Ethnography in the colonial era is often seen as a ‘softer’ front of domination, as colonial powers sought to control local systems so as to better subsume them under imperial power. ‘Animism’ as a term came from such efforts, as colonial ethnographers sought to categorise and understand local ways of life. But such encounters were always two-sided; colonial powers were not only the observers, but the observed, and just as theories of animism from Tylor to Durkheim to Freud fit local practices into European systems of thought, animists understood colonial power in their own way. Within Southeast Asia, Kaj Århem and Guido Sprenger (Animism in Southeast Asia, 2016) point out how animist thought posits and structures hierarchies—foreigners and spirits, for instance, share a categorical relation; power in one sphere bleeds into power in others. This ‘transformative animism’—the term is Shepherd's—is that which integrates new encounters and events into its own logic.

This two-way (mis)understanding is what frames Shepherd's excellent Haunted Houses and Ghostly Encounters. The book is organised into two primary sections, dealing with ethnographies done in East Timor under the aegis of Portuguese colonial occupation and those done by ‘professional’ ethnographers, respectively. The former follows writings from colonial governors, missionaries, explorers and the like, and the latter largely deals with anthropological writing. Each presents a particular confluence of characters, each with their own expectations and assumptions, and how actions by the Portuguese (and others) revealed their own interpellation into Timorese animist systems of thought.

As befits work where indigenous accounts are difficult to access, Shepherd is insightful at reading between the lines of his Portuguese literary interlocutors, revealing misapprehensions and misunderstandings between Timorese (themselves a collection of strikingly diverse people) and their would-be hegemons in ways that subvert both Eurocentric readings of ‘primitive superstition’ as well as Orientalist accounts of indigeneity ‘in tune with nature’. For instance, in ‘The Missionary’, Shepherd describes moments when Timorese surrendered their lulik, or ‘sacred’ objects en masse in exchange for blessings, conversions, and Catholic interventions upon ‘the devil’. While these moments were, for missionaries, clear examples of locals ‘seeing the light’, Shepherd takes careful note of the way in which missionaries were treated and to which tasks they were directed to make the compelling case that it was the Christians, and not the Timorese, who were operating under their supposedly opposed religion's logic—here, instead of embracing the logic of the gospel as preached by Catholic missionaries, locals understood priests as representatives of another source of a recognisable power (‘white lulik’) that could be mobilised for the locals’ own good, for instance, against difficult spirits that local practitioners had a hard time rooting out. In a similar way, Portuguese objects, such as flags, were taken not only as signs of political conquest, but as the sacred objects of a related creed, objects that had lulik power that could be appropriated by the Timorese themselves for their own purposes. It is a concept that is immediately familiar to those working in the region (and, indeed, elsewhere) on religious (I use the term cautiously, noting that the category of ‘religion’ in such instances bleeds readily into all spheres of being) notions of power, and evokes the omnivorous and fundamentally curious nature of thought.

Haunted Houses abounds with these illustrative and evocative encounters. In another, an American anthropologist is inducted into Makassae legends about the ancestors, but in his inattention seats himself on the headstone of the first ancestor's grave, leading to a chain of events in which the cause of misfortune remains constantly shifting, but which resulted in a slow disillusionment with the supernatural implications of the anthropologist's presence amongst both the living and the dead. In another chapter, another anthropologist, Elizabeth Traube, is encouraged to assemble a kind of tome of local cosmology, but is faced with the fact that mythical accounts are highly personalised and disparate; indeed, what emerges is that the Mambai with whom Traube worked had been understood by others to have such a synthesis, but this had been lost generations ago. Traube's appearance, then, offers a kind of prophetic moment, one complicated by the sexual interest (and rivalry) that Traube provokes amongst her key interlocutors. Ethnographers’ presence and witting or unwitting actions provoke animist reactions, new rituals to avert potential catastrophe from their blunders, the selective covering of particular lulik objects, or even the elevation of particular objects as local observers note what is of particular interest to foreigners.

The point here is to illuminate how ethnographers become a part of a story, as catalysts for animist events through the negative effects of, for instance, breaching taboos, or the more productive but still complicated promise of a revitalisation or validation of local ideas. Such a careful and humanistic reading of both sides of such an encounter recalls Nils Bubandt's The Empty Seashell (2014), which, in nearby eastern Indonesia, describes how Christian conversion does not ease the fears of witchcraft (as missionaries often promise it will), but presents new ontological concerns, inflaming doubt in ways wholly unpredicted, and, indeed, unnoticed by agents of conversion. Shepherd's careful reading is further helped by his writing style; in a lively and often comic manner, he interjects to suggest likely misunderstandings or omissions between the lines, constantly reminding us of the profound humanness of both the colonial subjects as well as their agents, and the limits of supposed-to-be totalising thought, especially in the too-neatly packaged written works that emerge from ethnography.

Haunted Houses also recalls the work of Marisol de la Cadena in understanding the ‘pluriverse’ of political actors in local politics, where colonial (national, in De la Cadena's case) officials, local leaders, and spirits themselves each act, though the nature of their influence (or, indeed, their actual presence) may not be accepted by all involved. We have here a play of symbols where the meaning and implications of each are not commonly held or fixed, but which impact how future encounters will subsequently be understood. Indeed, Shepherd, pulling from ethnographic work by Judith Bovensiepen, points out how decades of conscious mimicry of animist forms of power on the part of colonial officials paved the way for later anthropologists’ misapprehensions in the communities in which they worked—Portuguese officials had already established the lulik nature of foreign officials, and thus ethnographers presented local groups with yet more foreigners, who acted in radically different ways (and thus presented different dangers).

In so doing, Shepherd complicates notions of ethnographic authority, colonial power, and indigenous knowledge—none of these are absolute. In reading Haunted Houses, I am reminded of those same encounters through which I was ‘incorporated’ into a community: an identification with an old woman's son, dead by suicide; an affirmation of close friendship (siao) with a fisherman; a religious pledge to be a friend to another man's son when I returned to the capital; an affiliation with a political NGO; and, to echo Traube here, multiple offers of marriage (often extended by older relatives regarding women I had never met). Each of these, too, do not come from an abstracted ‘community’, but from individuals seeking their own connection and understanding with and of the ethnographer. But what are the histories and implications of such bonds? Shepherd addresses these questions and more by drawing not only upon the situations he presents, but also with a deep knowledge of debates within anthropological theory and an eye towards what a historical perspective can provide.

Shepherd's book, then, has value on multiple levels, and I enthusiastically recommend it to anyone interested in histories of ethnography, animism, or reflexive or ontological anthropology. It brings history into relevance for anthropology, and anthropology into relevance for historical research. Additionally, it is eminently readable, and each chapter can productively be used on its own in an undergraduate setting (to do so, of course, would be to miss his structuring insights, which would be a shame). Overall, the book is a sensitive and lively reading of history against the grain, of understanding the multiple ways that people seek to understand the role of the other, the unknown, the possible, and the unseen.