The terracotta object varyingly described as ovoid (Rao, 1985: 19, 26, 27), Bicones (Fairservis, 1993: 109) circular biconvex terracotta ‘cakes’ (Lal, 1997: 227) ‘round ones with a deep finger impression in the center’ (Dikshit, 1993: 399), terracotta mushtis (Mehta, 1993: 168), mushtikas (Dales & Kenoyer, 1993: 490; Nath, 1998: 41) and most recently as ‘idli shaped terracotta cake with a thumb impression (Rao, 2006: 40,41) have been commonly found associated with the mature Harappan remains.  A small Mature Harappan period site like Allahdino yielded 24000  bicones (mushtikas) and 2600 triangular terracotta cakes (Fairservis, 1993: 109). The mushtikas and triangular cakes were in vogue even at rural sites like Zekda (Mehta, 1993: 168)  during the mature period. Similarly, the triangular terracotta cakes are also generally known as most commonly associated with Mature Harappan Culture (Allchin, 1993: 235). However, the emergence of the terracotta cakes have been reported from the ‘pre Indus phase at Mehergarh and Nausharo’ (Jarrige, 1995: 21) and early Harappan phase at Bhirrana (Rao, 2006: 40,41) and Kalibangan (Madhu Bala, 2003: 231, 233). The triangular terracotta cakes in particular were also obtained from Hakra Ware cultural assemblage at Bhirranna, where it was collected as unbaked specimen (Rao et al., 2005: 63). Besides the mature Harappan period both variety of cakes have been reported from other time and space contexts also, albeit in much lesser numbers. The mushtikas have also been reported from late Harappan context at Hulas (Dikshit, 1993: 399) and as surface finds from Moti Pipli, a Harappan affiliated Chalcolithic site (IAR, 1992-93: 19). The triangular terracotta cakes have also been found in later context as at Lothal (Rao, 1985: 15) and Dholavira (IAR, 1991-92: 30). Thus both type of cakes although have been found from a wide span of temporal settings yet their overwhelming preponderance is clearly seen in the Mature Harappan Period, when urbanization, industrial activities and long distance trade were very much in vogue. Even though after the breakdown of the grand system the Harappans held on to some of the practices, the utility of terracotta cakes had probably reduced and therefore the sharp decline in their numbers as observed during the

Association with rituals

The mushtikas and the triangular terracotta cakes have mostly been associated with rituals.   From the early sixties the terracotta cakes were supposed to have been used in the performance of ‘fire altar’ rituals during the mature Harappan period as at Kalibangan (IAR, 1962-63: 30). Later, the presence of terracotta cakes, ash and the cylindrical blocks in fire places were reckoned as the usual contents of ‘fire altars’ (IAR, 1968-69: 31). Sankalia (1974: 350) also mentions that ‘in the center of the pit was a cylindrical or rectangular (sundried or prefired bricks)’ and around this central stele of ‘fire altar’, ‘flat triangular or circular terracotta pieces, known hitherto as terracotta cakes’ were placed. Further, according to Rao (1979: 121 & 1985:15,24,26,27) these ovoid balls and triangular cakes ‘were used for ritualistic purposes’ and found in different types of ‘fire altars’. The triangular terracotta cakes and mushtikas were noticed as offering in ‘fire altars’ at Rakhigarhi (Nath, 1999: 48). Terracotta cakes have also been reported from Tarkhanewala Dera as part of a square ‘fire altar’ (Trivedi & Patnaik, 2004: 31). Thus, mushtikas and triangular cakes now have been reportedly associated with the phenomenon of ‘fire altars’ for well over four decades. Pertinently, besides its association with fire, triangular terracotta cakes were earlier reported to have ‘special significance in connection with ritual bathing or other ablutions’ by several scholars including Gordon (Allchin, 1993: 235).

Contexts of findings

The mushtikas and the triangular terracotta cakes have been found from a large number of contexts other than ‘fire altars’. Triangular terracotta cakes have been found at the mouth of kiln at Harappa (Dales & Kenoyer, 1993: 490). At Sanghol terracotta cakes have been found in kiln that yielded unbaked pottery (Sharma, 1993: 157). Cakes have been reported from potters kiln at Tarkhanewala in association with ash etc. (Trivedi & Patnaik, 2004: 31). Rao (1985: 24) reported ‘an altar like enclosure with terracotta triangular cakes and a stone quern’. Nath (1998: 41) is of the opinion that excessive concentration of terracotta cakes including mushtikas at Rakhigarhi is due to the craft activity. At Nausharo, clay built containers ‘had terracotta cakes used as heat conservers in the fireplaces’ (Jarrige, 1994: 288). At Rakhigarhi, a jar filled with terracotta cakes in the base portion is supposed as an hearth for heating semi precious stones at different stages of workmanship in a lapidary workshop (IAR, 1999-2000: 32).  Terracotta cakes along with small vases, charred bones and ashes were found within ‘burial urns’ by Tessitori at Kalibangan, during his survey of Rajputana between 1916-1919 (Thapar, 2003: 13). Another place wherein the triangular terracotta cakes occur as decorations on walls (Rao, 1979: 215). According to Nath, (1998: 43) the mushtikas were ‘prepared to keep them in cowdung cake fire pans as heat absorbents, thereafter it was reused either in floor bedding or raising levels’. Successive mushtika beddings in massive mud brick fortification at Rakhigarhi and mushtika bedding in cutting of a street at Kalibangan (Nath, 1998: 41) show the various type of less than sacrosanct contexts these cakes are found.

Triangular cakes have been found in ‘houses and streets’ (Mackay, 1938: 429), ‘passages’ (Rao, 1979: 113), ‘surface of lane’, ‘road’ (Sant et. al., 2005: 53), floors of mud brick houses in association with ovoid terracotta balls ‘plastered with mud’ (Rao, 1979: 83), ‘rooms paved with bricks or fired terracotta cakes’ (Agrawal, 2007: 79), as soling material along with mushtikas for raising levels of store houses (IAR, 1997-98: 57), etc. The triangular terracotta cakes have also been reported to be found from bathrooms, prompting scholars to suggest that these were used in ‘ritual bathing’ (Allchin, 1993: 235). Pertinently, Agrawal (2007: 143) also points out that ‘most houses or groups of houses had private bathing areas and latrines, as well as private wells. The early excavators at both Mohenjodaro and Harappa did not pay much attention to this essential feature’.  According to him, the recent HARP excavations at Harappa are finding what appear to be latrines in almost every house. Agrawal mentions that,  ‘these sump pot latrines were probably cleaned out quite regularly by a separate class of labourers’. Pertinently, had these large jars or sump pots sunk into the floors in or near bathing platforms been identified as commodes earlier, the scholars would not have correlated the presence of triangular terracotta cakes in bathrooms with ‘ritual bathing’. Earlier the triangular terracotta cakes presence in bathrooms was recognized but since the latrines contiguous to the bathing platforms were not commonly known its presence obviously in the vicinity of commodes could not have also been known as such. Had these facts been known then, none would have given hallowed status to the triangular terracotta cakes now understood to be found in the conjoined latrine bathrooms. The less than sublime presence of these cakes could not have been given any use other than just plain bathing. For, if anyone insistent on ‘ritual bathing’ even after the triangular terracotta cakes being known to have been found in the vicinity of the earlier unidentified commodes would have to associate defecation also with part of rituals. 


Pertinently, the presence of mushtikas and triangular cakes in a bewildering range of contexts does not allow it to be associated with ceremonies associated with fire, even though they are more often associated with places of fire. Even though cakes of food items are offered to the gods in the fire altars the mushtikas and triangular cakes also could be construed as something similar, which incidentally did not find mention in the Vedic literature. However, problem arises due to the fact that these cakes are found in not only mundane contexts of industrial activities but also in such places that defiles their once hallowed status. The finding of mushtikas and/or its common associate the triangular terracotta cakes in contexts, like: as soling material, as part of floor, in streets, passages, and bathrooms and obviously in the conjoined latrines does not enable it to achieve a sanctified status. In fact, its presence in those fire places which otherwise could have been considered as ‘fire altars’, prejudices one about its defiling presence at a sacrosanct spot. In fact, those fireplaces without these cakes could yet be fire altars. Pertinently, not all ‘fire altars’ have the mushtikas and triangular terracotta cakes within them. At Rakhigarhi several ‘fire altars’ (Nath, 1999: 48 & IAR, 1997-98: 60) have been identified wherein cakes have not been reported. One of these has burnt shells of fruits, which formed part of the offerings. At Lothal also several ‘fire altars’ have been identified by Rao (1979: 117) in which although ash, pottery and or bones are reported but the cakes were not mentioned. On the availability of other concurring evidence, these and others like these could be verily declared as ‘fire altars’. However, those fire places with both or either of the terracotta cakes, being used, as heat conservers are definitely not ‘fire altars’.

Thus, the fireplaces with mushtikas and triangular terracotta cakes therefore has to be the run of the mill, hearths, ovens, kilns, etc. Pertinently, cakes have been found in diverse contexts associated with heat, namely: at the mouth of a pottery kiln at Harappa (Dales & Kenoyer, 1993: 490), with some unbaked pottery in a kiln at Sanghol (Sharma, 1993: 157), in the pottery kiln at Tarkhanewala Dera (Trivedi & Patnaik, 2004: 31), besides in jar identified as hearth at Rakhigarhi (IAR 1999-2000: 32). The findings do hint that the cakes were used in places where prolonged heating was required. Nath (1998: 41) has suggested that the “excessive concentration of terracotta cakes including the mushtikas” indicate to the “intensive involvement of the people in their craft activity”. The cakes therefore appear to be primarily used as heat conservers as reported at Nausharo (Jarrige, 1993: 288) allowing “air into the kiln and at the same time effectively sealing in the heat” as suggested by Dales & Kenoyer (1993: 490) with reference to the triangular terracotta cakes found in the pottery kiln at Harappa. It appears that the frequent finding of pottery along with the terracotta cakes reinforces the possibility that some of the many types of ‘fire altars’ were potters kiln and other industrial fireplaces for baking different types of pottery and processing variety of craft items. Since the cakes did not have any religious value, it could and did end up in streets, floors, bathrooms, etc. Even the reporting of the cakes being found in ‘burial urn’ by Tessitori at Kalibangan does not gain it any religious value as mundane objects of daily use are routinely found along with burial remains. Pertinently, Jarrige (1995: 21) mentions the ‘fireplaces filled with stones or terracotta cakes of the pre-Indus period at Mehergarh and Nausharo and the Indus period at Nausharo’. By extension of logic, if the terracotta cakes are deemed as offerings then the stones also have to be of the same class. Alternatively, if the stones found in the fireplaces are not deemed holy the terracotta cakes also have to be deemed as mundane objects used in hearths and kilns. These cakes, therefore, are nothing else other than what Allchin (1993: 233-238) has termed it, namely, ‘Substitute Stones’.

Further, unlike mundane things, objects of religious value do have a high degree of standardization in mediums used, forms of expression and the association of other objects. These are also found only in limited areas not anywhere and everywhere. Sacred objects even if they have outlived there use would never be used in bathrooms and latrines, nor thrown away on lanes and roads or used as soling in floors to be trodden under the foot of men and animals. Earlier, Rao (1979: 215) had decried that ‘it is not safe to attribute cult value to an object on the basis of its shape or just because no other satisfactory explanation is available in the present state of our knowledge.’ Specifically citing the example of triangular terracotta cakes, he wrote that these ‘were once considered to be cult objects are found to have been used in flooring and for decorating the walls of the houses’. Again those fire altars with the central stele, also does not have any defining attribute regarding the shape or material of the stele or the enclosure nor the range of objects found within. Thus a clay stele in one altar, a mud brick in another, a baked brick in yet another, a cylindrical yashti here and a square one in the adjacent ‘fire altar’ does not show any uniformity, so necessary for outlining religious practices. It is intriguing that scholars did not find ‘the anything would do’ mindset as inferred by the permutations and combinations of material remains of the said ‘fire altars’ strange. Such adhoc substitutions and varieties are seen in industrial activities where the end product has to be achieved irrespective of the construction of the workplace. Where as in matters religious, symbolism rules the roost and uniformity in religious practices and materials are invariably sought for. Albeit, one should say in the same breath, that there is no denying of the fact that those ‘fire altars’ with central stele could be ‘fire altars’. However, such ‘fire altars’ then should have some formal attributes with regard to the construction and materials including the stele, across several similar ones at least in the same phase of the site.


Thus in view of the evidence obtained from many sites wherein the mushtikas and triangular terracotta cakes were used as ‘heat conservers’ besides the later degraded contexts of association of the terracotta cakes supposedly used in ‘rituals’, it appears to the present author that those ‘fire altars’ having these cakes cannot be ‘fire altars’. Moreover, even the association of the two type of cakes in ‘fire altars’ with central stele also does not complement the evidence of ‘fire altars’, as the casual approach in construction of the stele is itself not above circumspection. In all probability, the ‘fire altars’ having the mushtikas and triangular terracotta cakes, both observed in less than sublime conditions, even if it/they be associated with such fire places which, for other reasons appear as ‘fire altars’ including those with the central stele are not actually ‘fire altars’. In fact, they were fire places built up for different type of industrial uses. Thus, only those fire places which do not have these terracotta cakes and are having food offerings with a standard type of stele or without stele could be ‘fire altars’ all others are hearths, ovens, kilns, etc.


I am thankful to Shri K.K. Muhammed, and to Dr Narayan Vyas, both Superintending Archaeologists, for always encouraging academic work and for being available for discussions, whenever needed. Thanks are due towards Smt Hemlata Ukhale, Librarian, Bhopal Circle, for not only providing the required literature at the earliest but also suggesting more sources that has been frequently found useful. Last, but not least the miscellaneous technical help rendered by Shri Vijay Mishra is acknowledged herein.  


Agrawal, D.P. 2007. The Indus Civilization: An Interdisciplinary Perspective, Aryan Books International, New Delhi.

Allchin, B. 1993. Substitute Stones, in G. L. Possehl eds. Harappan Civilization: A Recent Perspective, Oxford & I.B.H. Publishing Co. Pvt. Ltd., New Delhi, pp 233-238.

Dales, G. F. and J. M. Kenoyer 1993. The Harappan Project 1986-1989: New Investigation at an ancient Indus City. in G. L. Possehl eds. Harappan Civilization: A Recent Perspective, Oxford & I.B.H. Publishing Co. Pvt. Ltd., New Delhi.

Dikshit, K. N. 1993. Hulas and the Late Harappan Complex in Western Uttar Pradesh, in G. L. Possehl eds. Harappan Civilization: A Recent Perspective, Oxford & I.B.H. Publishing Co. Pvt. Ltd., New Delhi.

Fairservis, W.A. 1993. Allahdino: An Excavation of a small Harappan Site, in G. L. Possehl eds. Harappan Civilization: A Recent Perspective, Oxford & I.B.H. Publishing Co. Pvt. Ltd., New Delhi.

IAR Indian Archaeology, 1962-63:  A Review, Annual Bulletin of Archaeological Survey of India, 30.

IAR Indian Archaeology, 1968-69: A Review, Annual Bulletin of Archaeological Survey of India. 31

IAR Indian Archaeology, 1991-92: A Review, Annual Bulletin of Archaeological Survey of India. 30.

IAR Indian Archaeology, 1992-93: A Review, Annual Bulletin of Archaeological Survey of India. 19.

IAR Indian Archaeology, 1997-98: A Review, Annual Bulletin of Archaeological Survey of India. 57,60.

IAR Indian Archaeology, 1999-2000: A Review, Annual Bulletin of Archaeological Survey of India. 32.

Jarrige, C. 1994. The Mature Indus Phase at Nausharo as seen from a block of period III, Proceedings of the Twelfth International Conference of the European Association of South Asian Archaeologists, I:288.

Jarrige,J.F. 1995. From Nausharo to Pirak: Continuity and change in the Kachi/Bolan Region from the 3rd to the 2nd Millennium B.C., Proceedings of the Thirteenth International Conference of the European Association of South Asian Archaeologists, I: 21.

Lal, B.B. 1997. The Earliest Civilization of South Asia, 227, Aryan Books International, Delhi.

Mackay, E.J.H. 1938. Further Excavations at Mohenjo-daro, Govt. of India Press, New Delhi.

Madhu Bala 2003. Minor Antiquities, in B.B. Lal, J.P. Joshi, B.K. Thapar, Madhu Bala. eds Excavations at Kalibangan: The Early Harappans(1960-69), pp 231-233. Archaeological Survey of India, Delhi.

Mehta, R.N. 1993. Some Rural Harappan Settlements in Gujarat, in G. L. Possehl eds. Harappan Civilization: A Recent Perspective, Oxford &  I.B.H. Publishing Co. Pvt. Ltd, New Delhi. 

Nath, A. 1997-1998. Rakhigarhi: A Harappan Metropolis in Sarasvati- Drishdavati Divide, Puratattva 28: 41,43.

Nath, A. 1998-1999. Further Excavations at Rakhigarhi, Puratattva 29: 48.

Rao, L.S., Nandini B. Sahu, Prabhas Sahu, Samir Diwan and U.A. Shastri. 2004-2005. New Light on the Excavation of Harappan Settlement at Bhirrana, Puratattva 35: 63,65.

Rao, L.S. 2006. Settlement Pattern of the Predecessors of the Early Harappans at Bhirrana, District Fatehbad, Haryana,  Man and Environment  XXXI no 2 : 41,42.

Rao, S.R. 1979. Lothal: A Harappan Port Town, 1955-62, Vol I: 83, 113, 117, 121, 215, Archaeological Survey of India, New Delhi. 

Rao, S.R. 1985. Lothal, Archaeological Survey of India, Delhi.

Sankalia, H.D. 1974. The Prehistory and Proto history of India and Pakistan, Deccan College Postgraduate Research Institute, Poona.

Sant, U., T.J. Baidya, N.G. Nikoshey, N.K. Sinha, S. Nayan, J.K. Tiwari and A. Arif. 2004-2005. Baror - A New Harappan Site in Ghaggar Valley. A Preliminary Report,  Purattatva 35:53. 

Sharma, Y. D. 1993. Harappan Complex on the Sutlej ( India) in G. L. Possehl eds Harappan Civilization: A Recent Perspective,  Oxford & I.B.H. Publishing Co. Pvt. Ltd,  New Delhi.

Thapar, B.K. 2003. Discovery and Previous Work, in B.B. Lal, J.P.Joshi, B.K. Thapar, Madhu Bala eds. Excavations at Kalibangan: The Early Harappans (1960-69), Archaeological Survey of India, Delhi.

Trivedi, P.K. and J.K. Patnaik. 2003-2004. Tarkhanewala Dera and Chak 86 (2003-2004), Puratattva 34: 31.