The Stories of Others

As many other Amazonian indigenous peoples, the Aweti – a Tupi-speaking people who live in the upper reaches of the Xingu River – habitually name a specific kind of narrative, which we usually call myths, by an expression that could be translated as “stories of the ancient people”. Most of the time, however, they simply call them “stories” or, more precisely, tomowkap, which literally means “something that orients”, tales about an event that may have happened at anytime in the recent past or even in the present. This article is an attempt to explore the epistemological and ontological implications of this indiscernibility between myths and other kinds of narrative. The assumption here is that this may tell us something about the way the Aweti think, not only in what concerns the nature of what we call myth, but also about the nature of knowledge one can have about the world and, furthermore, about the nature of the world itself. keywords Myth, Aweti, Upper Xingu, Knowledge Theory, Ontology DOI http://dx.doi.org/10.11606/ 2179-0892.ra.2018.152164

In everything I wrote about mythology, I wanted to show that we never reach an ultimate meaning. For that matter, do we ever reach that in life?
Claude Lévi-Strauss, De près et de loin Myth lives for the immolation of its movement in a flash of realization; myths are the moths of a peculiar human tropism. They demand (and create) an ambience in which speech is lethal, ultimate predication is withheld, and conviction is blunted by a suave demurral.

Roy Wagner, Lethal Speech
If physical laws vary in the past as well as in our future, then the evolution of the universe is unpredictable, causing formal discomfort and more, shaking the core of the scientific edifice that presumed itself well structured and untouchable, producing uncertainties, questioning its foundations and provoking a crisis in the scientific construction of reality. Mario Novello, O universo inacabado Although I propose in this text a reflection on the myths told by the Aweti, a Tupi speaking people who live in the region of the Xingu river sources 1 , I will not dwell on the analysis of any particular myth. What interests me is not knowing exactly what is said in one myth or another, but rather to try to understand what the myths can tell the Aweti about the world they inhabit. For although listening to myths constituted a fundamental stage in my learning of the Aweti world, what led me to the considerations that follow were the countless comments I heard from my Aweti interlocutors about the mythical variations that I registered throughout our time together. 2 When I began my research, I quickly understood that the Aweti knew what I was doing there (although I did not have the same clarity): my job was to record stories. Their impression was probably largely due to the experience they had had with researchers before. Many anthropologists had passed through the village, but only the linguist Sebastian Drude 3 had stayed with them for an extended amount of time 4 , and most of them seemed to have focused on recording myths and songs. I could not wish for anything else, it seemed to them, so they promptly guided me to the narrators best equipped to help me. I accepted the path they presented to me and with certain formality I began to establish with each of the two nominated narrators the conditions for our relationship: when I said that I was interested in stories, they commented that many people had been there and had never given anything in return, and we finally agreed that I should pay for what I would learn. I did not realize at the time that I was making incompatible arrangements, for I should have chosen which stories I wanted to The area that is recognized in ethnological literature as "Upper Xingu", comprises a community of peoples speaking Aruak, Carib, Tupi and Trumai, an isolated language, united by ritual and matrimonial exchanges based on a common cosmology in general lines. The region is the subject of a relatively extensive bibliography. For some of the most significant or recent descriptions of the peoples of this region and the "upper Xingu multiethnic system," see Bastos, 2013 (Kamayurá); Basso, 1973, andGuerreiro, 2012 (Kalapalo); Gregor, 1977, andStang, 2009 (Mehinaku) ;Franchetto, 1986, 1993, 1998, and Heckenberger 2000, 2005; and Barcelos Neto, 2008 (Wauja). I have been working with the Aweti since 2004, but it was between 2006 and 2010 that I developed my most intensive fieldwork, adding up to about 13 months in the village at that time.

2
I re-present and develop in this article questions elaborated in the context of an ethnography on kinship and witchcraft according to the Aweti (Vanzolini, 2015: 306-317), to which I refer the reader for a more detailed presentation on the context of mythical narration and its relations with kinship. For considerations on the relation between the truth regime analysed here and the Xingu's pacifism see Vanzolini (2016).

4
George Zarur's thesis (1975), based on a period of only three months of fieldwork, was the only ethnographic study published on the Aweti until the completion of my doctorate studies. Following the hear from the start, that is, whose stories.
My situation got complicated when the narrators realized that I was simultaneously learning both their narratives, and that each of them had begun the narrative cycle of origins at different starting points. They continually criticized each other's stories, complaining mostly about incomplete versions and incorrect ordering of facts. Soon, criticisms started coming from the most diverse people who came in contact with my recordings: "This is not the correct noise of a man walking, he [the narrator] is telling you in kayayura! [Tupi-Guarani language Xinguana] "; "That's not what so-and-so said to so-and-so, he actually said..."; "Notice how he always repeats that phrase, it's very funny!"; "He makes up the stories, he keeps changing everything!", they often told me 5 .
In contrast with the tension that marked the narrating of myths to me, these stories are usually told among the Aweti in an almost comical and unconcerned tone. Not only the most acknowledged narrators, but also men and women who would not publicly admit that they know the myths, tell them daily to the children of the house, usually before bed. The narrative told each night is chosen according to the moment and according to everyday events. A man bitten by a snake in a neighbouring village prompts the memory of a story about the origin of the snakes. Children also have their favourite characters, such as the buffoon monster Awazá, or Kwalamiri, another bumbling type. Often, the nocturnal narrations stop in the middle: all the listeners have slept, or the narrator is feeling drowsy.
Stories circulate according to daily interactions and, along with other knowledge, words, goods and food, they condition the arrangement of groups of relatives, whose contours are scarcely defined. Stories of parents and grandparents are learned, but many men also spoke of stories learned from their in-laws 6 , revealing that the group of people who recognize a story in a certain way is quite circumstantial. In certain contexts, however, these people can effectively present themselves as a whole -when the stories they share are confronted with the versions of others.
Certainly people are aware that the stories known to them are told in other ways by some of their village neighbours and by people from other villages. Many times these variations are attributed to differences of linguistic origin: soand-so tells the story because he learned it from his grandfather, who was Kamayura, or his Mehinaku mother (the case of onomatopoeia that describes the walking sound of a character narrated above, illustrates this point). Eventually, a cumulative feeling may lead the hearing of stories of others, as if these represent some additional knowledge: "that's how they tell this story…!". It was always the desire to learn, not the intention to undo other versions, that my interlocutors demonstrated when asking to listen to my recordings of myths during sessions that nevertheless ended in harsh criticism of the narrator. Most likely, in this context, the dispute was not over the correct version, but rather over the position monographic style of the time, and certainly limited by the short duration of the fieldwork, Zarur's study presents an overview about the Aweti life and an important record of the group's situation in the 1970s, but it is lacking in several respects.

5
These and other reproductions of the Aweti speech presented below are free translations of dialogues that have taken place in the indigenous language. This narrative option, which follows the style adopted in my doctoral thesis, responds largely to the need to safeguard the privacy of my interlocutors in dealing with issues which are sensitive to those who share village life.

6
Given the preference for matrilocal residence during the first years after marriage, until the couple has at least one child. of the connoisseur before people like myself and other researchers who passed by. At any rate, it seemed striking to me how differences between versions could be summoned to underline, at times, complementary distinctions -as occurs when songs of distinct peoples are gathered in a Xingu ritual 7 -and at other times oppositions and incompatibilities.
Stories are usually learned at home, but they are retold outside the home with no help from the tape recorders anthropologists carry with them. If the case of an acquaintance bitten by a snake motivates the night time narration, told to grandchildren, about snakes, this same story may be told to a neighbour who happens to be visiting that afternoon, or to the shamans who gather in the center of the village at nightfall. At times like these, narrators present their stories as if they were the stories that relate to a certain fact, an attitude comparable to that of who presents oneself as spokesman of the Aweti mythology to a foreign visitor like myself. In this seemingly banal activity, which is performed with great enthusiasm, the narrators expose themselves to the same ironic reception dispensed to the versions I recorded. In public, the irony is momentarily repressed, but it emerges when commented at home: "So-and-so was telling this story, I've never seen this story told like that! He does not know this story, my father knew how to tell it well!".
Myths are a source of fundamental esoteric knowledge for the Aweti, which also explains, I believe, my interlocutors' insistence that, in order to understand their world, I needed to apprehend their stories, understood as a discourse passed down through generations that must be maintained unchanged in its narrative form. But while it presents itself as an established and codified knowledge -fixed in the special form that the narrative must assume -the myth is always the subject of variation and uncertainty. If the dissensions I have witnessed could be interpreted in terms of power disputes and other village disagreements, here I am interested in exploring the conditions of possibility and the metaphysical implications of mythic variation. My bet is that this may tell us something about what the Aweti think, not only with regard to the nature of what we call myth, but also concerning the nature of the knowledge one can have of the world in general and, additionally, about the nature of one's world.

what is a myth?
Much of the ethnology concerning the peoples of the South American lowlands registers the existence of a certain kind of narrative -sometimes called "stories of the ancient people" -that would distinguish what we call "myths" from other discursive modalities, such as chief speeches, chants, gossip, and stories about the recent past. Intersections and combinations are not, of course, excluded:

7
Most Upper Xingu rituals can be held locally, bringing together only people from the same village, or regionally, with the participation of one or more invited neighbouring groups, depending on the type of ritual. In the latter case, when the event involves ritual chants, usually singers of the guest groups join the local experts, presenting particular versions of the songs. The general effect of visitor participation is the affirmation of a complementary difference between the groups present at the gathering, even though their relationship can vary between competition and collaboration. A close comparison between this regime of collaboration between ritual specialists and the question of mythical variants would call for further work. mythical narratives can be presented in the form of chants, composing the shamanic repertoire, as is the case of the Pano peoples, described by Déléage (2010) and Cesarino, (2011), or may inform ceremonial discourses that substantiate the legitimacy of chiefs, as is the case in the Upper Xingu (Franchetto, 1993;Coelho de Souza, 2001). It is not surprising, in any case, considering that the same must be true for other regions of the world, that anthropology has given a particular significance to this type of discourse, although varying in the interpretation of its value, whether for the social life of the peoples or for analysis 8 . At first glance, it would seem that we all know what we are talking about.
No less remarkable, however, is the fact that a significant number of Amazonian peoples (and probably others) do not distinguish, or do not emphasize the distinction, between what we call myths and other types of discourse. Acknowledging this fact, some authors have argued for the pertinence of an analytic distinction between "myth" and "story" as different modes of collective consciousness in relation to time (see contributions to Hill's collection, 1988, which focus on a discussion concerning the Lévi-Straussian notion of cold societies, and also similarly Gallois, 1994). Others preferred to focus on the indiscernibility of this discursive genre, summoning a same theoretical tool to analyze narratives of varied character, some of which probably did not fit into a predefined category of myth -as is the case of Ellen Basso's (1995) interpretation of Kalapalo "Oral History" narratives as biographies, or, for a diverse ethnographic context, Herzfeld's (1985) analysis of the production of Greek folklorists in the nineteenth century 9 . While some sought to defend, others propose imploding (or expanding, which is the same), the myth as an analytical object. Closer to this latter alternative, the reflections that follow intend to elaborate on the consequences of the fact that the Aweti do not distinguish by a specific concept, as will be seen, what we call myth. This seems to be directly linked to the apparent paradox that these stories, being recognized by the Aweti as a fundamental source of knowledge passed down through the generations, are in constant transformation.
Rejecting the founding opposition of the very notion of myth, Lévi-Strauss recognizes it as an expression of a perfectly rational thought, albeit distinct from that of Western science itself: intellectual elaboration characterized by the formulation of abstract problems through empirical categories such as raw and cooked, high and low etc. A central point of novelty represented by this approach concerns the treatment of variants. Instead of looking for original versions, says Lévi-Strauss, one must compare the versions so that they confer intelligibility to one another. The method of analysis consists in making the myths "think among themselves" (2004b: 31), revealing something that is beyond the consciousness of a particular narrator or listener.
In interpreting myths in light of other myths, Lévi-Strauss not only develops a 8 Functionalist theories, as we know, take the myth as part of the self-replicating mechanism of a static sociology, that is, interpret its effects at the level of empirical reality (Durkheim, 1996, Malinowski, 1988. Lévi-Strauss transfers the question to a symbolic level in two senses: he understands myth as an exemplary representative of "savage mind" and, therefore, a privileged object for anthropological analysis; on the other hand, perceives the myth as an instrument of modulation of the temporal consciousness of the people who tell them. It is important to note the difference: for Lévi-Strauss, myth does not produce social stability, as functionalist theory does, but a sense of stability. I will return to this point.

9
In the light of a "discoursecentered" theory, Basso (1995) draws from a series of Kalapalo narratives that which we could understand as indigenous philosophy of the action and the person. Herzfeld (1985), using very different theoretical references, explores Lévi-Strauss' ideas on the continuity between myth and mythical analysis to interpret the work of the Greek folklorists as the production of a myth of the nation-state. By imploding the category of myth, these authors perhaps proceed in opposite ways: Basso prefers to treat myth as history (more precisely, he makes myth and history equivalent as "discourses"), while Herzfeld advocates for treating history as myth, that is, as an object subject to structural analysis. article | Marina Vanzolini | The Stories of Others Rev. antropol. (São Paulo, Online) | v. 61 n. 3: 162-184 | USP, 2018 11 Even though in The Savage Mind the question arises not only in terms of coherence, but also of the effectiveness of the logic of sensible qualities according to which myth operates (see the author's comment to that point in interview to Eribon, 2005: 159).

12
The variants in this case on are internal to the Xinguian multilingual set, but it would be interesting to investigate the Aweti appreciation of myths from other areas, as Stolze Lima does for the Yudjá. On the other hand, while Lévi-Strauss compares, above all, variants of distinct peoples, a reflection on the context of reception of the local variants can help elucidate the conditions of possibility of transformation of myths / mythical versions in varied ranges of distance.
13 This is in alignment with the works of Oscar Calavia (2002) and Daniel Pierri (2013) on mythological variation as cosmological reflection, between the Yaminawa and Guarani Mbya, respectively. And also, although starting from very different theoretical premises, in alignment with Pierre Déléage's (2010) ethnography on the constitution of the Sharanahua cosmological imagination based on ordinary experience. Nicòdème de Renesse's thesis (2017), recently defended in University of São Paulo's Social Anthropology Graduate Program, brilliantly elaborates this question in an ethnography of the Paiter Suruí of Rondônia. The author begins with the following question: what are the conditions according to which Paiter stories can vary and, at the same time, still be perceived as true by their narrators and listeners? method for making sense of the narratives, understanding them as transformations of one another, but he also reveals logical and existential problems of interest to the Amerindian peoples. The author deliberately leaves aside, however, the question of how the indigenous peoples think their myths, that is, what value they attribute to the discourses that we thus recognize 10 . The myth is sometimes approximated by Lévi-Strauss to Western science and philosophy as means of ordering the chaos of experience in intelligible terms (2004a: 32). The author's emphasis on the intellectual character of this thought can be seen, for example, by the idea that entities and procedures that populate mythical narratives do so in so far as they are "good to think with" (a recurring formulation throughout Mythologiques), allowing one to articulate abstract ideas through sensible qualities (2004b: 19) 11 . Given the way they were continually summoned by my Aweti interlocutors, however, I would say that their stories that I recognized as myths are not only good for thinking; they are indispensable for living (although they are also certainly linked to "a constant and always alert curiosity, a desire to know for the pleasure of knowing" (Lévi-Strauss, 2004a: 30). I believe that an ethnographic approach to how such knowledge is thought by the peoples who produce it is perfectly reconcilable with the recognition of the transcultural character (in the absence of a better term) of such narratives which, as Lévi-Strauss has shown, are objects in continuous circulation, crossing geographical and linguistic boundaries.
Stolze Lima (1999: 4) questions, in this sense, what would be the criteria of verisimilitude that would allow the Yudjá (Tupi people who inhabit the banks of the Xingu River) to recognize myths of other peoples as myths, that is, coherent discourses endowed with truth. The same author comments on the Yudjá reaction to the existence of different versions of the same myths: were the variants referring to distinct episodes, which can be sequentially linked to the same narrative? Or simply approximate versions, none completely faithful, of the same episode that occurred in the distant past? Her work demonstrates that mythical variations are not objects of interest only for structural analysis: as occurred when the Aweti questioned the recorded versions that I presented to them, the question seems to be a matter of speculation for other indigenous peoples 12 . But the myth must be understood in light of the native conceptions of knowledge in order to evaluate the implications of such judgments. I stick to the Aweti world 13 .

beginnings of this world
The narratives recounted by the Aweti are far from constituting a corpus of knowledge that is cohesive and orderly, or even amenable to ordering. The stories we would tend to classify as cosmogonic myths -but which are not distinguished in their language by any specific expression -illustrate this point well.
If we asked an Aweti about "our origin" (kajypy 14 ), we would probably hear one or several episodes pertaining to the cycle of narratives involving the demiurge twins, Sun and Moon. These stories do not really narrate the emergence of the universe, as apparently it has always existed, but the origin of its current settings. The recurring operation in these narratives refers to what Lévi-Strauss, in his extensive analysis of the indigenous mythology of the American continent, phrases as the "passage from the continuous to the discrete". Xingu mythology, in fact, thematizes this passage in very varied levels. The narrative cycle involving Sun and Moon 15 explains, for example, the transformation of a world of darkness -continuous night time -into a world with the presence of sunlight, marked by the alternation between day and night, after the making of a beautiful headdress with red and yellow feathers for the Sun. Mortality, the establishment of the generational succession, is the result of the inability of the twins to recover the life of their mother, killed by the jaguar-mother-in-law, an episode that also marks the origin of the quarup, the Upper Xingu funereal ritual. The same mythological saga explains the origin of the present humans and their division in to peoples with different characteristics and productive specialties, when different weapons and artifacts were distributed to the people created by Sun, from enchanted arrows. A series of episodes also accounts for the appropriation by humans of flatulence, sleep, male erection and jealousy, object-attributes previously belonging to certain animal people, thus establishing a series of biological and sociological alternations: the process of digestion; the variation between work time and rest time; sex and its corollary, generational alternation. And jealousy, which seems to have nothing to do with this, can not be there by chance: perhaps its message is the establishment of a principle of disorder, being it the main drive of Xingu witchcraft (Vanzolini 2015:252-260), at the same moment when order is established.
All the "originating" episodes listed above make up a narrative sequence that can and must, according to the Aweti experts, be told in a precise order (which, however, is a matter of divergence). But some myths that reveal facts of analogous effects have no explicit connection with the cycle of demiurge twins. To deal with the differentiation between peoples, for example, my interlocutors could also evoke the history of the great flood that forced all the indigenous people to take refuge in canoes, which were then tied to the top of the tallest trees, which were not underwater. When the liana that tied some of them broke, the people that were there were dragged away, giving origin to groups that live in other regions of the Xingu. In the very cycle of narratives involving the Sun and Moon twins, as told by the Aweti, or by some Aweti, an episode prior to the differentiation of the arrow-people accounts for the differences among the peoples of the Upper Xingu: during the quarup of their mother, the twins sang the funeral lament in all the Xingu languages, giving origin to the idiomatic differences between the peoples (and to a powerful spell: he who sings the funeral song in the vicinity of a village will cause, in anticipation, the death of all its population). The theme of mortality also emerges in the myth -highly widespread in the South American lowlands -that tells how humans refused to dive into a large pot of boiling water. If they had accepted the offer of the demiurge, today we would change skin like snakes, periodically becoming young again.
Incompatible with the intention of a continuous narrative, all these stories seem only to evidence the repetition of a same scheme of transformation. An episode of the cycle of narratives about Sun and Moon, described outside this context and significantly transformed, provides an example. An Aweti friend once commented to me that before the indigenous people were like the white peoples, and whites like the indigenous people. Sun made guns and arrows, offering them to both groups. The indigenous people (who were like the whites of today) found the firearms to be too very heavy and preferred the arrows, while the whites (former indigenous) were able to carry the rifles, and held on to them, becoming what they are today: the owners of firearms. In addition to the confusion in meaning of white and indigenous categories, what this variant of the mythical episode shows is that, even in the most complete version that explains the productive specialization of Xingu groups and their neighbours, the relevant and present fact in all versions is the differentiation, or some differentiation, between the peoples. And if the variations sometimes led to comments on the falsity of a given version, it does not seem to me that the problem perceived by the Aweti listeners is the lack of congruence between the episodes, as should be clear below.
Although they have no expectation of being able to establish, through mythology, a single version of the origins of the present order, the Aweti usually summon myths to explain the configuration and state of things in their world. The term by which these stories are designated in their language seems to me to be quite significant in this sense.

orient yourself
It may seem self-evident to say that for the Aweti the myths portray important, but perhaps mysterious aspects of the world we live in. For me, however, it was surprising to realize that when asked about the geography of the celestial world where the village of the dead is located or about the behaviour of the dead, for example, my Aweti interlocutors always resorted to the story that is well-known to all, of the man who accompanied his dead friend to heaven and returned to tell what it was like. "I've never been there, I do not know what it is like," a sha-man once told me when I asked him about it, and soon began to tell the myth (which I had already heard from other sources, in somewhat different versions) .
I believe that when the Aweti resort to a myth to explain what the life of the dead or the village of the birds are like, why we should not urinate on the harvest or why we should treat our mothers-in-law well, they are also making assertions about the meaning of these stories in their lives. On analyzing the contexts of mythical narration among the Piro, the Aruak people of the Peruvian Amazon, Peter Gow (2001) states that the myths among them are never told for a specific reason, to explain a situation or to advise the younger ones, but for the very pleasure of narrating. A similar remark might perhaps have been made for the Aweti, but the fact that myths are not explicitly told to instruct or explain does not mean that they have no explanatory potential, or that they do not serve as guidance.
The Aweti term which I translate as myth, indicates this function: the word tomowkap, which the Aweti usually translate as "story" is formed by the root mowka, "orient" or "guide", plus the substantializing prefix to, and the suffixp, which instrumentalizes the object. A tomowkap is, literally, as I mentioned above, "something that orients". The Aweti use the term tomowkap to designate both stories lived by unknown ancestors and passed down through the generations, as well as the narrative of something that happened ten minutes ago. A man who has information about a political meeting that happened the previous week, as well as a striking narrator of myths, can both be called the "story owner", tomowkap itat -temporary position in one case, relatively fixed in the other (see below).
What I designate here as "myth" is the discursive genre indicated by the expression that could be translated literally as "story of the ancients", mote mo'aza etomowkap 16 . In everyday speech, however, the Aweti do not make a point of distinguishing myths from other stories by this formula. In general, past and recent reports are just tomowkap: "I brought a tomowkap Marina 17 : it seems that you are having problems with Funai, we need to produce an authorization letter or they will take you away from the village", I was told once. But such observations are also common: "If you had children, Marina, when hearing this bird sing you should warn them not to want food that they cannot eat, because a lightning could strike them. This is what it's like, tomowkap." The notion of -mowka, "to orient", can refer to a chief's daily discourse addressed to the villagers pronounced at the center of the village, to a father advising his son, or to one person telling the other what happened during fishing. This instruction character that the Aweti associate with their myths is consistent with the moralistic effect some stories may have when they remind us that terrifying cosmological disturbances have occurred, in immemorable times, due to bad behaviour among relatives, for example. The fact that myths explain the 16 Mote mo'aza etmowkap: mote, refers to the remote past; mo'aza, "human", "anthropoform", "moral", "Xinguan", depending on the context of that which is said ; e-, possessive pronoun secondperson singular + tomowkap (see text).
17 "Tomowkap azezupeju Marina…". Azezupeju: a, prefix, first-person singular + -zezup, "have", "possess" + eju, gerund. current conditions of the world and thus guide rules of conduct can also be seen in a man's comment, once, about explanations that I received from his father about dietary restrictions associated with the pequi and corn harvest. Perhaps noting in me some expression of disbelief, the son made a point of noticing: temo'em e'ym, tomowkap! 18 , "This is not a lie, it's a story!".
The term tomowkap suggests that stories are not only accounts of what has passed, but also a guide to future actions. One of the Aweti storytellers who took care of my education asked me one day, pointing out very varied objects around us, which myth I would like to hear: "it's people [the door stop, for example], it has a story"; "It's people [a fly], it has story"mo'at, tomowkap oupeju 19 ..., he said. An entity that "has a story" is someone who was a character in a story in which it invariably appears living in human form. The story explains some transformation -how the fly has assumed its fly aspect, for example -as a result of a specific behaviour of the protagonist or of those with whom he relates. To say that something has a story, therefore, implies the recognition of a personhood (such entity is / has already been mo'at, a person), and therefore of the agentive power of its protagonist. Implicit in such accounts is the fact that if many things have already been people, they still are, in some dimension, and in that condition they can continue to act and produce effects on us 20 . By providing information on aspects of the world that are inaccessible to the average person, such accounts function as a basis for healing diagnoses (indicating that beings are potential human aggressors, that history can be repeated), can be summoned as justification for specific forms in ritual actions, and inform the exhortations of parents concerned about the behaviour of a child, for example.
Tomowkap, in short, is always an account and a warning, an information that will be considered as a basis for future action. As it allows one to get to know other agencies through the events in which such agencies were witnessed, and from which new actions can be predicted, every story is a source of knowledge: the chief went to the federal government and said such thing, which causes such effects; a hunter heard strange cries in the bush, one must be careful; they bewitched the daughter of a man from that other village. Therefore, being a "story owner" is a position of power, even if temporary, as in the case of the visitor who brings news from the city or neighbouring village. The association between leadership and mythological knowledge that Basso (1995) notes for the Kalapalo also occurs among the Aweti: the two men most recognized as story-tellers (tomowkap itat) in the village were also (more) recognized as chiefs, both parents of young men who also acted as chiefs 21 .
If a story from the times of the grandparents, a recent story 22 from the Aweti point of view, does not determine ways of being as does a story from the times of the ancients, it still holds a message, a warning: this may happen again. An ex-18 Te-, possessive pronoun, second-person singular; mo'em, "lie", "falsehood" (see text); e'ym, negative particle.
19 Oupeju: oup-, "to be" + eju, gerund. ample are the countless narratives of men attacked by jaguars on the pathways around the village, which refer not to the immemorial time of the demiurges, but to a near past, although impossible to determine. Tomowkapwan ekozoko 23 , "you will become history" is the typical phrase from one character to another in the closing of the narratives of the ancients: pronounced such sentence, the protagonist of a mythical event becomes a model for "the people of the future", amyñeza. This statement seems to remind listeners of myths that every agent of an extraordinary event can "become history" or, as we say in Portuguese, "go down in history", become the source and prototype of a new world order 24 . from novel to myth I do not want to produce the false impression that the Aweti do not recognize in this that we call myth any specificity in relation to other discourses. It was precisely this specificity that first aroused my interest in the myth as a system of knowledge as I saw certain versions being harshly criticized by their listeners, as mentioned above. Making it clear that the myth is not an account like any other, the reactions indicated that its veracity depends on the formal acuity of the narration. The Aweti would probably agree with Lévi-Strauss when he identifies in myths a precise structure in which the order of factors is highly significant, as opposed to common stories where, like novels, the value of form is inferior to that of content.
From what I have learned from the Aweti, mastering the mythological corpus implies mastering a narrative style, which in turn implies mastering as much detail as possible. Characters must be described through their body ornamentation; their names should be remembered; onomatopoeic terms describing movements such as walking, running, reaching someplace, bathing happily, eating, drinking, puffing, or entering a house must be reproduced; the tone of voice of each character, imitated; the warlike or shamanic songs sung by the characters, sung; the order of events precisely respected; myths that are part of a same saga must be told in a specific sequence.
Mythical narration also requires respect for certain language structures and the adoption of specific rhythm for narration 25 . The obligatory presence of indirect evidentiality in mythic narratives -with the use of the particle ti, "says" -signifies a fundamental character of any story: narration always refers to the experience of a subject, who may or may not be the "story owner" when it concerns the account of a recent fact, but it is necessarily undefined in the case of mythical narratives. It is significant, in fact, that through the use of the particle that underlies the externality of the experience which is being told, the narrator of myths attests his distance from the narrated event: "it is said that" this is how 23 Tomowkapwan: tomowkap + wan, "in the form of". The same wan prefix integrates, for example, the formula kwat'ypwan weijmo'apytewe, "made him disappear in the form of kwat'yp", in reference to the funerary effigies (kwat'yp) that occupy the center of a ritual (known as quarup) that has the purpose, as the Aweti explained to me, to take the souls of the dead permanently up to the sky. Ekozoko: e, personal pronoun, second-person singular + koz, "stay" + oko, indicates the continuity of an action.
24 I do not perceive that the Aweti suppose a genealogical continuity between mythical time and historical time, nor that they believe that a repetition of the actions of characters of the cosmogonic past in the present amounts to a kind of instantiation of cultural prototypes, as Sahlins interprets on the ethnography of Prytz Johansen, for the Maori and Hawaiian cases (Sahlins, 2008: 14-15). What is at stake in Aweti's apprehension of the myth, it seems to me, is less the problem of repetition than that of the eternal indefinition of the cosmos.
25 On the formal characteristics of some highly encoded discourse categories in the Upper Xingu, such as the "stories of the ancients" and the chief's speeches, see Basso (1995), Franchetto (1986and 1993), Ball (2007, and Guerreiro (2012). it happened, so he repeats after every sentence of his narrative. The idea of never being able to be sure seems to be implied there.
When the listeners characterized a story as "false" or "lying" they referred to the incomplete or inappropriate execution of its formal elements, although sometimes I have witnessed people questioning the existence of a whole sequence of actions in a mythical narrative, of a particular event, or even of an entire myth. When I heard of the two old men who instructed me the cycle of stories connected with the demiurge Wamutsini, for example, one of them maintained that the story should begin with the adventures of Tati'a, the Bat, which he claimed to be the father of the demiurge, while another narrator guaranteed me he had never heard of such a character.
But the boundaries between discursive regimes are not so rigid. It was curious to note throughout the fieldwork that the main narrators of Aweti myths were so impregnated with the narrative style required by the stories of the ancients that they used it in their more everyday accounts. The particle, ti, "it is said", obligatory in the stories of the ancients as well as the particular tone of mythical narration, could appear as mannerisms in worldly conversations and give rise to mockery among malicious audiences. Despite the humour generated by such occurrences, perceived as slips made by the speaker, these contributed to create an impression of permeability between narrative forms. Giving a "mythical" quality to diverse accounts, they seemed to very subtly suggest that the transformative power that characterizes the stories of the ancients may also permeate "recent" and current stories.
Evidently, not everything that happens to someone, and that can be told as a story, implies the establishment of a new world order. But the difference between cosmogonic stories and any other stories seems to me to relate to a variation of the degree of potency, a distinction which is obviously not insignificant, but nor is it absolute. The high level of formal codification of the stories of the ancients would hence correspond to the acknowledged importance of the events that are narrated, largely derived from the extraordinary power of the ancient beings, which makes these stories worth telling generation after generation.

all gossip is a story, for someone
The Aweti term which, in accordance with native Portuguese, I translate as "lie" mo'em, can designate either a deliberately misleading discourse, or an unintentional deception or a poorly executed representation of anything, such as a pattern of body painting reproduced without excellence or a myth half told. In contrast, there is no term that could be translated as "truth" or "true". Na ytoto, "him proper", "the true one", is what is said to indicate the truth in something.
Keeping in mind that ytoto also means "a lot", we see that "proper / true" is only a fuller version of the false, and that its distinction is a case of gradation 26 . Truth and lie, therefore (and it is clear that the terms are not the most appropriate to describe the thought in question), are only versions of one another. The prototypical -or mythical -is closer to the farce than we usually admit.
As I said, the Aweti idea of "lie" or "falsehood" does not emphasize the moral aspect of deception (although it may be used in contexts of deliberate deception), since a design or discourse may be "false" as a result of the inability of its creator, and not of an intention of hiding the truth. The "lie", in this case, concerns the mismatch between a model and its reproduction. But for both myths and body designs, for example, the expected correspondence is that between a drawing made in the past and the drawing made today, or between the narrative told by the ancestors and the narrative heard today, and not between facts in real history and the narrative that recalls them, or the path actually made by ants and the exact achievement of the body painting pattern that bears that name. To say that a myth telling is "lying" means to say that it does not perfectly reproduce the form by which it was told in the past, and should continue to be told today. A "false" myth is a myth that does not present the form, including the richness in detail, that a listener expects to find.
Myths are stories of a distant and unknown past, but they are stories whose strength lies in the fact that they originated, for a specific narrator and listener, from some identifiable point in the recent past: the stories told by daddy, uncle, grandpa. Thus, the question of representation in myth seems to me to be linked, for the Aweti, less to the relation of representation between narrative and world than to the political representation expressed in the narration: not the truth of something, but the truth of someone 27 .
Gossip, on the other hand, is not a discursive object with a definite form: the term that the native Portuguese translates as "gossip" (tui popy'i, literally "inclined chin") means the propensity to speak too much. To speak little, to take care of what is said, is a sign of nobility in the Upper Xingu, one of the main characteristics expected of a chief. Speaking and listening little, actually: not listening to what is said about him and his family. "Gossip" is therefore a category of accusation -no one says of oneself "I speak too much" -used to disqualify the words of others. But if all excessive talk is open to criticism 28 , only that which concerns sensitive issues regarding the lives of others tends to be described as gossip.
Perhaps the biggest problem is to whom it is spoken. I understand gossip as an effect of verbal incontinence not only in the sense that it would be better to say nothing, but also because gossip is always a domestic conversation that has leaked. It will only qualify as gossip to those who do not share the point of view it conveys, while others will just say it is a story, tomowkap. It is said that some-27 Point already referred to by Ellen Basso (1987,1995) on the contexts of Kalapalo narration. But the author, driven by another type of inquiry, does not explore the ontological implications, so to speak, of this fact. I am referring to how this way of dealing with knowledge projects a particular image of the (un)known world.
28 When asked, during the seminar Lições de Fala, (Universidade de São Paulo, July, 2018), about how his people teach children to speak, Mutuá Mehinaku (also Upper Xinguan) said one should not encourage children in this sense, as that would implicate in the risking them becoming blabbermouths. The seminar brought together 17 indigenous speakers to discuss the topic of speech. The importance of listening to the elders as a precondition for learning to 'speak well' , that is, to speak with confidence, sense of occasion and restraint, was repeated by almost all participants in the event. Another common point observed in this context was the perception of a verbal incontinence particular to the whites, possible only thanks to our lack of 'embarrassment' , a term whose meaning in the indigenous worlds far transcends the value we attribute to it in modern societies.
26 Similar logic was described by Viveiros de Castro (1977) concerning the Yawalapití, and also accounted on the Kuikuro notion of auréne, "lie", by Franchetto (1986). thing is tui popy'i as an alert: "do not pay attention to this". Tomowkap, by contrast, is something true that can and should be apprehended as the basis for action. The way in which stories of everyday life and gossip are told or the subject they deal with does not distinguish them. There is talk of strange events that happen during fishing, of a couple's quarrel, of the naming of a baby or of the illness of a relative. What distinguishes these comments is not their internal structure, but the listener's point of view: gossip is others' stories -more than their lives, their point of view, something impossible to define from a neutral standpoint. In light of what I have been discussing here -the relative indiscernibility of what we call myth in relation to other stories -we see that the distinction between myth and gossip also becomes much less clear. Although certain stories of the ancients are highly codified, since they are meant to reproduce an ancestral account passed down through generations, both what we recognize as myth and what some Aweti friends described as gossip may equally be called stories. Its assessment depends solely on the listener's point of view.
Stories are part of a flow of things which are mutually shared between relatives: the act of telling is co-extensive with other acts of giving things and knowledge that mark and constitute these relationships. This explains why in the relationship with the whites, too distant to integrate into kinship-sharing networks, stories have become subject of payment, just as it was with me at the start of my fieldwork. In contrast, neighbours who listen to a story which is told at the center of the village are treated as relatives, although the discomfort evident at their reception denounces how different (izetu 29 ) they are. Otherness apprehended as a flaw: "he doesn't know how to tell", one can always complain.
When they admit the possibility that different versions of a myth are understood as stories of other Xingu peoples, the Aweti reveal, I believe, that ensuring the perfect coincidence between a myth and the world one describes may not be a problem for them. But it can also be a problem, after all an ill-told myth is "lie." The issue only becomes really relevant when it involves important relations -when choosing one or another version of the story means to choose a side of a contention, a version of the narrative. The possibility of fixing an "official" version at a given time and place is out of the question: each new enunciation will set a context for dispute.
In commenting on the use of the particle of indirect evidentiality in mythical narratives I have noticed that if myths speak of the world, they never present it as definitive, but as that which can be known through a particular experience that lives through time through particular narrators, the story owners: we know that the village of the dead is as so because (it is said) so-and-so went there and came back to tell us ... but who knows, it might be different. To have the myth as the main source of certain knowledge is to admit, from the start, the inescapable 29 It is said, for example, of two spouses that they are izetu, different, which justifies that they do not follow the prescription of fasting when the other is sick, contrary to the real relatives generally designated by the term to'o.
incompleteness of knowledge -which puts the myth alongside the knowledge produced through shamanic journeys, a modality not very common in the present Upper Xingu, as far as my experience, but central in numerous Amazonian cultures. Myth privileges "flashes of knowledge" or "insights by the firepit," says Roy Wagner (1978), if we can use his understanding of a Melanesian semiotic system to think about the Xingu case. The understanding that it produces is not articulated in the form of a collective and cumulative framework of knowledge such as our science, or our culture 30 . More than ways of knowing, however, myths (and shamanic journeys) form the known world itself 31 .

the unpredictability of the world
Although the Xingu myth speaks of the creation of sunlight in a world of darkness, we must remember that the life of the ancient people, whose actions gave rise to the world we know today, was already fully social: there were villages, harvests, weddings, rituals, spouse betrayal and children learning to walk. Hence the narrator sometimes needing to remind his audience by describing the making of the feathered headdress by Sun: there was no light at that time, only darkness. Order that establishes itself as alteration of another given order, creation is always a transformation. But if things can transform, what guarantees the continuation of the created world? The myth, Lévi-Strauss defines (Eribon & Lévi-Strauss, 2005), is a story about the time when animals were people, and about how they ceased to be so. But as Viveiros de Castro (2006) recalls, this passage is not thought of as an irreversible transformation -a point confirmed by the statements of the Aweti narrator and shaman about things that were (and to some extent still are) people. Speciation solves the issue of communicability between beings only in a precarious way. This observation has important implications for thinking about the temporal system associated with mythical thought -a temporality that is not linear, because it does not postulate a single and determined direction for change, but neither a cyclical one, because it does not postulate the obligatory repetition of any order 32 .
We know that Lévi-Strauss never intended, in proposing the distinction between hot and cold societies, that some societies would actually experience historical changes while others would stand still in time. The distinction rather designates different ways of perceiving history: while some societies (like us) would imagine their existence as the product of a linear historical process (evolution, progress), others (like the Amerindians) would think of its existence as a repetition of ways of being established in the time of which the myths allude to. We also know the role played by the myth in this system of thought: by changing, myth creates the illusion that nothing has changed, that we continue to do 30 "Order, consistency and relational paradigma do not 'model' knowledge and understanding for the Daribi, as they do for the culture of science. It is not that the Daribi are incapable of abstract thinking, or that they have some predisposition to or preference for the irrational; their logic and common sense, when the occasion demands it, is as good as anyone's. It is rather that Daribi tradition allocates human responsibility to the realm of the incidental and the tropic, and Daribi do not build themselves orders of knowledge. Indeed, in a world where convention and relational order are held to be innate, such articulation would be a bit silly: shall man replicate the work of the creator, like a foolish younger brother? Instead of pooling their knowledge as 'culture' , and drawing from the pool, instead of linking fact to fact and making understanding part of a collective effort, Daribi articulate and prize their flashes of understanding as things-in-themselves, frozen lightning bolts to be thawed out in nightly sessions by the firepit" (Wagner, 1978: 29).
31 This is the point of the Wagnerian theory of myth, if I understand it well. The connection between this theory of myth and a Daribi theory of language or speech (po) is evident, although not emphasized by the author: "We can think of po as a kind of force that mobilizes the object through the speaker and vice versa; it is something, objectifiable in itself, not a mere codification or description of other things. It is not an 'instrument' , as we sometimes refer to language, but a type of use that combines the user and the used one. It is invention" (Wagner, 1978: 55).

32
Overing says something similar about the Piaroa, "The mythical time of Piaroa things like this because our grandparents did so. It is in this sense that, interpreting a myth collected by Thevet in the sixteenth century, Lévi-Strauss notes how Tupinambá mythology, only half a century after the arrival of European settlers, had already incorporated the figure of the white man into the scheme of asymmetrical dualism projected in their cosmogonic saga -a commentary on the unequal power relations between invaders and landowners.
Although the distinction between hot and cold historicities does not imply an effectively circular (or static) temporality, Lévi-Strauss seems to conceive a correlation between the dynamics of societies and their particular way of thinking themselves in relation to history. In an interview with Georges Charbonnier (1989), the ethnologist compares cold societies to clocks, that is, machines that ideally can function indefinitely without losing their initial energy, while hot societies would be like steam engines, experiencing change as acceleration toward exhaustion and disorder. The critical spirit of this observation seems to me to be absolutely pertinent, but it is curious that the indigenous, the peoples of the South American lowlands, at least as far as I know, have not invented machines to tell time. Repetition in their world seems to be much less precise, less predictable, more erratic than the 360 degree turn of a pointer.
This does not contradict Lévi-Strauss' observation on the obliteration of time operated by myth. This is perhaps the paradox inherent to mythic thought, promptly demonstrated in the structural analysis by the acknowledgement that the variants are all equally (at least from the analyst's point of view) "true": in solidarity with the continuity of the ways of life, myth escapes the question of essence, always admitting the irruption of the new 33 . In analyzing the transformations undergone by Piro narrative accounted by different researchers over three decades, Peter Gow questions the process of the emergence of mythical variants from the point of view of a particular narrator: "What can we make of these changes? (…) It seems to me that the only correct hypothesis is that we are looking here at an important feature of Piro myths and mythopoiesis: that is, that as people age, they tell myths in an increasingly confident and complex manner by spontaneously transforming versions they heard long ago and their own prior narrations. This suggests that the life-course process of mythopoiesis, while experienced as closer and closer fidelity to an ancient source, is in fact the ongoing genesis of new myth versions. (…) This is the process of mythopoiesis" (2001: 87).
It is in this sense that Gow presents his analysis as an ethnographic exploration of the Lévi-Straussian theme of the obliteration of history through myth (idem: 288). His observation of Gow seems interesting to account for the conditions of possibi-historicity is not past time (dead and buried), it is, in a certain sense, an omnipresent time, which has a continuous effect on the present. Moreover, since the gods and other beings of mythical time eternally have the power to act upon present time, the effect of mythical time on the present is as unpredictable as the specific intentions of individual mythical agents. Thus, Piaroa historicity can not presume a solid basis of linear events" (my emphasis, 1995: 133-134).
The idea is valid for the Aweti case even though the unpredictability of the world is not associated with the intentional action of mythical beings, but with the fact that in the world, as in the myth, anything can happen.
33 In a famous essay on the origins of philosophy and its relation to mythology in archaic Greece, Jean Pierre Vernant (2002) makes two interesting observations for comparison with the Amerindian case. The first observation is that first philosophy asks a similar question to that posed by the Greek myth. Both, mythology and philosophy, the author says, are asking about the emergence of order from chaos -which, in very general terms, applies to the Amerindian case, although, as we have seen, the question is somewhat different. The second observation, which interests me more, concerns the transformations of this question in first philosophy: the problem of the passage from chaos to order is reformulated, with Parmenides, in terms of questioning of the principle or essence of things, that from which order could be constituted. Now that is the question to which the Aweti really do not seem to be lity of mythic variation in long intervals of time when certain transformations may go unnoticed. But what about variation when it is not only perceived but questioned by those who tell and hear myths? There is more here at stake, I believe, than a way of making change apprehensible through a system of knowledge that masks its manner of operation. What if the Aweti understand that variation is inherent to the uncertain character of the things that are there to be known?
Although it is exaggerated to say that for the Aweti all history potentially determines the configuration of a new cosmic or social order, we can say that this system of knowledge is inseparable from the recognition of an intrinsic opening of the world. If nothing can be definitely known, it is because nothing is definitely given 34 . I want to suggest with this that despite recognizing, through their myths, the continuity of a way of being in the world -and also recognizing deep discontinuities in their lives, I must add, since obviously these two perceptions are not irreconcilable -the Aweti understand that the fact that something remains the same means that everything can be transformed. The Aweti myth not only enunciates the variation of the world, it produces it constantly, with its multiple versions. It creates the world as variation.  (1996,2002) and Stolze Lima (1996Lima ( , 2005.