Yielding a Performative Analysis of Kaija Saariaho's "Duft"

This post is based on a presentation I will be making tomorrow for my Analysis, Theory and Performance course at the Sibelius Academy in conjunction with the Eastman School of Music. The course consists of presentations made by all students that discuss analysis as it relates to performance. For the presentations, we are paired or grouped with student(s) from Eastman School. I am working with a saxophone player, and our presentations will be comparative analyses of solo works for clarinet. He is presenting on Joan Tower’s work, Wings, and I will be presenting Kaija Saariaho’s Duft. I have fortunately played Wings, so this presentation ends up being quite like the lecture recital recital I presented in Tallinn September 2018 with solo works by Markku Klami and Libby Larsen. Except this time, I only have to do half the playing!

Like my Royal Academy and Sibelius Academy study day presentation last October, I chose Duft for practical reasons - I can perform it alone - and artistic ones - it compliments and expands my work on Oi kuu, which I will be playing on ‘Elollinen’ in April. Duft and Oi kuu, however, are different in many ways. Oi kuu, composed in 1990, relates musically to the electronic works that Saariaho composed in the 1980s. It focuses in on harmony and timbre in ways very similar to Nymphea. Of the opening of Nymphea, Michael Rofe writes “not only does the opening gesture expand and diversify in harmonic and timbral content, but there is a broader, synergetic expansion of harmony timbre (Hargreaves 2017: 85). He could have just as well been writing about Oi kuu! But much less directly to Duft. In researching Saariaho, the various texts I studied helped me a great deal with my work on Oi kuu, but less so on Duft. I could recognize the sound/noise axis in Oi Kuu. I could identify moments of stasis and dynamism, and the temporal prolongation. But these elements, which were presented as integral to Saariaho’s compositional method and aesthetic ideas, were less “obviously” composed in Duft.

I think that my tendency when trying to traditionally analyze a work is to expect that my analysis should provide all the answers and clarify everything all at once. I also find that analyzes tend to narrow rather than broaden my conception of a work, or what a work can be in performance. And I also have felt that ‘analysis’ limits performance, in many ways, by setting invisible rules that one has to abide by. I am sure that many a theorist or musicologist would chastise my sentiments as juvenile or inexperienced, or simply the result of having produced an inadequate analysis. I have to admit that I was initially skeptical of the idea of generating an analysis relating to performance, feeling that if it related to performance, it must therefore be unacademic, ungrounded, or merely instinctual. But In titling this presentation - “Yielding a Performative Analysis of Kaija Saariaho’s Duft (2011)”, I will attempt to merge the ‘traditional academic scholarship’ I have done in terms of research and formal analysis, with my experience of practicing and playing around with the piece over the last six months or so. The goal is to yield an analysis that reflects and contributes to performance of the work, and hopefully to theoretical or analytical perspectives as well.

NARRATIVE, SOUND WORLD, INDIVIDUALITY

I framed my analysis of each movement in consideration of three elements - narrative, what I call ‘sound world’, and individuality. The first two are more purely analytical, they are also common analytical tools in many of the essays in the 2017 Hargreaves edited Saariaho book. The third, individuality, is based more in performance, but I argue that it is very much a composed quality in Saariaho’s work(s).

Duft is one of about seven or eight works out of 120 (based on my own assessment of her works list) with a German title. The language is due to it having been commissioned for the 2012 clarinet competition in Freiburg, Germany. Compared to Dancing Solo and Wings, and quite many of Saariaho’s other works, which were both composed for a particular player, this work was intended to be performed by younger clarinetists in the context of a competition.

The title, regardless of language, is important. Titles are very significant in Saariaho’s works, as they help “define the musical material which, according to [Saariaho], is the most complicated part of composing” (Moisala 2009: 60). Duft is a noun meaning smell or scent; the first movement “Blütenstaub” is also a noun meaning pollen, while second and third movements are adjectives: “Blühend” translates to flowery, in (full) blossom, or radiant, and the third movement “Flüchtig” translates to fleeing or in flight. In her analysis, Kathryn Vetter remarks that “each of the three moments, ‘Blütenstaub’ (‘Pollen’), ‘Blühend’ (‘In Blossom’) and ‘Flüchtig’ (‘Fleetingly’), refers to a motion of smell” (Vetter 2018”9). The three movements are very different in musical approach, and the titles are connected to these musical qualities.

When I am playing a work, especially by a composer I am unfamiliar with, especially a work that is contemporary, I find it very important to figure out the ‘sound world’ of a composer. I think it is particularly important for composers, like Saariaho, who work specifically with the acoustic properties of sound. Traditionally, ‘sound world’ was considered synonymous with ‘harmony’ or harmonic tendencies, but pitch organization, or formal pitch analysis, is much less important than timbre in Saariaho’s music.

Saariaho has written extensively on her own approach to harmony and timbre. For her, timbre generates harmony and timbre can be charted across what she refers to as the ‘sound/noise’ axis. She writes:

For some years, for some years now I have a tendency in my music to relate the control of timbre with the control of harmony. Initially I began to use the sound/noise axis to develop both musical phrases and larger forms, and thus to create inner tensions in the music. In abstract and atonal sense the sound/noise axis maybe be substituted for the notion of consonance/dissonance. (Saariaho 1987: 94)

She continues that she considers timbre to function vertically, while harmony functions horizontally (Saariaho 1987: 94). This perspective is quite the opposite of older western classical music, where harmony is often analyzed and considered vertically, while melodic line (and by extension, timbre, though it usually is not considered as a separate entity) is considered horizontally.

Finally, I propose that Saariaho composes individuality, or creative freedom, in this and many of her works. She worked very closely with particular instrumentalists and wrote many works specifically inspired by and for them. I have found that in works that have been written for particular players, there tends to be less ‘over notation’ and a greater amount of creative liberty composed in the work. Saariaho goes further by composing measures in all movements that offer the opportunity for improvisation and/or free interpretation of the written marking (vibrato, glissando, and slow pitch bending).

MOVEMENT 1: BLüTENSTAUB

How do we play like pollen smells? This was the narrative question I posed by the title of this movement. To capture the smell of a noun, what does that mean? Is it is the action of the noun? The way the noun makes us feel? As someone with seasonal allergies, I’m sure my reaction to pollen is quite different from some else’s!

All jokes aside, keeping in mind the idea of ‘pollen’ framed my approach to sound quality and color, as well as my treatment of the tremolos, trills and frullato in this movement. This movement is temporally suspended, there is little in terms of a formal ‘melody’, and the focus is on timbral and sonic expansion through trills and tremolos, which the entire movement consists of. The ‘clarinet’ tendency would be to play the trills and tremolos melodically, emphasizing the harmony, and probably putting a tenuto at the start of each one. Keeping in mind both ‘narrative’ and ‘sound world’, I try not to do that in my performance. The tempo, quarter note equals sixty beats per minute, is not quick but it is not slow either. It is steady and natural, connected to the organic quality of the narrative title.

The movement sonically expands by keeping the same basic harmonies, and developing sound and progressing in time through timbral expansion. For every movement, I listed the sonic and timbral qualities and notations, and I tried to decide whether they were predominantly harmonic or timbral in function, noise or sound. For this movement:

-tremolo (harmonic - horizontal) (sound)

-slow vibrato (timbre - vertical) (sound/noise)

-glissando (timbre - vertical) (noise?)

-flutter (timbre - vertical) (noise)

-grace notes (harmonic - horizontal) (sound)

Already, through the process of my analysis, I find myself changing my mind on whether something is harmonic or timbral, noise or sound, but the purpose here, for me, was not to unequivocally decide what was what. The purpose was to begin trying to think in terms of Saariaho’s concepts, and think more creatively how to translates these analytical concepts in performance, in conjunction with the narrative. For example, my approach the tremolos was that they were quick, but not technical, with warm air quality and color, but with a bit of air in the sound (a technical as well as musical consideration for maintaining consistency in the soft dynamics). The frulatto emphasized the tongue sound (or growl sound, for me) rather than the pitch, and the frulatto with the glissando were contrasts in timbral quality rather than harmonic or sonic development of a particular pitch.

Approaching performance from this perspective, has kept my creative attention and encouraged active participation in performance every time I put away and return to this work. None of the movements are particularly challenging technically, and the the notation throughout appears quite simple. But questioning and re-examining the musical content in this manner - forcing myself to think more critically about how I am playing things and why - prevents me from getting lazy or habitual in performance. It also would keep the process of working on this piece very active, with my translation of the score continuing to evolve and develop.

Decisions on sound quality (to put air in the sound, to make it pure, to have it fuzzier), the decision about tenutos on the trills, the pacing of the rests (are they metric or grand pauses?), reacting to the acoustic space to determine volume, trill/tremolo speed and sound quality, and the speed of the trills compared to tremolos are just a few examples of the creative individuality and freedom that Saariaho composes in this movement.

MOVEMENT 2: BLüHEND

The narrative elements implied by the title of this movement are also not obvious. Blühend is an adjective that means flowering or in bloom. The title implies something ‘spring-like’, but the tempo of the movement is very slow, quarter note is marked forty-eight beats per minute. It is as if Saariaho is ‘blossoming’ in slow motion or time lapse, much like her experiments slowing down tape samples to put sound under a microscope.

This movement has the most stylistic instruction of any movement, with fourteen markings over forty-one measures of the second movement: dolce, con amore (at the beginning), piu energico /sempre espressivo (mm. 4-5), con amore (mm. 8), piu doloroso (mm. 10), intenso (mm. 13), piu energico (mm. 16), doloroso (mm. 20), espressivo, calando (mm. 22), misterioso (mm. 26), doloroso (mm. 29), con amore (mm. 31), poco a poco più doloroso (mm. 33), and cantabile (mm. 38). These markings can be expressed through changing the tempo, but I think they are a guide to the clarinetist on how to express the narrative content of the movement; a guide to the slow motion ‘bloom’. There is no ‘right or wrong’ way to interpret the markings, but they serve to guide the clarinetist to vary the sound quality, vibrato and timbre within the written dynamics.

This movement is mostly ‘sound’, in terms of the sound/noise axis. The variety of harmonic and timbral notations in this movement are:

-vibrato (timbre - vertical) (sonic expansion, not noise?)

-improvised trills (harmonic - horizontal?) (sonic expansion, same harmonies as melodic lines)

-glissando (3, seem more harmonic - horizontal)

-flutter (timbre - vertical) (noise)

-grace notes (harmony - horizontal)

They are quite similar to the first movement, but much used much more sparingly, and to different musical effect. The vibrato, which in the first movement was a timbral variation on the written D-sharp (mm. 17), is notated as molto vibrato and vibrato ord. (mm. 35-36) as a timbral intensification to the ff climax of the movement in mm. 36. The vib ord. is also a reminder to the clarinetist that vibrato can be used throughout the movement and is not necessarily specified through notation (as it is in a work like, Oi Kuu, where vibrato is used very specifically for timbral expansion).

Similarly, there is one glissando in mm. 28 of the first movement, which intensifies the harmony and begins building intensity to the climax in mm. 29. In the second, and later third, movements, the glissandi are used more and more, developing as their own harmonic (or timbral?) sonic device until their final, most expressive iteration at the very end of the piece (movement 3, mm. 77-83). In the second movement, the glissandi are used at the end of the phrases in mm. 17 and 24, but also in mm. 35-36, again intensifying the horizontal harmony towards the climax (ff mm. 36).

In my performance analysis, the grace notes that are within the melodic line are horizontal/harmonic (like in the first movement), while the improvised grace notes with the trills and tremolos (mm. 3, 18, 28 and 40) are a combination of timbral (the trill/tremolo) and harmonic. The different ‘purpose’, to me, also means that the grace notes that are improvised in the trills have a bit more tenuto, as focal points, compared to the harmonic/melodic grace notes.

Analytically, the harmonic/melodic lines also development horizontally, with motion implied through the rhythmic development. In her thesis, Vetter applies an analytical method called stream segmentation to identify multiple voices in a single monophonic line. Taken from Dimitrios Rafailidis’ research, a stream segment is “a relatively short coherent sequence of tones that is separated horizontally from co-sounding streams and, vertically from neighboring musical sequences” (Rafailidis 2008: 84). Vetter applies stream segmentation to all the movements in Duft, and while analytically viable, I do not necessarily find the analysis applicable to performance. Guided by Saariaho’s narrative implications, I do not experience multiple voices within each phrase, but rather a single process of blooming in this movement.

MOVEMENT 3: FLüCHTIG

The third movement, ‘fleeing’ (‘in flight’, ‘brief’) is also an adjective and action of smell, rather than a quality. The narrative quality of ‘flüchtig’ is composed through much quicker tempo than the previous movements, quarter tone equals 104 beats-per-minute, and through rapid, short phrases. Most of the phrases have fermatas in the penultimate measure and are separated from the next phrase by only an eighth rest. The exceptions are measures 22, 33 and 58. To capture this narrative quality in performance, I do not allow myself (or the listener) to settle too long on the end of any single phrase. I force myself to go on, almost tripping over myself but giving, musically, a very breathless and brief quality. It is not frantic, Saariaho indicates energico, capricioso at the beginning, but not nervioso.

Along the sound/noise axis, I have been thinking in performance of moving almost from sound/consonance to ‘noise’/dissonance throughout. The piece does not end with ‘noise’ per se, but with a long, seven-measure glissando up to a high E, with slow vibrato indicated to non vibrato in the very end. The variety of musical notations (trills, slow vibrato, glissando, frulatto (notes), frulatto (scales), pitch drops and grace notes) are all prevalent the other movements, but with quicker changes between timbral and harmonic effect, such that their functions are constantly overlapping and changing.

For instance, the opening trill (ppp mm. 1) is both timbral/vertical and harmonic as it sets the timbre and leads to the melodic horizontal motion in mm. 2. To me, the trills also relate to the narrative of the first movement, where they can be almost bee-like (pollen -> blossom, bees, fleeting -> bees?). The frulatto in this movement, which were predominantly timbral in the previous movements, are both timbral and harmonic in this movement. The frulatto on individual notes are predominantly timbral/vertical, while the frulatto scales have also a harmonic or horizontal nature to them.

I find the performance analysis most challenging with the glissandi in this movement. Musically, I believe their purpose is to suspend and pull tension to the end of the movement/work. They are measured, such that it is indicated how long each half-step ascent should last (two beats), and I believe that the audience is meant to hear the chromatic ascent rather than a long glissando from A-sharp up to high E at the end (or F-sharp to A-sharp mm. 56-58, F-sharp to A-sharp mm. 61-64, E-sharp to A-sharp m. 69-74). The four longer glissandi phrases are developed from the short half step glissando in mm. 59-60 and the whole step glissando mm. 52-53. A half, or whole step, however is not a far distance and the musical execution leads me to believe that the glissandi are both harmonic and timbral.

Also challenging, musically, in this movement are the slow vibrato indications. While slow vibrato is used only once in the first movement (timbral variation coming out of the multiphonic in mm. 17), and the vibrato use in somewhat free in the second movement, the use of slow vibrato is very deliberately indicated in nearly ever short phrase of the final movement, usually on the final note of the phrase. I understand the musical ‘purpose’ to be timbral rather than harmonic, but I am challenged to find a slow vibrato that reflects a subtler timbral effect than a pitch bend or pulsation.

CONCLUDING REMARKS

Even as I have been writing this post, and the more I practice Duft, my analysis of performative aspects of this work are constantly evolving. In researching existing analyzes for this presentation, I found a number of formal methods others have used to understand Duft - contour analysis, stream segmentation, pitch organization charts, intervallic analysis. These are all valid, but none have enhanced my understanding of the work from a performance perspective, nor do I inherently understand how to apply them to my performance. Taking a few analytic elements - improvisation, narrative and sound - that directly apply to performance have yielded a very different kind of analysis, but one that I feel works in direct relation to playing the piece and continuing develop of my performance approach.

Bibliography

Dimitrios Rafailidis, Alexandros Nanopoulos, Yannis Manolopoulos and Emilios Cambouropoulos, “Detection of Stream Segments in Symbolic Musical Data,” International Society of Music Information Retrieval (2008): 83–8.

Hargreaves and Jon. 2017. Kaija Saariaho: Visions, Narratives, Dialogues. 1st ed. Routledge Ltd. doi:10.4324/9781315092218. https://www.taylorfrancis.com/books/e/9781315092218.

Howell, Tim. 2007. After Sibelius. Aldershot [u.a.]: Ashgate.

Moisala, Pirkka. 2009. Kaija Saariaho. Urbana: University of Illinois Press.

Raasakka, Mikko. 2010. Exploring the Clarinet. Helsinki: Fennica Gehrman.

Saariaho, Kaija. Duft. London: Chester Music Ltd, 2011.

Saariaho, Kaija. 1987. "Timbre and Harmony: Interpolations of Timbral Structures." Contemporary Music Review 2 (1): 93-133. doi:10.1080/07494468708567055. http://www.tandfonline.com/doi/abs/10.1080/07494468708567055.

Vetter, Kathryn. 2018. Slowing Down and the Vertical/Horizontal Spectrum in Kaija Saariaho’s Duft (Master’s Thesis). SHAREOK Joint Institution Repository.