Architecture and Space Design

By Gaudenz Domenig

Although in the last decade considerable progress has been made, the present state of research regarding these traditions is characterized by many blind spots and by grave insufficiencies, both on documentation and in making use of the source materials that are available in libraries and in museum collections. This is the more deplorable, as many of these traditions are now in the course of being abandoned in exchange for modern habitation styles. Recognizing the urgency of acting before it is too late, the Institute of Cultural and Social Studies of Leiden University plans to start a four-year research project in August 1995 entitled 'Design and Meaning of Architecture and Space among Ethnic Groups of Western Indonesia'. Based on a Dutch-Indonesian cultural agreement, the project is to be realized in co-operation with Indonesian counterparts. Supervision and co-ordination on the Dutch side will be in the hands of Professor R. Schefold (anthropology) in collaboration with Dr. P. Nas (urban sociology), both of the Faculty of Social and Behavioral Sciences of Leiden University.
The various research units, to be entrusted partly to senior researchers, partly to post- doctorates and to PhD candidates, will be devoted to the comparative study of settlement patterns, to linguistic aspects (e.g. building terminology), and to extended field researches in specific areas of Sumatra and western Java. The latter researches will focus on spatial, social and symbolic aspects, not only of traditional houses and settlements, but also of a city (Palembang). Although the subjects will be approached mainly from an anthropological perspective, it is planned to profit from the collaboration of researchers and experts in other fields, such as architecture, urban sociology and linguistics.

New hypothesis
Dealing with architecture in particular, the material and technical aspects will not be neglected, as they have so often been in anthropological studies of the house. Thus it is for instance intended to test, in the context of comparative studies a hypothesis that has recently been introduced independently by different scholars. It regards the evolution of architectural styles in Indonesia, claiming that many house types in this part of the world have originally been developed from relatively simple pile-structures such as used for the storage of rice.
This hypothesis is of considerable anthropological interest because it might lead to a more adequate recognition of the importance of the granary-type of structure, not only as a probable core element of early Austronesian cultures, but also as a building associated with various symbolisms, and perhaps even as the birth-place of some ides and rituals that later came to be transferred to the house -- where they may then have been changing in association with new elements, depending on how the house took form and how it was changed itself in the course of time.

Toba architecture
It is in connection with the above research project that a senior visiting fellowship of the International Institute for asian Studies (IIAS) allowed me to do library research in Leiden in February and March 1995. The main purpose was to study sources on vernacular architecture of western Indonesia with a view to evaluating the materials in a comparative and diachronic perspective. Although it is of course not possible to report in a few words what this study has produced, I will try to introduce one particular finding which relates to the above-mentioned hypothesis.
The Toba Batak of North Sumatra are now known to have two traditional types of dwelling house. Apart from the rumah, the dwelling of the old style, they have, probably since the nineteenth century, developed a second type that was initially produced simply by adding walls to the formerly open space below the roof of a granary, or sopo. As with as a sopo-granary in which the attic-floor had to carry the load of the harvest, this new type of building, although in form quite similar to a rumah, had a different structural system. The roof-supporting posts were placed, not under the ends of the attic-beams, as with the rumah, but a considerable distance further in towards the middle. This feature was usually preserved when a sopo was transformed into a dwelling house of the newer kind, and so the position of the posts has become one of the criteria that allow identification of the sopo-type dwelling even in cases where its architecture has otherwise been much modified in imitation of the older rumah-type.
So far the matter is probably familiar to any serious student of Toba architecture. What seems to have hitherto escaped attention is that there also exist sources telling us that even as an open building the sopo was occasionally modified, and in a rather interesting way that made its structural system similar to that of a rumah. In such cases the abandonment of the rice-storage function led to a simpler construction of the attic-floor and, combined with this, to a new placing of the columns nearer to the eaves of the roof. It is easy to imagine that a sopo of this different type could be equipped with walls and further modified so as to resemble a rumah more and more. However, as in this case the two types of dwellings would be similar not only in form, but also in construction, and this possibility opens up a totally new perspective: it raises the question whether the recent Toba custom of transforming granaries into dwellings might have had an older precedent in the earlier history of Toba architecture, in the sense that originally the rumah-type would have been derived from a sopo-type, but in a different way, resulting in its different structural system. To understand the Toba rumah in this way would be in harmony with the above- mentioned hypothesis, but whether or not the new idea will be tenable, depends of course on what further evidence future research can provide in its support. The new research project on architecture and space design in western Indonesia is much wider in its scope, but it should provide a good framework for dealing with this kind of problem.