David Maybury-Lewis was fuming. He had just returned from a six-week absence to find that his collection of wamnhono had disappeared. “Where are my dance masks?” he demanded, referring to the collection he had spent weeks gathering for his doctoral work in anthropology. The masks were destined for museums back in Britain. Maybury-Lewis’s anger was only accentuated by his disdain for Warodi, the eldest son of the cacique in the A’uwẽ aldeia of Wedezé, and the man he suspected of stealing them. Looking back some years later, Maybury-Lewis recalled that Warodi “came into our hut and stood there with his chest chucked out like a pouter pigeon” before going “through the performance of fettering for something,” that obviously was nowhere to be found. When pressed again, Warodi “thought for a while. ‘The spirits must have taken them,’ he replied at last.” Incredulous, Maybury-Lewis wrote that Warodi himself had stolen the masks and sold them to someone passing through the community in Central Brazil.1

A few days later, however, the aldeia Elders’ council confirmed unanimously that, in fact, wazepari’wa had spirited away the masks. These malevolent beings who live at the western edge of the sky were widely known to steal wamnhono. Prodded on by his injured pride, and despite otherworldly interference, Maybury-Lewis set about gathering new masks, “a difficult and wearisome business” given that they were “not lightly parted with at the best of times and especially not now that the spirits had intervened so dramatically with my first consignment.”2 Two of the masks now lie, wrapped in plastic, in the collections of the Pitt-Rivers Museum.

What am I to make of the material and affective traces of Maybury-Lewis’s research as I conduct my own? After all, historians love a misbehaving anthropologist.3 Shall I assign his insistence on collecting, in the face of human and spectral resistance, as a moral failing? Tempting. But what would that make me, in my effort to mine his experience for my own purposes? Maybury-Lewis represented some interlocutors as stupid and arrogant—even as he admired others’ intelligence, strength, or political ingenuity.4 As I face his characterizations and attempt to cast him as a character in turn, I confront my own ambivalences.

How satisfying it might be to dismiss Maybury-Lewis for his inability to see that his accusations about the masks were misplaced, that they were indeed stolen by spirits. This was my first thought, with a sliver of self-righteousness: Respect A’uwẽ ontologies, where these spirits are to be feared. Instead, my memory betrayed me. It called up my 2018 participation in the spiritual initiation darini in the same aldeia that hosted Maybury-Lewis. The women invited me to join them running water to the initiates. We doused the boys, pouring cool liquid into their parched mouths as they danced for hours under the sun. But the spiritual guards chased us. They knocked gourds and bottles from our hands, crushing them into the earth beneath their feet.

I had to be taught to fear them, my hosts admonishing me to respect the danger of too close an encounter. While I could behave as directed, I could not sense the risk, even if I willed myself to know it. And so, to claim a conviction that the spirits stole the masks would be to avow a knowledge that is still beyond me. Instead, I sit with the importance of uncertainty: neither knowing better than Maybury-Lewis, nor asserting access to a spirit world that eludes me.

I followed Maybury-Lewis, but is it possible to trace his steps and draw on his words without reproducing harms? The political implications of his work haunt my own. As white and foreign researchers, we are academic kin in a shared colonial genealogy. Maybury-Lewis conjured Warodi in vivid prose as cunning, petulant, proud. I use his words, even if they say more about the anthropologist than the man who later became a revered leader of A’uwẽ political struggles. Maybury-Lewis wrote without considering that Warodi’s family might request a future historian bring back his works. “I didn’t know my brothers, my parents are in that book,” Sidówi told me, “Why are they in that book?”5 As I work to make these writings available in the aldeia, I ask, what practices of care are required to engage such a precedent?6 I am pressed not only to write more carefully about my hosts but also about the scholars that came before me. With Sidówi’s question, how could I maintain a mantle of completeness and objectivity? I embrace partial knowledge—some things are not mine to know.

I trailed the masks, arriving in Oxford on a blustery November day, planning to bring photographs and film back to the aldeia. But the wamnhono eluded me, inaccessible due to an ongoing reorganization of the collections. Or perhaps they hid for some other reason, one neither I nor Maybury-Lewis could fully understand.


David Maybury-Lewis, The Savage and the Innocent, 2nd ed. (Boston: Beacon Press, 1988 [1965]), 242, 243. For the accusation of Warodi, see David Maybury-Lewis, Akwẽ-Shavante Society, 2nd ed. (Oxford: Oxford University Press, 1974 [1967]), 253.


Maybury-Lewis, Savage and the Innocent (n. 1), 247.


As a field science and site of colonial and postcolonial encounter, histories of anthropology offer compelling case studies in affectively charged ethical and political grey zones. For influential examples, see Warwick Anderson, The Collectors of Lost Souls: Turning Kuru Scientists into Whitemen (Baltimore: Johns Hopkins University Press, 2008) and Henrika Kuklick, The Savage Within: The Social History of British Anthropology, 1885–1945 (Cambridge: Cambridge University Press, 1993).


As Audra Simpson has shown, anthropological texts, perceived by the state and academic gatekeepers as authoritative, can be consolidated into “regulatory bodies of knowledge.” These canons create an authenticating loop by which new representations of Indigenous Peoples are judged, often to the political detriment of those Peoples in the present. See Simpson, Mohawk Interruptus: Political Life Across the Borders of Settler States (Durham, NC: Duke University Press, 2014), 67–93.


Sidówi Wai’azase Xavante, interview with Rosanna Dent, trans. Tsuptó Buprewên Wa’iri Xavante, 4 June 2014, Água Boa MT.


“Care-oriented scholarship” can help define standpoints from which one can both critique and develop knowledge. This is the goal. See Maria Puig de la Bellacasa, Matters of Care: Speculative Ethics in More than Human Worlds (Minneapolis: University of Minnesota Press, 2017).