Stretched across the Equator between the Southeast Asian mainland and the islands of Australia and New Guinea lies the world’s largest archipelago. Home to over 230 million people, this archipelago is now divided politically into five independent states, Indonesia, Malaysia, Singapore, Timor Leste (East Timor) and Brunei, while islands belonging to six neighbouring states – the Philippines, Thailand, India, Australia, Papua New Guinea and Palau – lie on its fringes.

Indonesian archipelago in its Asia-Pacific context

In size and population, the largest state of the region is Indonesia, founded in 1945 in the turbulent aftermath of the Second World War and based on the territorial outlines of the former Netherlands East Indies. Indonesia’s history since 1945 has been complex, with events spilling across the country’s borders on more than one occasion, and its history before 1945 is still more tangled. The Dutch colony was created by a long process of overthrowing and incorporating indigenous states, themselves based on widely varying social, religious, economic and cultural structures.

Main islands of the Indonesian archipelago

The purpose of this atlas is to present the history of the Indonesian region in map form, focussing on the territories which eventually became part of the Indonesian state, but paying attention also to regions – Malaysia, Singapore and Brunei – which were culturally and historically part of the Indonesian archipelago and which did not become part of Indonesia. Like all historical atlases, it is aimed at more than one audience: it is intended both as an introduction to Indonesian history for those especially interested in its geographical dimensions, and it is intended as a reference work for more established scholars in the field. In either case, it need hardly be said, the atlas is primarily a supplement to the many written works which illuminate various aspects of the archipelago’s history.

Western Indonesia: major geographical features

Eastern Indonesia: major geographical features

Maps of the Past

The use of maps to present the past is, of course, at least as contentious as the use of words. The sharp lines and edges which are characteristic of maps cannot easily convey degrees of imprecision and uncertainty in the way that prose does. Maps therefore often seem to claim a greater degree of accuracy than is warranted by the empirical knowledge on which they are based. Still more seriously, historical atlases have a tendentious pedigree: particularly where their focus is an existing national state, they often appear to annex the past to the present, claiming the legitimacy of antiquity for what may be much more shallowly rooted political units. The aim of this atlas – by starting with continental drift and with the region’s human inhabitants of half a million years ago – has been to let the reader see how different elements of what we now identify as ‘Indonesia’ gradually assembled, not to imply that there was some teleological necessity in the precise pattern they took. Indeed, one of the tests of the atlas will be how much its maps continue to inform readers in the future as these elements gradually reassemble in the processes of political, social and economic change.

The first historical atlases – standardized maps of a particular region, chronologically arranged and bound in a single volume – appeared in the 17th century. They were conceived as a supplement to written histories, and for a time the development of historical atlases paralleled the development of the historical discipline. Particularly from the early 19th century, as historians began to turn their attention from rulers and warfare to economic and social history, cartographers added thematic maps showing phenomena such as language, climate and economic standing to the once-standard fare of political boundaries and the movements of armies and travellers. Some historical atlases are still conceived as an adjunct to the whole body of historical writing on a region, but most now adopt the practice, used here, of providing a substantial text to link the maps, to provide context, and to say those things which no map, however well conceived, can say.

The question of what maps say and do not say has become a matter of some contention in recent years. In the first place, a map is always much more a snapshot in time than is historical prose. Even if the patches on a map are colour-coded to indicate different periods and the map is provided with arrows to show movement, and even when a series of maps is set in historical sequence, the effect is jerky. Few maps have a beginning and an ending in the manner of a prose passage: they demand instead to be explored, pored over, and revisited in a way that prose does not. A good map stops the reader dead in his or her tracks, subverting the narrative drive which lies at the heart of most history-writing.

A more important objection, however, arises from the fact that historians have become aware of limitations of maps as a vehicle for presenting some kinds of historical analysis. A picture may be worth a thousand words, but not every thousand words can be turned into a picture. These limitations arise especially from the fact that the information on which most historical maps are based is derived one way or another from the activities of the state – and this extends even to the outline of coasts and rivers, which is almost always the product of government-sponsored surveys. Can a map record structures of power without also embodying and legitimizing them? This question is particularly important in mapping regions which were once European colonies, for mapping was very often an integral part of both conquest and exploitation. ‘Give me a map’, says Tamburlaine (Tamerlane) in Marlowe’s poem, ‘then let me see how much is left for me to conquer all the world’.

The data which would be needed to present non-official views of the world may be largely or wholly lacking because those views are not of interest to those who sponsor the collection of official data. Sometimes it is possible to use official or officially-sanctioned sources as a mirror to read the minds of subordinate or resisting groups, but much of the material which historians have used in this way is geographically fragmented and not amenable to mapping. Women in the World: an International Atlas (Seager and Olsen 1986) is an imaginative attempt to use mapping technology to bring the experience of women to the fore, but it is able to do so only by using the nation as the main unit of analysis and at the cost of leaving large areas coloured in grey to denote that their status is ‘unknown or unclear’. The technology of mapping, moreover, is geared to precise classifications and lines of demarcation of a kind that suit bureaucracies. Transitional conditions and multiple meanings of the same reality are difficult – and sometimes impossible – to render clearly in map form. Like most modern authors of historical atlases, I have sought to make imaginative cartographic use of the material available to me, but most readers will soon notice gaps in the coverage of this atlas which cannot be adequately filled at present. Some critics have gone further in their deconstruction of maps to argue that maps say more about relations of power – political and intellectual – at the time of their creation than about their purported subject. To mark on a map, for instance, the extent of an early kingdom conjures up modern Western conceptions of kingship and territoriality which may be alien to the way in which people of the time viewed their world. To centre a map on an area of land, making the seas a blank periphery may have more to do with modern Western terrestrial conceptions of what is important than with the world-view of people living in that region. These critiques are a valuable corrective to sometimes-held perception that map-making is a technical and value-free exercise, but I confess that I have found the techniques of post-modernism more effective in deconstructing maps than in constructing them.

Indonesian archipelago and southern Europe on the same scale

Indonesian archipelago and the Caribbean region on the same scale

Earlier Atlases

This is not the first historical atlas to depict the Indonesian region. The western parts of Indonesia in particular have frequently appeared in historical atlases of the world or of neighbouring regions. In many such works, Indonesia is given no more than a peripheral place, but some provide an admirable level of detail, notably the Correlated History of the Far East: China, Korea, Japan (Penkala et al. 1966), the Historical Atlas of South Asia (Schwartzberg 1978), the Historical Atlas of Islam (Brice 1981), and the Times Atlas of World History (Parker 1993). In 1928, a historical atlas of the Netherlands (Stapel 1928) published six large, unbound maps showing the archipelago between 1619 and 1791 along with the Asian operations of the Dutch East Indies Company (VOC) during the same period; these maps were accompanied by a substantial text, but were too large themselves for easy use as a reference atlas and were very much focussed on the Dutch presence in the region. The great Atlas van Tropisch Nederland (1938) included a single opening with four detailed maps showing early kingdoms and the expansion of Dutch authority, but these had no associated text and were thus difficult to use.

Main island groups, seas and straits of the Indonesian archipelago

After the Second World War, Sellman’s rather crude Outline Atlas of Eastern History (1954) was soon overshadowed in detail and quality by Muhammad Yamin’s 1956 Atlas sedjarah (Historical atlas), a fine work which was taken less seriously than it should have been because of Yamin’s reputation as an unashamed historical propagandist. In 1964, the Dutch publisher Djambatan produced an Atlas of South-East Asia with eight historical maps on the end papers and a short general historical text by the renowned historian of Southeast Asia, D G E Hall. There was then a rather long interval before the appearance of Tugiyono’s 1982 Atlas dan lukisan sejarah nasional Indonesia (Atlas and illustrations of Indonesia’s national history), prolifically provided with simple, mainly hand-lettered black-and-white maps. This volume went through at least three editions before being superseded in Indonesian bookshops by Latif and Lay (1992). In 1985, the Indonesian Department of Education and Culture embarked on an ambitious programme to publish a series of historical atlases of Indonesia’s provinces, but unfortunately the value of these large volumes was reduced by the rigid use of a single base map for each province, by what appears to be an excessively precise delineation of early states, and by a coyness in dealing with recent, politically sensitive events. The Atlas of Southeast Asia (Ulack and Pauer, 1989) also contained a brief series of highly generalized historical maps.

Two works in recent times have gone beyond this relatively unimaginative pattern of historical maps. Hatley (1984) published a stimulating paper entitled ‘Mapping cultural regions of Java’ and provided several examples of how it might be possible to bring alive the diversity of Javanese civilization using maps to show patterns of language use, consumption, dress and so on. Pluvier’s magisterial Historical Atlas of South-East Asia (1995) vastly exceeded every preceding work in scope and detail. Its 64 maps offered a sequence closely spaced in time so that for the first time scholars could trace historical events from one map to another. Its rich detail, especially for the 17th to 19th centuries made it a valuable reference work for these centuries. On the other hand, prepared by traditional cartographic methods over more than twenty years, the atlas was unfortunately unable to take account of recent findings, especially on the prehistory of Southeast Asia; nonetheless, it forms a benchmark against which later atlases will be judged.

Major changes in place names, Western Indonesia

Major changes in place names, eastern Indonesia.

Terminology and Spelling

The places covered by this atlas have been referred to in the past using a bewildering variety of names and spellings. Indigenous place names changed over time; Westerners sometimes adopted and transformed local names which they heard, sometimes imposed one or more names of Western origin; and governments since independence have changed place names for practical and political reasons. Maps 0.9–0.10 show some of the most important changes and indicate the scale of the alteration in place names over the centuries.

For other maps in this atlas, the general rule adopted is that names current at the time referred to on each map are used but that these are presented according to the present system of spelling, which came into force only in 1973. Before 1973, the spelling system showed considerable Dutch influence, though Dutch spelling itself only became more or less standardized in the late 19th century. For the most part, the reader will not find non-standard spellings in this atlas, but a brief outline of the main features of colonial-era spelling may be useful when the reader refers from this atlas to other works.

The main changes introduced in 1973 were as follows: 

ch became kh
dj   j
j   y
sj   sy (pron. sh)
tj   c (pron. ch)

In an earlier reform (1947), oe was replaced with u.

In addition, the Dutch practice of occasionally doubling letters (Minahassa, Patti) was abandoned (Minahasa, Pati). At about the same time, ‘u’ also came to replace ‘o’ in a number of place names (Lampong, for instance, becoming Lampung). These and other changes mean that many place names can be found on older maps in four or more variant spellings.

I have made two significant exceptions to the general rule of using the modern Indonesian spelling of the place names current at the time referred to by each map. First, the names Java, Sumatra and Borneo have been used, rather than Jawa, Sumatera and Kalimantan, because the former are standard elements of English-language toponymy. And second, I have generally preserved colonial-era spelling for Dutch administrative names, mainly because many of them have no contemporary equivalent, and they cannot be modernized in the way that place names can. Thus, because Groote Oost has to be preserved in its original spelling, so too are the Lampongsche Districten and Bantam, insofar as these were administrative divisions.

Basic Geographical Information

Official figures concerning Indonesia’s exact size are not always consistent, but vary at most by a few thousand square kilometres. According to recent official figures, the Republic (excluding East Timor) has a land area of 1,904,443 km² together with 3,272,160 km² of sea and ocean, making a total area of 5,176,503 km². The country is situated between 94.45°E and 141.05°E and 6.08°N and 11.15°S, and stretches about 5,110 km from east to west and 1,888 km from north to south.

The Indonesian archipelago is formally considered to consist of 18,108 islands. This figure was decided in 2002, and replaced and early figure, 17,508 which had been announced in 1994 and which replaced the still earlier official figure of 13,667, set in 1963. It is not clear whether the current figure reflects the decision by the International Court of Justice to award the disputed islands of Sipadan and Ligitan to Malaysia just two months earlier.  Only about 3,000 of Indonesia’s islands, however, are said to be inhabited and only about 6,000 are officially named, though many more certainly have unrecognized local names. In reality, the number of islands – however an island is defined – is in constant flux. Siltation at the mouths of rivers creates new islands and joins old ones to the mainland. The mining of coral islands in the Pulau Seribu archipelago is reported to have caused several islands to disappear through erosion. Samosir in Lake Toba became an island only in 1906, when the Dutch cut a channel though the narrow isthmus which had connected it to the mainland. That channel is now silting up, and Samosir may soon cease to be an island.