The Indus Civilization c8: V. Copper and Bronze: Implements and Tools
The Indus Civilization c8
- Language
- English
V. Copper and Bronze: Implements and Tools
c8.2
THERE was evidently no scarcity of copper and bronze in the Indus valley round about 2800 B.C., for the latter has been found even at the lowest levels of Mohenjo-daro, where one specimen contained as much as 22·2 per cent. of tin. Dr. Frankfort and Dr. Woolley have also noticed that tin was very much used as an alloy to toughen copper in Sumer at a period contemporary with the Indus civilization.
c8.3
It is difficult to say where the metal-workers of the Indus valley obtained their tin, for although it is to be found in several places in India, such as Burma, the Bombay Presidency, Bihar, and Orissa, it is uncertain what sources were available in ancient times. It is a moot point whether the tin used by the early Sumerians and the Indus Valley people came from the same place, but the writer considers that this, in all probability, is not the case.
c8.4
With regard to the copper used by the ancient inhabitants of the Indus valley, supplies of this metal may have come from Rajputana or Baluchistan. Dr. Desch has shown, however, that the copper and bronze of the Indus valley, like that of the Sumerians, contains an appreciable amount of nickel, and as Mr. Peake has suggested that a possible source of the Sumerian copper was Oman, where old workings have been found in which the ore is of this type, it has been suggested that this district of Arabia also supplied the Indian cities. The presence of nickel has, however, also been detected in the copper from the old, but still worked, mines of Chota Nagpur in India, and it is perhaps safer assume for the present that the copper used at Mohenjo-daro and Harappa did not come from abroad.
c8.5
The use of tin to alloy copper has the effect of hardening the latter and also facilitates casting, but too much tin makes a brittle product. As a rule 9 to 12 per cent. of tin suffices for ordinary purposes, but in some of the bronze from Mohenjo-daro over 26 per cent. has been found. It would seem that the metal workers of the city, like the early Sumerians, either used no fixed proportion of tin to copper or obtained their bronze ready mixed from sources over which they had no control.
c8.6
The finding of a considerable amount of copper ore in a brick-lined pit at Mohenjo-daro proves that this metal was smelted there, although no furnaces have been discovered in which it could have been reduced. A very simple method would have been to heat the ore mixed with charcoal in a hole in the ground by means of a blast. The shape of several copper ingots from the city, one of which weighs two and a half pounds, suggests that this was done, and that the metal when molten ran through a channel into an even, semi-circular depression in the ground; a projection on one side of some of these ingots shows how the moulds were filled. The flat upper surfaces of all the ingots are very puckered owing to unequal cooling.
c8.7
No trace has been found of soldering in the case of copper or bronze work, although this process was employed on gold and silver. Rivets were generally used to join the two first-named metals, but in one instance a handle was fastened to a bronze cover with a rivet, which was then made more secure by running a little molten metal over it.
c8.8
With the exception of three very plain silver vessels, all the metal utensils yet discovered are made of copper or bronze, the majority of these being extremely simple in shape and in many cases following the forms of the pottery. They were either beaten out of sheet metal or were cast. It was not the custom to engrave these metal vessels in any way, and the only case of fluting occurs on a censer whose edge is also ornamented with indentations. Some of the jars with very narrow bases had the latter made separately and then lapped on, but in no case does an edge appear to have been turned over to stiffen the rim. With the exception of certain frying-pans no metal utensil has yet been found with a handle or spout.
c8.9
Blade-axes of very simple form appear to have been used for a number of purposes, and of these there are two kinds, the long and narrow (Pl. N, 2), and the short and broad. The longer ones probably served as adzes, and were supported by the handle more or less their entire length, while the shorter specimens are thought to have been set at right angles to a handle cleft to take the blade. Both types, whether of copper or bronze, have double-sloped, lunate edges, slightly sloping sides, and square-cut butts, and were cast in moulds and finished off with the aid of the hammer and an abrasive.
c8.10
Several defective castings that have been found show that the blades fresh from the mould were of the same shape as the finished articles, except for their rough surfaces, so that the metal must have been tempered and consolidated mainly with a hammer; even a bronze blade with a small amount of tin could be so hardened in this way that it equalled mild steel. Some of the blade-axes are of considerable size, and one, eleven inches in length, weighs four pounds three ounces. It cannot be assumed that these were used in war, as there is no evidence that their inhabitants were ever seriously threatened by outside enemies until the last phases of the existence of the Indus cities; but they could have been used in working wood, in the chase, and perhaps against dacoits, like the little battle-axe so often carried by the countryman of Upper Sind.
c8.11
One very fine bronze axe-adze, about ten inches long, has a hole in the head to take a handle, and is the first socketed implement to be found at Mohenjo-daro (Pl. N, 1). This axe-adze is very similar to some discovered in the basin of the Kuban river in the Caucasus, while its tubular collar resembles that of specimens from Tepe Hissar in North-east Persia. As it is at present the only one of its kind unearthed in the Indus valley, its date cannot be fixed until other examples are found. Socketed implements were, however, quite common in Sumer at a time contemporary with the Indus cities, although nothing has yet been found there of the same date that is strictly comparable with the specimen from Mohenjo-daro. A model axe with handle, in clay, which has been lately found in the city has the head coloured to imitate metal, and is quite unlike anything that has been discovered elsewhere. This, too, has a socketed head, and is a further proof that this type of weapon was known to the native metal-workers.
c8.12
One of the most interesting tools yet unearthed is a bronze saw, sixteen and a half inches long, which once had a broad wooden handle; three rivet-holes indicate where it was attached to the blade. The astonishing thing about this saw is that the toothed edge is undulated in order to prevent its binding in a cut, in contrast to the modern method of setting each tooth separately to achieve the same end. Saws with the teeth set in any way whatever have hitherto been unknown before the time of the Romans, and it is extremely interesting to find so much earlier a specimen, and one, moreover, in such a good state of preservation.
c8.13
Two copper swords, or dirks, the larger of which is eighteen and a half inches long without the missing handle, were among the most welcome finds at Mohenjo-daro. Both are considerably thickened down the middle for strength and are in excellent condition, but, to judge from their blunt points, they would have been more useful in slashing than in thrusting. Swords of so early a date are extremely rare at any site; at present the only specimen to be compared with them is one lately discovered at Tell el-Ajjul in Palestine by Sir Flinders Petrie, and assigned by him to the period of the Old Kingdom of Egypt, a period which is approximately that of Mohenjo-daro.
c8.14
The spear-blades used in the Indus cities are an unusual type, the largest, of very thin bronze, being over fifteen inches long and nearly five inches broad. A blade of this description must have been very liable to buckle, and to prevent this a mid-rib of wood, probably made in one with the shaft, was attached to the blade by thongs or wire passed through the metal. Even with this stiffening the edges of such a blade must frequently have been bent out of shape, and it is still a matter for conjecture why the Indus Valley people were so economical in the metal for their spear-blades when both copper and bronze were so lavishly expended on the manufacture of other tools and implements. Certain small, broad blades have also been discovered which are thought to be lance-heads.
c8.15
Daggers are sometimes difficult to distinguish from knives, and in some respects closely resemble the simpler forms of the latter. They are long and leaf-like in shape, occasionally with a rough mid-rib, but more often flat; they have an immature tang either with, or without, rivet-holes. Quite a number of the knives are of curious shapes, a very common type having only one edge and an upturned point; but this may have been made to serve a special purpose, perhaps leather-work. A broad, leaf-shaped blade with two edges was also very popular. Most of the blades appear to have been cut out of sheet metal, and many have been sharpened so often that the edges are pronouncedly concave. A portion of the original wooden handle of one knife, which is in a fair state of preservation, was made of a wood (*Dalbergia sisu*) that is commonly used at the present day for the handles of tools. Mention has already been made of the flint knives so much used by the Indus Valley people and of which so many specimens have been found.
c8.16
Arrow-heads fashioned from thin sheet metal with pointed wing-like barbs are very common, but no stone arrow-head has yet been found at Mohenjo-daro. The arrow-heads seem to have been strengthened in the same way as the spear-heads—that is, by the addition of a wooden mid-rib—but owing to the smaller surface they were cemented rather than tied in.
c8.17
The proximity of a river famous for its fine fish explains the number of fish-hooks found at Mohenjo-daro. These were made of bronze and have a single barb, and an eye at the end of the shank secured them to the line. Only in one or two cases is the point of a hook offset, but this may be accidental. Several hooks were discovered with cotton thread or cord wound round their shanks, and may quite likely have been the property of small boys.
c8.18
No less than four types of razor were used by the Indus Valley people. The most popular has opposite blades of dissimilar shape at the end of a long rod-like handle. Another kind is L-shaped, and a third has a handle which bends backwards parallel with the blade and terminates in a duck’s head. The fourth type, which is very rare, consists of a long, thin, straight blade with a rounded edge at one end. Possibly some of these razors were used for removing hair from the body as well as from the face, and, to judge from the numbers unearthed, it appears that they were used by both sexes.
c8.19
Chisels are very plentiful, all of them having double-sloped edges and made from round or rectangular rods of copper and bronze. Some of them resemble specimens found at other ancient sites, but one variety with a flat, substantial tang is peculiar to the Indus valley.
c8.20
In one hoard of copper found at an early level at Mohenjo-daro, some of the tools have pictographs incised upon them, and what appear to be numbers. Inscribed tools and weapons are, however, far more common at Harappa, a fact which with others has given rise to the suggestion that this city is slightly the earlier of the two, though it is impossible to say whether it was also the first to be deserted. The reason for this custom of inscribing tools and implements will, of course, remain uncertain until the script of the Indus valley has been deciphered, but it is thought that the hoard from Mohenjo-daro may have been temple or government property which was numbered to prevent theft. In ancient Egypt tools sometimes had numbers incised upon them for this very reason.
c8.21
Bronze is used for figurines as well as for models of small animals, and the casting in most cases shows great technical skill, the method employed being the *cire perdue* or wax process, a technique well known in Sumer and ancient Egypt. A small bronze figure of a dancing-girl is a perfect piece of casting, while closely approaching it in excellence is a little figure of a buffalo (Pl. K, 7). No examples of beaten work have been found other than the utensils; but it is quite possible that this kind of work was used for larger pieces of statuary of the type found in Sumer and in Egypt, and that these pieces have not survived or, more likely, are still to be unearthed. In any case, there seems no doubt whatever that the metal-smith held a prominent place among the inhabitants of the Indus valley and had plenty of material on which to exercise his skill.
c8.22
Lead was also known and worked, but only one leaden utensil, a small dish, has yet been discovered. Leaden figurines and animals are also non-existent; in fact, this metal is usually unearthed in the form of irregular lumps. It sometimes occurs as an impurity in the copper and bronze, but the metal-workers seem never to have realized that if it had been used sparingly as one of the alloys of bronze, casting would have been considerably facilitated. Lead has been worked in India from very early times, and there is no doubt that ample supplies were available for the Indus Valley cities. It is, indeed, possible that much, if not all, of the silver used was procured from lead, the process of extraction being exceedingly simple.
c8.23
Quite a number of stone mace-heads, either pear or lentoid in shape, indicate that this form of weapon was much in use. A hard, cherty limestone and alabaster are the most usual materials for the pear-shaped specimens, while the lentoid type is commonly made of hard, dark-green stone. The latter seems to have been the more popular, to judge from the number discovered, but no mace of either shape which bears any inscription or decoration has yet come to light.
c8.24
Saddle-querns made from a hard, gritty stone have been found in great numbers at Mohenjo-daro, and were used for grinding grain. These have a pronouncedly convex base, perhaps because this helped to set them firmly in the ground, in spite of the fact that models from both cities show that these querns were sometimes placed in large pans to prevent any waste of flour. The cereals known to the inhabitants of the Indus valley included both wheat and barley, for samples of these have been discovered which, although black and carbonized, were sufficiently preserved to be identified as *Triticum sphærococcum* and *Hordeum vulgare*. Other stones with large hollows perhaps served to prepare curry-powders, for the local labourers on the site declared that they used identical grinders for this purpose.
c8.25
Stone palettes, usually of slate, have also been found, though flat, natural pebbles were employed for the same purposes. These are all of the plainest description, and appear to have been used for rubbing down colours, as traces of a red oxide have been observed on some of them. This colour probably had a number of uses, including the slips on pottery.
c8.26
Stone weights are extremely common, and are almost invariably made of carefully polished, and sometimes veined, chert (Pl. K, 2). The cube is by far the most usual form, and weights of this type have been found ranging from minute specimens to those weighing as much as 274.938 grammes. Spherical weights with a flat top and base come next in popularity, and of these, too, quite large sizes have come to light; a third kind is the barrel-shaped weight which is not unlike those that have been found in Egypt, Sumer, and Elam. Some weights of exceptional size—one being over twenty-five pounds—are conical, and have two converging holes in the top for the ring of metal or rope by which they were lifted.
c8.27
Alabaster, limestone, quartzite, slate, and jasper are among the other stones which were used to make weights; but chert predominates and was always used for the cubical type. The stone was first roughly flaked into cubes, then rubbed down smooth upon an abrasive, and finally polished. No weight has been found with a number or any mark upon it, with the exception of one barrel-shaped specimen which bears two pittings, but these may only have been made to rectify its weight. Pebble weights are also known, but they are very rare.
c8.28
Mr. A. S. Hemmy, who has examined the weights from Mohenjo-daro and also some from Harappa, states that the system used was binary in the case of the smaller weights and decimal in the case of the larger ones, the succession being in the ratios 1, 2, ⅓ × 8, 4, 8, 16, 32, 64, 160, 200, 320, 640, 1600, 3200, 6400, 8000, 12,800, and that the unit weight had the calculated value of 0·8565 gramme. The same authority is of the opinion that, with the possible exception of a very few weights, the Indus Valley system bears no relation to that of the Babylonians, and though one group of weights strongly resembles the ancient Egyptian Beqa, he regards them, nevertheless, as quite independent. Very few of the weights from Mohenjo-daro, considering the great number that have been found, were bad or fraudulent, a fact which shows that stringent trading regulations existed in the city.
c8.29
The few examples of scales used with the weights are of very ordinary pattern, and consist of a bronze bar with suspended copper pans. Some earthenware scale-pans may have been toys made by children. Much larger beams, of course, must have been used with the heavy weights, but these were, no doubt, made of wood. There is no evidence that the steelyard was known.
c8.30
One measure only has been discovered, and this merely a fragment, consisting of a slip of shell marked with divisions of 0·264 inch, of which nine remain. The mean error of graduation on this scale is only 0·003 inch, and two different marks on it indicate that it was once part of a longer measure on the decimal system. Probably, in its original state the complete scale consisted of many pieces joined together by metal bands, and the choice of shell for this purpose was an excellent one, for it could neither warp, expand, nor contract in varying temperatures. It is certain that the decimal system of notation was also practised outside India at this time, or even earlier, as it occurs on the Proto-Elamite tablets and on others from Jemdet Nasr in Mesopotamia.
c8.31
Certain pedestals, particularly well made and finished, are objects for which no particular use has as yet been found. These pedestals are usually hemispherical in shape, with a circular recess in the top, which is sometimes bored with holes; they are about six inches in diameter and three in height. The material used is either limestone, white or coloured, or alabaster, and some specimens are ornamented, one very fine one bearing an incised trefoil design which was once inlaid. As this pattern seems to have been sacred in the Indus valley, it may be inferred that the pedestals were designed to support a cult object which was, perhaps, of perishable material, as it has not survived.
c8.32
Whetstones are by no means so common in the Indus valley as they are, for example, in Sumer, and it is difficult to understand how the edges of tools and implements were kept in proper order. Possibly bricks were used for sharpening tools, for the few hones that have been found are not particularly well made and are mostly of slate, though sandstone specimens exist. The ends and edges of some of the whetstones suggest that they were more often employed to polish metal than to sharpen an edge.
c8.33
A pottery candlestick (Pl. O, 5) provides an answer to the question how the houses were lighted; no dish or other receptacle has yet been found which, by definite marks of burning at the edge, can be identified as a lamp. It seems certain that some vegetable oil must have been used for lamps in the Indus valley, since lamps were undoubtedly known in neighbouring countries at that time; in any case, it is extremely interesting to discover that candles were also in use at such an early date.
c8.34
It has already been mentioned that cotton was spun and woven at Mohenjo-daro and probably also at Harappa. The numerous spindle-whorls with rounded tops and flat, or slightly concave bases found in both cities show that the women, if not the men also, spent much of their spare time spinning the thread; the latter must assuredly have been cotton, for the majority of the spindle-whorls are too small and light to spin an elastic fibre like wool. Specimens have been unearthed made of pottery, shell, faience, or a vitreous paste, and when they are ornamented, which is rare, the patterns are extremely simple. Whorls are found with one hole, or with two or three holes for a split spindle, while some have a groove round the edge which was probably used for rubbing the thread.
c8.35
The shell that was so largely used to make inlay, beads, and ornaments of various kinds was of a variety (*Murex chicoreus ramosus*, Linn.) that is common in the Indian Ocean and the Persian Gulf. A very efficient ladle with a handle could be produced by smoothing the rough outside of a shell and removing the interior core, and these utensils were much used in the Indus valley and also in contemporary Sumer. They probably served a variety of purposes, and differed appreciably in size, ranging from under an inch to over nine inches in length. Simple shell dishes are not so common as the ladles; they may have been too valuable to be left behind when the city was abandoned.
c8.36
Shell shapes were obviously popular, for certain pottery objects have been unearthed which are either round or rectangular in shape, but with convolutions inside like those of a shell. These have been identified as cake-moulds, and are of a type that are still being used in India.
c8.37
Of especial interest are the rolling-pins of pottery or stone, expressly made for the little flat wafer-cakes that are even now so popular in India. There are pottery net-weights which are ring-shaped, very much like a small bracelet; similar rings have been found by Dr. Frankfort at Tell Asmar with the remains of a fishing-net still adhering to them. In date, the Mesopotamian net-weights are contemporary with those unearthed in the Indus valley.
c8.38
Certain thin pottery plaques, rectangular in shape with a perforated lug at one end, may have been intended for writing-tablets. They are of small size, ranging from four to seven inches in length, and were doubtless once covered with a smooth substance from which the writing could be washed, after the fashion of the wooden tablets still used in India.
c8.39
Several round ivory rods—the largest nearly eight inches long—which are graded down to a blunt point, are still unidentified. These rods, which are well polished, are simply decorated at either end with an incised pattern filled in with a black pigment, and it is thought that the smaller sizes, at any rate, may have been used to fasten cloaks. Ivory was certainly not so popular as shell, perhaps because it was more expensive to procure, though there is reason to believe that the elephant was a native of Sind some 4500 years ago. Shell was certainly very popular, but it was, of course, impossible to make articles of any considerable size from it.