Everyday enacting of agents through bodily simulation , voicing , and familiarization of artifacts among the Arabela ( Peruvian Amazonia )

A habit caught my attention during my fieldwork among Arabela of the Peruvian Amazon – they consistently announced things they were about to do or commented on things others were doing . Moreover, such utterances were often aligned with witty references to other people’s peculiar behaviors or verbal expressions . The following is an example of such a humorous announcement of one’s action . On a chilly and humid morning Nuria, the eight-year old niece of my hosts, came to our house to return the shotgun her father borrowed from me the day before . She came wearing her father’s rubber boots (way too big for her), because she probably did not want to touch the cold, wet grass .1 Nuria left the shotgun and went back home . Later that same morning Elena (my host and Nuria’s aunt) asked her son Levi to go and offer a share of that morning’s catch of fish to her mother (who lived with Nuria and her family) . The boy took the fish from his mother, but before descending from the house (Arabela houses are built on platforms), he sat on the floor, and said (laughing): voy hacer Nuria (“I am going to do Nuria”) and put on a pair of large boots belonging to his father or one of his elder brothers; then he went away .

Everyday enacting of agents through bodily simulation, voicing, and familiarization of artifacts among the Arabela (Peruvian Amazonia) Everyday enacting of agents through bodily simulation… A particular habit caught my attention during my fieldwork among Arabela of the Peruvian Amazon -they consistently announced things they were about to do or commented on things others were doing . Moreover, such utterances were often aligned with witty references to other people's peculiar behaviors or verbal expressions . The following is an example of such a humorous announcement of one's action .
On a chilly and humid morning Nuria, the eight-year old niece of my hosts, came to our house to return the shotgun her father borrowed from me the day before . She came wearing her father's rubber boots (way too big for her), because she probably did not want to touch the cold, wet grass . 1 Nuria left the shotgun and went back home . Later that same morning Elena (my host and Nuria's aunt) asked her son Levi to go and offer a share of that morning's catch of fish to her mother (who lived with Nuria and her family) . The boy took the fish from his mother, but before descending from the house (Arabela houses are built on platforms), he sat on the floor, and said (laughing): voy hacer Nuria ("I am going to do Nuria") and put on a pair of large boots belonging to his father or one of his elder brothers; then he went away .
Thus to perform an unusual, conspicuous action, the boy evoked an other (his cousin), and presented what he was about to do as typical of that other's way of being . To a reader (and to the young ethnographer I was then) acquainted with contemporary Amazonian anthropology, Levi's joke strikes a familiar chord, bringing forth the Amazonian notions of body, agency, and conviviality . In my paper I will start with Nuria's rubber boots and explore how Levi's speech act and other similar Arabela ways of managing particular actions and engaging with human and nonhuman environment reverberate with the Amazonian indigenous ontologies .
Contemporary Amazonian ethnology depicts an indigenous world of intense human/nonhuman interactions that often shape social relations according to different schemes of predation, gift or exchange (Descola 2013: 309-312) . In these ontologies, a human shaman communicates with spiritual masters of animals of plants (Murphy 1958: 13-17;Guss 1989: 130), a woman needs to contain vampiric inclinations of her manioc offspring (Descola 1994) and a newborn's father who breaks the couvade rules by going hunting, risks being kidnapped by peccaries and turned into one of their kind (Fausto 2001: 313) . Nonhuman or other-than-human persons are crucial protagonists of ideologies concerning principal areas of human existence and are also made present in ritual contexts (notably with the help of psychoactive substances, themselves endowed with personhood and independent agency [Déléage 2005]) . There is also a more general orientation towards the environment that favors perception of the presence and action of nonhuman persons . Amazonian peoples have a tendency to interpret events and phenomena affecting them and occurring in their environment as the effects of action of nonhuman persons: from features of landscape, and meteorological phenomena, to abundance (or absence) of game animals or uncanny sounds of the forest (see for instance : Surrallés, García Hierro, eds . , 2005) .
Another conceptual tool of regional ethnology (and a complementary aspect of Amazonian ontologies) is a particular notion of the body, which has been one of the privileged ways of access to the understanding of Amazonian socialities (Seeger et al . 1979;Londoño Sulkin 2017), taking prominent place in the conceptions of animism (as "physicality" [Descola 2013]), perspectivism (Viveiros de Castro 1998;Lima 1999), and the aesthetics of everyday (good) life (Overing 1989;Overing, Passes, eds . , 2000;Belaunde 2001;Overing 2003) . The Amazonian body is the principal factor of singularity of different collectives of human and nonhuman persons -as a multitude of beings in the world lead or potentially may lead a personal existence (have intentionality, agency, communicative capacity, etc . ) and establish or may establish social relations with each other, the body is what accounts for different affects and ways of being characteristic to their collectives/communities . And it is also a body in continuous fabrication and transformation (Vilaça 2005), which constitutes a challenge and an opportunity -on the one hand, it allows for creating consubstantial communities of persons (through physical proximity, commensality, and exchange of bodily fluids through sexuality [Gow 1991]) and for developing resistant and efficient bodies through incorporation of attractive or useful nonhuman characteristics and dispositions (Santos-Granero 2012); on the other, it requires a constant prudence in inevitable corporeal engagements with beings of the environment, as these may trigger transformation into a nonhuman other, synonymous with being captured by an alien community and incorporated into it (Vilaça 2002;Opas 2005;Fausto 2007) . In this paper, I would like to demonstrate how that tendency to enact distant person hoods and agencies and manage bundles of transferable affects that constitute the human and nonhuman bodies resurfaces in the most minute everyday human to human engagements within a community .
Arabela live in two villages (Buenavista and Flor de Coco) on the eponymous tributary of the Curaray river, itself a major tributary of the Napo River . They are descendants of a handful of survivors from the Rubber Boom period, descendents of one of the Zaparoan-speaking groups, inhabiting the vast region between Pastaza and Napo (Villarejo 1953: 164-170;Rogalski 2016b) . In the 1950's, they settled permanently on the riverbank of the Arabela River and joined the regional social and political system of Mestizo and Kechwa communities . Today the larger of the two villages -Buenavista -is an important regional centre, having two-level schooling and a well-equipped medical post . Arabela subsist on slash-and-burn horticulture, hunting, fishing and raise chickens and sometimes pigs . To earn money, those Arabela who are not employed as teachers or nurses live on extracting natural resources (timber, fish and game meat) and seasonal labor for oil companies . The first language of communication is Spanish and only few persons are fluent in Arabela (some people speak also Kechwa) .

Enacting Others through behaviour
With this theoretical and geographical background in mind, I would like to return to the example of Levi, who presented the wearing of oversized boots as Nuria's typical (and emblematic) behaviour . From that moment (but, in this case, for a moment only -see other examples infra) "to do Nuria" became synonymous with wearing oversized boots . Levi, through his comment and action, endowed his cousin with a distinctive body and made wearing oversized boots into one of the elements of her "ethogram" (Descola 2013: 113) . That unit of behavioureasily identifiable and distinctive (because susceptible to singling out Nuria's particular way of being) may metaphorically be dubbed as an "etheme", by analogy to the linguistic notion of phoneme . By the same token, the behaviour also became available to Levi and his "micro-community", which was constituted of other witnesses of Nuria's behaviour and his enactment of her (these were Levi's parents, younger and older brothers, and the ethnographer) . From then on, they could potentially detach the behaviour from Nuria's body and use if for their own sake and purpose (just as Levi did) . 2 Arabela often perform such operations of making other individuals' behaviors into bodies and then use the elements of those bodies . When the extracted and adopted behaviour is not obvious, they make it explicit by giving additional explanations . For example Levi's older brother, Adan, while resting in a hammock, grabbed a copy of Selecciones (a Spanish version of Reader's Digest) that somebody had left on the floor and enacted their younger brother Emil saying: Voy hacer Emil -voy mirar puro dibujitos ("I am going to do Emil -I am going to look at the pictures only") . Then he started leafing through the magazine without reading it (it is likely that he could not read) . Whereas Emil's father, before going to the bushes to defecate, grabbed a machete and said: Voy hacer Emil -él no sabe ir al baño sin machete ("I am going to do Emil -he does not know how to go to the toilet without a machete") .
Sometimes Arabela apply the same operation to personal types, or members of other human or nonhuman collectives typically associated with some behaviors or dispositions . Thus, a youngster enthusiastically said: Voy hacer tomalón ("I am going to do drunkard", a personal type), and then went to serve himself a bowl of manioc beer (instead of waiting for a woman to serve it to him) . His brother once asked: Quién tiene lanza? Para hacer cashiquiori ("Who's got a spear? To do cashiquiori"), pretending that he would go out hunting with a spear, like the eponymous enemy group of ancient Arabela (see Rogalski 2016b) . On another occasion, his father Artemio, cleaning fish that fell into his net the night before, said to his wife, who worked with him and who spotted a half-rotten fish: Ponle aparte, para hacer mi gallinazo ("Put it aside to do my vulture") -announcing his intention of eating it . 3 The last example, where the Arabela man enacted a nonhuman, shows a continuity (grounded in the same set of practices) between adopting behaviors of other people (wearing oversized boots) and of nonhumans (eating rotten fish), which in a more direct way links the practice of "doing someone" to less casual, more objectified and often ritualized practices aiming at (more or less permanent) acquisition of characteristics typical of animals or plants . Among Arabela, as in other parts of Amazonia, these may be objectified in dietary or behavioral rules and other secretos, "secrets" . To give but one example, older women advise young mothers to put a pinch of powdered armadillo claw on a newborn's navel to transfer that animal's strength to the baby . I argue that the same logic, the same processes, and the same modes of engaging with the environment, underlie the above practices and ideas, only that here they operate on a micro-level of everyday interactions within one, cosmologically homogeneous, human collective and between closely related, "cosmologically identical" people .
As for the humans, Arabela enacted mostly children and older men -those in the prime of life (heads of important households and fathers to young men and women), as well as those in decrepit old age (divorced or widowers) . In general, women are not enacted -with the exceptions of old Julia (enacted in her supposed habit of not bathing, see infra) and a woman-type (voiced as enunciator of an interjection ay no!, see infra) -probably because the enactors seem to be mostly men (women often detect traces of other bodies in other people, but rarely enact others for their own sake) . 4 The dead people may also be enacted -I learned that when my companions, during a hunting trip upriver, one afternoon recalled that I was "doing Adolfo", a man from Buenavista, who died many years before: they referred to the fact that I walked barefoot carrying my boots under my arm . 5 It should be evident from the cited examples, that announcing one's action as taking over an Other's behaviour might be a kind of joking avoidance . 6 It is because conspicuous actions and behaviors -that become objectified, attached to another person's body and used by the subject -often break the Arabela rules of civility (which is particularly evident in the case of youngster grabbing unrestrainedly a bowl of manioc beer) . By making a witty comment and joking at the expense of the presumed source of that behaviour, the subject diverts other people's attention from his own engagement in the uncivil behaviour . Had Levi not announced that he would "do his cousin" and put on the oversized boots right away, he instead might have become the target of a teasing remark by his parents or brothers . It might have looked like the comment directed one day at a young man nicknamed Soldado ("Soldier") -he came from somewhere in the Amazon river and lived for some time in Flor de Coco, staying at our house (according to a rumor he was a defector from the army, hence his nickname) . That morning he was about to go out of the house and he started to put long trousers over his shorts . Observing him and seeing that, Levi exclaimed: Ua! Soldado está haciendo Murayari! ("Whou! Soldado is doing Murayari!"), and laughed at him .
Here, Murayari is an old man, who used to live in Flor de Coco (where the event took place) and was Levi's parents' brother-in-law (sister's husband to his father and sister's husband's father to his mother) . Levi presented wearing trousers over shorts as an element of that man's ethogram . His comment is a typical example of observational remarks through which Arabela detect presence 4 It should be noted, that the examples presented here are gender-biased, because I observed most of them during everyday life in my principal host household, composed of a couple with six sons (their two older daughters were already married and lived separately) . 5 The boots belonged to an Arabela man, who received them as part of personal equipment when he worked for an oil company . I borrowed them from him when we met on our way upriver, because my own boots somehow disappeared a few days before . Unfortunately, when we went to the forest, I soon had to take them off, because walking in them -too small and steel-capped -rapidly became an ordeal . 6 My use of the notion of joking avoidance should not be confused with that of Rupert Stasch (2002) . In his paper about the Korowai joking avoidance he used that term to describe an interpersonal relation that combined joking and avoidance: by calling each other humorous nicknames, two Korowai avoid using each other's names . Arabela, by making joking comments about other people, avoid being joked at by witnesses (see infra) . of other beings' characteristic behaviors in the people around them . It is worth noting that such remarks have a double objectifying character . First, in the content of the utterance, the target's existence in that moment (Soldado's projects, his reasons to put trousers, places he was about to go to), becomes reduced to a gesture (and at the same time the target is dispossessed of it by identifying its source as a distant Other) . It is easy to see here a repetition of the Amazonian scheme of mythical speciation -the process whereby undifferentiated primordial beings following some catastrophic events received the bodies they now have humans, animals, and plants that exist today (Viveiros de Castro 1998: 471-472, 2007 . This process also consists in a sort of petrification of beings into bodies with objectified and predictable behaviors . From a being freely developing his or her actions, not so much unconscious of his body, as simply bodiless, Soldado suddenly became embodied . The difference between the mythical embodiment and Soldado's is that the "embodifying" remark is targeted at two beings at the same time -the person whom the Arabela subject is observing (Soldado), and the absent body, that is presumably the source of the objectified gesture (Murayari) . It is not Soldado who -here and now, and from now on -comes to being as a distinct body; he becomes embodied as a copy of another body (Murayari's) .
One might ask whether there is a "mythical" origin of the body that wears trousers on shorts . Was there a primordial event, recorded in collective memory, transmitted in narratives, where Murayari dressed that way? I have never heard such a narrative (but I have also never had an opportunity to ask about it) . I am sure though that on the many occasions when I witnessed Arabela "doing someone" -either in the first or third person ("I am going to do Nuria", "Soldado is doing Murayari") -they never gave any narrative of origin . Declaration that one was "doing someone" was always made in a mode of shared knowledge, at most accompanied by additional explanation precisely identifying the affect or etheme in question . 7 Most probably that knowledge is acquired through participation in similar interactions and does not need to be accompanied by a narrative of origin . (I don't know if Levi ever saw Murayari don trousers on shorts . It is possible that his only knowledge of that particularity of his body came from participation in similar interactions -maybe he himself was once the butt of the joke?) .
Second objectification of a body-attributing comment is related to its pragmatic or conversational features, because the target is referred to and not talked to and therefore is not allotted a conversational slot for response . The effect is all the more powerful as the attribution of the body is expressed in an exclamative mode of surprise, marked by the opening interjection (Ua!) . Other such attributions that I observed were also introduced by interjections of surprise . Following Oswald Ducrot, I argue that the knowledge expressed in the exclamative mode (the identification of Soldado with long trousers worn on shorts, and the identification of long trousers on shorts with Murayari) is presented as a result of direct perception of the speaker, not mediated through speaker's intellectual, conscious reflection (Ducrot 1984: 186) .
While hacer alguien explicitly preserves the difference between the doer and the ethogram's owner -one is only "doing" the other -other linguistic forms make this identification stronger, by simply substituting the name of latter to the name of the former . A mother once exclaimed: Ua! Murayari!, when spotted one of her sons eating undercooked grilled fish (its blood was still visible along the spine), 8 and called another son "Julio Jaime", when noticed that he ate only the pieces of meat (presas) from his plate, without touching manioc nor broth . (Julio Jaime was an old man, about the same age as Murayari; he lived in Flor de Coco and was kumpa, ceremonial friend, to Levi's parents . ) This practice of name substitution links hacer alguien, as its transformation, to the Arabela onomastics, which is a subject I cannot develop here . On the other hand, the schema hacer alguien may be employed in negative sense, as an exhortation urging the addressee not to "do" or stop "doing" someone . For instance, the youngsters of Flor de Coco, when waiting for someone (a hunting trip companion) for too long, might call him from their canoe: No haga Sapo! ("Don't do Sapo!"), where Sapo ("toad") is a nickname of a man reputed to be an incorrigible slowcoach .
A comment is due here: as instances of joking avoidance and teasing, those interactions point to the Arabela structure of joking relationships, and tend to involve people in a particular relationship to each other (open-ended relations of potential affines, cross-cousins or locals vs . outsiders -see for instance De Vienne 2012) . I argue though that the Arabela other-enactions, albeit funny, may be better understood in relation to the Amazonian notions of body and agency, than as an illustration of a particular (general or local) theory of humor and joking practice . Although some other-enactions are purely humorous (and hence, for instance, might be linked to contexts of communal work or manioc beer drinking parties), they are, more often than not, tightly nested into people's everyday actions and utterances . The example of rubber boots is particularly telling -Levi's comment (Voy hacer Nuria) seems to be more about Levi, rubber boots and the people around him, than about his relationship with Nuria . Other examples, from a perspective of "joking relations", are far too complex to extract from them a particular, "joking" relationship . For instance, the comment about trousers worn on shorts may be tied to the relation between Levi and Soldado (beyond any doubt an open-ended relation), as well as to the relation between Levi's parents (particularly his father) and Murayari (Levi's father's ZH) . Besides, there are other more typical joking interactions, which are based on teasing, and Arabela would not necessarily point to other-enactions as examples of joking/laughing at someone (hacer broma/hacer reir a alguien) . Therefore, although there is some hint of teasing both in utterances of the Voy hacer alguien and Fulano está haciendo alguien types (as well as in the interactions discussed further), this paper is not intended as a paper about joking, but about acting, speaking, and naming things and people . 9 Enacting Others through voicing We saw up until now that Arabela are sensitive to behavioral idiosyncrasies of their relatives and neighbors . Now I wish to show that by the same token they also pay close attention to how people speak and what expressions they use . Here, they also exploit Others' vocal and verbal idiosyncrasies for various purposes .
In July 2009 I accompanied (and partly organized) a hunting trip upstream on the river Arabela . Its participants were: an old, experienced hunter Romario, knowing camping spots and hunting trails, his wife Julia, two teenagers (parallel cousins, MZS, grand-children of the woman, DS, one of them was also a paternal nephew of the man, BS), and me . In the afternoon of the third day of our trip we arrived at an old logging camp (set by a man from Buenavista) that consisted of a few shelters (tambo) and a house (casa) . 10 Romario, having chosen one of the shelters for himself and his wife, was about to put their luggage on the floor (bags containing mostly bedclothes and a mosquito net) . Consulting Julia where she would like them to have their bed, he asked: Dónde vamos mandar? -di Porojua (literally: "where are we going to throw? -says Porojua") .
To understand Romario's utterance and its context, we should note that mandar, "to throw", is not the verb Arabela commonly use to refer to making the bed . It is neither the standard expression (which is tender la cama), nor is it figurative, so his question was conspicuous and opaque and required us (me for sure, but I think that, actually, all of us present) to pay attention to what he was about to do and to look for clues . But, as we were still processing Romario's question and attributing its idiosyncrasy to him, after a short while he presented what he had just said as voicing a man nicknamed Porojua . (Porojua is an Arabela word for "worm, caterpillar" . The motive for that nickname is the man's reputedly insatiable appetite for tobacco . He is an old Arabela man, whose Spanish is sometimes difficult to understand, as it is influenced by the phonology of the Arabela, Porojua's first language . What adds an interesting twist to that interaction is that he is also Julia's former husband -it was Julia who left him for Romario . ) Thus, this kind of "reverse voicing" involves a process of reframing -an utterance (wording but also prosody) is first framed as originating from the speaker (Romario), but then, après-coup, it becomes reframed as voicing another utterer .
Arabela very often use this kind of reverse voicing, where they first utter an expression which stands out of the ordinary discourse, let it reverberate among the people present, and then assign the extravagant part of their discourse to a different utterer, with the reporting clause, framing it as voicing another utterer (see : Holt 2009: 194-195) . Most often, they employ this conversational move to produce interjections . For example, old Julio Jaime, in circumstances where a vivid reaction from his part was justifiable, used to say: Ay no! -dice la mujer . In that move, first he took a stance towards a state of affairs (Du Bois 2007), with an interjection indexical to feminine gender of the speaker (more precisely, to the speaker being a mestizo or ribereño woman from the Peruvian Amazonian region); that way, Julio Jaime not only showed his emotional engagement to the context, but also revealed, as it were, a body characteristic to another gender . After a short while, lowering his voice, he completed his move with the quotation segment assigning his foregoing utterance to a generic woman, and repositioning himself, taking a new stance this time towards mestizo women (Du Bois 2014) . (Other quoted interjections I recorded were for instance: Dios mío! -dice Lucho ["O my god! says Lucho"], or mildly vulgar La chucha! voiced through Rodrigo: La chucha! -dice Rodrigo . These interjections seem not to be enregistered voices, using the concept of Asif Agha, indexing stereotypic social personae, but indexically refer to individual persons, of Lucho and Rodrigo, respectively [Agha 2005: 39]) .
Arabela employ reverse voicing for the purposes ranging from joking avoidance (in lieu of euphemisms) to teasing . When Romario voiced Porojua, he evidently did it with the purpose of avoidance . He was not alone with his wife -her grandsons and I were present -and his utterance concerned an object related to their conjugal intimacy -their bed . By playfully mocking Porojua (pointing to his characteristic way of speaking) he diverted the attention of the people present from his action (and his dialogue with his wife), to a third subjectivity . Whereas the construction Ay no! -dice la mujer had a teasing character . The feminine interjection coming from a man would certainly provoke people to produce a teasing remark projecting on him a feminine identity . 11 Nevertheless, just before the rules of turn-taking (Sacks et al . 1974) would give the audience the space for a suitable teasing comment (or, in more technical terms, just before the transition-relevance place is established [Clayman 2013]), Julio Jaime anticipated their reaction and evoked the Other, whom he presumably just voiced . In both cases (joking avoidance and teasing), we may note an interesting transformation, where the speaker transitions from the position of a subject corporeally engaged in the world (Romario consulting his wife, Julio Jaime reacting to a situation), to the position of a meta-subject, making a meta-linguistic or meta-pragmatic (Agha 2006) comment on a virtual third subject embedded in the world (Porojua setting up a bed, a woman vividly positioning herself as disturbed by a state of affairs [Du Bois 2007]) .
If we put ourselves into the bystanders' position (bystanders who, in the end, become reframed as actual addressees), we realize that those interactions may have a real effect on their mental representation of the voiced figure (Porojua) . While the speaker proves his/her ability to discern characteristic elements of other people's idiolects and vocal habits, through reverse voicing he guides hearers' attention and exposes them to a particular perceptual and mental experience . The speaking subject -speaking to no one in particular (as Julio Jaime) or to an addressee (as Romario to his wife) -is outwardly unaware of the presence of the bystanders . He lets his speech divert from transparency and become salient index of his idiosyncrasy . He allows the bystanders to take notice of that transformation and "absorb" it, but just before they could stabilize it and attach it to some body-figure, which they could project on the speaker, he introduces an Other . Let's imagine the whole process in slow-motion: the bystanders first receive an observational input to build a new representation of the subject they watch and listen to . That input is epistemologically strong, because the bystanders observe the subject who is presumably not aware of their presence and attention . They are not bothered by requirements of an interaction in which they play an active part; they lower their guard . But then, just before the completion of the process of representation building, the subject suddenly substitutes the Other for himself . As a result, the observational input that had built up becomes attached to the evoked Other and feeds into that Other's image . Afterwards, the people who are present -at least the ethnographer -end up being persuaded that Porojua actually says mandar for spreading the bedmat, and women exclaim Ay no! (I will come back to that example later) .
But what about the second effect of that move? What about the representation of the speaking subject? I argue that he acquires a sort of complex identity . In the beginning, the speaking subject is just a being in the world among others -a man fixing a shelter for himself and his wife or sitting and drinking manioc beer with other people . His movements and words are coterminous with each other and with his identity . He is just another object inside the environment of the bystanders . Then, he suddenly starts to detach from that environment and becomes a salient figure sticking out from the background of interactions (Julio Jaime delivers the interjection, Romario says the word mandar) . Becoming something different, he attracts bystanders' attention . His movements and words still are coterminous, but do not correspond to his identity . Finally, at a third moment, he suddenly transcends the situation, and adopts an overarching point of view (saying the framing clause "says X") . As the whole process is relatively rapid, we could probably say that through a process of cumulative inclusion he occupies both positions: that of a being in the world becoming Other -some of his other-enacting (a part of the interjection Ay no!, a part of mandar) "sticks" to him -and that of a disembodied commenter . Would it be too far-fetched to see in that figure of a complex agent (cumulating the roles of doer and commenter) a variant of the shamanic complex enunciator cumulating contradictory roles (as the Kuna shaman analyzed by Carlo Severi)? 12 Familiarizing predation -transformation of objects into pet-subjects Until now, in the description of the relation between the Arabela speaker and his or her addressees or targets, I spoke of "objectification" both on a symbolic level (stabilizing their existence in bodies) and a conversational level (in the case of Soldado -speaking of him in third person, leaving him no space to retort) . Levi thus objectified Nuria, Soldado, and Murayari; Romario objectified Porojua . The flaw of such an interpretation is that it uses the Western category of object for the ethnographic context where the very existence of objects is problematic (Viveiros de Castro 2004) . In this section, I will propose a more suitable description of that relation, grounded in local Amazonian ontologies and relationalities, using the categories of familiarizing predation and mastery/ownership . These categories have a long tradition in the Amazonian ethnology, and their origins might be traced back to studies focusing on pet keeping and adoption (Erikson 1986(Erikson , 1987and Menget 1988 respectively, see: Allard 2019) . They were fully identified and developed by Carlos Fausto (as "familiarizing predation" and "mastery" [Fausto 1999[Fausto , 2008) and have become an important source of inspiration for scholars working in Amazonia (see for instance Costa 2017) . Briefly speaking, the notion of familiarizing predation -developed by Fausto (1999) in his analysis of warfare and shamanism -describes a process whereby vital elements coming from an exterior (acquired through warfare, shamanism or hunting) become contained, tamed and put at the service of the subject (a person or a group) . These may be captive enemies incorporated into the local group, shamanic powers acquired and controlled by the shaman or infant wild animals, captured during hunting, fed and raised as pets . Notions of ownership and/or mastery, which describe an asymmetrical relation between the masters and their pets/captives, are correlates of familiarizing predation .
I argue that a model of familiarizing predation and a relation of mastery/ ownership might also apply to what happens between Levi and Nuria, Levi and Soldado, or Romario and Porojua . First, the predatory component of that model can account for the agonistic dimension of Arabela joking at the expense of someone . Second, the idea of familiarization -acquiring something and taming it to put it to use by the subject -seems suitable to grasp the utility of acquired alien behaviors or voices . It is evident that Nuria's etheme of walking in oversized boots enables Levi to walk dry footed to his grandmother's house, and Porojua's particular way of talking about spreading the bed enables Romario to make reference to it in a context where it may otherwise be embarrassing to him . Third, the notion of capture, to which familiarizing predation is tributary, can also we have seen in the Mu Igala, who starts to sing about himself singing, is only the first (and, despite appearances, crucial) step in the same process that here becomes spectacular of accumulating contradictory identities of the image of the enunciator . The shaman then becomes a complex enunciator, a figure capable of lending his voice to different invisible beings" (Severi 2002: 36) . grasp the perceptual and cognitive ability to discern other people's idiosyncrasies of gestural or verbal expression . Instead of a lay anthropological theory of perception (using notions of discernment, recognition of patterns or similarities, etc . ) we could then use a notion developed on the ground of the Amazonian ethnology . 13 Besides, there is yet another set of enactions of other bodies/subjects, where the scheme of familiarizing predation and mastery/ownership is particularly evident . Here the other subject or body is enacted not as a co-agent (source of ethemes or vocal productions) but as a co-object of the speaker's action .
For some months of the 2008 in Flor de Coco there lived a group of mestizo construction workers, hired by an oil company to construct a medical post (as part of the compensation package for the firm's operations in the area) . Over the weeks, Arabela established bonds of friendship with the workers . Upon leaving the village, one of the workers, named Angel (diminutive: Angelito), left his blue sweatshirt to Artemio . One morning, sometime after the workers left, Artemio, choosing the clothing he would wear that day, picked up that sweatshirt and said: Voy a mudar mi Angelito, "I am going to wear my Angelito" (and laughed) . 14 Artemio's comment was partly a practice of joking avoidance . It seems to me that upon deciding to put on his sweatshirt, Artemio felt that it would be a conspicuous act that would call for evoking an other . It is very probable that for Artemio, the sweatshirt -as a gift from Angelito and as his former clothingwas imbued with its previous owner's subjectivity (a common phenomenon in the Amazonia, see for instance Santos-Granero 2012: 198) . Through his speech act, Artemio performed a double operation of subjectivation and familiarization . First, he subjectified the sweatshirt which -by referring to an item of clothing with the name of its former owner -became a "pet-person", so to speak, under the control of Artemio . Second, he put the construction worker into the asymmetrical relation owner-owned, humorously transferring his material and palpable mastery over a piece of clothing to his relation to (now absent) Angel . That second operation occurred specifically by coupling Angel's name with the first person possessive pronoun "my" -it is important to note that the Arabela parents often refer to their children precisely in that way (Artemio commonly referred to his sons as "my so-and-so"), and in Amazonia the relation of owner-owned often designates the relation parent-child and almost always the relation between parents and adoptive children (Fausto 2008: 333) . Hence Angelito -through his sweatshirt -became, as it were, Artemio's adoptive child . 15 13 A discursive and pragmatic dimension of the notion of mastery has already been underlined by Andrea-Luz Gutierrez Choquevilca in her analysis of ritual songs of Quechua Runa of the Upper Pastaza, where she observed a "coincidence between acts of musical nomination shutiyachina and the control acquired over designated entities" (Gutierrez Choquevilca 2016: 27 -translation from Spanish by the author) . 14 In Amazonian Spanish the verb mudar -that in standard Spanish refers to changing the clothes (take off one shirt and put on another) -is often used to the very action of putting an item of clothing . 15 This example may have political connotations, indexing disparities in access to wealth (clothing) . Nevertheless, "wearing others" also involves situations where people comment Just as "doing" and voicing Others, the comic evoking of persons to refer to artifacts was a common element of the local poetics of everyday life . The grounds for such identification of artifacts with people were multiple (all pertaining to metonymy) . First -the actual and singular relation of ownership as in the identification of one particular sweatshirt with its previous owner . Second -the association of a class of artifacts with a person (or a class of persons) singularly related to one item of that class (as in an identification of any baseball cup with a man who presumably always wears a cap -see infra) . Third -the association of objects with persons derived from symmetrical association of persons with objects expressed in their nicknames . For example, just as Melania was nicknamed Mecha, "wick", on the ground of phonological similarity, her nephews, when they were about to replace a wick in a kerosene lamp, used to say: Vamos a cambiar su Melania ("We are going to change its Melania") . 16 What is important here is that although those comments focus on artifacts, we also may be dealing with an imputation of ethograms . It is evident in the example where a baseball cap is associated with a man from Buenavista village: a man left his baseball cap on a bench; another man noticed that and said to him: Acá está tu Roger ("Here is your Roger"), referring to the cap with a name of a man who (presumably) always wore a baseball cap . Since clothing is an intrinsic part of Amazonian notions of the body, relating the baseball cap to Roger is a way of constructing his body . This example is also very instructive, because it allows us to highlight the process whereby an imputation of a body occurs . First of all, Roger definitely was not the only Arabela man who wore a baseball cap . Today the association of Roger with his baseball cap (and with baseball caps in general) is evident to me and when I evoke his image in my mind, he is wearing that cap . I also remember that when I heard the man in question addressing the cap's owner Acá está tu Roger, I immediately grasped that association . As if the speaker made explicit something I had known but wasn't aware of . But at the same time, it was the only time during my fieldwork when the association of Roger with his cap came up . Thus, it makes me wonder what mental image of Roger I would have, had I not witnessed the Arabela man making that association in that specific interactional context . This auto-ethnographic introspection and reflection points to the processes involved in the process of body-recognition/body-imputation . If those interactional events effectively lead to the formation of associations on similarities of their clothing with analogous items owned by their relatives . For instance, a boy putting on his own red T-shirt said to another: Voy a mudar mi Eusebio, and clarified, that Eusebio's T-shirt has black sleeves . (I may attest that Eusebio often wore that T-shirt during my stay among Arabela) . See also the example of a baseball cap -infra. 16 Note that in the last example a pet-person ("its Melania") is not pet to a human, but to another artifact . Actually, it is the operation of subjectivation of the lamp (see infra for another example of transforming a lamp into a subject) . Apart from clothing, the objects that were subject to that operation included a fishing spear and body parts . The elements included in that category brings to mind the notion of inalienable possessions, frequent in the Amerindian languages - Kockelman 2009 . The (in)alienability of the Arabela nouns has not been the subject of linguistic study yet . Usually, the Arabela nouns are not marked when non-possessed, but a few are -for instance "heart" and "bone" . between people and particular ways of being, we may ask what pragmatic and conversational properties make them effective . Here, it may be a combination of some observable substrate (I assume that Roger does often wear a baseball cap) and a pragmatic effect . When referring to a cap with the name of Roger in a context of a joking interaction, the speaker takes that association simultaneously as sufficiently evident for him and his addressee as to be comprehensible (presupposition of a common ground), and sufficiently surprising as to fuel the comic effect . It is always a revelation of some secret knowledge about somebody . Maybe that is what contributes to the establishment of a particular mental image . (An image which is not only associating Roger with his cap, but also placing him in a relation of a "pet" towards the cap's owner) .
By the way, although the association involved in the above example seems to operate only on the cap and the mentioned Roger -Roger becomes a man with a cap, and a cap becomes his owner's Roger -the speech act may also have another effect on the cap's owner, which would emerge afterwards . After collecting the abandoned cap, did its owner feel comfortable putting it on? Or did he feel that he was becoming Roger against his will?
Introducing the notion of familiarizing predation as a model to account for the other-enacting practices, discussed here, opens up a question if all cases of enaction of other persons/bodies follow the same scheme of relation . Although this problem goes beyond the scope of this paper, I will just signal two examples that in my view correspond to other schemes . (I am using the analytic distinction between relational schemes of predation, gift, and exchange made by Philippe Descola [Descola 2013: 309-321]) . The first is a comment made by Levi (with whom readers are already familiar): while watching his older brother refilling a completely used-up lamp with kerosene, he said: Ay me alegro! -va a decir el lamparín ("Oh, I'm so happy -the lamp is going to say") . The boy did not capture lamp's voice for his own sake, but instead lent the lamp his own voice . His speech act makes explicit and completes a nurturing subjectivation of the artifact, implicit in the care provided to it by the older boy . Levi's brother carefully removes the lid, slowly pours the kerosene so as not to spill it, all the time paying close attention to his movements, etc . (All that in the context of scarcity, where kerosene is sometimes lacking; when there is no kerosene, people just go to bed at dusk, skipping relaxing conversations, visiting, and storytelling altogether . ) Hence, in this case, we may be dealing with a scheme of caring (familiarizing gift), rather than familiarizing predation . 17 Another event, symmetrical to this "gift of voice", might be dubbed a "gift of etheme" . It was a comment made by a young man while contemplating a baby 17 Guilherme Heurich (2018: 55), in his article on voicing speech among Araweté, addressed the problem of ambiguity inherent in voicing other figures and argued that Amerindian speech acts do not stress the speaker's individual intentionality, because for Araweté every speech act is reciprocal . Although his interpretation may hold for Araweté, I argue that "reciprocity" (related to the scheme of exchange in the model developed by Descola) is merely one possible relational scheme in relations between the speaker and the voiced figure . boy (his nephew, BS) sleeping late in the morning . 18 He said: Está durmiendo hombre grande, se ha ido linternear ("The big man is sleeping, he's been out hunting with flashlight") . Through this comment the baby boy receives the affect of 'sleeping late in the morning', which is part of the ethogram of Arabela hunters . And my fieldwork confirms the observational ground of that operation -a night hunt (where the hunters went downstream, illuminating the river banks in search for glowing eyes of animals, especially the paca, Cuniculus paca, the most prized night prey) was one of the very few reasons authorizing Arabela men to sleep at this hour, while others were up and active . Again, it was humorous but not intended to mock the baby . It was the subjectivation of the baby through prefiguration of his being a hunter in the future . (This may be considered to be yet another example of an inculcation of affects to small children in view of their future gendered roles . The main difference is that it does not underline preconditions for those roles -as in the old days, when the Arabela fathers gave their sons little stones found in hunted alligators' stomachs to swallow, in order to promote their resistance to hunger -but accessory, incidental elements related to them . )

Other-enacting formulas
The linguistic expressions Arabela use for enacting Others -whether through reference to ethograms, voicing, or referring to artifacts with personal namesamount to a limited set of more or less characteristic and specific general constructions . The expression hacer X, apart from announcements <Voy hacer X>, appears also in a constative version <X está haciendo Y> (in the past tense: <X estaba haciendo Y>) or in an exhortative <No haga X!> ("Don't do X!") . Two other structures couple the name of the enacted other with the adverbs ya ("already") and así ("thus", "this way") in the form of a rhetorical question . They are symmetrical and serve to mark a difference (hence, presupposing similarity) between the addressee or the speaker, respectively, and the enacted Other . For instance, when Romario announced his intention to go upriver for a hunting trip, his adult niece Luisa half-mockingly asked him to bring her some ungurahui palm fruits (Oenocarpus batahua) . Para qué? ("What for?") -he asked . Para comer ("To eat")she responded . Pinsha ya!? -exclaimed he, with a mock indignation, identifying her with a toucan (Ramphastos spp . ) on the grounds of a presumably excessive appetite for the ungurahui fruit (which are, by the way, commonly consumed by Arabela) . For the second construction: Artemio, finishing a copious breakfast after getting a particularly abundant catch of fish, vigorously pushed his plate away and exclaimed with a mock irritation: Qué vale peje, lobo así?! ("What is the fish worth? Am I an otter?!", Pteronura brasiliensis) .
The voicing formulas are the most uniform -they all instantiate constructions <[utterance], dice X> and, seldom, <como dice X, [utterance]> . Whereas other-enactions that refer to artifacts with personal names share the general pattern that the personal reference is always placed at the end (and is preceded by a possessive pronoun) (see: Voy a mudar mi Angelito, Acá está tu Roger, Vamos a cambiar su Melinda) .
On the whole, all those constructions -some more than others -constitute a rather limited set of general patterns to generate other-enacting formulas . While formulas of the type <Voy hacer X> have low formal specificity (apart from the joking-avoidance context of their use, their sole singularity consist in coupling the verb hacer with a personal name), others are highly specific, as e .g . the construction <X así para [do something]!?> . Of those I recorded three instances, which were: Catalán así, para comer pescado!?, Maquisapa así, para tomar en pate!?, and Abuelo así, para cargar aguaje!?, linking, respectively, kingfisher bird (Chloroceryle spp . or Megaceryle torquata) with eating fish, spider-monkey (Ateles spp . ) with drinking from old Arabela gourds (instead of steel or plastic bowls), and an old man (person-type) with collecting peach palm fruits (Mauritia flexuosa), always following a pattern of the rhetorical question: "Am I so-and-so, to do something!?" .
The last series of other-enacting formulas offers an insight into their historical stability, because the utterance Maquisapa así, para tomar en pate!? comes from an old woman's narrative about her brother -now about sixty -who, as a child, used it to jokingly refuse drinking from gourds (although I do not know if the boy said it in Spanish or in Arabela) . That particular construction, therefore, might have been in use for half a century . As for a possible continuity with more ancient Arabela practices, the only thing I could establish was through a query of the Arabela dictionary compiled by Rolland Rich, from the Summer Institute of Linguistics (Rich 1999), where I found a group of verbs composed of a term relating to an entity and the morpheme /-shi-/, which connotes the idea of "behaving like" . For instance, the verb maajishiniu, according to Rich, means "to do woman's work [by a man]" (since maaji means "woman"), whereas among the meanings of the verb mueyashiniu (mueya, "child"), he includes "to bother someone asking for things" and "to behave like a child" . Similar relation links the verb nososhiniu, and the term nosuna (an unidentified tree, with edible fruit): nososhiniu means, according to Rich, "to urinate on someone as a prank (like the nosuna fruit that suddenly falls from the tree)" . 19 The morpheme /-shi-/ also encompasses the idea of "pretending" or feigning, since numueejushiniu, apart from "becoming deaf" (numeeju), refers also to "being disobedient" (pretending to be deaf) . We should note, though, that the same morpheme also refers to definite changes, where there is no idea of pretending: jiyasoshiniu means "to become old" (from jiyaso, "grand-father") and mashajashiniu, means to atteint the young-adult age (mashaja, "youngster") .
It is important to stress the formulaic character of those expressions and the effect of repetition and parallelism produced by their frequent use in the flow of everyday life . This is due to a relatively limited number of constructions that serve to generate a multitude of other-enacting utterances, and an announcement Voy hacer Fernando!, exclaimed by someone one day, will thus be parallel to (or dialogically resonate with) No haga Manuel! uttered some days before, and will prepare ground for a Vamos a hacer Rodrigo! uttered sometime in the future (Du Bois 2014) . Moreover, some of those formulas are very often repeated -like for instance Voy hacer Julia, which was for some time used every couple of days by the youngsters of one of the household where I lived (to announce that they would not bathe that evening) . 20 Other structures -like Catalán así, para comer pescado -are used more seldom, but instantiate linguistic constructions so specific, that it is doubtful that Arabela would not objectify them as particular forms of discourse .
In the formulas of enacting the other and in their use, we may hear a distant but audible echo of the formulaicity of Amazonian discourse, as explored, for instance, by Pedro de Niemeyer Cesarino in his analysis of shamanic discourse (Cesarino 2008(Cesarino , 2015, recovering the concept of formula elaborated by Milman Parry and Albert Lord (Lord 1960) in their studies of oral poetry . In the strict sense, the formula of Parry and Lord, understood as "a group of words which is regularly employed under the same metrical conditions to express a given essential idea" (Parry cited in Lord, cited in Cesarino, 2015: 21;Cesarino, 2008: 155), does not cover other-enacting expressions, because they do not form part of a formally delimited oral text with distinctive metrical features . On the other hand, we may treat the flow of Arabela's everyday conversations as a joint text produced by multiple speakers over weeks, months or years, and thus, instead of metrical conditions, consider situational and pragmatic settings . Just as formal properties of discourses of Marubo shamans, according to Cesarino, convey particular knowledge about beings of the world (and give the shaman the possibility to act upon them), the other-enacting formulas of Arabela, jointly, through parallelisms and repetitions, express a general idea that beings of the world (human and nonhuman) may have particular ways of behaving and talking, which may be adopted by others to perform certain actions and reach certain goals .
It is noteworthy that Arabela -through their practices -maintain those formulas in a state of certain exteriority in relation to other parts of their language . They combine parallelism and repetition with constant inventiveness . In fact, with time other-enacting formulas become "worn out" and have to be transformed (though a sort of "recursive displacement" [Keane 1997: 129ss]) or replaced . For example, during my fieldwork the formula hacer Educo, which announced voracious eating -where Educo is a diminutive of Eduardo, a man mocked because of his insatiable appetite (there were rumors that his appetite was so overwhelming, that after finishing his meal he picked pieces of meat from his young daughters' plates) -has been substituted by a new formula: hacer eduqueada, where eduqueada is a neologism created through verbalization and nominalization of the expression hacer Educo (hacer Educo → eduquear [not attested] → hacer eduqueada, do "educo-ering") . But it is also a peculiar kind of invention, because Arabela never indicate authors of those expressions, nor relate them to their initial use . Every use of hacer Educo is a repetition of its former occurrence, but every act of hacer Educo is new and exceptional .
We may note that the above transformation confirms that the other-enacting is not a simple imitation of the other but an objectification of a way of being . In fact, there is a strong analogy here with the notion of "becoming-jaguar", developed by Eduardo Viveiros de Castro in his analysis of the famous repartee by Cunhambebe, a Tupinambá chief, to Hans Staden, who was captive among the Tupinambá (Staden 2008(Staden [1557) . During a cannibalistic banquet, which he witnessed, Staden disgusted at Cunhambebe offering him a piece of killed enemy, asked rhetorically, how could he eat human flesh, since even animals do not eat their own species, to which Cunhambebe succinctly retorted Jauára ichê, "I am a jaguar" (or literally, "jaguar me" [Viveiros de Castro 1986: 626]) . According to Viveiros de Castro, the "Cunhambebe's equation", do not "concern either an imaginary 'turning into a jaguar' or a mere 'acting like' a jaguar -jaguars do not cook . But perhaps it refers to a jaguar-becoming, where 'jaguar' is a quality of the act, not of the subject" (Viveiros de Castro 1992: 271 -original emphasis) . Interestingly, both cases -an Arabela saying vamos hacer Educo (or eduqueada), and Cunhambebe saying Jauára ichê -refer to an extraordinary act of eating, and also in both cases, the constatation of other-becoming is made in a similar, humorous vein .
Interestingly, the exteriority in relation to "plain" language, the salience of those formulas, corresponds to the idea that the behaviors they denote, or the speech acts they voice are exterior in relation to the speaker-subject and his or her body . The linguistic expression hacer Julia stands out from the announcement Voy hacer Julia! uttered by a young man, just like the action it denotes (the action of not going to bathe that particular evening) departs from the speaker's default way of being (i .e . bathing every evening, just like any other proper human being) . Also, just as the expression hacer Julia does not have an author (a person who coined that expression), the etheme it denotes does not have a unique owner (a person who extracted and used that part of Julia's body), and may be used by every member of the collective (household, kin group, etc . ) .

Arabela bodies and the morality of living together
A closer look at the kinds of behaviour Arabela objectify in their practices of enacting others, shows that they constitute a coherent ensemble, covering a limited range of bodily functions or domains of practice . The most frequent are references to eating: consuming raw or rotten food, having excessive appetite for certain kinds of food (meat, palm fruits), eating voraciously/drinking thirstily . Idiosyncrasies related to clothing are next in line: oversized boots, trousers worn over shorts, walking barefoot (while carrying boots under one's arm), always wearing a baseball cup, wearing other person's clothing (previously worn by others or gifts) . Less common are references to other body functions, ways of being and activities . (Table 1 summarizes the enacted beings and corresponding behaviors and speech acts . ) The attention paid to those aspects of existence corresponds to the Amazonian perspectival registers of difference, where bodily ethogram indexes the place of a being in the socio-cosmos (see for instance Vilaça 2005) . It also brings to mind many Amazonian myths and anecdotes of encounters between humans and a nonhumans appearing in a human form, where -as Philippe Descola observed -there is always some detail that alerts the human protagonist to the animal nature of people with whom he or she is dealing: "a dish of rotting meat politely served reveals vulture-people, an oviparous birth indicates snake-people, and a cannibalistic appetite points to jaguar-people" (Descola 2013: 135) . We could almost imagine a narrative where Artemio's predilection for rotten fish (see supra), assumed in its entirety by himself, and not through the captured and familiarized body of a vulture, would reveal his complete belonging to the "tribe/species" of vultures .
But the Amazonian human body, as many ethnographers observed, also incorporates rules of moral behaviour towards one's kin (Gow 1991;Opas 2005) . That aspect is also present in the Arabela practices of enacting others, although it is not so explicit as the stress on bodily functions and movements . Unlike Wauja, who in similar register use names as "moralizing signs", linking people with general rules of living together (denouncing, for instance, giving something and later taking a part of it back [Ball 2015: 347-349]) -in this respect, Arabela only criticized marital jealousy by referring to overprotective and over-controlling husbands with a name of a man reputed to be jealous . Nevertheless, an implicit Arabela morality of living together becomes visible when we analyze speech acts that Arabela voice through other enunciators . For instance, they often treated that way interjections (La chucha! -dice Rodrigo, Dios mío! -dice Lucho), as if strong expressions of (negative) emotions violated the harmony of being together . Arabela also voiced other enunciators to formulate requests, especially requests for sharing food . For instance, the older brothers of little Eusebio (two-year-old boy), when they asked each other to share a fish roasted by one of them, used the expression ibalada, produced once by Eusebio and being transformation of igual, meaning "together" (as in: comer igual, "eat together") . (One of them would just ask Ibalada? and join his brother at the hearth . ) Similarly, a wife would voice another enunciator to urge her husband to join her in her bed (see infra) . Other enunciators were used for rejecting offers . (For instance, on several occasions Andres rejected someone's offers to join him and do something together, saying: No quiero, di gringo, with exaggerated pronunciation of "r" in quiero and gringo, probably in reference to my own rhotacism . ) The utterance Dónde vamos a mandar? -dice Porojua, as I interpreted it before, also points to a notion of shame . Just like in other examples, voicing of another enunciator goes hand in hand with the use of humorous euphemisms, e . g . when a man, talking about a menstruating woman, said: estaba con su periódico, dice Pancho, "she had her periodical, says Pancho" (where periódico, "periodical, magazine", is a humorous euphemism for período, "menstrual period", on the ground of paronymy between those words) . Towards an Arabela economy of affects It was late evening and the members of my household were already under their mosquito nets . All but Adan, who installed the mosquito net for his wife Nadia and their baby boy, and then laid down in a hammock . After some minutes, she called him from under their mosquito net: Adan, ahí vas a dormir? No vas a venir a la cama, di Tamani? ("Adan, are you going to sleep there? aren't you coming to bed, says Tamani?") . Nadia's utterance, apart from being yet another example of oblique formulation of an embarrassing request, is interesting because of the identity of the voiced enunciator . Who was Tamani, voiced by Nadia? Tamani is one of the common family names among the Quechua from the Napo river . I met once a man with this family name, who visited the Arabela communities, but I had never heard any comments or other mentions of him . Some three or four years earlier, during my first stay among Arabela, I heard children playfully calling each other that name, but later I found no other references to it . The expression venir a la cama is not conspicuous neither in itself, neither Nadia pronounced it a way that might have sounded peculiar . In this case, it was the situation that made it salient -Nadia was asking her husband to join her in her bed, and other people (her in-laws) could hear her . Given my inability to link Tamani to any singular, real person, it seems to me, that Nadia's utterance was, in fact, an enaction of a sort of a "wildcard" enunciator, to whom one can assign any verbal expression . (Note the similarity between Nadia's utterance and Dónde vamos a mandar? dice Porojua, both indexing marital intimacy in the presence of by-standers . ) On another occasion, I observed the analogous use of the name Tamani in the context of "doing" others . It was when Nadia's husband, Adan, setting off for a hunting trip upriver, ordered his younger cousin to operate the engine of the boat, saying: Juan, tú vas hacer Tamani ("Juan, you are going to do Tamani") . Again, since steering a boat is not a conspicuous or unusual activity among Arabela, Tamani was rather a "wildcard" agent/body, enacted to conceal or mitigate the directive issued by Adan .
The case of "wildcard" enunciators and doers provides a strong argument for not conceptualizing these practices as mere quoting or mimicking, but as active voicing and enacting . Quoting and mimicking imply a previous event where the quoted or mimicked person actually produced words of behaviors in question, which is not (not always) the case among Arabela . But the overall picture is even more interesting, and hacer Tamani seems not to be the "zero-degree" of enacting the other . It seems that the animated figure not only need not refer to a real being, but it may even remain unidentified . Another example of announcing voracious eating provides a hint for such a scenario .
As I wrote earlier, the young sons of my hosts used to enact other persons to enthusiastically announce or invite each other to have a big meal -expressing a strong appetite and an anticipation of a great pleasure . In such cases, they referred to their uncle Educo (diminutive of Eduardo), saying that they "would do Educo" (Vamos hacer Educo!), or later -as that expression "wore out" and lost part of its salience -that they would do eduqueada (see supra) . But during my stay, they also invented an interesting alternative to that enaction: in the same situations where they used to "do Educo", they started to "do kwaiau" (thus Adan, for instance, would invite his brothers to eat with him, saying: Vamos hacer kwaiau!) . Kwaiau was an ideophone, which I only recorded in reference to voracious eating and only in this context of those announcements/invitations . I have never encountered it in myths or other narratives . Moreover, I am pretty certain that it was invented by Adan and his brothers (and it was not intelligible to other Arabela) .
At first glance, there is no third figure being animated here and bearing some part of the responsibility for the "uncivil" behaviour of the youngsters . But, through the situational parallelism, Vamos hacer kwaiau! corresponds to Vamos hacer Educo! It seems that somehow a salient ("fresh") ideophone bears sufficient "charge of alterity" to introduce another person/body into the interaction . How is that possible? What kind of person or body would it be?
Ideophones and their use have already attracted attention of anthropologists and linguists working in Amazonia . Various scholars stress their particular status within language and point to Amazonian peoples' predilection for their use in ritual and everyday speech . They are seen as pertaining to a particular place in language and discourse, where speakers depart from referential and conventional use of language and use it for a more direct engagement with the reality . In this context, some authors use the notion of transformation or becoming, e . g . Janis Nuckolls, who argued, writing about the Runa from the Ecuadorian Amazonia, that through the use of ideophones "[t]he speaking self of the speech event communicates by imitating and thereby becoming the force that creates a movement, sound, or rhythm" (Nuckolls 2010: 31) . Similar observations have been made in relation to Amazonian musicality: Bernd Brabec de Mori and Anthony Seeger observed, for instance, that "much ritual music in Lowland South America is said to be received from, or directed to […] nonhuman beings" and when the Amazonian peoples sing a bird's song "they […] refer to the bird's person-entity as the song's source . Consequently, by singing, they become the bird" (Brabec de Mori, Seeger 2013: 271-272, 274) . Andrea-Luz Gutierrez Choquevilca offered a precise, pragmatic description of that transformation of the speaker in her analysis of the ritual discourse of the Quechua Runa of the Upper Pastaza . She showed that the transformation of the voice of the ritual speaker through the use of onomatopoeias, shouts, and whistles, plays a role analogous to a "mask", but in the field of sound: it has "the power to show the invisible face of invisible actors of ritual interactions" (Gutierrez Choquevilca 2016: 20 -translated from Spanish by the author; see also Gutierrez Choquevilca 2010 . Inspired by those analyses and starting from the example of eating "with" Educo or "with" kwaiau, we may advance a general model of the Arabela engagements with the world . In its logical beginning we should place an affect -here: an affect combining appetite, pleasure, engagement, voracity, etc . of a subject towards the food . I cannot say how that affect emerges in the subject (in Adan), but the subject becomes aware of its presence and of the fact that it does not correspond to the body of "real" people (i .e . Arabela) . It may be experienced as an intrusion of an alien subjectivity (that is similar to what happens with newborns affected by animals eaten by their mothers -they start to behave in a strange way, which is diagnosed as the cutipa of that animal) . Becoming aware of that affect would be tantamount to capturing it as an exteriority . As a second step, the subject needs to identify the affect's identity . There are (at least) two ways to do it . The first is to find a body to which the affect may be attached (Educo) . The other is to capture the affect in itself though an ideophone (kwaiau) . In the latter case, the affect is not attached to a pre-existing body, but is treated as a separate body, and the ideophone is the voice of that body (captured by a human), not an arbitrary name imposed by a human on it . (It is a "sononym", just like the names of birds derived from ideophones capturing their voices [Gutierrez Choquevilca 2012: 75]) . That is why, when Adan says Vamos hacer kwaiau! he does not invite his interlocutors to perform an action referred to with a word kwaiau, but to "do" a person/body that says "kwaiau" . (As a side remark, we may note that the formal equivalence between announcements/invitations Vamos hacer kwaiau! and Vamos hacer Educo! -both are used in homologous situations -shows that the meaning of enaction of other persons [Educo] cannot be reduced to joking at the expense of other people, for why would Adan be making fun of Kwaiau?) .
In this interpretation, an affect is something that emerges in the subject, to his or her surprise . I will perhaps be able to explain it more clearly using the example of Levi putting oversized boots to "do Nuria" . I argue that it is not that Levi first saw Nuria in her oversized boots, and then wanted to do the same, but to avoid being mocked, decided to announce his action as a mere "doing" Nuria (as if thinking: "I want to put on oversized boots, as Nuria did, but I don't want to be mocked, so I will say that I am going to 'do' Nuria") . It seems that the process was different, i .e . that Levi experienced the intention of putting on the boots not as something originating from him, but as something occurring to him, coming from an exteriority, which he experienced as his body . Note that, similarly to Levi's surprise at seeing Soldado putting on trousers over his shorts, here as well, the whole experience surprised the subject -it was expressed through the laugh with which the Arabela speakers made their announcements .
So far, I have used the notion of familiarizing predation to account for the acts, where the subject enacted other persons/bodies for his or her own sake (Adanto eat rampantly, Levi -to go in his father's boots, etc . ) . The subject captures an affect that she sees emerging from her and familiarizes it as an element coming from an external body . But this scheme applies also to another pragmatic variant: where the subject reveals an affect in an unaware and unwary Other (Ua! Soldado está haciendo Murayari!) . Here, the subject captures a disturbing, alien affect that is emerging in the body of another human and identifies it as part of another (although more distant) human . In both configurations (Voy hacer Educo!, Ua! Soldado está haciendo Murayari!) the theatre of predation and familiarization is located primarily inside the speaker's perception . Just as Voy hacer Educo was not about Educo, Soldado está haciendo Murayari is neither about Soldado nor Murayari . Instead, it is about the relation between the speaker and his or her own emerging affects . The difference is that in the first situation the speaker has to capture an affect that is emerging within her own body (a strong urge to eat), and in the second, she captures an affect that is emerging from another body (long trousers "putting themselves" on shorts) . Both cases bring a restoration of harmony, as an alien affect, non-identified and hence potentially leading to dangerous transformations, becomes identified (captured) and brought back under control of the subject (tamed) .
This model has also an implicit assumption that sensible qualities of the beings of the world are perceived by the Arabela subject as external intrusions . Such understanding is not uncommon in indigenous Amazonia . The stress on perception as a process of being affected is widely present and reported or implicit in the practices described in ethnographies . Among the most classic examples we may cite the Yanomami rituals, where men, before going to war, make the images of Other beings "descend into their chests", bringing them dispositions useful for fighting enemies (Albert 1985: 156-157), or the arutam quests of the Aents peoples (Descola 1993) . Most often, though -as in those two examples -the affects that people seek to incorporate are integrated into identified, stable wholes (in the Yanomami and Aents cases, these are animals and ancestors, respectively) . The Arabela case stresses the potentially independent mode of existence of affects that only in a second move are stabilized and associated with identified persons/ bodies . The Arabela affects -'wearing oversized boots', 'putting trousers over shorts', or 'eating raw fish' -are separate beings, similar to shamanic darts, flying through the world (see for instance Chaumeil 1983) . Every Arabela can always be traversed by those and other affects . The point of everyday life is to capture them and then familiarize them through a double operation of associating them with a person/body ('trousers over shorts' → Murayari, 'oversized boots' → Nuria), and then safely using them without incurring the risk of becoming an unidentified Other . 21 It is evident that in this model the dynamics of the predative familiarization of affects starts outside the Arabela subject, who is acting in response to an external stimulus, provoked -as it were -by the affect resurfacing in the subject (boots in Levi) or in another person (trousers/shorts in Soldado) . In this, it corresponds to other domains of the Arabela relationality, where more generally, actions of the Arabela subject are predominantly understood through the scheme of provocation (see Rogalski 2016b) .
The reader will notice that in this section I have departed from my first intuition to use the model of familiarizing predation to account for what happens between persons/bodies (Levi and Nuria, Artemio and vulture) . Now, it is no more about Levi who captured Nuria (and used her to safely wear rubber boots) or Artemio who captured a vulture (and enjoyed rotten fish) . Here the act of familiarizing predation occurs between Levi and the affect that "affected" him: it is not Nuria that gets familiarized, but the affect . Everything that concerns Levi's relation to Nuria (a real person, his cousin, living a few houses away) is a corollary of a more primary interaction between Levi and his affect . It was wildcard Tamani and then Kwaiau -faceless but audible -who took this analysis away from the domain of interpersonal relations to the realm of bodies and affects . At the same time, it is difficult not to see a kind of predatory malice in Romario's Dónde vamos a mandar? -dice Porojua, targeted at his wife's former husband . It is thus possible that the scheme of familiarizing predation operates in multiple dimensions, both at a person-to-person and person-to-affect levels .

Closing remarks
In this article I attempted to show how the Arabela use of language in certain situations corresponds to more general ways of engaging with the human and nonhuman environment outside and within the community, and how a common Amazonian notion of transmission of characteristics between humans and nonhumans (present, for instance, in couvade rules or hunting magic) also constitutes a matrix for thinking about ephemeral, half-serious relations between human members of one community . As such, it is another attempt to "anchor" the symbolic economy of alterity (mostly its body-related aspects and the notion of familiarizing predation) within the discourse and in the everyday (see Oakdale 2007) . In doing of Amazonian engagement with the world -other than subjectivity and body -which is a community of 'forms' or shared patterns and movements (Mezzenzana 2015(Mezzenzana , 2018 . Although I cannot discuss at length Mezzenzana's perspicacious and rich ethnography, I argue that, at least in the Arabela case, the classic notion of Amazonian body as a bundle of affects and ways of living (Viveiros de Castro 1998) is sufficient to account for this sensibility . We only have to admit an openness of the inventory of Amazonian bodies and that phenomena conceptualized as bodies may not correspond to identifiable and stable objects . Such is the case of the body, which contains the affect kwaiau, that Adan and his brothers brought to life . It is a body that does not have external, substantial, material existence, nor even visual form -it is just a bundle of one affect endowed with a voice (and hence also a name kwaiau) . so, it highlights aspects that might not have received enough focus in syntheses centered on the symbolic . We saw, for instance, that Arabela are very attentive to the presence in the others' bodies not only of nonhumans, but also of other humans, members of their own community . This paper also emphasizes a feature of the Amazonian body that is sometimes obscured within syntheses that focus on labile and malleable bodies, understood as wholes . The affects and dispositions bundled together into the Amazonian body do not constitute a uniform totality, but always preserve their independence . They are detachable and transferable . Moreover, the affects of Tamani and Kwaiau show that the inventory of Arabela bodies is open and thus phenomena conceptualized as bodies do not necessarily need to correspond to identifiable and stable external objects . A body containing the affect kwaiau, enacted by Adan and his brothers, does not have an external, substantial, material existence, nor visual form . It is just a strand of one bundle; it is one affect endowed with a voice (and hence also a name kwaiau) .
As for the voicing of each other, it stresses a notion of voice as a constitutive part of every person's embodied identity . As Arabela stress subtle differences in prosody or individual word choices, it is evident that, for them, language is a highly particularized medium, corresponding to particular bodies . It is interesting to note that the practices of enacting others reverberate with the conception of language inherent in Amerindian perspectivism . Since, as Viveiros de Castro observed (1998), in the multi-natural, perspectival worlds, different beings perceive the world in analogous manner and use vocabularies composed of the same words (both jaguars and humans designate some elements of their environment as "beer", or, as Arabela recount, both humans and peccaries engage in a collective practice of fishing in streams, that they call jiuushiniu), the meaning of those words always indexes the speaker's embodied perspective (jaguar's "beer" is the blood of his prey, human "beer" is a fermented beverage from manioc, while jiuushiniu, which for the humans designates damming up a stream and rapidly throwing the water away to collect fish caught in the mud, for peccaries it means digging in the stream bed with their snouts in search for snails) . Therefore, in a perspectivist world interpretation of utterances always comes along the observation of the speaking subject -to what exactly she refers? What is she going to do? This perspectival attitude towards language is also present in the Arabela other-enacting: when Levi says "Voy hacer Nuria", it is not immediately clear what hacer Nuria will mean for him . People need to pay attention to his subsequent actions, relate them to their memory of Nuria and discern what aspect of the girl's ethogram he is going to adopt .
The Arabela voicing of others clearly points toward the Bakhtinian notion of dialogicality (Bakhtin 1984) . But their sensibility to and use of other voices is linked to their sensibility to movements; dialogical relationships -so present in their voicing others -are also evident in their enacting of others in actions and movements (although it is always accompanied by a linguistic clue -the label <hacer X>), pointing to a broader, multimodal dialogicality . 22 On the other hand, the notion of dialogicality seems insufficient to encompass the other-enacting through artifacts (items of clothing) . Although the utterance Voy mudar mi angelito [+ puting on Angelito's sweatshirt] might seem to be commutable to *Voy hacer Angelito [+ puting on Angelito's sweatshirt], I have never found labels <hacer X> used in context of putting on clothes .
Another aspect to be emphasized is that in the practices of doing, voicing (and "wearing") Others, there are no references to a positive Arabela ethos, one that they would promote among children or "adopted" newcomers . Sensitive to appearances, movements and voices as they are, Arabela seem not to have any particular positive representation of how a true Arabela human should move, speak and look . The Arabela way of being emerges implicitly as that which is imperceptible, which passes unnoticed . At the same time, enacting others enables the Arabela subjects to depart from the usual, transparent ways of being, while still supporting the common (Amazonian) assumption that the local group forms a uniform community of humans sharing similar bodies . On the one hand, it gives people the means to scoff those among them who depart from established manners of behaving, and on the other, it allows skilled "enactors" to circumvent these norms .
I would also like to stress the fact that the resulting human world of bodies--ethograms is strictly bound to a poetics of everyday interactions . I have hardly ever witnessed other situations, were idiosyncrasies expressed in doing, voicing, and wearing others were explicitly thematized . I have never heard my hosts talking about or explaining the particularities of Murayari's behaviour (and linking them, for instance, to his community of origin or other inherent characteristic) . It seems that even children, who use those identifications with dexterity, learn them mostly by hearing their elders uttering them . This further confirms that this particular social ontology is rooted in everyday interactions .
The poetics of everyday interactions contributes to training people in discerning minor details of each other's movements, voices, and clothes . Through enacting others, Arabela socialize the difference that inevitably appears in every human community and may be especially disturbing in Amazonia, where the community is understood as a collective of persons sharing similar bodies . If someone starts to behave in a strange way, parodying him or her may be a means to cope with this idiosyncrasy . But the practices presented here relate to bodies not only on a symbolic or semiotic level -as they point to the body, recognize bodily idiosyncrasies, elaborate on them . It is evident that they allow for an active exploitation of new habits and for their inclusion into a "working" body of a collective -an Arabela community, a kin-group or a household . In that sense, "doing" Emil, Nuria, or Murayari, are also acts of the Amazonian production of the human body, once again, taken from a major ritual level to a minor domain of everyday interactions . At the same time, given the formal resemblances between the practices of enacting others and ritual configurations -for instance the complex enunciator involved in voicing others -it seems that when we move from the ritual to the everyday, the ritual moves with us .

SUMMARY
Everyday enacting of agents through bodily simulation, voicing, and familiarization of artifacts among the Arabela (Peruvian Amazonia) The Arabela -a group of Zaparoan origin from the Peruvian Amazonia -often claim to adopt other (human and nonhuman) persons' ways of performing actions, referring to things and expressing emotions . They do it through a variety of speech acts -from announcements of their own actions, to third-person comments about other people's actions, to exclamations -and to accomplish various interactional ends (from avoidance to teasing) . This paper shows that these different forms of enacting of others actualize a society consisting of human and nonhuman persons with different bodily ethograms, where relations between bodies and affects follow a scheme of familiarizing predation . Also, a specific concept of the Arabela agent emerges from this analysis, where the Other is individualized as a static ethogram of gestures and voices, while the speaking or acting subject has to prove his/her ability to singularize Others, using their presumably typical verbal expressions and actions . The ultimate goal of this paper is to stimulate reflection on the links between everyday interactions and ontologies in Amazonia .