Content uploaded by Bernd Brabec
Author content
All content in this area was uploaded by Bernd Brabec on Nov 13, 2017
Content may be subject to copyright.
Bernd Brabec de Mori
A Medium of Magical Power:
How to do Things with Voices in the Western Amazon
Abstract
This contribution is about functions of the voice, specifically the voice of singers among
the indigenous group known as Shipibo-Konibo who dwell in the Western Amazonian
rainforests. Among these people, much of curing and sorcery (commonly called ‘sha-
manism’) is achieved by singing certain chants or songs. This is done by specialists
(mdicos) who acquired a certain magical power for their voices during year-long training.
Voice electrification is treated in a twofold approach: first, the question is raised, why
recorded and reproduced voices do not work as carriers of said magical power ; second,
how electric devices can be summoned by the specialist singers in order to enhance this
power. Therefore it is necessary to explain how Amazonian magic is constructed. Lan-
guage (in this case the Shipibo’s) plays a crucial role, because its power of conceptualizing
the conceived environment allow for active manipulation by the means of uttering certain
songs or phrases. Thereby, the voice is viewed as a substantial item of transmission in an
animistic ontology. By recording, this substance cannot be reproduced, the recording is
but an image (like a photograph).
Introduction: Electrified Voices do not Work1
The regional radio programme Voz de la Selva (Voice of the Rainforest) has been
operating for more than a decade now in Pucallpa, a provincial town deep in the
eastern Peruvian rainforest. Pucallpa is situated on the shore of the Ucayali river,
a main tributary to the Amazon. Along the river’s shores, for about 350 kilo-
meters upriver and downriver, the Shipibo-Konibo dominate the demographic
1 I wish to express my gratitude to the University of Vienna and the Austrian Academy of
Sciences for grants they provided for my fieldwork in Amazonia. More specifically, I ac-
knowledge the invaluable help by my wife Laida Mori Silvano de Brabec, translating and
explaining Shipibo terminology and song lyrics. Artemio Pacaya Romaina likewise provided
song translations and information about the history of Shipibo radio programmes. Finally, I
thank Dmitri Zakharine for surprisingly inviting me to the conference Electrified Voices and
to contribute to the present volume.
In: Zakharine, Dmitri and Meise, Nils (Eds.): 'Electrified
Voices. Medial, Socio-Historical and Cultural Aspects of Voice
Transmission.' Göttingen: V&R unipress, pp. 379-401.
image of the region. The Shipibo-Konibo (or Shipibo) are an indigenous group,
calling themselves noa jonikon (“we, the Real People”). Their language non joi
(“our word”, “our language”, or “our voice”) has been identified by linguists to
pertain to the Pano linguistic family. The Shipibo comprise about 45,000 people,
and many of them listen to Voz de la Selva once a week. This programme
broadcasts in Shipibo language, and it is not the first to do so. Already in 1970,
Laurencio Ramrez founded the Manguar Shipibo, and many more are
broadcasting today.
Glorioso Castro, founder of Voz de la Selva and also its operator, talks into the
microphone at 5 a.m., telling international, national, and regional news, greeting
relatives, delivering messages to remote communities (a battery-driven radio
can be found everywhere), and he plays regional pop music between his talks.
Since around 2003, satellite telephones have been installed in many Shipibo
communities, and only a few villages remain dependant on radio communica-
tion. Radio communication means turning on Glorioso’s programme, or using
the radiofona. In the mornings and afternoons, radiofona is operated in most
villages, and it works better in the afternoon, when the batteries are well-charged
by solar panels. Radio communication has been increasingly common in Shi-
pibo settlements for about two decades now2.
It is therefore not exactly accurate to speak of the Shipibo’s voices as voices
that resisted electrification. However, they have a strange attitude towards
electrified voices: Pancho Mahua, a singer and specialist for magic (a mdico,an
indigenous doctor)3told me in 2005, that recorded magic songs would not work
properly. For instance, curing songs, songs for sorcery or for establishing (or
separating) love relationships can be regarded ‘magic’. But these magic effects
seem not to be caused by what we could define as the acoustic phenomenon, by
sound waves in certain frequency spectra at certain amplitudes extending
spherically through a medium, preferably air. There seems to be something else
at work as well. Maybe it is because recorded magic songs would not work
properly, that Pancho, as well as all his fellow magicians, healers, and sorcerers
did not hesitate to have their—presumably powerful—magic songs recorded
and burnt on CDs by the fieldworker. They would not work. Why?
Without initially paying much attention to this question, Itried to investigate
2 Before around 1990, indigenous people depended on radio communication provided by
missionaries, or logging and oil companies.
3 I prefer to use the term mdico instead of the more common chamn, “shaman”, because it is
more often used in native discourse. Both terms are nevertheless introduced from Spanish and
reinterpreted. However, they carry a slightly different meaning: mdico is usually more often
used to designate specialists who work incuring people by applying indigenous medicine and
magic, while chamn is most often used to designate (and also autodenominate) specialists
who work mainly with visitors, that is with researchers, tourists, and “shaman apprentices”.
Bernd Brabec de Mori380
whatever was the reason that made magic songs ‘work properly’. This was part of
my fieldwork in western Amazonia from 2001 to 2006, which I conducted in
order to provide an ethnomusicological documentation of the area. This served
for my M.A. and Ph.D. thesis, and as a process of safeguarding for the people who
dwell on the Ucayali river. Beside the Shipibo, I also have been working with
members of the Yine, Ashninka, Kakataibo, Iskobakebo, and Kukama-Kuka-
miria indigenous people, among a few other groups. Although peculiarities in
treating the human (and the non-human) voice could be observed among all
mentioned groups, most of what will be said here about the Shipibo-Konibo also
applies for them—in different languages, of course.
Understanding Amazonian Magic
Many cultural anthropological studies have been conducted in order to under-
stand how indigenous people perceive, (re-)construct, and manipulate their
world, but few have understood and studied the importance of music, songs, and
voices in this regards. Anthony Seeger can be considered a pioneer who un-
dertook ethnomusicological fieldwork among the Suy, a small indigenous
group of central Brazil. An essay he co-authored (Seeger et al. 1979) laid way to
many newer theories about Amazonian understandings of human beings, non-
human beings, and about the process of defining individual and social bodies as
well as cosmologies. In 1987, Seeger contributed with his book Why Suy Sing to
what he coined a “musical anthropology”. Music performance in ritual contexts,
Seeger argues, is a key process for constructing, affirming, and recreating social
structures as well as “social” relationships between the human Suy and non-
human others. Following this approach, ethnomusicologists and anthro-
pologists have ventured further into the topic of ritual (and also recreative)
music in South American lowlands (Menezes Bastos 1986, Hill 1993, Olsen 1996,
Lewy 2010, among others). In most of these studies, it appears that not merely
what we would understand as music (instrumentally or vocally performed), but
also speaking in certain circumstances may be charged with magical power. The
voice seems to be crucial. However, in order to foster interdisciplinary collab-
orations, and to understand what is meant here with humans and non-humans, I
will provide a short introduction to Amerindian ontologies as formulated in
recent anthropological theories, before tackling with voices.
One of Seeger’s former students, Eduardo Viveiros de Castro, investigated
various aspects of affinity and relationships among the Arawet in central Brazil,
elaborating a model of different perspectives, for examples comparing the very
Arawet’s point of view with the point of view of their cannibalistic gods.
Thereupon, he presented the concept of Amerindian perspectivism (Viveiros de
A Medium of Magical Power 381
Castro 1998). With this theory, he proposes that in Amazonian ontologies the
concepts of bodily-exterior form and mental-interior content works contrary to
the naturalistic Western conception. Westerners typically see a human being
defined by a certain body, and locate the principle of differentiation in the
mental-interior domain (for individuals, their ‘soul’ or ‘psyche’; for groups, their
‘culture’). That said, humans, as mammals, share a determined physicality, even
plants and ‘inanimate’ beings like rivers, or mountains share physical existence
with humans. However, Westerners distinguish humanity based on mental
abilities (‘reflexive consciousness’), and also distinguish humans among each
other by their mental individuality. Therefore, beings in a Western ontology
share physical similarity (a pig is very similar to a human, it is said), while the
mental-interior category serves for distinction. In Amerindian societies, on the
other hand, Viveiros de Castro argues, all beings would share humanness in
mental-interior (‘soul’, ‘social’, or ‘cultural’) categories, while the bodily-ex-
terior serves for distinction and for individuality. When Shipibo-Konibo, for
example, call themselves noa jonikon, ‘we, the Real People’; this does not mean
that all non-Shipibo are not human, but quite contrary, that all non-Shipibo
(neighbor indigenous groups, mestizos, whites, animals, plants, spirits, rivers,
and so on) are humans but dispose of different bodies.
In that sense, it is perfectly possible for Amazonian indigenous individuals
(with a certain intra-culturally determined training) to change their bodily ap-
pearance. Whereas it proves impossible to escape from humanness in a mental-
interior sense—quite contrarily to what is possible in our Western, naturalistic
understanding (we can change our ‘culturality’, while weare eternally determined
in our ‘physicality’). It is important here to understand that most Amazonian
indigenous people do not understand being human as a social state in an envi-
ronment of independent nature. The environment is part of the social. Animals,
plants, or rivers, are thought as similarly human persons living a human social, or
cultural life, in different bodies. In consequence, it is possible to contact these
persons and establish social relationships with them4. Such contact, however, is
not pursued by ‘common’ people, because it is understood as a very dangerous
endeavor. If someone would, for example, get in contact with the river dolphin
people, he or she would risk to be captured by them and become one of them.
Therefore, it is most important to maintain the order of things: ‘We Real People’
shall stay ‘really human’, while ‘other humans’ shall not disturb us. The main task
of trained specialists (mdicos, commonly referred to as ‘shamans’) is not to
4 Cf. the conceptualization of animistic, naturalistic, totemistic, and analogistic ontologies in
Descola (2005).
Bernd Brabec de Mori382
provide contact to non-human beings for their community but on the contrary, to
defend and prevent their community from the non-humans’ influence5.
Remarkably, in anthropology, the visual has been much more emphasised in
this context than the acoustic. Viveiros de Castro, for example, is right that
“Amerindian cultures evince a strong visual bias of their own—one not to be
confused with our own visualism” (Viveiros de Castro 2004 : 26). The connection
that he and others6draw between visual perception or terminology and the ritual
use of hallucinogenic drugs does not lack reason. In the Western Amazon, much
of the mdico’s work is done with the influence of the hallucinogenic plant
preparation ayawaska7.However, the same applies to acoustic phenomena, as
aural hallucinations are as often reported as visual ones. Although indigenous
people excessively use visual metaphors, the fact that these receive more at-
tention than the audible, I would like to nevertheless interpret as a result of the
Western researcher’s “visualism”. What startled me, the ethnomusicologist,
most in Shipibo people’s reports from their drug-induced perceptions was the
consistency of enunciations like “When I am singing, I sing like the Inka who I
hear in my mareacin (drug influence)”.
Lewy argues in this context that although seeing is emphasized as the sense for
distinction, the aural skills (hearing, listening, speaking, and singing) are used to
communicate among ‘others’, that is, inter-specific interaction. “Different see-
ing, similar hearing” (Lewy forthcoming: title) is the key for understanding the
mode of communication and in consequence of transformation between ‘Real
People’ and ‘other people’.
The Spirits’ Language
The important role of talking and singing makes oral language use8a more
delicate issue than in a Western naturalistic understanding. Whereas language
use in Western societies may be delicate, too, this delicacy is due to possible
5 For an explanation of border management in the ‘multiverse’ and the tasks of ‘shamans’
therein, see Halbmayer (forthcoming).
6 An important contribution to South American anthropology of the senses involving hallu-
cinatory experience is presented by Classen (1990).
7Ayawaska is cooked from two forest plants (Banisteriopsis caapi and Psychotria viridis in the
most common recipe). Preparation, pharmacology, and modalities of use of this preparation
has been studied extensively. For a comprehensive collection of essays about ayawaska, see for
example: Labate &Araffljo 2004.
8 That is, language use in general, because there was little to no written language in the Amazon
before the mid 20th century. For a highly interesting historical reinterpretation of written
language (and songbooks!) among the Shipibo, see the investigation about legendary Shipibo
‘design books’ in Tournon (2002 : 82 – 86).
A Medium of Magical Power 383
social or political consequences. In Amazonia, and among the Shipibo people,
language use however is considered an application of power, a power understood
as ‘magic’ by Westerners because we lack any scientific explanation of what
indigenous people are talking about.
“To speak”, in Shipibo language is expressed with the verb yoiti. This verb is
composed by its stem iti and the prefix yo-. The stem on its own, iti is generally
used to denote or (as a modal verb) substitute intransitive processes, and can be
most adequately translated with “to be” (to exist) or “to do” (with intransitive
acts). In Shipibo, “to sing” lacks a general term and can only be expressed
directly when adhering to the current genre performed: mash iti, for example,
has to be translated exactly with “to perform mash (repetitive song accom-
panying the mash round dance)”, or bewati with “to perform bew (song genre
with a certain descending melodic/rhythmic structure)”—in both cases “to
perform” is expressed with iti. The only way to translate “singing” as a general
term for melodic voice production is via iti,ori-xon-ti, the latter including a
morpheme of benefit (-xon-), meaning “to sing (be) for somebody”. Iti therefore
denotes a current mode of being, of existence; in case of song performance the
current mode of existence embodied by a performing singer.
Still, there is the prefix yo- in the verb “to talk”, yo-iti, that stands for power :
Illius describes this power as “charged with energy [energiegeladen]”, and
presents a list of words that are composed with this prefix (Illius 1999: 41– 42)9.
Therefore, “to speak”, yo-iti can be translated literally as “a mode of being
charged with (magical) power”.
This power involved in voice use implies a series of consequences. In spoken
language, the boman skill proves most important (Brabec de Mori 2009 : 132 –
140). Boman signifies direct or indirect manipulation of the world via ‘charged
speaking’. This way of speaking can be trained by the way of sam (“diet”,
“fasting”), and every trained mdico is thought to hold this power; they are
boman koshi,“boman powerful”. Although this skill can be applied for any
manipulation regardless of its moral implications, it is usually viewed rather
sceptically. In general, boman is thought to be mostly used as a tool of cursing or
even attacking:
“a sorcerer may utter ‘look, the girl over there stumbles !’ and when looking there one
will see the girl falling down. Or he could say, ‘a tree branch is falling on Juan’s head’,
and Juan will be brought home from the woodwork with a serious cranial problem”
(Brabec de Mori 2009: 143).
9 Illius (1999) mentions e. g. yo-ina, “wild animal” (as opposed to in, domesticated animal);
yo-shin, “demon, spirit”(from shin, “in darkness” or shin(-an) “thought, feeling”); or yo-b (a
dart magician or “shaman”).
Bernd Brabec de Mori384
In everyday life, the term boman is also often used with a certain ironic touch
when somebody’s assumption proves right. For example, one Shipibo art seller
may say “look at this foreigner, he will by our art”, and if this proves true, her
companion may reply, “you’re quite boman koshi!” From this can be read that
the act of the visitor buying artwork can be seen as an occurrence by chance
(Shipibo people would also normally tell so), but also as a consequence of prior
powerful uttering by the vendor. Consequently, some Shipibo art sellers train
their sam in order to develop boman skill, as it greatly improves sales figures10.
Singing, as was indicated above, is considered a special mode of existence, in
particular a ‘higher’ one than the normal being iti. This becomes obvious, for
example, when Shipibo people tell about the chaikoni jonibo. These are legen-
dary people who live far removed in the woods, invisible for untrained com-
moners (and whites), and who retain a way of life which is considered proto-
typically “correct” (kikin,jakon) by the Shipibo. The chaikoni jonibo would, as a
consequence, live in well decorated houses, wear most delicately embroidered
garments, use many perfumes only known to them (which also make them fairly
irresistible if they wanted to conquer a common Shipibo), have the power to
command all sorts of animals and to perform any rainforest magic. And, re-
markably, they do not talk to each other, they always sing.
Chaikoni jonibo may make their appearance among common Shipibo people,
in secret when visiting a Shipibo lover during night-time, but also publicly,
during the great drinking festivities ani xeati. These festivities were periodically
performed in Shipibo communities until a few dozen years ago. One of the main
features of such festivities was excessive singing—drunk (paena) Shipibo are
understood as entering another mode of existence, too, a mode wherein singing
is likewise considered the adequate form of communication. Having reached a
fair level of inebriation, all Shipibo present used to communicate only via
singing. In such a context it appears almost natural that chaikoni jonibo may join
the party and delight the audience with their especially refined singing.
Remarkably, outside of festivals or other communal or ritualized oppor-
tunities the performance of love songs, drinking songs, or dance songs is very
rare. Illius argues that singing in general means to apply the “spirits’ language”11
10 Besides boman, Shipibo (and other Western Amazonian) people use many different magical
techniques for negocio, “commerce”, like the popular perfume preparations pusanga which
are also used for love magic. At this point, it shall be mentioned that olfactory senses and
skills are likewise regarded crucial in Shipibo magic, but treating this issue lies beyond the
scope of this chapter. In some instances, olfactory and audible sensations or actions are
combined and complement each other even to the degree of synaesthetic perceptions ; see
Classen (1990), Santos-Granero (2006), or Brabec de Mori &Mori Silvano de Brabec (2009).
11 Illius (1997: 216): “Musik ist die ‘Sprache’der Geister; das Singen die angemessene Kom-
munikationsform mit ihnen”.
A Medium of Magical Power 385
and using their language would mean to summon them, or at least to be per-
ceived by them. This is definitely not intended by common Shipibo people
during everyday life.
However, what is it that makes a simple drinking song the spirits’ talk? There
is, on the one hand, the formalization of an ‘existence charged with power’:
speaking, yoiti. This means that using words, human talk (in particular, Shipibo
language) makes us (the Shipibo) something special—powerful. It is one of the
main factors that distinguishes the ‘Real People’ from ‘other people’, among
artwork, body paint, body deformation, and house or village structures (cf.
Halbmayer 2010). Viveiros de Castro indicates with “cosmological deixis” (Vi-
veiros de Castro 1998: title) that ‘we, the Real People’ (noa jonikon) do not only
exist as the others do (in Shipibo, the others do iti), but exist as ‘we’, as those who
use the “correct words” joikon or “correct tongue” janakon, both terms which
are synonymous to non joi “our language”. Applying these markers of cor-
rectness, we ‘yo-exist’, we exist charged with power. Speaking in Shipibo lan-
guage reveals our own ‘Real Humanness’. As a consequence, those more pow-
erful than we are supposed to use even more ‘correct’ words or tongues. The
chaikoni jonibo use the common Shipibo language, but they super-formalize this
language because their talk is song; with this they evidence their unquestionable
super-humanness.
Sung words are by definition powerful, and so they are thought to take effect
also in non-magical contexts like in drinking and love songs. I formulated this
concept with an example in a recent publication:
“a man, for example, may sing in order to have his desired girl fall in love with him. A
woman may sing to address her secret lover’s potential understanding that she would
like to flee with him to another village. As illustrated by these examples, numerous
songs—however “secular” they may be—are thought to cast effects upon persons sung
to. This understanding of effect is coded in metaphoric language. In song lyrics, for
example, people are referred to as certain animals. This ascription of animal identities
to human persons is not descriptive but prescriptive: the male singer who tries to
seduce a girl would, for example, name her bontoish. The bontoish is a beautiful small
bird, and in the code of Shipibo song lyrics it is used to address a young, good-looking
and marriageable person (whose sex is usually the opposite of the singer’s sex).
Mentioning bontoish in this context does not describe the girl’s behaviour but actually
prescribes it: the singer defines her ideal reaction for the near future through mean-
ingful naming” (Brabec de Mori 2011a: 170).
This function of sung music, especially regarding lyrics, does not appear magical
to Westerners, because from a naturalistic point of view, we can interpret the
process by sociological means. For Shipibo people however, both this socio-
logical and the magical effects of sung words are completely comprehensible and
Bernd Brabec de Mori386
rational12, because they are rooted in Western Amazonian ontology, where
bodily transformation is an ever present threat as well as a most effective tool.
How to do Things with Voices
Going further with a sociological, or rather linguistic interpretation, one could
refer to speech act theory13. Very similar to the singer’s prescription of the girl’s
behavior by naming, Austin defines performative speech acts as a specific var-
iant of illocutionary speech. As an example he presents the famous quote of a
man responding to the priest at a wedding, “I do (take this woman to be my
lawful wedded wife)” (Austin 1962: 12). This is illocutionary because it definitely
exceeds the mere act of description, and it is performative, because the very act
of utterance is part of the far-reaching consequences for the involved persons’
future agency. As far as I am informed, in linguistics there is no reasonable
explanation what is actually the power behind such a promise: there is no
measurable substance involved during and after such a performative speech act.
In Shipibo understanding, the voice itself is viewed as such a substance
charged with power respective to what it utters. Drawing upon the boman skill
mentioned before, one can see that this substantiality can be trained, or en-
hanced by ‘dieting’ (samati). With ‘dieting’, the voice’s power is ‘fed’, and in both
speaking and singing, a mdico’s voice unfolds different powers of manipulating
or changing the world. The mdico’s voice can therefore enter a patient’s (or
victim’s) body and operate on it like a surgeon’s scalpel; it is a common per-
ception in Shipibo rituals that a patient feels “pierced by the voice/word”, join
tsakana, or “pierced by the song”, bewan tsakana, as if the voice was a spear. For
comparison, a speared fish is “pierced by a harpoon” (pian tsakana).
In order to elude the substantiality of voice more clearly, in the following I am
going to explore the effective principles of three techniques of magical singing:
koxonti (to blow/whistle), yointi (to connect), and naikiti (to transform into a
non-human). Magical songs are not explicitely separated from secular songs in
Shipibo terminology but they can be distinguished by context, and by their
implicit purpose of actively manipulating the world. They are, as well, only
performed by trained specialists, by mdicos (although anybody might try, but
this could result in disaster). These techniques are usually performed in specific
12 Here, I am following the historical definition of magic by Tambiah (1990).
13 I refer to the terminology as coined by Austin in his influential work How to do Things with
Words (1962). Before and after Austin, many authors have written about speech and its
effects (magic or not), e. g. Wittgenstein, Burke, Searle, Turner, Tambiah, or Derrida, only to
mention a few. A comprehensive analysis of magic with a strong focus on performance and
speech can be read in Tambiah (1990).
A Medium of Magical Power 387
situations: koxonti is mostly undertaken during daytime, and in everyday sit-
uations. Naikiti, the transformative faculty, on the other hand is only performed
during controlled rituals, like sessions involving the hallucinogenic ayawaska
plant drug, or also without psychoactives, most prominently in legendary curing
rituals by master healers (merya). In all cases, yointi is the most important
aspect of the singer’s voice uttering. In the following, koxonti,yointi, and naikiti
will be explained in detail, focussing on the function of the voice.
The most discrete and least spectacular use of voice in magical singing is in
koxonti. Koxonti is always directed towards an object:14 a song is performed in
order to charge a substance (usually plant preparations, tobacco, perfumes, or
any objects used in indigenous medicine or sorcery) with power. This is achieved
by holding the object close to one’s mouth while whistling (in a sense of me-
lodious blowing) the appropriate melody. At the same time—most im-
portantly—the performer has to think the appropriate lyrics15. This substance is
then applied, for instance tobacco smoke is blown over the patient’s body, and
the power of the song is transmitted via this substance.
Koxonti is performed very silently and resembles a somewhat homeopathic
approach. After performance, the whistled voice and the imagined lyrics stay in
the object, therefore transforming the object’s semantic interiority. A tobacco
cigarette is just a cigarette before koxonti, but afterwards, it is the cigarette. The
‘song in the cigarette’ can be transported, and the cigarette can be smoked
elsewhere, much later, even by somebody else than the original singer (whistler).
Besides tobacco, liquids are often used as ‘voice carriers’, like perfumes which
are then blown over the patient in a fragrant explosion of olfactory experience.
The mestizo mdico Julin Vsquez often used koxonti (icarar in regional Lor-
etano dialect) on a glass of water which his patient then drank. Similarly, koxonti
is almost always applied if a mdico prepares a remedy of plants or other sources
which will then be ingested by the patient. Therefore, the mdico may blow on a
whole bottle or canister before handing it over, or just on one dose in a mug
which is then drunk at once.
The praxis of applying songs as ‘whistling with lyrics’ contributes to the
image that mdicos would only sing aloud during ayawaska sessions. Therefore
it appears that loud songs would not play a great role in indigenous magical
practice outside the ayawaska context. I think that this praxis indicates some-
thing completely different: ayawaska sessions (if not staged for tourists or re-
searchers) are practically reserved for ‘serious cases’ or for occasions when more
14 During ayawaska sessions, ayawaska drinkers may at times also perform koxonti at the start
of a session without focusing on a concrete target object.
15 This technique of imagined music or texts is also known among the Venezuelan Warao in
hoaratu magical songs, see Olsen (1996 : 259 – 260).
Bernd Brabec de Mori388
than one patients should be cured at the same time. Koxonti, on the other hand, is
used by mdicos in everyday situations anytime and anywhere, for example, for
healing the neighbor’s child of diarrhea, when handing over remedies as de-
scribed above, for taking away bad dreams, and so on. Such ‘small’ treatments
actually occupy most of a mdico’s time of practical labor (again, except those
who work with foreigners). It can easily be understood that—during such ‘small’
intervention—the mdico does not sing aloud, in order to not attract the an-
noying attention of non-humans summoned in this way. Non-humans, as argued
by Illius (1997: 216), understand the ‘spirits’ language’, song. On the other hand,
during long and intense rituals like ayawaska sessions, such confrontations are
often sought actively. In reports about the merya (the great healers of the past),
it has to be noted that they did often not ingest ayawaska, but performed intense
rituals including songs uttered loudly with pronounced lyrics, sometimes in
masked voices (see below). Therefore it is wrong to see a connection of loud
singing of magic songs with the ayawaska praxis. Instead, it is merely important
that in everyday situations it is crucial not to attract potential danger by (un-
intentionally) summoning non-humans, but to restrict the latter to controlled
and concentrated rituals. With that, it should also be clear that by concealing the
lyrics it is not intended to hide the semantic contents from human audience but
definitely from non-human audience (cf. Brabec de Mori 2011b : 429 – 430).
With this, the rare occurrence of loudly sung magic songs outside the aya-
waska context is clear. In former times, when the merya’s curing rituals without
ayawaska ingestion was common, obviously such songs were performed.
However, in some occasions until today, healers may sing loudly. In the two
examples that follow, both singers sang songs aloud for my recordings by re-
quest. Otherwise, they would perform these songs as koxonti on a liquid their
patients should then drink (water, or a plant remedy). The first excerpt is from a
song for healing a child from diarrhea:
joxo yawish ainbaonra The white armadillo woman
oripari yakayakakin still is sitting over there
jakon akin tsosinai and with great care she thickens,
jawen mewe tsosinai she thickens the mud by wallowing,
mewe tsosinai thickens the mud by wallowing16.
This is one sequence of three this singer performed. Here, he simply describes
what the armadillo woman does: wallowing. While wallowing, the mud thickens.
16 Excerpt from a bake benxoati song (lines 11 – 15), performed in an explorative setting by
Claudio Snchez Serafn in Caco Macaya, 2004. All my recordings are archived at the Vienna
Phonogrammarchiv, this one with signature number D 5346.
A Medium of Magical Power 389
In the two preceding sequences, he mentions that both the white ronsoco woman
(a big rodent) and the white agouti woman are sitting over there, eating sweet
potatoes. With that he resumed what healing diarrhea is about: the child should
eat proper food and his or her ‘mud’ should thicken, that is all. In the given
context the method is interesting, because the healer achieves (we assume)
healing success by mentioning these non-human people. That they are proper
people (though not ‘Real People’) is expressed with the use of ‘woman’ (ainbo)
instead of ‘female’ (awin) commonly used for animals. Remember that non-
humans can hear, listen to, and understand sung words, ‘the spirits’ language’. In
the present example text, these animal-persons are named and specifically at-
tracted. How this mere attraction results in actual attachment is described in the
following excerpt from a song for preparing a mother to give birth:
jishaman ik iki Watch carefully how it is :
jene xaman rakata she lies in the depth of the waters,
iwin tita rakata there lies the ray’s mother.
ishkai pacha bake
ataanan
After giving birth to hundreds of children
jawe iyamarairi she is still unaffected.
jakipari yoinbanon Towards her I am going to connect
ainboan yora
yoinbanon
connect the women’s body17.
In two more sequences, the singer mentions the mothers of the fish nosha and
wame, mentioning their perfectly slippery body—and connecting them to the
woman. This connection is achieved with the term yoinbanon, from the verb
yointi. The attentive reader remembers: yo-iti can be glossed as ‘charged.-
with.power-to.be’ and can be translated as a mode of existence expressing power,
or more simply as “to speak”. Yo-iti is intransitive, and yo-i-n-(i)ti is made
transitive with the transitivity morpheme -n- (Valenzuela 2010 : 71, 88). Imag-
ining a transitive powerful mode of existence proves fairly difficult: ‘powerful
existence’ which is usually expressed by speaking is applied on something or
somebody. Imagining, one may achieve animpression that the ‘voiced existence’
is extended over the one who utters (the singer), the one he addresses (the
mother) and the one he mentions (the ray). Grammatically speaking, this is a
raising-to-object construction, allowing for certain qualitites among those
17 Excerpt from a bakemati song (lines 6 – 13), performed in an explorative setting by Vctor
Ancn Cruz in Caimito, 2005. Signature number D 5533. The ray, as naturalistic biology
explains for us, does not give birth similar to mammals, but its spawn is kept in a portion of
the mother’s body and the offspring leave it only after hatching out; therefore it appears that
the mother-ray would give birth to living baby-rays.
Bernd Brabec de Mori390
mentioned to be shared, or more precisely, for their “active zones” constructed
by such speech to interact (cf. Langacker 1999 : 330 – 336). By yointi, transitive
speaking, the singer establishes what I call a ‘performative ontological con-
nection’ among them18. The ray-person is named and summoned (because the
words are sung), the same with the human woman. With that they ‘meet’ and
connect and the ray’s properties are transmitted (therefore ‘performative’). This
takes place in the ‘real world’ of dream and visionary experiences of the mdicos,
where space and time are unmade and which hides behind the ‘virtual reality’ of
physical existence (therefore ‘ontological’)19. When the song is over, the con-
nection lingers and stays effective. In a similar sense as with koxonti, where the
song stays in the substance, with yointi, the song stays in the ontological posi-
tions of addressees, which keep in active communication as long as the song
‘endures’ there. All this may appear fairly strange for readers not familiar with
Amazonian indigenous ontologies. I may only add here, that in Shipibo un-
derstanding this is quite logical and well embedded in their ontology20. In the
following, some processes will appear even stranger.
The third way of performing magic songs can be observed in rituals dedicated
to curing one or more patients, usually during night-time. The powerful specialists
of old, the merya, so it is told, would enter the spirits’ world or transform into
spirits themselves by the power of their concentration (shinan), by ingesting
tobacco juice (romoe poi), and by the power of their voice. Nowadays it is common
that the healers ingest the hallucinogenic plant preparation ayawaska.
An ayawaska session usually starts after nightfall, and when the common
mosquito assault (from ca. 5 p.m. to 8 p.m.) has somewhat diminished. Usually,
the healer, one or more patients, and relatives are present, sitting or lying on the
floor inside a house (traditional Shipibo houses consist of an elevated floor and a
palm-leaf roof but lack walls). At some point of talking about the patients’
problems, family issues, joking, and everyday dialogue, the healer will take a
draft from his bottle containing ayawaska. The dialogue continues, until the
drug shows first effects, indicated by the healer with initial masked sounds21.A
18 I am as ever grateful to my wife Laida Mori Silvano de Brabec who explained yointi among
uncountable similarly difficult Shipibo terms to me. She is native in Shipibo language and
her father was a mdico, so she provides necessary prerequisites for understanding these
issues which are seldom object of everyday Shipibo talk.
19 For the “real” vs. “virtual reality” see Santos-Granero (2006: 61) ; for the ontological concept
involving performance among human and non-humans see Descola (2005). This commu-
nication is performative in Austin’s (1964) sense, because it includes the speech act as part of
the effect, and the effect determines future agency.
20 Cf. Viveiros de Castro (1998), Descola (2005), Santos-Granero (2006), or Brabec de Mori
(2011a, 2011b, forthcoming), among others.
21 The concept of voice masking in spirit transformation is well describ ed by Olsen (1996 : 159 –
162, 192) in the context of musical curing among the Warao.
A Medium of Magical Power 391
hush falls and at some moment the healer starts singing. Usually a series of songs
is performed for each patient. Finally, the healer stops singing, and people go to
sleep without much fussing. Shipibo traditional ayawaska drinkers do not use
ritual paraphernalia or spectacular openings or closings as in ceremonies staged
for Westerners. The ritual time and space are defined by its mere application.
The basic conception within these rituals is the transformation (yoshinti,
naikiti) of the healer into a non-human being in order to obtain a subject
positionality different than, and a competence of manipulation more powerful
than the common position of Shipibo people. During this state of altered posi-
tionality, the healer will sing songs of power by non-humans which he receives,
or hears during his altered state of perception. He then imitates, translates and
performs these originally non-human songs for the audience (especially for the
patient). However, the main address of this singing is not the human audience in
the ritual but the non-human beings that can now be directly addressed by the
singing healer. Here, the singer’s mode of existence is again heightened due to
two effects: he is drunk (paena) with the hallucinogenic drug and he is singing.
While the singer may directly experience synaesthetic perceptions in his altered
state, the human audience is also confronted with a multimedia package: the
singer changes his shape in the dark, by moving his body from one place in the
ritual room to another, inhabitating different positions of transformation and
often indicating this with certain gestures and postures22. In most cases, the
healer also uses perfumes (in former times for example the mapichi fragrant
plant, nowadays most often industrially produced perfumes like the popular
Agua Florida) and provides massages (xeyti) to the patient. Although it is
usually pit dark at such rituals, almost all senses of the patients are stimulated.
The main source for orientation, however, are sounds, specifically the singer’s
voice. In both cases—in a merya’s performance like in the past, without drug
intake, and in ayawaska rituals—communication with and transformation into
non-humans plays the crucial role. Thereby, two different conditions have to be
distinguished: the singer can mimic (paranti) non-humans, or the singer can
transform into (naikiti) non-humans (or even into other ‘Real People’). This
distinction can only be undertaken by the executing specialist himself reporting
on his own experience, because the observable phenomenology of applied voice
production is similar in the conditions of paranti and naikiti.
The mentioned masked sounds at the beginning of an ayawaska session
indicate the healer’s transformation. The use of (visual) masks in Amazonian
22 This kind of postures must not be confused with Felicitas Goodman’s “shamanic” postures
in popular neoshamanism. Here, the posture is not taken in order to reach an altered state,
but it is an effect of transformation of the singer who already inhabits an altered state. The
posture is perceived visually (! ) by the audience in the gloom as if the singer would have
changed his shape into something non-human.
Bernd Brabec de Mori392
societies is explained by Viveiros de Castro (1998: 482): Wearing a mask does not
provide the mask with life, but it provides the wearer with an alternative body.
These masks are not like carnival masks but are used similar to diving suits or
astronaut’s space suits. They enable the wearer to navigate in regions usually
inaccessible for ‘Real Humans’. The same is true for voice masks, however with
said distinction: when mimicking, the healer retains his humanness (like
wearing a carnival mask) but deceives others with his voice’s sound. When
transforming, Viveiros de Castro’s description applies: the voice mask provides
a heightened power of yointi to the singer, dependent on who he transforms into.
These non-humans, according to animism, perceive themselves as humans, too.
In the healer’s self-perception, he therefore still retains humanness, but now,
because he is re-located in the world aspect (niwe) of the non-humans he
transformed into. But the healer’s audience now hears the non-human’s voice
and sees a different shape of the healer’s physical appearance in the gloom. He is
now something else: jara moa yoshina (he is now transformed into a spirit).
It is commonly known in literature about ayawaska singing, that the singers
claim to ‘receive the songs from the spirits’. However, by analyzing a broad
corpus of recorded performances by different mdicos, one will be aware that
actually the musical structures and basic lyrics are transmitted from father to
son, of from teacher to student. It was not yet recognised23 that ‘a song’ in
Western Amazonian ontologies is not defined by its musical and textual struc-
ture as in Western understanding (where most ‘songs’ are written down and
therefore fixed in structure). A ‘song’ is defined by its actual performance, and
every rendering of similar (musical or textual) structures is another ‘song’.
Consequently, it is not the structure, which is transmitted from the spirits, but
the form of the actual performance : additionally uttered unique lyrics phrases
and most importantly the performance style. The applied voice mask defines
pitch, intonation, timbre, tempo, and accentuation. A musical structure, ob-
viously a ‘Real Human song’,is therefore rendered in the way the respective non-
humans sing. Voice masks include for example a nasal singing style, high tempo,
or on the contrary, a grave voice with throat-generated roughness. The most
common mask is a high falsetto voice, with or without nasalization. The ‘bril-
liant voice’ (wirish) indicates current contact with powerful non-humans or
super-humans (like the chaikoni jonibo or the Inka).
Almost ridiculously high falsetto singing is also a common feature of ‘secular’
singing at drinking feasts, for example. One male singer, Benjamn Mahua,
performed a drinking song in a pitch reaching a4 (877 Hz), while his wife
Antonia Ahuanari still clearly pronounced her lyrics at a pitch of c5 (1,023 Hz).
23 I have presented a more detailed analysis of song transmission and performance styles in a
conference paper (Brabec de Mori 2007).
A Medium of Magical Power 393
This raises the question, if high falsetto singing is but ‘only a culturally de-
termined stylistic feature’ of Shipibo singing in general and therefore is also
applied in magical singing (dismissing voice masking as fantasy). On the other
hand, and I adhere to this interpretation, could voice masking also be performed
in non-magical singing, in drinking songs, or love songs? If yes, why?
In a prior section, I presented an explanation that the ascription of animal
identities in love songs is a prescriptive act, or a performative speech act. Here,
we have to consider that the singers are still rooted in Amazonian ontology : they
do not think like sociologists, but while singing, they use “the spirits’ lan-
guage”—this also applies for drinking songs. Non-humans would understand
this language and therefore be attracted by singing. The singers therefore make
sure that not the dangerous water demons, for example, are attracted (by ap-
plying a hollow, low voice with slight staccato, for example), but instead powerful
entities viewed as positive, benevolent, and superhuman (as the chaikoni jonibo,
for example, or bird-persons). If the male singer, in his love song, addresses a
beautiful girl with the bird-person identity of bontoish, he ideally performs this
song in high falsetto voice. This mask provides him with a voice that can be
understood by the bontoish-bird-people who therefore may establish an onto-
logical connection with the girl, in a way similar to yointi in magical singing.
However, there is a difference to be noted which sheds more light on the power
of voice and masked voice: In magical songs, the singer does not identify his
patient directly with the non-humans. In the example texts, the armadillo
woman, or the ray’s mother were mentioned in third-person perspective, not in
second-person address like a bontoish in a love song. This is due to the ‘pow-
erfully charged voice’ of magical specialists, who have the boman skill. In
magical singing for curing, the connection is established by yointi, but no direct
identification takes place. If the mdico would directly address for example the
child as armadillo, by the boman-power of his voice, he would initiate the child’s
transformation into an armadillo, resulting in madness, and death. This tech-
nique, however, can be applied in sorcery.
A non-mdico lacking boman power can therefore directly associate the de-
sired girl with a bird in order to establish an ontological connection. A mdico,
when applying his boman voice, on the other hand, has to be rather careful in
doing so in order not to transmit his own transformative faculty on his patient.
Considering this analysis of singing in Shipibo life, it should be obvious by
now why it is no good to carelessly sing along when working (worst if working
alone in the woods), or why children who play at the edge of the forest or on the
river’s shore should better not sing funny tunes. Singing, iti is a special mode of
existence which puts one into a position of possible communication with any
non-human beings, from the chaikoni jonibo who are generally seen as benev-
olent though however fierce enemies when annoyed, to animals, to forest ogres,
Bernd Brabec de Mori394
even to deadly diseases like smallpox brought by the smallpox-persons moro
jonibo. It is no good to enter this mode of communication if one does not know
very well about inter-species diplomacy.
Electrification without Electricity
Amazonian voices do not depend on electrification, nor do they usually suffer
from it or are violated by it. As was shown in the examples at the beginning of this
chapter, Shipibo people and their neighboring indigenous groups integrate their
use of radiofona, radio programmes, telephone and tape, CD or DVD recorders
or players into their daily life and their ontology quite easily. When living in the
Ucayali valley, I had the clear impression that most indigenous people had much
less problems with and were much less skeptical about technological in-
novations than for example, the Austrians (where cell phones are suspected to
cause brain tumors, high-speed trains may cause swollen feet, internet use will
damage children’s development, etc).
Electrification is used in many ways, but most interesting is voice elec-
trification without involving electricity in the narrow sense. In the following
example text, an old singer sung her remembrance of how she used to sing
decades ago at the big drinking festivities :
nokon joi rario My voice/song (was similar to the) radio
tskon ewa makina (like the) big machine for playing discs.
manetaitoninbi (My song was) the one maintaining (the rhythm)
tori ewa nakewe inside the great tower with designs on its interior,
chei xawan joyoya (where) macaws and parrots (danced) in a row.
manetaitoninbi The one who maintained (the rhythm),
tsoawanin xaranya was surrounded by tsoawa-birds24.
Here, loanwords from Spanish are used as metaphors for an impressive singing
style that does not falter and makes people dance. The radio and the disc player
are metaphors for the singer’s strong voice which sounds as if it were played
from a record. The great tower with designs on the inside stands for the situation
of the festivity where all people and objects are as fairly decorated as possible, in
particular for the circle of dancers when performing the mash round-dance : all
dancers face each other in the circle, so they can see the garments embroidered
or painted with designs and the likewise painted faces of their fellow dancers.
24 Excerpt from an itiki ik song (lines 19 – 26) performed by Car men Prez Medina in Pao-
cocha, 2004. Signature number D 5243.
A Medium of Magical Power 395
The macaws and other birds refer to the dancers, too (in the prescriptive sense
explained above).
Also in magical singing, electric apparatus appear:
inkanrai jonibo The royal Inka persons,
rios rai jonibo divine royal persons,
noara ikanai we are singing.
benebenebetani Together we enjoy ourselves,
makinanin bewai the machine is singing.
bewbewabetani We all sing together,
radionin bewai the radio is singing.
mochan bewabetani Together, (we are) singing a mocha song,
noara bekanairi as we are closing in25.
This is an excerpt from the lyrics of a ritual song called mochai. It is beyond the
scope of this chapter to explain the mochai ritual, but it shall be said here that
when it was still performed many decades ago it served for either summoning,
worshipping or healing (in the case of eclipse) the sun or the moon, and at the
same time for summoning all sorts of powerful spiritual beings (here, the Inka
persons) and for temporarily transforming into or merging with them (see
Brabec de Mori 2011b : 447 – 464). It seems that this ritual was a main instr ument
for the Shipibo to collectively negotiate their position among their peer in-
habitants of their animist ontology. The singer who sang this recorded song in
particular however provided an interpretation for current performance of such
songs: he said that he would use this kind of ritual songs still for healing cases of
severe witchcraft (saladera). By summoning and merging with the Inka persons
in a symbolical way (or during a night-time ritual in a fairly convincing way), the
singer bestows their superior healing and cleaning power upon his patient. Here,
the singer also mentions powerful Spanish loanwords, namely rey (king) and
Dios (God), as well as the machine and the radio again. In the same way as above,
the radio and machine are not meant as a description of actual reproduction of
recorded music in the same situation, but a prescriptive attribution of power to
the singing voice, which should sound firm and without errors as if it would be
played from a record.
Among the specialists for magic singing, in particular those who perform
their hour-long renderings under the influence of ayawaska, radio and re-
cording metaphors are frequently used. In their songs they often refer to light-
ning and electricity in general, to telephones, satellite communication, parabolic
25 Excerpt from a mochai song (lines 16 – 24) performe d in an explorative setting by Virgilio
Lpez Crdova in Vista Alegre de Ipara, 2004. Signature number D 5291.
Bernd Brabec de Mori396
antennae, cell phones, and internet chatting. These items are used as tools for
summons, mimesis and transformation in exactly the same way as chaikoni
jonibo,Inka, spirits, plants, animals, rivers, clouds and so on are addressed as
was described in the preceding sections of this chapter—and therefore, of
course, their use is efficient in the indigenous sense. If using electrification in an
indigenous way, without involving actual electricity as Westerners would un-
derstand it but summoning the power which is hidden in the ‘real reality’ behind
cables and bulbs, electrified voices do work well.
Conclusion: Why Electrified Voices do not Work
When I had recently started fieldworking I was surprised to encounter only very
few difficulties when asking indigenous people to record their songs. Contrary to
all my expectations, people were eager to have their voices recorded, even when
performing presumably secret magic songs, in explorative settings as well as
during current rituals. When I returned to the singers and their communities
with tapes and edited CDs, they were highly appreciated and listened to some-
times enthusiastically, if possible via the village’s loudspeaker.
I noticed that among the Shipibo, a broad category of song topics could be
called itiki ik (sung about singing). Many songs about other specific topics were
preceded or concluded with some phrases of itiki ik when performed in an
explorative setting in front of my microphone. I was particularly surprised that
this was not done among inexperienced singers, but mainly by renowned and
highly experienced ones. It seems that they cared very much about what hap-
pened to their voice:
makinanin biboxon je The machine receives it,
enkaya bibokin (like) I receive it myself,
jana rebon bibokin je (like I) receive it at the tip of the tongue,
pino jana rebonbi at the tip of the hummingbird’s tongue.
makexonbokin (In the same way as I) make it resound well
yamin ewan makekin the great metal makes it resound,
jakaya boai (the metal) carries (the voice) itself away
jawen jeman boai it carries (the voice) to his home country
abaanon yoikin and there, they will talk about
joi jakon bixonki je receiving this very good voice/song.26
26 Excerpt from an itiki ik song (lines 76 – 85) performed by Roberto Dvila Barbarn at
Sinuya river, 2005. Signature number D 5461.
A Medium of Magical Power 397
I translated biti as “receive” in the whole passage in order to underline the
parallel process marked by the singer who was also a healer, when choosing the
same word for describing how he receives songs from non-human beings during
sam ‘diets’ and rituals and how my DAT-recorder received his own very good
song. He also is concerned about the song’s fate : he predicts that after carrying it
to my home country, me and my fellow ethnomusicologists will talk about how I
received the song, which proved entirely correct. Predictions ranged from such
easily imaginable situations to the imagination that I will earn a lot of “round
brilliant metal” (which was not yet fulfilled), and also addressed some more
precise details of my fieldwork’s effects :
ramakaya ixonban Right now I am going to sing
non joi bixonki and our voice/song (the researcher) received
jaribi banexon (he will also) leave here (with us),
chin aniaibo (so that) the youngest ones,
chin aniaibaonki the youngest ones while growing
jakaya ninkanon can listen to this very (voice/song)27.
However, although the machine ‘receives’ the voice or song, and it’s re-
production may serve for research presentations, earning some coins, or edu-
cating children, recorded magical songs do not work. Why?
I have shown in this chapter, that profound analysis of indigenous singing
styles, singing contexts, and song lyrics can shed new light on concepts known in
anthropology (animism, mimesis, perspectivism). The anthropological theories
in general rely much on analyzing mythology and other narratives, on kinship
relations, warfare, eating and being eaten, and most of all on visual descriptions
and clues. Therefore, techniques like ‘existing charged with power’ (speaking),
‘using the spirits’ language’ (singing), or ‘establishing performative ontological
connections’ (naming in song) have hitherto been treated peripherically at most.
But analyzing the audible helps us to encounter elaborate models (like yointi)
which allow for magical action and manipulation.
I repeat what I said in the beginning: the term ‘magical’ I have to apply
because there is no scientific measure available that could grasp the sub-
stantiality of voice. For indigenous Shipibo people, this does not pose a problem,
and they do not perceive their voices’ effect as magical, but as something logi-
cally embedded within their ontology. Vice versa, for example, for Westerners it
is logical that substances contained in certain medicaments are received like keys
in a lock by enzymes produced in the human body, and therefore are effective—a
27 Excerpt from an itiki ik song (lines 22 – 27) performed by Antonia Ahuanari Medina in
Paococha, 2004. Signature number D 5230.
Bernd Brabec de Mori398
concept absolutely absurd for indigenous people who therefore perceive the
effectiveness of these medicaments as something ‘magical’.
A portrait photograph—this is even fairly obvious to Westerners—does not
contain the whole physicality of the depicted person. In the same way, a re-
cording does not contain the substantiality of sounds, or of voices. The recording
may carry the voice away and conserve it for the younger generation, but it is an
image of the voice, like a photograph. In both cases, also the image carries
meaning and a certain magical potentiality implicit to the mimetic faculty of
image-producing machinery28. The image or recording would be again a tool for
magical actions or manipulations. Often, for instance in love magic, a photo-
graph of the one to conquer is positioned on an altar, for example, or soaked in a
prepared perfume, or whatsoever. Anyway, it is a mimetic object but not an
agent. A recorded voice is definitely not apt to reproduce the original magical
effects of koxonti,yointi or naikiti by being played. When considering these
techniques, it should be clear by now why recordings of magical songs do not
work properly. The koxonti recording does not simultaneously think the proper
lyrics when played. A CD player’s existence is not ‘heightened’ when singing. The
singer has to be a person (human or non-human) to be able to musically socialize
with non-humans whose properties are then transmitted. When a CD player
reproduces nasal falsetto sound, no listener will mistake the machine with a
shape-shifted healer who embodies a tree-person apt to cure an illness. Like the
physical body escapes from being captured by a camera’s photon ingestion, the
substance of voice which is responsible for its magical potential is not by any
means affected by a microphone’s registration of periodic air pressure change.
Maybe, this substantiality of voices—and of sound in general—hitherto
unacknowledged in naturalism, results at the one hand in effective techniques of
voice-based magic in indigenous societies, and similarly, on the other hand, in
this great and well known difference between a live concert and a CD recording
in Western societies. Although a recording can be touching and emotionally
moving, it does not work properly, as music in live performance works.
References
Austin, J. L. (1962). How to Do Things With Words. The William James Lectures delivered at
Harvard University in 1955. London: Oxford University Press at the Clarendon Press.
Brabec de Mori, B. (2007). The Inka’s Song Emanates from my Tongue. Composition vs.
Oral Tradition in Western Amazonian Curing Songs, paper presented at the 39th World
28 This is brilliantly shown by Michael Taussig (1993) with many references to the philosophy of
Walter Benjamin.
A Medium of Magical Power 399
Conference of the International Council for Traditional Music (ICTM), Vienna, July
2007.
Brabec de Mori, B. (2009). Words Can Doom. Songs May Heal. Ethnomusicological and
Indigenous Explanations of Song-Induced Transformative Processes in Western Am-
azonia. Curare. Journal for Medical Anthropology 32 (1+2): 123 – 144.
Brabec de Mori, B. (2011a). The Magic of Song, the Invention of Tradition and the
Structuring of Time among the Shipibo (Peruvian Amazon). Jahrbuch des Phono-
grammarchivs der Österreichischen Akademie der Wissenschaften 2: 169 – 192.
Brabec de Mori, B. (2011b). Die Lieder der Richtigen Menschen. Musikalische Kulturan-
thropologie der indigenen Bevölkerung im Ucayali-Tal, Westamazonien, doctoral thesis
in ethnomusicology, University of Vienna.
Brabec de Mori, B. (forthcoming). About Magical Singing, Sonic Perspectives, Ambient
Multinatures, and the Conscious Experience. Indiana 29.
Brabec de Mori, B. &Mori Silvano de Brabec, L. (2009). La corona de la inspiracin. Los
diseÇos geomtricos de los Shipibo-Konibo y sus relaciones con cosmovisin y mfflsica.
Indiana 26: 105 – 134.
Classen, C. (1990). Sweet Colors, Fragrant Songs. Sensory Models of the Andes and the
Amazon. American Ethnologist 17 (4): 722 – 735.
Descola, P. (2005). Par-del nature et culture. Paris: Gallimard.
Halbmayer, E. (2010). Kosmos und Kommunikation. Weltkonzeptionen in der südamer-
ikanischen Sprachfamilie der Cariben. Vienna: Facultas.
Halbmayer, E. (forthcoming). Amerindian Mereology. Parts, Wholes, and the Multiverse.
Indiana 29.
Hill, J. D. (1993). Keepers of the Sacred Chants. The Poetics of Ritual Power in an Ama-
zonian Society. Tucson: The University of Arizona Press.
Illius, B. (1997). Ein Lied zur Haarschneidezeremonie der Shipibo-Conibo. In: Reli-
gionsethnologische Beiträge zur Amerikanistik, edited by E. Dürr &S. Seitz. Münster:
LIT, pp. 211 – 231.
Illius, B. (1999). Das Shipibo. Texte, Kontexte, Kommentare. Ein Beitrag zur dis-
kursorientierten Untersuchung einer MontaÇa-Kultur. Berlin: Dietrich Reimer Verlag.
Labate, B. C. &W. S. Araffljo (eds.). (2002/2004). O uso ritual da ayahuasca. Campinas/Sao
Paolo: Mercado de Letras.
Langacker, R. W. (1999). Grammar and Conceptualization. Berlin: Mouton de Gruyter.
Lewy, M. (forthcoming). Different ‘Seeing’—Similar ‘Hearing’. Ritual and Sound among
the Pemn (Gran Sabana/Venezuela). Indiana 29.
de Menezes Bastos, R. J. (1986). Mfflsica, Cultura e Sociedade no Alto-Xingu: A Teoria
Musical dos Indios Kamayura. Latin American Music Review 7 (1): 51 – 80.
Olsen, D. A. (1996). Music of the Warao of Venezuela. Song People of the Rain Forest.
Gainesville: University Press of Florida.
Santos-Granero, F. (2006). Sensual Vitalities. Noncorporeal Modes of Sensing and
Knowing in Native Amazonia. Tipit. Journal of the Society for the Anthropology of
Lowland South America 4(1+2): 57 – 80.
Seeger, A. (1987). Why Suy Sing. A Musical Anthropology of an Amazonian People.
Cambridge: Cambridge University Press.
Seeger, A., R. DaMatta &E. Viveiros de Castro. (1979). A Construżo da Pessoa nas
Sociedades Indgenas Brasileiras. Boletim do Museu Nacional, N. S. 32 : 2 – 19.
Bernd Brabec de Mori400
Tambiah, S. J. (1990). Magic, Science, Religion, and the Scope Ofrationality. (The Lewis
Henry Morgan lectures 1984). New York: Cambridge University Press.
Taussig, M. (1993). Mimesis and Alterity. A Particular History of the Senses. London:
Routledge.
Tournon, J. (2002). La merma magica. Vida e historia de los Shipibo-Conibo del Ucayali.
Lima: CAAAP.
Valenzuela, P. M. (2010). Ergativity in Shipibo-Konibo, a Panoan Language of the Ucayali.
In: Ergativity in Amazonia, edited by S. Gildea &F. Queixals. Amsterdam: Benjamins :
pp. 65 – 96.
Viveiros de Castro, E. (1998). Cosmological Deixis and Amerindian Perspectivism. Journal
of the Royal Anthropological Institute (N. S.) 4 (3): 469 – 488.
Viveiros de Castro, E. (2010). The Forest of Mirrors. A few Notes on the Ontology of
Amazonian Spirits: http://amazone.wikia.com/wiki/The_Forest_of_Mirrors (accessed
on September 23rd, 2010). Originally published 2007 in Portuguese in Cadernos de
Campo (USP) 14+15 : 319 – 338.
A Medium of Magical Power 401
