[Pramāṇa](https://indike.org/pramana/) › [Studies](https://indike.org/pramana/ins) › [Reference works](https://indike.org/pramana/ins.reference) › [An Encyclopaedia of Indian Archaeology](https://indike.org/pramana/ins.reference.ghosh-1989) › v1.c24 # An Encyclopaedia of Indian Archaeology v1.c24: 19 Weapons An Encyclopaedia of Indian Archaeology v1.c24 - **Language:** English ## Contents - [v1.c24.s1 19.1 Harappa](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1) — 10 parts - [v1.c24.s2 19.2 Copper Hoards](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2) — 19 parts - [v1.c24.s3 19.3 Later Periods](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3) — 35 parts ### 19 Weapons ### [v1.c24.1](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.1) \[ Multipurpose Stone Age tools, many of which could have been used as weapons, have not been dealt with here.\] ### 19.1 Harappa ### [v1.c24.s1.1](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1.1) \*19.1.1 Swords or dirks.\* Marshall, 1931, p. 35, was of the view that there were no swords in the Harappa civilization \[4.9\], probably basing his conclusion on the fact that the discovered blades were thickened in the middle and had blunt blades, thus making them ineffective for being thrust into the body of the enemy. However in 1930-1 were found double-edged swords in a hoard (Mackay, 1938). One of them, 40 cm in length and with a maximum thickness of 10 mm, has two rivet holes at the junction of the blade and the rectangular tang with its thickest part 17 mm wide. Another piece in the same hoard, about 46 cm long and 17 mm thick at the junction of the blade and the tang, also has two rivet holes, slightly chamfered, piercing the blade close to the tang, and has a stout medial rib on both sides of the blade. It was probably used for slashing and not for thrusting. Another broken copper casting appears to be the point of an unfinished sword. All these three, along with another of a similar type with a slight medial thickening and diamond-shaped section, probably represent short swords or dirks. Considering their foreign affinities Mackay, 1938, p. 467, observes that in India, Sumer and Palestine the short sword was used as early as or before 2500 B.C. \[the date of course depending on the date of the Indus civilization itself\]. ### [v1.c24.s1.2](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1.2) \*19.1.2 Arrowheads.\* They have been found in the smaller houses in DK area of Mohenjo-daro (Mackay, 1938, p. 461). Thin flat pieces of copper with long narrow barbs but no tang, they were probably inserted into split bamboos or wooden shafts. Perhaps the prototypes of these arrowheads are the swallow-type arrowheads of the Stone Ages. While a few stone arrowheads have been found at Harappa, there is no such specimen at Mohenjo-daro. ### [v1.c24.s1.3](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1.3) \*19.1.3 Knives.\* It is sometimes difficult to distinguish between a knife and a dagger, but the basic function of a knife is to sever and that of a dagger to thrust. Those that can be regarded as cutting weapons and therefore as knives can be divided into: (1) broad leaf-shaped blade (Mackay, 1938, pls. CXV, 11, CXVII, 9, CXIX, 5, CXX, 4, 5, 10, 11, etc.), the commonest type, in some cases a copper strip wrapped round the short tang in order to thicken it; (2) leaf-shaped blade with a curved tip, the only example being from Harappa (\*ASI-AR 1926-7\*, pl. XXII, 4, 5); (3) narrow and straight, comparatively small in size, the edges narrow and parallel and the blades shwoing signs of repeated sharpening; (4) triangular with upturned points, one from Harappa (\*ASI-AR 1926-7\*, pl. XXII, 12) and two from Mohenjodaro (Mackay, 1938, pls. CXX, 22, 23, CXXII, 4, 5), perhaps used for cutting leather; (5) broad curved-edged, three specimens coming from Mohenjo-daro, the tang riveted to the blade, one (Mackay, 1938, pl. CXXX, 23) still having a handle of wood, \*Dalbergia sissu\* Roxb., sticking to it; (6) narrow curved-edged, five specimens from Mohenjo-daro (Mackay, 1938, pls. CXIX, 3, CXX, 24, CXXII, 2, 3, CXXX, 22, 27), their backs straight and edges slightly curved; (7) hollow-backed, one specimen from Mohenjo-daro (Mackay, 1938, pl. CXX, 13), with a thick concave back from which it gradually fines down towards the convex edge; and (8) double-curved, again a single specimen from Mohenjo-daro (Mackay, 1938, pl. CXX, 14). ### [v1.c24.s1.4](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1.4) \*19.1.4 Daggers.\* The following varieties occur: (1) knife-cum-dagger, which could have been used either for cutting or thrusting or for both; (2) shouldered, with abrupt shoulders which would have withstood considerable pressure and must have provided ample bearing for the haft; (3) mid-ribbed, in some cases with a sharp tang; and (4) narrow double-edged, their concave sides suggesting their use for thrusting. No dagger sheath has been discovered. ### [v1.c24.s1.5](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1.5) \*19.1.5 Spearheads.\* It is difficult to distinguish between certain types of spear- and lance-heads and the larger knives and daggers. The longest one, from SD area of Mohenjo-daro, measures about 39 cm in length, about 13 cm in breadth and 4 mm in thickness: it has two small holes at the bottom of the blade to secure it in a split wooden shaft (Mackay, 1938, pl. CXX, 14). Another, about 25 cm long, 36 cm wide and 33 mm thick, is without tie-holes (Mackay, 1938, pls. CXXI, 33, CXII, 7, CXVII, 8, 11, CXX, 7, 9). A spearhead has become very short due to repeated sharpening and looks like a knife blade or lance (Mackay, 1938, pl. CXVII, 9). Another spearhead (Mackay, 1938, pl. CXVII, 11) has a thin and well-preserved blade and an edged tang-end to facilitate its insertion into a wooden shaft. The holes at the lower base of the blade are wide apart which suggests that the rib was of considerable width. In Egypt broad blades were used as a kind of long sword for slashing (Petrie, 1917, p. 31); on this basis a similar use may be assumed for these specimens also. The spearheads of the Harappa civilization are primitive in form and thin and broad in the blade without any mid-rib and have a tang instead of a socket unlike those from Egypt and Sumer. Most of them could buckle on impact. The leaf-shaped spearhead is universal. No barbed blade has been found, but it has a clear representation on a Mohenjo-daro seal (Mackay, 1938, p. 366, pl. 88, 279). ### [v1.c24.s1.6](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1.6) \*19.1.6 Lance-heads.\* The blades are similar to those of spearheads but being smaller in size they have been classed as lance-heads. More or less triangular in shape, double-edged and pointed, their average length is 8 to 10 cm. ### [v1.c24.s1.7](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1.7) \*19.1.7 Mace-heads.\* While copper specimens come only from Harappa, those of stone—alabaster, sandstone, cherty limestone and green-coloured stone—were doubtless used as weapons. Three shapes have been noticed: (1) pear-shaped, the commonest, as it was indeed in other ancient civilization (Marshall, 1931, p. 36); (2) lentoid; and (3) circular (Mackay, 1938, p. 460, where it is observed that the lentoid variety occurs at Susa and in pre-Dynastic Egypt but not in Mesopotamia). They are very suitable for hand-to-hand fight. The perforation is hourglass-shaped bored from both ends; they were probably lashed to a handle with leather thongs. 'The general type is widespread in time and space; it is found at Susa, in Egypt, in the Caucasus, and extensively in prehistoric Europe ... More distinctive is a bronze or copper mace-head of the late Harappa or Jhukar phase at Chanhu-daro' (Wheeler, 1968, p. 76). ### [v1.c24.s1.8](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1.8) \*19.1.8 Missiles.\* Clay missiles have been grouped into three types (Wheeler, 1968, p. 76): (1) pellets, either round, about 3 cm in diameter, or ovoid up to 5·5 cm in length; and (2) and (3) lumps first compressed in the hand and then lightly baked, the difference between the two being that (2) is lighter, about 170 g, and (3) heavier, about 340 g. Found in large numbers, their shape, material and findspot (near the Mohenjo-daro granary and in the defences of the citadel) make their use as weapons of offence and defence possible. It is however not clear whether they were thrown by hand or projected from a sling. \[At least some of them may be terracotta cakes, 6.27.\] ### [v1.c24.s1.9](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1.9) \*19.1.9 Armour.\* Several small domed pieces of copper, each perforated with two holes, have been found at Mohenjo-daro (Marshall, 1931, 2, p. 533; Mackay, 1938, p. 54). They were perhaps sewn on to a garment to be used as a coat-of-mail. ### [v1.c24.s1.10](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s1.10) Mackay, E.J.H., 1938, \*Further Excavations at Mohenjo-daro\*, 2 vols, Delhi, Marshall, J., 1931, \*Mohenjo-daro & the Indus Civilization\*, 3 vols, London, Petrie, F., 1917, \*Tools & Weapons\*, London. Wheeler, M., 1968, \*Indus Civilization\*, Cambridge.—G.N. PANT ### 19.2 Copper Hoards ### [v1.c24.s2.1](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.1) It is questionable if isolated copper weapons can be regarded as belonging to the Copper Hoard culture \[4.11\]. Further even the Copper Hoard objects include several specimens such as flat celts, also found at the Harappa sites, shouldered celts, bar celts and a unique socketed celt from RAJA KARNA-KA-TILA near Kurukshetra, which may not have been used as weapons. Only such objects as were more or less definitely weapons have been described below. ### [v1.c24.s2.2](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.2) \*19.2.1 Harpoons.\* Harpoons similar in shape to those from the Copper Hoards occur among bone objects of the Magdalenian culture, widely removed from the Copper Hoards in time and space. Though simple barbs from Transcaucasia, Talish and Luristan in Iran are known, they could scarcely have, in spite of the insistence of Heine-Geldern, 1936, 1937, inspired the Indian specimens. Similarly the multibarbed Mesolithic \[4.5\] horn harpoons mentioned by Gordon, 1958, cannot have anything to do with the Copper Hoard ones. The intermediate land-mass has not yet yielded specimens to suggest any cultural link. On the contrary nearer home the harpoons are comparable with the ones depicted on the cave paintings \[13\] at GHORMANGUR, Dt Mirzapur, and indicate the source of inspiration as suggested by Lal, 1951, p. 35. ### [v1.c24.s2.3](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.3) Six sites, all Uttar Pradesh, have produced them: PARIYAR has yielded a few, BITHUR and RAJPUR PARSU six and other sites one each. The harpoon from BISAULI, at present in the Bharat Kala Bhawan, Varanasi, 43 cm long, has a strong medial rib and a long tapering blade. The middle part has three pairs of incurved and pointed bards. At the junction of the barbed portion with the tang there is a pair of knobs of which one is perforated. A cord was inserted in this perforation, also called 'eye', to tie it to the shaft. Originally the weapon was cast but there was some hammering at the later stage. Lal's examination of one from SARTHAULI shows that it is possible that the barbs were cut out from the blade by chiselling and then trimmed up with a file, though they might as well have been cast as barbs and then sharpened by filing. A hook-barbed harpoon made in a mould, in the Horniman Museum, is of a fine massive casting of excellent workmanship (Gordon, 1958, p. 137). Although most of the harpoons so far discovered fall within the general category yet there is some technical difference between those from BISAULI and SARTHAULI: in the Sarthauli specimen the blade is not a well-developed entity and is only 7·5 cm long out of a 28 cm total length of the implement, its barbs being flat; ### [v1.c24.s2.4](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.4) the Bisauli blade is nearly half the total length of the implement and its barbs are incurved. While the Sarthauli piece was cast the Bisauli one was made by trimming a plain spearhead as there are indications of trimming on the outlines of the barbs. ### [v1.c24.s2.5](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.5) \*19.2.2 Antennae swords.\* Eighteen antennae swords have been recorded by Lal, 1951, 13 of them from FATEHGARH and one each from BITHUR and an unknown place. Four more such swords have recently been acquired by the National Museum, New Delhi, from a dealer of Mahesana, Gujarat. To sites outside the Ganga valley have thus to be added MODHERA, Dt Mahesana, with its four such swords and KALLUR, Dt Raichur, with its three. ### [v1.c24.s2.6](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.6) Antennae swords with a comparable shape have been found in the Koban region of upper Iran with the difference that the hilt and handle are separate pieces joined together subsequently, besides having a plain cross-section, as distinguished from the ridged one of the Indian specimens, and having a hole in addition. ### [v1.c24.s2.7](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.7) The antennae swords are long and tapering double-rapiers with a strong medial rib, the hilt bifurcating like the antennae of an insect. The two projections of the hilt stand at an angle of 45° to the base, i.e. the straight hilt. In some cases the other projection of the hilt is so much bent that it almost touches the hilt. The hilt and blade are of one make, with no hole in the hilt. However there is one specimen from SHAHABAD², now in the National Museum, with two holes of the blade, one on each side of the medial rib in the portion. The two edges of the blade were badly battered and irregularly twisted, probably due to their use against some hard object. A similar sword has been reported from n. Caucasus but there are some technical differences between the Caucasian and Indian swords. In the Caucasian specimen the hilt and the blade are two separate parts subsequently joined together. The hilt has a hole and the blade is flat-sectioned. All these characteristics make them entirely independent of each other. Heine-Geldern is of the view that the swords were inspired from the w. of the Koban where they occur between \*c.\* 1200 and 1000 B.C., or even a little later, but Lal, 1951, disagrees. ### [v1.c24.s2.8](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.8) \*19.2.3 Anthropomorphic figures.\* Of these bafflingly unique objects there are seven specimens; three each from BISAULI and FATEHGARH and one from SHEORAJPUR. Their purpose is not known but it is quite probable that they were an all-purpose tool. The curved ‘arms’, deliberately sharpened, could be used for cutting purposes. The comparatively long and blunt ‘legs’ might have served as a grip. Since they are very heavy, some as heavy as 5 kg or even more, they might have proved deadly when hurled from close quarters. The intentionally blunted ‘head’, hammered into a flanged ridge, is suitable for breaking the skulls of animals and enemies (Gordon, 1958, p. 137). D.P. Agrawal, 1971, p. 200, has also suggested that they might have been used as missiles. \[A ritualistic significance of these objects has been suggested and cannot be ruled out easily. If a heavy metal object was necessary to kill an enemy or an animal, such elaborate moulding was hardly necessary.\] ### [v1.c24.s2.9](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.9) The anthropomorphic figure from BISAULI is exhibited in the Allahabad Museum. Its total length is about 32 cm (from ‘head to foot’) and width 28 cm (from ‘arm to arm’). It was first cast and then hammered, as is evident from the visible marks of trimming. The ‘arms’ are incurved and have sharp outer edges. Two other such figures are kept in the Bharat Kala Bhawan, Varanasi, of which one is tall and the other dwarfish. The curved portions of the ‘arms’ of both these types are comparatively thin and the outer edges are very sharp. Hammer-marks visible on the ‘arms’ suggest that this portion was expanded by beating. ### [v1.c24.s2.10](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.10) In this connection reference may be made to a seemingly copper celt from LOTHAL, which has been regarded by van Lohuizen, 1960, as an anthropomorphic figure with arms and legs broken. This view has been upheld by some but disputed by others, \*e.g.\* Agrawal, 1971, p. 192, who doubts the identification on the ground that the Lothal sepcimen has a flat section at the top unlike the nail-head section of the Ganga valley specimens. ### [v1.c24.s2.11](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.11) \*19.2.4 Spearheads.\* Altogether eight spearheads have been discovered, five from SARTHAULI and one each from FATEHGARH, MANPUR and NIORAI. They have also been termed as ‘hooked swords’, though the hook of the Fatehgarh weapon, which is broad-bladed and is commonly called ‘Elliot sword’, is not very clear. All these weapons are more than 70 cm long and in spite of their long blades should be taken to be spearheads and not swords. Gordon, 1958, p. 137, says: ‘Both (Fatehgarh and Niorai) weapons have a projection curving outwards from the tang, and the same contrivance is found on three out of five spearheads found at Sarthauli. The ‘Elliot Sword’, has all the appearance of a sword but the three examples from Sarthauli with similarly furnished tangs are barb-headed and must be intended for spears’. This type of projection is seen on the harpoon-head found at the same site and was evidently a method of fastening alternative to the holed lug. ### [v1.c24.s2.12](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.12) The smallest of the five Sarthauli spearheads is 31 cm long. Like the other one it has leaf-shaped blade with a stout mid-rib and a small flat tang which is only 5 cm long. The longest of the lot is 52 cm long. The medial rib is very prominent. Smith, 1905, has also said, ‘In spite of its length it should be called a spearhead. I possess a Somali spearhead which is 2½ feet \[76 cm\] long without, and 3¼ feet \[98 cm\] long with, the socket. The hook on the side of the tang seems to have been intended for fastening the blade to the shaft by a thong.’ ### [v1.c24.s2.13](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.13) \*19.2.5 Battle-axe.\* A hatchet or \*paraśu\* from SARTHAULI in the Lucknow Museum has an unusual shape and is 17 cm long and 23 cm wide. Its cutting edge is not sharp. Hammer-marks are visible on its surface. It was originally cast but in order to give the final touch some retouching was done with a hammer. This is the only specimen of its kind. ### [v1.c24.s2.14](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.14) \*19.2.6 Double-headed axes.\* Copper double-axes have been reported from other cultural levels. At Harappa there are two specimens of a tool type that recalls the double-axe; the two working ends are not sharp and there are kinks between the edge and the waist. There is another double-axe of the Harappa culture from LOTHAL. Of a slightly different type, it is a rectangular piece of copper flattened at both ends and sharpened. The waist is narrow but not so prominently as in examples from the Hoards. A terracotta double-axe has been reported from Pd III of PANDU RAJAR DHIBI. It is also depicted in the form of a motif on a clay seal. ### [v1.c24.s2.15](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.15) There are three such pieces, one each in the Lucknow Museum, Haripada Museum and Patna Museum. These were apparently intended for use as battle-axes, the shaft being split at the end and the narrow neck of the axe-head firmly bound in the cleft. The largest axe measures 101 cm in length and 40 cm in breadth and might have been intended for ceremonial or sacrificial use. The other two measure 25·5 × 21·5 cm and 26·7 × 18 cm respectively. They are made by casting or hammering out a roughly oval disc of metal, sharpening the edges and then cutting out two more or less circular pieces to form the neck. Two such examples have been found at HALLUR, in the Neolithic-Chalcolithic \[4.6.9, 14.14.8\] level. Its butt end is sharp and rounded in shape as the primary working edge. The butt portion is shorter and narrower than the primary end. \[For BHAGRAPIR specimens see 6.2.\] ### [v1.c24.s2.16](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.16) \*19.2.7 Fort Munro sword.\* This ornamented bronze sword, 44·5 cm long, cast in one piece, from Fort Munro, Punjab (Pakistan), now in Edinburgh Museum, has a modern appearance (Smith, 1905, p. 243) and betrays w. Asian influence. It has a medial rib. The crescent-shaped hilt is well-made. \[According to Lal, 1951, this sword should not be regarded as belonging to the Copper Hoard culture.\] ### [v1.c24.s2.17](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.17) \*19.2.8 Miscellaneous.\* In 1965 the National Museum, New Delhi, acquired eight Copper Hoard objects of which the following three are definitely weapons: (1) a double-barbed harpoon, length 35 cm, with straight barbs, having four bars on each side of the shaft, a medial-ridged blade and a shaft with two projecting knobs, one on each side, one of the knobs having a hole; (2) a double-barbed harpoon with curved barbs, length 37 cm, similar to (1) except the curved barbs; (3) an antennae sword, length from the hilt to the tip-end 47 cm, with one portion of the hilt broken and missing, with two holes, one on each side of the medial ridge. ### [v1.c24.s2.18](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.18) \*19.2.9 Technology.\* According to Agrawal, 1971, p. 242, who has carried out spectroscopic and metallographic analysis of a number of Copper Hoard artefacts, their authors did not know alloying and while knowing the close-casting technique did not use cold work and annealing \[16.9.1\]. In metal forging the Copper Hoards artefacts are different from that of the Harappa and Chalcolithic cultures. The Harappans \[4.9\] deliberately used arsenic, lead and tin alloying, the AHAR² \[4.14.1\] people used lead only and the Malwa \[4.14.5\] and Jorwe \[4.14.6\] people employed lead and tin alloying, but the Copper Hoard objects are generally of pure copper. ### [v1.c24.s2.19](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s2.19) Agrawal, D.P., 1971, \*Copper Bronze Age in India\*, Delhi. \*A.R. Hyderabad Arch. Dept.\* 1939-40, pl. V (for Kallur). Gordon, D.H., 1958, \*Prehistoric Background of Indian Culture\*, Bombay. Heine-Geldern, R., 1936, in \*Jour. Indian Soc. of Oriental Art\*, 4, pp. 87-113; 1937, in \*Bull. American Inst. of Iranian Art & Arch.\*, 5, pp. 7-16. Lal, B.B., 1951, in \*AI\*, 7, pp. 20-39. Smith, V.A., 1905, in \*Indian Antiquary\*, 34, pp. 229-44; 1907, in \*ibid.\*, 36, pp. 53-5. van Lohuizen de-Leeuw, J.E., 1960, \*De Protohistorische Cultuuren van vor-Indie on hum Datering\*, Leiden, pp. 24-5. For sites see respective entries in vol. II.—G.N. PANT. ### 19.3 Later Periods ### [v1.c24.s3.1](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.1) \*19.3.1 Arrowheads.\* Literature refers to several types of arrowheads, such as \*ardhacandra\* (crescentic), \*kṣurapra\* (razor-edged) and \*sūcī-mukha\* (needle-headed). Three main types of iron arrowheads have been found at UJJAIN in levels of the second half of the 1st millennium B.C.: (i) tanged and leaf-shaped, (ii) tanged and bud-shaped with circular section, and (iii) multibladed. Such arrowheads have been reported from the NBPW \[11.16\] levels of several other sites such as BAHAL, TER, KAUŚĀMBĪ, HASTINĀPURA, SISUPALGARH, MAHESWAR, ERAN, etc. (Gupta, 1961, pp. 134-6). They also appear in levels of the 2nd-1st century B.C. and early centuries of the Christian era at HASTINĀPURA, NASIK, BESNAGAR, SAMBHAR, NEVASA, PRABHAS, PATAN TRIPURĪ, etc. They range in length from 2·45 to 13·9 cm. The two basic types, viz. 'leaf-shaped' and 'triangular-barbed', both socketed and tanged, are persistently met with. At UJJAIN they have been found in levels of 750 to 500 B C., 500 to 200 B.C and 200 B.C. to A.D. 500. At HASTINĀPURA they appear in Pds III, 600 to 300 B.C., and IV, 200 B.C. to A.D. 300. At MAHESHWAR-NAVDATOLI they occur in Pds IV, 400 to 100 B.C., and V, 100 B.C. to A.D. 100. At NASIK, VAIŚĀLĪ, etc., they come from levels ranging in date from 600 B.C. to A.D. 500. ### [v1.c24.s3.2](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.2) The arrowheads from KAUŚĀMBĪ, are of 11 types. The following types however need special mention: (i) leaf-shaped with two curved hooks at the junction of the blade and the tang; (ii) with projected points flanked by three-blade hooks; and (iii) with turned blades (if the turning is not accidental). ### [v1.c24.s3.3](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.3) The iron arrowheads from the s. India, Megalithic site \[4.17.9\] of ADICHANALLUR have been grouped under the following heads: taking the characteristics of the blade only, (i) concave in the middle, and (ii) convex in the middle; taking the characteristics of the barb only, (i) with incipient barb, (ii) with straight barb, (iii) curved barb, and (iv) ‘flying fish’ type with artistically curved blade; taking only the angle of the base of the blade in relation to the stem (the stem being always the base of the angle): (i) with out-turned base, i.e., obtuse-angled, (ii) with straight base, i.e., right-angled, and (iii) with in-turned base, i.e., acute-angled, and narrow leaf-shaped. The popularity of socketed stems is shared by spearheads, etc. ### [v1.c24.s3.4](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.4) Without exception, all the iron arrowheads found at Taxila (Marshall, 1951, 2) were intended for shafts made of reeds, not of solid wood, and were provided accordingly with long tangs instead of hollow sockets. (Indian arrow-shafts were of reed, cf. \*śara\*, though those of bamboo, \*veṇu\*, and wooden sticks \*śalākā\*, are also mentioned in literature.) A peculiarity of nearly all the earlier specimens is that there is a considerable length of solid shank intervening between the head and the tang. Perhaps they were a compromise between the socketed and tanged varieties. Their users were perhaps accustomed to socketed piles designed for solid wooden shafts, and when reeds were substituted for solid wood, the old form of pile was retained and a thin tang added behind the shank for insertion into the reed, the shank itself being made solid instead of hollow. Since the reed arrow was in general use in Iran and C. Asia (Stein, 1928, 1, pp. 95-6) as well as India, it may reasonably be inferred that these ‘double-tanged’ arrowheads, as Marshall calls them, were an adaptation of a w. socketed type rather than an e. innovation, and this being so it seems probable that these were introduced in Taxila by the Bactrian Greeks (Marshall, 1951, 2, p. 608, 3, pl. 71). ### [v1.c24.s3.5](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.5) The iron arrowheads of Taxila have been divided into eight different types (Marshall, 2, pp. 548-9), copper/bronze or horn specimens being absent: (i) triangular-shaped, with lozenge cross-section and double tang, length generally 82 cm; (ii) with triangular blade and double tang, length 6·82 cm; (iii) with square cross-section, single tang broken, length 8·90 cm; (iv) with rhombic cross-section, single-tanged, length 5·70 cm; (v) double-tanged with barbed and ribbed blade, length 7·82 cm; (vi) conical with circular cross-section, length 5.40 cm; (vii) three-bladed, with plain tang which is broken, length 5·10 cm; and (viii) four-bladed, barbed, with short shank and long tang, length 12·70 cm. ### [v1.c24.s3.6](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.6) Marshall says that the first five types occur in Bhir Mound and all the types occur in the later settlements of Sirkap. He assigns them to the 1st century B.C. In the Śaka-Parthian period (1st century B.C. to 1st century A.D.) two new types, viz. conical and three-bladed, also make their appearance. The older types also tend to become larger. In the 5th century an eighth type, i.e., with four barbed blades, was introduced (Marshall, 1951, 2, p. 547). The three-bladed type seems to have been specially favoured by the Śakas (Petrie, 1917, p. 34), to whom its introduction at Taxila was no doubt due. Many specimens of this three-bladed as well as of the four-bladed types have been found along with other types at the Dharmarajika and other monasteries destroyed by the Hūṇas in the 5th century. It is a reasonable surmise that the former weapons (four-bladed), which were traditional at Taxila, were used by the defending monks, and later (three-bladed) types by the attacking Hūṇas, though the three-bladed type, which had long before been introduced by the Śakas, may also have been used by the defenders (Marshall, 1951, 2, p. 548). ### [v1.c24.s3.7](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.7) The Taxila type of arrowheads has been unearthed from the bastions and gateway of KAUŚĀMBĪ (Sharma, 1960, pp. 45-7). Like Taxila, Kauśāmbī was attacked by the Hūṇas who were armed with these types of arrowheads. The ‘double-tang’ feature was copied in the spearheads in the e. up to SISUPALGARH and in the s. at various sites at different times. BRAHMAPURI has revealed the existence in the Bahmanī levels of a small collection of flat and thin arrowheads of the leaf-shaped and triangular types with minor variations in outline. The definite individuality of the group is a Muslim contribution to the older tradition (Gupta, 1961, p. 138). ### [v1.c24.s3.8](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.8) Arrowheads from C. Asia collected by Stein (Andrews, 1935, p. 331) furnish proof that Indian influence was carried across Chinese Turkistan with the spread of Buddhism. Arrowheads with or without tang, having obtuse point, hollow-sided and flat-headed, or concave-sided triangles are found there. The edge line of the blade varies in different specimens from straight to a beautiful cutting curve. In this type the shaft of the arrow must have been fitted to the tapering hollow stem of the arrowhead. ### [v1.c24.s3.9](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.9) The history of arrowheads in the post-Gupta period does not mark any significant change. All kinds of arrowheads, so-called ‘Indian’ and ‘foreign’ or ‘socketed’ and ‘tanged’, etc., have been used throughout India. Of that period very few arrowheads have been reported from excavations. This rarity seems, partly at least, due to the changeover in the mode of warfare from the chariot to the horseback which promoted the use of spears, javelins and swords against bows and arrows. Still the use of the bow and arrow, as is evident from literary sources, persisted throughout. ### [v1.c24.s3.10](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.10) The practice of the use of bone arrowheads in the Harappa and Chalcolithic cultures, as found at Mohenjo-daro, Chanhu-daro, LOTHAL and NAVDATOLI, was strengthened in the historical period. The variety in shape, the largeness in production and the vastness of the area of distribution—all combined to make this native practice the most important and popular one. The raw material, bone, in an almost finished form, was easily accessible to the craftsmen at all places, possibly free of cost. Again, it is easy to cut and polish bones and produce sharp points (Gupta, 1961, p. 135). Bone arrowheads have been found at: UJJAIN in periods ranging from 750 B.C. to 250 B.C.; PRABHAS PATAN, MAHESHWAR, ERAN, TER, NASIK, ROPAR (where it is called 'stylus') \[20.5\], etc., from the NBPW \[11.16\] levels; SONPUR in levels III to V; PĀṬALIPUTRA from Pd II; KARAD, TRIPURĪ and KOLHAPUR, from the Sātavāhana levels; KAUŚĀMBĪ, BANGARH, BESNAGAR, HASTINĀPURA, etc., from different levels; and ROPAR from Pd V. At UJJAIN and NASIK two large hoards of several thousand pieces each, belonging to the 3rd-2nd centuries B.C., have been found. ### [v1.c24.s3.11](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.11) The bone and ivory arrowheads of Taxila have been classified into the following four types (Marshall, 1951, 2, p. 664): (i) roughly shaped and sharpened at both ends, the point sharp and the tang tapering, length 7·77 cm; (ii) with smooth circular point and well-defined tang, divided from the tang by projecting flange, length 7·76 cm; (iii) with long circular point with hollow socket at back for insertion of shaft tenon, length 9·27 cm; and (iv) with trilateral point, circular shaft provided with socket-hole behind, length 5·39 cm. These bone and ivory arrowheads had been taken by Cunningham to be spillikins or 'tip-cats', but they are much too small for that game. Type (i) is found in both Bhir Mound and Sirkap and was in use from the 5th century B.C. to the 1st century A.D., but the earlier Bhir Mound specimens are distinguished from the Sirkap ones in having the point of the arrowhead somewhat better finished than the tang. The other three types are found exclusively in Bhir Mound and date from the 4th and 3rd centuries B.C. These Taxila specimens could hardly have been intended for serious warfare, unless they were poisoned. Possibly they were employed for the practice of archery or for shooting birds, or it may be that they were useful to the town police as being less deadly than iron ones. ### [v1.c24.s3.12](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.12) Bone arrowheads (Dwivedi, 1976) have basically been divided into two types: (i) bipointed (narrow lozenge-shaped), and (ii) tanged-pointed. On the basis of the cross-section of the body and the blade portion, these two have been further subdivided into: (i) circular, (ii) square, (iii) lenticular, (iv) planoconvex, (v) biconvex, and (vi) irregular (Gupta, 1961, pp. 136-7). The first type seems to be a multipurpose implement, used as tools for piercing holes in leather and other tough things. These might as well have been used as styli \[20.5\] for scratching letters or designs on pottery or terracotta figurines. Sankalia calls them bone points and believes that many of them might have been used as weaving needles, combs or even ornaments. Their use as arrowheads is however not in doubt. A specimen from UJJAIN had stains of blood of a bird on it. Probably the large number in which they were manufactured shows their use more as arrowheads than as styli. ### [v1.c24.s3.13](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.13) 19.3.2 \*Daggers and knives\*. The dagger is a weapon with a short blade designed for thrusting and used for close hand-to-hand fighting. Generally a dagger and a knife are not distinguished from each other. However by definition a dagger is essentially a weapon with a tapering double-edged blade, whilst a knife has only one sharpened edge, but in fact it is not often easy to be precise in the identification of some specimens. The origin of the dagger and knife can be traced back to the handaxes of the Lower Palaeolithic \[4.2\] Age, some of which might have been used for thrusting apart as for skinning and digging. The handaxe has as a rule a sharp point and a jagged edge with which the primitive man could easily cut down the branches of trees and also the organs of the body of the enemy. ### [v1.c24.s3.14](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.14) Of the two metal objects found at LANGHNAJ, one is a copper knife and the other an arrowhead. The knife, 30 cm long, is single-edged with a long and slightly thick tang. The thin, broad and straight blade, gradually tapering, has a square end with rounded corners. It bulges out a little and is imperceptibly thicker at the junction with the tang, which again is broad with a convex end. At the centre it measures 1.74, at the butt 1.27 and at the point 1.90 cm. It is a unique specimen and seems to be a hunting knife, also used for cutting up a carcass. It might have been hafted to a bone or wooden handle. It was not cast but finished when cold. An analogous knife, straight-edged and backed, comes from Dynastic Egypt; a specimen from Troy is also comparable (Petrie, 1917, p. 24, pl. XXVI, 141). ### [v1.c24.s3.15](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.15) A number of short daggers of bone of the Neolithic \[4.6.1\] period have been excavated at BURZAHOM. The great frequency of these miniature daggers makes it clear that they were manufactured locally. ### [v1.c24.s3.16](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.16) A most remarkable find in copper from CHANDOLI is a dagger with an antennae hilt. As is well-known, FATEHGARH and BITHŪR in Uttar Pradesh, KALLUR in the Deccan, MAHESANA in Gujarat, etc., have yielded swords with antennae hilts but all these are from the Copper Hoards \[4.11; 19.2.2\]. The Chandoli dagger is the first to have come from a well-defined horizon, ¹⁴C-dated to \*c.\* 1300 B.C. It has a leaf-shaped blade having a mid-rib on both sides and a rounded tip. The hilt is a plain strip of copper with rectangular section and with ends split up by means of a chisel as the cut marks at the splitting show. The whole piece is of one make. Its total length is 17·1 cm, the length of the blade 10·6 cm and the breadth of the blade 5 cm. Heine-Geldern, 1936, points out the similarity of the antennae hilts from BITHUR with those from the Koban culture. He links up the Indian specimens to the Aryan migrants round about 1200-1000 B.C. The Chandoli dagger may have thus some w. Asian affinity. ### [v1.c24.s3.17](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.17) A fragmentary knife made out of a rib-bone, 15·70 cm in length and 3·7 cm in breadth, has been reported from AHAR². It has a sharp edge showing a series of small flakes on the upper surface. One of its ends has a depression made by the grinding of the sharp-edged sides to serve as a handle. The underside of the weapon at the handle has some incised marks. It has been dated to \*c.\* 1250 B.C. ### [v1.c24.s3.18](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.18) Of the two fragmentary knives of copper with high tin conents from Pd II of RANGPUR, one is roughly rectangular in section and the other, its blade broken, has a sharp cutting edge. ### [v1.c24.s3.19](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.19) The Megalithic \[4.17.6\] habitational deposits of TAKALGHAT and the filling in the stone circles at KHAPA have yielded four dagger-blades and two knives. All daggers of Khapa are almost of the same type having a short pointed tang circular in section, medium long blade with pointed tip and biconvex section and a ringed base for the blade. The dagger of Takalghat is similar to those of Khapa but is without a ring at the base of the handle, having a solid cylindrical base instead. Knives are evidenced by two specimens, one from Khapa and the other from Takalghat, both alike but fragmentary. They have thin slightly curved blades and a short tang circular in section. ### [v1.c24.s3.20](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.20) An ivory dagger handle from Pd II of ROPAR has a chevron pattern at the lower and upper ends and that middle portion a criss-cross design. The uppermost portion is unfinished. ### [v1.c24.s3.21](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.21) Two knife-blades have been unearthed at PRAKASH. One, from Pd II, has a tang. The other comes from Pd III and has a socketed lower part. ### [v1.c24.s3.22](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.22) RAIRH has produced small light daggers and knives together with other weapons. The daggers are cast and some of these are provided with a raised band between the blade and the tang. Their thick blades slope towards the edges. Knives with curved or straight blades have also been found but on account of their highly corroded condition no idea of the sharpness of the cutting edge can be formed. The objects belong to the Mauryan period. ### [v1.c24.s3.23](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.23) Of the knives and knife from KAUNDINYAPURA, a blade, about 18 cm in length, has a straight side and curvilinear cutting edge, the latter being slightly broken. It was found along with two javelin-heads. Another, 7·6 cm in length, has a small leaf-shaped blade, comparatively flat in profile, and a tapering base. Both are post-Mauryan in date. ### [v1.c24.s3.24](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.24) A bone handle piece with a decoration of grooves on bands in relief has been found in the excavations at MAHESHWAR and NAVDATOLI and is datable to \*c.\* 400 to 100 B.C. It is 9·9 cm long and has a diameter of 2·8 cm and 1·8 cm. It might have served as a handle for a miniature dagger which had a round and tapering tang. A single specimen of what appears to be a dagger pendant has been found in Pd VI of NAVDATOLI. The find shows only the hilt portion of a dagger, blade not extant, and is decorated with incised irregular squares. It is 2·7 cm long and ·6 cm thick. A number of dagger-pendants of ivory and bone, all Mauryan in date, have been found at other sites (Dwivedi, 1973). ### [v1.c24.s3.25](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.25) Except in the matter of length the daggers of Taxila closely resemble the swords. All the six daggers or their parts are straight, double-edged and pointed weapons from about 20 to 30 cm in length and with a cross-guard attached to the blade at the blade of the hilt, except one from Bhir Mound. Only one complete dagger and one dagger-blade, both of the 3rd to 2nd century B.C., come from Bhir Mound, the others being from Sirkap. The Bhir Mound dagger differs from the Sirkap ones in that the blade is straight-edged almost up to the point where it tapers abruptly and there is no cross-guard at the hilt (Marshall, 1951, 2, pp. 544-5, 560). ### [v1.c24.s3.26](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.26) Apart from these some knife- or dagger-handles made of ivory have also been excavated at Taxila (Marshall, 1951, p. 660). All these belong to the Śaka-Parthian levels and are datable to the 1st century A.D. The 1944-5 excavation at Sirkap has yielded an ivory knife, wedgeshaped in section, from a post-structural pit (Ghosh, 1947-8, p. 21). ### [v1.c24.s3.27](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.27) Seven knife-blades have been recorded at MASKI in Pd II levels, except one found in a pit of Pd III. ### [v1.c24.s3.28](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.28) BHOKARDAN had produced seven knives and two complete specimens of small knives. Of the seven three each are broken blades and portions with tangs, while one is a full blade though with a broken tang. Two belong to Pd Ib and the others to Pd II. The complete small knives are types by themselves: one has a plano-convex blade with the convex portion abruptly ending near the tang which is broken; the other has its crescentic blade appearing like a small sickle, the convex edge portion ending abruptly to converge into the tang. ### [v1.c24.s3.29](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.29) BRAHMAPURI (Kolhapur) has produced two examples of hunting knives. One has a slightly curved blade and a tang. The second is only the front part of a broad-bladed dagger or knife. Both belong to the Sātavāhana period. ### [v1.c24.s3.30](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.30) PAUNAR has produced 12 badly encrusted blade fragments. Of these three each belong to Pd II A, five to Pd III and one to Pd IV. One is a complete blade of a knife with biconvex section, one endtruncated and the other having a tapering tang. It belongs to Pd III. ### [v1.c24.s3.31](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.31) Iron knives with straight blade and tanged haft have been reported from PIKLIHAL. The form is widely met at s. Indian sites. ### [v1.c24.s3.32](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.32) NAGARA has yielded knives of two types, one concave-backed and another with a straight back and tapering edge. Specimens of the first type occur at several places including KAUŚĀMBĪ, BESNAGAR, MAHESHWAR, NEVASA, NASIK and SISUPALGARH. There is also a specimen of this type in the Government Museum, Madras. The earliest specimen is perhaps from NASIK, dated to the 3rd or 4th century B.C. At KAUŚĀMBĪ the type occurs in the 6th century A.D. Specimens of the second type are noted at VAIŚĀLĪ, SAMBHAR, BANGARH and Kondivte. The specimen from VAIŚĀLĪ is dated to the 4th or 5th century B.C. Some specimens from NAGARA, however, differ from those noted at other sites in having handles made also of iron. The only dagger discovered at Nagara is fish-shaped and has parallels at ADICHANALLUR. All these knives and daggers of Nagara belong to Pd III. ### 19.3.3 Spearheads ### [v1.c24.s3.t1.1](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.1) The spear is a long-shafted weapon intended solely for thrusting. It is a class name and includes many named varieties like the javelin, lance, etc., and unnamed varieties. Wooden spears were used in the earliest times and are still used by many aboriginal tribes. Stone, bone, horn, ivory and shell have been used for spearheads, but in later times iron is the commonest material. ### [v1.c24.s3.t1.2](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.2) GANGAPUR, a Megalithic \[4.17.6\] site, has produced an almost intact iron spearhead. As compared to the dagger-blades found here, this spearhead has a longer and thinner blade and a longer tang. Its edges are rusted and are hence irregular, the blade thin and biconvex and the tang long and circular in section. It has been assigned to 8th-7th century B.C. ### [v1.c24.s3.t1.3](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.3) A notable iron spearhead has been unearthed at PRAKASH, Pd II. A part of the tang and tip of the blade are broken but what remains is a clear evidence of this object having been used as a spearhead. ### [v1.c24.s3.t1.4](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.4) Fifty-eight examples have been recovered at KAUŚĀMBĪ, almost all in a very good condition. They comprise five main types: (i) leaf-shaped with a small blade and lozenge cross-section; (ii) with a large flattened leaf-shaped blade and lozenge cross-section; (iii) with a solid point, plain tang and square cross-section; (iv) with a socketed head and circular cross-section; and (v) with a socket and rectangular cross-section. Among these type (i) appears to be the earliest. All the types are referable to Pd III, 605 to 45 B.C., according to the excavator. ### [v1.c24.s3.t1.5](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.5) KESARAPALLE has yielded a spearhead from the area of a Megalithic \[4.17.7\] burial. A fragment has also been recovered in Pd II, \*c.\* 5th century B.C. to the middle of 1st century A.D. ### [v1.c24.s3.t1.6](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.6) KAUṆḌINYAPURA has provided a few interesting specimens, mostly belonging to the post-Mauryan period, but one of them is medieval. The following specimens have been recorded: (i) long end of javelin, 40 cm long, having a sharp-pointed broad leaf-shaped blade about 20-30 cm in length, having a solid tenon with a splayed bottom and a solid ribbing in the middle; (ii) leaf-shaped javelin-head, 29 cm in length, and having a socketed tenon, found in association with (i); (iii) javelin-tip with a fish-shaped profile having a splayed base and a round tenon with a flattish circular appendage at the end, apparently meant for fitting it tightly into a long wooden pole, the circular end meant to prevent the iron portion from slipping; and (iv) long leaf-shaped blade of a javelin, lenticular in section, length 16 cm, breadth 4 cm, medieval. ### [v1.c24.s3.t1.7](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.7) Taxila has yielded several types of spear- and javelin-heads and a socketed butt. The former have been broadly classified into the following types (Marshall, 1951, 2, p. 546): (i) spike-shaped spearhead of iron, four-sided with tang for fixing into wooden shaft, of great penetrating power against plate or mail armour, length 19 cm, 3rd to 2nd century B.C.; (ii) four specimens, with the shaft and head in one piece, (a) one shaft broken, 98·50 cm long, the head three-flanged, 3rd to 2nd century B.C., (b) another, similar but with four-flanged head, shaft broken, length 63·50 cm, 1st century A.D., (c) similar to (b) and from same spot, and (d) a javelin-head of narrow leaf shape, broken, length 16·50 cm, probably of the same class of heavy javelins as (a) to (c), though only 7·50 cm of the shaft is preserved, 5th century A.D., (b) and (c), with four-flanged head, being from Sirkap; (iii) leaf-shaped with central rib and tang, tang broken, length 21 cm, 1st century A.D.; and (iv) socketed, two specimens, one dagger-shaped with hollow socket (broken), length 18 cm, 1st century A.D., and the other of knife-blade pattern, length 30 cm, 5th century A D The socketed butt is represented by a solitary conical spear-butt, with horizontal rim. ### [v1.c24.s3.t1.8](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.8) Two spearheads, both socketed and with leaf-shaped blades, are recorded from NEVASA, one 15·10 cm long, 3·40 cm broad, diameter of socket 2 cm, from Pd II (200 B.C. to A.D. 40), no details of the other being available, except that it has a length of 15·10 cm and a socket-diameter of 1·90 cm. ### [v1.c24.s3.t1.9](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.9) From MASKI have been recorded two fragmentary lance-handles and a spearhead: (i) handle of a lance, possibly with a round section; (ii) lance; and (iii) leaf-shaped spearhead. The first two are from megaliths \[4.17.8\] and the third is from an unstratified deposit in a rock-shelter. ### [v1.c24.s3.t1.10](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.10) PAUNAR, Pd II B, has produced an almost complete spearhead with squarish tang and pointed blade. ### [v1.c24.s3.t1.11](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.11) Eight pieces of spearheads come from NEVASA, two of which are socketed, three tanged and the rest blade fragments. Most of them are concentrated in the levels of Pd V, 50 B.C. to A.D. 200, whereas one each comes from Pd IV, 150 to 50 B.C., and Pd VI, A.D. 1400 to 1700. The distribution reveals that the socketed and the tanged varieties are contemporaneous. The maximum extant breadth of the socketed variety is 5·10 cm, while that of the tanged specimens is 3·50 cm, the maximum length of the latter being 17·7 cm. ### [v1.c24.s3.t1.12](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.12) RANG MAHAL has yielded a complete spearhead and sockets of two spears. The spearhead is leaf-shaped with a comparatively long shank, while one of the sockets has a wing-like downward projection over the sheath. All these specimens belong to the Kushan period. ### [v1.c24.s3.t1.13](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.13) NAGARA has provided spearheads of two types: (i) with thin lenticular section and convex-sided blade, tanged, from a late level of Pd III, first centuries of the Christian era; and (ii) with triangular section and convex sides from the middle level of Pd II, 3rd century B.C. to the beginning of the Christian era. Spearheads of type (i) are very common, being found at MAHESHWAR, SISUPALGARH and KAUŚĀMBĪ, from the 2nd century B.C. to the 5th century A.D. The form also occurs in several s. Megalithic sites including ADICHANALLUR. Parallels of type (ii) are difficult to find. ### [v1.c24.s3.t1.14](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.14) The lance and spearheads of RAIRH are cast and then shaped by hammering. To the first of the two types belong a number of pointed dagger-like heads with and without mid-ribs; the second type is represented by weapons which has almost flat blades. One is a tough lance-head, over 38 cm long, provided with a strong mid-rib and a tubular tang in which the lance was fitted. Another example, 23·50 cm, is also without mid-rib but has the same type of tang. Still another specimen is provided with a solid tang which was fitted into the lance. All these weapons belong to 2nd-3rd centuries. ### [v1.c24.s3.t1.15](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.15) SISUPALGARH has produced the following specimens: (i) fragment of a spearhead with fissured socket, from an upper level of Pd III, \*c.\* A.D. 200 to 350; (ii) spearhead, from the late accumulations of Pd III at the gateway; (iii) spearhead, from a deposit contemporary with Pd II B at the gateway, \*c.\* A.D. 100 to 200; and (iv) spearhead constricted in the middle, from the accumulations of Pd III outside the n. gateway-flank. ### [v1.c24.s3.t1.16](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.16) BRAHMAPURI has a few javelin-heads. They are leaf-like or triangular in shape and have simple tangs, but none with sockets. All belong to \*c.\* A.D. 300. ### [v1.c24.s3.t1.17](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t1.17) HASTINĀPURA has yielded a leaf-shaped spearhead with flattened section and with a tang from the latest level of Pd V, medieval. This shows that tanged spearheads were popular throughout. ### 19.3.4 Swords ### [v1.c24.s3.t2.1](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t2.1) The sword is a hand weapon with a longish blade of metal and a small grip to hold it. This definition does not differentiate it from all other weapons, the series of knife, sword, espadon and glaive being an unbroken one; there is no agreed point at which a division would be universally applicable. Neither the length of the blade nor the kind of the hilt nor the method of use would define a sword: all overlap. Definite occurrence of swords is limited. \[For Fort Munro sword see 19.2.7.\] ### [v1.c24.s3.t2.2](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t2.2) A single fragmentary specimen of copper of what possibly could be used as a sword has been recovered from a megalith \[4.17.6\] at KHAPA. Its length as also the sections of its long tang and blade make it distinctive. The tang was possibly accommodated in a wooden cover. The blade, heavily encrusted, has one edge sharp. ### [v1.c24.s3.t2.3](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t2.3) A pre-Mauryan sword, only a portion of its blade surviving, comes from KAUṆḌINYAPURA. It is lenticular in section, the length being 11·80 cm and breadth 4·40 cm. ### [v1.c24.s3.t2.4](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t2.4) RAIRH has produced a good number of sword-blades but their surfaces and edges are so corroded that it is impossible to determine whether they were double- or single-edged. The tang of one of them measures 9 cm and the width of the blade at its widest is 7 cm; its present length including the tang is 31·8 cm. Another fragment of sword-blade is only 15 cm in length. The pieces have been assigned to the 3rd-2nd century B.C. ### [v1.c24.s3.t2.5](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t2.5) MASKI has yielded a sword with blade, but other details are lacking, except that it belongs to Pd II, \*c.\* 200 B.C. to the middle of the 1st century A.D. ### [v1.c24.s3.t2.6](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t2.6) The earliest evidence of swords in s. India is the hoard of iron swords found at ADICHANALLUR, the well-known urn-burial site \[4.17.9\]. Only the blades, many with their spike tangs, are preserved, but the trappings and mounts have perished. The forms of the blades follow three types: (i) leaf-shaped and straight; (ii) parallel-edged and abruptly pointed; and (iii) straight and continuously tapering. All have strongly defined mid-ribs and are proportionately longer than those of n. India. The first two forms are known in n. India but the third type seems to have confined to the s., though all the types remain the fundamental pattern of the straight sword in the s. until modern times. ### [v1.c24.s3.t2.7](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t2.7) The iron swords found at Taxila (Marshall, 1951, 2, pp. 545-6) belong exclusively to the 1st century A.D. They are straight, double-edged and pointed weapons, from 53 to 87 cm in length and with a cross-guard attached to the blade at the base of the hilt, thus resembling the straight double-edged swords found at ADICHANALLUR and also the long \*spatha\* used by the Roman auxiliaries. ### 19.3.5 Caltrops ### [v1.c24.s3.t3.1](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24.s3.t3.1) The caltrop, also spelt 'calthorp', 'caltrap' and 'calthrop' (Stone, 1961, p. 158) and called \*gokharu\* in India, is composed of spikes, either planted singly on the ground or radiating from a common centre, and strewn on the ground to prevent the advance of elephant corp or cavalry, or in a breach in fortifications to hinder attacking troops. One of the commonest forms is four spikes radiating from a common point so that in any position one of the spikes would stand vertically. They are the direct ancestor of the barbed wire entanglements of the present day. Many of the aboriginal tribes like the Adis of Nefa and the Nagas of Nagaland use sharpened stakes called \*panjis\* for the same purpose. They occur in Roman military sites of the early centuries A D. and are reported from ancient China and Japan as well (Stone, 1961, p. 158). The \*Arthaśāstra\* of Kauṭilya, 2.3.21, perhaps refers to them as \*triśūla\*, to be placed outside the gateway of the fort along with other obstacles. _First 87 of 98 passages. [Read on](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24)_ ## Linked texts (39) - [Kauṭilīya Arthaśāstra](https://indike.org/pramana/in.niti.artha.kautilya.2.3.21) — Kauṭilya, Sanskrit - [Harappā](https://indike.org/pramana/inp.site.harappa) - [Mohenjo-daro](https://indike.org/pramana/inp.site.mohenjo-daro) - [Lothal](https://indike.org/pramana/inp.site.lothal) - [Chanhu-daro](https://indike.org/pramana/inp.site.chanhu-daro) - [Rangpur](https://indike.org/pramana/inp.site.rangpur) - [Hastināpura](https://indike.org/pramana/inp.site.hastinapura) - [Kauśāmbī](https://indike.org/pramana/inp.site.kausambi) - [Vaiśālī](https://indike.org/pramana/inp.site.vaisali) - [Pāṭaliputra](https://indike.org/pramana/inp.site.pataliputra) - [Sisupalgarh](https://indike.org/pramana/inp.site.sisupalgarh) - [Nevasa](https://indike.org/pramana/inp.site.nevasa) - [Nāsik](https://indike.org/pramana/inp.site.nasik) - [Maski](https://indike.org/pramana/inp.site.maski) - [Ādichanallūr](https://indike.org/pramana/inp.site.adichanallur) - [Burzahom](https://indike.org/pramana/inp.site.burzahom) - [Navdatoli](https://indike.org/pramana/inp.site.navdatoli) - [Maheswar](https://indike.org/pramana/inp.site.maheswar) - [Langhnaj](https://indike.org/pramana/inp.site.langhnaj) - [Prakash](https://indike.org/pramana/inp.site.prakash) - [Prabhas Patan](https://indike.org/pramana/inp.site.prabhas-patan) - [Bahal](https://indike.org/pramana/inp.site.bahal) - [Sonpur](https://indike.org/pramana/inp.site.sonpur) - [Saradkel](https://indike.org/pramana/inp.site.saradkel) - [Modhera](https://indike.org/pramana/inp.site.modhera) - [Khapa](https://indike.org/pramana/inp.site.khapa) - [Takalghat](https://indike.org/pramana/inp.site.takalghat) - [Pandu Rajar Dhibi](https://indike.org/pramana/inp.site.pandu-rajar-dhibi) - [Bhokardan](https://indike.org/pramana/inp.site.bhokardan) - [Dharmarājikā](https://indike.org/pramana/inp.site.dharmarajika) - [Bhir Mound](https://indike.org/pramana/inp.site.bhir-mound) - [Sirkap](https://indike.org/pramana/inp.site.sirkap) - [Kallur](https://indike.org/pramana/inp.site.kallur) - [Karad](https://indike.org/pramana/inp.site.karad) - [Shahabad](https://indike.org/pramana/inp.site.shahabad) - [Sarthauli](https://indike.org/pramana/inp.site.sarthauli) - [Sambhar](https://indike.org/pramana/inp.site.sambhar) - [Sheorajpur](https://indike.org/pramana/inp.site.sheorajpur) - [Brahmapuri](https://indike.org/pramana/inp.site.brahmapuri) --- Source: [https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24](https://indike.org/pramana/ins.reference.ghosh-1989.v1.c24) · Pramāṇa