through them, with Elam, as well as by sea, with Mesopotamia. Unlike Anatolia, where the intense metalworking activity does not seem to have produced any art specific to a given civilization or else highly customized before the 2nd millennium, Iran thus appears to hav been a huge community enlivened by a network of very long routes spreading out from the towns and villages of craftsmen who were creating a different art and using a wide range of techniques, perhaps simulated by Elam. These craftsmen worked copper and soft, colored stones, such as chlorite and alabaster, found locally, together with imported hard stones such as carnelian and lapis lazuli. They must have come into close contact with the transporters, presumably nomadic, according to the tradition of the bearers of the intercultural style. Shahdad lay at the crossroads of these routes, the one running north-south from Gorgan and Tepe Hissar and passing through Tepe Yahya on its way to the Persian Gulf, and those crossing the Lut desert or skirting it through Bampur, towards the north and south of the Hindu Kush and from there into India.” (Introduction, pp.8 - 10)
“The shaft is set on a 135 mm high pyramidal base. The thin metal plate is a square with curved sides set in a 21 mm wide frame. On the plate there is a figure of a goddess sitting on a chair and facing forward. The goddess has a long face, long hair and round eyes. Her left hand is extended as if to take a gift…a square garden divided into ten squares. In the center of each square there is a small circle. Beside this garden there is a row of two date palm trees…Under this scene the figure of a bull flanked by two lions is shown…The sun appears between the heads of the goddess and, one of the women and it is surrounded by a row of chain decorative motives.” (p.271, p.649). The inscriptional evidence discovered at this site which is on the crossroads of ancient bronze age civilizations attests to the possibility of Meluhha settlements in Shahdad, Tepe Yahya and other Elam/Susa region sites. The evolution of bronze age necessitated a writing system -- the answer was provided by Indus writing using hieroglyphs and rebus method of rendering Meluhha (mleccha) words of Indian sprachbund.
Shown are the glyphs of 1. zebu and 2. tigers which are also glyphs on Indus writing which I decode as related respectively to 1. blacksmithy on unsmelted metal (Adar Dhangar, zebu) 2. working with alloys (kol, tiger) !!! The tree is a smelter furnace (kuTi). The endless-knot motif is iron (meD, knot, iron).
The accounting system had advanced beyond bullae-tokens to a writing system to prepare stone-, metal-ware catalogs on thousands of inscriptions using mleccha language for Indus writing.
This is the Indian example. This is cited by Richard Meadow of the HARP (Harvard) Project which found it in Harappa. Meadow calls it the earliest writing system of the world.
This is the comparable image on Indus writing with five petals. This is dated to ca. 3500 BCE according to the HARP Harvard report.
This is a glyph showing five petals. Characteristic of tabernae montana tulip flower which is a fragrant flower used as hair-dressing is that it has five petals. So, the word tagaraka has two meanings: 'hair fragrance'; 'tabernae montana tulip' (Sanskrit). This glyph is what is reflected on Shahdad cylinder seal.
This is the comparable image on Indus writing with five petals. This is dated to ca. 3500 BCE according to the HARP Harvard report.
This is a frequently ocurring glyph.
This is a glyph showing five petals. Characteristic of tabernae montana tulip flower which is a fragrant flower used as hair-dressing is that it has five petals. So, the word tagaraka has two meanings: 'hair fragrance'; 'tabernae montana tulip' (Sanskrit). This glyph is what is reflected on Shahdad cylinder seal.
My evidence is the glossary of words of Indian sprachbund (linguistic union) where the words are commonly used across the set of families of languages (Indo-Aryan, Indo-Iranian, Dravidian, Munda). I have compiled an Indian Lexicon with about 8000 semantic clusters to prove the sprachbund. In my Indus Writing in Ancient Near East I have provided hundreds of examples of such semantic clusters in the context of bronze age metallurgy.
Inter-Iranian trade community from Harappa settled on the crossroads at Shahdad?
Object No. 0004 (p.26)
Figure 45. Proto-Elamite pictograms (“From a total of 606 different types of signs found on red ware of Shahdad, 331 are incised and 275 of them are impressed. The star is one of the most common signs, and it has been found in both incised and impressed signs. In the Sumerian and Elamite pictograms a star is an accepted figure representing gods.” p.67)
( Vorgelegt von David Mathias Philip Meier aus Mannheim,, 2008, Die metallnadeeln von Shahdad – eine funktionstypologische untersuchung, pp.82-199 present 121 tafels – plates -- of sets of metal pins and objects discovered at Shahdad.)
Copper lid or plate from Tepe-Hissar, Univ. Museum of Pennsylvania (After Schmidt 193, fig. 120).
Copper plate in Louvre Museum (After Amiet 1976b:no. 21)
Copper/bronze dish from Shahdad, Iran Bastan Museum.
Copper/bronze dish from Shahdad. Iran Bastan Museum.
Copper/bronze dish from Shahdad. Iran Bastan Museum.
Rounded shape copper/bronze dish from Shahdad. Iran Bastan Museum.
“Shahdad (Islamis Khabis) is to be remembered as one of the East Iranian Centres for making metal artefacts in the 3rd millennium BCE. During the years 1971-77 an archaeological mission working in the Lut area under my supervision discovered a wide variety of metal objects in the three main burial grounds A,B and C. Over a period of seven years about 700 broke or unbroken metal objects were discovered…From the study of the metal workshops in 1977, we learned that the base metal copper had been extracted and used in three stages: a) ore smelting, b) purification and c) moulding…The upper side of the plates have embossed moulded figures of living creatures, such as crabs, fish, snakes and gazelles…The abundance of rich copper mines and skillfully mad artefacts in the vicinity of Shahdad supports the view that the prehistoric people of Shahdad were peaceful artisans engaged in the art of producing earthenware, stonework and domestic metal artefacts…The large decorated metal plates discovered in the main cemeteries, especially, confirm that the Shahdad workshop did not just make a few modest articles, but was producing work of great historical significance in the 3rd millennium BCE. It can also be concluded that similar plates, mainly that found at Tepe Hissar, and perhaps some of the others presented in this work, could likely come from the southeastern region of Iran.” (Hakemi, Ali, 1997, Comparison between the plates of Shahdad and other plates that exist in a few museums, In: Taddei, Maurizio and Giuseppe de Marco, eds., 2000, South Asian Archaeology, 1997, Vol. 1, Istituto Italiano per l’africa e l’oriente, Rome, pp. 943-959).
See: http://bharatkalyan97.blogspot.in/2013/07/ancient-near-east-transition-fro-bullae.htmlAncient Near East archaeological context: transition to Bronze Age. Indus writing is for trade in this transition.
The finds of Shahdad; three plates are taken from the 1972 Catalogue: Note the pictographic writing on red ceramics (Plates XXIIB and XXIIC). These includes possible bullae with ‘tokens’ representing some articles being counted.
Plate XXIIIB includes picture of two footprints. This glyph occurs on Indus writing.
Disk seal (glyptic catalogue no. 58; 15 mm in dia. X 8 mm) Excavations at Tepe Yahya, 3rd millennium, p. 154 Double-sided steatite stamp seal with opposing foot prints and six-legged creature on opposite sides. Tepe Yahya. Seal impressions of two sides of a seal. Six-legged lizard and opposing footprints shown on opposing sides of a double-sided steatite stamp seal perforated along the lateral axis.
Lamberg- Karlovsky 1971: fig. 2C Shahr-i-Soktha Stamp seal shaped like a foot.
Shahdad seal (Grave 78). It is significant that a footprint is used as a seal at Shahdad. The glyph is read rebus as rebus word for 'iron':
Rebus readings:
Glyph: meṭṭu ‘foot’. Rebus: meḍ ‘iron’ (Ho.Mu.) dula ‘pair’ (Kashmiri); dul ‘cast (metal)(Santali). Six legs of a lizard is an enumeration of six ‘portable furnaces’ ; rebus: kakra. ‘lizard’; kan:gra ‘portable furnace’. bhaṭa ‘six’ (G.) rebus: baṭa = kiln (Santali); baṭa = a kind of iron (G.) bhaṭṭhī f. ‘kiln, distillery’, awāṇ. bhaṭh; P. bhaṭṭh m., °ṭhī f. ‘furnace’, bhaṭṭhā m. ‘kiln’; S. bhaṭṭhī keṇī ‘distil (spirits)’. Read rebus as : dul (pair) meḍ ‘cast iron’; kan:gra bhaṭa ‘portable furnace’.
Glyph: ‘foot, hoof’: Glyph: ‘hoof’: Ku. khuṭo ʻ leg, foot ʼ, °ṭī ʻ goat's leg ʼ; N. khuṭo ʻ leg, foot ʼ(CDIAL 3894). S. khuṛī f. ʻ heel ʼ; WPah. paṅ. khūṛ ʻ foot ʼ. khura m. ʻ hoof ʼ KātyŚr̥. 2. *khuḍa -- 1 (khuḍaka -- , khula° ʻ ankle -- bone ʼ Suśr.). [← Drav. T. Burrow BSOAS xii 376: it belongs to the word -- group ʻ heel <-> ankle -- knee -- wrist ʼ, see *kuṭṭha -- ](CDIAL 3906). Ta. kuracu, kuraccai horse's hoof. Ka. gorasu, gorase, gorise, gorusu hoof. Te. gorija, gorise, (B. also) gorije, korije id. / Cf. Skt.khura- id. (DEDR 1770). Allograph: (Kathiawar) khũṭ m. ʻ Brahmani or zebu bull ʼ (G.) Rebus: khũṭ ‘community, guild’ (Santali)
Alternative reading: meṭ sole of foot, footstep, footprint (Ko.); meṭṭu step, stair, treading, slipper (Te.)(DEDR 1557). Rebus: मेढ ‘merchant’s helper’ (Pkt.); meḍ ‘iron’ (Munda).
Sibri:
Source: Jarrige, Catherine, Jean-François Jarrige, Richard H. Meadow, and Gonzague Quivron, editors (1995/1996) Mehrgarh: Field Reports 1974-1985 - From Neolithic Times to the Indus Civilization. The Reports of Eleven Seasons of Excavations in Kachi District, Balochistan, by the French Archaeological Mission to Pakistan. Sindh, Pakistan: The Department of Culture and Tourism, Government of Sindh, Pakistan, in Collaboration with the French Ministry of Foreign Affairs.
The text on pg. 326 says:
6.6. Terracotta
Pawns, small wheels, spindle whorls, rattles (fig. 7.32C), sling-balls, and two crucibles, all in terracotta, were collected, together with a large number of discs formed from potsherds. One of the rattles with circular impressions on its surface is very similar to a specimen from a deposit of Period VIII at Mehrgarh, and another one, so far exceptional, bears incised signs and dots that could represent numbers (fig. 7.31C, 7.32C).
6.7. Seals
The seals are of two types. The most common type is the compartmented seal in bronze or in stone. Three specimens have a triangular shape while a terracotta cake bears several imprints of a square-shaped seal with a cruciform motif (fig. 7.31A). The second type is represented by a single piece, a black steatite cylinder seal with knob (fig. 7.31D). It was engraved with the representation of a zebu facing a lion and, on the base, a scorpion. This cylinder seal was found associated with two beads in black steatite and must have been part of a necklace as indicated by its suspension hole. This seal is very similar to a few cylinder seals found in Margiana, in particular at the site of Taip, where such objects are considered to reveal
Mesopotamian influence. One seal from Taip bears the representation of a zebu.
6.8. Copper/Bronze
In the same square (2K) where the cylinder seal was found, a bronze shaft-holed axe-adze of a type also often found in the Murghabo-Bactrian area was discovered (fig. 7.32B). A famous example of such an axe-adze comes from Mohenjo-daro. Other objects in bronze or copper include a few pins.
6.9. Figurines
Terracotta figurines, all made of sherd-tempered ware, were found in large numbers (fig. 7.32B). The main type is a "violin-shaped" female figurine. Eyes and breasts are "applique" as is the coiffure in some cases. Some of the figurines also bear necklaces or ornaments represented by small incised holes. Most of the time, however, only indications of sex are represented including applique breasts and small incised points marking the pubic area and the armpits. This violin-shaped type of figurine is quite original although it does have parallel among a few specimens from sites in the lower Murghab Delta and from later contexts at Pirak and in India (Navdatoli).
6.6. Terracotta
Pawns, small wheels, spindle whorls, rattles (fig. 7.32C), sling-balls, and two crucibles, all in terracotta, were collected, together with a large number of discs formed from potsherds. One of the rattles with circular impressions on its surface is very similar to a specimen from a deposit of Period VIII at Mehrgarh, and another one, so far exceptional, bears incised signs and dots that could represent numbers (fig. 7.31C, 7.32C).
6.7. Seals
The seals are of two types. The most common type is the compartmented seal in bronze or in stone. Three specimens have a triangular shape while a terracotta cake bears several imprints of a square-shaped seal with a cruciform motif (fig. 7.31A). The second type is represented by a single piece, a black steatite cylinder seal with knob (fig. 7.31D). It was engraved with the representation of a zebu facing a lion and, on the base, a scorpion. This cylinder seal was found associated with two beads in black steatite and must have been part of a necklace as indicated by its suspension hole. This seal is very similar to a few cylinder seals found in Margiana, in particular at the site of Taip, where such objects are considered to reveal
Mesopotamian influence. One seal from Taip bears the representation of a zebu.
6.8. Copper/Bronze
In the same square (2K) where the cylinder seal was found, a bronze shaft-holed axe-adze of a type also often found in the Murghabo-Bactrian area was discovered (fig. 7.32B). A famous example of such an axe-adze comes from Mohenjo-daro. Other objects in bronze or copper include a few pins.
6.9. Figurines
Terracotta figurines, all made of sherd-tempered ware, were found in large numbers (fig. 7.32B). The main type is a "violin-shaped" female figurine. Eyes and breasts are "applique" as is the coiffure in some cases. Some of the figurines also bear necklaces or ornaments represented by small incised holes. Most of the time, however, only indications of sex are represented including applique breasts and small incised points marking the pubic area and the armpits. This violin-shaped type of figurine is quite original although it does have parallel among a few specimens from sites in the lower Murghab Delta and from later contexts at Pirak and in India (Navdatoli).
A second type of figurine is represented by a seated callipyge individual while a third type is a standing, flat figurine with small applique breasts. In contrast to the large number of human figurines, very few animal figurines (three humped bulls and some others more difficult to identify) were found.
In Jarrige, Jean-François (1994) The final phase of the Indus occupation a Nausharo and its connection with the following cultural complex of Mehrgarh VIII. In: Asko Parpola and Petteri Koskikallio, eds., South Asian Archaeology 1993, Volume 1, pp. 295-313. Hesinki: Suomalainen Tiedeakatemia, Jarrige discusses relations between Central Asia, Balochistan, and the Indus Valley.
Scanned pages: pp. 360-361 (Sibri1996.pdf)
Rebus reading of + glyph with dots on four corners of the + glyph, on the bulla shown on 7.31a. The + glyph may denote a fire-altar (of temple). kaṇḍ ‘furnace, fire-altar’ (Santali) khondu id. (Kashmiri) kŏnḍ क्वंड् ‘a hole dug in the ground for receiving consecrated fire’ (Kashmiri) kunḍa ‘consecrated fire-pit’. ayaskāṇḍa is explained in Panini as ‘excellent quantity of iron’ or ‘tools, pots and pans and metal-ware’. [It is possible that there were allographs to depict the word: kāṇḍa. The allographs are: arrow-glyph; large dot; notch as a short numeral stroke (for example, ligatured on a fish-glyph or a 'rim-of-jar' glyph; dotted circle.]
Example of use of allograph on a seal from Banawali showing women acrobats leaping over a water-buffalo:
Impression and line-drawing of a steatite stamp seal with a water-buffalo and leapers. Buffalo attack or bull-leaping scene, Banawali (after UMESAO 2000:88, cat. no. 335). A figure is impaled on the horns of the buffalo; a woman acrobat wearing bangles on both arms and a long braid flowing from the head, leaps over the buffalo bull. Two Indus script glyphs in front of the buffalo.
Glyphs: ‘1. arrow, 2. jag/notch’:
1. kaṇḍa ‘arrow’ (Skt.) H. kãḍerā m. ʻ a caste of bow -- and arrow -- makers (CDIAL 3024). Or. kāṇḍa, kã̄ṛ ʻstalk, arrow ʼ(CDIAL 3023). ayaskāṇḍa ‘a quantity of iron, excellent iron’ (Pāṇ.gaṇ)
2. खांडा [ khāṇḍā ] m A jag, notch, or indentation (as upon the edge of a tool or weapon). (Marathi) Rebus: khāṇḍā ‘tools, pots and pans, metal-ware’.
The message of stone ore is reinforced by the glyphics of buffalo and overthrow of an acrobat woman (kola ‘woman’; rebus: kol‘smithy’):
Rebus: kāḍ ‘stone’. Ga. (Oll.) kanḍ, (S.) kanḍu (pl. kanḍkil) stone (DEDR 1298). mayponḍi kanḍ whetstone; (Ga.)(DEDR 4628).(खडा) Pebbles or small stones: also stones broken up (as for a road), metal. खडा [ khaḍā ] m A small stone, a pebble. 2 A nodule (of lime &c.): a lump or bit (as of gum, assafœtida, catechu, sugar-candy): the gem or stone of a ring or trinket: a lump of hardened fæces or scybala: a nodule or lump gen. CDIAL 3018 kāṭha m. ʻ rock ʼ lex. [Cf. kānta -- 2 m. ʻ stone ʼ lex.]
baṭi trs. To overturn, to overset or ovethrow; to turn or throw from a foundation or foothold (Santali) baṭi to turn on the ground to any extent, or roll; uaurbaṭi, to upset or overthrow by shoving or pushing; mabaṭi to overturn by cutting, to fell trees; baṭi-n rflx. v., to lay oneself down; ba-p-aṭi repr. V., to throw each other; baṭi-o to be overturned, overthrown; ba-n-at.i vrb.n., the extent of the overturning, falling down or rolling; baṭi-n rlfx.v., to lie down; baṭi-aṛagu to bring or send down a slope by rolling; baṭi bar.a to roll again and again or here and there; baṭi-bur to turn over by rolling (Mundari) Rebus: baṭi, bhaṭi ‘furnace’ (H.) Rebus: baṭa = a kind of iron (G.) bhaṭa ‘furnace’ (G.) baṭa = kiln (Santali). bhaṭa = an oven, kiln, furnace (Santali) baṭhi furnace for smelting ore (the same as kuṭhi) (Santali)
Sibri cylinder seal with Indus writing hieroglyphs: notches, zebu, tiger, scorpion?. Each dot on the corner of the + glyph and the short numeral strokes on a cylinder seal of Sibri, may denote a notch: खांडा [ khāṇḍā ] m A jag, notch, or indentation (as upon the edge of a tool or weapon). (Marathi) Rebus:khāṇḍā ‘tools, pots and pans, metal-ware’.
The + glyph of Sibri evidence is comparable to the large-sized 'dot', dotted circles and + glyph shown on this Mohenjo-daro seal m0352 with dotted circles repeated on 5 sides A to F.
Rebus readings of m0352 glyphs:
1. Round dot like a blob -- . Glyph: raised large-sized dot -- (gōṭī ‘round pebble);
2. Dotted circle khaṇḍa ‘A piece, bit, fragment, portion’; kandi ‘bead’;
3. A + shaped structure where the glyphs 1 and 2 are infixed. The + shaped structure is kaṇḍ ‘a fire-altar’ (which is associated with glyphs 1 and 2)..
Rebus readings are: 1. khoṭ m. ʻalloyʼ; 2. khaṇḍā ‘tools, pots and pans and metal-ware’; 3. kaṇḍ ‘furnace, fire-altar, consecrated fire’.
Four ‘round spot’; glyphs around the ‘dotted circle’ in the center of the composition: gōṭī ‘round pebble; Rebus 1:L. khoṭf ʻalloy, impurityʼ, °ṭā ʻalloyedʼ, awāṇ. khoṭā ʻforgedʼ; P. khoṭ m. ʻbase, alloyʼ M.khoṭā ʻalloyedʼ (CDIAL 3931)Rebus 2: kōṭhī ] f (कोष्ट S) A granary, garner, storehouse, warehouse, treasury, factory, bank. khoṭā ʻalloyedʼ metal is produced from kaṇḍ ‘furnace, fire-altar’ yielding khaṇḍā ‘tools, pots and pans and metal-ware’. This word khaṇḍā is denoted by the dotted circles.
Rebus readings of zebu and ‘tiger’? on the cylinder seal shown on 7.31d: khũṭ m. ʻ Brahmani or zebu bull ʼ (G.) Rebus:khũṭ ‘community, guild’ (Santali) kola ‘tiger’ Rebus: kol ‘working in iron’; pañcaloha, alloy of five metals (Tamil).
aṭar ‘a splinter’ (Ma.) aṭaruka ‘to burst, crack, sli off,fly open; aṭarcca ’ splitting, a crack’; aṭarttuka ‘to split, tear off, open (an oyster) (Ma.); aḍaruni ‘to crack’ (Tu.) (DEDR 66) Rebus: aduru ‘native, unsmelted metal’ (Kannada) aduru‘gaṇiyinda tegadu karagade iruva aduru’, that is, ore taken from the mine and not subjected to melting in a furnace (Kannada)
The numerical strokes on the seal may denote the number of ‘ingots?’ of iron made for the guild by the artisan who owned the cylinder seal. It may also denote that he was a worker in ‘iron’ for the smithy guild. An allograph to denote a guild is: footprint shown on some seals discussed in previous section.
Other glyphs used at Shahdad as evidenced by the drawings and artefacts unearthed by Ali Hakemi:
aya 'fish' (Munda) Rebus: aya 'metal (alloy)' (Sanskrit) (cf. Motif shown on copper/bronze plates).
On Shahdad standard there is an endless knot motif like a chain. This motif also appears on Indus writing.
bicha ‘scorpion’ (Assamese) Rebus: bica ‘stone ore’ (Munda)
Other glyphs used at Shahdad as evidenced by the drawings and artefacts unearthed by Ali Hakemi:
aya 'fish' (Munda) Rebus: aya 'metal (alloy)' (Sanskrit) (cf. Motif shown on copper/bronze plates).
On Shahdad standard there is an endless knot motif like a chain. This motif also appears on Indus writing.
If the date palm denotes tamar (Hebrew language), ‘palm tree, date palm’ the rebus reading would be: tam(b)ra, ‘copper’ (Pkt.)
But in one Indian language -- Kannada --, tamara means: tagarm tin (Ko.); tagara, tamara, tavara id. (Kannada.)
The endless knot motif on Indus writing is as shown on the copper plate of Mohenjo-daro. This is a lot different from the continuous endless chain shown on Shahdad standard.
Another comparable motif on Indus writing is a 'chain' like a beaded chain as shown on some seals.
If there is a word to describe the Shahdad glyph of endless knot motif, there are two possibilities: meDhA 'tangle in cord' rebus: meD 'iron'.
Another set of words from Indian sprachbund:
Glyph: kaḍī a chain; a hook; a link (G.); kaḍum a bracelet, a ring (G.) Rebus: kaḍiyo [Hem. Des. kaḍaio = Skt. sthapati a mason] a bricklayer; a mason; kaḍiyaṇa, kaḍiyeṇa a woman of the bricklayer caste; a wife of a bricklayer (G.)
Mohenjo-daro m1406 Seal Drummer vaulting over. Endless knot (chain like beads) motif.Kalibangan seal. k020 Glyphs: threaded beads + water-carrier.
m1457 copper plate Mohenjo-daro Endless knot motif
m1457 copper plate Mohenjo-daro Endless knot motif
मेढा mēḍhā A twist or tangle arising in thread or cord, a curl or snarl. (Marathi) Rebus: meḍ ‘iron’ (Ho.)
few samples of Indus writing glyphs of tree, zebu, knot, tiger -- the glyphs are not exactly comparable in style but the key is that these specific glyphs are deployed on seals. There are many more recording these glyphs.
Hieroglyphs from a vase in Tell Asmar (29-27th cent. BCE). Pair of tigers, pair of zebu; a person holding two snakes; eagle and lion attacking a zebu.
m0309 Mohenjo-daro seal (Tree and a person on a tree branch)
Harappa tablet h188A
meḍha ‘polar star’ (Marathi). Rebus:‘iron’ (Ho.)
ḍato ‘claws or pincers of crab’ (Santali) rebus: dhatu ‘ore’ (Santali)
Crab ‘kamaṭha’; rebus: kampaṭṭam ’mint’ (Tamil Malayalam) kamaṭa = portable furnace for melting precious metals (Telugu)
kuṭi 'tree'. khũṭ m. ʻstump of tree’ (Marathi) Rebus: kuṭhi‘smelter furnace’ (Munda)
tamar, ‘palm tree, date palm’ (Hebrew) Rebus reading would be: tam(b)ra, ‘copper’ (Prakrit)
Source: "Catalogue de l'exposition: LUT/xabis 'Shahdad'- Premier Symposium Annuel de la recherche Archéologique en Iran, Festival de la Culture et des arts, 1972," and published in Tehran. The text on p. 20 (French portion of the publication) identifies the bulla (No. 54 in the catalogue) as "Boule en terre cuite rouge creuse qui contient des cailloux. Décor estampé. Diam: 6 cm, Xabis "Shahdad" Kerman. 2ème moité du IV mill. av. J.-C. No. F.258/48."
Archaeometallurgical investigaions on bronze age metal finds from shahdad and tappeh yahya (I.R. Iran))more by David Meier (Iranian Journal of Archaeological Studies, 1:2 (2011)
http://www.academia.edu/2182599/archaeometallurgical_investigaions_on_bronze_age_metal_finds_from_shahdad_and_tappeh_yahya_I.R._Iran_
IJAS_1.2_2011_dmpm.pdf
Archaeometallurgical investigaions on bronze age metal finds from shahdad and tappeh yahya (I.R. Iran))more by David Meier (Iranian Journal of Archaeological Studies, 1:2 (2011)
http://www.academia.edu/2182599/archaeometallurgical_investigaions_on_bronze_age_metal_finds_from_shahdad_and_tappeh_yahya_I.R._Iran_
IJAS_1.2_2011_dmpm.pdf
http://www.scribd.com/doc/155933212/Archaeometallurgical-investigaions-on-bronze-age-metal-finds-from-shahdad-and-tappeh-yahya-I-R-Iran-more-by-David-Meier-Iranian-Journal-of-Archaeo
Bronze in Archaeology: A Review of the
Archaeometallurgy of Bronze in Ancient Iran
Omid Oudbashi, S. Mohammadamin Emami, and Parviz Davami This is dated 1980 and included in:
T. Wertime & J. Muhly (Eds.), The coming of the Age of Iron (pp. 229–. 266). New Haven: Yale University Press.
http://www.scribd.com/doc/155936341/Bronze-in-Archaeology-a-Review-of-the-Archaeometallurgy-of-Bronze-in-Ancient-Iran-1980-Omid-Oudbashi-S-Mohammadamin-Emami-and-Parviz-Davami
The Early Bronze Age of Iran, Tepe Yahya (Lamberg-Karlovsky&Kohl). Expedition, Spring/Summer 1971
http://www.scribd.com/doc/156061928/The-Early-Bronze-Age-of-Iran-Tepe-Yahya-Lamberg-Karlovsky-Kohl
BRONZE AGE, in Iranian archeology a term used informally for the period from the rise of trading towns in Iran, ca. 3400-3300 B.C., to the beginning of the Iron Age, ca. 1400-1300 B.C. It was originally adopted as part of a chronological system based on assumptions about successive changes in the use of raw materials for tool manufacture, but, along with Iron Age and other comparable terms, it has long since lost any precise meaning in relation to technology. More commonly today, however, it simply refers to archeological sites and events regarded as occurring after the Neolithic (more precisely, after the Chalcolithic) era and before the Iron Age, and this sense is the one that has been adopted here.
Archeological knowledge of Bronze Age Iran has been derived primarily from intensive regional studies in which systematic surface surveys have been combined with excavation at sites having long, well-defined stratigraphic sequences and with more limited excavations designed to obtain information on specific periods (Figure 29; for an outline of the results of these excavations, a detailed chronology, a discussion of chronological problems, and a full set of references, see Voigt and Dyson).
During the Bronze Age the populations of the Iranian plateau, bounded on the east by the Hindu Kush and the Himalayas and on the west by the lowlands of Ḵūzestān and Mesopotamia, prospered greatly, owing to rich natural resources and the overland trade routes between the western lowlands and the Indus valley, central Asia, and Afghanistan. There is evidence that at the end of the 4th millennium B.C. settlements throughout Iran were linked in a common cultural network, the “Proto-Elamite horizon.” Subsequently, however, distinct regional cultural and political systems and a major division between eastern and western Iran developed. As these regions exhibited strong cultural continuity throughout the Bronze Age, cultural development in each will be traced from the Proto-Elamite period.
Southwestern Iran. Modern archeological research on Iran began in the lowlands of Ḵūzestān, known in antiquity as Elam. This region passed from the prehistoric into the protohistoric period in the mid-4th millennium B.C. The most important site in the region is Susa, where in 1897 a French mission began work that continued intermittently until 1977. In the early years large settlement areas were excavated; more recently the focus has been on detailed stratigraphic analysis (see Carter and Stolper). The results of intensive surface surveys on the surrounding Susiana plain have been summarized for this period by J. Alden (1987) and R. Schacht (cf. Wright, for the adjacent Deh Luran [Dehlorān] plain).
At Susa a great many texts in Proto-Elamite script (including both pictograms and numerical symbols) have been found on small clay tablets dated to the end of the 4th millennium (Meriggi, 1971). This script was superseded by cuneiform writing borrowed from Sumer in about 2300 (Carter and Stolper). The pottery of the earliest Proto-Elamite level (Susa III) is quite different from that of the underlying (Susa II) deposits, which are contemporary with the Late Uruk period in Mesopotamia (ca. 3500-3100 B.C.); in contrast, the Susa III pottery has parallels with that of the Jemdet Nasr (Jamdat Naṣr) and Early Dynastic I period in Mesopotamia (ca. 3100-2800 B.C.). Proto-Elamite Susa is estimated to have had a total area of about 11 ha, but the excavated architecture provides little information on community organization. Elsewhere on the Susiana plain there were only small, scattered settlements.
The influence of Susa, revealed through the presence of Proto-Elamite tablets, cylinder seals, products bearing seal impressions, and selected pottery types, extended far to the east and north, where trade in raw materials and manufactured goods among a series of cities and towns was well established by 3100 B.C. The geographic range of this Pro-Elamite network encompassed the plateau as far east as Shahr-i Sokhta (Šahr-e Sūḵta) in Sīstān and Tepe Hissar (Ḥeṣār) on the Damghan (Dāmḡān) plain in the north. The archeological evidence from Proto-Elamite sites differs, and the exact nature of the economic and political ties among them therefore remains problematic. Nevertheless, such settlements as those of Susa III, the Banesh (Baneš) period at Tal-e Malyan (Tall-e Malīān), Sialk (Sīalk) IV:2, Tepe Yahya (Yaḥyā) IVC, and Shahr-i Sokhta I/II produced Proto-Elamite texts and glyptic finds that suggest both shared ideology and economic ties (Carter and Stolper; Alden, 1982; Amiet, Dyson, 1987; Finkbeiner and Rollig; Lamberg-Karlovsky, 1977; idem and Tosi; Weiss and Young).
To the east of Ḵūzestān the prehistoric period is well documented in the Kor river basin of Fārs province. The major excavated Bronze Age sites in this region are Malyan (Sumner) and Darvazeh (Darvāza) Tepe (Jacobs). Surface surveys conducted by Louis Vanden Berghe, William Sumner, and others have shown significant changes in settlement patterns and economic life during this period (Sumner, with references). In the Banesh (Proto-Elamite period) there was a smaller settled population in this region than in previous times, probably as a result of a broad shift from sedentary farming to pastoral nomadism. Malyan itself was a city, with a built-up area of about 50 ha. (In the Late Banesh period this area and about 150 ha of open space were enclosed by a wall.) Excavation has produced evidence of craft specialization, for example, production of small personal ornaments from imported raw materials. A large number of Proto-Elamite tablets, cylinder seals and sealings, and ceramics from Banesh Malyan are directly related to those in Susa III, evidence of strong contact between the two regions. There is, however, no evidence of political domination by Susa, and Sumner has suggested that Malyan was “the seat of a local tribal khan who exercised some form of political authority over the settled population and the pastoral nomads [of Fārs]” (p. 317). During the later 3rd millennium, when Susa and the lowlands were under the domination of Mesopotamian rulers, Ḵūzestān and Fārs showed greater cultural divergence. In Fārs after the Banesh phase there was a “severe depopulation” of the Kor river basin, lasting approximately from 2600 to 2200 B.C. There is no evidence for agricultural settlement, but the area is assumed to have been used by pastoral nomads.
Settlement data from the succeeding Kaftari phase (2200-1600 B.C.) in the Kor river basin suggest a state organization centered on the walled city of Anshan (Tal-i Malyan) and the reestablishment of ties with the lowlands. The rulers of Fārs also played a role in political developments in Mesopotamia: Both the Akkadian king Maništusu (2269-2255 B.C.) and Gudea of Lagash (2143-2124 B.C.) claimed to have defeated Anshan, and subsequently the city became part of the Elamite political sphere (Carter and Stolper, pp. 13-16; Sumner, pp. 316-18). Little is known about Fārs from 1600 to 1300 B.C.; the population again declined, and the remaining settlements were divided into two geographically distinct cultural groups, named Qale (Qaḷʿa) and Shoga Teimuran (Šoga Teymūrān) by William Sumner (Sumner; Jacobs).
The southeastern plateau. Excavations at Tepe Yahya in Kermān province have uncovered occupation levels dating from the end of the 4th and the 3rd millennium (Lamberg-Karlovsky, 1970, 1977; Potts, 1980); intensive surface surveys have yielded further data (Prickett). In the Proto-Elamite period (IVC) Yahya was a large village or a small town in a sparsely populated region. Excavation has revealed a large building with a number of rooms that contained artifacts associated with economic administration: inscribed Proto-Elamite and blank tablets, seals and sealings, and pottery vessels apparently imported from Elam. This structure has been interpreted as an enclave for foreigners, because contemporary domestic structures on the mound contained artifacts and ceramics typifying a continuous indigenous cultural tradition. During the middle and late 3rd millennium (IVB) Yahya specialized in production of vessels and other small objects from chlorite, a soft stone that is abundant locally (Kohl). These items were exported to Mesopotamia, probably by way of Susa, and to settlements along the Persian Gulf, which conducted a flourishing sea trade extending as far as the Indus valley.
North of Yahya, on a deltaic fan at the western edge of Dašt-e Lūt, Shahdad (Šāhdād, historic Ḵabīṣ) has been explored by means of excavations in a cemetery and a surface survey of the settlement (Hakimi; Salvatore and Vidale). The site was apparently occupied during the Proto-Elamite period, though the evidence has not yet been fully reported. In the second half of the 3rd millennium it was an active production center for artifacts of copper and semiprecious stones like agate, carnelian, and chalcedony. A large cemetery of pit graves yielded metal tools, vessels, and ornaments. Quantities of ceramics bear incised and stamped signs related to the older Proto-Elamite script. Similar pottery with some of the same signs was found in Yahya IVB and A (Potts, 1981). Several cylinder seals from Shahdad, apparently depicting a vegetation goddess, are also paralleled in contemporary levels at Yahya. Unique modeled clay busts of men and women have been compared with the stone sculptures of Early Dynastic II in Mesopotamia (ca. 2700-2600 B.C.). Two types of artifact from the cemetery, compartmented copper stamp seals and miniature columns of limestone, are common at sites of the same period in eastern Iran and central Asia and provide evidence of long distance trade on the eastern plateau. No doubt Shahdad served as a point of departure for the dangerous journey across Dašt-e Lūt to northeastern and eastern Iran.
The farthest eastern extension of the Proto-Elamite network has been documented by a single tablet, seals, and sealings from period I at Shahr-i Sokhta. This site on the Helmand delta, which has been explored by means of surface surveys and extensive excavations of both the settlement and cemetery areas, was founded around 3200 B.C. (Tosi, 1983). By the mid-3rd millennium (periods II-III) it had grown into a major urban center covering 80 ha, surrounded by rural villages each with a surface area of between 0.5 and 2 ha. At the height of its development Shahr-i Sokhta was divided into functional zones, with an area devoted to public and administrative buildings, residential quarters, and a cemetery covering 21 ha. Productive activities were initially scattered but were later concentrated in what may have been a craftsmen’s quarter. Crafts documented at the site include the working of lapis lazuli, turquoise, chalcedony, quartz, and flint, as well as of copper. Pottery was manufactured at a small specialized kiln site (Rūd-e Bīābān) located about 30 km away; the styles of painted pottery were shared with central Asia and Baluchistan. During the mid-3rd millennium Shahr-i Sokhta was apparently the largest settlement on the eastern Iranian plateau. Whether or not a state organization had been achieved remains a matter for speculation. Nevertheless, given the size of the settlement and the complexity of its spatial organization, the presence of a state apparatus in periods II-III seems likely.
The central and northern plateau. The influence of Proto-Elamite Susa can also be seen in the mountains of central western Iran and along the northern east-west overland route via Sialk (near Kashan/Kāšān) to Tepe Hissar. In the central Zagros the best known Bronze Age sequence comes from excavations at Godin (Gowdīn) Tepe in the Kangāvar valley (Young and Levine). In the last quarter of the 4th millennium an enclave of lowland traders (or indigenous administrators with strong ties to the lowlands) had been established there (periods VI-V; see Weiss and Young). A complex of buildings in an open court was surrounded by an oval wall. Within the enclosure such exotic items as tablets (all numerical except for one example inscribed with a single non-numerical character), seals and sealings, and types of ceramic vessels identified with the lowlands were found, as were objects of local manufacture. The latter included pottery typical of preceding occupation levels (period VI) and of contemporary settlements on the surrounding plain and in adjacent valley systems as far north as Bījār and south into Luristan. The oval enclosure at Godin was apparently abandoned in some haste, for numerous pots and other objects were left on the floors of the buildings. Following a brief (?) hiatus the settlement was reoccupied around 2700 B.C. by people with a very different material culture (Godin IV, “Yanik period”; see “Northwestern Iran” below), including dark, burnished pottery with incised and white-filled decoration. These people had apparently migrated to the Kangāvar area (and to the Qazvīn and Malāyer plains) from northwestern Iran and ultimately from across the Caucasus (Burney and Lang, p. 59).
The later Bronze Age is well documented for the Kangāvar region, owing to excavations over a large area at Godin (III: 6-2) and to extensive surface surveying (Henrickson, 1987, with references). Surrounding valleys, including the Māhī Dašt, or Kermānšāh plain, are known only from surface surveys and limited soundings (Henrickson, 1987; Schacht). For Luristan survey data are supplemented by excavations at a series of cemeteries in the Pusht-i Kuh (Pošt-e Kūh; Vanden Berghe); these burial grounds are not associated with settlements and may indicate the presence of nomadic pastoralists in the area. Historical sources from Susa and Mesopotamia attest that in the middle and late 3rd millennium the Zagros valleys were occupied by ethnic groups called Guti and Lulubi and were under the control of the Elamite dynasties of Awan and Shimashki (Carter and Stolper, pp. 10-23; Gadd, pp. 429ff.; Schacht). The archeological evidence (see Henrickson, 1987) indicates that at the beginning of this period, during the occupation of Godin III:6, large parts of the central Zagros shared a distinctive ceramic tradition, with more distant links to Ḵūzestān (Susa IV) and Fārs (Late Banesh). This pattern is generally interpreted as an indication of shared contact and economic (rather than political) ties. When Susa came under the control of the Akkadian dynasty, diverging ceramic styles within the mountains reflect isolation from the lowlands. This isolation appears to have persisted after Susa became part of the Ur III state, though both peaceful and military contacts have been documented in texts. Finally, in the early 2nd millennium settlements like the town designated Godin III:2 were linked in a broad cultural zone, attested by elements of a ceramic style that extended throughout central western Iran. This common style may reflect a degree of economic and political unity as well: It has been suggested that the central Zagros was the location of the kingdom of Shimashki, contemporary with the Sukkalmah dynasty at Susa (Henrickson, 1984).
Farther east the Proto-Elamite occupation of Sialk IV:1-2 was contemporary with Godin VI/V, though it lasted into a slightly later period. Like Shahdad, Sialk is located on a deltaic fan at the edge of the central desert. Limited excavations in the ruins of several small mud-brick structures produced diagnostic artifact types (tablets, glyptic, and ceramics) that clearly demonstrate contact with Godin and Susa (Dyson, 1987). As at Godin, however, other elements of material culture show a continuing local cultural tradition. The importance of Sialk within the Proto-Elamite network may have been owing to its proximity to a major source of copper at Anārak; the geographical location of Sialk was equally critical, for it lay on the route from Susa to the north via Fārs (Amiet). Shortly after the beginning of the 3rd millennium Sialk was abandoned; it was not resettled until the Iron Age, late in the 2nd millennium.
Still farther east, along the northern edge of the desert on the northern east-west route, often called the “high road,” is Tepe Hissar (Schmidt; Dyson and Howard), near Damghan. It is also located on a rich deltaic fan, and its population was able to draw on the natural resources of both mountains and plain. In about 3000 B.C. a Bronze Age town (Hissar Middle and Late II) evolved from the earlier settlement (Hissar I-Early II). It consisted of small houses of mud brick separated by open spaces and unpaved walks. About a third of the town was given over to craft production, especially smelting of copper and production of copper objects and working of large quantities of lapis lazuli, a raw material imported from the area that is now northern Afghanistan. Unoccupied parts of the mounds were used for burials. A major innovation characterized this period of town life: the introduction of reduction kilns for the mass production of burnished gray pottery imitating metal vessel forms. Soon this gray ware had almost entirely replaced painted pottery. Although copper technology was already known in Hissar I, more extensive smelting of copper ores led to an increase in the number and types of metal objects produced in Hissar II. The importation of lapis lazuli and turquoise demonstrates links with the east, but at the same time blank clay tablets of the size and shape characteristic of Proto-Elamite tablets, clay tokens (cones, balls, and other forms), and a single cylinder seal show continuing contact with the west. The large number of burials at Hissar from the middle of the 3rd millennium is evidence of considerable wealth within the community. Although the town was somewhat reduced in area, it contained a special, well-built structure filled with rich materials: copper, gold, and silver vessels and weapons. This building housed a small fire altar in one corner of the main room and may have been a shrine. A compartmented bronze stamp seal with a stepped-square design links it to Altyn Tepe, a contemporary urban center in southern Turkmenia. The building at Hissar was destroyed by fire, clearly as the result of violent attack: Remains of a number of bodies were found sprawled on the floor, and the surrounding debris was filled with stone arrowheads. Little is known about the town at Hissar during the remainder of the Bronze Age. In the last phase of its occupation (III) yellow alabaster or calcite objects increased in quantity. Among them were miniature columns with grooved ends, which have now also been found at Tureng (Tūrang) Tepe in Gorgān, in southern Turkmenia, in Bactria, in Sīstān, and at Shahdad. The contexts of these finds can be interpreted as religious, suggesting that some kind of cult practice linked all of eastern Iran at the end of the 3rd millennium.
North of Hissar, across the Alborz (Elburz) range at the southeast corner of the Caspian plain, lay the town of Tureng Tepe (Deshayes, 1977, with references). Like Hissar it had been founded much earlier and remained occupied into the 2nd millennium B.C. Although the pottery and artifacts of Tureng and Hissar II differ somewhat in style, there are many similarities, and both centers participated in the lapis lazuli trade. The outstanding feature of Bronze Age Tureng was a major terraced mud-brick structure built around 2000 B.C. It was 80 m long and rose 13.50 m into the air, in two stages. It was thus comparable in scale to the contemporary Ur-Nammu ziggurat at Ur. Miniature columns of Hissar type were found on the upper story of this building, together with pottery of the Tureng IIIC1 period. Comparable brick structures have been identified at Altyn Tepe (the High Terrace, 12 m high) and at Mundigak (Mondīgak) in Afghanistan (the Monument Massif of period V). Deshayes concluded that toward the end of the 3rd millennium central Asia and eastern Iran were part of a cultural community that was influenced by Mesopotamia. These terraced structures were certainly cult centers of the type mentioned in the legend “Enmerkar and the Lord of Aratta” (Jacobsen, 1987, pp. 275-319).
Northwestern Iran. Throughout the Bronze Age northwestern Iran, or Azerbaijan, constituted a separate cultural zone, more closely related to adjacent regions to the north and west than to the Iranian plateau. Although geographically a unit, this region often comprised two separate cultural provinces, northern and southern Azerbaijan, divided by Lake Urmia. For the Bronze Age the key site for northern Azerbaijan is Haftavan (Haftavān) Tepe (Burney, 1976; Edwards); important supplementary data have resulted from earlier excavations at Geoy (Gūy/Gök) Tepe (Burton-Brown, 1951), nearby Gijlar (Gejlār) Tepe (Pecorella and Salvini, 1987), and Yanik Tepe (Burney, 1961, 1962). Late in the 4th millennium people with a distinctive material culture including round houses and burnished dark pottery migrated into the area, apparently from the north; closely related material (the Early Transcaucasian, or Kur-Araxes, assemblage) has also been found in eastern Anatolia (Sagona, 1984). In Iranian Azerbaijan the earliest excavated settlement yielding this kind of material is at Geoy Tepe (K:1), cleared only in a deep sounding. Settlements dated to the 3rd millennium are better documented, for example, the Early Bronze Age I and II occupations at Yanik, Gijlar, and Geoy K:2-3 (Burney and Lang, pp. 59-66). In the 2nd millennium these sites were characterized by a very different ceramic assemblage consisting of monochrome- and polychrome-painted buff wares. At the same time Haftavan (period VIB) was experiencing its greatest prosperity. The town was built on a series of terraces, and there is some evidence of functional differentiation of space (Burney, 1974, 1975). Although there are a few parallels with sites in the southern part of the Urmia basin, these northern sites are most closely related to settlements in the Trans-Caucasus and Anatolia, continuing the pattern established at the beginning of the Bronze Age.
To the south of Lake Urmia only the Ošnū and Soldūz valleys have been well documented archeologically (Dyson, 1983, with references). Following a period of abandonment that appears to have lasted through most of the 4th millennium and well into the 3rd, this region was reoccupied by agricultural groups living in sizable towns like Hasanlu (Ḥasanlū) VII. The distinctive pottery is only distantly related to that of northern Mesopotamia and the central Zagros. In the 2nd millennium the presence at sites like Dinkha (Denḵā) Tepe IV and Hasanlu VI of ceramics typical of the Khabur (Ḵābūr) region in ancient Mesopotamia (modern north Syria) reflects strong economic or political ties with the west, particularly the kingdom of Shamsi-Adad (Kramer, p. 105). Northern Mesopotamia and Syria are easily accessible from the Ošnū valley through the Kelešīn pass, and these Iranian sites may have participated in the tin trade, which was dominated by Assyria in the early 2nd millennium. Massive mud-brick walls at Dinkha suggest an urban settlement, but the architecture and settlement layout of this period are not well known because of limited excavations.
The end of the Bronze Age. In the late 1960s, in the absence of regional surveys, careful excavations, and analytical studies of resources, technology and subsistence, the apparent abrupt decline of urban centers in the east, from southern Turkmenia to the Indus valley, was attributed to violent invasion and mass migration. Current research suggests, however, that the decline of urban centers and long-distance trade was a more gradual process, beginning as early as 1850 B.C. and continuing for several centuries at varying rates in different regions (Tosi, 1986). Some areas remained unoccupied, for example, the vicinity of Tepe Sialk, whereas others, like the plain around Hissar, were now abandoned. Gorgān and southern Turkmenia remained inhabited but with greatly reduced populations. The area later known as the Bactrian plain, on the other hand, appears to have been resettled; there towns were replaced by scattered rural villages and administrative centers established along natural water courses or man-made canals (Biscione, 1977).
In the Helmand basin shifting hydrological conditions probably played a role in the abandonment of Shahr-i Sokhta and the immediately surrounding territory. The town appears to have been abandoned gradually, for in each succeeding occupation level more open space occurs until finally, in period IV, only one large building stood on the site. At the same time, however, about forty small nearby villages remained occupied, indicating a change in social and political organization, rather than a depopulation of the area (Tosi, 1980). Subsequently these villages also shifted, probably following the water supply. In southern Baluchistan there is also evidence of continuity of occupation (Jarrige, 1983). The introduction of new crops (rice and sorghum) and of double cropping were among major economic changes that took place late in the 2nd millennium B.C. (Costantini, 1981).
Reconstructions of the linguistic and historical geography of eastern Iran suggest that the area was occupied in the 3rd and 2nd millennia by proto-Indo-Aryan speakers (Burrow, 1973) and that Iranian-speaking groups began to move in between about 1400 B.C. and the early 1st millennium (Gnoli, 1980), three or four centuries after the beginning of the decline of the cities. It is relevant to this problem that horse bones and equestrian figurines have been found for the first time in late 2nd-millennium contexts in southern Baluchistan (Jarrige, 1983). Furthermore, sherds of Andronovo pottery, derived from southern Siberia and traditionally linked by scholars with Iranian tribes, appear for the first time in central Asia at the end of the Bronze Age (i.e., the end of the Namazga/Namāzgāh VI period), half a millennium after the onset of urban decline (Biscione, 1977; L’Asie centrale, 1988).
The transition from the Bronze Age to the Iron Age in western Iran is still extremely difficult to trace and has recently been discussed by Young (1985) and Levine (1988).
Bibliography:
J. Alden, “Trade and Politics in Proto-Elamite Iran,” Current Anthropology 23, 1982, pp. 613-40.
Idem, “The Susa III Period,” in The Archaeology of Western Iran, ed. F. Hole, Washington, D.C., 1987, pp. 157-70.
P. Amiet, “Archaeological Discontinuity and Ethnic Duality in Elam,” Antiquity 53, 1979, pp. 195-204.
L’Asie centrale et ses rapports avec les civilisations orientales des origines à l’âge du fer, Actes du colloque franco-soviétique Paris, 1985, Mémoires de la mission archéologique française en Asie centrale I, Paris, 1988.
R. Biscione, “The Crisis of Central Asian Urbanism in the Second Millennium B.C. and Villages as an Alternate System,” in Le plateau iranien et l’Asie centrale des origins à la conquête islamique, Centre National de la Recherche Scientifique, Colloques Internationaux 567, Paris, 1977, pp. 113-28.
C. Burney, “Excavations at Yanik Tepe, North-West Iran,” Iraq 23, 1960, pp. 138-53.
Idem, “Excavations at Yanik Tepe, Azerbaijan, 1961,” Iraq 24, 1962, pp. 134-49.
Idem, “Excavations at Haftavan Tepe 1971.
Third Preliminary Report,” Iran 11, 1973, pp. 153-72.
Idem, “Excavations at Haftavan Tepe 1973. Fourth Preliminary Report,” Iran 13, 1975, pp. 149-54.
Idem, “The Fifth Season of Excavations at Haftavan Tappeh. Brief Summary of Principal Results,” in Proceedings of the IVth Annual Symposium on Archaeological Research in Iran, 1975, ed. F. Bagherzadeh, Tehran, 1976, pp. 257-71.
Idem and D. M. Lang, The Peoples of the Hills, New York, 1972.
T. Burrow, “The Proto-Indoaryans,” JRAS, 1973, pp. 123-40.
T. Burton-Brown, Excavations in Azarbajan, 1948, London, 1951.
E. Carter and M. W. Stolper, Elam. Surveys of Political History and Archaeology, University of California Publications, Near Eastern Studies 25, Berkeley and Los Angeles, 1984, pp. 103-313.
L. Costantini, “Paleoethnobotany at Pirak,” in South Asian Archaeology 1979, ed. H. Hartel, Berlin, 1981, pp. 271-78.
J. Deshayes, “Tureng Tepe et la période Hissar IIIC,” Ugaritica 6, 1969, pp. 40-163.
Idem, “A propos des terrasses hautes de la fin du IIe millénaire en Iran et en Asie centrale,” in Le plateau iranien et l’Asie centrale des origines à la conquête islamique, Centre National de la Recherche Scientifique, Colloques Interationaux 567, Paris, 1977, pp. 95-112.
R. H. Dyson, Jr., “Introduction. The Genesis of the Hasanlu Project,” in M. M. Voigt, Hajji Firuz Tepe, Iran. The Neolithic Settlement, Hasanlu Excavation Reports 1, University Museum Monograph 50, Philadelphia, 1983, pp. xxv-xxviii.
Idem, “The Relative and Absolute Chronology of Hissar II and the Proto-Elamite Horizon of Northern Iran,” in Chronologies de Proche Orient/Chronologies in the Near East. Relative Chronologies and Absolute Chronology 16,000-4,000 B.P. II, ed. O. Aurenche, J. Évin, and F. Hours, British Archaeological Reports, International Series 379, Oxford, 1987, pp. 647-78.
Idem and S. M. Howard, eds., “Reports of the Tappeh Hessar Restudy Project, 1976,” Mesopotamia (forthcoming).
M. R. Edwards, Haftavan, Period VI. Excavations in Azerbaijan (North-western Iran) I, British Archaeological Reports, International Series 192, Oxford, 1981.
U. Finkbeiner and W. Rollig, eds., Ğamda Nasr. Period or Regional Style?, Beihefte zum Tübinger Atlas des Vorderen Orients B62, Wiesbaden, 1986.
C. J. Gadd, “The Dynasty of Agade and the Gutian Invasion,” in CAH3 II, pp. 417-61.
G. Gnoli, Zoroaster’s Time and Homeland, Serie Orientale Roma 52, Rome, 1980.
R. Ghirshman, Fouilles de Sialk I, Musée du Louvre, Départment des antiquités orientales, Série Archéologique 4, Paris, 1938.
A. Hakimi, “Découverte d’une civilisation préhistorique à Shahdad au bord ouest du Lut, Kerman,” in Memorial Volume of the VIth International Congress of Iranian Art and Archaeology, Oxford, 1972, Tehran, 1978, pp. 131-50.
C. (Kramer) Hamlin, “The Early Second Millennium Ceramic Assemblage at Dinkha Tepe,” Iran 12, 1974, pp. 125-53.
H. Hartel, ed., South Asian Archaeology, 1979, Berlin, 1981.
R. C. Henrickson, “Simaski and Central Western Iran. The Archaeological Evidence,” ZA 74, 1984, pp. 98-122.
Idem, “Godin III and the Chronology of Central Western Iran circa 2600-1400 B.C.,” in The Archaeology of Western Iran, ed. F. Hole, Washington, D.C., 1987, pp. 205-28.
L. Jacobs, Darvazeh Tepe and the Iranian Highlands in the Second Millennium B.C., unpublished Ph.D. dissertation, University of Oregon, 1980.
T. Jacobsen, The Harps that Once . . . Sumerian Poetry in Translation, New Haven, 1987.
J.-F. Jarrige, “Continuity and Change in the North Kachi Plain (Baluchistan, Pakistan) at the Beginning of the Second Millennium BC,” in South Asian Archaeology 1983, ed. J. Scholtsmans and M. Taddei, Naples, 1985, pp. 35-68.
P. Kohl, The Seeds of Upheaval. The Production of Chlorite at Tepe Yahya and an Analysis of Commodity Production and Trade in Southwest Asia in the Third Millennium, unpublished Ph.D. dissertation, Harvard University, 1974.
C. Kramer, “Pots and Peoples,” in Mountains and Lowlands. Essays in the Archaeology of Greater Mesopotamia, Bibliotheca Mesopotamica 7, ed. L. D. Levine and T. C. Young, Jr., Malibu, Calif., 1977, pp. 91-112.
C. C. Lamberg-Karlovsky, Excavations at Tepe Yahya, Iran, 1967-1969. Progress Report 1, American School of Prehistoric Research Bulletin 27, Cambridge, Mass., 1970.
Idem, “Foreign Relations in the Third Millennium at Tepe Yahya,” in Le Plateau iranien et l’Asie centrale des origines à la conquête islamique, Centre National de la Recherche Scientifique, Colloques Internationaux 567, Paris, 1977, pp. 33-44.
Idem and M. Tosi, “Shahr-i Sokhta and Tepe Yahya. Tracks on the Earliest History of the Iranian Plateau,” East and West 23, 1973, pp. 21-53.
L. Levine, “The Iron Age,” in The Archaeology of Western Iran, ed. F. Hole, Washington, D.C., 1987, pp. 229-50.
P. Meriggi, La scrittura proto-elamica, 3 vols., Rome, 1971-74.
E. Negahban, “Preliminary Report of the Excavations of Sagzabad,” in Memorial Volume of the VIth International Congress of Iranian Art and Archaeology, Oxford, 1972, Tehran, 1976, pp. 247-72.
P. E. Pecorella and M. Salvini, Tra lo Zagros e l’Urmia. Ricerche storiche ed archeologiche nell’Azerbaigian iraniano, Rome, 1984.
D. Potts, Tradition and Transformation. Tepe Yahya and the Iranian Plateau During the Third Millennium B.C., unpublished Ph.D. dissertation, Harvard University, 1980.
Idem, “The Potter’s Marks of Tepe Yahya,” Paléorient 7, 1981, pp. 107-22.
M. Prickett, Man, Land and Water. Settlement Distribution and the Development of Irrigation Agriculture in the Upper Rud-i Gushk Drainage, Southeastern Iran, unpublished Ph.D. dissertation, Harvard University, 1996.
A. G. Sagona, The Caucasian Region in the Early Bronze Age, 3 vols., BAR International Series 214, Oxford, 1984.
S. Salvatore and M. Vidale, “A Brief Surface Survey of the Protohistoric Site of Shahdad (Kerman, Iran),” Rivista di archeologia 6, 1982, pp. 5-10.
R. Schacht, “Early Historic Cultures,” in The Archaeology of Western Iran, ed. F. Hole, Washington, D.C., 1987, pp. 171-204.
E. F. Schmidt, Excavations at Tepe Hissar, Damghan 1931-1933, Philadelphia, 1937.
W. M. Sumner, “Tall-e Maljan (Anshan),” RIA VII/3-4, 1988, pp. 306-20.
M. Tosi, “The Archaeology of Early States in Middle Asia,” Oriens antiquus 3-4, 1986, pp. 153-87.
Idem, ed., Prehistoric Sistan I, Istituto Italiano per il Medio ed Estremo Oriente, Centro Studi e Scavi Archeologici in Asia Reports and Memoirs 19/1, Rome, 1983.
L. Vanden Berghe, “Pusht-i Kuh, Luristan,” Iran 11, 1973, pp. 207-09.
M. M. Voigt and R. H. Dyson, Jr., “The Chronology of Iran, ca. 8000-2000 B.C.,” inChronologies in Old World Archaeology, 3rd ed., ed. R. Ehrich, Chicago, forthcoming.
H. Weiss and T. C. Young, Jr., “The Merchants of Susa. Godin V and Plateau-Lowland Relations in the Late Fourth Millennium B.C.,” Iran 13, 1975, pp. 1-18.
H. T. Wright, ed., An Early Town on the Deh Luran Plain. Excavations at Tepe Farukhabad, University of Michigan Museum of Anthropology Memoir 13, Ann Arbor, 1981.
T. C. Young, Jr., “Early Iron Age Iran Revisited. Preliminary Suggestions for the Re-Analysis of Old Constructs,” in De l’Indus aux Balkans, Receuil à la mémoire de Jean Deshayes, ed. J.-L. Huot, M. Yon, and Y. Calvet, Paris, 1985, pp. 361-78.
Idem and L. D. Levine, Excavations of the Godin Project. Second Progress Report, Royal Ontario Museum, Art and Archaeology Occasional Paper 26, Toronto, 1974.
Figure 29. Bronze Age sites in Iran and Afghanistan
(Robert H. Dyson, Jr., and Mary M. Voigt)
Originally Published: December 15, 1989
Last Updated: December 15, 1989
This article is available in print.
Vol. IV, Fasc. 5, pp. 472-478
http://www.iranicaonline.org/articles/bronze-ageVol. IV, Fasc. 5, pp. 472-478
PROTO-ELAMITE HISTORY
By: R. K. Englund
Figure 1. Proto-Elamite administrative account of four sheep herds. (Scheil, 1905, no. 212; scale 1:2) Figure 2. A proto-Elamite account of cereal rations for labor gangs of two superv isors (Scheil, 1905 no. 4997; (Click to enlarge) Figure 3. Complex rotation of the proto-Elamite Figure 4. Numerical sign systems attested in the proto-Elamite text corpus (Damerow and Englund, 1989, 18-30; the numbers located above the arrows indicate how many respective units were replaced by the next higher unit). In the capacity system, the basic sign (= "1" in the systems qualifying discrete units) may have represented ca. 25 liters of grain. (Click to enlarge) . |
"Proto-Elamite" is the term for a writing system in use in the Susiana plain and the Iranian highlands east of Mesopotamia between ca. 3050 and 2900 B.C.E., a period generally considered to correspond to the Jamdat Nasr/Uruk III through Early Dynastic I periods in Mesopotamia. This span is represented in Iran by levels 16-14B in the Acropole at Susa (Le Brun, 1971), as well as Tepe Yahya (Yahyâ) IVC, Sialk (Sîâlk) IV2, and Late Middle Banesh (Baneš). Proto-Elamite tablets are the earliest complex written documents from the region; the script consists of both numerical and ideographic signs, the latter sometimes assumed to represent a genetically related precursor of the Old Elamite language (see iv, below). This supposed precursor language is, however, unknown, and the script itself has been only partially deciphered. Nevertheless, conclusions about the contents of the Proto-Elamite texts can be drawn from contextual analyses and formal similarities to proto-cuneiform tablets from Mesopotamia. In particular, the structure of published documents containing accounts and the use of numerical signs and of certain signs for objects in bookkeeping can be somewhat clarified.
History of decipherment
Since the first Proto-Elamite documents were discovered at the turn of the century (Scheil, 1900, pp. 130-31; Friberg, I, pp. 22-26) approximately 1,450 Proto-Elamite tablets from Susa have been published. Recent excavations at other sites have proved that the script and numerical systems known from Susa were in use at administrative centers ranging across Persia as far as the Afghan border, including the sites of Sialk, Malyan (Malîân), Yahya, and Shahr-i Sokhta (Šahr-e Sûkhta; Damerow and Englund, 1989, pp. 1-2; Stolper, 1985, pp. 6-8; Sumner, 1976; Carter and Stolper, p. 253; Nicholas, p. 45). The texts, written on clay tablets, seem without exception to be administrative documents: receipts and transfers of grain, livestock, and laborers; rationing texts; and so on. There are neither literary nor school texts of the sort known as "lexical lists" from contemporary Mesopotamia. The earlier "numerical tablets" from Godin (Gowdîn) Tepe V and Chogha Mish (Chogha Mîš, q.v.), generally dated contemporary with Uruk IVb and level 17 in the Acropole at Susa, lack ideographic signs and are thus not classified as Proto-Elamite (Weiss and Young, pp. 9-10; Porada, p. 58)
Some scholars have attempted to demonstrate a link between the Proto-Elamite and Linear Elamite scripts (see v, below; Hinz, 1975; Meriggi, 1971-74, I, pp. 184-200; Andre‚ and Salvini), but adducing syllabic values proposed for Linear Elamite has not led to successful deciphering of Proto-Elamite. A preliminary graphotactical analysis of the Proto-Elamite texts has also met with only modest success (Meriggi, 1975; idem, 1971-74, I, pp. 172-84; Brice, 1962-63, pp. 28-33; Gelb, 1975). Other scholars have attempted to establish a connection between Proto-Elamite and proto-cuneiform, which first appeared in Uruk IVa (ca. 3200-3100 B.C.E.) and thus seems to predate Proto-Elamite by about a century (Langdon, p. viii; de Mecquenem, p. 147; Gelb, 1952, pp. 217-20; Meriggi, 1969; Damerow and Englund, 1989, pp. 11-28).
Advances in the decipherment of Proto-Elamite have been hindered to a certain degree by the absence of necessary philological tools. A first step would be a sign list sufficiently dependable and cleansed of redundant variants to offer an approximate idea of the number and frequency of signs in the scribal repertoire, as well as providing a transcriptional instrument for analysis of sign combinations and simple contexts. Such textual work is a pre-requisite for a complete edition of the Proto-Elamite texts.
Sign lists provided by early editors (Scheil, 1905; idem 1923; idem, 1935; de Mecquenem; Meriggi, 1971-74) have proved wanting (Damerow and Englund, 1989, pp. 4-7). The first serious attempt at a formal description and decipherment of Proto-Elamite script was undertaken in the 1960s and early 1970s (Brice, 1962-63; idem, 1963; Meriggi, 1971-74; Vaiman, 1989a). Most recent advances have resulted from a new understanding of the structure of the numerical sign systems, which has provided a powerful tool for semantic identification of a number of ideograms, including those for grain products, animals, and, it seems, human beings (Vaiman, 1989a; Friberg, I; Damerow and Englund, 1989).
Format and semantic hierarchy
Proto-Elamite texts are written on clay tablets similar in general shape and proportions to Mesopotamian clay tablets of the 3rd millennium B.C.E., including Uruk III proto-cuneiform tablets of the later phase. The tablets are thick oblongs, their height and width normally in a ratio of 2:3. Following the convention established in the earliest proto-cuneiform phase, Proto-Elamite scribes used both sides of the tablet. Regardless of the space remaining after two or more entries on the obverse, the scribe usually rotated the tablet around a vertical axis and recorded the totals along the upper edge of the reverse. Larger accounts could have a more complex format (Brice, 1962-63, pp. 20-21; Vaiman, 1989a, pp. 130-32; Damerow and Englund, 1989, pp. 11-13; Figure 1).
Three features distinguish Proto-Elamite tablets from proto-cuneiform documents, however. First, the Proto-Elamite documents were written in a linear script. Second, the first signs on a tablet, the heading, have approximately the same function as the proto-cuneiform "colophon," which is usually inscribed together with the final total on the reverse of the tablet; Proto-Elamite headings never contain numerical notations, however. Third, each entry normally includes an ideogram followed by a numerical notation, a divergence from the strict sequence of numerical sign followed by ideogram in proto-cuneiform texts.
The heading of a Proto-Elamite tablet generally specifies the purpose and authorizing person or institution; the best known such ideographic designation is the so-called "hairy triangle", which seems to represent a leading institution or possibly kin group in Elam. Qualifying ideograms were inscribed within this sign, apparently to designate subordinate institutions or groups (Dittmann, 1986a, pp. 332-66; Lamberg-Karlovsky, p. 210; Damerow and Englund, 1989, p. 16). Following these introductory sign combinations are the individual entries, in horizontal registers without regard to formal arrangement into columns (Figures 2 & 3). The ideograms in Proto-Elamite text entries seem almost exclusively to denote persons, quantified objects, or both; sign combinations seeming to designate persons invariably precede those designating quantified objectswhen both appear in one notation. A sign or sign combination representing a person or title is often introduced by a sign representing his position. Objects are generally designated by ideograms in combination with qualifiers; as yet, however, there are no statistical means of testing the probability that certain signs functioned as qualifiers of presumed substantives.
In Proto-Elamite documents there can be multiple entries with different levels of internal organization. A text may consist simply of a sequence of entries of exactly the same type; an example would be a list of grain rations for a number of different recipients. A text may also embody a hierarchical order of transmitted information, as in the oft-encountered alternation of two different types of entry, perhaps a number of workers followed by the amount of grain allotted to them. In this instance the two entries may be considered to be combined in a more comprehensive text unit. A text may also, however, be highly structured, with many identifiable levels, reflecting, for instance, the organizational structure of a labor unit (Figures 2-3; Nissen, Damerow, and Englund, pp. 116-21).
That all entries seem to contain numerical notations suggests that they represent a bookkeeping system, rather than the distinct sentences or other comparable semantic units of a spoken language. This semantic structure is evidence of a close relation between Proto-Elamite and proto-cuneiform texts. Proto-Elamite headings correspond to the "colophons" that often accompany totals on proto-cuneiform texts. Entries in Proto-Elamite documents correspond to the physically encased notations on proto-cuneiform texts; curiously, the hierarchical structure of individual Proto-Elamite entries is not reflected in a syntactical structure, whereas in Mesopotamian texts this hierarchy continues to be represented in some measure by the graphic arrangement of cases and subcases. Despite different graphic forms, Proto-Elamite texts thus exhibit the same general semantic structure as that of proto-cuneiform texts. This relationship must be considered a strong indication of their relative chronology: The more developed linear syntax apparent in Proto-Elamite texts, in which the graphical arrangement of semantic units has been dispensed with, implies that proto-cuneiform is earlier. This conclusion is in full accord with the established stratigraphic correspondences between Susa and Uruk (Dittmann, 1986a, pp. 296-97, 458 table 159e; Dittmann, 1986b, p. 171 n. 1).
Numerical sign systems. Early work on the numerical notations in Proto-Elamite texts was hampered by inadequate identification of individual signs and in particular of sign systems, which were applied in Mesopotamia and Elam to record different types of objects. Initially there was an attempt to combine a large number of what are now recognized as incompatible numerical notations into a single "decimal" system (Scheil, 1905, pp. 115-18; idem, 1923, p. 3). This attempt was abandoned in 1935, when it was recognized that different numerical systems had been in use in Mesopotamia, particularly for enumeration of discrete objects and for measuring grain by capacity (Scheil, 1935, pp. i-vi). It was, however, mistakenly assumed that the sign had the same decimal value 10 x (instead of 6 x) when representing grain measures as when representing numbers of discrete objects (Thureau-Dangin, p. 29; Langdon, pp. v, 63-68; Vaiman, 1989a), which prevented understanding of capacity notations until the late 1970s (Friberg, 1978-79). Although detailed documentation of the various numerical systems has not yet been undertaken, the formal structure of these systems and their dependence upon the older proto-cuneiform systems are now clear (Damerow and Englund, 1987, pp. 117-21, 148-49 n. 12; idem, 1989, pp. 18-30).
As the semantic analysis of Proto-Elamite is largely dependent upon examination of the contexts in which signs are used, the close connection with proto-cuneiform sources in the numerical systems has been helpful in establishing correspondences between Proto-Elamite and proto-cuneiform ideograms. For example, the sexagesimal system used in Meso-potamia for most discrete objects, including domestic and wild animals, human beings, tools, products of wood and stone, and containers (sometimes in standard measures), is also well attested in the Susa administrative texts, though the field of application seems limited to inanimate objects like jars of liquid and arrows (Damerow and Englund, 1989, pp. 52-53). A decimal system used in Proto-Elamite texts for counting animals and human beings has no proto-cuneiform counterpart.
Bisexagesimal notations qualify barley products, as in contemporary Mesopotamian documents. The numerical system for indicating grain capacity involves signs from the sexagesimal system but with entirely different arithmetical values. This system is well attested in both Proto-Elamite and proto-cuneiform sources and seems to have had the same area of application. In particular, the small units inscribed below are qualifying ideograms for grain products, thus denoting the quantity of grain in one unit of the product. The Proto-Elamite system differs from the proto-cuneiform system in that below the sign only units that are multiples of one another appear (e.g. 1/2, 1/4, 1/8), a simpler system than the somewhat cumbersome use of fractions in proto-cuneiform texts (Damerow and Englund, 1987, pp. 136-41). As with the proto-cuneiform texts, in the Proto-Elamite texts there are numerical systems graphically derived from the basic systems but perhaps applied to different sorts of discrete objects or grain (Figure 3). All these similarities together suggest that the Proto-Elamite systems, with the exception of the decimal system, were borrowed from Mesopotamia; even signs in the decimal system were apparently borrowed from the Mesopotamian bisexagesimal system to represent the higher values 1,000 and 10,000.
Ideograms
Semantic analysis of the objects counted by the decimal system has led to the probable identification of a number of ideograms. The most important are the two signs (Symbol 3) and (Symbol 4) . The graphic form, as well as the association, of the ideogram (Symbol 3) with other signs strongly resembling proto-cuneiform signs known to represent domestic animals, in particular sheep and goats (Symbol 5), suggests the interpretation of this sign as "sheep" (Figure 1). In texts from the essentially rural economy of ancient Persia the large numerical notations qualifying this ideogram and related signs seem to confirm the identification. The fact that the signs are on the whole abstract forms may suggest either a set of symbols for domestic animals common in Mesopotamia and Susiana before the inception of written documents or, more likely, signs borrowed in altered form from Uruk (Damerow and Englund, 1989, pp. 53-55).
It appears that the very common sign (Symbol 4) was used to qualify personal names. All signs or sign combinations in a text may be introduced by it, though more commonly it introduces only the first entry (Damerow and Englund, 1989, pp. 53-55). The same sign was used as an ideogram for objects, together with decimal notations commonly used for counting animals. This double function suggests that the sign denotes a category of workers or slaves. The use of the sign in both ways is firmly established in the text illustrated in Figures 2-3 (Damerow and Englund, 1989, pp. 56-57; Nissen, Damerow, and Englund, pp. 116-21). In the same text numbers of objects represented by this ideogram correspond to a regular capacity measure of barley of 1/2 (Symbol 2), parallel to texts known from contemporary Mesopotamia. Finally, the sign is often used parallel to signs that may thus also be interpreted as referring to persons. One of them is a clear graphic equivalent of the proto-cuneiform sign SAL (Symbol 6), so that both the graphic and semantic correspondences of proto-Elamite (Symbol 4) to proto-cuneiform (Symbol 7), meaning "male slave/laborer" (Vaiman, 1989b), seem clear.
Bibliography
B. Andre‚ and M. Salvini, "Re‚flexions sur Puzur-Inšušinak," Iranica Antiqua 24, 1989, pp. 53-72. | |
W. Brice, "The Writing System of the Proto-Elamite Account Tablets of Susa," Bulletin of the John Rylands Library 45, 1962-63, pp. 15-69. | |
Idem, "A Comparison of the Account Tablets of Susa in the Proto-Elamite Script with Those of Hagia Triada in Linear A," Kadmos 2, 1963, pp. 27-38. | |
E. Carter and M. Stolper, Elam. Surveys of Political History and Archaeology, Berkeley and Los Angeles, 1984. | |
P. Damerow and R. K. Englund, "Die Zahlzeichensysteme der archäischen Texte aus Uruk," in M. Green and H. Nissen, eds., Zeichenliste der Archäischen Texte aus Uruk, Berlin, 1987, pp. 117-66. | |
Idem, The Proto-Elamite Texts from Tepe Yahya, American School of Prehistoric Research Bulletin 39, Cambridge, Mass., 1989. | |
R. Dittmann, Betrachtungen zur Frühzeit des Südwest-Iran, Berlin, 1986a. | |
Idem, "Susa in the Proto–Elamite Period and Annotations on the Painted Pottery of Proto-Elamite Khuzestan," in U. Finkbeiner and W. Röllig, eds., G¦amdat Nasár. Period or Regional Style? Wies-baden, 1986b, pp. 332-66. | |
J. Friberg, The Early Roots of Babylonian Mathematics, 2 vols., Göteborg, 1978-79. I. Gelb, A Study of Writing, Chicago, 1952. | |
Idem, "Methods of Decipherment," JRAS, 1975, pp. 95-104. W. Hinz, "Persia ca. 2400-1800 B.C.," in I. Edwards, C. Gadd, and N. Hammond, eds., The Cambridge Ancient History I/2, Cambridge, 1971, pp. 644-80. | |
Idem, "Problems of Linear Elamite," JRAS 1975, pp. 106-15. C. C. Lamberg-Karlovsky, "Third Millennium Structure and Process. From the Euphrates to the Indus and the Oxus to the Indian Ocean," Oriens Antiquus25, 1986, pp. 189-219. | |
S. Langdon, Pictographic Inscriptions from Jemdet Nasr, Oxford Editions of Cuneiform Texts 7, Oxford, 1928. | |
A. Le Brun, "Recherches stratigraphiques a l'Acropole de Suse, 1969-1971," | |
CDAFI 1, 1971, pp. 163-216. | |
R. de. Mecquenem, Î, Me‚moires de la De‚le‚gation en Perse 31, Paris, 1949. | |
P. Meriggi, "Altsumerische und proto-elamische Bilderschrift," ZDMG Suppl. 1, 1969, pp. 156-63. | |
Idem, La scrittura proto-elamica, 3 vols., Rome, 1971-74. | |
Idem, "Comparaisons des systemes ide‚o-graphiques mino-myce‚nien et proto-e‚lamique," in M. Ruipe‚rez, ed.,Acta Mycenaea II, Minos 12, 1972, pp. 9-17. | |
Idem, "Der Stand der Erforschung des Proto-elamischen," JRAS, 1975, p. 105. | |
I. Nicholas, "Investigating an Ancient Suburb," Expedition 23, 1981, pp. 39-47. | |
H. Nissen, P. Damerow, and R. Englund, Frühe Schrift und Techniken der Wirtschaftsverwaltung im alten Vorderen Orient, 2nd ed., Bad Salzdetfurth, Germany, 1991. | |
E. Porada, "Iranian Art and Archaeology. A Report of the Fifth International Congress, 1968," Archaeology 22, 1969, pp. 54-65. | |
E. Reiner, "The Elamite Language," in B. Spuler, ed., Altkleinasiatische Sprachen, Leiden, 1969, pp. 54-118. | |
V. Scheil, Textes e‚lamites-se‚mitiques, Me‚moires de la De‚le‚gation en Perse 2, Paris, 1900. | |
Idem, Documents en e‚criture proto-e‚lamite, Me‚moires de la De‚le‚gation en Perse 6, Paris, 1905. | |
Idem, Textes de comptabilite‚ proto-e‚lamites, Me‚moires de la De‚le‚gation en Perse 17, Paris, 1923. | |
Idem, Textes de comptabilite‚ proto-e‚lamites, Me‚moires de la De‚le‚gation en Perse 26, Paris, 1935. | |
M. Stolper, "Proto-Elamite Texts from Tall-i Malyan," Kadmos 24, 1985, pp. 1-12. | |
W. Sumner, "Excavations at Tall-i Malyân (Anshan) 1974," Iran 14, 1976, pp. 103-15. | |
F. Thureau-Dangin, "Tablettes a signes picturaux," RA 24, 1927, pp. 23-29. | |
A. Vaiman, "Die Bezeichnung von Sklaven und Sklavinnen in der protosumerischen Schrift," tr. T. Götzett, inBaghdader Mitteilungen 20, 1989a, pp. 121-33. | |
Idem, "Über die Beziehungen der protoelamischen zur protosumeriscnen Schrift," tr. I Damerow, in Baghdader Mitteilungen 20, 1989b, pp. 101-14. | |
H. Weiss and T. C. Young, Jr., "The Merchants of Susa. Godin V and Plateau-Lowland Relations in the Late Fourth Millennium B.C.," Iran 13, 1975, pp. 1-17. |
Source/Extracted From: Encyclopaedia Iranica |
http://www.cais-soas.com/CAIS/History/Elamite/proto_elam_history.htm
cf. Mirrored at http://cdli.ucla.edu/staff/englund/publications/englund1998a.pdf
See: http://cdli.ox.ac.uk/wiki/doku.php?id=proto-elamite
Susa. "Copper came from Magan and later Dilmun through the Persian Gulf (Tallon). Tin reached Mesopotamia through Susa and probably also through some route(s) through the central or northern Zagros to Assur (Larsen; Cleziou and Berthoud; Tallon). The Habur ware assemblage at Dînkhâ Tappa (q.v.; Hasanlû VI) in northwestern Persia reflects strong contact with northern Mesopotamia in the early second millennium (Hamlin). " Henrickson, Robert C., Economy of Ancient Iran, Economy in Pre-Achaemenid Iran http://flh.tmu.ac.ir/hoseini/mad-hakha/articles-1/79.htm
One Meluhhan village in Akkad, 3rd millennium BCE.
meluhhanvillage
See: "This review of recent archaeological work in Central Asia and Eurasia attempts to trace and date the movements of the IndoIranians—speakers of languages of the eastern branch of ProtoIndo-European that later split into the Iranian and Vedic families. Russian and Central Asian scholars working on the contemporary but very different Andronovo and Bactrian Margiana archaeological complexes of the 2d millennium b.c. have identified both as Indo-Iranian, and particular sites so identified are being used for nationalist purposes. There is, however, no compelling archaeological evidence that they had a common ancestor or that either is Indo-Iranian. Ethnicity and language are not easily linked with an archaeological signature, and the identity of the Indo-Iranians remains elusive." C.C. Lamberg-Karlovsky, Archaeology and Language, the Indo-Iranians, in: Current Anthropology, Vol. 43, Number 1, Feb. 2002, pp.63-88.
http://mapageweb.umontreal.ca/tuitekj/cours/IE/LambergKarlovsky.pdf
Potts, DT, A new Bactrian find from southeastern Arabia, in: Antiquity 67 (1993), pp. 591-6.
http://faculty.ksu.edu.sa/archaeology/Publications/Arabia/Bacterian%20Camel%20in%20Arabia.pdf
http://www.scribd.com/doc/156075547/Bactrian-Find-in-southeastern-Arabia-Potts-DT-1993
https://en.unesco.org/silkroad/sites/silkroad/files/knowledge-bank-article/vol_I%20silk%20road_the%20bronze%20age%20in%20iran%20and%20afghanistan.pdf
The bronze age in Iran and Afghanistan, (M. Tosi, S. Malek Shahmirzadi and M.A. Joyenda, pp. 184-216)
http://www.scribd.com/doc/156079084/The-bronze-age-in-Iran-and-Afghanistan-M-Tosi-S-Malek-Shahmirzadi-and-M-A-Joyenda-pp-184-216