One of the intentions of this thesis is to find ways of looking at the transmission of the honkyoku tradition that are valid from a musicological perspective as well as being in tune with the tradition in an ’emic’ sense, that is, in a way that is tradition-based. With this in mind, issues concerning the transmission of the honkyoku “Reibo” of the Ôshû lineage are examined through the simultaneous application of two analytical methods to transcriptions of recordings of ten performances.

The first method is as follows: detailed transcriptions are made of ten performances of pieces that oral histories and genealogies suggest are versions of the piece “Reibo”; an orthographically simplified transcription is made so that the melodic contour of one performance can be compared with melodic contours of the other performances. It will be demonstrated from these comparisons that the pieces chosen for analysis are indeed versions of the same piece. It will also be shown that the lines of transmission of the pieces indicated by sources such as oral histories and genealogies are confirmed by the patterns of similarities and differences that arise from the comparison. At this stage, the formal structures and melodic formulae of honkyoku become observable. This method, which relies upon musicological analytical, or ‘etic’ concepts for its departure point can only be successful if achieved nonetheless through a tradition-based approach.

The second method of inquiry used in this thesis is completely different from the first in that it draws upon traditional ways of analyzing honkyoku. It was conceived by observing how people within the tradition talk and write about honkyoku in an analytical way, in particular during lessons and in various written accounts. In contrast to the large scale structures and melodic formulae highlighted by the first method, this method deals primarily with small and frequently subtle details. Details of this kind are, to performers of honkyoku, the major issue discussed in the course of transmission; an examination of such details in performances of the “Reibo” pieces contributes insights into how insiders to the tradition view the music, and show what elements are likely to be or not be transmitted from performer to performer.

Both the fundamentally musicological ‘etic’ approach and the detailed, tradition-based, more ’emic’ approach used in this thesis rely largely upon data derived from the transcriptions of performances of “Reibo” pieces by various performers. Philosophical and technical problems arising both from the transcription of honkyoku and from the above two analytical approaches are discussed below.

6.1 Musicological approaches to honkyoku analysis

In the following discussion, I will distinguish between formal analyses, by which I mean analyses that use musicological methods and look primarily at the form or structure of the music, and informal analyses, by which I mean analyses that use traditional methods to examine detailed features of the music that most concern the members of the tradition. I will first deal with formal analyses, and then discuss informal analyses.

Among the limited number of transcriptions and formal analyses of honkyoku – as distinct from the informal process-oriented analyses made by members of the tradition and described below – that exist in the literature, the overwhelming majority are by non-Japanese scholars in sources written in languages other than Japanese. To my knowledge, Tukitani is the only musicologist writing in Japanese to have extensively transcribed honkyoku pieces, though most of these transcriptions remain unpublished.1 She has also used data from transcriptions in analyzing large-scale structures of honkyoku and examining hierarchies of pitches as related to theoretical scales.2 Since the mid-1970s, however, Tukitani’s attention has shifted from transcription-based formal analysis of honkyoku, to other issues such as genealogies of performers and lineages of pieces. This appears to indicate a shift of interest from the study of structural elements of the music itself to aspects of transmission of more concern to the honkyoku tradition. There are no other examples, to my knowledge of transcriptions of honkyoku performances being made or used as data for analysis by and for members of the honkyoku tradition.

1. In 1989, when I asked about transcriptions of honkyoku, Tukitani showed me several binders full of transcriptions that she had done during her university days, under the tutelage of the late Koizumi Fumio. Though Tukitani appears to use data derived from these transcriptions, she has rarely discussed or included transcriptions in her recent publications.
2. For a more detailed discussion of Tukitani’s work, see below (pp.316319).

One reason for the relative lack of interest on the part of shakuhachi performers in formal analysis of the honkyoku is the emphasis placed upon the ‘process’ or act of performance in the present moment. If one’s attention is constantly focused on the here and now of performing a piece, formal structures become relatively unimportant. Thus large scale structural forms are of little interest to honkyoku performers.

Marett (1992) has suggested an analogy to this in Zen practice. The chanting of sutras are an integral part of Zen practice. These sutras have semantic structure; their words are profoundly meaningful. Yet during the rhythmical chanting of the Heart Sutra during a week long sesshin or retreat for example, the consciousness of the chanter tends to focus down to the level of each individual syllable of the chant. The syntax of the sentences and even the meanings of the words in the sutra may on that level, be lost in a way similar to what happens to formal structure in the mind of a honkyoku player during performance.

Another example of emphasizing the present moment can be seen in the Zen practice of breath counting during zazen (, seated meditation). Breath counting is the practice of counting ‘one’ for the inhalation, ‘two’ for the exhalation, ‘three’ for the next inhalation, etc., until reaching ‘ten’ on the exhalation. The subsequent inhalation is then counted as ‘one’ again and the entire process repeated. In breath counting, the formal structure is clearly the repetition of ‘1, 2, 3, 4…10’. But as Aitken (1982:11, 24) explains, “you must devote all your attention just to ‘one’, just to ‘two’…. Become each point, each number in the sequence of counting. You and the count and the breath are all of a piece in this moment. Invest yourself in each number. There is only ‘one’ in the whole universe, only ‘two’ in the whole universe, just that single point”. The formal structure of breath counting falls into the background, evaporating into a universe of “just ‘one’, just ‘two'”.

This is not to imply that in traditional honkyoku, structures cannot be perceived. Contrary to Gutzwiller’s insistence that “it is extremely difficult to observe higher order formal structure in the pieces of honkyoku” (Gutzwiller and Bennett 1991:58), there are a number of honkyoku with readily observable formal structure. A prime example of this is the piece “Kyorei” (, Empty Bell) as transmitted through the Myôan and Chikuho lineages. The overall structure of the piece, as indicated by an analysis of the score, is A – A’, with A’ being a slight variation of A, but played in the upper octave. Both A and A’ have a musical contour which starts on the pitch G (when played on a 1.8 shaku instrument), climaxes on the pitch D above the starting pitch G, and ends on the pitch D below the starting pitch G. Figure 16 presents the score of “Kyorei” as transmitted in Chikuho ryû, transnotated from the traditional shakuhachi notation into staff notation and annotated to show the structures discussed above. Examples of other honkyoku which exhibit easily recognizable structural forms include “San’an” (, Safe Delivery), and the Myôan version of “Honte chôshi” (, Original Fingering Searching).3

3. The Watazumi version of “San’an” has the overall form AA’BB’C. The Myôan version of “Honte chôshi” has the overall structure of AA’B.

There is also in honkyoku a process of structuring through patterned repetition of particular techniques. For example, in the piece “Ajikan” (, Seeing the Letter ‘Ah’), the process of performing a number of yuri techniques (various playing techniques which produce a variety of pitch oscillations) is repeated throughout the piece. The yuri technique is thus a defining and unifying feature of the the piece. The final phrase of the piece is sometimes played with an extended yuri technique, completing the unifying process. As will be shown below, formal structure can also be observed in the Ôshû family of “Reibo” pieces (see pp.360365).

Despite the existence of formal structures in honkyoku, however, they are nonetheless given little attention by members of the tradition. Gutzwiller is accurate in stating that “In Japanese traditional music the essence of music is not an abstract structure whose beauty could be understood even without playing the music at all, but it is the performance, the way music is played” (1974:88), and that “We may therefore expect that ‘structure’ occupies a different place in the mind of a Japanese musician than it does in the case of his Western counterpart” (1974:87).

This should not be taken to mean that structure has no place at all in the Japanese musician’s mind, or that formal structure cannot be observed in honkyoku beyond the individual phrases (Gutzwiller 1974:128). Gutzwiller (1974:130) is inaccurate in concluding that “vagueness and indeterminancy form the key factor in the construction [my emphasis] of honkyoku”. Formal structures exist and are frequently labelled in traditional honkyoku scores. These named sections will be the starting point of reference from which the comparative analysis of the “Reibo” pieces in this thesis will begin.

6.1.1 Transcribing honkyoku

From the above, it is reasonable to conclude that traditionally based methods are unlikely to be useful in detecting many of the kinds of formal structures that might exist in honkyoku. If one desires to examine them in any detail (a desire that is not found within the tradition), one must rely on musicologically (rather than traditionally) inspired analytical methods, such as the isolation and subsequent comparison of melodic contours. Although musicologically based methods can be problematic in their dependence upon either transcriptions of performances or transnotations of traditional shakuhachi scores into staff notation, it is not necessary to go as far as at least two western musicologists, Gutzwiller (1974:138) and Stanfield (1977:190), who conclude that transcriptions of honkyoku are impossible. It will be argued that as long as due regard is given to the aesthetic parameters of the honkyoku tradition, transcriptions can contribute to a valid understanding of transmission. I will begin by summarizing the arguments against transcription of honkyoku and then present counter arguments.

Gutzwiller has consistently avoided transcriptions of honkyoku because of philosophical considerations and what he sees as technical difficulties. According to Gutzwiller (1974:138-142), since the tradition maintains that the process of performing honkyoku cannot be separated from the product (i.e., the resultant musical sound), an understanding of any sort of product or ‘form’ in honkyoku can only exist in the process of performing. This being the case, meaningful transcriptions, which are possible only if the result is separated from its process, are a violation of the central aesthetic of honkyoku. Gutzwiller (1974:134) calls the central aesthetic fukikata ( ‘way of playing’), which “comprises [of] technique and its result”. Analyses of transcriptions are consequently considered worthless by Gutzwiller. Furthermore, leaving aside the question of process and form, Gutzwiller (1974:138-142) argues that the precision of Western staff notation inevitably distorts the indeterminacy of honkyoku, both in terms of pitch and rhythm.4

4. Staff notation is even less suited to symbolize timbre, an element of honkyoku at least equal in importance to pitch and rhythm.

Though accurate in his assessment of the interrelationship of process and product within the tradition, Gutzwiller, in concluding that transcriptions of honkyoku using staff notation violate the spirit of the tradition, risks objectifying, and consequently distorting the essence of honkyoku just as much as he accuses transcriptions of doing. Gutzwiller’s statement that “Honkyoku is a non-public music and it will be understood by exactly as many people as are playing the music” (Gutzwiller 1974:142) is only true in terms of a narrow doctrinaire definition of ‘understanding’ as the sole prerogative of a performer. For a mature performer, however, there are many levels of understanding. While the level of understanding one might gain from an analysis of a transcription of a honkyoku is completely different from the understanding of a performer during the act of performing, I would argue, nonetheless, that it is a kind of understanding.

Gutzwiller (1974:134-136) equates the Zen kôan (), which he defines as “a problem – usually stated in the form of a question – that has in itself no logical meaning”, with honkyoku. He argues that if honkyoku is used together with meditation as “an activity to reach enlightenment” then it must function as a kôan. Aitken (1990:330), however, defines the word kôan not as a sort of functional tool of meditative practice, but as ‘universal particular’. He further elaborates that a kôan is “a presentation of the harmony of the universal and the particular;5 a theme of zazen to be made clear”. Clearly a kôan is not a didactic strategy or tool at all. The same can be said of honkyoku.

5. Though Gutzwiller’s emphasis on “function” misses the point, he is accurate in equating honkyoku with kôan. There is no better definition of honkyoku than “a presentation of the harmony of the universal and the particular”.

The doctrinaire view of honkyoku held by Gutzwiller early in his career almost twenty years ago (1974), resembles the view of Zen students who are stuck on the ‘one hundred foot pole’ and have not stepped off into the ‘worlds of the Ten Directions’6. With a more matured understanding of honkyoku comes the realization of living the ‘everyday’ or ‘ordinary mind’7, and not trying to hold onto the peak experience of truth.

6. The “hundred foot pole” and the ‘worlds of the Ten Directions’ are referred to in Case 46 in Mumonkan. See Aitken (1990:273-277) for a complete reference to this case.
7. The ‘ordinary mind’ is touched upon in Case 19 of Mumonkan. See Aitken (1990:126-131) for a complete reference to this case.

The technical difficulties in transcribing honkyoku that Gutzwiller sees as violating the ‘way of playing’ are equally less of a problem than they are made out to be. Though the precision of Western staff notation is not ideally suited to represent the indeterminacy of honkyoku, it is able to represent a single performance of honkyoku. The performer of honkyoku is faced with infinite possibilities of performance, but once the performer has chosen and the performance takes place, there is no longer any indeterminacy, and no problem of representing indeterminacy. What must be remembered is that a transcription in staff notation of a single honkyoku performance is just that. It is not honkyoku itself, nor is it a representation of the infinite number of possible performances of honkyoku.

Transcriptions are analogues, like maps and words. A map cannot be identical with ‘the real thing’ without becoming ‘the real thing’, in which case it is no longer a map but the landscape itself. Yet maps are extremely useful in gaining an understanding of ‘the real thing’, for example when negotiating from one location to another. Likewise, the word ‘love’ is not what it represents, but functions only as an analogue. As with maps and words, transcriptions can only represent aspects of the recorded sound (itself already an analogue) of one performance of honkyoku. A single performance is not honkyoku, but rather only one of an infinite number of manifestations of honkyoku. A recording of a honkyoku performance is even farther from the actual honkyoku, in that it cannot reflect the many dimensions of the performance that are inseparable from the moment of the act of performance (Marett 1992). A transcription of a recording removes one a step further from honkyoku. Nevertheless, when one is clear about what they are, transcriptions, like maps and words, can be useful in understanding certain aspects of honkyoku.

Stanfield (1977:190)) has adopted the same stance as Gutzwiller on transcribing honkyoku, stating that “a detailed transcription of any one performance runs counter to the ‘gestalt’ of the music”. Instead, he attempts to represent “an ideal application of performance practice details” as taught by his teacher. Ironically, the make-believe result – a representation of something that has never existed – is even further away from the reality of honkyoku than transcriptions of recordings of performances that actually took place. The idealized transcriptions are like maps based upon one’s idealized memory of the streets of a city rather than upon the city streets themselves.

Contrary to both Gutzwiller’s and Stanfield’s unyielding position, a transcription of honkyoku in staff notation can be a useful source of data if the limitations (inherent in any analogue) are always taken into account in deriving conclusions from the data. Like the image generated by an electron microscope, which is not the object itself, a transcription of a recording of a single honkyoku performance, though in no way honkyoku itself, can still be a valuable source of data from which conclusions about honkyoku can be made.

It is interesting to note that in his most recent article on shakuhachi honkyoku (Gutzwiller and Bennett 1991:36-59), Gutzwiller presents a computer aided analysis of the acoustical characteristics of a few honkyoku phrases, which he believes represent some of the most common tone cells8 in honkyoku of the Kinko lineage. The data used in the analysis was generated by actual performances of the phrases, in this case by Gutzwiller himself. Though Gutzwiller (1974:138) earlier states that transcriptions of honkyoku are an impossibility, his conclusions in this later article are based upon the analyses of computer generated images, which are, like transcriptions in staff notation, analogues of performances of honkyoku.

8. Gutzwiller uses the term ‘tone cell’ to denote what he believes to be a fundamental musical unit in the honkyoku. See below p.323 for his definition of ‘tone cell’.

A parallel paradigm with regards to words (analogues) versus the experience of realization (honkyoku) exists in the Zen Buddhist tradition. On the one hand, Suzuki (1966:81) states that “Buddhist faith is not looking outwards but inwards. It denies any reality in words, in concepts, in language.” It follows that there are many instances of Zen masters who “sometimes utter a primitive cry or burst out in a meaningless ejaculation or gesture” (Suzuki 1966:50). Yet Zen literature abounds with the words of the masters. Twelfth century Zen master Dôgen was especially prolific in his writings. Kim (1985:58) explains that for Dôgen, “words, like deeds, have at once limiting and liberating functions, or, as Dôgen himself put it, ‘discriminating thought is words and phrases, and…words and phrases liberate discriminating thought”, and “In spite of inherent frailties in their make-up, words are the bearers of ultimate truth. In this respect, words are not different from things, events, or beings–all are ‘alive’ in Dôgen’s thought.” On this level of thinking, transcriptions of recordings of honkyoku performances are also ‘alive’ with meaning.

6.1.2 Existing musicological analyses that use transcriptions

Objections to the transcription and analysis of honkyoku and questions regarding their compatibility with the central aesthetics of the tradition were addressed above. Before embarking upon the transcription and subsequent analysis of the “Reibo” pieces, a brief review of existing musicological analyses of honkyoku may suggest analytical methods to explore or avoid. The following critique of existing analyses looks at their effectiveness, firstly, as analogues and a means of communicating an understanding of honkyoku and, secondly, at the degree to which they are accurate analogues.

As mentioned above (p.306), and in contrast to Gutzwiller and Stanfield, Tukitani transcribes and analyzes honkyoku. Tukitani first used data from transcriptions of performances in analyses of honkyoku in her M.A. thesis (1969a), and in an article derived from that thesis, entitled (Shakuhachi koten honkyoku no kenkyu^; kôseihô ni tsuite, Research into the Shakuhachi Classical Honkyoku; a Constructive Analysis) (1969b).

The initial section of Tukitani’s analyses focuses upon ‘the phrase’, which is defined within three parameters: scale, dynamics, and rhythm. By scale, Tukitani means the stability derived from what she calls the core notes or nuclear tones (, kakuon, in contrast to ‘tonic’, with its harmonic implications of western music theory). In honkyoku performed on the standard 1.8 shaku length flute, these core tones are d, g, a, and c (1969b:44), of which d and g are the more fundamental.9 The dynamics are the patterns of relationships between loudness and softness and the melodic lines. Rhythm is the patterns of relationships between note durations and the melodic line. Examples of dynamic and rhythmical patterns found in honkyoku are also described, but their sources and the methods of deriving them are not explained in this particular article.

9. The term ‘core tone’ and Tukitani’s definition is adopted in the analysis of this thesis (see p.363).

While this sort of analysis is valuable, it is not as germane to the main theme of this thesis, the transmission of honkyoku, as the second part of Tukitani’s analysis, which looks at the overall structures of particular honkyoku. Unlike Gutzwiller, she has no problem about describing the formal structure of honkyoku she studies. An example of a structure found in many individual honkyoku is described as having a beginning, a climax or ‘high sound’ (, takane), a ‘shift’ or ‘change’ (, ten), being a secondary climax, and an ending. Other structural forms given as examples are more detailed. For example, the piece ‘Kokû’ as performed at the temple Myôanji in Kyôto, is said to have the six part structure: A, A’, B, C, D, and E. The section A is made of the phrases a+b; section A’ is a’+b’; section B is c+b’+d; section C is e+b’+d; section D is f+e+b’; and section E is g+b’+h. Tukitani notes that if the b phrase is thought of as the core, then the sections could be labeled A, A’, A”, etc. (Tukitani 1969b:50-51).

An article written four years later, entitled (Shakuhachi koten ni okeru dômei ikyoku no mondai “Questions Concerning Classical Shakuhachi Pieces of Different Melodies with the Same Titles”) (1973:225-250), points even more to the direction that this study takes. Tukitani analyzes three performances of the piece “Mukaiji” representing three lineages, Myôan Taizan, Chikuho, and Kinko. In contrast to the previous work, analyses in this article are comparative rather than structural. In other words, the versions of each piece are compared with one another rather than individual pieces being taken apart to see how they work. The comparative analysis between versions of pieces representing different lineages directly addresses the issue of transmission, a characteristic that becomes increasingly evident in Tukitani’s later work.

Though Tukitani’s analysis in this article is based upon transcriptions of the entire piece, transcriptions of only the first seven to eleven phrases of each version are presented. In the same article, notes representing the melodic contours of versions within the groups of pieces “Kokû” and “San’ya” are also compared. These notes are said to be derived from transcriptions, but the transcriptions are not shown and it is unclear how the melodic contours are derived. The derivation and analysis of ‘prominent notes’ in Tukitani’s later article suggests a method of analysis adopted in this thesis.

Both of these early analyses of Tukitani suffer somewhat from the lack of explanation of the methodology of both transcription and analysis. This thesis, on the other hand, attempts to make clear the methodology of transcription and analysis. Furthermore, in the comparative analysis, Tukitani looks at only three lineages, and again does not clearly state the performer and performance used in the analysis. The basis for the analysis in the present thesis is broadened to represent particular performances by ten clearly identified performers who represent at least five lines of transmission.

Finally, in leaving aside the area of formal analysis and concentrating upon comparative analyses of pieces representing different lineages, the trend in Tukitani’s later research toward issues almost entirely related to transmission can already be seen. As stated earlier, this trend suggests that over time, Tukitani is intuitively focusing more upon the issues that are most pertinent to the performing members of the shakuhachi tradition. The direction Tukitani’s research has taken validates the emphasis this thesis places upon the issue of transmission,

Two other analysis based in part upon transcriptions of actual performances of honkyoku are by Lee (1986) and Takahashi (1990). Transcriptions of performances of the piece “Kokû” (, Empty Sky) by three generations of performers within the Chikuho lineage are analyzed by Lee. Transcriptions of entire performances of pieces are analyzed and the methodology employed discussed. Solutions to some of the technical problems in transcribing honkyoku are suggested. For example, the use of a time line together with the spatial representation of duration of notes is adopted to reflect the ‘free rhythm’ of honkyoku. Other problems, such as the relationship between the process of playing notes and pitch, as well as between process and the important element of timbre, are not, however, addressed. In the present study, a method of reflecting the relationships between process, pitch, and timbre in the transcriptions is developed. While similarities and differences in melodic content, durations, and other components between the three generations of performers are described by Lee (1986), structural forms and organizational patterns such as tonal cells are not discussed.

Though implicitly about transmission, the subject is not explicitly addressed. Most of the discussion is about change within the tradition, but not about change and transmission. In addition, the study is limited in that it discusses the transmission of only one piece within a single lineage, in this case, the piece “Kokû” of Chikuho ryû.10 As mentioned above, the present study looks directly at an issue central to the tradition, that of transmission, making use of data derived from recordings of ten performances by players representing at least two versions of a piece and five lines of transmission.

10. Almost all other honkyoku analyses written in a musicological context also deal with only one lineage of transmission, most commonly being limited to Kinko ryû.

Takahashi (1990:295-309) makes use of both transcriptions and scores written in original notations in his analysis of “Honte chôshi” (, “Original Fingering Tuning”). He concludes that analyses based on traditional scores alone are impossible. Takahashi’s arguments are discussed in detail below (pp.327328).

Two of the earliest musicological analyses of honkyoku in non-Japanese literature were attempted by Malm (1959:158-162) and Weisgarber (1968:313-343). Malm transcribed and analyzed the piece “Hi-fu-mi Hachi Kaeshi” (, “One Two Three, Returning the Bowl”). Weisgarber also transcribed and analyzed “Hi-fu-mi Hachi Kaeshi”11, as well as the pieces “Banshiki-no-Shirabe” (), and “San-ya Sugaki” [sic] (, should read “San’ya sugagaki”). One of the problems with Malm’s transcription and analysis stems from an apparent ignorance of the fact that “Hi-fu-mi Hachi Kaeshi” is actually two pieces, “Hifumi chô” and “Hachi gaeshi”.

11. Malm and Weisgarber call the piece “Hi-fu-mi Hachi Kaeshi”; “Hachi gaeshi” is linguistically more correct.

An even greater problem is Malm’s unqualified use of European musical terminology such as ‘tonic’ and ‘dominant’ together with such conventions as time signatures, and bar lines. In doing so, Malm arbitrarily forces honkyoku into theoretical boundaries that are meaningless to traditional ways of viewing the music. The harmonic framework and hierarchy of pitches implied by Malm’s analysis are not applicable to honkyoku.12 Likewise there are no beats in honkyoku, much less a hierarchy of beats of the type that time signatures and bar lines imply. The result is a transcription that is culturally inappropriate as an analogue of honkyoku performance. Malm (1959:160) uses the term “breath phrase” once, but only in the context of a discussion of “grace notes”. Like Weisgarber (see below), the importance of the breath in performing honkyoku, as well as the phrase as a formal unit of honkyoku, escaped him.

12. In the transcriptions and analyses, I will also make use of key signatures, but with qualifications which are clearly stated (see p.356).

Despite the above, it is to Malm’s credit that his observations in a footnote (Malm 1959:162) regarding the fluctuating of the tonal centre hint at the concept of core notes or tonal centres used by Tukitani, Gutzwiller, and others to describe honkyoku.

There are also a number of problems with Weisgarber’s discussion of shakuhachi honkyoku, including his analyses and transcriptions. First of all, Weisgarber assumes that the “tonic” is ‘g’ just because ‘g’ is the final note of the piece, though the “tonic” could equally be ‘d’. Like Malm, Weisgarber ignores the distinction between the two pieces “Hifumi chô” and “Hachi gaeshi”. The most important unit of honkyoku, the phrase as delineated by breaths, is also disregarded. Weisgarber uses the word ‘phrase’ twice but does not define it. In his transcription he uses dotted lines to indicate “what seem to be the phrase, period, or sectional endings” (Weisgarber 1968:322); these dotted lines do not, however correspond to breath phrases.

Central to his analyses are what Weisgarber calls “patterns” or “cells”. These patterns, which are arbitrarily labelled with numbers on the transcriptions, are usually segments of phrases and encompass most of the musical material. Some sections of phrases are inexplicably ignored, however, being neither part of one of Weisgarber’s “patterns” nor labelled a pattern in themselves. One of a number of examples of this in Weisgarber’s transcription and analysis of “Hi-fu-mi Hachi Kaeshi” (Weisgarber 1968:319) occurs on the sixth line of between what is labelled patterns no. 10 and no. 7 (see figure 17, no. 1).

Also left unlabelled by Weisgarber are segments of phrases which are identical to segments that are labelled as “patterns”. An example of this is the first phrase of what Weisgarber calls section 2 (it is in fact the penultimate phrase of “Hifumi chô”), which is identical with “pattern 1”, the first phrase of the piece, yet is not identified as such (see figure 18, no.2).

Weisgarber’s arbitrary and inconsistent labelling and non-labelling of segments of the transcriptions as ‘phrases’ brings to mind the objections made by Treitler to the labelling of segments of plainchant as ‘formulas’ in an analysis of plainchant by Levy (Treitler 1975:15-23). Treitler argues:

“If there is any point in speaking of formulas at all, it is that they play into the transmission, or the composition if you will, of the chants to which they belong….we must have some boundary criteria for identifying a formula and distinguishing it from other passages. For the identification of a formula is an assertion of the existence of a piece of more-or-less fixed or stereotyped stuff-material that the composer put into this or that place or that the singer held in readiness for performance upon reaching a certain point in the melody. Either way, we want to be able to say with some confidence, here is that formula, there it is not, and there again is a variant of it. If we cannot do these things with confidence and with criteria that are demonstrable, we will have lost the point of the analysis, for then we would have no reason to think that the formula was any more distinct in the mind of the composer or the singer than in that of the analyst. And in those circumstances we could not make the assertion we meant to make by identifying the formula in the first place.”

(Treitler 1975:15-17)

Treitler’s point that formulae must be able to be identified and distinguished from other passages with demonstrable criteria before they can be used in the analysis is applicable to Weisgarber’s article. With his arbitrary and inconsistent “patterns”, chosen with no demonstrable criteria. Weisgarber cannot, as Treitler says, “make the assertion [he] meant to make by identifying the formula in the first place”.

Gutzwiller is one of the most widely published non-Japanese scholars of shakuhachi (1974, 1983, 1984, 1991). In contrast to Malm and Weisgarber, he writes with the insight of a shakuhachi performer, and is accurate in many of his observations. Some of his conclusion, however, are inconsistent or based upon an incomplete understanding of the honkyoku tradition. Among his arguments against transcriptions of performances of honkyoku, discussed above (p.311), is his belief that an understanding of any sort of product or ‘form’ in honkyoku can only exist in the process of performing. Yet having stated that it cannot be done, he proceeds to convey an understanding of honkyoku outside of the process of performing, that is, in his writings.

According to Gutzwiller and Bennett (1991:36), “the primary vehicle by which musical meaning is conveyed” in honkyoku is what he calls the ‘tone cell’. Similar to the ‘patterns’ observed by Weisgarber, Gutzwiller’s tone cells are in their simplest form a main note, which is kari, and an auxiliary note, which is meri.13 Within these simple structures are patterns of intricate relationships between pitch movement, dynamics, and timbre. The pitch of kari notes is fixed, while that of meri notes moves in specific ways. Both timbre and dynamics develop within a tone cell as determined by the movement from meri to kari and back to meri again. Tension and release is achieved in honkyoku through the interplay of the three elements of pitch, timbre, and dynamics within the tone cell and from one tone cell to the next (Gutzwiller 1983:347-348). More specifically, Gutzwiller’s tone cells have three parts, a preparatory note, which is unstable in pitch, softer in dynamics, and meri in timbre; the main note, which is stable in pitch, louder in dynamics, and kari in timbre; and the ending, which is unstable in pitch, softer in dynamics and meri in timbre (Gutzwiller and Bennett 1991:54).

13. The terms kari and meri denote performance techniques that change the pitch being produced without changing the fingering, by altering the angle and distance from which one blows onto the blowing edge of the mouthpiece. The kari technique raises the pitch, while the meri technique lowers the pitch. The pitches produced with the standard, open-hole fingerings (i.e., D, F, G, A, C on the 1.8 shaku length flute) are also called kari notes. The latter meaning of the word is used here.

One must remember that Gutzwiller’s studies are based solely upon honkyoku of the Kinko lineage, which make up a fraction of the entire honkyoku tradition. It is not surprising that, in some cases, his conclusions are applicable only to the Kinko lineage. They become problematic when applied to studies, such as the present thesis, which examine transmission over a number of lineages. While his tone cells may accurately describe Kinko honkyoku, they are less relevant to honkyoku of other lineages.

For example, elements of Gutzwiller’s tone cells cannot be observed in many of the phrases in the transcriptions of the ten versions of “Reibo” analyzed in this study. The most conspicuous examples of this are the reibo no te phrases, many of which have no stable kari main note at all (eg., phrase 58 of Watazumi’s “Reibo” and phrase 43 of Yokoyama’s “Reibo”), or consist of very long successions of many meri notes followed by one or two concluding kari notes (eg., phrase 31 of Watazumi’s “Furin”, phrase 33 of Yokoyama’s “Shôganken reibo”, phrase 32 of Iwamoto’s “Furin”, phrase 39 of Jin’s “Shôganken reibo”, and phrase 39 of Sakai’s “Shôganken reibo”). Phrases over twenty notes in length that are entirely meri notes frequently occur (eg., phrase 23 of Yokoyama’s “Shôganken reibo”). Other phrases may have as many as twenty-six meri notes with one or two final kari notes at the very end (eg., phrase 31 of Watazumi’s “Furin”). It becomes extremely difficult to determine in Gutzwiller’s terms what might be called the tone cell in many of the phrases in the “Reibo” samples. The usefulness of the tone cell concept in this case is much less than it is with Kinko pieces familiar to Gutzwiller.

As stated above, it is rare for a performing member of the shakuhachi tradition to transcribe and analyze honkyoku performances. In most of the analyses of transcriptions and transnotations of honkyoku that have been undertaken by others, the analyses are comparative, dealing with the similarities and differences that may exist between versions of the same honkyoku piece or type. Clearly and in marked contrast to the Western musicology tradition, the act of formally analyzing honkyoku with the use of either transcriptions or scores is not an established and integral part of the shakuhachi honkyoku tradition. Musicological methods, nevertheless, facilitate the examination of elements of the honkyoku, for example music contour and formal structures. These elements may be of little interest to honkyoku performer, but a knowledge and understanding of their existence is one kind of insight into the process of transmission.

6.1.3 Analyses of honkyoku scores

In contrast to the few analyses of transcriptions of honkyoku performances in the literature, there are a number of examples of analyses of scores of honkyoku which rely solely upon either scores written in traditional notation or transnotations of such scores from traditional shakuhachi notation to staff notation. In this thesis, I do not, however, rely heavily upon traditional scores or their transnotations as sources of analytical data. Analyses of this type do not address the process of performance in what remains largely an oral tradition. I have outlined elsewhere (Lee 1991:18-35) some of the problems which arise in attempting to interpret honkyoku scores as representative of performances; for example, I draw attention to the numerous discrepancies between what notation symbols mean and how they are realized in performance. Takahashi (see below, p.327) has also examined these problems.

Though the methodology of such analyses have little bearing upon the present study, they illustrate the importance placed on the tracing of lineages and genealogies in the shakuhachi tradition, thus helping to validate the focus of the present study, transmission. The importance given to lineages and genealogies is further emphasized by the amount of literature on the subject, including oral and written genealogical histories.

In the present study, where data is taken from traditional scores (i.e., the labelling and placement of formal divisions of the “Reibo” pieces), interpretation of notation symbols is not required. Furthermore, the data derived from the traditional scores is directly confirmed by transcriptions of performances of the pieces (see p.363).

Of the existing studies which are based upon scores, only one, by Takahashi (see below), addresses the process of performance and deals with the discrepancies between traditional scores and their realization. Other score analyses do not deal with the process of playing honkyoku nor do they attempt to elucidate aspects of the music such as formal structure. Instead, almost all of the analyses of this type attempt to show similarities and differences in the notation which are then explained by similarities and differences in either lineages or historical periods.

Takahashi (1990:295-309) makes use of both transcriptions and scores written in original notations in his analysis of “Honte chôshi” (). He also discusses an area of transmission that brings into question the validity of analyses based upon transnotations, namely, that of the construction of the instrument itself and how it has affected pitch in honkyoku playing. He describes how variations in traditional shakuhachi construction methods, including two systems of placing finger holes, have affected the musical realization of the scores, in particular the phenomenon of tonal ambiguity. The variations in construction, which can be grouped into two methods corresponding to the two major shakuhachi centres of Edo Japan, Edo and Kyôto, result in variations in pitch production.

Takahashi also observes the ambiguity of pitch in traditional notation which is manifested in the performance of honkyoku. He concludes that transnotations into staff notation of the piece Honte chôshi cannot be made by using the data presented in the original written score alone (1990:301), a conclusion which has implications for other transnotations. Traditional honkyoku scores may give only one tablature symbol where in fact there are two pitch possibilities. For example, the finger position (tsu) (only the bottom finger hole open) produces the pitch f on a standard length shakuhachi. Two pitches, e and e-flat, can be produced with the same finger position by using different degrees of the meri technique.14 Yet in some honkyoku scores of Honte chôshi, only one type of tsu meri symbol is used. According to Takahashi, whether to play e or e-flat is part of the orally transmitted tradition.

14. Deviations in pitch up to a perfect fourth can be produced on the shakuhachi with no change in finger positions through what are called the meri/kari techniques. These techniques involve subtle changes in the distance and to a lesser degree the angle from which the air stream leaves the lips of the performer and is deflected by the blowing edge of the instrument. The meri technique lowers the pitch, while the kari technique raises the pitch of any given fingering.

Tukitani’s treatment of two versions of the piece “Tsuru no Sugomori” (, Nesting of the Crane), namely those of Kinko and Tozan lineages (Tukitani 1976:80-87), is an example of total reliance on transnotations of traditional scores for analysis. In an article entitled (Shakuhachi koten ni okeru dômei ikyoku no mondai, “Some Questions on Shakuhachi Pieces of Different Melodies under the Same Titles”), Tukitani looks at three to nine versions of the four pieces “Kyorei”, “Kokû”, “Mukaiji”, and “San’ya”. Segments of transnotations of the versions are used as a basis for comparison. Though some conclusions regarding the transmission of these pieces are made, the problem of notation versus performance is not addressed. Furthermore, the transnotations from traditional scores to staff notation adds a layer of interpretation between the data and the conclusion.

Gutzwiller (1984), writing in German, also presents a comparative analysis of honkyoku using the transnotations of scores written in traditional notation. Three lineages within Kinko ryû are represented in his analysis, those of Araki Kodô (), Miura Kindô (), and Kawase Junsuke (). The piece he compares is the classic “Shin Kyorei” (). As with the above example, the subject of transmission is a focus of the study, but the issue of discrepancies between scores and performances is not addressed.

Another score analysis was published in 1984 by Toya Deiko in his book Komusô shakuhachi shinan (, Teachings of the komusô shakuhachi) (Toya 1984:97-101). Toya gives a brief analysis of scores of Hifumi chô (, One Two Three Searching), in an attempt to show how the piece has changed over time in mode and ornamentation. In contrast to Tukitani and Gutzwiller, whose transnotations insert a layer of interpretation, Toya worked directly from the traditional scores written in shakuhachi notation. Like Tukitani, he does not consider the discrepancies between notation and realization in performance.

Yamaue is a shakuhachi honkyoku performer who amassed a large collection of his own hand-written honkyoku scores representing many lines of transmission, some of which have been published. In the book, Yamaue Getsuzan shûshû shakuhachi: Ôshû hen and kyûshû hen (, Yamaue Getsuzan’s collection of shakuhachi scores; Ôshû and Kyûshû editions) (Yamaue 1984), are a number of comparisons of traditionally notated scores. Yamaue has written out the scores of different versions of the same piece parallel to each other, lining up what he assumes are the same phrases in different versions. In one case, Yamaue (1984:54-59) compares two versions of “Reibo”, that of Jin Nyodô (written in red) and that of Orito Nyogetsu and Takahashi Kûzan (written in black). In a second case, three versions, that of Sakata Tôsui (written in red), Gotô Tôsui (written in black), and Uramoto Setchô (written in red) are lined up together (Yamaue 1984:60-66). The titles of formal divisions of the pieces, such as take shirabe (, bamboo searching) and hachigaeshi (, returning the bowl)(see p.361) are prominently notated.15

15. Relationships between the titles of formal divisions that can be found in traditional scores of “Reibo” pieces are discussed in this study (see pp.360366).

As is the case with many honkyoku performers, Yamaue was particularly fascinated with the issue of transmission, as evidenced not only by the above comparisons, but also his extensive genealogies (see chapter four, p.177). As with the previous examples, however, Yamaue does not address the relationship between scores and performances. A further problem with these comparisons is that all of the scores were originally written by Yamaue, presumably after he had learned the pieces, and consequently may display more similarities than actually exist between the lineages represented. Finally, Yamaue does nothing beyond lining up the notes of the scores; no conclusions are drawn from the collation of the scores.

In all of the above examples of shakuhachi honkyoku score analyses, the main theme is transmission, particularly in terms of lineages and transmission lines. Similarities and differences between individual transmission lines are compared on the level of individual notes. There are no cases of the traditional scores or transnotations being used to formally analyze the music in a western musicological sense. In only one of the examples of analyses of scores (Takahashi) is the relationship between the notation and the process of the performance discussed. The aspect of the honkyoku tradition that is orally transmitted is ignored altogether in studies such as the above examples, which are based entirely upon scores. Though the focus of analyses of this type helps to validate the emphasis placed upon transmission in the present study, the methodology and the conclusions have little relevance to this thesis.

6.2 Tradition-based approaches to honkyoku analyses within the shakuhachi tradition

While musicological approaches to analyzing honkyoku yield insights into such elements as the structural organization of the music and lines of transmission, the analytical approaches that members of the tradition use during the process of transmission yield another kind of insight, which cannot be drawn from musicological methods. In particular, tradition-based approaches to honkyoku analysis increase an understanding of the experiential elements of honkyoku transmission as it occurs within the tradition. More significantly, tradition-based approaches show what the tradition sees as the most important elements of the honkyoku during the process of transmission. It is primarily for this reason that this thesis incorporates a method of analysis based upon traditional methods.

The following translations of excerpts of written tradition-based analyses and discussions of oral analyses are presented in order to evolve strategies that can be used in this study to reveal some of the experiential elements of honkyoku transmission as it occurs within the tradition.

One type of tradition-based analysis is the written descriptions of pieces or parts of pieces which adopt an approach similar to the teaching method a teacher would use during a shakuhachi lesson: explaining the details of performance of a honkyoku one phrase or performance technique at a time. In contrast to analyses of European art music tradition, as well as the majority of ethnomusicological analyses of other music cultures, these types of analyses of the honkyoku are based upon neither transcriptions of performances nor scores of the pieces. Though they may annotate in detail existing scores written in traditional shakuhachi notation, the scores themselves are not what is being analyzed. Instead, the scores are used only as a form of reference, as opposed to an object of analysis, to show which phrase or performing technique is being discussed.16

16. See Lee (1991) for a detailed description of traditional shakuhachi notation and its relationship to performance.

Most importantly, only a shakuhachi performer already able to perform a given honkyoku is able to analyze the piece in such a manner. Furthermore, the function of the analysis is to assist the reader in performing the piece, rather than to enhance his/her intellectual understanding or appreciation of the piece and its form. The emphasis is on the ‘doing’, the experience of performing the piece, rather than the intellectual manipulation of theories to verbally describe the piece.

A prime example of this type of analysis can be found in Ichi On Jôbutsu (; One Sound Becoming Buddha), a periodical catering to shakuhachi honkyoku enthusiasts. In Issue No.5 (June 1, 1983), Okamoto Chikugai () (1983:2-22) examines the method of performing the piece “Ôshû Reibo” as part of an article entitled “About Ôshû Reibo” (, Ôshû reibo ni tsuite). The following translation is taken from the third section of the article, called “The method of blowing Zen for ‘Ôshû reibo'” (, Ôshû reibo no suizenhô). Terms for specific shakuhachi performance techniques will be defined in the translation only if the context warrants it.

This presents a summary of the ‘method of blowing Zen’ (, suizenhô), as transmitted by the shakuhachi player Konashi Kinsui () and inherited by his student Daimon Nyogen() (See genealogy chart, p.192). This offering of instruction is respectfully dedicated to ‘blowing Zen’ (, suizen).

(Bamboo Searching)17

This sound is blown two times.18

The first thing in the act of ‘tuning’ the breath and the heart, according to one way of thinking, is the warming up of the bamboo, which also has the effect of normalizing the melody (, chôshi). This is especially true in the case of long instruments. In the process of playing the instrument the pitch (, onritsu) will become higher, so it is necessary to consider this area.

17. This is the first of three divisions notated in the score used by Okamoto. The three sections are Take shirabe (. ‘Bamboo Searching’ or ‘Bamboo Melody’) Reibo (, ‘Yearning for the Bell’); and Hachi gaeshi (, ‘Returning the Bowl’).
18. (Sono oto fukikomu nikai). The piece is treated as follows: first the tablature notation of one or more phrases is given, and then explanations of the performance of the phrase or phrases are presented. The first phrase of the score, however, is the performance command “This sound is blown twice” rather than tablature notation.

[The second phrase of the piece.]

The fingering position (tsu) is played from a chû meri (, middle meri)19 position of pitch. Using the wedge blowing technique20 until a lingering tone remains, one then moves to the next note, (ro). With this line [in the score, which also includes phrases three and four], unless the spirit (, kiai, literally ‘spirit uniting’) is full or replete, one is justified in repeating this melody any number of times. There is an oral tradition that this melodic figure (, senritsu) must be called the ‘bamboo melody (, take shirabe, literally, ‘bamboo searching’) of the tuning of the breath and heart’.

19. Meri () is a technique which flattens and mutes the sound. Chû meri (, ‘middle’ meri) refers to the degree of flattening.
20. , kusabi buki. An appropriately named blowing technique which begins with the note or phrase played loudly and fully and gradually decreasing in dynamics and energy until the sound disappears. This technique is especially common in certain lineages of the Myôan style of playing.

[Okamoto does not discuss individually the third and fourth phrases of the piece. The fifth phrase is divided into two sections by Okamoto.]

[The fifth phrase of the piece.]

The symbol [of the first section] is the yuri technique (, a bending of the pitch by changing the angle and distance of the lips from the blowing edge) which is executed by revolving the chin as if drawing circles. Also, the tail end of the sound is played meri. The symbol [of the second section of the phrase] indicates a yuri sound which is an exhalation that is alternatively strong-weak-strong-weak, produced while blowing in a manner that is exactly like eating food.

Regarding the yuri sound, when it is written (yuri oto, literally ‘yuri sound’), one shakes one’s neck from left to right as with a boat rolling side to side and pitching up and down. Of course, besides this there are the jôge yuri and the fune no kubi yuri.21

However, it is easy to understand why the yuri of the Tohoku lineage is written with the characters .22 Grains which have been put into a container overflowing with water can be selected out while shaking the container. Likewise, the sound is shaken when playing the technique. It is like the water that falls from the container, and is not the so called (yuri, shake, sway) technique.

These two yuri are given the following expedient names: (mawashi yuri, turning, revolving yuri) and (hira yuri, common or flat yuri).

With the fingering (ru) at the end of the hira yuri of the fingering, the third finger hole is hit. The is written as being chû meri (middle meri). However, with bamboo whose hole placements are determined in the old way of flutes made without ji (, filler)23, the first hole [from the bottom] is lower than on modern instruments. Therefore, [the first hole open] will produce the pitch of chû meri. There is nothing to say otherthan one must listen with discriminating ear.24

21. Jôge yuri: , up and down yuri). Fune no kubi yuri: , literally ‘ship neck yuri’. The fune no kubi is the spar in the very front of the boat.)
22. The character is normally pronounced tô, and has the meaning ‘select’.
23. See pp.274 for a discussion of flutes constructed with and without ji.
24. The point Okamoto is making here is that in aiming for the same pitch, the performer needs different techniques with different flutes.

[The sixth phrase of the piece.]

[Okamoto also divides the sixth phrase into two parts for discussion.]

[The first part of the sixth phrase] is applying the meri technique to the fingering downward a little, returning to the original [pitch] and then moving on to the (re) fingering. With the note , kari and meri are quickly scooped up with the chin, moving on to . Next, the hira yuri of the yuri sounds is played. The (oshi, literally ‘push’) technique is applied to the fourth hole. After this, the oshi technique is also applied to the three hole. In this case, however, a special fingering technique is used. The finger of the three hole is slid up and down so as to produce a sound like the nayashi technique. In the latter part of phrase six, the is given a large nayashi, after which the oshi technique is applied to the four hole. The oshi is the same as before.

(Okamoto 1983:6-7)25

25. See appendix 4 for the original Japanese text of this passage.

Okamoto continues in the same manner, eventually analyzing the entire piece. At the end of the analytical section is a reproduction of the entire score of the “Reibo” under scrutiny.


Four main points of Okamoto’s analysis that are useful in arriving at an analytical method for this thesis can be seen in the above excerpt. Firstly, the importance of the line of transmission to which the described performance methods belong is evident from the initial sentence and elsewhere. Throughout the text, performance details are discussed as they are related to transmission. For example, in his discussion of the fifth phrase, Okamoto explains the manner and meaning of the way the technique yuri is written by a specific line of transmission, the Tohoku lineage.

Secondly, the method of describing the piece in terms of the minute details of the process of playing is central to the analysis, especially the various meri techniques and special finger movements. The emphasis of process can be seen in Okamoto’s method of dealing with pitch, which is important not only as specific frequencies of sound waves, but also in its relationship to the spirit of the performance and performer. In fact, the particular pitches that are produced by the performance techniques are not specified directly.

Thirdly, Okamoto’s analysis of “Reibo” assumes that the reader has a working knowledge of both the symbols and the terminology used in discussing and notating basic shakuhachi performance techniques. His analysis is a commentary directed to insiders to the tradition. It is not a step-by-step ‘how to’ description of the piece. Okamoto does not comment at all on places in the score where he assumes that the notation gives enough information on its own.

Fourthly, Okamoto’s analysis in some instances assumes a knowledge of the piece which only a member of his lineage, or one who has already been taught the piece, might have. For example, the first line of the score is a performance command, which reads, “This sound is blown two times”. This sentence on its own is so ambiguous as to be meaningless. What sound is to be repeated? It could be the first sound of the first phrase, the entire first phrase, or the entire section. The meaning of this phrase cannot be known without explanation, yet Okamoto offers none.

It is important to emphasize the distinction between Okamoto’s use of the score as an adjunct to performance. It is clearly not the object of analysis.

An earlier example of this type of honkyoku analysis is to be found in the writings of Uramoto Setchô,26 noted honkyoku performer of this century (see p.192). Uramoto’s writings are complied and edited by Inagaki and published in a book entitled Master Uramoto Setchô, Who Loved Shakuhachi Honkyoku and old Flutes (Inagaki, ed. 1985). In an article written in 1929 (in Inagaki, ed. 1985:156-170), Uramoto analyzes six pieces which he calls the Take Shirabe (, Bamboo Searching) pieces27. The six pieces are Hifumi chô (, One, Two, Three, Searching); San’ya shirabe (, Three Valleys Searching); Sugomori chôshi (, Nesting of the Cranes Tuning); Yamato chôshi (, Yamato Tuning); Take shirabe (, Bamboo Tuning); and Honte Chôshi (, Original Fingering Tuning).

26. Uramoto is one of six honkyoku players whose performances form the basis for the analysis in this thesis.
27. These pieces are also called Shirabe (, “Searching”) or Chôshi (, “Melody” or “Tuning”) pieces. They frequently act as preludes to larger honkyoku.

In some instances, Uramoto gives little more information than a brief explanation of the piece as transmitted by various lineages, the number of minutes and seconds within which the pieces should be performed, and a sentence or two about the performance practices of several of the phrases of the piece. With other pieces, he gives more elaborate performance instructions similar to those of Okamoto. For example, in his discussion of the piece Yamato chôshi (), he writes:

“An area where care must be taken in the manner of playing is the places where the spaces between phrases are cut off and not connected with a trailing off [of the ending note]. For example, the (ha i ha; producing the pitches ‘c d c’ on a 1.8 shaku flute28) at the end of the second phrase and the (re chi re; ‘g a-flat g’) at the end of the fourth phrase are like this. Among the six Take shirabe [pieces] mentioned herein, this is a unique [melodic] figure. Also, in melodic lines such as (tsu ro – ha ha – ha i ha; ‘e-flat d – c c – c d c) or (tsu – tsu ro ha ha – ha i ha; ‘e-flat – e-flat d c c – c d c’), the (ha; ‘c’) after the (ro; ‘d’) being played brightly as well as strongly are places from which the distinctiveness of this piece comes out.”

(Uramoto, in Inagaki, ed. 1985:164)

28. The pitches produced on a 1.8 shaku flute by playing the fingerings noted by Uramoto here and elsewhere are given in order to facilitate an understanding of what Uramoto is saying.

In his discussion of the Take Shirabe pieces, Uramoto frequently uses the German word ‘thema'(, te-ma) (Inagaki, ed. 1985:157, 160, 163,). He states, for example, that most of the pieces introduce a theme that is presented in the larger honkyoku that is performed afterward. The themes of three of the pieces are identified. For example, in one instance, Uramoto writes, “What must be called ‘the theme’ of the piece, Yamato Chôshi are the two phrases (tsu re – re chi re; ‘e-flat g – g a-flat g’), and (tsu ro – ha ha – ha i ha; ‘e-flat d – c c – c d c’). These phrases are repeated twice, the second time being the variation (tsu – tsu ha ha – ha i – ha i ha i ha i ha ha – i ha; ‘e-flat – e-flat c c – c d – c d c d c d c c – d c’)” (Inagaki, ed. 1985:163). Uramoto does not elaborate on the ‘themes’ that he identifies. He does not use the idea of a theme in any way to describe the structural form of the pieces.

Elsewhere in the same book (Inagaki, ed. 1985:184-188), Uramoto discusses and analyzes the piece Yamato chôshi a second time. He also presents two different scores of the piece, not to give a comparative analysis of those scores, but rather to help illustrate two of the many ways that the piece is performed.

Another example of this kind of written analysis of honkyoku is given by Toya Deiko (, b.1920), who analyzes the piece “Honte Chôshi” (, Original Fingering Tuning) (1984:149-155). As was the case with Okamoto’s analysis, Toya explains in detail how to execute the techniques involved in performing a phrase or section of a phrase, and ignores a few phrases that require no explanation.

A number of elements of Uramoto and Toya’s analyses are useful in establishing an analytical methodology for this thesis. Some of them also reinforce what was observed in Okamoto’s analysis. First of all, the process of performance at a minute level is paramount. Secondly, the similarities and differences between versions of what are considered to be the same piece are important. In Uramoto’s analysis, the elements of dynamics and duration are mentioned. Finally, the phrase or sections of phrases are the primary level of analysis rather than large-scale form or structure.

Similar analyses of honkyoku have appeared in English publications in recent years, and are aimed primarily at the large ‘teach-yourself’ market outside of Japan. Because their readership is, for the most part, novices unfamiliar with most of the standard technical devices which, for example, Okamoto and Toya took for granted, these analyses are more detailed than their Japanese counterparts. Every phrase is discussed, and as much of the performance techniques as possible is described. Two examples of honkyoku analyses of this type in English are by Samuelson (n.d.:31-39) and Grous (1978:40-72-99). Samuelson describes the first phrase of the Kinko piece, “Hifumi hachigaeshi” (, “One Two Three Returning the Bowl”) as follows:

“A. TSU-MERI. Delayed hit attack with #2 [finger]. Hold a bit.
B. ORI-SURI. Bend TSU-MERI pitch down (lower head), then slide up (slide up #1 and #2; #2 is crucial).
C. Come down on RO and hold. (In the movement form the TSU-MERI SURI to RO, #2 actually comes all the way off the hole.)
D. Take a breath.
E. Nayashi. Begin with a delayed hit attack on RO-MERI (#2), then move smoothly up to RO pitch and hold.
F. Ori. Cut-off at end (head down).”

(Samuelson n.d.:36)

An innovation introduced by Samuelson is the graphic display of pitches and pitch changes by the up and down placement and movements of a single line, with timings indicated as the line progresses horizontally left to right (see figure 18). The English language analyses are like the Japanese language analyses in their focus on the process of performance at the level of detailed technique, and in the importance placed on the phrase in organizing the analysis.


Analyses of shakuhachi honkyoku found within the tradition, whether written in Japanese or in English, deal with the performance of the piece or pieces in question. Attempts to examine honkyoku as an independent musical entity, separate from the act of performance, as typically occurs in the analysis of European art music, are rare in traditional shakuhachi contexts.

The analyses described above can be characterized in the following ways. They emphasize the transmission of the piece, both in terms of the lineage and the act of transmission from the writer to the reader of the analysis. They do not analyze scores of the pieces as separate from the performance of the pieces. They focus upon technical details involved in performing the piece, in particular finger movements and the meri technique. Most of the analyses also examine the similarities and differences between pieces that are considered versions of the same piece or family of pieces. Most importantly, they all analyze the pieces from the standpoint of a performer. None of their analyses would be possible without the knowledge and experience of one who performs the pieces. As will become evident later (pp.370418), all of the above elements are useful in evolving a methodology of analysis appropriate to this thesis.

6.2.1 Informal oral analyses

Another type of tradition-based analytical procedure can be seen in the shakuhachi honkyoku tradition. Within my experience, members of the shakuhachi tradition frequently analyze orally the similarities and differences between performance practices of various performers in general, as well as individual performances. These verbal analyses are similar to the ‘blow by blow’ description of the piece “Ôshû reibo” given by Okamoto and partially translated above. The manner of execution of particular phrases or, even more commonly, performance techniques encompassing less than a single phrase is the topic of discussion. As with most of the written analyses discussed above, the emphasis is on details. Rarely have I encountered discussions of formal structures of either honkyoku or phrases within them.

The following is a description of one such analytical discussion which illustrates the above. One of the most evocative and frequently performed honkyoku is the Kinko lineage version of Shika no Tône (, Distant Sound of Deer, a duet). My first teacher, Chikuho II, taught me the piece as transmitted by the Myôan lineage, which is very different from the version transmitted through the Kinko lineage. Eventually, I learned the Kinko version with the help of shakuhachi playing friends who were of the Kinko lineage, and from numerous available recordings by various performers.29

29. See p. 247 for a discussion of the ways in which readily available recordings have become part of the transmission process of the honkyoku tradition over the past several decades.

About ten years ago, I performed in concert the Kinko version with Osaka based Tajima Tadashi (, b.1942). At that time Tajima said that it was interesting how I had incorporated performance practices and techniques from Kinko performers who represented a number of distinct sub-lineages of the Kinko lineage (see p.251). These performers each had unique, recognizable ways of playing certain phrases of the piece. While each individual’s performance practices might change with every performance, those variations are usually much less than the differences between performers representing various sub-lineages.30

30. Note the emphasis Tajima places upon lineages and transmission in his interest in my performance of “Shika no tône”.

Tajima proceeded to informally analyze my performance. He said, for example, that he recognized a Yamaguchi Gorô influence when I entered one phrase in the uppermost register from an almost imperceptibly quiet and smooth pianissimo. Other performers attack the note more loudly. Elsewhere, Tajima recognized in my playing the powerful breathiness combined with a certain pattern of pitch bending for which Yokoyama Katsuya is noted when performing that particular phrase. Finally, in a series of phrases that are said to represent the guttural coughing of the deer, Tajima said that he detected in my playing the embellishment techniques which he associates with Aoki Reibo. This kind of informal, oral analysis or critique is fairly common amongst honkyoku performers.

It appeared that where differences occurred between performers of various sub-lineages of the Kinko lineage, I had, in part unconsciously, incorporated the performance practices which most appealed to me. My eclectic version of the piece was ultimately validated amongst members of the honkyoku tradition in 1989, when I was requested to record the piece with Yokoyama on a CD released by the prestigious Ongaku no Tomo Sha label (Yokoyama 1989b) thereby establishing my version as an orthodox one within the tradition. Tajima’s analysis of my performance of “Shika no tône” is especially relevant to the theme of transmission since it shows how a specific version of a piece came to happen.

Informal oral analyses such as the above example are similar to the written analyses discussed earlier in the following ways: they pay particular attention to lineage; they treat the music from the viewpoint of an experienced performer rather than a listener, however knowledgeable; they deal with details of performance at the level of the phrase or, more commonly, at a level of performance technique which constitutes only a part of a phrase; they consider pitch, timbre, dynamics, duration, and ornamentation as elements of the performance; finally they look at the similarities and differences between versions of the same piece.


The following is a summary of the main points gleaned from both the written and oral tradition-based analyses discussed above:

  1. The importance of the transmission is stressed, both in terms of the lineage and the act of transmission from the writer to the reader of the analysis. Performance details are discussed as they are related to transmission.
  2. The method of describing the piece in terms of the minute details of the process of playing is central to the analysis, especially the various meri techniques and special finger movements.
  3. The traditional analyses assume that the reader has a working knowledge of both the symbols and the terminology used in discussing and notating basic shakuhachi performance techniques. The pieces are analyzed from the standpoint of a performer. Not all of the phrases are analyzed.
  4. In some instances it is assumed that the reader possesses a knowledge of the piece which only a member of the lineage might have.
  5. The honkyoku is not examined as an independent musical entity, separate from the act of performance.
  6. Many of the analyses also examine the similarities and differences between pieces that are considered versions of the same piece or family of pieces.

These approaches will be incorporated as much as possible in the analytical methods used in the present study of honkyoku. Regarding points three and four, however, it is not practical to assume that the readers of this thesis have a working knowledge of the symbols and terminology used in the honkyoku tradition, or that they possess the knowledge of members of particular shakuhachi lineages. Explanations and annotations will be given where deemed necessary. Nonetheless, care will be taken to examine honkyoku as much as possible from the knowledgeable performer’s point of view.

The preoccupation with comparing pieces which share similar names and/or melodic material seen in the examples of analyses presented above, and of which this thesis is also an example, may be attempts to define the honkyoku itself. It is indicative of the elusive quality of honkyoku, which is changing with every performance and whose form is vagueness. It is appropriate that the analysis in this thesis is primarily a comparative one. In the following section, versions of the piece “Reibo” will be analyzed in a number of ways that will take into consideration most of the elements which have been observed in the analyses discussed above.

6.3 Methodology of transcription and analysis

As stated above, the analyses in this thesis compare the transcriptions of ten recordings of single performances of honkyoku whose relationships have been determined by genealogies based upon oral and written histories of honkyoku transmissions. Two methods of comparative analysis have been devised, a musicologically inclined method and a more tradition-based method.

First of all, the ten recordings are transcribed in detail, in a way that allows ‘outsiders’ to relate them to the recordings. The transcriptions attempt to embody the process of performing the pieces as well as the product of performing the notes. In particular, the processes involved in producing the meri and kari notes (see pp.269270) are incorporated in the transcriptions through a system using colour as a code (see below, p.353). The meri and kari notes are significant in terms of both pitch and timbre. They are also embedded with philosophical meaning.

Secondly, the transcriptions are orthographically simplified by replacing frequently occurring musical elements with signs or symbols and by omitting certain elements such as embellishments used to articulate or re-articulate notes. For the purpose of comparision, these orthographically simplified transcriptions are arranged in such a way that corresponding material is lined up according to principles outlined below (p.359), in order to show similarites and differences between the transcriptions. Once the individual phrases and notes are lined up, the transcriptions are compared on a number of levels.

On the basis of these comparisons, observations can be made between specific performers representing a number of lines of transmission. The central questions to be addressed in the analysis are: do the transcriptions of performances support or refute the lines of transmission shown in a lineage chart based upon evidence such as oral histories and genealogies; and in what ways do the lines of transmission manifest themselves in the analysis of transcriptions of recordings by performers on the chart?

Comparisons will be made between a number of categories: between the Futaiken and Shôganken groups of pieces, between lineages, and between individual performers. Comparisons will be made on the basis of elements such as large scale structures and melodic contours (as encapsulated in orthographically simplified versions of the transcriptions), distinctive melodic, cadential formulae, and finally, details of the performances.

Patterns of transmission between the pieces represented in the ten performances are thus brought to light. These patterns occur in the ways in which musical elements are retained, omitted, or changed during transmission according to the performer, the lineage, and/or the group. At the highest level of detail, the largely musicological methods applied to the first levels of analysis become less appropriate. The second method of comparative analysis applied to the details of the performance has been developed from tradition based models of describing honkyoku. This second method is similar to the way in which the phrases might be explained in a traditional teaching situation. It entails examining the process of performing the details of the pieces as they occur phrase by phrase.

I will show, largely by musicologically-based methods, that large scale structural analysis of the transcriptions of the performances of ten honkyoku examined in this thesis confirms the genealogy constructed in Chapter Four: that the ten performances are essentially manifestations of what is conceptualized as “Reibo”, but that the ten performances can be divided into two groups which are in turn transmitted by basically two lineages or lines. In some cases, this commonality between the ten pieces has been forgotten or overlooked by some members of the honkyoku tradition. It is appropriate to use non-traditional methods to establish their shared identity.

The tradition-based method of analysis facilitates the observation of the elements of honkyoku most important to the members of the tradition. Meaningful comparisons of the “Reibo” pieces using this second method can be made, however, once the identity of the pieces has been established by the first method of analysis outlined above.

Before either the musicologically based or the tradition based comparative analysis can be made, it is necessary to explain the method of acquiring and organizing the analytical data. The following discussions explain the methods of transcription and how the transcriptions are used as data for comparison, afterwhich the analyses of the data will be presented.

6.3.1 The recordings

The data used to analyze the “Reibo” pieces at the three levels of the first, musicologically based method outlined above are derived from transcriptions of recordings of performances of the pieces. Both philosophical and practical issues pertaining to the transcribing of honkyoku have been discussed above (p.310315). A detailed examination of the transcription methods used in this thesis is undertaken below. Transcriptions of recordings of six performers, Uramoto, Watazumi, Yokoyama, Iwamoto, Jin, and Sakai playing pieces known as “Reibo”, “Futaiken den reibo”, “Futaiken reibo”, “Shôganken den reibo”, “Shôganken reibo”, and “Furin” form the basis for the analysis in this thesis.

According to the lineage chart constructed in chapter four, all of these pieces, with the exception of Watazumi’s “Furin”, can be traced back to one player, Hasegawa Tôgaku of the Ôshû district of northern Japan, who performed a piece which may have been called “Reibo”. Watazumi’s “Reibo” is also an exception because he does not acknowledge Uramoto as his teacher, although the genealogy chart indicates taht Uramoto taught Watazumi the piece. Implications of Watazumi’s exceptional position are manifested in many ways in the analysis (pp.404418).

As explained in chapter four, the genealogy tracing these pieces and their exponents is based upon reliable data from within the shakuhachi tradition. In the discussion that follows, I will show that this genealogy, including Watazumi’s “Furin” and “Reibo”, is confirmed by the analysis. Through the analysis, I will further argue that some of the specific lines of transmission recorded in the lineage chart are reflected in many aspects of the performances.

Chart 7

Genealogy Chart 7
(click to enlarge)

Performers whose recordings are transcribed are marked “(T)”.

The ten recordings represent two groups of pieces and two lineages of honkyoku performers. Groups “A” and”B” are two distinct versions the piece “Reibo” of the Ôshû district. These two versions or groups of pieces will be referred to in the analysis as the “Futaiken reibo” group and the “Shôganken reibo” group in accordance with the findings of the genealogy. The transmission lines of [Konashi] – Uramoto – Watazumi – Yokoyama31 and [Konashi] – Jin – Sakai32 are represented in group “A”. The transmission lines of Watazumi-Yokoyama-Iwamoto and Jin-Sakai are represented in group “B”. Chart 7 gives the genealogy of the ten performances in question. The details of the ten recordings are presented below:

31. As discussed above (p.349), Watazumi refutes the suggestion made by many members of the tradition that Uramoto taught him the piece “Reibo”. The analysis to follow bears out, to a certain degree, his claim that Uramoto was not his ‘teacher’, as well as the seemingly contradictory belief that the piece “Reibo” was transmitted from Uramoto to Watazumi.
32. Jin did not teach Sakai directly, but rather transmitted both “Futaiken reibo” and “Shôganken reibo” through Sakai’s immediate teacher, Moriyasu Nyotô (). Regrettably, recordings of Moriyasu do not appear to exist.
Recordings in Group “A”
1. Uramoto Setchô “Reibo”
TRS-5084 Privately released LP recording.
2. Watazumi dôso “Reibo”
Crown SW-5006. LP recording.
3. Yokoyama Katsuya “Reibo”
Ongaku Tomo no Sha OMC1912.
Cassette recording.
4. Jin Nyodô “Futaiken den ‘Reibo'”
Teichiku GM6007. LP recording.
5. Sakai Chikuho II “Futaiken reibo”
Adamu Êsu AAC-2001. LP recording.


Recordings in Group “B”
6. Watazumi dôso “Furin”
Phillips PH-7503.
LP recording.
7. Yokoyama Katsuya “Shôganken reibo”
Ongaku Tomo no Sha OMC1912.
Cassette recording.
8. Iwamoto Yoshikazu “Furin”
NHK (Japan National Radio) 17 November.
Radio broadcast.
9. Jin Nyodô “Shôganken den ‘Reibo'”
Teichiku GM6007. LP recording.
10. Sakai Chikuho II “Shôganken reibo”
Columbia KX7002. LP recording.

The recordings are part of a collection of twenty-nine recordings of “Reibo” pieces from the Ôshu district in my possession.33 The ten recordings chosen for analysis from this group were selected because they represent two lineages over two or three generations which have each transmitted two versions of the “Reibo” piece. Another major consideration in their selection is my being the recipient of both the two groups of pieces and the two lineages of performers.34 Besides the obvious personal interest and satisfaction I stand gain to from such a choice, I would argue that my intimate understanding as a performer of each of the two lineages and two groups is not only desirable but imperative in order to successfully analyze them (see p.369). This is particularly important for the second tradition-based method of analysis. The necessity to rely in part upon one’s knowledge as a performer for the musicologically based method of analysis will also become clear.

33. Many of the recordings were collected in Japan in 1988-1989, with the assistance of Ms. Tukitani and the “Syakuhati kenkyûkai” (, Shakuhachi Research Group).
34. Sakai taught me his versions of what he called “Shôganken reibo” and Futaiken reibo” between the years 1973 and 1975, and Yokoyama taught me his versions of what he called “Shôganken reibo” and “Reibo” in 1989.

Throughout this thesis, the importance of the experience of performing honkyoku has been emphasized. If a transcription is to provide accurate data about a honkyoku, it must represent the process of the performance as well as the sound product. Such a transcription demands an intimate knowledge possessed only by one who can perform the piece.

6.3.2 Transcription method

The transcriptions of the ten recordings of the Ôshû family of the piece “Reibo” attempt to embody the process of performing the pieces as well as the notes produced, in particular the processes involved in playing meri and kari notes (see p.265). There are, for example, at least five different combinations of fingerings and meri/kari techniques all of which produce the pitch g’ (on a standard length 1.8 shaku flute) (table 1). Each of the combinations produce sounds that are the same pitch but that are distinct in timbre and other qualities such as the sound produced when they are re-articulated.

In notation systems such as western staff notation, which denote pitch and rhythm but in which timbre and fingering technique are not integral to the system, the differences between the five playing methods mentioned above are not easily represented. While it might be possible to notate some of the playing methods with the use of double sharps, double flats, and a combination of sharps and flats, the result would be unclear and misleading in terms of tonality. Figure 19 illustrates some of the problems of staff notation using only sharps and flats in attempting to transcribe passages of honkyoku that contain meri/kari techniques.

In the transcriptions in this thesis, the meri/kari processes are reflected in the notation by the use of coloured note heads; blue note heads are meri notes and orange note heads are kari notes. The important distinction is immediately apparent in the transcription between notes that are identical in pitch but that are produced with contrasting meri/kari techniques. Figure 20 shows transcriptions of the preceding examples, but with the use of colour to indicate the meri/kari processes. Not only is the transcription more readable, but one aspect of the process of performance, the meri/kari techniques, becomes readily apparent.

It is important to point out that the choice of colours, blue for meri and orange for kari, is consistent with the philosophical symbolism of the yin/yang duality underlying the meri/kari techniques. As explained elsewhere (p.269), meri notes represent yin (earth, night, darkness, female, softness), while kari notes represent yang (heaven, day, light, male, hardness). The colour blue is appropriate for representing the meri (yin) notes, while the colour orange is appropriate for representing the kari (yang) notes.

The second major weakness of staff notation in transcribing performances of honkyoku is that of rhythm. The precise rhythmical markings based upon metrical beats of the staff notation system are inappropriate for notating the ‘free’ rhythms of indeterminate meter of classical honkyoku. In the transcriptions in this thesis, durations are indicated both spatially and by using modified rhythmical markings of staff notation. Durations are indicated linearly on the staff with one centimetre representing one second in time. Accordingly, a note that is held eight seconds, is separated from the subsequent note by eight centimetres. One staff line, being about sixteen centimetres in length, represents approximately sixteen seconds of real time performance, while one page of transcription represents a little over two minutes of performance. The accuracy of this method of representing duration is obviously not exceedingly high, especially in the case of notes of very short durations. It does, however, give an immediate and easily understood indication of the relative lengths of notes and phrases.

The spatial representation is reinforced by the use of the standard rhythmical markings of staff notation, which have been arbitrarily assigned the following values. Semibreves are used to indicate notes with durations of over four seconds. Minims indicate durations between two and four seconds. Crotchets indicate durations between one and two seconds. Quavers indicate durations between one and one-half seconds and semiquavers indicate durations of less than a half second.

The modified rhythmical markings, which make no differentiation between a note held four seconds and one held fourteen, are less precise than the spatial representation of durations, but only when judged by western musicological criteria. Their very imprecision in fact mirrors the ‘insider’s’ way of looking at duration: in the performer’s mind, a note with a ‘long duration’ is held a ‘long time’, not ‘four seconds’ or ‘eight seconds’. How long the note ends up being held depends upon the circumstances of the individual performer and performance. The two complementary systems of notating duration in the transcriptions allow for both an ‘objective’ musicological and a ‘subjective’ insider way of viewing the music.

The performers of the ten recordings used a variety of lengths of shakuhachi. In order to facilitate the comparison of the performances, the transcriptions are transposed so that the fundamental pitch (the lowest pitch produced with all finger holes closed) of all instruments equals the pitch D above middle C. Honkyoku, when performed on a standard length 1.8 shaku, employ primarily the miyako bushi mode, consisting of the pitches D, E-flat, G, A-flat, and C, or the pitches D, E-flat, G, A, and B-flat. In other words, when the pitches D, G, and C occur, they are almost always natural, and when E and B occur they are almost always flattened. The pitch A occurs both as A-flat and A-natural.

The consistent flattening of the pitches E and B suggests the use of a key signature in order to eliminate the majority of accidentals that would otherwise occur in the transcriptions. Therefore, the two flats, E-flat and B-flat, appear in the key signature of the transcriptions, but not in the order traditionally assigned to them. Writing the E-flat before the B-flat is my attempt to avoid the standard interpretation of key signatures and their harmonic implications. Finally, A-flat does not appear in the key signature even though the pitch A-flat frequently occurs in the transcriptions, because A-natural also appears with regularity. In contrast, E-natural and B-natural are almost non-existent.

The transcriptions have also accommodated the honkyoku performing tradition by not notating any ‘grace note’ caused by the standard technique used in shakuhachi playing to repeat or re-articulate a note. In performing traditional shakuhachi music, this technique replaces tonguing to repeat a note.35 The technique is nearly identical to what is used in the performance of bagpipes (where tonguing is impossible), and involves the rapid opening and closing, or hitting, of a finger hole, producing an extremely brief sound in between the notes being repeated. Although these sounds could be considered grace notes and notated as such in the staff notation system, they are neither conceived of nor heard as grace notes in the honkyoku tradition. Non-standard techniques used to repeat notes are noted in the transcription.

35. Tonguing is used only in playing certain modern pieces dating from this century.

In the course of performing, the finger hole opened and closed, or hit, is determined according to the note being repeated. This standard system is one of the first things a shakuhachi player learns as a beginner. The standard repeats are neither notated nor discussed in traditional shakuhachi notation. They are not considered ‘notes’. For these reasons, and in order to make the transcriptions more legible, the standard re-articulations are not notated.

In honkyoku performance, however, there are alternative re-articulations. Finger holes other than those used for standard re-articulations are opened and closed, or hit, or combinations of techniques are used to repeat notes. In contrast to the standard re-articulations, which pass without comment in a lesson, these alternative re-articulations are emphasized in honkyoku lessons and are frequently notated in traditional scores. Sounds produced by re-articulations other than those that are standard will therefore be notated in the transcriptions.

Likewise, articulations used to stress the beginning of notes are executed by striking or lifting one or more finger holes. Again, tonguing is never used in honkyoku performances to articulate a note). These articuations are common, but unlike re-articulations, there are no standard methods of articulating notes which are internalized by the beginner shakuhachi player. These articulations are notated in the transcriptions.

With the above minor changes to the staff notation system, it is possible to transcribe honkyoku in such a way that the important features of the music are retained in a form that is relatively easily and widely understood. The transcriptions of the recordings of the ten performances mentioned above can be found in appendix 1.

6.3.3 The orthographically simplified transcription

Once the transcriptions are completed, the second step towards a comparative analysis is the simplification of the transcription for the purpose of comparison. This simplification allows the phrases and the notes within the phrases to be lined up. In this thesis, the transcriptions are simplified by a systematic replacement of frequently occuring musical elements with signs or symbols, and by the elimination of details that are not musically significant. It is important to note that the result is a simplified orthography that facilitates comparison but does not eliminate significant detail.

Many of the details of the original transcriptions are utilized in the final analysis, but tend to hide the elements being looked for at this level. The simplified transcription is derived by making the following changes in the original transcriptions:

  1. First, the embellishments produced by articulation techniques, and by standard and non-standard re-articulations techniques are omitted. These embellishments are not musically significant at this stage of analysis (figure 21a).
  2. Rapid oscillations between pitches are notated with the following notation symbol: (figure 21b).
  3. Turns, inverted turns, and similar ornaments are notated with the following sign (figure 21c).
  4. Repetitions are notated with symbols where possible (figure 21d).
  5. Linear or spatial enhancement shown in the transcriptions of duration symbols are omitted. The duration symbols themselves, however, i.e., quavers, semibreves, etc., are retained.

Coloured note heads indicating instances of meri and kari techniques are also retained in the melodic contour. With the above changes in the transcribed score, the broad melodic contour of the pieces can be more readily observed and the pieces in question compared.36 Similarities and differences between the transcriptions on a macro level are used to draw conclusions about the large scale structure of the “Reibo” pieces, about the degree of change and the plausibility of the claim of shared origins between pieces. It will be seen that both similarities and differences between the transcriptions of the ten performances support the findings of the research into lineages as shown in the genealogy chart. Furthermore, the location of similar and different musical material in the transcriptions validates the claims by many members of the shakuhachi tradition that the transmission of honkyoku in general and the Ôshû “Reibo” pieces in particular is largely oral in nature. The orthographically simplified transcriptions of the ten performances of “Reibo” are found in appendix 2.

36. See pp.569594 for the othrographically simplified transcriptions of the “Reibo” pieces.

6.3.4 Lining up the orthographically simplified transcriptions

Once the ten transcriptions representing the recorded performances are simplified orthographically, attempts can be made to arrange the transcriptions so that a location in one transcription can be compared with a corresponding location in another transcription. The lineage chart discussed in chapter four (p.192) suggests the possibility of arranging the transcriptions so that one might assumme that phrases in one transcription are lined up with the corresponding phrases in the other transcriptions. A successful attempt to line up most of the data provided by the transcriptions would largely validate, independently of any data used to create the lineage chart, the conclusion of the lineage chart that the ten versions all share a common identity, which in this case can be traced back to the “Reibo” as transmitted by the komusô Hasegawa Tôgaku.

The musical material of the ten transcriptions at first seemed hopelessly confusing in both the degree of divergence and the repetition of material. The method of organization I found most successful relies heavily upon my practical knowledge of the pieces as a recipient of four of the versions.

The first step toward arranging the simplified transcriptions in the manner described above is to look at those traditional scores which label the placement of formal divisions. Beginnings and ends of formal sections are marked in the traditional scores of three of the pieces in one group and two of the pieces in the other group. These traditional markings can be used to determine the beginnings and ends of the formal sections in all ten of the pieces, which in turn can become the starting point from which to line up the transcriptions.

In the traditionally notated scores of “Shôganken reibo” as transmitted by Jin, Sakai, and Yokoyama (transcribed by Furuya Teruo ), six sections are clearly marked and named. The names of these sections, which are identical in the three versions are: ‘Take shirabe’ [] (bamboo searching), Honte’ [] (original pattern), ‘Takane’ [] (high sound), ‘Takane gaeshi’[] (repeating the high sound), ‘Hachi gaeshi’ [] (returning the bowl), and ‘Musubi’ [] (closing).

Tukitani describes the six sections as follows:

“‘Take shirabe’ [] (bamboo searching): a sedate introductory section, suggesting the atmosphere of the entire piece, and introducing certain motifs.

‘Honte’ [] (original pattern): the section presenting the main motifs of the piece such as the Reibo theme. This theme varies slightly with each piece.

‘Takane’ [] (high sound): a beautiful melodic line beginning in and descending from the highest register, followed by a repetition of the main theme of the honte section.

‘Takane gaeshi’[] (high sound returning): a repeat of the takane section, with a variation of the honte section.

Here, the komusô would return the bowl, having received the alms, and perform the following section as a sign of gratitude.37

‘Hachi gaeshi’ [] (returning the bowl): a transposition38 of the theme, using the tamane [] (ball-like sound) technique39. This section is a ‘change’ within the piece as a whole.

‘Musubi’ [] (closing): a short concluding section, played in the high register, though many variants finish in the low register.”

(Tukitani 1990b:15)

37. Tukitani here refers to the practice of mendicancy performed by komusô. The most common form of alms received by the mendicant priests was uncooked rice, brought out in a bowl (hachi). The priest, stopping in the middle of performing the honkyoku, would empty the bowl of rice into his collection bag, and then return the bowl to his benefactor. The komusô would then continue playing, performing the section, “Returning the bowl”.
38. Tukitani is not clear in her use of the word “transposition” (, tenchô) here. In musical terms, the word can also mean ‘modulation’ and ‘transition’.
39. The tamane technique is similar to flutter tonguing, except that the uvula is ‘fluttered’ instead of the tongue. The result is a softer sound.

The location of these divisions in the transcriptions corresponding to the scores that indicate them is facilitated by the close correlation between the performances and the corresponding scores. For example, the number of phrases notated in the score used by Sakai is identical to the number of phrases in the transcription of Sakai’s performance.

Once the divisions are located in the transcriptions of the three “Shôganken reibo” pieces performed by those using traditional notation with the divisions labelled, it is possible to look for the formal divisions in the transcriptions of the remaining two performances included in the “Shôganken reibo” group, those of Watazumi and Iwamoto.40 On the basis of the conclusions made upon examining the endings and beginnings of the divisions in the above three transcriptions, one can determine where the divisions exist in the other two pieces. The remarkable extent of agreement between these two transcriptions and the first three transcriptions above, including cadential formulae shared by all five transcriptions, is sufficient to locate the formal divisions in the latter two transcriptions with relative certainty. The cadential formulae will be discussed below (pp.379382)

40. Though Watazumi and Iwamoto use the title “Furin” for their pieces, they are initially included in the “Shôganken reibo” group. Their inclusion is based on the oral and written information used to assemble the genealogy chart (see chapter four). Data derived from the transcriptions, beginning with the initial and final phrases of the formal divisions, discussed here, validate their placement in the “Shôganken reibo” group.

Figure 22 shows parts of the transcriptions of the five performances of the “Shôganken reibo” pieces, which correspond to segments of the initial and final phrases of the formal divisions of the piece, whose location has been substantiated by the three traditional scores of “Shôganken reibo” mentioned above. In the transcriptions of all five performances, all but one of the sections ends on the pitch D, which is the fundamental note of the instrument (on a 1.8 shaku length flute), that is one of the primary core tones (kakuon) in the honkyoku (see p.316). The final musubi section is the exception. In the case of the ending of the musubi section (which corresponds to the end of the piece), there is likewise a sustained D in all five transcriptions. In the three transcriptions representing the Watazumi lineage (Watazumi, Yokoyama, and Iwamoto), however, a final phrase, which ends on the pitch A and which is not present in the Jin-Sakai transcriptions, is added after the phrase with the sustained D. The implications of this difference will be addressed below (p.378).

For the most part, the formal divisions, as notated in the traditional scores, substantiate the hypothesis that the five transcriptions represent pieces that are closely related, if not the same. The similarity of material that can be found at the beginnings and endings of what are believed to be the formal divisions clearly establishs their location in all five transcriptions.

Having established the location of the formal divisions in the five transcriptions of the “Shôganken reibo” group, the same process can be applied to the “Futaiken reibo” group of pieces.41 In contrast to the case of “Shôganken reibo”, the formal divisions of the piece “Futaiken reibo” are labelled only in the traditional scores of Jin and Sakai. Furthermore, there are only five sections instead of six. The five sections of “Futaiken reibo” as found in the scores of Jin and Sakai are ‘take shirabe’, ‘honte’, ‘takane’, ‘hachi gaeshi’, and ‘musubi’. As in the case of “Shôganken reibo”, the location of these divisions in the transcriptions of Jin and Sakai’s performances and the corresponding locations in the transcriptions of the remaining three performances can be pinpointed with relative certainty due to the high degree of similarity between the five transcriptions.42 Figure 23 shows the initial and final phrase segments of the divisions in the five “Futaiken reibo” transcriptions as marked in traditional scores.

41. The five pieces grouped under the name “Futaiken reibo” are Jin’s “Futaiken den reibo”, Sakai’s “Futaiken reibo” and Watazumi, Yokoyama, and Iwamoto’s “Reibo”. The grouping together of these five pieces is initially dictated by the written and oral information used to derive the “Reibo” lineage chart (see chapter four), and, as with the “Shôganken reibo” group, data from transcriptions validate the grouping.
42. Though the label takane gaeshi does not occur in either of the traditional scores of “Futaiken reibo”, musical material that corresponds to varying degrees to the takane gaeshi sections of the “Shôganken reibo” pieces can be found in all five versions of the “Futaiken reibo” transcriptions. Discussion of this will be deferred until later (see p.374)

Similar divisional markings can be found in traditional scores representing other lineages of “Reibo” pieces, although frequently with different titles. Scores of eight versions of the “Reibo” piece are included in the published honkyoku scores of Yamaue (Yamaue 1984). They exhibit the following variations from the above division of formal sections:

  1. The piece “Reibo” of Futaiken (, a sub-temple), as transmitted by Sakata Tôsui () through Sanô Tôkai (), is divided into take shirabe, ichi no takane (high sound 1), ni no takane (high sound 2), san no takane (high sound 3).
  2. The piece “Reibo” of Futaiken, as transmitted by Konashi Kinsui () through Goto Tôsui (), is divided into take shirabe, taki otoshi (, falling waterfall), take ne (taki otoshi no uchi) (high sound within the falling waterfall [section]), hachigaeshi, and tsuyu kiri (, literally ‘cutting the dew’43).
  3. The piece “Reibo” of Futaiken, (also called “Miyagino Reibo” ) as transmitted by Konashi Kinsui through Kobayashi Shizan (), is divided into take shirabe, taki otoshi, takane, and hachigaeshi.
  4. The piece “Reibo” of Futaiken, as transmitted by Konashi Kinsui through Uramoto Setchô () and then Tanikita Muchiku (), is divided into take shirabe, taki otoshi, ichi no takane, ni no takane, and hachigaeshi.
  5. The piece “Reibo” of Kinjôji (, a temple) and Garyôken (, a sub-temple), as transmitted by Hasegawa Tôgaku () through Onodera Genkichi () and then Orito Nyogetsu (), and also transmitted by Takahashi Kûsan (), is divided into take shirabe, reibo (honte), takane, honkyoku (main piece), and hachigaeshi.
  6. The piece “Reibo” of Shôganken (a sub-temple), as transmitted by Jin Nyodô () is divided into take shirabe, honte (reibo), takane, honkyoku, and hachigaeshi.
  7. The piece “Ôshû reibo” as transmitted by Tsuruta Nandô () has no formal divisions.
  8. The piece “Bôno reibo” () is divided into an unnamed introductory section, takane, a second takane, and an unnamed concluding section (Yamaue 1984:12-35).

Finally., in the book, Kimpû ryû shakuhachi honkyoku den (, Transmission of Kimpû ryû shakuhachi honkyoku) (Uchiyama 1972:77-78), the score of the piece “Miyagi reibo” () as transmitted by Onodera Genkichi is given as divided into shirabe, reibo, honkyoku kaeshi, and hachigaeshi.

43. The term tsuyu kiri is occasionally used to denote the final section of a honkyoku. The ‘dew’ refers to the condensation that forms inside the shakuhachi while playing.

The majority of the above names of formal divisions of “Reibo” pieces coincide with those of the “Shôganken reibo” and “Futaiken reibo” groups examined earlier. Table 2 summarizes the sections as found in the ten versions of the “Reibo” pieces studied in this thesis and comparing them with the sections named in the nine traditional scores mentioned above. Similarities between the formal divisions are apparent, despite differences in naming the divisions. In particular the use of the names “Reibo”, “Honte” and “Taki otoshi” to indicate the section after the “Shirabe” section, further substantiate the hypothesis that they are all pieces belonging to the “Reibo” family.

Once the beginnings and endings of the various sections of all ten transcriptions have been located, it is possible to begin organizing for comparative analysis, the remaining portions of the transcriptions, that is, the material between teh beginning and ending of each division. Immediately apparent are the notes that are core tones (see p.316), that is, the pitches D, G, A, and C. A number of these core tones are sustained for over two seconds. Figure 24 shows a page of the lined up orthographically simplified transcriptions with a box drawn around all of the core tones held for two seconds or more.

In a few cases, especially in passages in the upper register, there are sustained notes in the different transcriptions that are also easily lined up. These notes, however, being the pitches A-flat, B-flat and E-flat, are not core tones. These pitches, which are usually performed with the meri technique, are almost always played using the kari technique when in the context of obvious correlation.


After the beginnings and ends of formal sections and the core tones are lined up, the remaining intervening notes can then be scrutinized. In many cases, distinctive melodic formulae can be used to line up the transcriptions. These distinctive features or melodic formulae act as the “signposts” which are known to operate in remembering oral music (Treitler 1974:344-345) (see p.220). The most obvious melodic formula is the reibo no te (, the ‘”Reibo” pattern’ or ‘fingering’).

The term reibo no te is used by members of the honkyoku tradition in talking about “Reibo” in the context of lessons and elsewhere. The reibo no te is a pattern of notes which may be less than a phrase, an entire phrase or more than one phrase. It is always in the upper octave range (, kan), is predominately performed with the meri technique, is usually characterized by a density of notes far greater than the average, and has elaborate ornamentation.

The pattern of notes centres primarily around the pitch g” and to a lesser degree a-flat”, and frequently concludes with the pitches e-flat” and d”. One difficulty in defining a phrase as reibo no te is the frequent occurrence of what could be called fragments of reibo no te, e.g., without the e-flat” – d”, often within phrases that contain non-reibo no te material as well (see the latter half of S:Y39)44. In this study, material will be termed as reibo no te only in places that were so defined to me.

44. In the analysis, specific phrases of individual performers are denoted by either F or S followed by the initial of the performer, and then the number of the phrase or phrases. For example, S:Y39 is phrase 39 of Yokoyama’s version of “Shôganken reibo”.

Variations of reibo no te occur repeatedly in all ten versions of “Reibo” transcribed in this study. Figure 25 gives two examples of reibo no te from the transcriptions used in this study. The two examples are from recordings of performances by the same person (Watazumi), and represent both the “Shôganken reibo” and the “Futaiken reibo” groups. They show the extremes the reibo no te pattern can have in length and complexity.

Melodic passages considered reibo no te are labelled and discussed as such in the honkyoku tradition, for example during lessons. Because the notes in reibo no te are usually not core tones and are of extremely short duration, the placement of the reibo no te passages are most helpful in collating material that are neither beginnings and ends of formal divisions nor core tones. All of the reibo no te are marked with brackets

in the completed lining up of the orthographicaly simplified transcription.

There is some material in the orthographically simplified transcriptions which does not fit in any of the three categories mentioned thus far: beginnings and ends of sections, core tones, and distinctive melodic formulae. The less obvious the correlation between notes becomes, the more arbitrary the decision becomes to determine their placement. Notes that do not seem to line up with any notes in other versions are given an isolated position.

It soon became apparent that my attempt to line up the transcriptions would not succeed without relying on my intimate and intuitive understanding of the piece as a recipient, performer, and teacher of four of the ten versions. The choice of notes or phrases to use as correlative ‘signposts’, which most likely corresponded with each other and between which the other notes had to fall, could in many instances only be made with such an understanding. Yet once those choices were made, the high degree of correlation between the transcribed material which remained to be lined up reinforced the original decisions.


In all phases of the arrangement of the transcriptions the two groups of pieces, “Shôganken reibo” and “Futaiken reibo”, are dealt with separately, as explained in the case of lining up of the beginnings and endings of formal divisions. Only after the correlation of the pieces within each group is independently established are the notes from the pieces of both of the two groups lined up. It should also be noted here that the various levels of material used to line up the simplified transcriptions confirm each other. For example, the placement of the core tones confirms the locations of the formal divisions and distinctive melodic formulae. Likewise, the locations of the distinctive melodic formulas confirm the placement of both of the previous levels of core tones and formal divisions.

The successful arrangement of the ten transcriptions confirms a common identity shared by the ten pieces represented in the transcriptions, as is strongly suggested by the genealogy chart. To what extent and in what ways the common identity is manifested in the ten transcriptions, and how they are related to the lines of transmission are addressed in the analysis.

The completely lined up orthographically simplified transcriptions of the ten recordings can be found in appendix 3.

6.4 Comparative analysis of “Reibo”

Although the impetus behind a comparative analysis of the melodic contours and the details of the “Reibo” pieces comes largely from a musicological perspective, the methodology of the analysis relies very much on a tradition-based understanding of the pieces. The method of analysis used in establishing contours is based upon knowledge of, and processes found within, the shakuhachi honkyoku tradition in its execution.

With the large data base of the transcriptions of ten recordings, representing the lineages, generations, and individual performers outlined earlier, comparisons will be made on a number of levels, and between a number of subjects in a number of combinations:

  1. between the two groups of “Reibo” pieces (the “Futaiken reibo” group vs. the “Shôganken reibo” group) in order to show similarities and differences which define the two groups.
  2. between the two lineages within each of the two groups of pieces (the FR versions of the Watazumi-Yokoyama lineage vs. the FR versions of the Jin-Sakai lineage, and the SR versions of Watazumi-Yokoyama-Iwamoto vs. the SR versions of the Jin-Sakai lineage) in order to show the level of consistency between performers within the same lineage.
  3. between individual performers of different lineages within the same group (Jin’s FR version vs. Uramoto’s FR version). In this and the following combination of comparisons, patterns of variation between individuals can be seen.
  4. between individual performers within the same lineages (Uramoto’s FR version vs. Watazumi’s FR version).

The comparisons focus upon the similarities and differences between such elements as formal divisions, the placement and frequency of core notes (see p.366), the location and execution of cadential and other distinctive melodic formulae, details of performance such as attacks and re-articulations and details of pitch.

Through the comparisons, the following questions can be addressed: In what ways can the two groups of “Reibo” be defined? Given the ten different performers representing two different lineages within the two different groups of pieces, what patterns of similarities and differences can be seen? How are these patterns related to the transmission between the performers, and to the issue of stability and change? To what extent do patterns occur which support or refute what is known about individual performers and their beliefs concerning the nature and essence of honkyoku?

Before such comparisons can be made, however, it is necessary to clearly establish that the ten performances represented by the transcriptions share similarites and differences to a degree that indicates that they are all identifiable as valid manifestations of “Reibo”. This has already been suggested by the genealogy charts (see pp.173196) and ahs been further substantiated by the successful arrangement of the orthographically simplified transcriptions (see pp.569594).

6.4.1 Establishing a common identity for the ten pieces using data from the transcriptions

In the context of theories of orality (pp.218223), “Reibo” is not a single piece. There is neither an ‘original’ “Reibo”, nor versions of an ‘original’ “Reibo” piece. Thereis no single underlying text of “Reibo”. In trying to establish whether or not the performances represented by the ten transcriptions are all identifiable as “Reibo”, one must look at the amount and placement of shared material. The more material that is common to all ten of the transcriptions, the more likelihood of a shared identity. If the shared material includes such important material as core notes, beginnings and endings of formal sections and distinctive melodic formulae, then this likelihood gains even more credence.

A great deal of shared material between the ten pieces can be seen in the arrangement of the orthographically simplified transcriptions described previously (pp.359369). Particularly significant are notes sustained for long durations. In figure 26, it can be seen that certain notes of long duration in the first seven to thirteen phrases of the initial shirabe section are shared by the majority of the pieces. These are arranged in the same way as in the orthographically simplified transcriptions in the previous section. Notes appear only if identical notes are found in at least six among all ten of the “Reibo” versions, and the note in at least one of the versions has a minimum duration value of a minim (i.e., at least two seconds in length). Most of the phrases in the shirabe section of all ten transcriptions share at least one of these notes. A similar degree of shared material can be seen throughout the entire piece.

In addition to the notes of long duration discussed above, other material supports the idea of a common identity of the ten pieces, in particular the material at the beginnings and ends of formal divisions (table 3). The following observations can be made regarding the main divisions of “Reibo”.

In all ten performances of the “Reibo” piece except that of Uramoto, the take shirabe section ends on the fundamental note of the instrument (d’) (figure 27). The honte section, which follows the take shirabe section, begins on the first octave above the fundamental note (d”) in all performances. This division is one of the most easily recognized in the transcriptions in both groups.

It was stated earlier (p.364) that the “Futaiken reibo” group has only five sections marked in the traditional scores, while the “Shôganken reibo” group has six. This is because there is no formal division marked in the traditional scores at the location in the FR pieces which corresponds to the location of the division between the honte and takane sections in the SR pieces (see table 3). The final phrase of the honte section of the SR pieces ends on d” (figure 28). The FR phrase, which should correspond to the final phrase of the “Shôganken reibo” honte section, ends on g”. Both of these notes are ‘core notes’ (see p.316), and are important in their function as cadential endings or ‘tonics’ (see the discussion of orality, pp.218223). These notes are held for over four seconds in all of the FR pieces and for over two seconds in all of the SR pieces. Therefore, it is reasonable to draw a parallel between the g” of the FR pieces and the d” of the SR pieces at this point. Furthermore, the following phrase in FR follows the same melodic line as the corresponding SR phrase, that is, the beginning phrase of the SR takane section. One may therefore conclude that there is a division in the FR pieces at a point which corresponds to the end of the honte section and the beginning of the takane section in the SR pieces, even though this division is not labelled in the traditional scores. If we count this unmarked division, the number of formal divisions of the “Futaiken reibo” pieces become equal to the six marked divisions of the “Shôganken reibo” pieces.

As shown in figure 29, what is labelled the final phrase of the honte section of FR in the traditional scores, corresponds to the final phrase of the takane section of SR (see table 3). Though the names of the sections differ between the two groups at this point, the cadential and initial formulae at this formal division are related to each other. In all ten versions, there is a sustained (over four seconds) d” at this point, followed by a short melodic figure in all of the SR versions, and a brief final note in Jin’s FR version.

What is called the takane section of FR is labelled the takane gaeshi of SR (see table 3). In the five FR pieces, the main note of the initial phrase of this section is g”. In the SR pieces, the g” phrase does not occur. There is, however agreement between the SR group and three versions of the FR group in the subsequent material, beginning with the note b-flat”, followed by a long c”‘. These three are the third phrase of the takane section of both Watazumi (F:W52) and Yokoyama (F:Y51) and the second phrase of Sakai (F:S31). After the long c”‘, which is found in eight of the ten transcriptions, all ten versions coincide with phrases that play between d”‘ and e-flat”‘, resolving on c”‘.

The final phrase of the takane section in the FR versions (F:U33, etc.) and the takane gaeshi section in the SR versions (S:W 47, etc.) (figure 30) resolves on a sustained d”. In all ten versions, the d” is followed by the short melodic figures discussed in more detail below (see p.379).

In all ten versions, the section following the above division ending is labelled the hachigaeshi section (figure 30). Also in all ten versions, this section begins with a distinctively intense b-flat” played with the kari technique, followed by variations of an oscillation between a” and b-flat”.

In the FR versions, the ending of the hachigaeshi section and the beginning of the musubi section occurs at F:U38, according to divisions mentioned in the scores of Jin and Sakai (figure 31). Though there is no sectional division labelled at this point in the SR versions, the corresponding phrase in the SR versions (S:W59, etc.) also begins with the pitch g’. In two of the versions (Watazumi and Jin) the note is sustained over four seconds. Material in this phrase in the SR group parallels that of the FR group. One might therefore conclude that there is a weak unlabelled division at this point in the SR group, which corresponds to the beginning of the musubi section of the FR group (see table 3).

Though there is less correlation between the beginnings and endings of the musubi section of the two groups of pieces than can be observed in the rest of the piece, in the context of this discussion it is important to observe that both groups have final sections identically labelled as the musubi section.

The data in the transcriptions at the locations of the beginnings and endings of formal divisions manifest to a great extent a common overall structure shared by all ten of the performances. The amount and type of similarity seen between the ten transcriptions at these points also correspond to characteristics in tune with theories of orality regarding structural points (see pp.218223). The degree of similarity at the points immediately before and after the divisions between large scale structures in the pieces, especially the frequent occurrence of shared cadential formulae, is also in accord with theories of orality, in particular where pertaining to the use of salient features and the beginnings and ends as focal points in the process of constructing the oral performance (see p.221).

The material at the formal divisions substantiates the observation concerning notes of long duration shared by all ten performances discussed above. A third indication that the ten performances are all manifestations of “Reibo” is the existence and location of the distinctive melodic formula described earlier, reibo no te (p.368). This formula can be found in all ten of the transcriptions. Table 4 summerizes the occurrences of reibo no te within the formal divisions of the ten performances. Though there are differences between the details of the various reibo no te, for the most part they occur in the same places in all ten of the pieces.

Two of the discrepancies between the ten trascriptions in the occurrences of reibo no te underline the differences between the two groups of pieces. In the hachigaeshi section, a single reibo no te occurs in all five of the SR versions, but in none of the FR versions. Also in the musubi section, one reibo no te occurs in four of the FR versions, but not in any of the SR versions.

Some of the other differences in the occurrences of reibo no te indicate patterns of variation other than those related to the two groups. For example, the second occurrence of reibo no te in both the honte and the takane sections of the FR group is performed by the Uramoto-Watazumi-Yokoyama lineage but not the Jin-Sakai lineage. Differences relating to lineages and other elements of transmission are discussed in the analyses to follow.

The above discussion presents ample evidence that the performances represented by the ten transcriptions used in this study are all manifestations of the piece “Reibo”, with both a great number of shared similarities and a substantial amount of possible cross-fertilization where differences occur. This was shown by the notes of long duration, the material at the formal divisions of the pieces, and the distinctive melodic formula known as reibo no te.

6.4.2 Comparison of the two groups, “Futaiken reibo and “Shôganken reibo”

The previous section established a commonality between the performances represented by the ten transcriptions. This section focuses upon the differences that can be found between the two groups at the locations of the formal divisions, between cadential formulae and between other material that is present in one group but not the other. These differences are analyzed for relationships between the ten performances in order to show to what extent similarities and differences indicate patterns of variation that can be explained in terms of transmission.

Of all the divisions noted in traditional scores, the endings of the hachigaeshi sections and the beginnings of the musubi sections appear to bear the least correlation between the SR and FR groups. According to divisions marked in the Jin and Sakai scores, the final phrase of the FR hachigaeshi section (figure 32) ends with a sustained d” followed by a short c”, similar to patterns of other section endings. At this point, there is no division marked in any of the SR scores. The SR versions also have phrases ending on d” and c”, but the d” in the SR versions is short in duration.

Variation amongst the SR versions can also be seen at this point. In Watazumi’s SR, this phrase (S:W58) does not have the final c”, while in Iwamoto’s SR version (S:I58), the phrase ends on c”‘, an octave higher than the others. As will be seen in the analyses that follow, Watazumi’s versions of both the FR and SR groups most frequently diverge from the other versions. The versions of the other members of his lineage, that is Yokoyama and Iwamoto, also frequently exhibit differences.

In the FR versions, the following phrase (F:U39, etc.) (figure 32), marks the start of the FR musubi section. In all but one of the FR versions the phrase begins with a sustained g’, which is preceded by one or two notes of brief duration. Watazumi’s FR version is the exception (F:W78), omitting the g’. Immediately after the difference in the initial phrase, the Watazumi version parallels the other FR versions except Yokoyama’s version, which omits a number of phrases at this point. The differences seen here between the performances not only differentiate the two groups, but also point to a number of other patterns of variation, discussed below, such as divergences within a single lineage, in this case, that of Uramoto-Watazumi-Yokoyama.

The division between the hachigaeshi and musubi sections in the SR versions, as labelled in the Yokoyama, Jin, and Sakai scores, does not occur until several phrases later (S:W60) (figure 33). The final phrase of the SR hachigaeshi section conforms with the cadential pattern c”-g’ (see below, p.379). There is nothing in the FR versions to suggest a similar pattern or a cadential ending, a clear case of variation following the parameters of the group.

The SR musubi section begins with a b-flat”, played with the kari technique, which then alternates in various patterns with a”. In Yokoyama’s version, the initial note is a”, but the overall pattern is the same (S:Y67). No such kari b-flat” – a” pattern can be seen in the FR versions at this juncture, except in Watazumi’s version (F:W88), which, once again, exhibits a high degree of divergence from other FR versions.

In all of the beginnings and ends of sections of the piece, the ending of the “Reibo” pieces exhibits the least conformity between the two groups. In the SR versions of the Watazumi-Yokoyama-Iwamoto lineage (S:W62, S:Y68, S:I63) (figure 34), the final phrase of the piece ends on a sustained a-natural”, which is preceded by b-flat”. In the Jin (S:J81) and Sakai (S:S80) versions, this entire phrase does not occur. These two versions conclude on the phrase corresponding to the penultimate phrase of the Watazumi-Yokoyama-Iwamoto versions, which ends on the notes c” – d”, with the d” sustained. A virtually identical phrase is notated in the Yokoyama score as the final phrase of the piece, but is not performed.45 The pattern of variation seen here clearly reflects transmission between members of the same lineage within each group.

45. When Yokoyama taught me “Shôganken reibo” in 1989, he said that the c” – d” phrase was one way to end the piece, but that he preferred ending on the a-natural”.

The FR versions do not have the b-flat” – a-natural” phrase that concludes the SR versions. Instead, up to eight phrases later (figure 35), all versions conclude on a sustained d’, preceded by a shorter c’ in Uramoto’s (F:U51) and Watazumi’s (F:W99) versions and the pitch g’ in Yokoyama’s (F:Y82), Jin’s (F:J66), and Sakai’s (F:S60) versions. Thus, the final note is d’ in five of the ten versions (the five “Futaiken reibo” pieces), d” in two versions (S:J and S:S), and in one version the final phrase is d” in the notated score but not in the performance (S:Y). The piece ends on a” in three of the versions (S:W, S:Y, S:I). The variation seen in the endings is not confined to the final note, but is evident in the entire final phrase.

The differences observed in the endings of the piece can be classified in the following manner. The greatest differences can be seen between the two groups of pieces. Within both groups, the differences coincide with the different lines of transmission, with the single exception of Yokoyama’s FR version.

The data in the transcriptions at the locations of the beginnings and endings of formal divisions not only manifests to a great extent a commonality between all ten performances, but also reflects lines of transmission defined by both the two groups of pieces and the lineages within each group. The greatest degree of similarity is found in comparisons of the FR Jin and Sakai pieces, the SR Jin and Sakai pieces, the FR Uramoto, Watazumi, and Yokoyama pieces, and the SR Watazumi, Yokoyama, and Iwamoto pieces.

In addition to the material at the formal divisions, another feature which distinguishes the two groups of pieces is the distinctive melodic formulae ‘d-c’ (hereafter referred to as cadential formula 1) and ‘d-c-g’ (hereafter referred to as cadential formula 2) (see figure 36). These two melodic formulae are particularly significant because they are often cadential phrases and are found in a number of locations throughout all ten transcriptions. Cadential formula 1 is found only in the FR versions, while cadential formula 2 occurs most often in the SR versions. The absence or presence of a final g’ differentiates the two formulae, and hence the two groups of pieces. The following discussion compares the two groups of the “Reibo” piece, focusing upon these distinctive melodic formulae.

Figure 36 shows the first example where melodic formula 1 or melodic formula 2 is found in all ten versions (F:U14, etc.). This phrase concludes with a sustained d”, followed by either formula 1 or 2. In all versions the phrase after the melodic formula culminates in a sustained g”, the first occurrence of this core note in the upper octave. Thus an important cadential point occurs here.

Looking first at the FR group, variations of cadential formula 1 are found in all five FR versions. In the SR group, variations of cadential formula 2 are found in all five SR versions. As stated above, the main difference between these formulae is the presence of the g’ in all of the SR versions and the absence of the g’ in all of the FR versions.

Figure 37 (S:W21, etc.) gives the second and third occurrence in the SR versions of the cadential pattern (formula 2). There are no corresponding cadential formulae in any of the FR versions. This pattern, which is found with minor variations in all five of the SR group, is preceded by a phrase dominated by a sustained d” and marks the end of an example of a distinctive melodic pattern called reibo no te (see above, p.367).

Cadential melodic formula 2 occurs again in the SR versions during the final phrase of the honte section (S:W23, etc.) (see figure 38). This section concludes with a sustained d”, followed in all of the SR pieces by cadential formula 2, except S:W. The Watazumi version ends with a single sustained d”, another of many examples of variation between Watazumi and the other performers within the group. As was the case in the previous example, there are no corresponding cadential formulae in any of the FR versions, though there is a sustained g” marking the corresponding point in the five versions of the FR group. As explained in the discussion of large scale structure, there is no formal division labelled in the traditional FR scores at this point, but an important cadential point does occur.

The fifth and sixth occurrences of cadential formula 2 in the SR group are shown in figure 39, marking the end of the takane section. The formula occurs only once in Jin’s SR version (S:J44). In contrast, in the FR versions, cadential formula 1 is found in the location corresponding to the fifth occurrence in the SR pieces of formula 2. No cadential formulae is found in the FR pieces in the location corresponding to the six occurrence of SR formula 2.

The section marked takane in the SR group (S:W36, etc.) (see figure 40), once again concludes with cadential formula 2. In this case, the pattern in the Watazumi-Yokoyama-Iwamoto SR versions omits the c”, being only d” – g’, while the Jin-Sakai SR versions retain the c”. In the FR pieces, this point is the final phrase of the honte section and ends with one or more sustained d”s. A variation of cadential formula 1 occurs only in Jin’s FR version. The phrasing at this point corresponds to a pattern of variation frequently observed, with these notes making a separate phrase in the Watazumi-Yokoyama versions (F:W49, F:Y48). In contrast, they conclude a longer phrase in the Jin-Sakai versions (F:J29, F:S29), as well as the Uramoto version (F:U24).

The final phrase of the takane section in all of the FR versions and the final phrase of the takane gaeshi section in all of the SR versions (figure 41) resolves on a sustained d”. In all of the pieces in the SR group, this note is followed by cadential formula 2. These notes form a separate phrase in the Jin-Sakai SR versions (S:J62, S:S62). In all of the FR versions, the d” is followed by a variation of cadential formula 1.

As explained in the discussion of large scale structure, there is no formal division labelled in the traditional SR scores at this point, but an unmarked minor division does occur. The final phrase of all of the FR hachigaeshi sections also concludes with cadential formula 1, (see figure 42). Uramoto’s version differs from the other versions by omitting the c” (F:U38). In all of the SR versions at the corresponding point, the phrase ends on a d”, followed in every version except Watazumi’s with a short c”. Watazumi again is the exception, omitting the c” (S:W58).

The phrase preceding the SR musubi section, that is, the final phrase of the SR hachigaeshi section (see figure 43), conforms with the previous examples by end-ing with a sustained d” followed by variations of formula 2. As seen earlier, these notes form a separate phrase in the Jin-Sakai SR versions, and conclude a larger phrase in the Watazumi-Yokoyama-Iwamoto SR versions. The cadential formula does not occur in the FR versions at this point.

In the above instances, the consistency and variation of melodic formulae almost always corresponds to the two groups of pieces and the different lineages within the two groups. Figure 44 illustrates this correspondence. It is possible to conclude that the final g’ of cadential formula 2 is a defining feature of the “Shôganken reibo” group. There are no instances of the g”, which differentiates formula 2 from formula 1, occurring in the cadential patterns of any of the “Futaiken reibo” pieces.

Besides the differences that are found at the formal divisions and between the form and number of occurrences of the cadential formula, other material differentiates the two groups of pieces . There are a number of instances of material of one or more phrases that is present in all five versions of one group that is not found in any version of the other group. There are also a number of instances of a series of phrases present in most or all of the versions of one group which are not found at all in the versions of the other group.

The first major example of this is found in the honte section at the location corresponding to S:W21 (figure 45). Three phrases that occur in all five of the SR versions are completely absent in all five of the FR versions. Secondly, in what corresponds to F:W57 in the takane section of the FR versions (corresponding to the takane gaeshi section of the SR versions), a phrase of reibo no te material (see p.367) that occurs in all of the FR versions (except that of Uramoto), does not occur in any of the SR versions (figure 46). The opposite phenomenon occurs a few phrases later, with another reibo no te occuring in the SR versions (figure 47) but not found in any of the FR versions.

It is the hachigaeshi section, however which contains the longest occurrences of material found in the versions of one group but not in those of the other. At the point corresponding to F:U36 (figure 48), between two and six phrases of material is found in all five of the FR versions and in none of the SR versions. In the same relative location (S:W54, etc.) (figure 49), up to four phrases present in all five of the SR versions are not found in the FR versions. Several phrases later (S:W58, etc.) (figure 50), a reibo no te occurs in all of the SR versions but does not appear in any of the FR versions.

There are also a few occurrences of the above phenomenon in the musubi section. A phrase in the SR versions at what corresponds to S:W59, which is part of the hachigaeshi section in the SR versions but still a part of the musubi section in the FR versions (figure 51), has no counterpart in the FR versions. At the same relative location, beginning at F:W79 (figure 52), there are up to five phrases of material in the FR versions that does not occur in any of the SR versions. Finally the final phrase in SR versions of Watazumi, Yokoyama, and Iwamoto (S:W62, etc.) (figure 53) does not occur in the FR versions, and up to eight phrases at the very end of the FR versions have no counterpart in the SR versions (figure 54).

These instances of material shared by all of the versions of one group and none of the other group, like the note g’ in the cadential formula discussed above, can be used to define the two groups.

Despite the similarity exhibited by all ten pieces on many levels, there is a high degree of variation between the two groups. Variation between different lineages and even between pieces of the same group and the same lineage was also observed in the above discussion. The existence of the innumerable variations between all performances of “Reibo” is a characteristic of honkyoku frequently ignored and even suppressed in the more rigid lines of transmission and in some of the literature. (see pp.288295) In the remaining analyses, numerous patterns of variation between all of the ten performances will be seen repeatedly. Many of these variations may illustrate the incorporation of the concept of honnin no kyoku (see pp.229,255,267)in honkyoku performance.

The previous sections compared the first and broadest classification of transmission, the two groups of pieces, first by looking at the similarities and then the differences of the FR and SR pieces. The following section looks at the second classification of transmission, that of lineage.

6.4.3 Comparisons within a single lineage within one group

Not surprisingly, the performances of “Reibo” whose transcriptions exhibit the least difference and the most similarity between each other are those within a single lineage in a single group. Nevertheless, varying degrees and kinds of differences can be found within these lineages. Many of the similarities and differences between transcriptions of performances of the same lineage and group have been seen in earlier comparisons. The following discussion further explores the relationship between the versions of Uramoto-Watazumi-Yokoyama FR lineage, the Watazumi-Yokoyama-Iwamoto SR lineage, and the Jin-Sakai FR and SR lineages.

The discussion of performers within a single lineage will first examine the transmission within the Jin-Sakai FR and SR lineages, followed by an examination of the Uramoto-Watazumi-Yokoyama FR and the Watazumi-Yokoyama-Iwamoto SR lineages. As explained previously (p.349), although Uramoto and Watazumi belong to the same lineage and group according to the genealogy chart, their relationship is an exception; the transmission between Uramoto is not acknowledged by Watazumi. In contrast, the Watazumi-Yokoyama-Iwamoto transmission is similar to the Jin-Sakai transmission in that it is clearly acknowledged. In the final part of this section a detailed look is taken at the straightforward, clearly acknowledged Watazumi-Yokoyama-Iwamoto lineage. The more ambiguous Uramoto-Watazumi transmission will be dealt with separately (p.404418). Jin-Sakai lineage

The greatest similarity between performances can be seen in the Jin-Sakai lineage of both the FR and the SR versions. This is the case in spite of the fact that the two shakuhachi players do not represent a direct transmission. Instead, Sakai is one generation removed from Jin, with Jin’s student, Moriyasu, the intermediary between them (see genealogy chart 7, p.351). Although not direct, the transmission is openly acknowledged by all parties concerned. Sakai in particular made it clear that the transmission of the “Reibo” pieces should be traced back to Jin (and beyond). It is therefore not surprising to observe more similarities than differences between both the FR and the SR versions of Jin and Sakai, although variation between the performances of the two does occur. In general, Sakai’s versions of both groups of pieces adhere closely to those of his predecessor Jin. A similar degree of similarity can be observed between Watazumi and Iwamoto, also performers of the same lineage but one generation removed on the genealogy chart (see below, pp.389397).

The similarity between the performances of Jin and Sakai is especially apparent in the melodic line and in the phrasing. The total number of phrases in Jin’s SR version is eighty-one, compared to eighty phrases in Sakai’s SR version. There is more discrepancy between the FR versions of the two players, where a total of sixty-six phrases in Jin’s FR version contrast with only sixty phrases in Sakai’s FR version. The difference between the number of phrases in Jin and Sakai’s FR versions is still, however, significantly less than the difference between any two of the three members of the other FR lineage: the total number of phrases in Uramoto’s FR version is fifty-seven, compared to ninety-nine phrases in Watazumi’s FR version and eighty-two phrases in Yokoyama’s FR version.

The differences between the versions of Jin and Sakai occur primarily where Sakai’s version becomes more complex or elaborate than that of his predecessor. For example, the initial phrase of Sakai’s FR version is much more developed than Jin’s initial FR phrase (figure 55). The same pattern can be seen in the third phrase of Jin’s and Sakai’s SR versions (figure 56), in the fifth phrase of Jin’s and Sakai’s SR versions (figure 57), in the ninth phrase of Jin’s and Sakai’s FR versions (figure 58), and again in the twenty-third phrase of Jin’s SR version (phrase twenty-five in Sakai’s SR version) (figure 59). In one instance, Sakai performs an entire phrase which has no counterpart in Jin’s version (S:S45) (figure 60). In contrast, there are very few, if any instance of either Jin’s FR or SR versions being significantly more elaborate than Sakai’s FR and SR versions.

The same tendency towards elaboration in Sakai’s performance compared with Jin’s performance is reflected in all aspects, including minute details. Because the similarities and differences between the Jin and the Sakai performances are so clear in the comparison of their large scale features, these details will not be discussed here.


Sakai Chikuho’s elaboration of Jin’s performance may be part of the elements of playing that he used to give his honkyoku a flavour unique to Chikuho ryû and that his successor, Sakai Shôdô, later so emphatically repudiated (see p.241). Yet, other than the type of elaboration seen in the above examples, there is very little deviation between the transcriptions of the Jin and the Sakai versions of the “Futaiken reibo” and “Shôganken reibo” pieces. This may be indicative of the more ryû oriented context in which Sakai operated, in contrast to the recusant nature of Watazumi’s transmission, the second lineage of this analysis. Uramoto-Watazumi-Yokoyama-Iwamoto lineage

The second lineage represented in the ten transcriptions is that of Uramoto, Watazumi, Yokoyama and Iwamoto. Though the transmission between Uramoto and Watazumi is not acknowledged by Watazumi, there is some evidence that supports the genealogy chart, which traces a line of transmission through them. Examples of this evidence, which are presented below, incorporate Uramoto’s performance in the examination of the entire lineage, and compare the performances within the lineage with those of the Jin-Sakai lineage.

There are many examples of similarities shared by the Uramoto-Watazumi-Yokoyama-Iwamoto lineage that are not present in the Jin-Sakai lineage. These examples demonstrate a simple correlation between persons within one lineage in a single group, indicating a direct transmission. For example, in the phrase which corresponds to F:U8 (figure 61), Uramoto plays a series of notes (a-flat’-g’, etc.) that are also played by Watazumi and Yokoyama, but which are omitted completely by Jin and Sakai.

Although essentially the same material, there is slight variation between Uramoto and Watazumi with regard to phrasing and the complexity of notes within the phrase. For example, the beginning of Uramoto’s phrase eight begins a note earlier than the beginning of the same phrase in Watazumi’s transcription. Also, Watazumi takes a breath at F:W13 and F:W14, while Uramoto does not. These differences fit the general pattern of variation between the two players, with Watazumi performing more phrases (a total of ninety-nine) than Uramoto (a total of fifty-seven) throughout the piece.

Uramoto and Watazumi exhibit slight differences between the short notes in what corresponds to Watazumi’s phrase fourteen. Even more variation occurs in the version of Yokoyama, who performs a number of short a-flat’ and g’ notes (F:Y13) rather than the single sustained a-flat’ found in both Uramoto and Watazumi’s versions.

A clear pattern of transmission between three generations of the same lineage can be seen. There occurs some variation in the complexity of the phrases from Uramoto to Watazumi, with more complexity occurring from Watazumi to Yokoyama. This tendency towards elaboration is similar to that which was seen in the Jin-Sakai lineage.

The similarity which exists between the Uramoto-Watazumi-Yokoyama-Iwamoto performances is thrown into strong relief when compared with the Jin-Sakai performances; representing a different lineage or line of transmission. For example, at the point corresponding to the sustained a-flat’ and g’ in Uramoto’s phrase eight and Watazumi’s phrase thirteen, Jin and Sakai do not play anything. Instead there is no correspondence until the end of F:U8, etc. Turning to the SR groups, one finds a similar pattern of variation differentiating the Watazumi lineage from the Jin lineage. For example, at a nearby point in the SR versions, (see figure 61), there is no material in the Watazumi-Yokoyama-Iwamoto SR performances (at the end of S:W7, etc.) corresponding to phrase 13 in the Jin-Sakai SR performances. These and other examples found in the transcriptions support the genealogy chart in classifying within a single lineage the performances of Uramoto, Watazumi, Yokoyama, and Iwamoto. Watazumi-Yokoyama-Iwamoto lineage

The lineage chart and the examples presented above notwithstanding, Watazumi refuses to acknowledge Uramoto as his teacher of “Reibo”. In contrast to the unacknowledged transmission between Uramoto and Watazumi, the Watazumi-Yokoyama-Iwamoto lineage corresponds to the Jin-Sakai lineage in the straightforward, acknowledged nature of the transmission it represents. The Uramoto-Watazumi-Yokoyama-Iwamoto lineage will be further scrutinized by looking at just its Watazumi-Yokoyama-Iwamoto component. The unique Uramoto-Watazumi tranmission will be discussed separately (pp.404418).

The pattern of similarities and differences corresponding to the Watazumi-Yokoyama-Iwamoto lineage is especially noticable in the details of performance, an area of analysis as yet untouched. In order to compare the details of performance between performers and lineages, the method of analyzing them must be introduced and explained.

While analyses of written transcriptions using the methods outlined above can shed light on certain questions relating to transmission by showing large scale structures, melodic contour, and distinctive melodic formulae, they do not address the elements which, from the point of view of the performer, most distinguish one performer, generation, or lineage from another. These elements are the details of performance; these include embellishments and other minute performance techniques. The existing written and oral analyses of honkyoku by members of the tradition, described above (pp.331342), confirm that it is these elements that most concern the performers themselves. In terms of the shakuhachi tradition, this level of analysis is central to an understanding of honkyoku as a process rather than a product.

The method of analyzing the details of performance that is used below is based upon the traditional methods of discussing honkyoku. With this method, the transcriptions into staff notation function as points of reference rather than a source of analytical data. Such a detailed commentary on a particular technique, note, or segment of a phrase, is based on the language used in teaching, and in tradition-based discussions of the honkyoku tradition.

Because of the nature of the analysis, it is necessary to confine the discussion to a small portion of the pieces. A detailed analysis of all ten transcriptions based upon tradition-based methods is not practical due to the large amount of manipulation of data and analytical discussion it would entail. The only option was to choose a limited amount of material from the transcriptions, basing the choice as much as possible on my own knowledge and understanding of the “Reibo” piece as a performer of the piece and recipient of the tradition represented by the two groups of pieces and the two lineages within each group.

Consequently attention is focused on a single melodic formula only. Although representing only a fraction of the piece, I believe that the sample is large enough to draw a number of conclusions regarding elements that are and are not transmitted from one generation of performers to another in the honkyoku tradition. Although the detailed analysis is limited to one melodic formula, corresponding examples, observations, and cross-references will be drawn from the entire piece.

I have chosen to use as the basis of this analysis, elements of the reibo no te pattern (see p.367), a melodic figure which is found throughout the piece. In particular, I will focus on reibo no te found in the honte section (figure 62). My reasons behind this particular choice for analysis are as follows:

  1. Reibo no te is the most important melodic pattern in the “Reibo” piece. It is given a specific name within the tradition, and is often discussed in the process of transmission, for example during lessons.
  2. Reibo no te contain minute embellishments and performance techniques, including pitch and timbre variations generated by the meri and kari techniques. These details and performance techniques too are frequently discussed in tradition-based analyses, and in the context of teaching.
  3. This particular reibo no te pattern is present in all ten of the transcriptions, thus allowing observations to be made across the whole sample of recordings.

As stated previously, (p.367), reibo no te is a pattern of notes which may be less than a phrase, an entire phrase or more than one phrase. It is always in the upper octave range (, kan), and is predominately performed with the meri technique. It is usually characterized by a density of notes far greater than found in the rest of the piece.

The pattern of notes centres primarily around the pitch g” and to a lesser degree a-flat”, and frequently concludes with the pitches e-flat” and d”. The note g” is the primary note in all instances of reibo no te in nearly every version of the piece. There are, in the context of reibo no te, three methods used to produce the pitch g” (see table 5). These methods are distinct from each other and, for the most part, are not interchanged in the context of reibo no te. Which method should be used in a particular instance is an important aspect of what is taught and learned in the course of transmitting the piece “Reibo”.

The first and most common method of playing the note, which I will call ‘method A’, entails covering (counting from the bottom finger hole) the second, fourth, and fifth (back) finger holes and using the meri technique.46 The second method, which I will call ‘method B’, is the same as the first except that the second finger hole is left open. This raises the pitch, so the application of far more of the meri technique than is employed in the first method is required to lower the sound to the proper pitch. The third method of producing g”, which I will call ‘method C’, is by using the standard fingering (finger holes one and two open, with finger holes three, four and five closed) and the kari technique. Method C is not used in the main body of reibo no te. Instead, it is frequently used to produce the g” immediately after reibo no te, providing a sense of resolution to the reibo no te phrase.

46. See p.265 for a definition of the meri technique, and its opposite, the kari technique.

Similar usage of the above-mentioned three methods of playing the note g” in reibo no te are most apparent in versions by performers of the same lineage. These similarities manifest themselves over the same lineage regardless of group, FR or SR. In the Jin-Sakai performances of both the FR and the SR groups, only method A is used in the main body of reibo no te. In contrast, in the Watazumi, Yokoyama, and Iwamoto versions of both the FR and SR groups, both methods A and B are used. Method B is utilized most often to produce the initial g” in the reibo no te in the Watazumi-Yokoyama-Iwamoto performances. Data from the level of detail described above and conclusions based thereon will be incorporated in the remainder of the analysis where appropriate.

Turning now to more large-scale features of the performances, transcription-based analysis at times show that Iwamoto (third generation) adopts and further adds to variation that Yokoyama (second generation) has added to the version played by Watazumi (first generation). This pattern is also frequently the case in the Futaiken reibo Uramoto-Watazumi-Yokoyama lineage. But, as well as tending towards elaboration as seen in the Jin-Sakai lineage and the FR Uramoto lineage, the variation can also be one of simplifying. An example of this type of pattern of transmission can be seen at S:W15 (figure 63). Yokoyama abbreviates a series of e-flat”-d”s found in Watazumi’s transcription. Iwamoto adopts Yokoyama’s abbreviated variation and abbreviates further by omitting the f” that follows the e-flat”-d” pattern in both Watazumi and Yokoyama’s versions.

Another, quite different pattern can be seen in the transmission of “Shôganken reibo” between Watazumi, Yokoyama, and Iwamoto which may, in fact, best typify these three versions. This pattern is one whereby variations added by Yokoyama to Watazumi’s versions are not adopted by Iwamoto. In other words, in many cases Iwamoto’s version is more similar to Watazumi, his teacher’s teacher, than it is to Yokoyama, his teacher.

This can be seen from the beginning of the transcriptions (figure 64). Here Yokoyama plays in three phrases what Watazumi plays in two. Yokoyama’s additional breath, which occurs after the initial sustained d’, is not adopted by Iwamoto. Although Iwamoto’s version incorporates the additional sustained d’ found in Yokoyama’s version at the end of phrase two but not in Watazumi’s version, it again does not follow Yokoyama’s version in the next phrase (S:Y3). A single sustained e-flat’ in Watazumi’s version is performed as an elaborate pattern of e-flat’ and d’ notes in Yokoyama’s version. In Iwamoto’s version, the sustained e-flat’ appears again, followed by only two notes, e-flat’-d’, to suggest the elaborate pattern of Yokoyama’s version.

The variations seen here could be explained in the following manner. Yokoyama appears to have added the extra d’s to the end of Watazumi’s first phrase, which he then transmitted to Iwamoto. The extra d’ also appears in Yokoyama’s FR version, indicating a possible cross-over from one group to the other. Yokoyama may have added the extra phrase after transmitting Watazumi’s version, without the extra phrase, to Iwamoto. In taking the extra breath, Yokoyama might have been reflecting the performance practices of the FR version, in which he, and all of the four other players, take the extra breath. Equally possible is his being influenced by the performances of the SR version by Jin and Sakai, which has the extra phrase. This same scenario may explain the g’ played by Yokoyama before the e-flat’ in his phrase three. There are no notes before the e-flat’ in either the SR or the FR Watazumi performances.

There are numerous other examples in the transcriptions of what seems to be a borrowing or cross-over from the FR to the SR, or the SR to the FR pieces. The borrowed material may be from the performer’s own version of the piece in the other group, or from another performer’s version in the other group. A possible reason as to why Yokoyama’s elaborate pattern of e-flat’-d’s was not transmitted to Iwamoto will be given below.

Another example of this pattern of transmission between Watazumi, Yokoyama, and Iwamoto is in phrase S:W11 (figure 65). Yokoyama’s version varies from Watazumi’s version at the end of this phrase with the addition of a sustained c” followed by the two notes c”-g’ of short duration. These three notes are not, however, incorporated into Iwamoto’s version.

A final example of this pattern of transmission is located in S:W25 (figure 66). First of all, the breath that marks the beginning of this phrase in Watazumi’s version is not present in Yokoyama’s version. This variation in phrasing is not adopted in Iwamoto’s version, which follows that of Watazumi. Secondly, the pattern in which sustained and short e-flat”‘s and d”‘s are found in Watazumi’s S:W25 is varied in the corresponding phrase of Yokoyama’s version. This variation is not adopted by Iwamoto, which follows the original pattern found in Watazumi’s version.

Finally, the phrasing in Watazumi’s SR version is followed more closely by Iwamoto than by Yokoyama, as evidenced by the total number of phrases. Watazumi’s SR version has a total of sixty-two phrases, compared to sixty-eight phrases in Yokoyama’s version and sixty-three phrases in Iwamoto’s version.

Because of these and other similar examples of this pattern of transmission found in the transcriptions, one might, if one did not have access to genealogical data, assume that rather than Yokoyama being Iwamoto’s teacher, the line of transmission was Watazumi-Iwamoto-Yokoyama. The anomaly in the relationship between variation and transmission can be explained by an element of change and stability in the honkyoku tradition as yet not touched upon in the above analyses, namely that of date of transmission and date of recording.

Though it is not known when Yokoyama learned “Shôganken reibo” from Watazumi, we do know that Yokoyama taught Iwamoto the piece before the late 1970s, when Iwamoto went to England, where he has lived since. The recording used as the basis of Watazumi’s transcription was made in 1971. Iwamoto’s recording was made four years later, in 1975. That is, Iwamoto’s version reflects fairly consistently the version that Watazumi was playing just before Iwamoto left Japan for England. Yokoyama’s version on the other hand was not recorded until 1985.

Taking this chronology of the recordings into consideration, the following seems possible. Yokoyama learned the piece from Watazumi, and then taught the piece with only minor variation to Iwamoto. Iwamoto then performed the piece for recording in a way that reflected the similarity between Yokoyama’s and Watazumi’s performance at that time. By 1985, however, when the recording of Yokoyama’s performance was made, Yokoyama’s way of playing the piece has evolved to become more different from Watazumi’s recorded performance than Iwamoto’s recorded performance.

It is possible that by the date of Yokoyama’s recorded performance, Iwamoto’s performance would still have closely resembled that of Watazumi. Iwamoto, in England being relatively isolated from his teacher and the rest of the shakuhachi tradition in Japan for over a decade, might exhibit in his performance of honkyoku the conservatism seen in the language patterns and the performance of traditional music by groups of peoples and their successive generations who have migrated away from their native country.47

47. As one example, Ryan (1989:327) describes many of the performances of traditional music by Brazilian migrants in Sydney as “marked by rhythmic simplification and standardization”.

Yokoyama’s (1985:116-120) account of his experience of being taught the honkyoku entitled “San’an” (, “Safe Delivery”) by Watazumi further illustrates the relationship between the passage of time and and the transmission of honkyoku. Before he became Watazumi’s student, Yokoyama had in his possession a recording of the piece “San’an” as performed by Watazumi. Because it was one of Yokoyama’s favourite honkyoku, he was naturally pleased when Watazumi finally decided to teach the piece to him. But when Watazumi began to teach “San’an” to Yokoyama, it had changed. In Yokoyama’s ears, Watazumi was not teaching him the “San’an” that Yokoyama had grown fond of by listening to the recording. Consequently Yokoyama resisted learning the piece as Watazumi was teaching it. Because Watazumi would not accommodate Yokoyama’s desire to be taught “San’an” as performed on his recording, it took Yokoyama three years to learn the piece.


It might be possible to address the element of passage of time in relation to the transmission of honkyoku with a comparison of transcriptions of a set of recordings of, for example, three generations of performers made over a span of time. A comparison of the similarities and differences between transcriptions of several performances by the same person over many years, and between transcriptions of performances representing different generations might show patterns of variability that could be attributed to the effects of the passage of time, as well as other variations that could be attributed to the transmission process from generation to generation. In the case of this present thesis, however, with the ten transcriptions described above as its subjects for analysis, little more can be said than has been conjectured above.

6.4.4 Comparison between the transcriptions of Uramoto and Jin

The previous section focused upon the most direct lines of transmission, firstly by looking at the Jin-Sakai lineage. Secondly, the Uramoto-Watazumi-Yokoyama-Iwamoto lineage was examined. Finally, the component of this second lineage which is openly acknowledged by both teacher and student, that is, that of Watazumi-Yokoyama-Iwamoto, was further discussed. Relatively simple and direct correlations between performances and lineage were observed, especially in the Jin-Sakai versions. Variations between performers within the same lineage and group could be found, however, especially in cases involving Watazumi, and to a lesser degree Uramoto.

The unique relationship between Uramoto and Jin has yet to be studied. According to the genealogy chart, they both learned “Futaiken reibo” from the same teacher, Konashi Kinsui. They are therefore of the same generation and, in terms of their teacher, the same lineage. For this reason, a comparison of their performances adds another dimension to the discussion of transmission of the “Reibo” piece.

Although there appear to be no recordings of Konashi Kinsui, a comparison of Uramoto’s and Jin’s “Futaiken reibo” versions could indicate characteristics of the transmission that occurred between their common teacher and themselves. Comparing Uramoto’s and Jin’s versions may show ways in which pieces change and remain constant during transmission from a single teacher to different students. Among the ten transcriptions used in this study, this kind of relationship is found only with Uramoto’s and Jin’s FR versions.

Reference must also be made to the variable date of transmission. As was seen in the example concerning the Watazumi-Yokoyama-Iwamoto lineage, the dates of transmission between teacher and student, as well as the dates of the recordings from which the transcriptions are made, may have a signifcant bearing upon the patterns of similarities and differences between performances. It is not known when either Uramoto or Jin learned their FR versions from Konashi. Though the dates of the recordings are 1985 for Uramoto and 1980 for Jin, both of these dates represent when the recordings were released, not the much earlier dates of the performances. Therefore, in the following comparison, it is not possible to take into account any similarity or difference which may have arisen from differences in the times of transmission from Konashi and the times of recording the two performances.

The performances of “Futaiken reibo” by both Uramoto and Jin begin in a similar fashion. By the third phrase (figure 67), however, the two begin to vary. In Uramoto’s version, the third phrase begins with a sustained e-flat’, a note that does not occur in Jin’s version. The third phrase of Jin begins on the note g’, which is followed by the note d’. Both of these notes are core notes and are sustained for over four seconds. There is neither the beginning of a phrase nor the note g’ at this point in Uramoto’s version, though a note d’ does appear, which might correspond with the sustained d’ in Jin’s version. Uramoto, however sustains the d’ for less than two seconds. In other words, there is musical material present in both Uramoto’s and Jin’s FR versions which is not present in the other’s version.

Once again transmission lines seen in the genealogy chart are reflected in the comparisons, in that Watazumi’s and Yokoyama’s versions basically follow that of Uramoto, with the note e-flat’ beginning their third phrase. Likewise, Sakai’s version follows that of Jin, his predecessor on the genealogy chart, by not playing the note e-flat’.

In this case, Sakai’s version once again parallels that of Jin, his predecessor, almost note for note, by omitting the e-flat’ of Uramoto’s version in phrase three, and instead performing a sustained d’, followed by a sustained d’ in a single phrase. The closeness seen here between Jin and Sakai’s versions has been observed previously.

In contrast to the similarity between Jin’s performance and that of Sakai, material that is performed by Jin but not Uramoto, as well as material played by Uramoto but not Jin can be found in Watazumi’s performance. Comparison of the performances of “Futaiken reibo” by Uramoto, Jin, and Watazumi, accentuates patterns of divergence in the performances. These divergences, anticipated by orality theories, are not compatible with literate-based concepts such as ‘original’ or ‘authentic’ versions or performances.

The corresponding point in the SR versions further substantiate this conclusion. The e-flat’, which occurs in the third phrase of Uramoto’s FR version but not in Jin’s FR version, occurs in all of the SR versions, including that of Jin. The g’ which occurs in the third phrase in Jin’s FR version but not in Uramoto’s version, occurs in Sakai’s SR version. It does not occur in any of the other SR versions, including that of Jin. The sustained d’ which follows the g’ in four of the FR versions, and the shorter d’ at the corresponding point in Uramoto’s FR version, is found in phrase four of Jin and Sakai’s SR versions. The note does not occur in any of the other SR versions. Again, both the e-flat’ not played by Jin and the g’ not played by Uramoto, are represented in at least one of the SR versions, and in the case of the e-flat’, in all five of the SR versions.

In the previous example, the FR versions of Uramoto and Jin did not agree with each other, while Watazumi’s FR version contained all of the core notes found in both Uramoto’s and Jin’s versions and Sakai’s version agreed with that of Jin’s. An example of similarities occurring between Uramoto and Jin, which are also played by both Watazumi and Sakai, can be found in the initial few phrases of the takane section in the “Futaiken reibo” group (F:U25, etc.) (figure 68), implying a simple transmission from Konashi to both Uramoto and Jin and on to Watazumi and Yokoyama and to Sakai. In both Uramoto’s and Jin’s FR versions, there is an initial phrase centering around a sustained g”. This material occurs in the other three FR versions, though in a more elaborate form. In the FR versions of Uramoto and Jin, this initial phrase is followed by a phrase beginning with the distinctive notes d”‘ and e-flat”‘, the highest pitches in the piece. These notes also occur in the other three FR versions.

An example of similarities occurring between Uramoto and Jin, which are played by neither by Watazumi nor by Sakai, can be found in the same initial few phrases of the takane section in the “Futaiken reibo” group (F:U25, etc.) (figure 68). Deviating from the Uramoto and Jin FR versions, the other three FR versions (Watazumi, Yokoyama, and Sakai) all share what appears to be additional material between the initial phrase and what is the second phrase of the takane section in the Uramoto and Jin versions. This additional material also creates a difference in phrasing between the Uramoto and Jin versions and the other three versions. It must be pointed out that the additional material found in the versions of Watazumi, Yokoyama, and Sakai, but not in the versions of Uramoto and Jin, corresponds to the notes that are the beginning of the first phrase of this section in all of the SR pieces (including Jin’s SR version), that is the note b-flat”, followed by a long c”‘. After the long c”‘, all ten versions coincide with phrases that play between d”‘ and e-flat”‘, resolving on c”‘. The multiple patterns of transmission seen here can best be described in terms of the cross-over and variation which occurs among performers within an oral tradition, in which the most ‘authentic’ performance is the one most recently performed.

In the example above, the performances representing the first generation among the ten transcriptions agree with each other in the FR group in both material and phrasing at an important juncture in the piece, the beginning of a formal division. The three transcriptions representing the subsequent generations in the FR group, that of Watazumi, Yokoyama and Sakai, all have material not found in the transcriptions of the previous generation. The ‘added’ material is, however, shared by not only the two different lineages in the FR group, those of Watazumi-Yokoyama and of Sakai, but also by all of the SR versions, including Jin. In this case, new material was neither created nor added to the FR piece, by either Watazumi or Sakai, because variations of the material are found in all of the SR versions. The cross-over between both different lineages (Watazumi-Yokoyama and Sakai) and the different groups of pieces (FR and SR) is again evident here.

There is only one instance where a long and complex passage is found only in Uramoto’s FR version. Located in the musubi section (F:U42-44) (figure 69), the material is performed over three phrases. Only the initial note, c”‘, is found in Watazumi’s FR version (F:W83). The rest of the passage, which contains a number of core notes as well as a reibo no te, is not found in any of the other nine transcriptions. This is the only instance in the entire “Reibo” piece where a substantial amount of musical material occurs in only one of the ten versions. Watazumi is thus not the only performer to exhibit characteristics unique to his performance.

There are in fact, relatively few major musical events, such as phrases, melodic segments, beginnings of phrases and even solitary core notes, which occur only in one performance. This characteristic implies a cohesiveness or unity within the “Reibo” tradition that transcends lineage or grouping.

The transcriptions of Uramoto and Jin, who are of the same ‘generation’ on the genealogy chart and were both taught by Konashi, exhibit the non-uniform and relatively high degree of variation that is found in comparisons between transcriptions of other performers who do not have the same close relationship in terms of teacher. There is less agreement between the Uramoto and Jin transcriptions than is found in the more stable, acknowledged teacher-student transmission of both the Watazumi-Yokoyama-Iwamoto lineage and the Jin-Sakai lineage. As stated earlier, it is not known how much the disparity observed between Uramoto’s and Jin’s performances, (or between any of the performances) is due to differences in the dates of transmission and recordings. The influence and effect that the numerous teachers both performers appear to have had may be another factor in the variation between their performances (see p.160). Finally, in the comparison between Jin and Uramoto’s performances, a number of instances of cross-over between the two groups of pieces and the two lineages can be observed. For example, there is material that occurs in all of the SR performance which also occurs in the FR versions of Watazumi and Yokoyama, yet is not present in the FR performances of Uramoto and Jin.

6.4.5 Analysis of the Uramoto-Watazumi transmission

Of all of the lines of transmission on the level of individual performers of the same lineage, including that of Uramoto and Jin (see above), the greatest degree of disparity is found between Uramoto and Watazumi’s FR performances. A comparison of their performances testifies to the recusant nature of Watazumi’s ideology, discussed above (p.303304). An extreme example of this disparity can be seen in the initial phrase of the musubi section (F:U39, etc.) (figure 70). Uramoto’s version is not only more similar to that of Yokoyama than it is to that of Watazumi, but is also more similar to Jin’s and Sakai’s versions, which completely omits the material in the other versions.

The relationship between Uramoto and Watazumi is a special one, deserving particular attention in the present examination of the processes of transmission operating in all ten transcriptions. Because the transmission that is said to have occurred between Uramoto and Watazumi is not clearly acknowledged, at least by Watazumi, it is more problematic than the distinct, acknowledged transmissions between Jin and Sakai, or between Watazumi, Yokoyama, and Iwamoto. Thus, a more focused comparison of Uramoto and Watazumi, looking at both similarities and differences, is presented below.

The pattern of incongruity between the performances of Uramoto and Watazumi can be expected for two reasons. First of all, Watazumi claims that Uramoto did not teach him the piece “Reibo”. As discussed earlier (p.303), this is in keeping with Watazumi’s philosophy regarding the nature of honkyoku, i.e., the essence of honkyoku cannot be taught or learned; Watazumi claims that no one did or was even able to teach him honkyoku. Secondly, Watazumi’s reputation for eccentricity in expressing his individualism and his repudiation of hierarchical and frequently constraining systems of transmission, such as the iemoto system (see p.288), are consistent with his honkyoku performance.

Nevertheless, there are numerous similarities between Watazumi’s performance and that of Uramoto. The single most outstanding agreement between Uramoto’s and Watazumi’s performances of “Futaiken reibo” can be seen in the melodic line. The high degree of correspondence between Uramoto’s and Watazumi’s orthographically simplified transcriptions reveals agreement between the two versions at this level. The similarity is particularly evident in the case of core notes that are sustained over two seconds.

Another similarity between Uramoto’s and Watazumi’s performances, but one which cannot be seen in the transcriptions, is the degree of accuracy of the intervals between pitches, or relative pitch.48 Both Uramoto and Watazumi perform intervals between notes are fairly consistently, especially when compared to Jin, whose relative pitch is quite variable, and Sakai. Not surprisingly, the relative pitches of Watazumi’s student Yokoyama, and Yokoyama’s student Iwamoto are also quite accurate and consistent.

48. See chapter five (pp.271282) for a discussion of pitch and interval relationships in the honkyoku tradition, and how these are transmitted though lineages and individuals.

A similar pattern can be seen in a detailed analysis of one instance of reibo no te (see p.367). In both methods ‘A’ and ‘B’ of producing the note g” (see p.388), the meri technique must be applied to a certain degree in order to produce the desired pitch. If the meri technique is not sufficiently applied, both the g” and the a-flat” in reibo no te will be sharp relative to the final g” produced by method ‘C’. The relative pitch of the meri notes in this and all reibo no te in the versions of the Uramoto-Watazumi-Yokoyama-Iwamoto lineage is consistent, relative to the g” produced by method ‘C’. By contrast, both the FR and SR versions of the Jin-Sakai lineage exhibit a-flat” and g” meri notes that are sharp relative to the kari note g” of method ‘C’. The degree of sharpness is generally greater in Jin’s versions than in Sakai’s versions.

Of the two performers Uramoto and Watazumi, however, Watazumi’s relative pitch is more consistent than Uramoto’s pitch, particularly in the case of some meri notes. Uramoto performs these notes relatively sharp on occasion, especially during the distinctive reibo no te phrase. An example of the differences in pitch control can be heard in the initial phrase of the piece. In Uramoto’s performance, the initial e-flat’ is slightly sharp relative to the following d’. No such pitch variation can be heard in Watazumi’s performance.

This is not to imply that the Uramoto lineage is more ‘correct’ in producing the meri g” than the Jin lineage, or that Watazumi’s pitch is generally ‘better’ than Uramoto’s. As discussed in chapter five, the importance placed upon the accuracy of relative pitch varies from performer to performer and from lineage to lineage. In terms of the values of Uramoto and of the Jin-Sakai lineage, the pitch of the notes in question is not incorrect or ‘out of pitch’. The above observations regarding pitch amongst the ten versions are judgementally neutral, and are made for comparison, just as observations regarding the number or length of notes are made for comparison.

It should also be pointed out that a small part of the variation in pitch heard in the performances of Uramoto and Jin may be attributed to their being of an older generation. Their lives were closer to the historical period in Japan which predated the wide-spread introduction and consequent influence of western music and its standardization of pitch than were those of the generations that followed them. Also, the shakuhachi instruments available for much of their lives generally produced less standardized pitches than those instruments which became available to later generations (see p.281).

As stated above, although Uramoto nominally transmitted “Reibo” to Watazumi and thus is technically part of the Uramoto-Watazumi-Yokoyama lineage within the “Futaiken reibo” group, Watazumi claims that Uramoto exerted minimal influence over his playing. We have already observed a number of examples of differences in performances between Uramoto and Watazumi in the various comparisons made above. An examination of formal divisions shows a number of other differences between the transcriptions of the two performers that, because of their location, are particularly conspicuous. They become even more so when compared to the similarities that occur between other performers within direct teacher-student relationships.

For example, at the formal division between the shirabe section and the honte section (figure 71), Watazumi’s performance differs quite noticeably from Uramoto’s performance in both the final phrase of the shirabe section and the initial phrase of the honte section. There are no major differences at this point between any of the other performers within a single lineage, that is, between Watazumi’s and Yokoyama’s FR version, between either Jin’s and Sakai’s FR or SR versions, and between Watazumi’s, Yokoyama’s and Iwamoto’s SR versions.

Characteristic differences between the performances of Uramoto and Watazumi can also be seen from the initial notes of the transcriptions at the very beginning of the piece (figure 72). In this first phrase and throughout the piece, Watazumi makes use of more embellishment than Uramoto does. A second major difference is the number of phrases. Uramoto plays his version of “Futaiken reibo” in fifty-one phrases, while Watazumi performs the piece in ninety-nine phrases. The total elapsed time of this particular performance of Uramoto’s version is 10 minutes 34 seconds, compared with Watazumi’s 16 minutes 2 seconds. In most cases, Watazumi incorporates the pauses or breaths made by Uramoto, and adds many of his own.

The length of the versions both as expressed in elapsed time and in number of phrases demonstrates a pattern which occurs throughout the piece. Watazumi’s version of “Futaiken reibo” is far more complex than Uramoto’s version not only in embellishment, but also in melodic line and phrasing. Again an example of this can be seen soon after the beginning of the piece. The material between the initial series of the note d’ and the first instance of the sustained note a-natural’ in Uramoto’s version consists of two notes, e-flat’ and d’ (not including embellishments), performed in a single phrase (F:U3). The corresponding material in Watazumi’s version covers five phrases (F:W3-5) and consists of seven notes (also not including embellishments) (see figure 73).

Besides the elements discussed above, little else can be said about the similarities and differences between the FR versions of Uramoto and Watzumi without reference to the other versions, not only within the same FR group, but also from the other SR group.

The importance of basing all but the most rudimentary conclusions on comparisons of transcriptions of more than any two single performers, such as Uramoto and Watazumi, can be seen in the following examples. There are numerous occasions where the transcriptions of Uramoto and Watazumi do not correspond at all. For example, in the final phrase of the first formal division, take shirabe (figure 74), Uramoto plays two sustained g’. In Watazumi’s version, there are no g’s, either in the final phrase or the preceding phrase. Instead, his final phrase consists of a series of sustained d’ notes alternating with shorter c’s.

This clear example of Watazumi’s version differing markedly from Uramoto’s version is all the more prominent because it corresponds to the end of a formal section. Taking into account only the transcriptions of Uramoto and Watazumi, one could assume with reasonable confidence that the difference was consciously and deliberately effected by Watazumi.

It is incorrect, however, to assume that the many conspicuous and seemingly deliberate discrepancies between the two FR versions discussed above are Watazumi’s creations or compositions. In the first example (figure 74), the final d’ found in Watazumi’s FR transcription is found in all of the other FR and all of the SR transcriptions. Furthermore, the c’ is also found in the SR version of Jin and Sakai. On the other hand, the sustained g’s which Uramoto performs occur in both the FR and the SR transcriptions of Jin and Sakai, while they do not occur in the FR and SR transcriptions of Watazumi and Yokoyama, and the SR version of Iwamoto.

In a second example (figure 75), musical material that is found in Watazumi’s FR version, but not in Uramoto’s FR version, is performed in an even more elaborate form in all five of the SR versions (S:W8, etc.).

Thus, it cannot be concluded that, in his FR version, Watazumi either made up or deleted the material mentioned above since the SR versions of the Jin-Sakai lineage either have or lack the same material as Watazumi’s version. Watazumi might still be credited with (or ‘accused of’) changing the FR version of “Reibo”; he may have (perhaps even unconsciously) borrowed additions and omissions from the SR version and inserted them into the FR version transmitted to him by Uramoto. Many such crossovers between the two groups of pieces have been observed elsewhere.

Major examples of the ways in which the performances of Watazumi and Uramoto differ are found by comparing phrases eight through ten of Uramoto’s FR version with the latter part of phrase twelve and phrases thirteen though nineteen off Watazumi’s FR version (see figure 76). The difference in phrasing and the higher number of phrases in Watazumi’s is consistent with what has been observed before. Other major differences in the transcriptions of Uramoto and Watazumi begin at this point.

The final note of phrase fourteen in Watazumi’s version is the core note g’, sustained for over four seconds. This prominent note is not found in Uramoto’s version. Uramoto also does not perform the material that makes up Watazumi’s next phrase (F:W15). Yokoyama’s FR version parallels his teacher in both phrasing and material played. The Jin and Sakai FR versions, like that of Uramoto, do not have this material.

Both Uramoto and Watazumi perform the note g’ at the point corresponding to the beginning of phrase nine of Uramoto’s FR version (figure 76). This note is not present in any of the other eight versions. Watazumi’s agreement with Uramoto, however, is soon over. Watazumi begins a new phrase (F:W17) with the note a’-natural. Uramoto does not begin a new phrase here, nor is there an a-natural in his melody. Yokoyama follows Watazumi, also beginning a new phrase here with a’-natural. Thus, we can observe an element of variation between Uramoto and Watazumi that is transmitted to the next generation, Yokoyama.

This material does not occur in the FR versions of Jin and Sakai, nor is it heard in the SR versions of Watazumi, Yokoyama, and Iwamoto. The a’-natural and the beginning of a new phrase is, however, found in the SR versions of Jin and Sakai. Once again, the patterns of similarities and differences suggest a cross-fertilization that overrides the immediate influence of one’s teacher, especially in the case of the Uramoto-Watazumi transmission. Such an overriding cross-fertilization is an important feature of transmission within the honkyoku tradition.

At the point corresponding to F:U10 (figure 76), Uramoto plays two g’s, which together are sustained for over four seconds. Uramoto’s version is the only one in which these notes end the take shirabe section. These highly visible core notes are not performed by Watazumi, in either his FR version or his SR version. In Watazumi’s FR version, the take shirabe section ends with a new phrase centering on three sustained d’ notes. In Yokoyama’s FR version, there is also a new phrase, with two sustained d’ notes. There is a single d’ note in Jin and Sakai’s FR versions, although they are not in a separate phrase.

The variation at this point between Uramoto and Watazumi, and between the Watazumi-Yokoyama-Iwamoto lineage and the Jin-Sakai lineage is particularly significant because it occurs at a boundary in the piece. In the light of theories of orality, it is unlikely that the variations at this point occurred unintentionally. Furthermore, the patterns of variation suggests that Watazumi was largely the initiator.

Table 6 summarizes the manner in which the elements of the phrases F:U8-10, in question line up with each other in the ten transcriptions. Besides the differences between groups, the most noticeable features of the table are the similarities between versions belonging to acknowledged lines of transmission (Watazumi-Yokoyama-Iwamoto; Jin-Sakai), and the differences between both the Watazumi-Yokoyama-Iwamoto line and Uramoto, and the Watazumi-Yokoyama-Iwamoto line and the Jin-Sakai line. A similar pattern can be observed throughout the entire transcriptions.

Another musical element which reflects the nature of transmission between Uramoto and Watazumi, as well as the other lines of transmission, is reibo no te (, “Reibo” pattern’) (see above p.367).

Figure 77 (corresponding to F:W36-37) shows an instance where Watazumi characteristically performs reibo no te quite differently from his nominal teacher Uramoto, especially regarding complexity. There are only ten notes altogether in this Uramoto version of reibo no te (F:U16). The sparseness of this instance of Uramoto’s reibo no te is reflected most in the FR versions of Jin, and the recipient of Jin’s transmission, Sakai (F:J22. F:S22). Other examples of similarities between Uramoto and Jin, who according to the genealogy chart had the same teacher (Konashi Kinsui), were given above (p.187).

At the next occurrence in the transcriptions of reibo no te, Watazumi’s FR version again bears little resemblance to Uramoto’s FR version. Although in this instance, two reibo no te patterns (figures 78 and 79) can be identified in all three versions of the Uramoto-Watazumi-Yokoyama lineage (in F:U19 and F:U22, in F:W42 and F:W43-44, and in F:Y42 and F:Y43), there are noticeable inconsistencies between Uramoto’s version and Watazumi’s version.

For example, this occurrence of reibo no te begins in the middle of a phrase in Uramoto’s version (F:U19), while it corresponds to the beginning of a phrase in Watazumi’s version, (another example of the phrasing differences in the two versions). The only other performance to begin this instance of reibo no te in the middle of the phrase is Sakai’s FR (F:S25). Watazumi’s more complex and dense reibo no te compared to Uramoto’s also follows previously observed patterns. As observed earlier, in general, Watazumi plays more phrases with more notes and more embellishments than does Uramoto.

Instances in which the Uramoto version differs from versions of the Watazumi-Yokoyama lineage more than from versions of the Jin-Sakai lineage, in spite of Uramoto nominally belonging to the former, are also apparent on the level of performance details. Differences between the Uramoto and Watazumi FR versions (same lineage, same group) are at times not only greater than differences between the Uramoto FR version and the Jin-Sakai FR versions (different lineages, same group), but are also greated than differences between the Uramoto FR version and the Jin-Sakai SR version (different lineages and different groups). A partial reason for this might be that Uramoto and Jin, though considered belonging to different lineages in this analysis, both learned from the same teacher, Konashi Kinsui (see p.187).

Two clear examples of Uramoto’s performance differing from Watazumi’s performance more than from Jin’s performance can be seen in the details of performance. First of all, Uramoto produces all of the g”s in reibo no te by method A (see p.392), covering the second finger hole. This is also the case in Jin and Sakai’s versions of both the FR and the SR groups. By contrast, in the Watazumi and Yokoyama versions of both the FR and the SR groups, as well as Iwamoto’s SR version, some of the g”s in reibo no te are played with method B, requiring more meri technique than method A.

The difference noted here is especially significant. Both Uramoto and Jin, who learned from the same teacher (Konashi), do not use method B, while Watazumi, and the other performers of his lineage, Yokoyama and Iwamoto, do employ method B. The changes in technique involved in going from method A to method B are not likely to have been made unintentionally. Method B is, therefore, almost certainly an addition made by Watazumi.

The second example of differences between the performances of Uramoto and Watazumi being greater than differences between the performances of Uramoto and Jin/Sakai can be seen in the sequence of notes of reibo no te. In his reibo no te version, Uramoto plays a number of sequences of notes, the basic pattern being a-flat” – g” – e-flat” – g”, which is often preceded by the grace notes b-flat” or a-flat”. This series of notes, or variations of it, occurs in reibo no te of all ten versions of the piece (bracketed in figure 80). In Uramoto’s version, the initial occurrence of this series at the beginning of reibo no te is, in effect, two series, which are defined by the two occurrences of the note e-flat” (marked ‘X’ in figure 81). The e-flat” notes are not present, however, in the initial series of notes in any of the versions that are nominally considered to be the same lineage as Uramoto, that is, the FR versions of Watazumi, Yokoyama, and the SR versions of Watazumi, Yokoyama, and Iwamoto (figure 80 ‘A’). The note, e-flat”, is, on the other hand, present in all of the versions that are considered not the same lineage as Uramoto, the FR and SR versions of Jin and Sakai (figure 80 ‘B’). In this instance, the data again points clearly to Watazumi as the source of innovation in transmission. In this case the two occurrences of the note e-flat” are part of what was transmitted from Konashi to Uramoto and Jin, but was later omitted by Watazumi.

In Uramoto’s version, if a g” follows an a-flat”, the next note is always an e-flat”. The same rule applies to both the FR and the SR versions of both Jin and Sakai (bracketed in figure 82). The simple a-flat” – g” – e-flat” series of notes does not occur in the other versions of Uramoto’s FR lineage, those of Watazumi and Yokoyama. In contrast, in both Watazumi’S and Yokoyama’s FR versions, a variety of combinations of a-flat”, g”, c”, and d” notes are performed before the e-flat” note (figure 83 ‘A’).

The simple a-flat” – g” – e-flat” series of Uramoto’s FR version can be found, however, in the SR versions of Watazumi, Yokoyama, and Iwamoto (figure 83 ‘B’). But the complex series that were not found in Uramoto’s FR version also occur in the SR versions of Watazumi and Yokoyama, though not in Iwamoto’s SR version (figure 83 ‘C’).

This series of notes in Uramoto’s version is more similar in its simplicity to Jin and Sakai’s versions (of both groups) than it is to the more complex series in Watazumi’s and Yokoyama’s versions (of both groups). The exception to the above pattern is Iwamoto’s SR version, which, is more like Uramoto’s, Jin’s, and Sakai’s versions than like Watazumi’s and Yokoyama’s versions. This exception cannot be explained by either lineage or by such elements as the passage of time between transmission and performance.

The ending of phrase F:U22, and the corresponding phrases in the other versions, (figure 84) further illustrates the complexity of similarities and differences not only between Uramoto and Watazumi, but also between all ten of the transcriptions. Looking first at similarities, a reibo no te ends with a sustained a-natural” in the FR versions by Uramoto, Watazumi, and Yokoyama (a similarity within a single lineage in a single group). This also corresponds to the a-natural” in the SR versions of the Watazumi-Yokoyama-Iwamoto lineage (a similarity within a single lineage in two groups). The a-natural” is also the final note in Sakai’s version, but it is extremely short in duration (slight similarity between different lineages within the same group). The short notes after the a-natural” in Watazumi’s and Yokoyama’s FR versions parallel their SR versions (similarities within the same lineage but in different groups). These are some of the ways in which the identity of “Reibo” is maintained in various performances.

Turning to the differences between the performances, the a-natural” is the final note of the phrase in F:U22, but is followed by three short notes (b-flat” – a” – g”) in F:W and two short notes (g” – g”) in F:Y (differences within a lineage within a group). The corresponding note in the F:J version is also the final note and is short in duration, but is an a-flat” instead of a-natural” (differences within a lineage within a group).

One can only conclude from reibo no te that while similarities and differences most often correspond to shared lineages and/or groups, this is frequently not the case. The manner in which similarities and differences between reibo no te fall sometimes along the lines of the group, sometimes along the lines of lineage with a disregard to group, and sometimes along a combination of the lines of both group and lineage can be observed throughout the transcriptions of the piece and on all levels of analysis.

Throughout the above analyses, one of the most commonly recurring patterns is the frequent involvement of Watazumi when variation between shared lineages or groups do take place. As noted elsewhere, this is to be expected in light of Watazumi’s rejection of Uramoto as his teacher of “Reibo”. Watazumi’s assertion that he did not learn “Reibo” from Uramoto might best be interpreted, however, less as an assertion of fact and more as a statement concerning the relative freedom one must have to develop one’s own version into honnin no kyoku or ‘one’s own piece’, not bound by tradition (p.229). Watazumi rejects the doctrinaire approach to honkyoku performance with the spectacular incongruity between Uramoto’s performance and his own. By making it difficult to establish connections between himself and others, both in his performance of “Reibo” and in his giving the title “Furin” to one version of the piece, Watazumi emphatically affirms his insight into the nature of performance and transmission of classical shakuhachi honkyoku

Yokoyama addressed this issue, and at the same time, provided a partial explanation to the high degree of variation between the ten versions with the following statment:

“Playing honkyoku is a way of searching for one’s own ideals. So when you transmit a honkyoku, you are also trying to transmit your ideals. In doing so, it is essential to eventually cut yourself off from your teacher, while at the same time still respecting your teacher. You must be as strict as that, otherwise the transmission of your own ideals will not occur.”

(Yokoyama 1989)

Applicable as this may be to the above analyses and to the honkyoku tradition in general, Watazumi goes much further than what Yokoyama suggests in the above quote. Yokoyama’s instruction regarding ‘cutting off yet still respecting one’s teacher’ is not even an issue with Watazumi. Since Watazumi emphatically insists that he does not have a ‘teacher’, there is neither a teacher to cut off, nor one to respect. Watazumi takes the discussion one magnitude higher by stating that he does not even play honkyoku. Instead, he performs dôkyoku (, pieces of the Way). Furthermore, his instruments are not shakuhachi. Instead he plays dôgu (, tools of the Way). The authority Watazumi exercises over his performances of dôkyoku with his dôgu is absolute. They are just “That”.