Monday, 26 November 2018

Visual storytelling in Khmer reliefs

ICONOGRAPHY OF IMAGES No.22                                                                                        

Visual storytelling in Khmer reliefs

By Vittorio Roveda

(Text and pictures copyright)

Murals are illustrated in most scholarly books rarely explaining their position in the art history of ancient Cambodia’s. I have tried to put some order amongst narrative murals of various temples, to look at the beginning, development and evolution of visual storytelling.

After almost 25 years spent studying Khmer iconography, (and publishing 7 books) I think it is the time for me to summarize my idea about the emergence of visual narrative or visual storytelling and its evolution of images in Khmer temple art. I did tentatively expressed my theory of narrative evolution in my book Images of the Gods

(2005: 46). Several decades have passed since art historians devoted their talents to the Art History of Khmer Culture: Jean Boisselier,  Philippe Stern and Madeleine Giteau were perhaps the last of most illustrious scholars; since their time, temple conservation and restoration (and now extraordinary archaeological discoveries)  became more important while  epigraphists transformed themselves in art historians with the publications of various articles in specialised periodical papers (out of reach to common people) or indirectly to lush guidebooks. Scholars of Khmerology at large have forgotten that iconography has been the basis for the understanding the structure of Khmer Art. Narratology looks at the organization of visual events (images) in cooperation with the date of temple’s making.

I must clarify what I intend for visual storytelling, or visual narrative: the visual telling of a story which is an open ended event, with an indefinite beginning and an indefinite end, unrolling in time (past, present, future) and making sense. It is assumed that the viewer takes an active and creative role in the dialogue with the artwork, to re-invent the experience that the images communicate to him/her.

Whereas Iconography is basically a descriptive process of identification of images, narratology is an analysis of the organisation if images, indicating ‘how’ the images of a story are combined to narrate as opposed to ‘what’ is narrated, following through the system of narratology[1]. Traditional iconographic interpretation can reduce the image to a question of treatment (style) that confuses the meaning. The power of narrating is an integral part of visual art.

The invention of narrative was initially motivated by the need to communicate with the divine, but soon it assumed a variety of aspects. Visual narratives could be used to express symbolic meanings, to visualise allegories. This may have been at the root of the creative process when the religious-learned people needed to express metaphysical and religious concepts. The priests and the monks wanted to manifest the gods, to make them active in their own lives in order to achieve their own goals. In Khmer culture, the religious community that exercised a cultural monopoly increased their stronghold and supremacy over the temporal power of the king (visual power)[ as in Western art].

               On his side, the king himself wanted to use visual storytelling to emphasise his own power, to impress his people, Khmer overlords and rulers of adjacent countries. For example, the royal performances of the Ramayana’s narratives depicting events of the life of Rama, the most humane incarnation of Vishnu, alluded clearly to the divine nature of kingship. The approach to the divine was important also for the king; he could e seen as the one whose soul had achieved the closest possible relationship with Shiva or Vishnu or Buddha by virtue of his rightful and ascetic efforts.

Brief history of visual storytelling in Khmer art.

The first signs of story-telling in Cambodian temple art come from the 6 to 7th century (Sambor Prei Kuk).  Strangely, this episode was limited to the area and narrative storytelling disappeared until the end of 9th century, with Bakong in 883 and about 100 later at Banteay Srei (967).

The narratives comes from monuments from the presumed capital of Zhen La, Sambor Prei Kuk (late 6th - early 7th century). Some oldest examples have some narrative representations sculpted on sandstone lintels (see Benisti’s  1970: Fig. 2 a ) of which one copy is deposited at the Museum of Phnom Penh (re-pained replica?). Further examples are on the lintel of the N towers representing Shivaite rituals or celebrations.

At Sambor P.K., the eastern face of tower ‘S’, has a now much eroded high lintel. Under a very robust arch, enveloped by an undulating naga, is the scene of 3 men (Benisti 1970) probably rishis or priests going to pay respect to Shiva with his consort Uma. I consider narrative this lintel since it has worshipers approaching two divinities to interact spiritually with them.

Decoding the narrative possibility of the “flying palaces” hosting groups of figures is harder. In the event of the figures being involved in a ceremonial or ritual celebration, the palace’s images narrate an event. On the contrary, if the personages are simply an heraldic (hieratic) group of static figures or deities, then the representation on a “flying palace” is not narrative.

Always at Sambor Prei Kuk, the figures on the first enclosure’s wall, carved inside the large circular brick frames seem involved in violent narrative actions, as described in texts (Ramayana, etc.). 

A good example of narrative storytelling is on the lintel of Vat Eng Khna (c.650-700), kept at Cambodia National Museum, illustrating the origin of the linga (Jessup & Thierry 1997: 173, fig. 20), with a procession of many figures carrying a bow (offerings?), converging toward a shrine in which a seated king or statue (?) is being anointed. This happens below the large image of a muchalinga, carved on the complex frame of the lintel between Brahma and Vishnu. All the carvings of the 6-early 7th centuries show a great influence from India, perhaps due to the influx of Indian gurus and people of culture.

Narrative signs can be found in the lintel of Prasat Han Chei (Bénisti 1970: fig.87) carved with the images of sages (rishi) in hieratic style over the central smaller figure probably of Vishnu on Garuda. Similarly, the pediment of Neak Ta Po Noreay shows the scene of Mahishasura-mardani (? 6th-7th century; Bénisti 1970: 122).

Towards the end of the 7th century, in Southern Cambodia many brick and laterite single tower were erected, but none have narrative reliefs but only a common lintel with large vegetal branches. Exception is Prasat Andet tower that has a very unusual empty space at the top of the tower apparently carved with what could be  palace of the gods.

The temples of Phnom Kulen (end 7th century), show a difficulty to have narrative on the early pediments[2] carved on the brick’s temples . In some Kulen temples, above the lintel there is a central figure of a god flanked by minor goods or worshipers, thus of a non-narrative storytelling.

  I regret not to be aware of the new discoveries done by the Australian team of Sydney University at Phnom Kulen. What is I saw are eroded non narrative pseudo pediments
It is worth mentioning here the  drawings (filled with white paint) of Neang Khmau, now almost unreadable but that Marchal was able to reconstruct in early 20 century and publish in 1955 (Page 153), Some drawing are narrative such as Krishna lifting Mt. Govardhana, Vishnu making the three step. They were attributed to the  10th century.

The first evidence of narrative may by be seen on a single slab decorating the 5th step of the stepped-pyramid of Bakong initiated in laterite by Jayavarman III in 881 and finished later by Indravarman I (2102-1050) (Jacques 1999: 198). If one slab has a well identifiable group of Yakshas preparing for war,  the other ones following or preceding this one, are difficult to identify because the rock slabs were already of poor quality that was quickly degraded by tropical monsoons and time.) One can perceive figures of men and rishis in meditation, and running men. They make me to assume they may have been part of a narrative.  The original top shrine was totally destructed and a new one added, probably during the 12th century, by Yashovarman I, but restored in the 12th century by Yashovarmnan II (Jacques 1999: 198). In either event, the rock used was unsuitable. 

 The last temple of the Roluos area was the Lolei, initiated by Indravarman I but completed by Yashovarman II in 893 (Jacques 1999: 202).  Although it presents some of the largest and finest lintels of Khmer art (of the hieratic type), the overlying pseudo-pediment made of bricks, show a single palace flanked by worshipers. It is highly eroded now, almost unrecognisable.
This hieratic, non-narrative tradition continues in the majestic mountain-temple of Bakheng, consecrated by Yashovarman II in c.907 (Jacques 1999: 69).  The Bakheng’s lintels and pediments are very similar to those of the Bakong, except the single true pediment over the southern door of the tower probably depicting a story or eulogy of Shiva and his followers. This is the first large well framed pediment of Khmer art, made in sandstone, not of bricks.

My my knowledge on possible narrative developments at Koh Ker temples (c.928-.941) is extremely poor having visited in the distant 2003 only dozen of temples known before.
The first evident narrative example is to be found on the neatly carved lintel of the small brick temple of Preah Enkosei in Siem Reap, built during the second half of the 9th century, presenting a sandstone lintel over the door of the central shrine. Its upper band - rather high in comparison with the main lintel of hieratic type - depicts on the right side, the Churning of the Ocean of Milk in horizontal layout and, on the left, the figures of Brahma, Shiva and men.  It seems the sculptors wanted to combine two episodes: firstly the gods asking Brahma to persuade Vishnu to help producing the elixir of mortality (amrita), then the actual Churning carved to the right of this scene.  This lintel may thus be seen as the oldest attempt at a deliberate storytelling. Possibly a little earlier age.
Attributed to the date of 921 are the famous reliefs of Prasat Kravan, showing scenes on Vaishnava mythology.

               A significant change in architectural style and construction method appears in the temples built in the Koh Ker region by Jayavarman IV from 921 to 937. However I will express my opinion when all the contemporary dynamic excavation and restorations will be made public. After 940 CE., the Khmer capital was reinstated at Angkor, initially in the area of the Eastern Baray.  The East Mebon temple was built for Rajendravarman II in 953 by the architect Kavindrari-mathana (Jacques 1999: 161); I did not recognise any narrative relief in the shrines of this large stepped pyramid.

In the contemporaneous Preah Rup unfinished temple I was able to detect the first tentative sign of narrative carved on a high lintel fallen to the ground unfinished. One of them is carved with Rama shooting his arrow towards a much sketched Ravana. This uncertain stage will be surprisingly replaced by the fully narrative pediments of Banteay Srei.

The dating of Banteay Srei has been problematic since the day of its discovery in 1914. Initially (1919) the temple was considered by Parmentier as an example of the art of Indravarman (late 9th century).   In 1925, the same author, in association with Victor Goloubew and Louis Finot completed a detailed monograph (published in 1926) on Ishvarapura – the old name of Banteay Srei - and concluded, on epigraphic evidence, that the temple had been erected in 1307.  However, this dating was definitively contested in 1929 when George Coedès[3] re-interpreted the inscriptions and concluded that the temple, including the 3 shrines and the two gopuras of the first enclosure, were not later than the 10th century.1 This was dramatically confirmed in 1936 by the discovery by Henri Marchal of a foundation stele buried in the earth by the gopura of the 4th enclosure.3 It was soon interpreted and publicised by George Coedès, together with three other inscriptions (1937: 143).  The year of the temple consecration recorded on the stele as 967 was thus definitively taken as the date of the whole temple. The new attribution of Banteay Srei to 967 was widely accepted.  Little concern was expressed about how this date could apply to various buildings of different styles and construction techniques within the temple compound.  However, some scholars were uneasy about the mature style of the reliefs carved in the temple, particularly on the libraries pediments, as well as an array of other advanced architectural and decorative techniques (Parmentier 1939: 153; Glaize 1944: 228).  Little or no consideration was given to the relationship of the stele date and the dates provided by other inscriptions.  The possibility that several patrons contributed to the completion and enhancement of Banteay Srei over a period of time was ignored[4].
In 2007, Olivier Cunin has made a perfect visual summary of the birth and evolution of B.Srei, on behalf of the Apsara National Authority, publically visible at the entrance of the temple. It can’t be missed! The reliefs carved in the various edifices of Banteay Srei are refined and beautiful, of a very high level of visual storytelling. For me they are the best of Khmer art of all age.

At the state-temple of Baphuon (c.1066), the decoration consists of images of nature (trees, wild and domestic animals) and many mythological and epic stories of the classic Hindu literature, mainly of Krishna. Although the narrative of individual panels can be understood instantly, often the association with another panel has to be made to make sense. This is network reading. This results in an extraordinary richness of narrative stories (Roveda 2004 and 2016) and messages. At Baphuon, by networking reading Mahabharata’s stories becomes evident to belong to the same text used at Angkor Wat. Equally, for the Northern face of Gopura III was devoted to Rama’s epic. Altogether the quality of carving is extraordinary.

Copies of the Baphuon panels are found on the Eastern gate of the Western Baray temple and in the remote and Prasat San Kew, in a severe state of degradation. Narrative pediments became common at Pimai (1080) in modern Thailand. At Preah Vihear a few pediment are narrative, including that of the renewed Churning of the Ocean of Milk (my photo of 1999) that has to be combined with the underlying lintel showing Vishnu on Ananta meditating before participating to the Churning. I doubt to classify narrative the sandstone pediment  with the story of Shiva and Uma passing by under a marvellous tree, riding the mythological bull Nandi, because no action is involved apart the presumed “passing by” which is a personal interpretation.

I am showing some example of networking narrative from the photographs I have taken during Baphuon restoration works, before the final state of today. The stories carved in reliefs have been identified in many publications, including Maurice Glaze, 1993. Recently, in 2007, Olivier. Cunin has made a perfect visual summary of the birth and evolution of Banteay Srei, on behalf of the Apsara National Authority, publically displayed at the entrance of the temple. It can’t be missed!

The network narratives of the panels of Bapuon were not seen before in Khmer art, or after and the mental origin remain inexplicable to me. I like to put forward the theory that the relief’s carvers belonged to different workshops. They were not concerned with the chronological development of the events, creating thus a random display of images requiring network reading. At Baphuon the influence of southern India art is considerable (K. Evans 1007, A. Dallapiccola 2010), possibly brought in by gurus hired by the king.

With the colossal temples of Angkor Wat visual narrative reached its apex with the full deployment of complex narrative programs.  The need to illustrate mythological and epic events on a large surface, obliged the architects to construct long galleries with enormous smooth walls for relief sculpture.

The construction of the temple of Angkor Wat was initiated by king Suryavarman II, who reigned from 1113 to 1145 (or 1150; the exact date is unknown), as a temple in honour of Vishnu and as the state temple.  I think that it was conceived and planned by the royal Brahmin Divakarapandita who had served two previous kings and himself descending from a family of illustrious brahmins (see their history on the inscription of Sdok Kak Thom).  The construction of the temple started around 1113-15 but was probably halted after about 30-35 years, when the king died. It may be, as Mannikka suggested, that the central sanctuary was installed in 1126, and that perhaps all decorative works including splendid narrative reliefs were suspended after the king’s defeat by the Chams in 1136 (?).  Most architectural elements and sculptural reliefs were left unfinished at the time of the death of Suryavarman II in mysterious circumstances. 

The overwhelming richness of themes, taken from the classic Indian texts is not so astonishing when considering that the close advisor of the king was Divacarapandita. This vrah guru of the king had a solid Indian background and a great culture. I believe that he had the ability to conceptualise, visualise, draw and probably supervise the carving of reliefs all over the temple.

I use the term conceptualizing for the mental process of transforming a story narrated in a text into a visual story, or more simply to know how everything of a textual story would look like on a carved wall, like at Angkor Wat or Bayon

 I believe that Divacarapandita was an artist comparable to the great geniuses of the Western Renaissance. Besides full knowledge of the Puranas (especially the Bhagavata), as well as of the Mahabharata and Ramayana, he had a structural mental plan on where and what use for each part of Angkor Wat, obviously prioritizing Vishnuite imagery. There has been other illustrious vrah guru, but none could achieve what Divacarapandita did, in my opinion.

He had an exceptionally fine brain to preview how a story could be illustrated in in a defined space, its layout and quality and technique of carving.   To execute his mental vision, Divakarapandita must have drawn on paper his perception and then use it for “pochoir” technique (B. Ph. Groslier 1961: 162), a sort of stencils. The Chinese traveller Zou Daguan mentions that official documents were written on buckskin parchment dyed black, using sticks made with something like chalk, corresponding to Thai dinsò, (pencil ; Pelliot 1992 : 27). Manuscripts have a limited life and the maintenance of the texts they contain need frequent recopying (introducing mistakes). The decline of writing Sanskrit started to decrise in the 13 century and disappeared at the 16th century.

Following the narrativity’s theory, a simplifier view of the panel illustrating the Historic Procession would reveal the hero, (the real protagonist of the story) Suryavarman II, sitting on his field-throne, surrounded by his courtier, ministers and priests, is in full control of his power. Then, after descending Mount Shivapada and having joined the procession, he reconfirm hi position of power (of hero) another time by standing on his elephant surrounded by complacent subalterns, his generals (but that to all effect are soldiers) in charge to lead a group of mercenaries (the Siam Kuk) that in case of war will be the first to die to protect the hero of the story (Suryavarman II).

The long reliefs of Angkor Wat have the peculiarity that the two enemies (the hero and the villain, Krishna (Vishnu) against asuras are never depicted face-to–face, but there are always intermediaries’ figures between the two. In the case of the panel of the northern gallery eastern wing (Krishna against the asura Bana. The image of Krishna is repeated 7 times, each to reaffirm his hero position until he get closer to the villain (Bana) that he destroys re-establishing pace and order on earth.

In the long panel of the northern gallery western wing (Vishnu/Krishna against the asura Kalameni), between the hero (Vishnu/Krishna) and the villain (Kalameni) that they illustrate, there are 13 intermediate elements (gods and deities, protectors, helpers?) before Krishna gets closer to the villain Kalameni appearing with lots of arms brandishing swords, symbol of their un-natural destructive power.

 In the battle of Lanka panel, Rama (the hero) stands proudly on the shoulders of Hanuman, flanked by his brother Lakshmana in a much lower position (of power). The heroic Rama can see far away the figure of the villain (Ravana) on his chariot. Between the two there are several cruel battles delaying the re-establishment of equilibrium with the death of the villain Ravana.

There is no evident structural guidance for reading the mural of  the Battle of Kurukshetra depicting two fundamental events of the Mahabaratha The relief starts by showing an hero (the great Bishma),dying on a “bed of harrows”, but having the time to narrate the need for all kings to cease hostilities and remember theirs duties.

On the relief there is no visual presence of evil but only the Kaurava infantry attacking the Pandava infantry. The villain may be interpreted the hidden Kaurava or visually by Arjuna (a Pandava) who was having a crisis of conscience, being unable to kill his Kaurava cousins of the same blood to win the battle. Arjuna is illustrated on his war chariot driven by Krishna, who reminds him that his highest duty of a ksatryia: accept god’s will and keep fighting (all narrated in the Bhagavad-Gita). After Arjuna understood Krishna’s spiritual teaching, peace was restored and the real evil (ignorance) defeated. I do not know if at that time the symbolism of ending fratricides wars was understood.

In conclusion, the Angkor’s reliefs are tentatively dated 1115-1145, and art historians use the term “Angkor Period’, for the period that is contemporaneous or that follows, until the making of the Bayon reliefs.

Art Historians did establish a “style of Khleang” as an intermediary between Angkor Wat’s style and Bayon’s style. I do not see any iconographic reason to have done so. I have not seen in the two Kleangs or on the Ta Keo pyramid, anything particular apart of the rich decoration of some friezes and a damaged pediment; certainly there were no narrative reliefs at the time of my visit in 2002. It is not clear if the temple of Beng Mealea was built, before or after Angkor Wat. Probably it was started a little before Angkor Wat.

To the period going  ‘Suryavarman II - Jayavarman VII’ can be considered as the period in between Angkor Wat and Bayon I place the narrative reliefs of Banteay Samrè, Thomannon,  Chau Say Tevoda, and other “gites d’étape” of Dagens (3003: 26) , having in common  the new architectural structures of mandapa and antarala. The temples of  Banteay Samré, Pimai and Phnom Rung (in modern Thailand), both having some Tantric Vajrayana motifs mixed with Hindu myths.

At the extraordinary Bayon temple, the themes carved on reliefs were popularized, involving lay people and daily life’s scenes. Already at Ta Prohm wall’edges were carved with many topics of the life of common people: a man carrying a bucket, a woman carrying flasks on a pole, a man laying fishing nets, mothers holding their baby, embracing lovers and sometimes some Buddhist figures of Jatakas. This interest in popular activities is further illustrated at Bayon in the market scenes and other popular activities carved on the southern side of the Outer Gallery. Another novelty is the display of cruel wars, both having dead and wounded men.

At Bayon the desire for gigantism appears to have been coupled with an increased need to communicate religious and secular messages.  The rapid passage from the relatively naïf narration of the Baphuon or the severity of Angkor Wat to the Bayon’s full-blown narratives of the Outer Gallery, is quite astonishing.

The views of Bayon’s Outer Gallery make visitors interact with imagery mentally and physically, extending the power of the narrative in real life. The lively interaction with life appears in some contemporary ceremonies in Thailand and Laos.

The panels of Bayon’s Inner Gallery are all with mythological scenes. They reveal diversity, possibly due to different age of making or certainly different carving workshops. They are comparatively in worst conditions due to the weathering by the humidity of the small rooms.

Furthermore, the reliefs suffered defacing by the iconoclastic reaction after the death of Jayavarman VII.

According to the theory that the Bayon is the perfect centre of Angkor Thom (which is not), the temple functioned as Mount Mandara, and the giants at the entrance gates were pulling the naga for producing the elixir of happiness for the King and his nation (Dagens 2003: 32, with figure). The king positioned at the centre of the Churning would legitimize his position of Universal Sovereign (Chakravartin). This theory is controversial by many factors that are behind the scope of this paper.

The focus of Bayon’s religious worship was originally Mahayana, having images of gods brought in from all parts of the country, dwelling in the temple’s shrines built around the circular section of the structure, the gigantic face-towers creating an ideological message with the heart in the Bayon, centre of the capital-city. The central image of worship was a colossal statue of Buddha on naga (characteristic of Mahayana Buddhism).

In general, all complex or extravagant reliefs are narrative, with lay images on the Outer Gallery and religious narrative in the Inner Gallery of the first level (Roveda 2006).

 It seems to me that Khmer kings used gigantic architecture to express the position of power of the ruling king (the hero) and his reverence to a special god (Shiva, Vishnu, etc.).

In the Bayon, it appears that the decoration was unplanned, executed randomly by artists of different workshops. Therefore, also at Bayon, trying to understand the reliefs by circumambulation the temple is meaningless, especially a systematic pradakshina (counter-clock wise) for funerary ceremonials.

To complete the study on visual narrative’s development in Khmer was imperative to look at two more elements of the development of narrative: lintels and pilasters.

On them Jean Boisselier did an extensive excellent study of lintels in his 1966 book (Le Cambodge). In my opinion the first lintels that I have seen with narrative scene are to be found at Phimai.

Concerning narrative sculpture on the door pilasters, especially at their base, started to be manifested at Angkor Wat, developing at Beng Mealea, then at Banteai Samre and all the temples of Jayavarman VII.

The progress of the development and evolution of Khmer visual narrative (9-14th centuries) was an exclusively Khmer success, achieved through the geniality of the artist-designers and the indefatigable application of Cambodian sculptors. 

It appears that after Prah Palilay, visual narrative art started to decline firstly in the carving of the reliefs, and secondly by the ignorance of new Buddhist stories. Another example of this type is temple No.438.   No more stone-temples were built eliminating visuals storytelling It is possible that this lacune was filled by verbal storytelling, theatre and music.

From the 16th century with the development of Theravada, there must have been simple wood pagodas or viharas, gradually being built with stone walls, and -much later- decorated with painting, totally disregarding Brahmanical iconography. Therefore I believe that Khmer art did not die with the successors of Jayavarman VII. It was recuperated by Buddhism, with monks worshiping in sparsely decorated (with paintings?) walls until the 18th century when colour painting appeared. It survived only in Thailand with the Thai murals of Wat Rajapurana of Ayutthaya (c. 1250) and the graffiti of Wat Si Chum in Sukhothai (mid-14th century). Only in late 18th, but mainly 19th century, both in Thailand and Cambodia, painted murals made a visual triumphant entrance. Another art gap was produced by wars, internal strife, political and social disorders all overdo continental Southeast Asia.

Narrative development

I have tried to explain the development and evolution of visual narrative carved in Khmer temples.

 It is difficult because there are only few points of reference, spaced in time: the poor reliefs of Bakong and Bakheng, the shocking beautiful Banteay Srei, the still unknown Phnom Kulen and Koh Ker reliefs, the extravagant Baphuon, the marvellous classic Angkor Wat and the messy revolutionary Bayon.
The beautiful Banteay Srei murals have the first well-developed visual narrative on all the pediments and lintels. There are also pediments of different making and “style”, one over the inner pediment of the second enclosure and three others in custody of museums. I am not surprised to see at Banteay Srei the visual narrative was so well established, because the input of the Brahmanical guru Yajnavaraha contributin to the erection of the temple. Its refined style was never reproduced after, or copied by other image-makers, or visual narrative creators, for almost 100 years.
The next step in my search on the development of visual narratives is the temple of Baphuon built and designed by Udayadityavarman’s architects’ and terminated by 1060. The profusion of interesting new modes of visual representation and reading seems to have superficially called the attention of Divakarapandita, later totally reorganised at Angkor Wat.  After this, Divacarapandita became involved at Prah Vihear, Wat Phu, etc. The surge of the need to decorate temples brought new emphasis on visual storytelling, a very significant fact because it generated a new appreciation of images.

               The network narrative of the murals of Baphuon was unique in Khmer carving  and its conceptualization a total novelty. At Baphuon images unroll like that of storytelling for children, although many of its thousands images require the effort of network reading, just the opposite of what will be at Angkor Wat. 
Separate by around 30-40 years, are the reliefs of Angkor Wat. The long panels of the third enclosure are all of classic narrative type with long panels full of orderly energy, expressing with clarity the visual language.

At Angkor Wat the Khmer visual storytelling was completed, revealing emotions to the eye of the viewer. The Vishnuite and other Brahmanical images of Angkor Wat became the standard Hindu references for all the temples until the Bayon. The architect and designer was the King’s guru Divakarapandita who conceptualised and had carved the enormous panels executed with the help of Cambodian artists. They released energy to the reliefs, followed an orderly plan, the lot supported by the innovative spirit of Suryavarman II and his wealth.
Divakarapandita contributed to the creation of a revolutionary mental change for visual storytelling from the random Baphuon networking narrative to the linear classicism of Angkor What. I believe that Divacarapandita was a competent and gifted artist like to the geniuses of the Western Renaissance. He was familiar with events written in the Ramayana of Valmiki and the Mahabharata and Puranas great sources on mythological stories.

The Buddhist king Daranindravarman II ( 1050-?1065) successor of Suryavarman II, started to build several temples to the east of Angkor, Beng Mealea, Banteay Samré, Thomannon, Chau Sei Tevoda and the distant Prah Khan of Kampong Svay. Mature visual storytelling is evident in these temples and on fragments of pediments now preserved at the Musée Guimé, Paris.
The last step of my study on the evolution of visual narrative in Khmer art is that of Bayon.

To me, the temple reflects disorder, either in its planning or construction (see Olivier Cunin on Internet pages). It was built in stages, with changing iconography.

The dark Inner Gallery has Brahmanical mythological themes, not always narrative.

The Outer Gallery has only lay and religious narrative reliefs carved at different scale, different techniques, certainly by different workshops (Roveda, 2006), and at different times. All this reflects lack of planning and execution’s coordination of all the decoration of the Bayon.

Personal conclusions:

1. Khmer art was new each time a new King raised to power. This renovation demanded a change of style or a different degree of visual language.

From the 10th century there must have been some small illustrated manuscript on various topics circulating in Cambodia.

2. In early Imperial Cambodia, visual narrative functioned always as a demonstration, a visual display of the King’s a power visible from everywhere in the great city. It satisfied the ideology of power that obsessed Khmer kings, an ideology that- in part - materialised in monumental architecture.

3. The development or evolution of Khmer visual narrative seems to be erratic probably due to the lack of temples which could have reliefs. The visual narrative that started with honour in some lintels of Sambor Prei Kuk; then it jumps to the presumed narrative of the Bakong. Thereafter there is an extraordinary evolutionary jump to the mature visual narrative of the extraordinary reliefs of Banteay Srei, a fact difficult to explain in Khmer art.

 This first narrative display was followed by the impressively extravagant Baphuon. Then we arrive to the needed classicism of Angkor Wat, clear and creative. Many temples built around the time of Angkor were influenced by the art of Angkor Wat, with a variety of visual narratives carved mainly on pediments. After the structured classicism of Angkor Wat, the narratives of Bayon seem decadent, with loss of coherence. With the 14th century everything changed and the expanding Theravada needed Buddhist images. Of these, few originals stone reliefs remain (Preah Pithu “X” and Mangalartha).

We will never know when and how Hindu narratives on stone were replaced by Buddhist visual narratives in painting because they were executed on perishable materials, except the statue of Lord Buddha. Mural painting started to flourish only in from the end of the 19th century.

4. It seems to me that Khmer art, from Sambor Prey Kuk to Bayon, was based on the freedom that the Khmer king allowed to each new sage (guru) as his personal guru, to use elements of Indian culture. Some gurus were of Cambodian genealogy (born and raised in Cambodia, others Indian) this imported knowledge allowed new visual narratives, conceptualized, absorbed and localized by the Khmers artists and carved with the dedication and ability of Cambodian artists and sculptors. I conclude in believing that Indian culture was the source of all our storytelling.

5. In my research (2017) on the development of visual narrative in mainland South-eastern Asia, I have provisionally reached the conclusion that the Khmers were the first (after Borobudur and probably Prambanan) to carve in stone visual storytelling from the end of the 10th century, well ahead of other countries of the same region, showing a progressive interest in the history of images. This is not surprising considering that  the Khmer was the largest empire of Southeast Asia until the end of the 13th century or the death of Jayavarman VII.

My comments on Myanmar are very scanty limited to the knowledge only of some stone tablets of the Payagyi temple (18-19th century) that have nothing in common with Khmer temple’s reliefs. However a well-documented explosion of visual storytelling happened with the murals of the Buddhist temples of Pagan (11-14th centuries) and later in the 18th century caves of Powin Taung (Munier &Aung, 2007).

Most monuments and their reliefs were done by slaves of the king, of which hundred-thousand died in the construction of the monuments, giving all of themselves to the King and his ideology

All my research is based on the chronology and kink’s genealogy established by EFEO scholar in the 20th century. To them I am greatly indebted.

Concerning the beauty of Angkor Wat reliefs I totally agree with what B.P. Groslier wrote in the distant 1961 (Indochine). I give here a summary free translation from the French at page 162.

The reliefs look more like mural paintings (“fresques”) because we can perceive more the brush of a painter rather than the chisel of a sculptor.   The Khmer were familiar with painting, having painted the decoration of some of their temples. Evidently the sketches of these reliefs were made on paper, then utilised as “pochoirs”. Furthermore, originally gold and colour were applied for the main personages, their jewels ant setting high lightening the main pictorial effect.

The composition or layout was also coming from drawing techniques. The perspective is quite simplified and the layout of a disconcerting audacity, especially thinking at the small panels of Baphuon. The composition is continue for each panel: 49 metres long on the East and West faces and 100 m on the North and South faces. The episode is built dynamically by the movement of the actors, rarely driving the gaze on the main personage. This parsonage is of larger size on his mount; duel of the heroes on the first plan; soldiers turning their head towards the leaders, a horse raising his neck. This emphasis produces the tempo of a majestic recitative, avoiding slow down. It is a superior art that allows the viewers the possibility to create in their mind a personal vision, though guiding them unconsciously.

The making of the reliefs seems to have ben done by chiselling away successive thin layers(“films”) of rock, evidencing again the following or directing . The personage are firstly carved flat, then modelled on a few centimetres of thickness, only millimetres for the vegetation. He space is defined by semi-tones thanks the play of different carving.

BP Groslier concluded (page 164) that there is nothing better in the world than these narrative reliefs. Thy can be flanked to the best frescos of the Italian Renaissance. On this basis of this alone, Angkor Wat is one of the marvels of the world.

Bénisti, Mirelle., Rapports entre le premier art Khmer et l’art Indien, Publ. EFEO, Mem. Arch. V, Paris, 1970

Bhattacharya, Kamal The theme of Churning of the Ocean in Indian and Khmer art, AA, VI, 1959: 121-134

Boisselier, Jean, Asie du Sud-Est, Tome I, Le Cambodge, Picard, Paris, 1966

Coedès G. & Dupont P., Les stèles de Sdok Kak Thom, Phnom Sandak et Prah Vihar, BEFEO, XLIII, 1943: 56-154

Coedès, George., La date du temple de Banteay Srei, BEFEO, 29: 289-330, 1929

Coedès, George, Les bas-reliefs d’Angkor-Vat, Bulletin de la Commission Archéologique de l’Indochine, Paris, 1911: 170-220

Coedès, George, La date d’exécution des deux bas-relief tardif d’Angkor Vat, JA, II, 1962: 235-243

Cunin,Olivier, posters for Banteay Srei, explaining the history and how to visit the temple. Project on behalf of APSARA Authority, Siem Reap

Cunin,Olivier, The Bayon, on Internet

Dagens, Bruno, Etude sur l’iconographie du Bayon (Frontons et linteaux), Arts Asiatiques, XIX, 1969, 124-158

Dagens, Bruno , Les Khmeres, Les Belle Lettres, Paris,2003

Dallapiccola, Anna, Ramayana in Southern Indian art, in Vergese &All.,  Themes and variations, Art and Archaeology, Oxford University Press, New Delhi 2010

Dehejia, Vidya., On modes of visual narration in early Buddhist art, The Art Bulletin, LXXII/3, 1990

Dimmitt C. & van Buitenen J.A.B., Classical Hindu Mythology, Temple University Press, Philadelphia, 1978

Dufour, Henry , Le Bayon d’Angkor Thom, Ministère de l’Instruction publique et des Baux Arts, Paris Leroux, 1913 (Folio)

Evans Kristi, Epic narrative in the Hoisala temples. Brill, Leiden 1997

Finot, Louis , Goloubew, Victor. & Coedes, George, Le Temple d’Angkor Vat, Publ. de l’EFEO, Memoires Archeologiques, II, Paris, 1927-32

Finot Louis, Parmentier Henry et Goloubew V., Le temple d’Ishvarapura (Banteay Srei), Publ. de l’EFEO, Mem,oires archaelogiques, I, Paris, 1926

Finot Louis, Les bas-reliefs de Baphuon, Bulletin de la Commission Archéologique de l’Indochine, Paris, 1 Giteau M., & Guéret D., L’Art Khmer, Reflet des civilisations d’Angkor, ASA Edit., Paris, 1997.

Glaize, Maurice , Le Guide d’Angkor: les monuments du groupe d’Angkor, Paris Maisonneuve, 1963

Goloubew, Victor, Artisans chinois à Angkor Wat, BEFEO, XXIV, 1924, 513-519

Groslier, Bernard Philippe, L’Indochine, in L’Art dans le Monde, Edit. A.Michel, Paris, 1961

Holliday, P.J. (edit.), Narrative and event in ancient art, Cambridge University Press, 1993

Jacques, Claude, Angkor, River Books, Bangkok 2002

Le Bonheurm, Albert, Angkor, temples en péril, Herscher, Paris 1989

Le Bonheur, Albert, Of Gods, Kings, and Men, Serinda, London, 1995

Loizeau Rachel, Indic Epics in Khmer Art, Pdf available on internet

Mannikka, Eleanor. , Angkor Wat, Time, Space, and Kingship, University of Haway’i Press, 1996

Marchal Henry, The temples d’Angkor, 5th edit., A. Guillot, Paris 1955

Press, Chicago, 1987

Mitchell W.J.T., Iconology, University of Chicago 1987

Munierm Christophe & Myint, Aung, Burmese Buddhist Murals, Vol.1, White Lotus, Bangkok 2007

Mus, Paul, Le sourire d’Angkor, Artibus Asiae, 24, Ascona, 1961, 363-381

Parmentier Henry., L’art khmer primitif (2 vol.), Publ. de l’EFEO, 21&22, 1927

Pelliot, Paul, The Customs oF Cambodia, The Siam Society , Bangkok,1992

Poncar, Jaro and Maxwell,Thomas, Of Gods ,Kings and Men, Edition Panorama, 2006 

Rodriguez, Natalie, Variation autour d’un theme iconographique: Arjuna et le Kirata. EASAA, 6th Internat. Conf, Leiden, 1996

Roveda, Vittorio, The use of drawings for the making of the reliefs from Angkor Wat, in Southeast Asian Archaeology 1998, Centre for South-East Asian Studies, University of Hull & Ethnologishes Museum, Staatliche Museen zu Berlin, Hull, 2000, 169-175

Roveda,Vittorio, The archaeology of Khmer images, ‘Aséanie’, 13, June 2004
Roveda,Vittorio, images of the Gods, River Books, Bangkok 2004

Roveda. Vittorio., The world of Khmer mythology, APSARA, Phnom Pewnh 2016

Stern, Philippe, Le s monuments du style du Bayon et Jayavarman VII, Publication du Musée Guimet, P.U.F, Paris 1965

Van Nooten B.A., The Mahabharata, Twayne Publishing, New York, 1871

Zephir, Tierry, L’art Khmer, in L’art de l’Asie du Sud-East, Citadelles & Mazenod, Paris 1

 – Sambor Prei Kuk. Lintel of tower N showing a Shivaite scene below a robust naga.

Photo by M.Benisti before 1970

 – Sambor Prei Kuk. Pediment wit the scene of acolits dancwe around Shiva. National Musem iof Phmnlom Penh.
(painted plastercast?)

– Wat Ek Khna. Procession of worshippers converging to the central shrine, probably for the anointment of a statue or a king. On the robust band over the scene are a Shivalinga at the center,Vishnu to its  left an Brahma to its right.

Artefact preserved at the National Museum of Phnom Penh (photo2007).

- Sambor Prei Kuk. Typical flying palace

– Sambor P.K. The degraded wall of the central enclosure with

brick wall carved with narrative figure

- Kulen. Pseudo-pediment with figure.            

   PrasatThma Dap (Kulen).   Pseudo-pedimeny with a figure in a shrine over the lintel. All white-washed.

– Bakong towers. Traces of a human figure on an animal (Shiva on Nandi?)

Carved at the top of a proto prdiment.

 – Bakong towers. Traces of a human figure on an animal (Shiva on Nandi?)

Carved at the top of a proto prdiment.

 Bakong towers with a proto-pediment, the outline of a pediment with niches for images of gods.

 – Bakong. - The only preserved panel of the             
The frieze on the 15th step of the pyramid                                                      

        – Bakong. This northern door of access to thy temple. It is analyzed in the paper on pediments’ evolution.
Bakheng, southern pediment of the central tower tower with figures of Shiva and Uma on the bull Nandin; the rest of the images  are obliterated.

 – Pr.Krvan. Vishnu making the three steps (Vishnu Trivicrama)
 – Pr.Kravan. Lakshmi in hieratic pose

Prasat Wat Einkosei. Top band of the large lintel sdhowing the major Brahmanical Gods to the left. On the right the Churning is going on with uneqal pulling powers. One of the first example of visuasl storytelling in Khmer art.

Banteay Srei. Pediment of the northern shrine of the causeway illustrating Vishnu
in the semblance of a lion, tearing apart the stomach of the demon Hiranhyakasipu.

Lintel narrates one episode of the Arjunakirata  legend, by which Shiva and Arjuna argued who had been the first to kill a wild boar, represented on the lintel under their feet.
 *It is important to notice that at Banteay Srei also the lintels became narrative, gently and elegantly.

Banteay Srei, This lintel narrates the episode of Ravana kidnapping Sita,
while Rama was far away, in the forest chasing Marica and Lakshmana also in the
forest searching Rama. The brothers are carved at the extreme edges of the lintel,
 too far and too late to reach Ravana with their arrows.
Banteay Srei. This small pediment fallen on the ground narrates the episode of Virada attempting to kidnap Sita, but I soon killed by Rama an Lakshmana with their Arrows.
Banteay Srei. Also this lintel narrates the story of Rama killing the monkey king Valin when he was fighting with Sugriva. Rama is shown twice with bow and arrow.
Detail of the previous pediment showing frighten emotional states of Kama’ courtiers and girls of the harem.
Banteay Srei. Complete pediment illustrating the Mahabharata legend of Bihma
(Pandava) fighting Duryodhana (Kaurava) in the presence of Krishna, Balarama and Arjuna. The duel ended with Krujisna punishing  Bhima for killing his enemy by hitting him below the navel, an unacceptable act amongst kshatryia.
National Museum, Phnom Penh
Banteay Srei. Pediment over the gate of the Third Enclosure with the Ramayana event of Rama killing Valin, shown dying on the ground (left corner of pediment).

The massacre of all the boys born the same day of Krisha
The plot to exchange the ne-born male baby for new born female
Khrisna swapped for the Devaki (baby girl)
Baphuon, Mahabharata scenes
Baphuon, Ramayana scense

Baphuon Hanuman offering to Sita Rama's ring
Baphuon.Yudhishtira Desolate to have lost everything (left) at a chess game
Baphuon Vishnu Trivikrama
Angkor Wat. The Royal Procession. King Suryavarman II is shown on the largest elephant surrounbded by parasols and insignia.  Thed narrative is fully devloped in showing the pogressive march of th army in the fist ground, the King in the intermediate gound and the forest in th background. (Photo J.Pomcar)
Angkor Wat.A large pediment of the western tower of the First Enclosure showing the battle
of Rama against Ravana.

A series of pediments over the staircase
leading from the Cruciform terrace to the north
side of the Second Enclosure.
Preah Vihear. Pediment of the Third Level showing the Churning with the protagonist surrounded by several holy men (Rishis), Garuda and Indra on his elephant.
It may be that this pediment was carved after the construction works awere terminated, or during some restoration phase, being in such an advanced stage of visual storytelling.
Preah Vihear. Pediment of the third level showing Shiva and Uma
on the bull Nandi, passing by or resting under a large tree, interpreting thus the pediment as narrative storytelling 
Preah Vihear. Pediment of the Gopura of the Third level displaying a large pediment frame, which is polylobate and surrounded by foliage. It includes a pyramidal decoration of flowers and laves over a Kala’s head.
Notice the deep holes for inserting wooden posts to create a roof for
protecting a passage

Bayon. Two views of the famous face-tower. I personally wander if some Khmer architect of Bayon imagined that the faces of the towers acted as pediments of intense symbolic meaning.
Bayon. The army is followed by provisions on a cart, followed by a soldier with his family.
The visual storytelling has reached  a stage of realism never seen before.
Bayon. Soldiers at war                     
Bayon. Soldiers at war                     
Bayon. Men eating together using their hands

Fig.51 – Bayon.  Men playing chess
Bayon. Boy soldiers       
– Bayon. Special soldiers carrying the Sacred Arc
Bayon, Inner Gallery. Mythological scenes
Bayon, Inner Gallery. Mythological scenes

Bayon, Inner Gallery. Mythological scenes
– Bayon. Inner Gallery facing the sanctuary.                                                  
Panel carved with probably with a scene of the Vessantara Jataka.   
Panel carved with a scene of the Vessantara Jataka. (Photo J.Poncar)
                           Fig.58 -  Bayon. The Avalokiteshvara that allowed the attribution of the temple to Mahayana Buddhism.
Preah Palilai.Pediments with Buddha in meditation
Preah Palilai.Pediments with Buddha in meditation
Pra’ Pitu X. Rows of Buddhas

Pra’ Pitu X. Rows of Buddhas
National Museum Phnom Penh. Pediment allegedly coming from Tep Prahnam (Angkor Thom)
The Buddha is enthroned over the scene of evil Mara attacking The Buddha at the moment of enlightenment. 
He used his devil forces lead by generals on elephants. for the need to narrate the full story.
Between the images of the 2 elephants heads there is the figure of Preah Thorani


[1] Holliday P.J., Narrative and event in ancient art, Cambridge University Press, 1993, p.8.
[2] See my paper “Pediment’s evolution” in this series Archaeology of Images (No.22)
[3] In his 1929 paper’s (page 163) Coedès made no comment on the inscription of Suryavarman I at the temple of Sek Ta Tuy, a temple that Finot used to confirm the 967 date of Banteay Srei, on the ground that the same deity (Tribhuvanamaheshvara) was worshipped in both temples. All different opinions were dissolved by the stele found under the temple displaying the date 967.
[4] V.Roveda, in  Aseanie