An Encyclopaedia of Indian Archaeology v1.c19.s15: 14.15 Wells and Ring-wells
An Encyclopaedia of Indian Archaeology v1.c19.s15
- Language
- English
14.15 Wells and Ring-wells
v1.c19.s15.1
Wells with or without lining of brick, stone or terracotta rings have been uncovered in excavations at sites in almost the entire subcontinent showing characteristic features and types in terms of their spatial and temporal distribution. While those with or without brick or stone lining were used as draw-wells for water, the ring-wells, so called because of the lining with terracotta rings, are generally not considered to be draw-wells (see below).
v1.c19.s15.2
The earliest example of wells are of the Harappa culture [4.9] as evidenced at Mohenjo-daro, Harappa and other sites in Pakistan, and KALIBANGAN, LOTHAL and other sites in India, were brick-lined wells have been found in the habitation areas of these sites. An interesting fact about the Harappa wells is their continuous use, barring exceptions, practically throughout the long period of the existence of the settlements, as is evidenced by the raising of the lining of the wells from time to time or by the construction of a short staircase to lead down into the well-room consequent upon the raising of the level of the mound (Mackay, 1938, p. 165). At Mohenjo-daro most buildings of any size have wells of their own (Marshall, 1931, p. 16) besides wells for public use provided in private houses with an entrance to the well-chamber direct from the street or ‘placed in a cul-de-sac between two houses’ (*ibid.*, pp. 16, 270).
v1.c19.s15.3
At Harappa only six wells situated at long distances from each other have been found; they are all public wells (Vats, 1940, pp. 13-14, 74). One of them is on the terrace of a house beside a lane and was perhaps available for use to outsiders; another one has three large jars placed beside it and was probably meant for the wayfarers.
v1.c19.s15.4
The wells at Mohenjo-daro vary in diameter from 33 cm to 2·26 m, the average diameter being between about 91 cm, and 1·22 m, while the wells at Harappa range in diameter from 55 cm to 2·13 m. According to Vats, 1940, p. 14, at Harappa the two wells situated far away from the river Ravi are much larger in diameter than those in Mound F whose diameter is so small—56 to 84 cm—that ‘two people cannot draw water from them at one and the same time.’ It has therefore been inferred that ‘people living far from the riverside had to depend on well-water for all and sundry purposes’ for which wells with a larger diameter which could be simultaneously used were a necessity. The well in Chamber 16 e. of the Great Bath at Mohenjo-daro (Marshall, 1931, pp. 26, 139, 142) is 1·87 m in internal diameter and is surrounded by another circular brick wall which is separated from the coping of the well by a distance of 33 cm and is provided around the well to strengthen its coping.
v1.c19.s15.5
At Chanhu-daro only two wells have been found, one in the trial excavation of Majumdar, 1934, p. 36, and another in the subsequent excavation of Mackay, 1943, p. 17, their diameters being 1·07 m and 86 cm respectively.
v1.c19.s15.6
In contrast to Mohenjo-daro, the number of wells at KALIBANGAN is very small. They have been located in both the Citadel and the Lower City where there is often a well in the courtyard of a house. One of the wells which, according to excavators, is ‘associated with a fire altar . . . showed novel structural features: in the foundation, the ring-wall of the well constructed of wedgeshaped baked bricks, was found to be supported on four sides by (normal type) radial walls of single-brick course' (*IAR 1966-7*, p. 32). However from the published photograph (*ibid.*, pl. XXIII) it appears that the radial walls are not of 'single-brick course' but of single-brick thickness.
v1.c19.s15.7
At LOTHAL, according to the excavator, 'sometimes a private well and more than one bath-room were provided for a large wealthy family' (Rao, 1973, p. 111). However the published town plan and plan of Block B in the Acropolis at LOTHAL (*ibid.*, pp. 63, 65, figs. 12, 14) show only a brick-built well in Block B in the Acropolis, located at the corner of the main drain of the Acropolis, and does not form part of a house. The well is about 2 m in internal diameter as seen from the published drawing.
v1.c19.s15.8
It is quite likely that the Harappans mainly used water from the adjacent rivers as is evident from the limited number of wells at sites like HARAPPA, KALIBANGAN and LOTHAL and their total absence at sites like SURKOTADA (J.P. Joshi, personal communication).
v1.c19.s15.9
Most, if not all, Harappa wells are circular except one or two oval examples (Marshall, 1931, p. 150; Mackay, 1938, p. 166). Mould-made wedge-shaped brick of different sizes is used in the construction of these wells (Marshall, 1931, p. 270; Mackay, 1943, p. 17; *IAR 1966-7*, p. 32); some of them are constructed of ordinary-shaped brick used conjointly with wedge-shaped one (Marshall, 1931, p. 270). Alladino has a series of three stone wells aligned with one another (Shaffer, 1979, p. 25). At Harappa however except in the inner ring of one well wedge-shaped brick is not used (Vats, 1940, p. 14). In some cases, as at Mohenjo-daro (Marshall, 1931, p. 270), the wider ends of the brick in the steening of some wells are carefully curved so as to present a smooth surface on the outside of the well, while in one case the narrower end of the brick is curved. In a few examples the coping of the well is slightly projecting above the floor (Marshall, 1931, p. 270; Mackay, 1938, p. 165) and a brick edging around it provided to prevent waste water from re-entering the wells. In one of the wells at Mohenjo-daro the coping is raised to a height of 40 cm above the floor (Mackay, 1938, p. 154).
v1.c19.s15.10
It is difficult to say whether windlass was used for lifting water from the wells. In some private wells at Mohenjo-daro there are no marks caused by friction of ropes suggesting thereby either the use of some form of windlass which, if any were used, might have been made of wood (Marshall, 1931, p. 270) and for only the larger wells. Some of the public wells at Mohenjo-daro show marks caused by friction of ropes on the coping (Mackay, 1938, p. 113). In the smaller private wells water was probably lifted by standing straddle-legged over the well. Sometimes a channel or pavement is provided to drain away the drippings from the pots (Mackay, 1938, pp. 113-4, 154; Vats, 1940, p. 61).
v1.c19.s15.11
Wells or ring-wells have so far not been reported from any site belonging to the Chalcolithic cultures [4.14], whether late contemporary with or posterior to the Harappa culture. Nor is there any evidence about the presence of wells or ring-wells at sites of the early Iron Age cultures [4.16] of the last quarter of the 2nd millennium and early half of the 1st millennium B.C.
v1.c19.s15.12
After this interval wells are reported from sites of the middle of the 1st millennium B.C., almost coeval with the emergence of the NBPW [4.18.1]. Significantly enough ring-wells also make their appearance around this period (Pande, 1966, pp. 207-19). At most sites of the early historical period wells are generally lined with baked brick. There are, however, also instances of stone-lined wells, particularly in the n.-w., e.g. at Sirkap and Kalawan, a large Buddhist establishment at Taxila (Marshall, 1951, pp. 116, 322). At Sirkap in the main entrance hall there are remains of two small wells constructed of rubble stone. One of these, circular on plan, is 'partly engaged in the eastern wall' and is earlier than the other one, square on plan and close to the former. However Marshall regards both to be refuse wells like the wells 'unearthed in the Sirkap and Bhir Mound cities, where every drop of water, other than rain-water, had to be fetched either from Tamranala stream or from wells in the low-lying ground outside the walls.' However he has not recorded details about the 'wells in the low-lying ground outside the walls'. The two wells have been dated by him to the Śaka period, i.e., about the middle of the 1st century B.C. when, according to him, the stone fortifications of Sirkap were built. In one of the chambers in Block D of Sirkap has been found a well, the top of which is square and is constructed of stone while the portion below is circular and faced with lime and river sand (Marshall, in *ASI-AR 1914-5*, p. 18). The well at Kalawan is located about 140 m from the foot of the hill, 3·5 m in depth with a stone lining about 91 cm at the mouth (Marshall, 1951, p. 322). At Zaro Dheri or Zar-ki-Dheri, about 21 km from Mansehra, a circular pit, about 1·98 m in diameter, has been found to the n. of the monastic buildings; it is made of large diaper masonry lining and is in all likelihood a well (*ASI-AR 1922-3*, p. 96).
v1.c19.s15.13
While the wells datable from the latter half of the 2nd millennium B.C. are generally circular on plan with lining of wedge-shaped brick, there are also examples of circular wells lined with terracotta rings in the lower portions and brick on top. Similarly there are instances of circular wells with lining of terracotta rings below and square tops paved with brick. Wells of either variety are ubiquitous in this period right up to the middle of the 1st millennium A.D. or even later at all sites particularly in the plains and are generally found as parts of building complexes such as monasteries, houses, etc.
v1.c19.s15.14
Among the early historical sites with brick-lined wells are: UJJAJN, NBPW levels [11.16], Pd II, datable between *c.* 200 B.C. and the beginning of the Christian era; RAJGHAT⁴, where one of the wells was raised at a later period by the addition of terracotta rings and seven brick courses, another had a brick pavement, and particularly a third one, of the 4th-5th century remarkable for the thickness of its wall, the inner diameter 2·35 and the outer 7·76 m and for its lime-plastered surface; PRAHLADPUR, NBPW levels; SAHETH-MAHETH, late Kushan levels, with an internal diameter of 2·28 m, 2·14 m broad and 46 cm high parapet; KUŚĪNAGARA, many examples, one of them square and surmounted by a low circular platform 3·12 m and another circular 1·73 m in diameter with a well-head 2·60 m in diameter; KUMRAHAR, Gupta period (Spooner, in *ASI-AR, 1912-3*; Altekar and Mishra, 1959, p. 31); BHITA, where each house has a well, datable between the 1st and 5th centuries A.D. one of them composed of wedge-shaped brick and terracotta rings respectively on top and lower portions and others of wedge-shaped brick; Brahmanabad in Sind (Cousens, 1929, p. 50), with two wells made of 'curved moulded bricks' having a diameter of about 2·44 m; VAIŚĀLĪ (*ASI-AR 1903-4*, p. 92), three square masonry wells adjoining each other of the Gupta period, the smallest one having two holes in each of its eastern sides which seem to have held two wooden beams laid across the well at about half its depth; SAMBHAR, where a well with a steening wall was sunk in one of the rooms of a house; NASIK, brick-lined wells datable between 300 and 200 B.C.; NĀLANDĀ, where amongst the many wells one excavated in a corner of the court of Monastery 6 is octagonal on the upper part and circular below; PEDDABANKUR, Megalithic [4.17.8], circular wells lined with wedge-shaped brick; VADGAON-MADHAVPUR, a square brick-lined well; and TRIPURĪ, where a large brick well with brick-on-edge flooring around its periphery was excavated outside a *vihāra*.
v1.c19.s15.15
Ring-wells, so called because of the lining with terracotta rings and variously identified as soak-pits, draw-wells or granaries, show a vast area of distribution extending from Shahbazgarhi, (*ASI-AR 1921-2*, pp. 65-6) in the n. of ARIKAMEDU in the s. and Brahmanabad in the w. to Mahasthangarh in the e. The maximum occurrence of ring-wells is in the Ganga-Yamuna doab with a lesser concentration in the Taxila region in the n. and in the w. Deccan. Their presence in the rest of the area in c. India and the e. coast is sporadic. The earliest examples do not go back to a period earlier than *c.* 5th-4th centuries B.C. and are almost coeval with the NBPW. Their use continued in subsequent periods and the tradition lingers on till today in some parts of the country (Pande, 1966, pp. 207-19).
v1.c19.s15.16
The most characteristic feature of the ring-wells is the use of terracotta rings inside a circular shaft to forestall the collapse of the sides. For the construction of a ring-well a large pit having a much bigger diameter than the rings to be placed was dug and the rings were then piled inside it and fitted into each other; the space in between was filled with rubble and compact earth and carefully rammed near the opening at the ground level (Dikshit, 1955, p. 23; *IAR 1970-1*, p. 8). While ring-wells are generally found singly, there are instances also of twin ring-wells, as at TRIPURĪ and NASIK, besides clusters of them as at RAIRH where these are present in a large number within a considerably smaller area, and at NAGARI. Similarly, while single-ring casing is the general rule, at ARIKAMEDU (Casal, 1949, Annexe III) double-ring casing can be noticed; in some examples at Taxila (Marshall, 1951, 1, p. 94) and RAIRH pots have been placed upside down with their bases perforated and telescoped into each other.
v1.c19.s15.17
Generally the terracotta rings used in the ring-wells have a uniform diameter from bottom to top varying between 46 cm to 1·37 m and can be classified into two diameter-groups: (i) with diameter ranging between 46 and 76 cm, and (ii) with diameter over 76 cm. The variation in diameter can be noticed *vis-à-vis* their spatial distribution, and on the basis of their spahe the rings fall into four broad categories: (i) rings with mouth splayed in a way as to fit into the lower edge of the corresponding upper ring; (ii) rings with wide and flat top and bottom; (iii) rings with grooves at both the edges for fixing the upper ring into the lower one; and (iv) rings with a narrow top and broad bottom. The shape of the rings thus obviated the use of any mortar or binding medium. Besides complete rings there are also examples of rings made of three segments as at ROPAR. The height of the rings is generally between 10 and 22 cm, while at ARIKAMEDU (Casal, 1949) they are exceptionally large ranging from about 22 cm to 1·04 m. Similarly there is also a variation in the thickness of the individual rings. Mention is also made to ornate rings at VAIŚĀLĪ (Krishna Deva and Misra, 1961, p. 11).
v1.c19.s15.18
Apart from ring-wells composed entirely of terracotta rings from bottom to top, there are also examples of ring-wells lined with wedge-shaped brick and terracotta rings respectively on the top and lower portions at BHITA or wells cut into hard marl and rivetted with pottery rings only on the lower portion at Charsada (Wheeler, 1962, pp. 28-9).
v1.c19.s15.19
The present evidence shows ring-wells reaching to a depth varying between 31 cm to 7 m, the number of rings ranging between 2 and 47; in some cases the bottom of the wells could not be reached, as at HASTINĀPURA. Not all ring-wells are excavated sufficiently deep into the natural soil and the bottom of some is within the cultural deposits e.g. at PRAKASH. At HASTINĀPURA, NEVASA, Brahmanabad (Cousens. 1929, p. 51), RAIRH, BANGARH, etc., the wells have been dug to a sufficient depth into the natural soil.
v1.c19.s15.20
Some of the ring-wells have well-heads in the form of either a parapet wall or a stone slab placed over the mouth of the well or domical covers of terracotta. At HASTINĀPURA, BHITA and BANGARH the upper courses of the ring-well are made of wedge-shaped brick which probably formed part of the well-head (Pande, 1966, p. 214). At SISUPALGARH and BHITA a square stone is placed on the mouth of the ring-well having an aperture of the same diameter as that of the ring-well. Rope-marks are also noticed on the inner edge of the stone cover of the ring-well at SISUPALGARH. The solitary example of a detachable domical lid of terracotta placed over the mouth of a ring-well is from TRIPURĪ.
v1.c19.s15.21
Evidence of the association of ring-wells with structures such as brick platforms or pavements is available from KUMRAHAR, ARIKAMEDU, BANGARH, NEVASA, BESNAGAR, RAIRH, Mahasthangarh, Paharpur (*ASI-AR 1922-3*, p. 118) etc.; at NASIK stones are dumperd close to the ring-well which, according to the excavators, the occupants might have deposited ‘as an embankment against the rings or as an approach to the wells.’
v1.c19.s15.22
At AWRA (Trivedi, 1962, p. 19) a drain made of terracotta pipes joins the top of a ring-well, while at RAIRH pottery drains are found in the vicinity of the ring-wells. At the latter site brick wells are found around a cluster of ring-wells. At HASTINĀPURA, RAIRH and Taxila (Marshall, 1951, pp. 94-5) large-sized jars with perforated bases and telescoped into one another have been found close to the ring-wells.
v1.c19.s15.23
Most of the ring-wells seem to be located inside houses. At Bhir Mound (Marshall, 1951) each house has at least three such wells—one each in the courtyard, bath-room and the kitchen. At BESNAGAR, Brahmanabad (Cousens, 1929, p. 51) and UJJAIN (Garde, 1940, p. 18) ring-wells are generally found inside the house. At RAIRH, NASIK, ŚRĀVASTĪ, TRIPURĪ, AWRA and NEVASA ring-wells are located near the houses. Kitchen wells fitted with rings have been reported in different parts of Dt Bergaum (*ASI-AR 1914-5*, p. 76).
v1.c19.s15.24
Because of their constructional features it is difficult to identify the purpose of the ring-wells. But those that are too shallow or have thin bottom within habitational deposits are perhaps soakage pits. However,, on the basis of the well-heads, location inside rooms, pavements around them, use-marks in some, *e.g.* at BANGARH and SISUPALGARH, and their being excavated sufficiently deep into the natural soil, some of them could have served the function of draw-wells. Such wells for drawing water are still common in certain parts of the country.
v1.c19.s15.25
Altekar, A.S., & Mishra, V., 1959, *Rep. on Kumrahar Excavations 1951-5*, Patna. Grade. M.R., 1940, in *An. Administration Red. of the Arch. Dept., Gwalior State, for Vikram Samvat 1995, year 1938-9*, Gwalior, Casal, J.-M., 1949, *Fouilles de Virampatnam-Arikamedu*, Paris. Cousens, H., 1929, *Antiquities of Sind with Historical Outline*, ASI, New Imperial Series, 46, Calcutta. Dikshit, M.G., 1955, *Tripuri-1950*, Sagar. Deva, K., & Mishra, V., 1961, *Vaiśālī Excavations: 1950*, Vaishali. E.J.H., Mackay, 1938, *Further Exacavations at Mohenjo-daro*, Delhi; 1943, *Chanhu-daro Excavations 1935-6*, New Haven. Majumdar, N.G., 1934, *Explorations in Sind*, Mem. ASI, 48, Delhi Marshall, J., 1931, *Mohenjo-daro and the Indus Civilization*, London, 1951; *Taxila*, Cambridge. Page, J.A., 1930, in *ASI-AR 1926-7*, Calcutta. Pande, B.M., 1966, in *Bull. Deccan College Res. Inst.*, 25 (for ring-wells in general). Rao S.R., 1973, *Lothal and the Indus Civilization*, Bombay Shaffer, J.G., 1979, in Agrawal, D.P., & Chakrabarti, D.K., ed., *Essays in Indian Protohistory*, Delhi. Trivedi, H.V., 1962, in *Jour. of Madhya Pradesh Itihasa Parishad*, 4, Vats, M.S., 1940, *Excavations at Harappa*, Delhi. Wheeler, M., 1962, *Charsada—a Metropolis of the North-west Frontier*, London. For other sites see respective entries in vol. II.—B.M. PANDE