Save

Religious Beliefs in ‘Aboriginal Taiwan’

In: International Journal of Taiwan Studies
Author:
Christopher Joby Research Associate, Centre of Taiwan Studies, SOAS, University of London, London, UK

Search for other papers by Christopher Joby in
Current site
Google Scholar
PubMed
Close
Open Access

Abstract

There are various approaches to building a picture of the ‘spiritual’ entities in which indigenous Formosans believed in ‘Aboriginal Taiwan’. This article does so by studying what sources written in the seventeenth century tell us about them. One source was written by a Chinese observer, and others by two groups of Europeans: Dutch East India Company employees and Spanish missionaries. Thus, one methodological issue is that these authors looked at Formosan belief systems through the different lenses of their own religious experience and tried to fit the Formosan belief systems into their own ‘existing knowledge grids’. A related problem is that the authors’ usage of terms may differ and indeed does differ from modern usage in the anthropology of religion. Despite these methodological issues, the article argues that these sources indicate that different Formosan tribes believed in different spiritual entities and were therefore marked by their heterogeneity.

1 Introduction

As Leigh Jenco and Birgit Tremml-Werner have recently demonstrated, in attempting to construct a picture of what Raleigh Ferrell (1969) calls the age of ‘Aboriginal Taiwan’, that is, before the arrival of Europeans in the seventeenth century, it is instructive to study early accounts written by those who had contact with indigenous Formosans, which contribute to a ‘historiography of the other’ (Jenco & Tremml-Werner, 2021).1 Generally, there is often a tendency to conceptualise the ‘other’ as a homogenous group (Williamson, 2014: 123). However, as Jenco and Tremml-Werner argue, the authors of these accounts manage to avoid what Geoff Wade references as the ‘topoi’ of otherness (Wade, 1997).

While some scholars have added to historiographical knowledge by accessing colonial or precolonial indigenous lives through material objects and oral histories, this article aims to examine what the authors that Jenco and Tremml-Werner reference and other authors write about the religious beliefs of the Formosan ethnic groups with whom they had contact (Ziomek, 2019, quoted in Jenco & Tremml-Werner, 2021: 219). These authors were Chinese, Spanish, and employees of the Dutch East India Company (Vereenigde Oost-Indische Compagnie, voc), including Dutchmen, a Swiss man, and a Scot. One methodological issue that my article therefore needs to address is that these authors or groups of authors observed Formosan belief systems through the different lenses of their own experience of religion and tried to fit the Formosan belief systems into their own ‘existing knowledge grids’ (Jenco & Tremml-Werner, 2021: 220). One consequence is that the authors’ usage of terms such as those rendered in English as ‘deities’ and ‘idols’ differed and indeed differs from modern usage in the anthropology of religion. This article attempts to help us understand how these authors used these terms with the aid of a typology of spiritual entities proposed by the French social anthropologist Philippe Descola. To contextualise this typology, it offers a comparative analysis of the belief systems of several Formosan ethnic groups in the present and recent past.2

From the seventeenth century, we have texts referencing three Formosan ethnic groups: the Siraya, the Favorlang, and the Basay. This article attempts to demonstrate that, just as the authors of these texts and their existing knowledge grids were heterogenous, so too the religious beliefs of these different Formosan ethnic groups in ‘Aboriginal Taiwan’ were by no means homogenous but were, on the contrary, marked by their heterogeneity.3 The first group that the article analyses are the Siraya. They lived in villages on the southwestern plains of Taiwan. They are almost certainly the subject of a short report that the Chinese Chen Di (陳第) wrote in 1603 (Jenco & Tremml-Werner, 2021: 219). In 1623, some voc employees were sent to Taiwan to construct a preliminary fort. One of them, the Swiss captain Elie Ripon, reported on the Siraya he encountered, above all in the village of Mattau (Andrade, 2009: 33, n. 1). This fort was destroyed. However, the following year (1624), the voc established a permanent base on the Bay of Tayouan near the Siraya villages and remained in Taiwan until 1662. Several authors, including two Dutch missionaries and one Scot working for the voc, provide details of the Siraya religious beliefs. One feature of their belief system that more than one author describes was a pantheon of deities.

To the north of the Siraya was a region inhabited by the Favorlang. From the mid-1640s onwards, the Dutch had intensive contact with the Favorlang. Texts compiled by Dutch missionaries working with the Favorlang suggest that, in contrast to the Siraya, they worshipped a supreme deity. Third, the Spanish established a colony in northern Taiwan, which functioned between 1626 and 1642, where they had contact with the Basay. Letters written by one of the missionaries do not mention deities but reference the many ‘idols’ of the Basay. The article investigates what this term means and compares it with the use of the term in Dutch texts.

In short, this article investigates the religious beliefs that formed an intrinsic part of the oral tradition and cultural practice of three Formosan ethnic groups prior to intensive contact with non-Formosan religious beliefs and practice and argues that, far from being homogeneous, they were marked by their heterogeneity, and thus challenge what Wade called the ‘topoi’ of otherness.

2 Methodological Issues

A significant challenge that scholars face in describing ‘Aboriginal Taiwan’ is that the Formosans had no writing system prior to the arrival of Europeans in Taiwan.4 Therefore, in using written sources to build a picture of the religious beliefs and practices of the Formosans, we must rely on non-Formosan sources. While authors no doubt attempt to get close to the ‘real’ state of affairs in their observations about Formosan belief systems, they nevertheless view these systems through the different lenses of their own religious experience and try to fit them into their own ‘existing knowledge grids’ (Jenco & Tremml-Werner, 2021: 220; Kennedy, 2006: 290). In other words, behind these non-Formosan sources are very distinct religious and linguistic forces that influence how the ‘other’ is perceived and depicted. Therefore, just as it is important to avoid the ‘topoi’ of otherness in relation to the Formosan ethnic groups, we must also avoid falling into the same trap when describing the accounts of non-Formosan authors. One of these is the Chinese Chen Di, mentioned above. The other authors are European. Again, though, we must disaggregate this term, for the religious systems that informed the Dutch and Spanish authors—Reformed Christianity and Counter-Reformation Catholicism, respectively—were also marked by difference.

One consequence of the heterogeneity of the religious backgrounds of these authors is that their usage of terms, such as those rendered in English as ‘deities’ and ‘idols’, differed. Furthermore, their usage differs from that of the discipline of the anthropology of religion. There are, in fact, at least three major classes of entity in the classical literature on religion, such as that of Émile Durkheim (Wallwork, 1985). Scholars are, however, not consistent in using terms to denote these entities. One scholar who has attempted to bring some order to the terminology in this field is Philippe Descola. He identifies three major classes of entity as ‘spirits’, ‘deities’, and ‘antecedents’, and positions them within his earlier work on analogism and animism (Descola, 2014: 129–130, 201–202; 2013: 37). The reader’s attention is therefore drawn to how Descola defines these three classes, for in cases of ambiguity, it will be instructive to use these definitions to help us understand how these terms were used by the seventeenth-century authors under review. First, spirits are ‘the typical incarnates in … animist ontologies, that is, where a continuity of souls and a discontinuity of bodies is assumed’. Each animist body, be it an animal or a tree, or even a shadow, is animated by a spirit. Spirits exist phenomenally as ‘presences’, which can be fleeting. One such ‘presence’ is the chirp of a bird, used in ornithomancy by Formosans, or the gaze of an animal (Descola, 2013: 39–40). The evidence suggests that spirits were present in all three Formosan religious systems under review.

Second, ‘deities … are specialized agencies specifically assigned to social units, to subdivisions of space … and of time … to regimes of practice … and to kinds of techniques, temperament, and life habits’. Creative deeds and exploits are attributed to deities in an unspecified past. ‘Deities’, Descola writes, ‘are generally firmly attached to places, where they are the object of genuine cults’. They are less immanent than spirits, for ‘they are affiliated to a segment of the collective from which are eventually issued the ritual experts entrusted with their celebration, and specialized fields of intervention are assigned to them’. Whereas spirits are defined as ‘animist collectives’, deities are defined as ‘analogist collectives’. As such, they are ‘alone in having veritable pantheons … because … one finds the same diversity and profusion in the little community of deities as there is elsewhere in the world at large’ (Descola, 2013: 40–43). An essential element of analogist collectives is sacrifice. This ‘establishes an operational continuity between intrinsically different singularities’ (Descola, 2013: 42). The Siraya and Favorlang both established a link with their deities by making sacrifices to them.

Third, antecedents ‘are literally what one has to get back to in order to understand, and accept, the conditions of the present order’. Descola identifies two main kinds of antecedents: ancestors and totems. Although totems play a role in the contemporary religious systems of Formosan groups such as the Amis, they are not referenced in the sources under review. By contrast, ancestors are. These ‘are humans from previous generations, neither really dead nor entirely alive, often materialized in domestic or lineage shrines, whose descendants depend on them for almost everything: their status, their temperament and dispositions, their means of subsistence’ (Descola, 2013: 43–44). Part of the cult of ancestors is to conciliate them and dispel their anger. ‘Despite behavioural differences between deities and ancestors—both, anyway, emerge in analogist ontologies—they share an identical mode of presence’ (Descola, 2013: 44). Finally, although Descola makes a clear distinction between these three classes of entities, he recognises that there is a certain fluidity of ontological boundaries (Descola, 2013: 46).

One term authors use to describe such entities that does not occur in Descola’s typology is ‘idol’. Dutch and Spanish authors use this term to reference both deities and spirits. This illustrates well that these authors’ usage of such a term was by no means consistent and, to some extent, reflects their own prejudices as they attempted to map the Formosan belief systems onto their own ‘existing knowledge grids’.

3 Comparative Analysis

Shortly, this article examines the textual evidence for the belief in spiritual entities by Formosan ethnic groups in ‘Aboriginal Taiwan’. Before doing this, however, it will be instructive to contextualise Descola’s typology by offering a comparative analysis of the belief systems of various Formosan ethnic groups in modern Taiwan.

As with Siraya, Robert Blust places the Amis language in the East Formosan group of languages (Blust, 1999: 45).5 This may suggest a historical connection between the two ethnic groups. Before their Christianisation, which gained momentum after World War ii, the Amis used the term kawas to reference supernatural beings in general. They can be divided into good kawas and bad kawas (S. Huang, 2008: 24–25). The Amis have many deities. For example, they have the protector gods, Fasonihar, Kakacawan, and Faydongi (S. Huang, 2008: 27). A famous Amis flood myth brought to the West by James Frazer references four sea gods, Mahahan, Mariyaru, Marimokoshi, and Kosomatora (Vermander, 2017: 261–262). The making of sacrifices also points in this direction. They make sacrifices of animals, with the domesticated pig being the largest. ‘Other essential gifts include betel nuts … and three kinds of rice food’ (Liu, 2011: 125, quoted in Vermander, 2017: 271–272).

The Paiwan belief system includes deities, ancestors, and spirits. According to one author, the Paiwan have at least eight gods or deities. These include creator gods; Sakina, a mountain god; Volaluval, a river and sea god; and Salamati, a soil god who helps crops to grow (Tung, 1984: 13–15). Another author recounts a Paiwan story that refers to two deities, ‘the god of heaven’ and ‘the god of earth’ (Vermander, 2017: 269). Furthermore, the Paiwan have ancestral spirits (T. Hu, 2017: 338). In the Paiwan village of Kulalao, the shaman enters a state of ecstasy referred to as tjetju tsemas (lit. ‘inhabited by the spirits’) to deliver the message of the ancestral spirits (T. Hu, 2017: 349). Some of these ancestral spirits are good, some bad. In the secondary literature, no firm distinction is made between ancestral and ‘animist’ spirits. Both inhabited natural places such as mountains and seas (Tung, 1984: 15). Benoît Vermander observes that the Rukai and Tsou have shamans who commune with the spirit world, although likewise does not distinguish between ancestral and animist spirits (Vermander, 2017: 270, 274).

The Puyuma, too, believe in deities, ancestors, and spirits. A central concept here is the birua. Some birua are ancestors, while others are deities, such as miadaran, ‘master of the roads’, and miawakal, ‘master of the paths’. The Puyuma also practise divination by dreams (oneiromancy), bird song (ornithomancy), and bamboo splinters. These ‘presences’ point to a belief in animist spirits (Cauquelin, 2004: 50–51, 204).

The Bunun make a distinction between bunun (human), hanitu (spirits), and dihanin (a deity or natural phenomena) (Vermander, 2017: 264–265). More specifically, hanitu include both animist and ancestral spirits. Hanitu are, in fact, the spirits of any living creature or natural object, animate and inanimate. This points to animist spirits. However, it also denotes the spirits of the deceased. When a person who has made a good contribution to the household and village dies, he/she is buried near the door of the house, and the Bunun believe that his/her hanitu will protect the members of the household (Y. Huang, 1988: 97, 111). In the Bunun language, dehanin means ‘sky’. As a deity, it is associated, above all, with celestial phenomena and natural disasters (Y. Huang, 1988: 174–175).

The Tayal (also Atayal) have deities, ancestral spirits, and animist spirits. The sky spirit Utux Kayal is a supreme deity. The Tayal ancestors become good or bad spirits. The Tayal have revived an ancestor-spirits festival, Mahoc (Vermander, 2017: 274–275).6 They also have ‘oracles’ such as a bird, dreams, and bamboo. These can be likened to Descola’s presences (Van Reeuwyk, 2012: 40–41).

Finally, there are Siraya villages near Tainan, although these have been heavily Sinicised. Some villages worshipped a fertility deity called Ali-tsu (also Ali-zu 阿立祖), which has been incorporated into the Han Pantheon. This name probably derives from Alid, a name that referenced the Judaeo-Christian God in translations of biblical texts made by Dutch missionaries in around 1650 (Adelaar, 2011: 300). This is used with the final element of the Sinicised rendering, tsu (祖), referencing ‘ancestor’, as in the name of the popular Taiwanese deity Ma-tsu (媽祖) (Shepherd, 1984: 30).7 This notion is strengthened by the fact that an alternative name for Ali-tsu was T’ai-tsu (Tàizǔ 太祖), or ‘supreme ancestor’. If so, it suggests that the name Alid had gained some traction among the Siraya before the Dutch were expelled from Taiwan. The Siraya in Toushe (頭社) Village, Tainan City, maintain a kuva and worship Ali-tsu.8 Furthermore, some Taivoan, who are closely related to the Siraya, moved to eastern Taiwan, where they have a communal shrine in Dazhuang (大庄) community in Dongli (東里) Village, Hualien (花蓮) County. Here, they worship Ali-tsu. However, Ali-tsu is a female deity, and so would clearly have undergone radical changes from the Christian God presented to the Siraya by Dutch missionaries (Shepherd, 1984: 7).

While this evidence is suggestive of monotheism, other evidence points away from monotheism. Other local gods have the prefix, Ali-. For example, in Donghe (東河) Village in Dongshan District (東山), Tainan City, which was known as Kabuasua, the villagers worshipped Alimu, a deity representing all ancestral spirits in the village.9 The suffix -mu means ‘mother’. The sacred annual ritual celebrating the birthday of Alimu is a major event at Kabuasua Village. Pork and wine offerings are important rites. The villagers also persist with the worship of ancestral spirits called arit in Siraya.

Caution should be exercised here, for some of these beliefs have been influenced by Sinitic cultures or Christianity. Furthermore, most of the authors cited are non-Formosan. While their aim is, of course, objectivity, there may be an element of cultural bias. Nevertheless, from the evidence presented, there seem to be examples of belief systems in contemporary or recent Formosan societies that include all three entities in Descola’s typology: deities, animist spirits, and ancestors. It is thus reasonable to expect to find examples of these entities in the belief systems of Formosan ethnic groups in ‘Aboriginal Taiwan’, which are the focus of the rest of this article.

4 The Siraya

Of the Formosan ethnic groups with whom non-Formosans had contact, we know most about the Siraya.10 Chinese author Chen Di and several voc employees described the religious practices and beliefs of the Siraya. Although there are differences, in many respects, they agree, and their accounts allow us to build a reasonably accurate picture of the entities in which the Siraya believed. One problem we face, however, is that some of the names Dutch authors give for the Siraya deities differ, partly because of orthographical inconsistencies.

4.1 Chen Di’s Dongfan Ji

After a naval expedition to Taiwan in 1603 to deal with wako pirates, the military adviser, scholar, and author Chen Di (1541–1617) compiled a short ethnography of Formosans that the expedition encountered (Shepherd, 1993: 35). The ethnography, entitled Dongfan Ji 東番記 (An Account of the Eastern Barbarians), was published in block-print.11 Chen’s essay, which comprises only 1,438 characters, was the first significant first-hand written account of the indigenous peoples of Taiwan (Jenco & Tremml-Werner, 2021: 227). Chen does not mention a specific Formosan tribe. He does, however, reference Da Yuan (大員). This is thought to correspond to Tayouan, the name used in Dutch sources to reference the spit of land and bay where the voc established its base at Fort Zeelandia, which covers the area occupied by modern-day Anping (安平). This indicates that Chen is most probably describing the Siraya, who inhabited this area.

Chen’s approach is ‘overall remarkably factual and detail-oriented, adhering closely to the historicist (even iconoclastic) spirit of the philological work for which he was known among his contemporaries and later scholars’. Recently, scholars in Taiwan studies have used Chen’s account to reconstruct the island’s precolonial or ‘pre-invasion’ history. The present article uses it for the same purpose (Jenco & Tremml-Werner, 2021: 226, 222). In trying to get close to the ‘real’ state of affairs, Chen was facilitated by his encounter with an indigenous person, whose name he transcribes into Chinese as Dà Mí Lè (大彌勒). Here, we can perhaps talk in terms of ‘indigenous agency’ (Jenco & Tremml-Werner, 2021: 223, 220).

Chen describes various aspects of the Siraya social life, including the division of labour between men and women and how they selected marriage partners. He also devotes a few sentences to their religious beliefs. At one point, he references Tian (天) and shen (神) (Fang 1956: 69):

When the crop has ripened, they resume their original behaviour. They say that if they do not act thus, then Tian will not help them, the shen will not bless them, and they will have a bad harvest for the year.12

In another section, Chen references gui (鬼) (Fang 1956: 69):

[When] a wife dies [the husband] remarries, [but when] a husband dies [the wife] does not remarry: [she] is called a gui.13

Tian, shen, and gui are complex concepts loaded with meanings across time and religious traditions. The questions of what Chen was signalling with these terms, what his readers might understand by them, and how closely they represented the Siraya beliefs are not easy to answer. In terms of understanding their meaning within the context of Chinese religious life, it may be instructive to begin by returning to an account of ‘classical’ Chinese religion as set out in Li Ji 禮記 (Book of Rites), part of the Confucian canon. C. K. Yang (1961) observes that the core of this classical religion, which predates the ‘foreign’ influence of Buddhism and the rise of Daoism, was the worship of Heaven, its pantheon of deities, and ancestor worship: ‘Heaven was the supreme anthropomorphic power of the universe directing the operation of the spiritual world.’ ‘Heaven’ here is Tian. This can be understood as a simple principle within a monistic philosophical/theological system. The spiritual world is dichotomous. At the top was the pantheon of deities (shen).14 Each of these had a definite function in governing humans and spirits. Below these were many spirits or ghosts (gui). The gui were ‘the common subjects of the spiritual world’ (Yang, 1961: 23–24).15 Stephan Feuchtwang also describes the treatment of shen and gui in Li Ji. He observes that shen and gui (he writes kuei) are both used variously in the sense of spirit and god or demon (i.e. good or bad deities). Specifically, he writes that ‘any dead human is a kuei, unless he is worshipped as an ancestor or as having lived a life worthy of commemoration, in which case he is a shen’ (Feuchtwang, 1978: 115–116). So, for both Yang and Feuchtwang, in Li Ji, shen rank above gui/kuei.

This belief system would later be incorporated into the theological system of Yin and Yang (Yang, 1961: 23–24). Broadly speaking, shen are yang-spirits and gui are yin-spirits, or shen positive or ‘light’ spirits and gui negative or ‘dark’ spirits (Chan, 1963: 514, 780).16 In this context, gui could be translated as ‘demons’. Interpreting Chen’s use of shen and gui in the context of Yin and Yang has merit, for they are central to Daoist cosmology and, as recent scholars note, there are clear Daoist influences on Chen’s thought, which can be identified in Dongfan Ji (Teng, 2004: 60–71).17 Baozhu Hu (2020) writes that shen and gui are, in some sense, two sides of the same coin. Although antithetical, they have much in common. In fact, they are sometimes written together as guishen, which has ‘the connotation of unseen supernatural beings or deities who intrinsically have a kind of divine power’. This concept is evident in I Ching 易經 (Book of Changes) (Chan, 1963: 505).18 In latter-day sacrifices, guishen referenced ancestors (Chan, 1963: 780). Feuchtwang, by contrast, writes that over time guishen came to reference the spirits of the lowest class in society, that is, of common people whose lives had not been noteworthy to any degree (Feuchtwang, 1978: 116).

Let us try to map these terms onto the typology of entities proposed by Descola (2013: 37). ‘Heaven’, understood as ‘the supreme anthropomorphic power’, has something in common with a supreme deity. Indeed, this is how Tian was understood in the Shang dynasty (1600–1046 bce). Furthermore, in the early Qing period, a little after Chen wrote his text, Tian and Shangdi 上帝 (Highest Deity) were used interchangeably (Feuchtwang, 1978: 117). As for shen and gui, these, like Tian, are polysemous terms, and the text’s meaning depends on how Chen intended to use them. If he used them in the sense of classical Chinese religion, shen were deities and gui spirits. However, within the theological system of Yin and Yang, both shen and gui are deities, spirits, and ancestors, with shen being good or positive spirits and gui being negative ones (Chan, 1963: 780).

The question then arises of what Chen’s account might tell us about the Siraya belief system. It suggests that the Siraya believed in a supreme deity and many other spirits, some good, some bad, which in the anthropology of religion could be deities, spirits, or ancestors (Descola, 2013: 37). A related question is how close this brings us to the ‘reality’ of this system. Here, it will be instructive to compare this account with those of the Dutch, to which the article turns shortly. In both cases, however, we need to remember that Chen is limited both by his use of what I term an ‘allolect’, that is, a language, in this case, wenyan (文言) or Classical Chinese, other than that of the culture he was describing, and by a belief system which differed significantly from that of the Siraya. He was, perhaps unavoidably, attempting to fit the Siraya belief system into his own existing knowledge grid.

Before turning to the Dutch texts, one other term in Chen’s account requires our attention, she (社). Chen writes that ‘thieves are slain at the [village] altar’ (有則戮於社) (Fang, 1956: 69). He uses she to reference a village altar (cf. Thompson, 1964: 175). The root meaning of this character (社) is ‘god of the soil’ or ‘god of earth’. By extension, it references altars to this god (Yang, 1961: 96–99; Dean, 1998). Two points are in order here. First, while Chen uses a word that denotes an altar, the Siraya altar was probably not merely one to the god of the soil, so there is a slight shift in meaning. Second, the presence of an altar points to sacrifice. Therefore, this was probably where the Siraya made sacrifices, including human ones, to their deities (Thompson, 1964: 175; Lin, 2014: 26–28).

4.2 Reports on the Siraya by voc Employees

We have reports on the Siraya which reference their belief system by voc employees of several nationalities. Alongside Dutchmen, one Swiss man and a Scot made reports that require our attention. It is probable, though not certain, that the Swiss and the Scot were Calvinists. This section analyses their reports in chronological order.

4.2.1 Elie Ripon’s Report on the Siraya Village of Mattau

In 1623, while voc forces were still based on the Penghu Islands to the west of Taiwan, they made reconnaissance missions to Taiwan. On one of these, led by Commander Cornelis Reyersen in October 1623, a Swiss man working for the voc, Captain Elie Ripon, was required to help construct a preliminary fort. He wrote an account in French of his experiences in Taiwan, which was discovered in a Swiss attic in 1865 (Ripon, 1990: 7; Andrade, 2009: 33, n. 1).

There were several Siraya villages in relatively close proximity, which, according to the Dutch missionary Georgius Candidius, shared the same language, manners, and customs. One of these was Mattau (Campbell, 1903: 9).19 In a short description of this village, Ripon does not mention spirits or other entities. He does, however, mention a ‘temple’ twice. First, while they were still active as warriors, husbands lived separately from their wives. In a passage on the husband visiting his wife at night, Ripon observes that the man ‘goes to the temple on the cemetery, he jumps three or four times, and then goes to his wife’ (Ripon, 1990: 106).20 A little later, Ripon describes the temple. He observes a round building in the middle of a large square. Inside are many heads of their enemies, as well as heads of deer, wild boar, and jaw bones. In front of the temple, a lamp burns night and day. When they return from war, the men make their devotions: they angrily cry and scream (Ripon, 1990: 107). Reference to ‘devotions’ is suggestive of deities, such as a god of war, to whom the men pray, although this is not explicit.

4.2.2 Report on the Siraya by Constant and Pessaert

Two Dutch voc employees on the same expedition in October 1623 were Jacob Constant and Barend Pessaert. Assisted by Chinese intermediaries, they had contact with another Siraya village, Soulang. They had little to say on religion but did note that each house had an altar where the Siraya placed the skulls and bones of enemies defeated in wars (Blussé, Everts & Frech, 1999–2010, vol. 1: 9). The presence of an altar in each house seems to point to household gods. Further research may reveal whether, like the lares in ancient Rome, they performed a tutelary role.21

4.2.3 Report by Georgius Candidius

Two Dutch missionaries, Georgius Candidius (in Taiwan 1627–1631 and 1633–1637) and Robertus Junius (in Taiwan 1629–1643), provide details of the Siraya deities. These and an ethnography compiled by David Wright, a Scot who worked for the voc, point to a pantheon of deities for the Siraya. In addition, Candidius wrote an ethnography which included a brief description of the religious beliefs and practices of the Siraya, first published in 1635 and again in 1639 (Van Rechteren, 1635: 53–59; 1639: 66–71).22

Candidius observes that the Siraya knew nothing of ‘one Supreme Creator’ (Campbell, 1903: 75). Rather, they had many deities, which Candidius sometimes references as ‘afgoden’, often translated as ‘idols’ (Campbell, 1903: 14–15; Grothe, 1884–1891, vol. 3: 11; Andrade, 2009: 26). He names several important deities in the Siraya pantheon. Tamagisangack, who lives in the south, makes people beautiful.23 His consort, Takarckpada, lives in the east. The fact that these deities had consorts indicates that they were anthropomorphic.24 Another deity, Sarisang, who lives in the north, is not considered very important. In contrast to Tamagisangack, this god makes people ugly with pockmarks. The Siraya call on two other gods when preparing to wage war: Talafutta and Tapatiap. Candidius’s account was republished several times. In Olfert Dapper’s Gedenkwaerdig Bedryf der Nederlandsche Oost-Indische Maetschappye, first published in Amsterdam in 1670, the names are slightly different: Tamagisangack becomes Tamagisanhach; Takarckpada becomes Taxankpada Agodales; Sarisang becomes Sariasing, while the gods of war become Talafula and Tapaliape (Dapper, 1670: 15–16). An English version of Candidius’s report appeared in Atlas Chinesis, published the following year, 1671, by Arnoldus Montanus (Montanus, 1671). The names of the deities it gives are identical to those in Dapper’s publication.25

Candidius then describes the inibs (shaman/priestesses), who communicate with the gods and bring sacrifices to them and banish evil spirits. He also observes that each house has a place where the inhabitants call upon the gods (orig. Dutch: ‘goden’) and bring them sacrifices (‘offeren’). This observation points to the altars for household gods identified by Constant and Pessaert. Furthermore, it indicates that the Siraya had other gods besides those that Candidius and other Dutch authors named. In this regard, Candidius wrote that ‘they have a great number of other gods besides-too many to mention here, on whom they call and whom they serve’ and ‘on the roads and byways, too, everywhere they place sacrifices for their gods’ (Van Rechteren, 1639: 71; Grothe, 1884–1891, vol. 3: 27; Campbell, 1903: 24–25; Ferrell, 1969: 16).26 Finally, Candidius observes that the Siraya listened to the singing of birds before deciding on important matters (Campbell, 1903: 23). Birdsong is suggestive of the ‘presences’ of spirits in Descola’s typology (Descola, 2013: 39).

Candidius reported that when the Siraya returned from a successful headhunting expedition, the severed head would be taken to a building he describes as a church (‘kercke’). Fifteen or sixteen families had such a building. They would then slaughter many pigs in honour of their deities (‘afgoden’) (Campbell, 1903: 14–15; Grothe, 1884–1891, vol. 3: 11; Andrade, 2009: 26). The ‘churches’ to which Candidius refers are probably separate buildings close to the houses of more prominent villagers, which could be likened to a temple, where sacrifices were made to a deity.27 It is noteworthy that Candidius uses a Christian term to reference these buildings used for pagan rituals, an example of him trying to fit the Siraya belief system into his own ‘existing knowledge grid’ (Jenco & Tremml-Werner, 2021: 220).

4.2.4 Texts Compiled by Robertus Junius

Candidius’s successor, Robertus Junius, produced several catechisms, prayers, and sermons in Siraya. While much of this material has been lost, translations into Dutch survive. For example, in an undated sermon on Psalm 50:15, Junius argues that those who continue to worship the Siraya gods (orig. ‘goden’) and practise idolatry (‘afgoderyen bedrijven’) should be censured. He specifically mentions three gods: Tekaroupouda, Tamagisangan, and Tapaliapeang (Grothe, 1884–1891, vol. 3: 315; Campbell, 1903: 333). Tekaroupouda is a fertility goddess, while Tamagisangan is the principal deity who lives in the west of the sky.28 Tapaliapeang probably equates to the god of war Tapaliape mentioned by Candidius, with -an/-ang being a common Siraya affix (Adelaar, 2011: 4). In another sermon, on Psalm 116:12, Junius tells the Siraya not to worship ‘Tekarapada or tamagisangachang’ instead of the Judaeo-Christian God.29

4.2.5 An Observational Report Compiled by David Wright

Another source is an observational report compiled by David Wright, a Scot who worked for the voc in Taiwan for several years.30 Wright’s report was first published in Dutch in 1670 in Olfert Dapper’s Gedenkwaerdig Bedryf, mentioned above (Dapper, 1670: 17–51). It was subsequently published in English in 1671 in Montanus’s Atlas Chinensis (Montanus, 1671: 10–49; Shepherd, 1984: 56–81).31 Dapper’s work describes Taiwan and China. The section on Taiwan is divided into two parts. The first part, which describes aspects of Siraya religion, was based on the ethnography compiled by Candidius. The second part was written by Wright, probably in about 1655 (Lin, 2014: 32). It describes the culture and religion of the Siraya. It provides the most comprehensive account of the Siraya religion compiled in the Dutch period, describing its deities, festivals, and taboos or laws, and deserves our close attention.

Wright observes that the Siraya have 13 principal deities (see Table 1). He writes, ‘The Formosans (except those by the Hollanders converted to Christianity) believe not in God, the Creator of Heaven and Earth, but Worship thirteen Idols’ (Montanus, 1671: 32; Dapper, 1670: 33). Reference to ‘idols’ (orig. Dutch: ‘valsche Af-goden’) betrays Wright’s Judaeo-Christian perspective on the Siraya belief system and the notion that the worship of deities other than the Judaeo-Christian God, the Creator of Heaven and Earth, amounted to idolatry. John Calvin, whose theology most probably informed that of the Scot, frequently warns his followers against worshipping idols instead of ‘the one, true God’, that is, the Christian Trinitarian God (see e.g. Calvin’s Institutes, Book I, xi, xii).32 A more objective, or perhaps late modern, even postcolonial, way of describing this state of affairs would be to say that there were 13 principal gods or deities in the Siraya pantheon (Montanus, 1671: 32–33).

T1

Wright records the first deity as Tamagisangang, who resides in the west of the sky (Dapper, 1670: 33).33 He names the second deity as Takaraenpada, the consort of Tamagisangang, who lives in the east.34 The Siraya feared that if they were slow to worship these two deities, they would become wrathful and destroy the Siraya villages. The third and fourth deities are also a couple: the male, Tamagisangak, who lives in the south and his consort, Teckarupada, who lives in the east. Tamagisangak shapes beautiful people, while Teckarupada is a fertility goddess with power over corn and fruits of the field. Together they have sovereignty (orig. Dutch: ‘gewout’) over human life. For this reason, offerings are made to them, above all by women. The fifth deity is called Tagittellaegh, and the sixth is his consort Tagisikel. They can cure the sick and are worshipped and celebrated by them. The seventh deity, Tiwarakahoeloe, and his consort Tamakakamak regulate hunting. The ninth deity, Tapaliat (Montanus 1671 has Tapaliape), and the tenth, Tatawoeli, regulate the waging of war. They are principally honoured by men, who then go to war. The eleventh deity, Takarye, and the twelfth, Tamakading, preside over the seven annual feasts of the Siraya and mete out punishment if the Siraya fail to observe their established customs. The thirteenth and final principal deity was Farihhe, Fikarigo Gougosey (Lin, 2014: 62).35 He was said to live in the north and to be an evil deity. He had the power to make humans deformed, whom Tamagisanghach (recte Tamagisangak) had made beautiful, so he is only worshipped to prevent this from happening. This quality is like that of Sarisang, mentioned by Candidius. However, we cannot currently say anything more about the relationship between Sarisang and Tamagisanghach/Tamagisangak.36

According to the Siraya tradition, Farihhe had been a man in the village of Sinkan (also Sinkam) who was transformed into a deity. This may suggest that Wright was using Sinkan as his principal source, although he does not name a village at the start of his account. This was the Siraya village near Fort Zeelandia, where Candidius built himself a hut and where Junius spent much time. Having been made a god, Farihhe returned to earth with 27 ‘articles’ to which the Siraya had to adhere. These articles were in fact taboos or laws, such as not building houses, which the Siraya had to observe during the monthly solemn period of Karichang (Montanus, 1671: 34–37; Lin, 2014: 62).

Wright also observed that when the Siraya went to war, they would take the skull of a dead enemy warrior, fill it with rice and proclaim that if the spirit of the warrior were to accompany them into battle and they were to win the battle, they would make continual sacrifices to him and count him among their gods (Montanus, 1671: 27).

Finally, Wright observed that the Siraya believed in the devil as schytinglitto, whom they blamed for any incurable disease (Shepherd, 1984: 67; Chiu, 2008: 188). In their Siraya translations of the Lord’s Prayer, both Junius and a later Dutch missionary, Daniël Gravius (in Taiwan 1647–1651), used litto to reference the devil, emphasising that the Judaeo-Christian God could deliver the Siraya from the devil and all the bad things that he brought on them (Müller, 1680: 27; Adelaar, 2011: 339).37 Candidius had earlier recorded that inibs would chase ‘the devil’ (orig. ‘duyvel’) until he jumped into water and drowned. He also references ‘many devils’ (‘veel duyvelen’), which may point to evil spirits (Grothe, 1884–1891, vol. 3: 27; Campbell, 1903: 25).

4.2.6 Comparative Analysis of Texts on the Siraya Belief System

Concluding, it will be instructive to use the typology described by Descola to identify what these accounts have in common and whether we can get close to the ‘real state of affairs’ of the Siraya belief system (Kennedy, 2006: 290). First, the Siraya most probably believed in deities. Candidius, Junius, and Wright name entities that seem to correspond to deities closely connected to the principal activities of the Siraya, such as hunting and waging war. Wright’s account points to a pantheon of the most important deities. References to altars on which sacrifices were made by Constant and Pessaert, Candidius, and indeed Chen Di indicate deities and tell us that there were more than the 13 that Wright mentions. Before the arrival of the Dutch, the Siraya villages lacked chieftains (Campbell, 1903: 15; Ferrell, 1969: 16). One possibility is that having a pantheon of deities, and not one supreme deity as Candidius observed, may reflect this social structure.38

Above, I problematised the terms used by Chen Di. Shen and gui may refer to deities, but they may also refer to other entities, that is, animist spirits and ancestors. Candidius’s references to birdsong and ‘many devils’ point to the Siraya belief in spirits, in the latter case, evil ones. The focus of the voc authors is nevertheless on deities. This may reflect their own belief system, which focused on one God and the desire of the missionaries to replace the Siraya pantheon with their own Judaeo-Christian God. Questions remain, such as what Chen was referencing with Tian. This may have more to do with his own worldview, however, than with a reality that the Siraya would recognise and worship, for they clearly did not have a supreme deity, and, furthermore, there is no evidence that they developed a philosophical system or understanding of the world that could be described as monistic. One possibility is that Tian references nature or the gods that the Siraya believed controlled natural phenomena.

Finally, the reader’s attention is drawn to the fact that as the Siraya had no writing system before the arrival of the Dutch, they relied on oral tradition. One consequence of oral tradition was that Formosan culture was mutable; that is, it underwent change. Candidius observed that the religion of the Siraya had been undergoing decay for the last 60 years. Furthermore, he suggests that even without the intervention of Christian missionaries, it would again undergo much change in the next 60 years. He bases his claim of changes in recent history on the evidence of conversations with older people who state that things seemed quite different now from how they remember them (Campbell, 1903: 90; Kang, 2020: 122). One consequence of this is that the belief system described above pertained in the first half of the seventeenth century. It may have differed from what went before, and indeed, if we remember that Chen Di wrote his report in 1603 and David Wright his report around 1655, it may have changed during the period under review.

5 Dutch Texts in Favorlang

The Favorlang (also Favorlangers) lived to the north of the Siraya in modern-day Yunlin (雲林) County by the Zhuoshui River (濁水溪).39 The main village of the Favorlang, which Tonio Andrade describes as ‘one of the largest and most powerful aboriginal villages in Taiwan’, was probably located near present-day Huwei (虎尾) in Yunlin County (Andrade, 2005: 295, 306). They spoke varieties of Favorlang, a now-extinct Formosan language, which Robert Blust places in the Western Plains language group along with Taokas, Babuza, Papora, Hoanya, and Thao (Blust, 1999: 45).

Because of tensions between Favorlangers and Chinese settlers, in the early 1640s, the voc had increasing contact with the Formosans and tried to subdue them with a combination of force and social control. One element of the social control was the attempt to convert the Favorlangers to Reformed Christianity. Three missionaries were stationed in the Favorlang district: Simon van Breen (in Taiwan 1644–1647), Jacobus Vertrecht (1647–1651), and Gilbertus Happart (1649–1653). Vertrecht and Happart compiled what are almost the only surviving documents we have in the Favorlang language, which most probably built on earlier work by Van Breen (Li, 2019).40

Happart is named as the author of a Favorlang-Dutch lexicon, which remained unpublished in the seventeenth century (Happart, 1650). The manuscript was identified in Batavia (Jakarta) in 1839 by Wolter (W. R.) van Hoëvell. In 1840, the English missionary Walter (W. H.) Medhurst published a Favorlang-English edition, and two years later, Van Hoëvell published a Favorlang-Dutch edition (Happart, 1840, 1842). This had entries for more than 2,500 words. Each edition has words in a different order, and each omits words that the other includes.

The reader’s attention is drawn to the Favorlang word Haibos and its orthographical variants. In the Dutch edition, Haibos is given as ‘de duivel’ (the devil) (Happart, 1842: 369). In the Medhurst English edition, Chaibos or Haijbos is defined as ‘a name given by the natives to the devil’ (Happart, 1840: 38). The Scottish missionary William Campbell republished the lexicon in 1896. He included both Chaibos and Haibos with the definitions just given (Campbell, 1896: 129, 137). This may lead one to the conclusion that Haibos was an evil deity like the devil, who brought bad things upon the Favorlang. However, a different picture emerges from a close reading of doctrinal texts compiled by Vertrecht (Li, 2019).

Here, we see that the Favorlangers believed that, on the one hand, Haibos was kind to those who worshipped him, while, on the other hand, he could also oppress the Favorlangers, make them sick, and take away their souls. If they experienced something good, they would call Haibos ‘good’, saying ‘mario Haibos’. The Favorlang-Dutch lexicon records mario for ‘good’ (orig. Dutch: ‘goet sijn’) (Happart, 1842: 165–166; Campbell, 1896: 153). However, if they encountered suffering or illness, they would say ‘rapis Haibos’, rapi[e]s being the Favorlang word for bad or evil (Dutch: ‘quaad’) (Happart, 1842: 272; Campbell, 1896: 179). So, Haibos has good and bad. To frame the Favorlang belief in Haibos in theological terms, it is, in some sense, dualist. However, unlike dualist religions such as Zoroastrianism, they believed that good and evil were contained within one spiritual being, that is, Haibos. It could, of course, be argued that Haibos has something in common with the Judaeo-Christian God in that even though he is understood to be a good God, the Bible includes accounts in which he seems to inflict evil and suffering on people, as appeared to be the case, for example, in the Old Testament story of Job. Often, though, the bad things are ascribed to the devil, and herein lay the solution to how the Dutch missionaries addressed the Favorlang worship of Haibos.

Several texts and sermons compiled by Vertrecht attack the Favorlang belief in and worship of Haibos. It is, however, in a dialogue between a Favorlanger and a ‘Stranger’, that is, a Dutchman, in which the Dutch approach to undermining belief in Haibos is stated most clearly (Vertrecht, 1888: 17–39; Campbell, 1896: 12–32; Lin, 2014: 132–135, 180). The dialogue is based on a syllogism: (a) the world has not always existed, and therefore some being must have created it; (b) the Christian God created the world as revealed in the Bible, so (c) Haibos’s claim to create the world is false and therefore Haibos is a deceit (Lin, 2014: 133).

The dialogue begins with an attack on Haibos, with the Stranger asking the Favorlanger why he worships a deity who causes bad things to happen. The Favorlanger then states that just as Christians praise their God for good things, so the Favorlangers praise Haibos, calling him ‘mario Haibos’ (good Haibos). The Stranger replies that the deity responsible for good things can only be called Deos, i.e. the Christian God (Li, 2019: 102–103). In fact, the Stranger argues, it was Deos, a Latinate word most probably derived from Portuguese, who made the heavens and earth and, as a providential God, still cares for and engages with his Creation.41 He argues that, by contrast, Haibos is the deity responsible for bad things. Later, he asserts that Haibos is revealed in the Bible under the name of Satan. In other words, Vertrecht’s approach divides Haibos into two deities and argues that the good part of Haibos is, in fact, Deos, while the bad part of Haibos is the devil or Satan. As noted above, in the Favorlang-Dutch lexicon, Haibos is rendered as duivel (devil) (Happart, 1842: 369; Campbell, 1896: 137).42 Vertrecht strengthens the equivalence between Haibos and Satan by saying that Haibos was formerly a good angel who served Deos but that he and his followers rebelled against Deos, and he was therefore cast out of heaven and punished for his disobedience (Li, 2019: 108–110).

In constructing his argument, Vertrecht takes advantage of the fact that the first man in the Bible and the name of the small bird used for divination by the Favorlangers have the same name, adam. The Favorlangers would use the call of the small bird as a portent. The Favorlang-Dutch lexicon goes to some length to describe how the number of calls that the adam made could be interpreted as a good or bad portent (Happart, 1842: 38; Campbell, 1896: 122). These calls could be understood as ‘presences’, which point to a belief in spirits to return to Descola’s typology (Descola, 2013: 39). The Stranger says that Haibos has deceived both the bird and the first man in the Bible (Li, 2019: 104). The Favorlanger does not simply give up in the face of these arguments. He says that he believes in and worships Haibos as the creator of heaven and earth because he follows the ancient belief of his forefathers. The Stranger responds that they did this because they did not know Deos, the true God, for they did not know the written Word of God (Fav.: ‘o ranied o Deos kabinidoan’), that is, the Bible (Li, 2019: 97). So, central to Vertrecht’s argument that the Favorlangers should reject Haibos and accept Deos is that knowledge of Haibos has been passed down orally, whereas knowledge of Deos has been transmitted in written form. Implicit in this view of things is that the written source is more reliable than an oral source or oral tradition. This accorded with the view that correct knowledge was textually dependent, which emerged during the European Renaissance when the construction of knowledge and transmission of learning in written form took a definite shape (Heylen, 2001: 210).

One other defence against the acceptance of Deos that the Favorlanger offers is that if he does reject Haibos, Haibos will be angry: ‘We fear too much the wrath of Haibos’ (Li, 2019: 116). The Stranger retorts immediately, asking rhetorically whether the Favorlanger does not fear the wrath of Deos even more. He asserts that the wrath of Deos is greater than that of Haibos. He returns to identifying Haibos with Satan, saying that Haibos is a poisonous snake, echoing Genesis 3, and that anyone who worships him will end up in the fire of hell (Li, 2019: 117). In one of Vertrecht’s texts—‘A short catechism before receiving Christian baptism’—the baptismal candidate is asked whether he forsakes Haibos and his vain superstitions. The answer is yes. However, he rejects not only Haibos and all vain superstition but also the bird, adam (Lin, 2014: 129; Li, 2019: 189). Vertrecht clearly felt that an important step towards rejecting their previous practices was the rejection of divination, which is not accepted within the Christian tradition. For Vertrecht, it was not appropriate to try to gain access to the spiritual realm via divination or, more specifically, ornithomancy.

So, a fundamental part of the Dutch missionary strategy among the Favorlangers was to undermine their belief in Haibos. A major problem that they faced was that there were certain similarities between Haibos and the Judaeo-Christian God. The Favorlangers believed, or said they believed, that Haibos created the world, just as Christians claim that their God created the world. Furthermore, in the Christian and Favorlang mind, their supreme deities can cause good and bad.

One way to understand Vertrecht’s approach to converting the Favorlangers to Christianity is to see it as integrating their belief system, above all in Haibos, into the Christian story. In doing so, Vertrecht argued that they had received the Christian message incorrectly and that he and his colleagues were now providing them with the correct version. They did this above all by bifurcating Haibos and claiming that his bad nature was, in fact, none other than Satan and that his good nature was Deos, the creator and providential Judaeo-Christian God. They therefore attempted to turn Haibos’s potentially problematic nature as a dualist supreme deity to their advantage to graft the Favorlang belief system onto their own. Looking at things from another perspective, Vertrecht and his colleagues infused the Favorlang religion with the biblical message, not dismantling its structure but reconfiguring it so that it conformed to what the Bible revealed about the Creator God and the devil (Lin, 2014: 131–132).

Dutch sources do not mention other deities, leading Lin (2014: 128) to conclude that the Favorlang religion was monotheistic. In response, I would argue that because no other deities are mentioned, this does not necessarily mean that the Favorlang religion was monotheistic. It does, though, suggest that Haibos was the supreme deity of the Favorlang.43 As for spirits, reference to ornithomancy points in this direction. The Favorlangers probably practised ancestor worship, although there is no direct reference to this in the Dutch texts.44 The lexicon gives ‘ancestors’ (Dutch: ‘voorouders’) as boeboe or boebos but provides no further details (Happart, 1842: 364; Campbell, 1896: 128).

6 Spanish Texts on the Basay

The Spanish established a base in Taiwan in 1626. José Borao summarises the rationale for the Spanish presence in Taiwan as ‘trade, the counterbalance of Dutch power, and missionary access to China and Japan’ (Borao, 2020: 286). To this list, we can add missionary work in Taiwan. In 1631 the Dominican Fr. Jacinto Esquivel, a native of Vitoria in Spain, arrived in Taiwan from the Philippines. He had reached the Philippines in 1625, where he taught at the University of Santo Tomás in Manila, founded as a college seminary in 1611 (Borao, 2020: 269–270; Hernández, 1996: 42). In 1632 Esquivel wrote a long memorandum which included a short ethnography of the indigenous Formosans he encountered. Among these were the inhabitants of Quimaurri and Taparri, both near modern-day Jilong (基隆).45 He observed that the villagers of Quimaurri and Taparri were of the same stock and had the same customs and traits (Borao, 2001–2002, vol. 2: 166). In the same year, Esquivel recorded that many villages, including those in the settlements of Quimaurri and Taparri, shared a common language called Basay (orig. Spanish: ‘una lengua común y general, que es la baçay’), although some villages or group of villages also had their own language (‘una lengua particular’).46 It is likely, therefore, that the ethnography mainly concerns the Basay.

The question of whether the Basay had a supreme deity or a pantheon of deities is not an easy one to answer. Japanese linguists have recorded a Basay word for ‘god’, samiai, and indeed ‘god table’, samiai takan (Tsuchida, Yamada & Moriguchi, 1991: 240). This was, however, in the first half of the twentieth century. It is therefore not possible to understand precisely what this word references.47

Esquivel does not mention deities in his account of the Basay. Nevertheless, he observes that they had many idols (‘ídolos’). They may not, however, be idols in the sense of material objects, nor in the sense of pagan deities, which seems to be how David Wright uses the term in relation to the Siraya, but rather things connected to the spirit world, which they used to regulate their lives—that is, ‘presences’, to return to the term used by Descola (Descola, 2013: 39). This includes the chirping of a small bird, the call of herons, good or bad dreams … even sneezing. The Formosans that Esquivel encountered believed that there were good and bad spirits, the former bringing good health and prosperity, the latter illness and misfortune (Borao, 2001–2002, vol. 1: 178–179). As well as being common among other Formosans, the idea of the world being inhabited by good and bad spirits was also found among the indigenous peoples of the Philippines (Scott, 1994: 78–81).

In his report, Esquivel observes that the Basay have thousands of omens or marnaches as they call them. For example, they believe that if they eat ‘this food and not that food’, the pigs will come and eat up the rice or the rice fields will dry up (Borao, 2001–2002, vol. 1: 180). As with the Siraya, older women played an important role in the praxis of the social life of the Formosans that Esquivel encountered. He observes that although they had festivals, they had no rites and made no sacrifices (‘No tienen género alguno de ritos o sacrificios’). Elsewhere, however, he writes that they had drunken feasts (‘borracheras’) that lasted 15–20 days. Here, they killed pigs and practised superstitions (‘supersticiones’). Men in an outer circle danced around women in an inner circle, and in the centre were the old women performing the ritual slaughter of a pig. These women, called majuorbal, functioned as healers whose purpose was to exorcise evil spirits from the infirm (Borao, 2001–2002, vol. 1: 179–180; Chiu, 2008: 207). Another Spanish Dominican missionary, Diego Francisco de Aduarte, recorded his observations of the Basay in 1634. These were published in 1640 in Historia De La Provincia Del Sancto Rosario (History of the Province of the Holy Rosary). Like Esquivel, he references the superstitions to which the parents and elders of the Basay are very much given (‘las supersticiones a que son muy dados sus padres y mayores’) (Borao, 2001–2002, vol. 1: 220). Aduarte does not, however, refer to any deities of the Basay. Mention of ‘superstitions’ by both Esquivel and Aduarte does, of course, betray their Christian view on the Basay belief system.

Based on Esquivel’s account, we can at least say that the Basay believed in spirits. The fact that neither Esquivel nor Aduarte mentions deities does not necessarily mean that the Basay did not believe in them. However, the explicit mention of a lack of sacrifices by Esquivel may be suggestive of this. Currently, we can say nothing more concrete.

7 Conclusion

This article has examined what authors in the seventeenth century wrote about the religious beliefs of the Formosan ethnic groups with whom they had contact. One significant methodological problem we face is that although the authors attempted to describe the entities in which the Formosans believed as accurately as possible, very distinct religious and linguistic forces influence how they perceive and depict these entities. An important element in this story has been avoiding the ‘topoi’ of otherness. The non-Formosan authors are marked by their heterogeneity and come from various historiographical traditions (Jenco & Tremml-Werner, 2021: 219). One, Chen Di, was a Chinese military adviser and scholar informed by religious systems quite different from those he encountered in Taiwan. Others were voc employees. Some were merchants, while others were missionaries who viewed the beliefs of the Siraya and Favorlangers through the lens of their own religious experience, informed primarily by Reformed Christianity. As well as Dutch, one author was a Scot, David Wright, who, we might assume, was a Calvinist but about whom we know very little. Another author, Elie Ripon, was Swiss. He, too, was probably a Protestant, although he writes little about the religion of the Siraya. Our information about the Basay comes from the reports of Spanish Catholic missionaries, above all the Dominican Jacinto Esquivel, who viewed the religious beliefs of the Formosans through yet another lens. Despite this, all these authors tried to fit their observations into their own ‘existing knowledge grids’.

The article has tried, nevertheless, to mitigate what Jenco and Tremml-Werner (2021: 237) describe as ‘serious biases in their accounts’ by understanding and interpreting the religious and linguistic forces at work in these authors’ accounts, above all by placing their descriptions of the entities in which the Formosans believed within the framework of the typology of spiritual entities proposed by Philippe Descola. Another issue is that the sources available for each Formosan group vary in quantity. While our information about the Basay is based above all on the observations of one Spanish missionary, the information for the Siraya is based on observations by a Chinese, several Dutch, a Scot, and a Swiss. This range of nationalities points both to the heterogeneity of these authors and reminds us that they wrote in what I have termed different allolects: wenyan, Dutch, and French.

Considering these factors, we can nevertheless conclude from this analysis that Formosan religious belief is marked by its diversity. Whereas the Siraya worshipped a pantheon of deities and had other deities besides, the Favorlang worshipped one supreme deity, Haibos, which embodied both good and evil. As for the Basay, Esquivel, writing in another allolect, Spanish, does not refer to the worship of a pantheon or supreme deity but rather to the many ídolos (idols) of the Basay. As I have suggested, this does not refer to idols in the sense of material objects but rather to what Descola describes as ‘presences’ connected to the spirit world, which they use to regulate their lives, such as the chirping of a small bird. This belief has something in common with the sounds of birds used for ornithomancy by the Siraya and Favorlang. Indeed, apart from the named deities, the Formosan cultures shared a general belief in a world inhabited by good and bad spirits. ‘Superstitions’ framed as omens and taboos or laws also played a role in regulating interaction with the spirit world, with shamans such as the Siraya inibs and Basay majuorbal acting as gatekeepers to that world.

As for ancestor worship, the case is less clear. Ancestors may be covered by Chen’s use of the term shen, but this is such a broad term that we cannot be certain of this. Nevertheless, the analysis of modern-day Formosan religious practices does point to ancestor worship, and we could tentatively read this back into the historical context. I have suggested that the absence of ancestor worship from the Dutch texts may be a function of the fact that they focused on the deities that the Formosans worshipped.

Interestingly, two of the three ethnic groups analysed in this article, the Siraya and the Basay, have languages which have been placed in the same language group, East Formosan, by Robert Blust (1999: 45).48 If Blust is correct, then this illustrates that even though some groups were related regarding one aspect of their culture, in this case, language, they diverged in another regard, that is, religious belief. Finally, what this article has attempted to do is illustrate that, perhaps as a result of the conceptualisation of the ‘other’ as homogeneous, though one might expect Formosan religious beliefs to be similar, they were, in fact, quite diverse (Williamson, 2014: 123). This has, in part, resulted from the authors under review avoiding what Geoff Wade describes as the ‘topoi’ of otherness (Wade, 1997). One possible explanation for this heterogeneity, suggested in the case of the Siraya, is that it may reflect a diversity of social structures.

Concluding, it is hoped that the analysis presented in this article can be placed alongside analyses of Formosan religions using other methodologies, such as those of anthropology and archaeology, to build a more complete picture of religious belief in ‘Aboriginal Taiwan’.

Acknowledgements

This article has been supported financially by a grant from the Chiang Ching-kuo Foundation (rg002-U-19).

Notes on Contributor

Christopher Joby is Research Associate at the Centre of Taiwan Studies, soas, University of London, and a Visiting Scholar at the Institute of Linguistics, Academia Sinica, Taipei. Much of his research focuses on the intersection of the Dutch language and culture and other languages and cultures in a historical context. He has also published articles on theology and a book: Calvinism and the Arts: A Re-assessment (Peeters, 2007). He is writing a book on the reception of the Christian Gospel in seventeenth-century Taiwan, under contract with Brill.

References

  • Adelaar, K. Alexander (2011) Siraya: Retrieving the Phonology, Grammar and Lexicon of a Dormant Formosan Language, Berlin: De Gruyter Mouton.

    • Search Google Scholar
    • Export Citation
  • Aduarte, Diego Francisco de (1640) Historia De La Provincia Del Sancto Rosario de la Orden de predicadores en Philippinas, Iapon, y China [History of the Province of the Holy Rosary of the Order of Preachers in the Philippines, Japan, and China], Manila: Beltran.

    • Search Google Scholar
    • Export Citation
  • Ahuan, Kaisanan (2019) ‘The Central Taiwanese Plains Indigenous Peoples absent from textbooks’, Indigenous Sight (26 December): 3039.

    • Search Google Scholar
    • Export Citation
  • Andrade, Tonio (2005) ‘Pirates, pelts, and promises: The Sino-Dutch colony of seventeenth-century Taiwan and the Aboriginal Village of Favorolang’, Journal of Asian Studies 64(2): 295321.

    • Search Google Scholar
    • Export Citation
  • Andrade, Tonio (2009) How Taiwan Became Chinese: Dutch, Spanish and Han Colonization in the Seventeenth Century, New York: Colombia University Press.

    • Search Google Scholar
    • Export Citation
  • Blussé, Leonard (2003) ‘The eclipse of the Inibs: The Dutch Protestant Mission in 17th century Taiwan and its persecution of native priestesses’, in Chuen-Rong Yeh (ed.), History, Culture and Ethnicity: Selected Papers from the International Conference on the Formosan Indigenous Peoples, Taipei: smc Publishing, 7188.

    • Search Google Scholar
    • Export Citation
  • Blussé, Leonard and Roessingh, M. E. (1984) ‘A visit to the past: Soulang, a Formosan Village Anno 1623’, Archipel 27: 6380.

  • Blussé, Leonard; Everts, Natalie; and Frech, Evelien (eds) (1999–2010) The Formosan Encounter, 4 vols., Taipei: Shung Ye Museum.

  • Blust, Robert (1999) ‘Subgrouping, circularity and extinction: Some issues in Austronesian comparative linguistics’, in Elizabeth Zeitoun and Paul Jen-kuei Li (eds), Selected Papers from the Eighth International Conference on Austronesian Linguistics, Symposium Series of the Institute of Linguistics (Preparatory Office) No. 1, Taipei: Academia Sinica, 3194.

    • Search Google Scholar
    • Export Citation
  • Borao, José (2001–2002) Spaniards in Taiwan, 2 vols., Taipei: smc.

  • Borao, José (2020) The Spanish Experience in Taiwan 1626–1642: The Baroque Ending of a Renaissance Endeavor, 2nd ed., Hong Kong: Hong Kong University Press. First edition 2009.

    • Search Google Scholar
    • Export Citation
  • Bowman, Marion (2003) ‘Vernacular religion and nature: The “Bible of the Folk” tradition in Newfoundland’, Folklore 114(3): 285295.

    • Search Google Scholar
    • Export Citation
  • Calvin, Jean (1961) Institutes of the Christian Religion, ed. John T. McNeill, trans. Ford Lewis Battles, 2 vols., London: scm Press.

  • Campbell, William (1896) The Articles of Christian Instruction in Favorlang-Formosan Dutch and English from Vertrecht’s Manuscript of 1650 with Psalmanazar’s Dialogue between a Japanese and a Formosan and Happart’s Favorlang Vocabulary, London: Kegan Paul.

    • Search Google Scholar
    • Export Citation
  • Campbell, William (1903) Formosa under the Dutch, Described from Contemporary Records, London: Paul Kegan. Reprinted 1992 in Taipei.

  • Cauquelin, Josiane (2004) The Aborigines of Taiwan: The Puyuma – From Headhunting to the Modern World, London: Routledge.

  • Chan, Wing-tsit (1963) A Source Book in Chinese Philosophy, Princeton, NJ: Princeton University Press.

  • Chang, Hsun (2017) ‘A resurgent temple and community development: Roles of the temple manager, local elite and entrepreneurs’, in Hsun Chang and Benjamin Penny (eds), Religion in Taiwan and China: Locality and Transmission, Taipei: Institute of Ethnology, Academia Sinica, 293332.

    • Search Google Scholar
    • Export Citation
  • Cheng, Weichung (2013) ‘淺釋牧師甘迪留斯論述 (Discours) 版本與標題之流變’ [A brief account of diverse versions and titles of Rev. Candidius’ ‘Discourse’], 臺灣學研究 [Taiwan Studies] 15: 171182.

    • Search Google Scholar
    • Export Citation
  • Chiu, Hsin-hui (2008) The Colonial ‘Civilizing Process’ in Dutch Formosa: 1624–1662, Leiden: Brill.

  • Dapper, Olfert (1670) Gedenkwaerdig Bedryf der Nederlandsche Oost-Indische Maetschappye op de Kuste en in het Keiserrijk van Taising of Sina… [Memorable Activity of the Dutch East India Company on the Coast and in the Empire of Great Qing or China…]. Amsterdam: Jacob van Meurs.

    • Search Google Scholar
    • Export Citation
  • Dean, Kenneth (1998) ‘Transformations of the she 社 (altars of the soil) in Fujian’, Cahiers d’Extrême-Asie 10: 1975.

  • Descola, Philippe (2013) ‘Presence, attachment, origin: Ontologies of “incarnates”’, in Janice Boddy and Michael Lambek (eds), A Companion to the Anthropology of Religion, Oxford: Blackwell, 3549.

    • Search Google Scholar
    • Export Citation
  • Descola, Philippe (2014) Beyond Nature and Culture, trans. Janet Lloyd, foreword by Marshall Sahlins. Chicago: University of Chicago Press. English translation of Par-delà nature et culture, Paris: Gallimard, 2005.

    • Search Google Scholar
    • Export Citation
  • Durkheim, Émile (1915) The Elementary Forms of the Religious Life: A Study in Religious Sociology, trans. Joseph Ward Swain, London: George Allen & Unwin. English translation of Les formes élémentaires de la vie religieuse, Paris: F. Alcan, 1912.

    • Search Google Scholar
    • Export Citation
  • Fang, Hao (1956) ‘陳第東番記考證’ [Text-critical analysis of Chen Di’s Record of Formosa], Bulletin of the College of Arts of Taiwan National University 7: 4176.

    • Search Google Scholar
    • Export Citation
  • Ferrell, Raleigh (1969) Taiwan Aboriginal Groups: Problems in Cultural and Linguistic Classification, Taipei: Institute of Ethnology, Academia Sinica.

    • Search Google Scholar
    • Export Citation
  • Feuchtwang, Stephan (1978) ‘School temple and city god’, in Arthur P. Wolf (ed.), Studies in Chinese Society, Stanford, CA: Stanford University Press, 103130.

    • Search Google Scholar
    • Export Citation
  • Funk, Leberecht (2014) ‘Entanglements between Tao People and Anito on Lanyu Island, Taiwan’, in Yasmine Musharbash and Geir Henning Presterudstuen (eds), Monster Anthropology in Australasia and Beyond, New York: Palgrave MacMillan, 143159.

    • Search Google Scholar
    • Export Citation
  • Gesterkamp, Lennert (2020) ‘Red-haired barbarians: The Dongxi Yangkao (1617) and its portrayal of the Dutch in China’, in Thijs Weststeijn (ed.), Foreign Devils and Philosophers: Cultural Encounters between the Chinese, the Dutch and Other Europeans, 1590–1800, Leiden: Brill, 5781.

    • Search Google Scholar
    • Export Citation
  • Gravius, Daniël (1662) Patar ki tna-’msing ki christang, Ka Tauki-papatar-en-ato tmaeu’ug tou Sou Ka Makka Si-Deia. Ofte ’t Formulier des Christendoms. Met de Verklaringen van dien inde Sideis-Formosaansche Tale [The Formulary of Christianity: With Explanations of it in the Sideis-Formosan Language], Amsterdam: Michiel Hertogh [in Siraya and Dutch].

    • Search Google Scholar
    • Export Citation
  • Grothe, J.A. (1884–91) Archief voor de geschiedenis der Oude Hollandsche Zending [Archive for the History of the Old Dutch Mission], Vols. 1–5, Utrecht: C. van Bentum.

    • Search Google Scholar
    • Export Citation
  • Happart, Gilbertus (1650) Woord-boek der Favorlangsche taal, waarin het Favorlangs voor, het Duits achter gestelt is [Dictionary of the Favorlang Language in which Favorlang is Given First and Then Dutch], unpublished manuscript.

    • Search Google Scholar
    • Export Citation
  • Happart, Gilbertus (1840) Dictionary of the Favorlang Dialect of the Formosan Language…, trans. W. H. Medhurst, Batavia, printed at Parapattan.

    • Search Google Scholar
    • Export Citation
  • Happart, Gilbertus (1842) Woord-boek der Favorlangsche taal, waarin het Favorlangs voor, het Duits achter gestelt is door Gilbertus Happart [Dictionary of the Favorlang Language, in which Favorlang is Given First and Then Dutch by Gilbertus Happart], Batavia: Bataviaasch Genootschap van Kunsten en Wetenschappen.

    • Search Google Scholar
    • Export Citation
  • Hernández, Vicente S. (1996) History of Books and Libraries in the Philippines 1521–1900, Manila: National Commission for Culture and the Arts.

    • Search Google Scholar
    • Export Citation
  • Heylen, Ann (2001) ‘Dutch language policy and early Formosan literacy (1624–1662)’, in Wei-ying Ku (ed.), Missionary Approaches and Linguistics in Mainland China and Taiwan, Leuven: Leuven University Press, 199252.

    • Search Google Scholar
    • Export Citation
  • Hu, Tai-li (2017) ‘The decline and revitalisation of shamanic traditions in a Paiwan village in Taiwan’, in Hsun Chang and Benjamin Penny (eds), Religion in Taiwan and China: Locality and Transmission, Taipei: Institute of Ethnology, Academia Sinica, 333364.

    • Search Google Scholar
    • Export Citation
  • Hu, Baozhu (2020) Believing in Ghosts and Spirits: The Concept of Gui in Ancient China, Abingdon: Routledge.

  • Huang, Ying-kuei (1988) ‘Conversion and religious change among the Bunun of Taiwan’, PhD thesis, University of London.

  • Huang, Shuin-wey (2008) ‘Coin and healing ritual among the Amis in Taiwan: State, images of others, and socio-cosmic order in early 1930s Iwan’, Journal of Ritual Studies 22(2): 2335.

    • Search Google Scholar
    • Export Citation
  • Jenco, Leigh (2021) ‘Chen Di’s record of Formosa: An alternative Chinese imaginary of otherness’, Historical Journal 64(1): 1742. Retrieved 19 February 2022 from https://www.cambridge.org/core/services/aop-cambridge-core/content/view/24FEA686C6AD63301A680EA20EF741DD/S0018246X1900061Xa.pdf/chen_dis_record_of_formosa_1603_and_an_alternative_chinese_imaginary_of_otherness.pdf.

    • Search Google Scholar
    • Export Citation
  • Jenco, Leigh and Tremml-Werner, Birgit (2021) ‘Historiography of the other: Global history and the Indigenous pasts of Taiwan’, International Journal of Taiwan Studies 4(2): 218247.

    • Search Google Scholar
    • Export Citation
  • Kang, Peter (2020) ‘A brief note on the possible factors contributing to the large village size of the Siraya in the early seventeenth century’, in Leonard Blussé and Yung-ho Ts’ao (eds), Around and about Formosa: Essays in Honor of Professor Ts’ao Yung-Ho, reprint, Taipei: Ts’ao Yung-ho Foundation for Culture and Education, 111127. Originally published 2003.

    • Search Google Scholar
    • Export Citation
  • Kennedy, Duncan (2006) ‘Afterword: The uses of “reception”’, in Charles Martindale and Richard F. Thomas (eds), Classics and the Uses of Reception, Oxford: Blackwell, 288293.

    • Search Google Scholar
    • Export Citation
  • Kottak, Conrad Phillip (1974) Cultural Anthropology, New York: Random House.

  • Li, Paul Jen-kuei (2010) Studies of Sinkang Manuscripts, Taipei: Institute of Linguistics, Academia Sinica.

  • Li, Paul Jen-kuei (2019) Text Analysis of Favorlang, Taipei: Institute of Linguistics, Academia Sinica.

  • Lin, Changhua (2014) ‘Christian contextualization in Formosa: A remarkable episode (1624–1662) of reformed mission history’, PhD thesis, Vrije Universiteit Amsterdam.

    • Search Google Scholar
    • Export Citation
  • Littlejohn, Ronnie L. (2009) Daoism: An Introduction, London: I.B. Tauris.

  • Liu, Pi-chen (2011) ‘Encounters with deities for exchange: The performance of Kavalan and Amis shamanic chants’, Shaman 19(1–2), 107128.

    • Search Google Scholar
    • Export Citation
  • Montanus, Arnoldus (1671) Atlas Chinensis: Being a Second Part of a Relation of Remarkable Passages of Embassies from the East-India Company of the United Provinces, to the Viceroy Singlamong and General Taising Lipovi Konchi Emperor of China or East-Tartary, London: White Fryers.

    • Search Google Scholar
    • Export Citation
  • Montanus, Arnoldus; Dapper, Olfert; and Ogilby, John (1671) Atlas Chinensis, Nederlandsche Oost-Indische Compagnie, United Provinces of the Netherlands. Retrieved 12 January 2022 from https://quod.lib.umich.edu/e/eebo/A36730.0001.001/1:7.1?rgn=div2;view=fulltext.

    • Search Google Scholar
    • Export Citation
  • Müller, Andreas [pseud. Thomas Lüdecke] (1680) Oratio Orationum [Prayer of Prayers], Berlin: Officina Rungiana.

  • Primiano, Leonard N. (1995) ‘Vernacular religion and the search for method’, Western Folklore 54: 3756.

  • Ripon, Elie (1990) Voyages et aventures du capitaine Ripon aux grandes Indes: journal inédit d’un mercenaire (1617–1627) [Travels and Adventures of Captain Ripon to the Great Indies: Previously Unpublished Journal of a Mercenary (1617–1627)], ed. Yves Giraud, Thonon-les-Bains: Editions de l’Albaron.

    • Search Google Scholar
    • Export Citation
  • Scott, William Henry (1994) Barangay: Sixteenth-century Philippine Culture and Society/ William Henry Scott, Manila: Ateneo De Manila University Press.

    • Search Google Scholar
    • Export Citation
  • Shepherd, John R. (1984) ‘Sinicized Siraya worship of A-li-tsu’, Bulletin of the Institute of Ethnology, Academia Sinica 58: 181.

    • Search Google Scholar
    • Export Citation
  • Shepherd, John R. (1993) Statecraft and Political Economy on the Taiwan Frontier 1600–1800, Stanford, CA: Stanford University Press.

  • Shepherd, John R. and American Anthropological Association (1995) Marriage and Mandatory Abortion among the 17th-century Siraya, Arlington, VA: American Anthropological Association.