See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/339165696 “Talking Jew’s Harp” and Its Relation to Vowel Harmony as a Paradigm of Formative Influence of Music on Language Chapter · February 2020 DOI: 10.1007/978-981-15-4250-3_8 CITATIONS READS 0 800 1 author: Aleksey Nikolsky Braavo Enterprises 56 PUBLICATIONS   60 CITATIONS    SEE PROFILE Some of the authors of this publication are also working on these related projects: Tonal Organization of Music as a Reflection of Cognitive Development View project Timbre-oriented music View project All content following this page was uploaded by Aleksey Nikolsky on 11 February 2020. The user has requested enhancement of the downloaded file. 1 "Talking Jew's Harp" and its relation to vowel harmony as a paradigm of formative 2 influence of music on language 3 Abstract 4 A very popular, yet barely researched, musical instrument is the Jew's Harp (aka Jaw Harp, JH). Its 5 earliest archaeological occurrences date back to the early Bronze Age, but its simplest constructions, made 6 of tree twigs and bark, along with its cross-cultural connection to shamanic beliefs, suggest its prehistoric 7 use. Archaeological evidence points to Northeast China and the Amur basin as the source of the JH early 8 dissemination (B.C.E.) over a vast area, from the Volga steppes to Japan, as being an important musical 9 instrument in the traditional musical culture. For some indigenous ethnicities such as the Yakuts, this is 10 the only non-percussive musical instrument. Its uniqueness is manifested in the peculiar tradition of 11 articulating speech-like sounds. The importance of this tradition is evident by its enormous geographic 12 span from Western Europe to Melanesia. Its use to camouflage romantic communication between a young 13 male and a young female is especially common. Little known is the similarity between the vocal system 14 of articulations in the "talking JH" tradition and vowel harmony found in most languages of the Turkic, 15 Mongolic, and Tungusic families of the Transeurasian (aka "Altaic") family, as well as in many Uralic 16 languages - all of which are spoken by peoples that regularly use the JH. This paper outlines a possible 17 scenario in which the spread of the cult of ancestral plants across this vast region, from Altai to Sakhalin, 18 and the cult of the "singing mask" of the Tuva-Amur area may have given special importance to musicking 19 on the JH, initiating its spread along pastoralism to the neighboring regions. Once established, the JH 20 tradition may have bifurcated into two types: the framed idioglot, usually made of organic materials, that 21 was sustained in north and northeast, and the bow-shaped heteroglot metallic type that spread to west and 22 southwest. The ethnicities that preferred the framed JH construction retained the JH as their sole (or one 23 of the very few) musical instruments within their timbre-oriented music culture. The ethnicities that 24 adopted forms of frequency-oriented music developed a rich assortment of musical instruments which 25 transformed their JH traditions and reduced the importance of the JH in their music cultures. It is JH's 26 unique status as a primary traditional instrument that may have granted it formative influence over the 27 existing languages of the Transeurasian family and the neighboring Uralic family. 28 Keywords: Jew’s harp, protomusic and protolanguage, musilanguage, tonal organization, “timbre- 29 oriented” music, “timbre-class,” vowel harmony, agglutination, morphonology, Transeurasian family, 30 “Altaic” languages. 31 "Talking Jew's Harp" and its relation to vowel harmony as a paradigm of formative influence of music on 32 language ...................................................................................................................................................... 1 33 1. Current views on emergence of language and music .......................................................................................... 2 34 2. Jew’s Harp and its sound production ................................................................................................................... 4 35 3. Jew’s Harp as the world’s earliest vocoder and its cultural ramifications ........................................................... 8 36 4. The repertory of JH’s sounds ............................................................................................................................. 10 37 5. Modal nature of tonal organization of JH vocalizations .................................................................................... 16 38 6. Where and when did Jew’s Harp emerged as a principal instrument ............................................................... 18 39 7. JH tradition and genesis of languages in Northern Eurasia ............................................................................... 29 40 8. Nivkh language’s ties to the emergence of Jew’s Harp’s Eurasian tradition ..................................................... 33 41 9. Jew’s Harp and the origin of vowel harmony .................................................................................................... 42 42 10. Conclusion ..................................................................................................................................................... 50 43 Language and music have been seen by generations of scholars as semiotic systems unique to human 44 species. Recent research, however, has questioned this view. Not only the delineation between animal 45 communication and human language/music became blurred out, but the traditional opposition between 46 “referentially-oriented” language and “affect-oriented” music as well. Music’s capacity to convey 47 referential information is employed in military, hunting and herding calls (Monelle 2006). Language’s 48 capacity to convey emotional information is high enough to justify it as language’s “second order” 49 function (Jensen 2014), especially important in poetic speech (Kraxenberger and Menninghaus 2016). 50 Cross-disciplinary investigation of speech, music and animal communication has led to conceptualization 51 of a single communicative system used to convey referential and affective information, parental to both, 52 music and language, and derivative of animal communication - musilanguage (Brown 2000a). Since 53 language and music, both, possess shared and domain-specific features, it would be logical to accept that 54 the shared features were the first to evolve, followed by domain-specific, turning language and music into 55 homologues (Brown 2001). This paper examines the shared features between the languages and the music 56 systems practiced in northeastern Eurasia and argues that the indigenous tradition of Jew’s Harp 57 musicking that remains vital, important, and surprisingly homogenous across many local ethnoses, 58 constitutes a common ground for both, tonal organization of music and vocal systems of languages of this 59 region. 60 1. Current views on emergence of language and music 61 The musilanguage model allows to distinguish 6 possible ways for music and language to be formed, 62 as determined by the differences in vocal learning and known ethological typology of auditory signals. 63 These 6 ways are as follows (Cross et al. 2013): 64 a) autonomy of music and language components within musilanguage and after its extinction, 65 b) their mutual origin in musilanguage, followed by their bifurcation and autonomous development, 66 c) their inseparable origin and development, 67 d) their autonomy within musilanguage, followed by their convergence, 68 e) music emerging as language’s offshoot (Spencerian view), 69 f) language emerging as music’ offshoot (Darwinian view). 70 Of these possibilities, b) (Brown 2000b), f) (Fitch 2010), and to a lesser extent, e) (Bowling 2013) find 71 support amongst modern scholars. The consensus leans towards the b) scenario (Cross and Morley 2009). 72 The recent update of the musilanguage model further elaborated the b) framework, distinguishing two 73 evolutionary stages (Brown 2017): 74 1) formation of the affective prosody from the innate conspecific animal calls, produced collectively 75 in an uncoordinated manner by a group of animals based on emotional contagion; 76 2) formation of the intonational prosody from common heterospecific traits of affective prosody, 77 produced individually and branching into “musical” and “linguistic” domains, as determined by different 78 approaches to temporal organization of frequency modulations: 79 a) musical intonation is faceted by tightening the synchronization of sustained pitch levels in collective 80 vocalization, aimed primarily at mediating the affective state of each of the participants within a group; 81 b) linguistic intonation is faceted by the alternation of vocalizations in a duetic setting in a freer timing, 82 without strict timekeeping, aimed primarily at exchange of propositional information, and only secondary 83 at manifesting speaker’s affective state and attitude towards the conveyed data. 2 84 Temporal organization here plays the formative role, directing the entire tonal organization (TO) of 85 proto-music and proto-language, including frequency, amplitude and timbre-related aspects of vocal 86 expression (e.g., pitch contour, dynamic contour, registers) towards different targets. Inherently 87 uncoordinated collective production of “affective prosody” engages a peculiar musilinguistic texture – 88 “isophony”: brief calls, continuously reproduced by multiple performers with irrational deviations in 89 timing and pitch, where each participant retains idiosyncrasy of rhythmic, timbral, and directional 90 attributes of the pitch contour (Nikolsky 2018). “Intonational prosody” imposes TO on “jumbled” 91 isophonic texture, transforming it into synchronized and definite-in-pitch progressions of musical pitch- 92 classes or asynchronous and essentially atonal indefinite-in-pitch progressions of linguistic phonemes. 93 • The paradigmatic model of “musical texture” becomes monodic choral – simultaneous production 94 of the same melody by multiple singers, expressing the same musical emotion in certain timing. Most 95 known forms of music adhere to this paradigm even in complex heterophonic, polyphonic and 96 homophonic textures. Unlike language, any music usually contains some kind of a recognizable 97 thematic material (Drabkin 2001) confined to a single melodic line (stream) and reproducible by 98 numerous performers: synchronously or solo - at some other point in some other suitable 99 circumstances (Val’kova 1992). “Musical” TO specializes in integrating the expression of multiple 100 performers throughout multiple performances, called to repeatedly reproduce the “same” affective 101 expression in redundant and long-lasting way. 102 • The paradigmatic model of “verbal texture” becomes individualized dialog – successive production 103 of different sentences by two interlocutors in loose timing, in response to each other, where prosodic 104 features similar to musical rhythm and pitch are used without incremental quantization, and where 105 interlocutors focus on informing each other of the constantly changing whereabouts. This paradigm 106 retains its formative power over the individual use of speech, e.g., in internal speech (Winsler 2009). 107 The dialogic setting promotes dialogic thinking - “ongoing interplay between different internalized 108 perspectives on reality” (Fernyhough 2009). “Verbal” TO specializes in segregating the expression 109 of few performers: “actual” in conversation or “imaginary” in internal speech. In both cases, 110 opposing viewpoints stay non-redundant, constantly updated and relatively compact. 111 Distance plays a key role in transforming musilanguage into protomusic and protolanguage. Close 112 proximity favors audio signals that are brief, diverse and rapidly changing, relying on quick processing, 113 innate reflexes, short-term memory and integration of multi-sensory data - since auditory expression is 114 often accompanied with non-auditory cues (Wallin 1991). Long-distance communication relies 115 exclusively on the auditory data, requiring long-term memory, retrieval and interpretative skills bound to 116 a particular environmental frame of reference (ibid.). Such communication forges sender/receiver 117 conventions that operate via ritualization of behavior (Wiley 1983). Conventionalization opens doors to 118 dishonest signaling that works to the advantage of the sender, thereby devaluating the acoustic markers 119 characterizing a particular call-type and causing the signaler to eventually modify the signal. The chain of 120 ritualization and devaluation goes on until a state of stationary equilibrium is reached within a sizeable 121 population, at which point the convention becomes permanently fixed (Maynard-Smith 1976). Close- 122 distance communication allows receivers to verify the auditory information by non-auditory cues and 123 detect the dishonest use of a signal. 124 Both, protomusic and protolanguage, seem to represent two different methods of identifying dishonesty 125 by means of cross-relating the acoustic markers of a signal type to the prosodic traits that convey the 126 affective state of signaler and to the framework of a “speech act” or “music act” within which a signal was 127 produced. If animal communication tends to be relatively long-range, stereotypical, with coarse 128 gradations of very few expressive parameters - depending on the acoustic properties of the environment, 3 129 human communication is mostly close-range, diverse, using incremental gradations of multiple 130 parameters and quite independent from the environmental acoustic properties (Morton 1977). 131 Semiotically, protolanguage was shaped by complexity of production and reproduction, intention to 132 communicate, sound symbolism, parity of encoding and decoding, hierarchical organization, accounting 133 for past and pre-planning future speech acts (Arbib 2012). All of these fit into the “face-to-face” dialogic 134 exchange that refers to the shared environment, supported by deictic gesticulation and mimics – such 135 communication is found in human infants as well as amongst primates (Kita 2003). 136 Protomusic differs from protolanguage in engaging gestures of another kind – not deictic but 137 locomotive: histrionic bodily expression and technical motions necessary for sound production (Arbib 138 and Iriki 2012). Hence, “music act” constitutes gestures that are inherently expressive on their own, not 139 needing real objects to which to point (Godøy 2004). Such gestures become abstracted early in infancy 140 throughout the baby’s interaction with the caretakers and form the foundation for perceiving melody, 141 harmony and rhythm (Trevarthen et al. 2011). These gestures are appreciated holistically on par with 142 “sound objects” (Nazaikinsky 1973), which requires extra time for their semantic processing, thereby 143 setting an upper limit for how fast and for how long the music can go before its intelligibility becomes 144 compromised (Nazaikinsky 1972). Gestures that accompany speech do not require such mediating 145 interpretation. They are not autonomous “objects,” but rather aids in support of faster transmission of the 146 propositional information (Lieberman 2008). 147 • The goal of verbal communication is to maximize information output, whereas musical 148 communication limits it in sake of achieving emotional contagion. 149 Protolanguage opposes language across “the divide between biology and culture” (Arbib 2012), where 150 “closed” auditory qualities of phonemes are more limited than “open” qualities of natural auditory sources 151 in establishing parity between sender and receiver (Liberman and Mattingly 1989). Likewise, music 152 differs from protomusic by abstracting the auditory qualities of pitch and rhythm, and using them to 153 construct histrionically expressive musical movement in a virtual space of vertical (harmonic) and 154 horizontal (melodic) axes (Nikolsky 2015). Abstraction of phonemes and using them to construct syllables 155 and words trigger the syntax creation process that marks the transformation of protolanguage into 156 language, opening great number of roads that all lead to the genesis of a great variety of languages. 157 Similarly, abstraction of pitch- and rhythm-classes generates a great wealth of music systems - in contrast 158 to homogeneity of protomusic. The gist of this diversification can be grasped from observing how music 159 skills advance throughout childhood (Hargreaves 1986). All newborns sound more or less the same in 160 their narrow assortment of cries and coos (Loewy 1995) which becomes progressively diversified 161 (Wermke and Mende 2009) until the child’s vocal musicking obtains clear traits of its native musical 162 culture (Louhivuori 2006). There is no reason to believe that this course of development is limited to 163 ontogeny. Anatomic organs necessary for production and perception of language and music are present in 164 Homo Heidelbergensis (Wurz 2010). Its newborns must have shared music/language capacities with the 165 modern newborns, except that their parents did not expose them to musicking as we know it today. 166 However, the latter circumstance would not revert the course of development of vocal skills from its 167 direction towards greater diversity and cultural specificity of vocalizations. 168 2. Jew’s Harp and its sound production 169 Traditional forms of Jew’s harp (JH) music fit the second, “intonational,” stage of the musilanguage 170 model, containing traits that are likely to have formative influence on the genesis of both, TO of music 171 and language. 4 172 JH is a unique musical instrument (Fig.1) that somehow evaded the attention of scholars until the mid- 173 20th century despite the fact that for at least 21 indigenous ethnicities of northeastern Eurasia (the largest 174 of which is Yakut) JH still remains a principal musical instrument in the scarce instrumentarium that 175 features primarily the instruments of “timbre-oriented music” (Nikolsky et al. 2020a). Such music engages 176 timbre as the primary expressive means and differs from “frequency-oriented music” that prevails in the 177 West by relying on “timbre-classes” rather than “pitch-classes” in its TO. A JH music piece is built from 178 the limited set of sounds made from onomatopoeic imitations of environmental sounds, speech-like 179 articulations, isolated harmonics and special effects. Changes in pitch in timbre-based music are usually 180 indefinite: performers take in consideration only the direction and the rough estimate of the interval of the 181 pitch change (e.g., step vs. leap). Evidently, people 182 habituated to such music have no need in “frequency- 183 oriented” musical instruments, so that JH completely 184 satisfies their musical needs. 185 Fig.1. The typical construction of JH on the example of 186 the favorite instrument by Ivan Alekseyev, made by Ivan 187 Kolodeznikov (Khampa, Vilyi ulus). All parts are numbered 188 according to their contribution to typical sound production 189 of the JH. Their Yakut names are provided by Shishigin 190 (2015). 1) Lamella, or “JH’s tongue” (tyl) is usually 1-2 mm 191 thick, 20-25 mm wide and 9-11 cm long. 2) Left and right 192 arms (syngaakhtar) are on average 5 mm thick and are 193 separated from the lamella by the mean gap of 0.1 mm (0.04- 194 1.6 mm). 3) The inner cheek (iss iedes) is usually the widest 195 of all 4 sides of an arm and comprises a wide angle in 196 relation to the upper front side of lamella. 4) The outer cheek 197 (tas iedes) usually forms an acute angle in relation to the 198 inner cheek. 5) The knee of lamella (khakhuora) is typical 199 for metallic instruments, where it supports stronger striking, 200 and usually makes a quarter of the lamella’s length. 6) The 201 tip of lamella (chyychaakh) supports more effective 202 plucking. 7) The ring (tierbes) can be held with various extent of firmness/looseness. 8) The frame (iedes) 203 can be held in a variety of ways. Together, these 8 parts determine the timbral characteristics of the JH 204 sound. 205 Amazingly, JH remains one of very few instruments whose sound generation is not yet fully understood, 206 leaving problems in its organological classification. According to Fox (1988), already the earliest 207 scholarly attempt to classify JH remained inconclusive: in 1636, Mersenne pronounced JH “percussive” 208 (Mersenne 1957), whereas circa 1640 he called it “pneumatic.” This controversy has not yet been resolved 209 (Kolltveit 2016). The classic work by Sachs and Hornbostel (1914) qualified JH as a plucked idiophone, 210 based on the plucking action of the hand on the lamella (Fig.1) (Sachs 1917). This classification lasted 211 until Crane (1968) pointed out that the lamella alone is incapable of generating melodies made of the 212 sounds of the harmonic series – therefore, reclassifying JH as a free aerophone. Not all organologists 213 accepted Crane’s correction (Montagu and Burton 1971). Ledang (1972) confirmed Crane’s classification 214 by experimentally demonstrating that the distance between the JH’s arms and the lamella is no less critical 215 for sound production than plucking of the lamella, which per se produces a bleak monotone when plucked 216 away from the mouth, and that the range of JH’s tones is determined by the configuration of the player’s 217 mouth. Adkins created an electric analog of the JH and demonstrated that JH’s sound is generated by the 5 218 resonances of the mouth cavity which can be varied continuously at a high frequency range (Adkins 1974). 219 Nevertheless, Kolltveit argues that JH should not be regarded as a reed instrument because it requires hand 220 action (Kolltveit 2016). Such arguments zoom into the mere mechanics of playing JH, ignoring what kind 221 of music is produced on JH. 222 • Qualifying JH as an idiophone makes it a percussive instrument, which subsequently emphasizes 223 rhythm as a primary means of its expression. 224 • Qualifying JH as an aerophone makes it a wind instrument, which emphasizes pitch as a primary 225 means of its expression. 226 The use of JH as a rhythmic instrument is exceedingly rare and is limited to the South-Indian and 227 Indonesian ensemble music (Morgan 2008). Across the vast region of Northeastern Eurasia, its most likely 228 homeland (Fox 1988, 45), JH is used as a melodic solo instrument (Yesipova et al. 2008). Absolute 229 majority of ethnicities of former Soviet Union consider JH purely melodic – obvious in its use to play 230 traditional songs (Alekseyeva 1986). A number of ethnic traditions, even those far away from North, 231 consider JH a wind instrument, because one has to blow into it, e.g., Khmer (Sam 2008). Huli of Papua 232 New Guinea explain JH’s sound production as thoughts formed in person’s chest and carried up to the 233 mouth through breathing, where thoughts roll off the tongue as words, spoken out with the help of JH 234 (Pugh-Kitingan 1977). 235 Breathing technique is more important for JH music than hand technique. It is simply impossible to 236 play on JH without breathing into it. But there are many JHs that can be played without any engagement 237 of hands (Mazepus 1989). Interaction of the strength and speed of inhaling/exhaling on the one hand and 238 the frequency and amplitude of lamella’s vibrations on the other hand can stop the sound or excite it 239 without plucking (Shishigin 2015, 47). There are entire traditions of handless performance on JH. Thus, 240 Altaic women used to play JHs while doing housework or handicraft, when their hands were busy, holding 241 the instrument by their jaw and controlling the lamella with their tongue (Konchev 2004, 13). The JH 242 designed for such performance was called t’ak komys (Tiukhteneva et al. 2006). Its name, translated from 243 Altaic as “jaw harp,” reflects its handless design (Sheikin 2002, 469).1 244 Breathing rhythm is not any less important for JH articulations than plucking rhythm (I. Alekseyev 245 1988). In Indonesian tradition, JH players even employ a special solfege to teach JH, where pitch levels 246 are named according to whether they are produced by inhaling or exhaling (Morgan 2008). Similar 247 approach is used in Bashkiria, where performers master the repertory of syllables in combination with 248 various patterns of breathing (Zagretdinov 1997). The hand-rhythm constitutes only one and the most 249 undemanding means of rhythmic expression – after the rhythms of breathing, mouth articulations and 250 pitch changes (E. Alekseyev 1991a). 251 JH’s sound is generated in 3 stages. Sound production starts with the excitation of lamella generating 252 transverse waves (Ledang 1972). Unlike longitudinal vibrations, transverse cantilever vibrations do not 253 form harmonically related modes, though they produce standing waves (Fletcher and Rossing 1998, 58– 254 67). Their propagation depends on the vector of the shear force applied to the lamella, causing different 255 states of polarization (Bettini 2017, 235–236). Each of the modes of transverse, longitudinal and torsional 256 vibrations features its own frequency, altogether comprising a complex tone that consists of numerous 257 partials of different intensity, some of which form periodic relations, while others do not (Fletcher and 258 Rossing 1998, 624). This can be thought of as a “spectral chord” where some component-tones are more 259 salient than others. Performer can activate a partial vibration by varying the angle of excitement of the 260 lamella, the tightness of fixing the clamped end and the jaw/lip pressure on the JH’s ring. 6 261 The slit between JH’s arms and the lamella acts as a filter, dampening most of the vibration frequencies 262 that lay above the lamella’s FF (Ledang 1972), comprising the second stage of sound production. Gaps 263 >0.2 mm cause the reduction of the periodic content in the spectrum, impoverishing it beyond any musical 264 use at gaps exceeding 1 mm. Absolute majority of commercially available JHs have the clearance size 265 between the lamella and the arms about 0.1mm.2 Ledang specifically states that it is this geometry that 266 explains why the enormous variety of JHs made of different materials retains the same characteristic JH 267 sound, easily recognizable by ear. 268 The mechanism of the second stage was identified by Adkins (1974). The turbulence of air, excited by 269 lamella’s vibrations, interacts with the tightly spaced arms as in an accordion where the valve drives the 270 air through the reeds. The lamella pumps air through the slot of the frame, increasing the air pressure in 271 the front and decreasing it in the rear, then goes through that slot – at which point the air begins to stream 272 into the low-pressure area. The thinner the arms, the faster the lamella travels through the frame and the 273 fewer partial vibrations are dampened by the arms, enriching the sound. As the amplitude of vibrations 274 recedes with the decay of sound, lamella’s oscillations slow down, and the radiated sound loses its 275 harmonic richness – from the smaller vibration modes to the larger. 276 The third stage in sound production is initiated when the air turbulence along the frame reaches the 277 player’s mouth and finds the Helmholtz resonator in the oral cavity. The coupling of sound generator to 278 resonator occurs through the air rather than the solid material of the JH body (Crane 1968). The player 279 continuously reconfigures his/her oral cavity, tuning it to different resonant frequencies: the more reduced 280 the cavity’s volume, the higher the frequency (Dournon-Taurelle and Wright 1978). JH is peculiar in its 281 design that delegates the frequency and dynamic controls to different stages: lamella generates partials 282 but cannot selectively alter any of the partials’ amplitude (only all of them together), whereas the player’s 283 vocal apparatus amplifies or attenuates specific partials while unable to generate any additional frequency 284 components (Trias 2010). 285 • The musical instrument itself is responsible for producing changes in frequency content within the 286 JH’s spectrum, many of which are not audible even to a close observer; 287 • The player’s vocal apparatus is responsible for making a particular frequency audible to observers 288 and the player. 289 Integration of instrumental and vocal production makes JH a hybrid musical instrument – what Eduard 290 Alekseyev, one of the main specialists on JH music, calls “centauric” (1991b) – akin a centaur that 291 inseparably combines anthropomorphic and non-anthropomorphic features. What this merging introduces 292 is a paradoxical combination of the monotonous complex tone, generated by JH’s lamella, and extremely 293 diverse articulations, produced by the vocal apparatus. In essence, the performer uses his/her mouth in a 294 way identical to speaking a language or singing a song – but silently, without engaging the vocal cords – 295 instead relying on the JH’s monotone. This has two very important consequences (Nikolsky et al. 2017): 296 1. JH’s monotone replaces the player’s voice, which does not allow the listener to retrieve personal 297 information about the player (gender, age, state of health, etc.). Effectively, JH makes every 298 person sound “the same,” thereby making a player anonymous. JH also conceals the affective 299 state of a player by cutting off expressive modulations of player’s voice. JH replaces 300 individualized human conspecific “affective prosody” with generic “intonational” prosody of 301 JH. 7 302 2. JH player’s articulations compensate for JH’s monotony by engaging much wider range of 303 motions of vocal anatomical organs than that of singing and speaking. JH maximally enriches 304 vocal articulation to support all sorts of naturally prototyped as well as “fantastic” sounds.3 305 Audio example 1. Onomatopoeic imitations of bird songs, horse trot, etc., in the khomus improvisation “A 306 summer morning” by Petr Ogotoyev. Used with his permission. http://chirb.it/AszLqB 307 3. Jew’s Harp as the world’s earliest vocoder and its cultural ramifications 308 The combination of monotone and diverse articulations, in effect, makes JH into a vocoder (Leipp 309 1963). Lamella generates the “carrier” signal, whereas the vocal apparatus passes it through the 310 “modulator filter.” Such vocoding is related to secrecy. Electric vocoders were invented to conceal the 311 speaker’s identity while delivering that speaker’s words in military communications (Tompkins 2010). 312 Adoption of vocoders in Western popular music also answered the need to conceal singer’s personal traits 313 (gender, age, ethnos, class, temper), perceived as potential vulnerability (Dickinson 2001). The same 314 applies to two cross-cultural uses of JH, typical for many indigenous traditional music cultures. From 315 North-western Europe to Papua New Guinea, across the entire Eurasia, JH is employed as a ciphering 316 device in private romantic serenading of young couples (Hauser-Schäublin 1995; McLean 1999, 265; Hsu 317 2001; Levin and Süzükei 2006, 116–17; Sam 2008; Uchida and Catlin 2008; Kartomi et al. 2008; Matusky 318 2008; Canave-Dioquino, Santos, and Maceda 2008).4 Amongst the speakers of tone languages JH is 319 effective in imitating not only linguistic phonemes but tone intonations as well (Poss 2012). 320 Another cross-cultural tradition, probably of greater antiquity, is using JH to camouflage one’s voice in 321 rituals dealing with supernatural forces. JH excels in the production of “grammelot” effect (Jaffe-Berg 322 2001). Such enigmatic “foreign” speech, interspersed with onomatopoeic imitations, is used by Siberian 323 shamans to represent messages from spirits (Alekseyenko 1988). Common people, too, use JH as a “vocal 324 mask” for protection from evil. The animistic belief system, surviving in Altai, holds that masters of 325 landmarks (mountain, river, forest) can steal a person’s voice, which equates with that person’s death 326 (L’vova et al. 1989, 90). In many indigenous cultures of Eurasia, voice is regarded as an immanent live 327 object, tied to the person through breathing and vulnerable to invasion by a spirit (Pashina et al. 2005, 50). 328 Hence, JH is seen as a shield protecting its possessor.5 329 “JH magic” most likely comes from the totemic ideology that is still very much alive in Siberia and Far 330 East, especially in Altai (Nikolsky et al. 2019). Many ethnicities draw their ancestry from a specific plant: 331 Nivkhi – from larch, Oroks – from birch, Ulchi – from cedar, Ainu – from fir (Duvan 2003). Plants were 332 believed to breathe and to possess soul (L’vova et al. 1989, 89), therefore capable of nourishing or even 333 giving birth to human kin-founders (Tadysheva 2018). As of today Khakassian and Altaic kins still hold 334 different tree species for their ancestors (Sagalayev and Oktiabr’skaya 1990, 50–59). A dying tree near a 335 human settlement is seen as an ill omen for the death of someone from the corresponding kin: Kuzen – 336 for pine, Tubalar – for aspen, Jyc – for fir, Todosh - for honeysuckle (Kypchakova 2006). Therefore, 337 cutting of ancestral trees is tabooed across Altai. 338 In Siberia, Ugric and Turkic ethnicities use trees to identify genealogical “trees”: they believe that bones 339 of people from the same kin are made of the same “wood” type in accordance with the paradigm of the 340 tree species and the forest – so that every person corresponds to some tree, while their kin – to the same 341 tree species within the forest that represents a particular ethnos (Sagalayev and Oktiabr’skaya 1990, 43– 342 57). Traditional identification system involves a totemic animal, an ancestral tree and a sacred mountain 343 (Tadysheva 2016). This system is still in use: in 2010, there were 82 kins (seoks) registered by the census 344 in Altai Republic (Tyukhteneva 2015).6 Turkic cosmology envisages the Seok Tree as the global origin 8 345 of life (Turan 2006). In such belief system JH made of the ancestral tree would secure its possessor against 346 misfortune with the ancestral protective power. Thus, in Buryatia, after a shaman died, no one played his 347 personal JH – it was hung on the sacred tree nearest to the shaman’s place of burial (Agapitov and 348 Khangalov 1885). 349 The association of JH with supernatural powers might have originated in the “Aeolian harp” naturally 350 made by the lightning striking an ancestral tree. Turkic ethnicities of Siberia consider such trees purified 351 of evil spirits and have a custom of keeping their chip as a talisman, using such chips to make JH (Duvan 352 2003). Modern Yakut master-makers of musical instruments explain that the lightning strike dries and 353 fragments the tree so that its wood can “sing” unlike the regular wood (Sheikin 2002, 68). Yakuts call 354 such JHs mas-khomus (Yakut, “chip Harp”) and today use it as a children toy (V.Dyakonova 2017).7 355 Siberian myths attribute the invention of JH to the bear, a sacred animal, related or ancestral to humans, 356 who accidentally discovered that a splinter of a tree makes a peculiar sound (Startsev 2017, 104). Ugrian 357 ethnicities worship a bear-looking deity Kaigus’, who made the first JH from the birch’s chip and used it 358 to imitate animal calls in order to gain power over these animals – later teaching human hunters to play 359 JH before hunting to secure prey (Sheikin 2002, 126). It is quite possible that this myth indeed faithfully 360 reflects the invention of JH – albeit by humans - Ude hunters of Primorye still use wooden JHs right after 361 setting hunting traps despite its reputation of a children toy, unlike the “artistic” metallic JHs used to make 362 “serious” music (131). 363 The vocoding capacity of JH is perhaps of greatest value to shamans whose job in traditional societies 364 consists of dealing with evil spirits on behalf of their clients (Balzer 2016). In fact, the idea of putting on 365 an “auditory mask” complements putting on a facial mask that Siberian and Altaic shamans wear for self- 366 protection (S.Ivanov 1975). Not surprisingly, JH has earned the reputation of shamanic accessory all over 367 the Eastern part of Eurasia: from Bashkiria (Shchurov 1995) to Malaysia (Canave-Dioquino, Santos, and 368 Maceda 2008).8 A distinguished turkologist, Saul Abramzon, held that ancient Turkic shamans used two 369 musical instruments: JH and tambourine (Suleimanova 1994). Across Altai shamans receive their 370 “standard” musical instrument, a tambourine, only after reaching maturity in their supernatural skills, prior 371 to which they were allowed to use only JH (Rouget 1985, 126). This applies to Buryat and Mongol 372 shamans (Pegg 2001), as well as Ulchi in Amur and Sakhalin (Duvan 2000). Noteworthy, Chukchi, Kereks 373 and Koryaks call JH “a mouth tambourine” – probably because of the shared shamanic affiliation 374 (Podmaskin 2003). 375 Some shamans never qualified to receive a consecrated tambourine and were left with a JH until their 376 death (Vainshtein 1991, 248). This circumstance led Vainshtein to believe that JH was the primordial 377 musical instrument of shamanism, and tambourine was introduced at a later point (273). Vasilevich (1969, 378 209) considered JH to have proto-Tungus origin and belong to female shamanism. Amongst Ugric 379 ethnicities, JH retains the reputation of female shamanic accessory (Alekseyenko 1988). The reputation 380 of JH, especially its more ancient, non-metallic varieties, as belonging to the female and children domain, 381 is exceedingly strong amongst nearly all Middle Asian, Ugric, Turkic Altaic (Emsheimer 1986) and Far 382 Eastern ethnicities (Mazepus and Galitskaya 1997) - to the extent that male players would abstain from 383 playing such instruments (Sheikin 2002, 131). This could be the remnant of an ancient shamaness cult, 384 rooted in a special need for vocoding by wives within the circle of husband’s relatives. The traditional 385 practice amongst Uralo-Altaic ethnicities, ancient enough to have left a genetic impact, is for men to marry 386 foreign women (Pakendorf 2007). According to widespread behavioral code, the wife remained a stranger 387 to her husband’s kin and therefore was tabooed from calling his kin’s totemic entities by name under the 388 threat of harming the kin (Tadysheva 2018). Using JH to articulate their names or hint at them by sound 389 imitations would circumnavigate this taboo. This may explain why most archaeological finds of JHs in 9