There are more than 1,100 rock-cut caves across India, which once played an important role in Buddhist, Jain and Hindu religious traditions. In this essay, Dr Phyllis Granoff , Professor of Religious Studies at Yale University, provides a nuanced perspective of India’s religious traditions and makes a subtle distinction between ‘Caves’ and ‘Cave Temples’
In one of his commentaries, the 6th-century polymath Buddhaghosa drew a clear distinction between a cave, which was an unsuitable place for the Buddha to be, and an elaborately carved and adorned rock-cut temple, which he seems to have thought was a more fitting setting for the Buddha’s august presence. Buddhaghosa wrote commentaries on all the suttas written in Pali, and the passage in question comes from his commentary to the Sakkapanna Sutta, ‘The Questions of Sakka’, in the Dighanikaya.
In this text, Sakka, the king of the gods, seeks out the Buddha, who is meditating in the Indasala cave outside the city of Rajagriha. He takes with him the celestial musician Panchashikha, who carries a harp, and whose presence is often the key to identifying the scene in its many representations in early Buddhist art.
In Gandhara, the Indasala cave is represented as unadorned and its status as a natural cave, remote from human habitation, is often indicated by a rocky landscape and the presence of wild animals among the rocks and the gods and sages. The scene seems somewhat domesticated in a relief from Nagarjunakonda by the presence of an elaborate throne on which the Buddha sits, but the location remains a bare cave, indicated by the rocky outcrop above the seated Buddha.
At Sanchi, the focus of the attention is not the cave, but Indra and his retinue, who occupy two-thirds of the picture space. The “cave” is set in a rocky landscape and wild animals and fantastic beasts peek out from the rocks at either side of the “cave”, which is clearly differentiated from the natural rock around it.
In this relief, the structure is not a natural cave at all, but is closer to what we might call rock-cut architecture. The stone has been smoothed and a decorative chaitya arch cut into the facade, reminiscent of the well-known Lomas Rishi cave in the Barabar hills, Bihar. Another early relief from Bodhgaya shows a similar small structure with an inverted-U-shaped door; the rocks below indicate that the structure has been built into a mountain.
It is possible that the artists of this scene at Sanchi and Bodhgaya were simply following the practice of many artists of the period, who seem to take delight in reproducing examples of contemporary architecture. I would like to suggest, however, that the choice to depict the Indasala cave as a man-made structure may also reflect certain beliefs about caves and their uses.
There are hints even within the Sakkapanna Sutta itself that a natural cave was not considered to be the ideal setting for the presence of the Buddha. When Indra finally summons up the courage to approach the Buddha and enters the cave, something marvellous happens. The cave is transformed; indeed, the features that are most cave-like about it are eradicated. The uneven floor of the cave becomes levelled and the cramped space opens up. The natural darkness of the cave is dispelled and the newly formed expanse is brightly illuminated. All of this, the sutta tells us, happens through the magical power of the gods.
Buddhaghosa in his comments on the Indasala cave transforms it even more than the gods had done. He tells us that originally the Indasala cave (guha) was a natural formation in between two mountains. There was an Indasala tree in front of the opening to the cave, which gave the cave its name. But this is not where Sakka found the Buddha. A donor turned this natural cave into a rock-cut temple, a lena, by having doors and windows made, having subsidiary structures added all around it, and having it whitewashed and lavishly adorned with sculpted and painted garlands and creepers.
The donor then gave this lena to the Buddha for his use, and it was here that Sakka encountered the Buddha. In fact, this description corresponds closely to what we find in the caves at Mihintale, Sri Lanka, which are roughly contemporary to Buddhaghosa’s comments. The natural rock formations have been carefully outfitted with doors and windows and traces of elegant paintings remain on the interior walls.
The use of the term ‘lena’ as opposed to ‘guha’ in Buddhaghosa’s brief account is telling. An examination of inscriptions from cave temples reveals that these structures were not called “caves”, or guha, but were most often called either lena in Prakrit, or layana, the Sanskrit equivalent. They could also be called houses or dwelling places (ghara, griha, veshma, bhavana). What these terms all share is that they refer to man-made and not naturally occurring structures.
I will argue that caves, with few exceptions, were either considered uncanny and frightening places, or were entrances to fantastic other worlds. In either case, they were not suitable to be temples in which the god dwelt, where images were donated and where people could come for worship.
Natural caves often figure in religious texts as places for solitary meditation or the practice of certain rituals, which we often call by the name of esoteric or tantric rituals. They were suited to these purposes because of their remoteness from civilization; the fearsome setting of the cave corresponded in many cases to the inherently dangerous nature of the rituals or the need in meditative practice to encounter and transcend terrors, external and internal.
In a very different vein, caves in poetry and stories have marvellous and enticing features, and can well be described as elaborately constructed edifices rather than natural caves. But they are still not temples. They are, for example, fantastic cities built by demon architects, who magically outfit them with palaces and gardens. These caves on the surface may seem to have more in common with Buddhaghosa’s lena than with the original Indasala cave or guha that the anonymous donor so greatly altered, but this essay will propose that they are still in nature and function different from rock-cut temples.
When we look at some of the inscriptions in rock-cut temples or the textual descriptions of lenas, the aptness of Buddhaghosa’s comments cannot fail to strike us. The long inscription outside “cave” 16 at Ajanta describes the structure as a veshma, a dwelling place, adorned with windows, balconies, handsome corridors, turrets, and images of heavenly damsels. It is also adorned with beautiful pillars. These are some of the very elements that Buddhaghosa had singled out as features of the lena that the donor had built. Somewhat later in time, they would be incorporated into the definition of a layana in an architectural treatise attributed to King Bhoja in the 10th century, the Samaranganasutradhara.
Caves: Real and Imagined
Among the most frequent references to caves in religious texts is their appearance in a varied list of places where a practitioner is either to meditate or perform a particular ritual. The texts specify that all of the places should be deserted, without anyone there. Thus the Buddhist Manjushrimulakalpa tells us that some of its spells are best accomplished in places like mountains, particularly mountain caves.
The Sadhanamala, a collection of diverse Buddhist rituals, in its opening instructions allows its rituals to be performed almost anywhere; the initial list includes mountains, forests, caves, houses, gardens and rock-cut temples, as long as they are deserted. Once the specific instructions for rituals are given, however, deserted mountain caves seem often to be preferred.
We are familiar with meditation caves from Central Asia; in his detailed instructions for meditators, Buddhaghosa, writing from Sri Lanka, recommended that some monks meditate in remote mountain caves. A cave was also the location for the first conclave of Buddhist monks. According to the legendary account in the Mahavastu, some monks became so despondent at the death of the Buddha that they wanted to kill themselves.
Kashyapa explained to them that they had a responsibility to preserve the Buddha’s teachings. He told them to gather in a cave called the Saptaparna cave in Rajagriha near Bodhgaya, to recite the teachings together. Nothing is said about the cave itself, but the mountain is described as forested with various kinds of trees, and seems to have been in an isolated location, very much like the caves for meditators or those caves that ritual specialists would later recommend.
The Mahavastu tells another story about a cave that illustrates both the potentially dangerous aspects of caves and their possibly marvellous powers. One member of the Buddha’s clan, the Shakyas, has a daughter afflicted with leprosy. Physicians are summoned and they try everything, but to no avail. Her entire body is an oozing and festering wound. Just to look at her makes people feel disgusted.
Her brothers decide to abandon her and so they load her into a cart and take her into the Himalaya. There, in the foothills, they excavate a cave (guha) and provision it with food and drink and bedding. They then wall her up in the cave, the existence of which is disguised by the heap of refuse that they place in front of its entrance.
The cave acts like the kind of sweat-chamber described in Ayurvedic texts, and the girl is cured from the still air and the heat. She becomes exceedingly beautiful. A tiger happens by and smells the scent of a human. With his paws he scrapes away the refuse in front of the cave. Eventually a sage finds the cave and the girl, and fathers children with her. This is the origin of the branch of the Shakyas known as the Koliyas.
The cave in this story is in the wilderness and wild beasts are an ever-present danger; but it also has healing properties and reminds us of the many caves in special mountains, which in Sanskrit literature are said to have healing herbs. There is also a common association of caves with alchemical treasures. I will return to these points below.
Caves: Rama & Lakshmana
In the story of the ancestor of the Koliyas, I would argue that the healing properties are not really of the cave itself; the cave is a frightening place, totally removed from human habitation, discovered by a man-eating beast despite efforts to conceal it. In Sanskrit literature, caves are often frightening places. In the Ramayana, Lakshmana tells his brother Rama that caves are horrifying. They are also difficult to get to and protected from public access.
Rama himself had asked Lakshmana to hide Sita in a cave that was durga, difficult for anyone to get to, and hidden by an overgrowth of forest. Such a place would hardly be a fit place for a lavish temple in which worship was to be carried out on a regular basis, as we are told was the case with the rock-cut temple made by the vidyadharas in Karakanda’s story reviewed earlier.
In fact, even the most marvellous of caves in Sanskrit literature are not described as temples. Rather, they are magical mini-worlds. One of the best known is in the Ramayana, the cave-world of Kishkindha, where Rama’s monkey allies live.
This cave is made of jewels and filled with flowering trees. It is strewn with jewels and crowded with palaces. There are marketplaces and more trees that provide everything that the residents need. The fragrance of sandalwood and aloes mingles with the sweet scent of lotuses.
The streets are broad and there is plenty of wine and mead for everyone. The luxurious homes have many floors and the splendid mansions of the wealthy monkeys line the main road. The facades of the houses are whitewashed, making them look like so many white clouds. They are adorned with divine garlands and filled with money and grain. The royal palace has seven sections and a well-guarded harem. It is equipped with furniture made of gold and silver.20 The description of the wonders of this cave-world goes on. This cave is an alternate world; it has everything that our world has and more, but it is not a temple.
Such caves are often associated with demonic creators. Elsewhere in the Ramayana, the monkey Sugriva’s cave, probably the same cave as the cave-world of Kishkindha, is said to have once belonged to the demon or daitya Maya; it is dark and frighteningly large, but inside is the golden palace of Maya.21 The demon or rakshasa Kumbhakarna also lives in a magical cave, which has a sweet fragrance and golden palaces inside it. The cave has pillars of crystal and gold and gates of ivory. It glitters with cat’s-eye gems and is like one of the marvellous caves of Mount Meru.
There is something uncanny about all of these caves, which are not incidentally associated with demons. In a Buddhist text, the Gunakarandavyuha, the Bodhisatva Avalokiteshvara preaches to the demons in a cave called the Vajrakukshi cave. These caves may share an abundance of jewels and precious substances with the layanas or rock-cut temples described in the inscriptions and Jain texts, making them all very much like the caves on Mount Mandara, the abode of the gods, but they are nonetheless very different from the rock-cut temples. They are hidden wonder-worlds with palaces and bazaars, but they are not temples in which to worship. The same is true of the next cave, which lacks the architectural splendours of the Ramayana caves, but is also the gateway to another world.
In this essay, I have argued that the concept of a ‘cave temple’, one of the most frequently used terms to describe rock-cut temples, combines two different types of structures that were not combined in Indian religious or secular literature. By the early medieval period, temples are carefully constructed structures, with refinements that may include painting, sculpted ornamentation, latticed windows, stairs, turrets – in fact, all the varied architectural elements that beautified secular palaces.
Temples or prasadas are, after all, dwelling places for the gods or the Buddha or the Jina, and as such are to be richly appointed. Caves, on the other hand, are uncanny natural structures, that may be either fantastic by their very nature, leading to other worlds, or frightening in their lack of refinement and their distance from any civilizing influence. Combining the two mixes what I believe were very separate topoi with very different associations.
Excerpted from the essay "What’s in a Name? Rethinking Caves" by Phyllis Granoff, published in Living Rock: Buddhist, Hindu and Jain Cave Temples in the Western Deccan (June 2013). Republished in Readings on the Temple from 75 Years of Marg, curated by Naman P. Ahuja, Marg, Volume 73 Numbers 2 and 3, December 2021-March 2022. Find the publication at bit.ly/ReadingsTemple.
If you enjoyed this article, you will love LHI Circle - your Digital Gateway to the Best of India's history and heritage. You can enjoy our virtual tours to the must-see sites across India, meet leading historians and best-selling authors, and enjoy tours of the top museums across the world. Join LHI Circle here.