Back to article list

December 29, 2003



Since the inception of systematic ethnomusicological research in the Philippines during the 1950s, a number of studies on the music of indigenous Filipinos have been published. Most are descriptive, but they have dealt almost exclusively with musical instruments, listing genres that are practiced or actively remembered by ethnolinguistic groups. In general, the study of songs lags behind the numerous works in instrumental music.1 As stated in the book Gongs and Bamboo (1998) written by the pioneer of modern ethnomusicological scholarship in the country, José Maceda, the gap in understanding indigenous Philippine vocal musics is immediately noticeable. Maceda ironically states that “vocal music is a very important part of the music cultures of both Northern and Southern Philippines which will not be fully discussed in this book devoted principally to music instruments” (viii, italics mine). The glaring exclusion and absence of such an important type of cultural expression therefore behooves us to study vocal music in the Philippines. As I shall demonstrate in this paper, song is a speech of the body, an inquiry of which can give us rare insights into salient local notions of self and personhood.

Filling in a part of our knowledge of Philippine musics, this paper shall investigate an indigenous song still heard in middle Agusan Valley.2 This is a solo, improvised type that descendants of the Agusan Valley’s original inhabitants, the Agusanen Manobos, call ted-em.3 The paper will discuss the performance context of the genre and the ideology of song vis-à-vis mimetic theory that will explain why Manobo voice in song is ambiguous (i.e., why voice cannot be so precisely located within or outside the singer’s body). Hence, rather than presenting formal aspects of the song,4 the paper will investigate song from a more encompassing stance that will be concerned with meanings. In particular, it will examine how song relates to the metaphysics of presence, self, and other.5

Stylistic features of Agusanen Manobo song

Manobo ted-em is a marked genre of speaking. It is a “high” ritual speech that, as Manobos explicitly verbalize, possesses melic or “singable” (dandanen) qualities. This style endows song with a certain degree of “ceremonial” formality and seriousness that metacommunicatively frames the genre for a special speaking event. Furthermore, this style is not evident in less marked, “light-hearted,” and “entertaining” types such as sugilen (spoken story), buwa-buwa (metered indigenous song), sura-sura (metered indigenous song accompanied by hand gestures and dance rhythms played on the bamboo “parallel zither”), and kanta (acculturated song with either Cebuano, Tagalog, or English text).

Structurally speaking, Manobo ted-em song is sung to a text with phonic, syntactic, and semantic parallelisms (Buenconsejo 1993). In performance, a singer repeats verse lines to form couplets, the second line of which morphophonemically alters or “doubles” (daduwa ne kalitokan) the first. Verses are “artfully” interpolated with rhetorical particles, a process similar to that which Elkins (1968) had found out with regards to the particle /s/ or /es/ of the “poetic” Western Bukidnon language. Indeed, the alterity of ted-em texts makes the genre different from other speech types. Some of the latter also exhibit versified text forms, but these are used in speaking events that do not place emphasis on sentiment during interpersonal meetings, the context in which ted-em is sung. Instead, other non-ted-em speech genres are associated with the settlement of debts and obligations and hence involve “political” rhetorical speech such as bisaya (marriage negotiations and advice) and tampuda (arbitration).

Singers extemporaneously set their improvised verses to set tunes (tonada). These are unique to each of them.6 Beyond the easily recognizable couplet structure, singers create “speech paragraphs” out of the verses. Musically-foregrounded sections called adudungan, i.e., “places in song where the sentiment of pity is expressed,” separate the speech paragraphs from each other. Via these musical sections—the felt adudungan—sentiments that are difficult to express are conveyed.

The emotions felt by singers in song result from, at the moment of singing, their being in front of their others, i.e., as they face the guests. This is the fundamental meaning of the act in song, which obviously points to the context of self-other encounter. A singer literally “confronts” her or his listeners/interlocutors to display a communicative routine called pamalibad (taking back one’s words). This act almost always forms the basis of song performance, as we shall see. Suffice it to say at this point that successful song performances normatively regenerate social ties and cordial relations among singers and audiences. During a “sacred” curing ritual séance, for example, the patient’s family plays the role of guest. They are spoken to by the possessed medium using the voice of her or his spirit helpers called baylan. In a parallel context, these guests are also played by human visitors who are received by their kin during ubiquitous “secular” interpersonal meetings realized in everyday social life.7

Manobos say that song is a form of disclosing (nawnangen) the singer’s thought (ginhawa). At the metalingual level, it is a reflexive act that points to the immediate interpersonal encounter in which such song is enacted. In ritual séance, a possessed medium sings to reveal her or his thoughts or visions that only s/he can see to ritual participants. S/he describes images of movements in which “seen” beings cross the boundaries of self and other, a vision of human souls and detachable spirits crossing the metaphysical divide of life and death. Ritual audience would inject particles to express pity in response to hearing the images that song offers.

In “secular” context, personal songs likewise embody the act of intimacy. They state messages of hospitality, respect for the katenged (i.e., human right or the person in front of the singer), politeness, friendship, and so on. Messages contained in personal songs perform the act by which the singer’s “inner self” (be-et) speaks to another inner self, i.e., that of one’s interlocutor. Like messages of spirit song which embody presence and absence (i.e., images of spirits and humans moving across the cosmos), singing in secular context is always already a part of that image—a human being connected to another one via song.

While the distinction between the voice of the “spirit” and “secular/personal” is clearly perceived by the Manobos at the discursive level, Manobo voice in song, in actuality, is ambiguous. Personal voice is spirit voice itself. It should be noted that there exists no formal textual and musical stylistic difference in using one voice from the other voice. This ambiguity strongly indicates that song is a mimetic language that is meaningful in self-other encounters. The fusion enables Manobo singers to efface borders. In song, the origin or “place” of speaking is neither outside nor inside the singer, a matter evident in why singers were reluctant to sing the personal songs that I, nonetheless, recorded in the field in 1991, 1993, 1996, and 1997. Singers rationalized their fear by saying that spirit familiars would possess them the moment they used their voices. As documented in audio recordings, some singers, in fact, went into a state of spirit possession (yana-an) after a few seconds of singing ted-em in what were perceived to be personal voices in the beginning. No doubt, this indicates that song can breach or transgress compartment-alization of voices between personal/spirit, even rendering the categories “secular/sacred” inoperative.

Moreover, the either/or of voice confirms why the Manobo cannot verbalize where the fast vision spirit (daligmata) resides. This is a spirit associated with clairvoyance who lives in any place where the double of that medium resides.8 The fast vision is a singing spirit who is said to reveal or “prophesize” things for no apparent reason to the person owning that spirit during sleep. It is this mysterious place of song, the speech of the fast vision spirit, which distinguishes poetic song from prosaic speech of everyday life. In the latter, the parameter of intonation (tonada) enables a listener to pinpoint the geographic origin of a speaker (cf. Mirano). Conversely, in song, the origin of voice is enigmatic, a feature that suggests the relationship between voice and performance. In what follows, I will elaborate on the notion of this mystery, delving more into the problematic of the “difference-then-fusion” of self and other. I will also suggest that this problem—the ambiguity in the persona of the singer during moments of singing—has to do with utilizing ted-em voice to perform authoritative power or legitimacy. Yet, because that voice/power is itself ambiguous, voice can be construed as a “double,” creating an effect that potentially subverts that very same authority and power. To understand this, we need to know what the concept of self is for the Manobos.

The Manobo self and its components

Within the self:
breath/life (ginhawa)
and soul (umagad)

Unlike the concept of the atomistic self that is corollary to “modern” and historically naturalized material practices, Manobo self is ostensibly sociocentric. Instead of a monolithic bounded self, Manobos perceive self to consist of two components: the breath or the vital principle, which is located inside the body, and the double, which is located outside of it. Like most durable Filipino cultural constructions of a self, Manobos call the “essence” associated with the “inner” person ginhawa. This is the inherent, vital force of a person, the quality of which makes up the person’s charisma (bantog) or power (gahem) (cf. Rosaldo 1980). This ginhawa has a far deeper meaning than simply “physical breath.” It is an integrated and multi-dimensional concept that includes notions that are physiological, emotional, agentive, and psychological. Moreover, Manobos call that component of the self that is believed to inherently reside outside of the person kadengan-dengan or double. When this entity desires to possess humans (kiham), the spirit is given a human name—thus transforming the outside nature of the spirit into near-human. The spirit becomes the baylan (spirit helper) who dances, speaks, and sings during séances.9

Table 1 (see Appendix A) is a list of a few of the representative contexts of enunciation where the words ginhawa and umagad were used. Both of these components are associated with the inner person. These statements are culled from actual recordings of myths, rituals, and songs performed in situ during fieldwork. The table shows the complexity of the Manobo concept of ginhawa and the comprehensiveness of its semantic field. For this reason, it is heard very often in everyday discourse and during rituals that reflexively look into it. Symptoms of illness, for example, are couched in terms of the state of the place where the ginhawa circulates (ginhawa-an).10 Illness etiology is deciphered through the malfunctioning of the breath inside the body. The coordinated working of ginhawa-an is analogous to the synchronized machine parts; for the whole to function, each of the parts is needed.11 The circulation of the ginhawa has been likened to walking, an image that spirits refer to as they participate in rituals. Manobos read the state of each other’s ginhawa through the face (kiley) (cf. Covar 1988). Upon seeing the face of another, they almost always express endearment with the formulaic particle, “I pity.”

The word ginhawa, however, seems to have often been misconstrued as soul (umagad). Data gathered during fieldwork, however, show that this is the form that the ginhawa takes only when it gets separated from the inner part of the physical body during times of individual sicknesses, dreams, and, of course, death. All of these phenomena are understood as movements of the ginhawa or consciousness outside or away from the body. They are perceived to be temporary, or in the case of death, permanent soul loss.12 Therefore, the words ginhawa and umagad do not refer to entirely different spiritual “substances” inside the person. Both terms refer to the same “substance,” the terminological difference indicating the particular state of the “substance” at a given moment. Ginhawa refers to the “substance” (i.e., breath ) in its normal waking/living state, while umagad refers to the same “substance” when this enters into or is in a different realm of the cosmos. The following statements uttered by participants during actual ritual séances in Table 2 (see Appendix A) will clarify this proposition.

Outside the person:

the kadengan-dengan (double)

and the baylan (spirit familiar)

While the concept of ginhawa refers to that component located inside the person’s body, kadengan-dengan or double refers to an entity that is perceived to be naturally located outside of the person’s body.13 Like the ginhawa which can move away from the inner person—i.e., lost or transformed into a state of absence or death—the double can also move, although this time in a reverse direction, toward the inner person. The double can be made present or incarnated inside a body, a belief that forms the basis of spirit- possession.14 Indeed, spirit-possession is a mimetic process in which human society draws affinity with its external reality or nature.15

Like the illness that has been taken to mean the loss of the vitality of a person’s ginhawa, the disengagement of the spirit-double or an inherently detachable spirit from nature as it desires to alight or “clothe” persons also makes the hosts’ bodies ill.16 The encroaching spirit from the environment is believed to gradually “eat up” the host. The person becomes ill and rituals are held to counter the harmful spirit force. A chicken or a pig is symbolically killed for this purpose.

In most cases, the type of spirit that wishes to eat up the human body is the double of the ill person. In a few cases, however, it is a detachable spirit of nature. A person and its double are born at the same time. The double can begin to attack the person’s body even at a very early stage. An irritable baby, for example, is said to be bothered by its double. The double is believed to “tease” the baby. This can lead to illnesses. To offer a chicken sacrifice is to recognize the double and its owner, who live in the skyworld. Manobos say that such an act protects the baby from further harm because an animal has been substituted for the baby.17

Almost all Manobos experience various kinds of spirit attack intermittently throughout their lives.18 If a particular person is persistently or severely attacked, a senior medium holds a series of divinatory rituals. These will negotiate the terms of the attacking spirit/s. Spirit attacks often result in near-death experiences of the persons concerned or members of their families. Human beings respond by slaughtering animal symbolic substitutes.19

Like the suggestion that the ginhawa and umagad are similar in substance, the double and the spirit helper resemble each other as well. As I have mentioned, ginhawa and double primarily contrast in state with umagad and spirit helper (i.e., the latter two are transformed states). Umagad is the ginhawa when it is dead or is outside of its embodied, living state. Similarly, the double is an outsider presence that will become a spirit-helper when it is inside or has already been acting the role of the familiar to the person.20 What moves toward death in one is a move toward presence or life in the other. This in-and-out movement is a fundamental physical fact. Out of this, a metaphysical discourse has been built.

When spirit doubles cross the human world or human breaths “escape” from bodies, the metaphysical ontological divide between life and death is breached. In both instances, the medial language of song is articulated. It is the language expressed at the gaps between two states, a matter evident in how a sick person’s soul (umagad) is recovered through the all-night ritual singing (gud-guden). Moreover, persons believed to have been persistently attacked by their doubles and who are undergoing the series of divinatory rituals will speak through song only during the final moment of his or her “initiation.” This takes place when the spirit double of that medium-to-be has been fully negotiated or when the ill-person-turned-medium begins to heal. Throughout these initiatory rites, the medium-to-be gradually acquires the rich lore behind Manobo rituals. In the early stages of the series of divinatory rituals, the medium only learns to speak the language of dance. This represents spirits as incarnated in the medium’s body or ritual space. It is only toward the end of the divinatory cycle that the medium sings. This belief is interesting because it suggests that song has to do with time, during which the crossing from death to life is about to begin or when the ontological gaps are about to be crossed. Song is therefore an expression of becoming. It is sung by an ill person whose death has now been transformed into one of health and presence. How is this metaphysical crossing materialized at the concrete level of empirical Manobo social life?

Nawnangen, pamalibad, and the fusion of self and other

The notion of crossing is a significant metaphor used in both Manobo ritual and everyday worlds. It is the word that Manobo spirits—speaking through their corporeal counterparts—use to describe their sensation as they participate with human beings during rituals. At the empirical level, crossing also means the actual meeting of people from different households who act as hosts and guests during rituals. The medium’s families and ritual sponsors perform the roles of hosts and guests or vice versa (depending on the local circumstances). In this ritual space, conversations among households occur and there is an intimation of thoughts or ginhawa. This speech of intimacy, nawnangen, in its ordinary usage, is a way of “revealing a secret.” Obviously, the act creates interpersonal connectedness between speakers and listeners.21

Manobo personal songs would also have the same effects of engagement and intimacy. Ted-em texts do not have “stories” per se because they reflexively or metonymically account for immediate interpersonal encounters in which those texts emerge. In Appendix B are found the texts of two such songs sung to the author in 1997. In order to situate the meaning of these songs, I would describe first what a pamalibad is because it is a corollary of the nawnangen style of speaking.

Pamalibad means “to take back what one said.” It is a face-to-face confrontation that usually attempts to address a misunderstanding that has transpired between persons through song. Literally, the term means refusal, but since outright rejection of another’s request seems impolite, the person who has been put in a compromising situation sings his or her real feelings instead. This is meant to clarify one’s position, as exemplified by Edgardo Dumalinao’s song (see Appendix B).22 This was sung for the author on the veranda of his house on December 15, 1996, after he failed to keep a promise he made to me. The following detailed account is given to illustrate vividly that the “fusion of self-other” is concrete or immediate.

Edgardo Dumalinao is a male medium who lives in Macopahen, Barrio San Vicente, Loreto. I interviewed him a couple of times and recorded six of his songs in 1991. I met him again in October 1996 to negotiate documenting a ritual to commemorate the completion of his new house.

Edgardo was originally reluctant to hold the ritual, since he was still mourning the death of his oldest son, who had died earlier that year. He believed his son had been poisoned by the Visayan parents of his daughter-in-law. However, as our conversation progressed, Edgardo told me he would build a new house away from the former one, in a forest clearing which he still owned.23 He said that there would most likely be a ritual if I could find a gong, a ritual musical instrument he did not own. I promised to give him one in December, and this I did, but to my dismay, Edgardo’s house was still unfinished then. Instead of holding a ritual for the inauguration of his new house, he invited me and my fieldwork guide to his old one on December 15 to celebrate a ritual to honor his daughter’s marriage to a Visayan furniture factory worker from the neighboring barrio. This marriage had been arranged by Edgardo and the Manobo boss of the furniture worker, who was supposed to provide the pig for the ritual sacrifice.

When we arrived at his house on the set date, we were once again disappointed. No wedding feast was held. We were exasperated, since we had had a difficult time bringing the drinks, candles, cigarettes, and many other things for the ritual to Edgardo’s house, which was more than a mile away from the town road. Since he was put in a compromising situation, Edgardo promised to hold another ritual. At this point, the ritual seemed more for my needs than his family’s.

That afternoon, Edgardo thought of going to the furniture worker’s boss to get the promised pig. He voluntarily sang the song before he left his house to get the pig so that the ritual could proceed. It was already dark when Edgardo returned to his house, where we waited patiently. He had failed again and explained that the furniture factory owner did not have paper money but a personal check. That evening, he sang another song (not included in the appendix) expressing his difficult situation.

In the song included here, Edgardo tried to compensate for the inconvenience he had caused us, which as mentioned above, is in the context of pamalibad, i.e., taking back one’s own words that have created confusion or breached proper social relations. In the song, Edgardo described his relationship with the song event. He began by saying how long he had been “conveying the difficulty,” how he had failed to comply with my request (metaphorically expressed as a “journey”), and how I had been wishing to witness a ritual performed by him. He seemed to complain about our conversation in October, and the many more meetings which came after that, each time to negotiate the time of the ritual. He stated that we wanted to make the ritual more festive and elaborate each time. He then finished his song, reporting that he was ashamed because he was not able to fulfill the promise (his house was not yet finished, his daughter’s marriage had failed, and he had no pig to honor his spirit helpers). In effect, these events served to degrade his person, because he failed to fulfill the promise he had made in exchange for the gong I gave him.

Song, power, and authority

In the anecdote above, Edgardo uses song to express what is difficult to convey in plain speech. By resorting to this special language form, Edgardo uses the voice that, because it is associated with no specific place, authorizes the presentation of his message. Not all Manobos know how to sing the ted-em. As already mentioned, it is a form of language that is perceived to have no fixed place of origin in Manobo cosmos that persons can hear, nevertheless, from within their own selves. To be able to use a language that is outside of the self is to unleash the power of that language via performance. Through its use, the appropriation of the outsider-language can influence the speaking event in which such language is used. This is the ability to assume an authority, a matter evident in the belief that it is only when the medium has undergone existential problems like death that a novitiate medium can sing. Song is the voice of the person’s double, yet because it does not originate from a single place, it can also be used to undermine authority at the same time. I will present another song that illustrates the final proposition of the paper. By using this song to end this work, I also wish to appropriate the voice of the Manobo singer for my own voice, i.e., to re-iterate a crucial point that I wish to make in this paper: the doubleness of song because it is performed. Indeed, song is a risk and it can frame authority as much as it can critique it.

This is the ted-em sung to me by Edna Tambis, who heard about my research one day as she went to sell her vegetables in town (see Appendix B). She agreed to meet me at the prodding of a male Manobo government official living in town. I recorded her song beside the Catholic church in the town square on March 27, 1991. Edna’s song is a “refusal” in the context of reluctance, though her very strong words went beyond the usual politeness associated with the song genre. Edna Tambis questions the intended hearer, myself, indirectly as the “person over there” (note again that this indirectness had to do with speaking to the “inner” person of the listener as discussed above). She wonders why he needs the “tune of the natives” and what the significance of “copying” such tunes is. She sings against “coercion” and “downgrading” and “pities” the tune, the spirit familiars, and her own perplexed body. These things are synecdochic of that thing called “Manobo.” This indirect questioning prompted the government official to respond to her by saying that recording Manobo songs as “souvenirs” is beneficial to them. (Note the motivation here: the official’s remark is informed by the outsider’s ideological effort of conservation and the joining of the “ethnomusicological” project with the outsider’s ideology.) This Manobo official was a town councilor at that time and head of Manobo Foundation. He had gone to various urban centers where he had seen the value of festivals, showcasing “traditional” performances, and the need for conserving Manobo artifacts as “souvenirs.”


In this paper, I discussed the Manobo notion of voice in song and attempted to situate it within a broader ideology that broaches on the topic of the Manobo sociocentric self. I discussed how this self is emplaced within its others and the various metaphysical presences in which such a self can become in the world (kalibutan). Locution in Manobo voice is not tied to a fixed “place” or origin. This proposition has to do with song being a mimetic language that clearly emerges in medial self-other encounters or during in-between, liminal processes of becoming. Persons possess the ability to sing or voice ted-em by hearing distant voices outside of themselves, but which can only emanate from their inner persons in the last instance.

Because of these in-and out-fusions, Manobo voice in song is ambiguous. There exists a rich lore surrounding song performance that indicates the in-betweenness or “neither-here-nor-there” character of song. First, it is the channel by which the lost ginhawa of an ill person is transformed to presence during an all-night ritual séance done in darkness. Second, mediums-to-be sing only during the last stage of the divinatory cycle before they are accepted by their kin to heal. Song therefore marks a passage; it lies in a juncture in which a past state begins to move into another. Third, personal songs reside not in the self’s monolithic subjectivity but in a combination of the singer’s self with others. Because of this “outsideness,” voice in song enables persons, via that exteriority, to don authority or legitimacy. However, as the last song in this paper showed, it is also that same exteriority of song—its non-fixedness of origin—that enables singers to paradoxically subvert power. Thus, it is the “doubleness” of voice in the Manobo ted-em that makes it powerful and rhetorical. In this paper, I have shown that voice resides in performance, in an act that without which texts and structures are meaningless.


1. First, this may have to do with the fact that vocal music contains texts that would demand more extensive linguistic preparation on the part of fieldworkers. Second, vocal color and production of indigenous songs—especially improvised ones sung without meter and with ornamentations during rituals—appear more difficult to represent and describe on paper using words. And finally, the materiality of songs, like mime, is deeply connected to the body; to study it would entail a more complex work on the hermenuetics of the human body as an instrument. Vocal music therefore creates more demands on ethnographers than instrumental music, since dance rhythms and melody can be easily appropriated by them for their own projects.

2. The song is now heard less frequently because of Visayan settler hegemony.

3. The following orthographic conventions are used to represent the ritual texts. The five letters (1) a, (2) æ, (3) e, (4) i and (5) u represent seven vowel phonemes of the (Agusan) Manobo, Umayam dialect, as follows: The letter æ represents the phoneme /æ / as in the English word “cat.” The letter e represents either the vowel phoneme /e/ or /i/. The letter i represents either vowel phoneme /i/ or /ü/. The sixteen consonant phonemes will be represented by the following letters and symbols: (1) b, (2) k, (3) d, (4) g, (5) h, (6) l, (7) m, (8) n, (9) ng (i.e., /h/), (10) p, (11) r,(12) s, (13) t, (14) w, (15) y, (16) j, and (17) q. The last phoneme represents the glottal stop phoneme in word final position, after either a stressed or an unstressed vowel. In mid-position, the glottal stop is represented by – , i.e., (1) between two consecutive vowels as in ka-iling and (2) between a consonant and vowel as in ted-em, udas-udas and huna-huna-on. There is no attempt to represent vowel length and stress in the orthography.

4. See, e.g., Prudente 1984.

5. For clarification, I use the word “ideology” to refer to concrete material practices that produce or reproduce what are, more or less, shared ways of doing “things” called “culture.” I will connect these practices to a local understanding of the metaphysics of life, death, spirit, and soul. However, I will address metaphysics—not in the tradition of idle philosophical thought or dis-embodied idea—but in terms of how it has become concrete and embodied as ideological practices.

6. It is not known how these melodies may form types or “families” and how these may be shared or distributed among the various Manobo groups living in Middle Agusan Valley.

7. The hundreds of personal songs that the author recorded during fieldwork (intermittently in 1991, 1993, 1996, and 1997) exemplify the latter.

8. Nonetheless, Manobo song can only be sung in Manobo, thus suggesting that it is metonymic of Manobo indigineity.

9. Etymologically speaking, umagad and baylan are affinal terms, not consanguinal. Spirits and actual persons who come outside of Manobo human society are outsiders who have breached the boundaries of “places” in which they are associated. These outsiders establish links of affinity with the insiders.

10. This refers to the upper torso of the body: the head (uyo), heart (dageha), diaphragm (giheb) and liver (atey). The significance of head and heart is evident in the blowing of the possessed medium’s breath on the fontanel and the chest of a person during the curing ritual. The liver is an important site because it is where the breath is supposed to walk.

11. Undesirable cold air is understood as being trapped inside the body, producing numbness, pale complexion, and what is commonly called “cold sweat.” This constricted circulation of the ginhawa is often attributed to the improper interaction of the humors inside the permeable body with the hot and cold elements of the weather and the environment. This experience of constricted ginhawa draws a parallel to living in cramped places in which Manobos have to live side by side with Visayan settlers. The image of the porousness of the human body, which contains the sociable or sociocentric person, is also the image of a house made of light, biodegradable materials never intended to be sealed for permanency and privacy. Thus even though some Manobos make statements about personal states using specific terms such as thought (hena-hena), will (be-et), or emotion (Ceb. pagbati), one should bear in mind that any of these components may refer to the undifferentiated inner component or ginhawa of the person.

12. Life historical narratives about the Manobo experience of souls that I have documented suggest that when the ginhawa becomes umagad, the rationality of the living being is lost; what is left of the umagad is the dimension of emotion. This is consistent with the general Manobo perception that umagad of dead relatives can always harm the living and, like spirits, desire to take the living with them.

13. Elsewhere, this is called the “twin-soul.”

14. Those of us who grew up as Christians may find this distaseful, but I suggest that we suspend disbelief in order to appreciate what I will say about this phenomenon.

15. The Manobo belief that outsider doubles have an effect on the inner person is analogous to the belief in snake twins in other parts of the Philippines in which a snake and a human being are paired, each resembling the life of the other (cf. Rudolph 1988).

16. The act of spirit transgression has been called by other scholars “spirit attacks.”

17. I have shown in my dissertation (1999) that the sacrifice also “marks” the baby whose self has been transformed to one capable of participating in symbolic exchange between beings of the world. The child therefore becomes a sociable or “cultured” being.

18. A classification of these spirit attacks goes beyond the scope of this present paper.

19. There is a huge body of literature in comparative religion on this subject.

20. Spirit helpers have human names. To have a name is to possess its attributes: near-human characteristics. Spirits can be spoken to and propitiated with rituals gifts that counteract their malevolent power. It is the capacity to be spoken to that permits spirits to likewise speak to ritual participants. The capacity enables the spirits and humans to cross each other and negotiate the distance between them.

21. In the context of storytelling, nawnangen is a metacommunicative device that storytellers use to key their performance. In this device the storyteller keys his engagement with the story he is narrating. In one of my documentaries of these full performances of stories, a storyteller presented the dialogue of the characters he was narrating during moments when characters were in very pitiable situations. Instead of using indirect speech, the narrator assumed the speech of the character concerned, as if that speech were the narrator’s words. In effect, the indirectness of speech was gone; the words of the characters became the words of the narrator and the roles of characters and narrator became one. What is revealed as some kind of secret here is not necessarily a message but the form of engagement that narrators make towards their stories.

22. True name withheld to protect the research associate.

23. Manobos generally abandon their houses when a family member has died in it.

Works Cited

Buenconsejo, José. 1999. Songs and Gifts at the Colonial Frontier: The Aesthetics of Agusan Manobo Spirit Possession Ritual, Philippines. Ph.D. diss., University of Pennsylvania.

——————. The Ted-em among the (Agusan) Manobo, Mindanao Island, Southern Philippines: Musical and Textual Characteristics of a Discourse Sung in In- and Out-Group Context. Master’s Thesis, University of Hawaii (Manoa).

Covar, Prospero. 1988. “Faces.” In Three Essays. Manila: Cultural Center of the Philippines.

Elkins, Richard E. 1968. Manobo-English dictionary. Oceanic linguistics special publication. no. 3. Honolulu: University of Hawaii Press.

Maceda, José. 1998. Gongs and Bamboo: A Panorama of Philippine Music Instruments. Quezon City: University of the Philippines Press.

Mirano, Ma. Elena. “In Search of a Lost Kingdom: the Music Traditions of Batangas.” TS.

Prudente, Felicidad. 1984. Musical process in the gasumbi epic of the buwaya Kalingga people of Northern Philippines. Ph.D. diss., University of Michigan.

Rosaldo, Michelle. 1980. Knowledge and passion: Ilongot notions of self and social life. Cambridge: Cambridge University Press.

Rudolph, Ebermut. 1988. “Snake-twins of the Philippines: observations on the alter-ego complex: second and last installment.” Philippine Quarterly of Culture and Society 16.3-4: 250-280.