What we do is what we blog:

The project aims at using digital technology – sometimes in a very simple and basic manner – for the documentation of cultural heritage in the region of the Western Himalaya and its foothills, a region wherein the former kingdom of Chamba was a major political and cultural factor for more than a millennium. We further aim at using digital technology to develop specific methods of representation for the different kinds of objects and data collected in our surveys. These methods include – but are not limited to – modelling of 3D-objects, 3D-printing, diagrams, plans, turntable movies, films as well as AR (augmented reality) and VR (virtual reality) environments. The goal is a contribution to digital cultural heritage preservation – a creative approach and an invitation to learn more about the cultural history of the region where the plains of Northern India border the Himalayas.

27.03.2023 – The Temples of the Kharura Area of Chamba 1: The Temple with the Stepped Roof:

© Gerald Kozicz – Temple with a “stepped roof”, Chamba, India, 2016


Among the temples listed by Sethi and Chauhan in their Temple Art of Chamba, we find six temples in a neighbourhood called Kharura located to the East of the palace. The authors mention these temples belonging to Brahman families. Unfortunately, not a single image illustrates the brief description. Neither do the mentions contain a sitemap or a sketch that allows to ascertain the exact connection between the textual description with the various monuments in the area. One description which can still be clearly identified with a building on the ground refers to a structure with a “stepped roof”.

Sethi and Chauhan (2009: 102) highlight several interesting aspects of the shrine which – according to the authors – had suffered severe damage in the 1905 Kangra earthquake but also show significant traces of insufficient maintenance. These include the “niches like Rajput balconies with cusped arches”. Today, only the western niche can be viewed since the other perspectives are completely blocked by a neighbouring house to the South and stored building material on the eastern side. Sethi and Chauhan further note the design of the columns and capitals, which reflect the impact of the art of the Mughal court. They also draw attention to the kīrtimukha on the capital.

Gaṇeśa is depicted in the lalatabimba position above the entrance, i.e., the central field of the first lintel. As usual, the elephant-headed god is heavily coated with orange colour which hardly allows his identification. The figure appears to be four-armed. Above, the Nine Plates (navagrahas) are lined up. They too, can only be identified by the lotus symbols held by the far-left figure which signal Sūrya. Otherwise, they show severe traces of abrasion.



Sethi, S. M. and Chauhan H. (2009) Temple Art of Chamba, New Delhi.

20.03.2023 – Workflow Update 3: How to plan for / in less than ideal situations:

© Gerald Kozicz – reconstruction process, overlay between lineart and photograph, Graz, 2022


This week’s update to our workflow tab will show how we deal with situations, where our previously mentioned photogrammetry pipeline – either though incomplete photographic documentation or difficult situations in the field – fails or is just not applicable. It will show the remarkable results and the potential that can be achieved with the help of simple sketches and manual reconstruction.

The update can be accessed either via the site menu at the top of the page or directly HERE.

13.03.2023 – The 1905 Kangra Earthquake: A Disaster for Communities and Architectural Heritage:

© Gerald Kozicz – Kangra Fort, Kangra, India, 2016


The regions along the Himalayan Range have always been exposed to natural threats due to tectonic factors, causing earthquakes and landslides in the aftermath. One of the worst disasters of the recent centuries was the so-called Kangra Earthquake on April 4th 1905. This earthquake also hit the region of Chamba right to the North of Kangra. We may assume that the damage already noted on the portal of the Chamesan Champāvatī Temple of Chamba was caused by that event. Quite similar patterns of damage can also be noted at the Gauriśankara Temple inside the Lakṣmī-Nārāyaṇa Complex. There the lintels of the outer face of the antarāla (porch) show the same kind of cracks. They run quite through the architraves that carry the load of the śukanasa above. A major vertical crack is also visible along the śukanasa between the central field and the left lateral part (see the images of the December 26th 2022 entry, and the sketch below). The śukanasa of the Lakṣmī -Damodara has a similar albeit narrower crack through the right lateral part of its śukanasa ‘s front – and the Chandra-Gupta Temple’s śukanasa displays abrasion which quite reminds on the Chamesan Champāvatī Temple.

Likewise, the main temple of the complex after which it had been named, the Lakṣmī-Nārāyaṇa Temple too, has a whole portion of its śikhara tower with fillings of comparatively new, non-original stone elements on its southern face. It is very likely that this repairs were carried out after 1905.

© Gerald Kozicz – Sketch drawing showing the crack along the śukanasa of the Gauriśankara Temple of the Lakṣmī-Nārāyaṇa, Graz, Austria, 2023

As for Kangra, J.P. Vogel provided a summary of the devastating event in the Archaeological Survey Annual Report 1905-6 (finally published 1909). Therein he provides a number of comparisons of states of preservation of monuments before and after the earthquake (Plate 2). His first account is on the Kangra Fort and Town and includes a description of two major temples inside the fort, which had been the strongest fortification of the Western Himalayan Hill States in medieval times. Vogel refers to the two temples as the Śitalā and Lakṣmī-Nārāyaṇa but also notes, that there were no actual hints to confirm the dedication to any of the respective deities as there were neither idols or inscriptions. The temples were standing next to each other in a yard close to the top of the fortified area facing North. Regarding their architectural concept and layout, these temples completely differ from the Chamba temples and this entry will therefore focus on a brief comparison between the two systems by having a look at the remaining rear wall, i.e. the wall portion from ground level to the kaṇṭha (recess below śikhara) of the larger of the Kangra Temples, the Lakṣmī-Nārāyaṇa Temple. 

On first sight, the Kangra temples look quite similar to the Chamba temples as the decorative patterns and motifs are widely identical – only at Kangra the patterns are even more elaborate. The ornaments were the dominant visual factors on the Kangra temples. Also, the major architectural elements of which the facade was composed of are widely identical, e.g. the design of the aedicules and niches. But otherwise there are significant differences. These differences primarily result from the fact that the Kangra temples had flat pyramidal roofs – not to be compared to the śikhara-like but pyramidal design of some small Chamba temples such as the temple near the Court (November 11th 2022 entry) or the pyramidal temple in Chauntra (forthcoming). The technology of the almost flat roof structure demanded a square plan for the large-scale temple. Thus, the first major difference between the Kangra Fort temples and the Chamba temples is the plan lay-out which does not allow the proliferation of the central bhādra field. It is also interesting to note the lack of attention this temple has received. The particularities of its floor plan have not been mentioned and one may suspect that this is because it does not fit into the stereotype categorisation on which previous discussions of temple architecture in the region have been based (s. e.g. Laxman Thakur The Architectural Heritage of Himachal Pradesh) – a categorisation which almost exclusively restricts itself on the counting of proliferations of niches and a widely incomplete listing of elements of the vedibhanda with its horizontal layers of different mouldings, i.e. those lower parts of the buildings that are easily investigated without any further technical efforts (see e.g. Thakur 1996). These purely descriptive lists of elements employ Sanskrit terminology that is problematic since sometimes it is not even confirmed as being original. Such discussions not only have an exclusive character. Above all, they have not yet resulted in substantial understanding of the ideological or functional – functional in a ritual sense – context or explain the architectural complexity of actual nagara temples in the region.

Despite its fragmented state of preservation, the last wall of the Lakṣmī-Nārāyaṇa Temple may at least serve as a starting point for a proper architectural discussion.

In accordance with our established workflow, a 3D-model could be generated from a set of photographs that were taken in passing during a short visit in 2016. Based on the model, which clearly shows the recesses and proliferation of the façade, an exact plan was prepared. The drawing now clearly shows how the proportional rhythm was established.

The Kangra temple contrasts the Chamba temples in regard to the structural and proportional design of the façade. While the elevation of all the larger Chamba temples have two-partite jaṅghās (façade of the chamber where the niches are placed) above the vedibhanda, the Kangra temple has three. The demand for an additional layer results from the central aedicule (i.e. the proliferating high-relief temple structure in the central field termed A (bhādra) in the elevation drawing). In the nagara architecture of Chamba, the central aedicule extends into the śikhara tower. Such extension is impossible in the case of a flat roof which is why the Kangra temple has three jaṅghās. Because of curvilinear shape of the śikhara, the available spaces on both sides of the amalaka and final were used for the implementation of two small aedicules.

The flanking fields (B-left and B-right) display aedicule which expand over two fields only. The third, top field was split into two and dedicated to two smaller aedicules.

The outer flanking jaṅghā field was clearly divided into three fields and display a vertical line of three aedicules.

The three jaṅghā levels are of different height, diminishing from bottom towards the top. To archive that, the number of layers of two elements of the pilaster structure was changed, to be more precise: the shaft rising from the bottom base-element bearing the vase-of-abundance motif. On the lower jaṅghā, the shaft measures three layers, then two in the middle and finally just one on top (indicated by colours in the bars on the left side of the drawing). One interesting feature concerns the cornice element which is found above the first jaṅghā but missing above the second.

A strange abnormity is the non-symmetric design of the two large aedicule of B-left and B-right, as well as the aedicule of the middle fields of the outer jaṅghā s (C-J2, left and right). In both cases, there is a difference in height of exactly one horizontal layer of brickwork. In addition, the C-J2 aedicule to the right has the cornice that is missing elsewhere on this level. Non-symmetry along the main axis of a temple – which is the case here since this was the rear wall – is usually a no-go in nagara architecture. This leads to the question about the reason for this abnormality. If the design is original, then it would hint at two workshops working on the same wall without proper coordination of works. Otherwise, the a-symmetry might also result from repair works carried out after the earthquake, i.e. a composition of re-used elements. It is not possible to answer this question from afar. Nevertheless, a careful study of the material data can at least lead to questions which might direct attention to the crucial aspects of the ruin and finally lead to a better understanding of the architectural history of these temples.



Vogel, J.Ph. (1909) Ancient Monuments of Kangra Ruined in the Earthquake. In Archaeological Survey of India Annual Report 1905-6, 10-27, Calcutta.

06.03.2023 – Neil Howard and his work on fortifications in the Western Himalayan Region: A tribute

© Gerald Kozicz – Fortification of Nurpur, Kangra, India, 2016


Neil Howard passed away in January 2023. He was a civil engineer by training. When trekking the Western Himalayas with his wife Kath in the early 1980ies, they became deeply fascinated by the rich cultural history of the region. Coming from outside the classical fields related to Himalayan studies (such as religious studies, linguistics, art history or Tibetan and Indian studies), they selected their own topics of research. While Kath focused on the early stupas of Ladakh and produced a seminal paper on the stupas (chörten, mChod-rten) of Ladakh, Neil immediately focused on the fortifications and castles. This topic was to become his academic mission in a way, as his interest would persist throughout his life. It made him the outstanding expert in the field, in particular in the region of Ladakh, and his results got soon published in an extensive article in East&West. Both Kath and Neil decided for specific premises during their fieldwork, one of which was communication. Their works were based on long-lasting relations with local experts and scholars, open-mindedness and kindness, always seeking constructive discussions and willing to share their material. Neil’s interest was not only the local fortresses of Ladakh, but also the military camps of the Dogra’s who invaded Ladakh in the middle of the 19th century and whose camps he accurately tried to trace.  Less known is Neil’s research on military structures and fortifications outside Ladakh such as the fortified camps erected by the Gurkhas during their campaigns into the Himachal region. He also did field research on the fortresses of the local kings in Himachal. One fortress he was particularly attracted to was the ruined fort of Nurpur, a fortified royal palace in Kangra, that had been in use as a fortification until the 19th century. Neil spent more than a week at the site, studying the remains of the broken and dilapidated walls and towers when the place had partly been turned into a school yard and park for the local people of the small adjacent villages. Neil took measurements by pacing the rocky area – a method he was well aware of as neither being highly accurate nor time efficient. While generally he used this method as a measurement, for single smaller objects, like towers, he took a more precise approach and used a scale. He basically decided to use whatever worked best under the given circumstances.

In 2013, Neil forwarded his notes to me with the request for a digital site sketch map. This map was meant to be incorporated into a paper on the Nurpur Fort which he was going to write. Neil particularly focused on the southern part of the fortification walls since the northern and western sections were well protected by steep cliffs overlooking the Jabhar Khud, a tributary to the Chakki River, that runs east-to-west north of the rocky spur on which the fort was built. The southern and eastern sections were thus the strongest fortified, and this is also where the gates had been erected. The other remains and ruins along the plateau were of lesser interest to him, which is why he marked the areas of the plateau where there are still remains of palace architecture simply as “debris”.

Neil’s paper addressed an unusual subject and it turned out difficult to find a journal that would accept such a topic which was definitely outside the mainstream. This paper never got published and Neil’s work on Nurpur has never been made accessible. The map based on Neil’s notes might not meet up with modern standards of cartography, but his observations on the castle reflect expertise and a deep understanding of the functional logic behind military structures. His original field research is unique – and the architectural evidence might have faded away in the meantime. This makes his notes an invaluable contribution to the studies of the cultural heritage of Himachal.

Neil Howard was a true British gentleman scholar with a good sense of British humour, a highly respected colleague and a wonderful host to everyone who visited him and his wife and collaborator Kath at their home in Birmingham. An appreciation by John Bray is also accessible at the Ladakh Studies website.



Howard, Neil (1989) “The Development of the Fortresses of Ladakh c. 950to c.1650 A.D.”East & West 39, no. 1-4, 217-288.

Howard, Kath (1995) “Archaeological Notes on mChod-rten in Ladakh and Zanskar from the 11th-15th Centuries”. In Osmaston, Henry and Philip Kenwood (eds.) Recent Research on Ladakh 4&5, 61-68. London: SOAS Studies.


27.02.2023 – The Inner Portal of the Śaktidevī Temple of Chatrarhi:
A Critical Note on Methodology of Field Research:

© Gerald Kozicz – Inner portal of the Śaktidevī Temple of Chatrari, Chamba, India, 2019


Besides the inner portal of the Lakṣaṇā Devī Temple of Brahmapura, the inner portal of the Śaktidevī Temple of Chatrarhi is the second original wooden doorframe from the very early phase. Dated to the ca. 8th century, the Chatrarhi portal displays close similarities with the outer portal of the Brahmapura Devi Temple regarding composition, style and construction methods. This entry however, is not about the art historical relevance of the portal itself, but rather about its historic discussion as an academic subject.

Chatrarhi itself is a place not easy to get to today. It lies half way between Brahmapura and Chamba Town and to get there one has to take a single lane road that climbs along the steep slopes from the Ravi ravine to the small plateau where the hamlet and its compound are located. Climatic conditions – monsoon in summer and snow in winter – and landslides make travel impossible during most time of the year, which is just one reason why this temple has not yet been surveyed completely. Whoever wants to study its architecture and art but has no chance to see the site itself naturally has to rely on publications as it is the common practice in such case. This is of course academic practice, but the Chatrarhi portal may serve as a useful example to exemplify some critical issues that come with that – issues that become very obvious when one examines some of the previous discussions of this masterpiece in detail. 

The first study that appears as a comprehensive discussion of the portal was done by Hermann Goetz and published in his The Early Wooden Temples of Chamba (1955: 87), a publication referred to regularly in every major publication on the art and architecture of Chamba. Goetz begins his discussion of the multi-layered frame from the outward jamb towards the opening but provides no number of the frames but just mentions “the next frame”. He mentions three standing deities on each of the outer jambs plus kneeling gaṇas in minor compartments that separate these four fields (which would be the second layer). He identifies Karttikeya, Indra and (tentatively) Śiva on “the left side” and “Brahmā on the right side”. Now this is correct if the portal is viewed from the perspective of Devī, i.e., from the innermost sanctum. For the viewer’s perspective, Brahmā is on the left and the previously mentioned deities are on the left, i.e., the directions are reversed. No other deities are mentioned on “the right side” here. No position is given for the four deities noted, i.e., the reader is not provided with any information about at which level the deities are positioned and who mirrors whom. Ignoring the “next” jamb which displays ornamental imagery, Goetz immediately continues with the no.4 pair of jambs which contain four standing deities each. He starts on the left jamb at the bottom and identifies the three figures as (tentatively) Vāju or Yama, Durgā Mahiśamardanī and Viṣṇu. Goetz writes that he starts from the bottom, but the figure of Gaṅgā, the River Goddess is then mentioned to be placed below the three deities mentioned above – i.e., Goetz contradicts his own descriptive structure. But what is even worse and more confusing is the fact that this jamb, labelled as “on the left” is at the same side as the outer jamb displaying Brahmā – previously labeled “at the right side”. Goetz changes the perspective – or more likely just confuses right and left. Then Goetz continues on “the right side” and lists “an unidentified goddess(?), a god with a club (Bhairava?), again a god or goddess, and finally the river goddess Yamunā”. One may conclude that the mention of Yamunā at the end points at a list top to bottom, but actually one cannot be sure since the order was confused on the opposite side as well. The publication does not provide any photographs.

A comparatively clear photograph of the portal is provided by Postel, Neven and Mankodi in their “Antiquities of Himachal” (1985: 45-46, fig. 45 and 48). Unfortunately, the left (viewer’s perspective) jambs are not depicted and the details are naturally hardly visible in the total image. The authors provide the exact number of jambs and lintels. They mention Gaṅgā and Yamunā on the “third band” and one may conclude from that – with the help of the photograph – that they start their counting from inside towards the outside. “On the left” they further note Viṣṇu Caturārana, under him Mahiśamardanī followed by a two-armed deity with a ringed club. They hesitate to suggest identifications for the damaged figures on the opposite side. The figures on the outer figurative jamb which would be the “fifth band” do not find any mention in their description, leaving the discussion incomplete. Due to this unexpected omission, the mistake made by Goetz remains undisputed.

Cinzia Pieruccini, who discusses the portals of the Laksanadevi Temple of Brahmapura extensively and with great detail, uses the Chatrarhi Temple as the major comparative monument. Unfortunately, when it comes to the portal she confines her discussion to style, e.g., by comparing the “lions” (1997: 217), and completely avoids the topic of iconography and composition. This is actually surprising since she does not avoid that topic in her discussion of the outer Brahmapura portal. Thus, Goetz’s analysis remains unquestioned once more.

O.C. Handa provides another contribution to the discussion. His study is published twice at least – almost verbatim – in his Temple Architecture of the Western Himalaya: Wooden Temples (2001: 165-66, pl. 22) and Ancient Monuments of Himachal Pradesh (2010: 54-55, pl. 12). His description of the jambs of the six-layered frame starts from the outside towards the inside. He begins with (o)n the left vertical, from bottom upwards are Karttikeya, with six faces and peacock; Indra; and possibly Shiva with his mount, according to Goetz (2001: 165). He next mentions Brahmā on the “right” side. But then he continues with the “left” side where he locates Gaṅgā, Yama (or Bhairava), Durgā and Vaikuṇṭha from bottom to top correctly. This exactly reflects Goetz*s confusion of the orientation. This is also striking because Karttikeya‘s six heads are among the very few details of all the imagery along the whole frame that can be identified even on the low-quality reproduced images of the frame – and these six heads are not with the bottom figure but the top figure of this element as can be clearly be seen on the photograph (2001, pl. 22) in the same publication – which shows the figure on the right side (viewer‘s perspective). It must also be recalled that Goetz does not provide the vertical order as quoted by Handa. For the inner element with the figurative imagery Handa follows Goetz’s switch of perspective. Handa still provides a few new details such as the Viṣṇu Vaikuṇṭha, previously termed Viṣṇu Caturārana by Postel, Neven and Mankodi. Such identification should be the result of his in-situ inspection – which triggers the question why in his published study the mistake made by Goetz is not critically reviewed and corrected.

Finally, Sangram Singh in his The Art of Mountain Temples (2015: 118-22) provides a clear, well-structured discussion of the portal. The reader is well guided through the complex six-layered composition which is described in detail layer by layer. Singh maintains a clear orientation keeping “right and left” constantly in accordance with the viewer’s perspective. He also contributes to the discussion by suggesting the tentative identification of the two decayed figures as forms of Viṣṇu.

To sum up, the discussion by Sangram Singh provides us with a solid description of the portal’s jambs as complete as possible given the state of preservation. But this not yet brings the discussion to a satisfying end, as a few questions now arise. How is it possible to ascertain that Singh is correct? Unfortunately, the reproductions of his photographs (ibid: 204-205, pl. 6.10 – 6.13) are rather poor and the identification of details almost impossible. He also avoids mention of the confused orientation found in all previous publications noticed above. It is impossible for a reader who is without a proper personal documentation at hand, to evaluate the correctness of each author. It is also important to ask now whether it was the status of “Godfather of Chamba temple studies” held by Goetz that hampered critical approaches towards his work?

Whatever the reason, this blog entry provides some material that will support the analysis by Sangram Singh. The visual material is confined to the jambs so far and will be completed at a later point.

The reader of this present summary of earlier publications might easily get the impression that this entry is about criticising or even dismantling the fame of Hermann Goetz. This is actually not the case in the first place. The actual topic is the question: How could such mistake happen – and how could it have been avoided or corrected properly?

First of all – and this is critical after all – the description by Goetz as such is confusing even where the directions are correct. But then, how did the mistake happen in practice? And why did no one notice the mistake and correct it? Both questions are related to a single problem and this is viral: Historians – including art historians – do not sketch. Everyone makes mistakes. But such mistakes do not remain unnoticed while sketching. This process of reproduction means permanent interaction between draughtsman and object – be it a stele, an interior space, an architectural object or even a landscape. This interaction is a slow-motion process of appropriation of the qualities of the object through the capturing of the significant features on paper. It is a permanent process of filtering, evaluation, perception of structure – a method of understanding that cannot be replaced by photographs or textual notes. Mistakes on the paper become immediately apparent and trigger corrections. Sketches are reflections of a cognitive process – and as such they also work as media to convey the result of a survey to the reader. They are not perfect in a technical sense like a photograph. But sketches are visual studies and they have the potential to communicate the result directly.

Trying to understand and creating a picture of the whole – both metaphorically as well as visually – from Goetz’s description is a frustrating process that does not yield any useful result. A simple sketch would be of tremendous help. And if Goetz would have produced one, he would probably have realized his mistake immediately and provided all following researchers with a correct and solid basis for further studies. Even Singh’s analysis can only be properly understood and appreciated if one produces a sketch from his notes. Otherwise, the complexity of the door frame is almost impossible to imagine.

This blog is of course primarily dedicated to digital media and methods of presentation of the cultural heritage of the Himalayan region. However, the sketch is at the very beginning of this process which starts with field research and documentation.

© Gerald Kozicz – Line drawing of the deities along the jambs plus the figure of Lakṣmī in the lalatabimba of the first lintel, Graz, Austria, 2023

L: Lakṣmī

L3-I: Gaṅgā

L3-II: Bhairava(?)

L3-III: Mahiśamardanī

L3-IV: Vaikuṇṭha

L5-I: Yama(?)

L5-II: Śiva(?)

L5-III: Brahmā

R3-I: Yamunā

R3-II: Nandin(?)

R3-III: Deity with club (Viṣṇu?)

R3-IV: Deity with club (Viṣṇu?)

R5-1: Maheśvara (?)

R5-II: Indra

R5-III: Karttikeya



Goetz, H. (1955) The Early Wooden Temples of Chamba. Brill, Leiden.

Handa, O.C. (2001) Temple Architecture of the Western Himalaya: Wooden Temples. Indus Publication, New Delhi.

Handa O.C. (2010) Ancient Monuments of Himachal Pradesh. Museum of Kangra Art, Dharamshala.

Singh, Sangram (2015) The Art of Mountain Temples. Agam Kala Prakshan, Delhi.

Pieruccini, C. (1997) ‘The Temple of Laksana Devi at Bharmaur’, East and West 47, nos. 1–4, 171–228.

20.02.2023 – Orphaned Objects 2:
A collection of stone works assembled in the Chatrarhi courtyard:

© Gerald Kozicz – Courtyard of the compound of Chatrarhi, Chamba, India, 2016


The significance of the courtyard of Chatrarhi Temple complex has widely been overshadowed by the main sanctum of the compound with its wooden architecture and the main idol of the goddess inside. The courtyard appears rather insuspicious on first sight. What appears unusual is the different level of the courtyard area. While the temple itself is on roughle the same level as the surrounding open space, one has to descend to a lower level in front of the temple in order to climb some stairs again to reach the outer portal.

On the opposite side of the lower terrain, a platform was erected. On the platform and facing the sanctum, a bull is placed in the most prominent position. This sculpture is certainly of an early age. It may hower be doubted that is is original since the appropriate vahana (animal mount) in front of a Devī temple would be the lion.

Otherwise, the platform has become another collecting point for orphaned stone images. these images are quite randomly arranged. They include several lingams and curvilinear-shaped cones, a kind of saurapīṭha slap centering on a lotus flower with a goose having a pearl-chain in its beak, and several steles from different periods. One figure literally stands out from the collection. It is a standing figure of a male holding a spear in the right hand. The figure has a large circular halo that shields half of its back. The style reminds on the two dvārapālas (door-keepers) of Lākhamaṇḍal discussed by T.S. Maxwell (1980:15-18, fig.4 and 5) and dated on stylistic grounds to the 7th-8th century, i.e., the post-Gupta era. The figur at Chatrarhi differs from the Lākhamaṇḍal sculptures since they lack the halo and lean against clubs. The halo hints at a deity and the spear are the major attribute of Skanda-Kārttikeya/Kumara, the warrior god of the Brahmanic pantheon and son of Śiva. The muscular chest and the posture with the left hand on the waistband mirror images of this deity elsewhere, e.g., the two-armed Skanda from Apsidal (Rangarajan2010, Fig. 39). Still, there are several features that would be expected with a depiction of the warrior god which are absent from this stele. First of all, Kārttikeya‘s animal mount, the peacock, is not depicted. The bird is usually depicted behind the legs of the deity when standing or under the deity when seated. In this case, the deity is just standing on a square pedestal. Second, Skanda-Kumara’s typical hair-do that mimics the two horns of the ram, are also absent. Instead, the figure wears a bejewelled hat-crown. Finally, the necklace of tiger-claws that is sometimes worn by Kārttikeya is not there, either. Still, what is depicted in this stele clearly hints at Skanda-Kārttikeya/Kumara. Even if the figure would display a different deity or semi-god, its art historical significance would not drop. This stele is among the oldest stone steles of this part of the Western Himalayan Hills, if not the oldest.



Maxwell, T. S. (1980) ˋLākhamaṇḍal and Triloknāth: The Transformed Functions of Hindu Architecture in Two Cross-cultural Zones of the Western Himalayasʻ. In Art International, Volume XXIV/1-2, 9-74.

Rangarajan, H. (2010) Images of Skanda-Kārttikeya-Murugan: An Iconographic Study. Sharada Publishing House, Delhi.

13.02.2023 – Dressed with Conches: Attire unique to the Ears of Gaṇeśa:

© Gerald Kozicz – Gaṇeśa of the Baśeśvara temple of Bajaura, Kulu, India, 2017


While attributes and physical aspects such as body colour, number of arms and heads as well as postures and mudras may be considered the major categories of visual language in Indian iconography – Hindu/Brahmanic, Buddhist and Jain – there are also minor elements which are distinctive and may refer to specific groups or even individual deities or demi-gods. These include certain dresses but also jewelry and attire. The necklace of claws worn by Skanda/Kumara/Kārttikeya and also by the Buddhist bodhisattva Manjusri is a prominent example.

This entry addresses a tiny detail in the iconography of Gaṇeśa, a detail which is connected to the ears of the elephant-headed god and thus related to his specific physiognomy. It is a pair of conches that hang right in front of the opening to the ear canal. It is the visual similarity between the shape of the conch and this part of the human ear that has resulted in the term „Ohr-Muschel“ (lit. „ear-conch“) in German, and thus catches attention especially by a someone speaking German. These conches do not appear in every depiction of Gaṇeśa. Since the conches constitute for quite a tiny part of the god’s attire, they can not be shown in small sculptures or carvings but only represented within large-scale works. In Chamba, they have so far only been detected on the Gaṇeśa stele embedded in the compound wall of Swai. They are also found on steles of Gaṇeśa in neighbouring regions. One such example is the well-known Gaṇeśa enshrined in the southern chamber of the Baśeśvara temple of Bajaura in Kulu to the east of Chamba. Another, further to the South-East but still in Himachal, is the Gaṇeśa in the southern niche of the Temple III at Parahat near Hatkoti, a Śiva temple that clearly displays post-Gupta stilistic elements and may be approximately dated to the 8th century. The conches as elements of Gaṇeśa’s jewelry are not limited to North-West India. Several specimens are now in various museums in Bangladesh and published in Sculptures from Bangladesh (Haque and Gail 2008: 516-520, Pls. 432, 435, 436, 44 and 446). The catalogue entries however do not make any reference to the conches.

The wide-spread appearence of the conch-jewelry signals a well-established practise but does not yet inform us about a specific meanic or symbolism of these tiny members of the iconographic programme. Did they have any meaning at all? Or was it just convenient to place them there – a matter of visual relation between physiognomy and shape, which in turn invited their inclusion? The conch is otherwise clearly related with Viṣṇu and also with Durgā, who receives the weapons from all the male gods when she sets out to fight the asura armies. The conch (śanka) also forms a pair with the lotus (padma) as niddhi (little treasure), i.e. Śankaniddhi and Padmaniddhi. A connection with niddhi symbolism appears far-fetched for the ear jewelry but can not be completely ruled out since Gaṇeśa is also related to aspects of abundance and wealth.

For the time being, the pair of conches as attire to the ears of Gaṇeśa may be considered a secondary element of the god’s iconography that was known across all over Northern India at least.



Haque, Enamul and Adalbert Gail (ed.): “Sculptures in Bangladesh”. The International Centre for Study of Bengal Art, Dhaka (2008)

06.02.2023 – Orphaned objects 1: Four Gaṇeśa steles and one broken Durgā:

© Gerald Kozicz – Gaṇeśa Mahiśāsuramardanī at the Lakṣaṇā Devī Temple of Brahmapura, Chamba, India, 2019


The term “orphaned” has been coined to define the decontextualized status of a cultural object, mainly in the field of museology when the provenance or original setting to which an object once belong is unknown (Motoh, forthcoming). It appears however, also as a perfect term to describe the situation all over Northern India where steles and sculptures have been removed from broken temples the traces of which have sometimes completely disappeared. Little information is sometimes available about the history of single objects, where steles and sculptures had been originally placed, which configuration they had been part of, where they had actually even come from exactly. Very often, for example, when temples collapse or when stone steles simply resurface from the ground in the course of construction activities, these steles are either arranged under sacred trees or simply taken to the next intact temple and enshrined either in the sanctum or one of the bhadra niches. Sacred trees are also inside temple compounds, and many temple complexes have thus become hoards of orphaned steles. Many of these steles show severe traces of decay and have often escaped scholarly attention.

One such example is the Gaṇeśa inside the Gauriśankara Temple of Chatrarhi where the stele of the elephant-headed god is placed against the left side wall (viewer’s perspective from the entrance). The four-armed deity is partly covered by a textile and shows quite some traces of abrasion. Another Gaṇeśa stele is placed inside the Śiva temple of Swai. The stele is not as fine as the Gaṇeśa stele embedded in the courtyard wall (see the 01.08.2022 entry). The elephant-headed son of Śiva and Pārvatī is four-armed, holding axe and radish in the right hands, and a cup with sweets and a lotus(?) in the left hands. His head is decorated with an orange flower – and a flower is also carved above his head into the large round backslab (nimbus?), just like the flower above the head of the Gaṇeśa outside. In a comparatively poor state of preservation is the Gaṇeśa inside the Lakṣaṇā Devī Temple of Brahmapura. The god is again shown four-armed. At least the axe and the sweets can be identified with certainty. This stele is placed next to a broken stele showing Mahiśāsuramardanī. This image of Durgā differs from the larger steles presented in the previous entries since it shows the deity with the right foot on the demon’s bull head and grabbing the tail, thereby lifting the demon’s body. Both the steles show clear remains of orange colour.

Touching the foreheads of deities, the attributes and the breasts in the case of female deities is an act of religious veneration in daily ritual. However, in the case of Gaṇeśa the application of orange colour all over the body seems to have been an old tradition. With new paints available on the marked this tradition has however caused quite some damage to stone sculptures. Another modern fashion is the application of silver plates on the eyes – a fashion that is not yet that popular in Chamba as it is in other regions of India such as Bihar. A rare example from Chamba is the four-armed Gaṇeśa inside the dilapidated nagara shrine of Rajnagar.

The application of orange paint on Gaṇeśa images is also eye-catching on door frames. It is a regular feature since Gaṇeśa is in the central position of the first lintel of almost every temple affiliated with the cult of Śiva and Pārvatī. Even inside the Lakṣmī-Nārāyaṇa Complex, the steles of Gaṇeśa have been covered with thick layers of paint. These layers sometimes even completely mantle the details of the iconography of the respective figure and the identification of a Gaṇeśa is only possible through the shape of the head and the trunk.



Motoh, Helena (forthcoming) Orphan(ed) Scroll: the Case of Contextualizing a Late Qing Object in a Slovenian Museum. In Ming Qing Yanjiu 24 (2020) 139–158. 

30.01.2023 – Kangra, Kulu and Shimla – Regions category added:

This weeks update sheds some light on the geographical situation as well as the correlation between the different regions existing along the southern face of the Himalayan chain. The newly added Regions category is meant to provide an overview of the different regions discussed in this blog and serve as a guide for our visitors accross the different topics and articles discussed on this website.

The new category can either be found by clicking on it in the top menu or directly by clicking HERE

Kangra – a hub for the transmission of the artistic tradition of Mughal art:

© Gerald Kozicz – Kangra, India, 2018


The Dhauladhar Range is the geographical order between Chamba and its southern neighbour Kangra. Kangra is by far less mountainous than Chamba and its ties to the major cultural centres of Northern India were much closer than those of the remote valleys of Chamba – which of course made it more vulnerable in times of war and migration. In contrast to Chamba, Kangra got exposed to severe pressure from the West when Muslim invaders from the Ghazni Region of Afghanistan raided the region from the turn of the first millennium onwards. During the Mughal period it got under control of the central government which of course had a major impact on the art and architecture of the Mughal Court. Kangra served as a hub for the transmission of the artistic tradition of Mughal art to Chamba where it was adopted especially in painting and the decorative arts in temple architecture.

In the 8th and 9th centuries CE, Kangra was probably part of the Gurjara Pratihara Empire of Kannauj who was in a competitive conflict with Kashmir (Meister 2006: 45). The rock-carved temple – never completed and badly damaged in the Kangra Earthquake of 1905 – bears witness of their political power and religious agenda. The list of major religious monuments inlcude the temple of Baijnath, and the fortifications of Nurpur and Kangra Town.