1
Multiphonics for the Oboe Nora Post
Reproduced from INTERFACE by arrangement with the editors.
ABSTRACT This paper explores four aspects of multiphonic sounds as they pertain to the oboe definition, acoustical properties, notation, and compositional usage. The article provides a detailed multiphonic fingering chart as well as a brief discussion of (1) the notation of oboe fingering generally and the multiphonic fingering chart in particular; (2) slurred multiphonics; and (3) the factors which affect the reliability of oboe multiphonics.
DEFINITION AND NOTATION
The impact of multiphonics on today's repertoire is undisputed. Of all the recent expanded techniques for the woodwinds, the interest in, and development of, multiphonics has been far more extensive than that of any other "new" sound. This development has been advantageous for the oboe, which possesses an extraordinary ability to execute a great variety of multiphonic sounds. The oboe can make an almost imperceptible pianissimo multiphonic entrance - and it can produce an immediately incisive fortissimo attack. Well-chosen multiphonics can be articulated at the same speeds as the most rapid single sounds, including double and triple tonguing. More than anything else, it is this responsiveness which accounts for the reliability - and consequent attractiveness - of oboe multiphonics. There are several types of multiphonics for the oboe. But first, a working definition is essential. What, exactly, is a multiphonic? Bruno Bartolozzi (1967, p. 35) offers one explanation: "... the generation, at one and the same time, of a number of frequency vibrations in the single air column of an instrument." Or, as Arthur Benade (1976, p. 565) describes it: A multiphonic oscillation is made up of a collection of components whose frequencies are connected to one another by an elaborate set of heterodyne relationships. The ordinary tones of woodwind instruments also fit this description, but the frequency components in normal tones are limited to those belonging to a single harmonic series. As to the process involved: "These chords result from fingering patterns which in principle provide several tube- lengths on which to provide composite tones." (Read, 1976, p. 150). Does singing and playing simultaneously constitute a multiphonic? Technically speaking, singing/playing and multiphonics are both the same acoustic phenomena, although they are produced by different means. On some wind instruments - the flute
2 and saxophone, for example - singing and playing sometimes cannot be distinguished from multiphonics. [1] (This is not the case with the oboe, due primarily to its limited capacity for singing and playing.) But, while singing and playing is the result of one tone being played and one being sung, multiphonics are produced exclusively by the vibrating system of the instrument. Since singing and playing do not satisfy this criterion, they will be excluded from my comments. There is considerable diversity about the manner in which multiphonics are notated. From the performer's viewpoint, any notational system is acceptable if it efficiently conveys accurate information. This is no easy task. Bartolozzi sought clarity via numbers representing the keys of the woodwind instruments (Fig. 1). His system had a substantial influence upon composers during the years immediately following the publication of his book. Of late, this influence has waned. Performers find his notation awkward because it requires that arbitrary numbers representing each key of an instrument be memorized. It is quite natural for an oboist to see "B" written in a fingering; it is time consuming to see "9" written in a score, forcing the performer to consult Bartolozzi's book to decipher the composer's intention and then to translate Bartolozzi's notation into standard fingering symbols. It is unfortunate that a number of outstanding composers employed a system which creates unnecessary difficulties for the performer. I have devised what is, in my opinion, a simpler and clearer notation for multiphonic, as well as other, fingerings. It entails depicting the six main tone holes of the oboe, including a dividing line between right and left hands, and then giving letter names for all other keys. Fig. 2 indicates two fingerings of the same multiphonic - the left utilizing
the Bartolozzi system, the right using mine. Aside from variants of the two systems of multiphonic notation described above, other multiphonic notation deals with indeterminate or only partly-determined multiphonics. David Cope [2] employs the sign to indicate indeterminate multiphonics. Vinko Globokar [3], on the other hand, utilizes several different multiphonic notations, depending upon the context. Most often, however, he uses the symbol "M" above a note, and places under it either a number specifying how many pitches should be in the multiphonic or alternatively, provides pitchless noteheads representing the total number of tones desired. (Fig. 3)
3 There are two groups of multiphonic sounds for the oboe. The first is what Bartolozzi calls "homogeneous chords". These are "...all sounds of approximately the same volume and colour..." (Bartolozzi, 1967, p. 42). The second group, which he calls "broken sounds", he describes as: ...chords containing sounds of different tone colour are obtained by emitting simultaneously two sounds which are close together, with their relative harmonics. In this way it is possible to exploit the phenomenon of "beats" caused by the interference in sound vibrations occurring when two sounds are emitted which are very close together, within the maximum space of a semitone. There are substantial differences between the sound and production of "beating" and normal oboe multiphonics. For that reason, beating multiphonics will be treated as a special subset. COMPOSITIONAL USAGE
The uses of multiphonics are almost as numerous as they are diverse. The examples discussed below illustrate the vast variety of available multiphonic possibilities. Lucas Foss and Peter Schat have each written works which utilize oboe multiphonics for virtually their entire duration. The oboe part for Schat's Clockwise and AntiClockwise, [4] written in circular form, consists of twelve groups of multiphonics for three oboes. Beginning at any point, the players proceed either clockwise or counterclockwise, changing multiphonics at the conductor's signal. The effect is a homogeneous wash against which the rest of the piece is set. While the oboe part includes a catalogue of fingerings and pitches, the actual "boxes" of multiphonics show only fingerings (Ex. 1). Foss's The Cave of the Winds, [5] uses multiphonics, borrowed in toto from Bartolozzi, for all four woodwinds. Included in the score are Bartolozzi's drawings of the four instruments with his numbering of the keys. An unusual version of multiphonics has been the "orchestration" of their pitches - that is, having certain pitches doubled by other instruments in the ensemble. Ursula Mamlok [6] employs this device to reinforce the individual pitches of several sustained multiphonics in her Concerto for Oboe and Orchestra. In similar fashion, Paul Zonn [7] requires each of the other instruments to enter separately on one of the pitches of a sustained oboe multiphonic. Multiphonic ostinati patterns occur occasionally. Peter Salemi [8]calls for numerous ostinati in his Riff 70/71, each of which is played at one of three recurrent speeds. Paul Earls [9]demands something more difficult - the English Horn begins a slow, soft ostinato between several multiphonics; each time the pattern is repeated it must crescendo and accelerate. Furthermore, a new multiphonic must be added at each repetition. When all the indicated multiphonics have been added, the ostinato ends. Multiphonics are frequently employed in conjunction with other sounds, conventional or otherwise. A simple combination is the arrival at, and/or departure from, a multiphonic via one of its pitches. Although often encountered in the repertoire, this is possible only with certain multiphonics (see Remarks on Multiphonic Fingering Chart). However, one cannot always precisely predict when these multiphonics will sound.
4 Composers usually allow some time, consequently, for the embouchure to achieve the desired effect. Ton de Kruyf takes this into account when he adds F5 to the C6 (Ex. 2).
In contrast to most multiphonics, which speak tutti, there are some whose pitches can be phased in one by one (see Beating Multiphonic Fingering Chart). Occasionally, fluttertongue is added to a multiphonic. The result depends on the multiphonic chosen, for the denser multiphonics (i.e., those comprised of many pitches) are so complex that the added effect of fluttertongue is negligible. Two of the more successful examples of this technique are found in Toru Takemitsu, who adds singing to a fluttertongued multiphonic and then deletes fluttertongue from a multiphonic (Exs. 3 and 4). Multiphonics can sometimes be slurred together [10], although this demands a thorough knowledge of the oboe's fingering system. Iannis Xenakis does it quite expertly in the example given below (Ex. 5). [11] Composers often do not grasp the difference between a multiphonic trill and a multiphonic tremolo. A trilled multiphonic is one in which all the pitches of the multiphonic move in the same direction by a semitone. If the pitches do not change in this symmetrical fashion, what we have is, technically speaking, a tremolo. Consequently, many multiphonics marked "tr" are not actually trills, but tremolos. While most multiphonic trills and tremolos are executed the same way - i.e., only one finger moves - this does not necessarily produce a symmetrical semitone trill of all the pitches in the multiphonic. Takemitsu clearly distinguishes between trills and tremolo added to multiphonics (Ex. 6). In most cases, though, composers use the "tr" indication regardless of the internal construction of the multiphonic. Oboe multiphonics do not have much pitch flexibility. Individual notes within a given multiphonic cannot be tuned. Although some multiphonics lend themselves to semitone pitch bend, most do not. Recognizing this limitation, Earls asks simply for whatever degree of multiphonic pitch bend possible (Ex. 7). Xenakis uses special fingerings to extend the limited range of multiphonic pitch bends (Ex. 8), calling for a semitone glissando via fingering change. (He also requires that fluttertongue and trill be added.) Some composers ask the oboist to sing while playing multiphonics, as illustrated by the Takemitsu example cited earlier (Ex. 3). Globokar seeks a more complex effect by employing a series of vocal entrances, including vocal glissandi (Ex. 9). However, as with simultaneous singing and playing, the effectiveness of this technique is less than overwhelming. Beating multiphonics are rarely encountered because few composers and performers are familiar with them. Most instances consist of two predominant pitches, usually a semitone apart (see Multiphonic Fingering Chart). If the performer begins the sound
5 quietly enough, one of the two main pitches can frequently be produced alone, and then the beating pitch added. The speed of the beating is influenced by the dynamic level of the multiphonic- the beats tend to be slower when the dynamic level is soft, and to increase in speed as the dynamic increases. Therefore, some degree of beating control is possible. Furthermore, the speed of some beating multiphonics can also be controlled by the embouchure. The multiphonic shown below, based upon a predominant F4, is an excellent case in point (Fig. 4). One of the few published examples of beating multiphonics is found in Foss's The Cave of the Winds which, incidentally, also provides a particularly good fingering (Ex. 10).
REMARKS ON MULTIPHONIC FINGERING CHART 1. Fingerings
Fingerings will be based upon the six major keys (tone holes) of the oboe, which will not be identified by letter name:
Any other keys, when utilized, will be given written symbols (consisting of letter names and/or numbers), which will appear in the location of their position on the oboe. These symbols will be shown to the right or left of the six main tone holes, as if the reader were looking directly at an oboe. Duplicate keys will not be specifically indicated, since their location will identify them as such. The symbols for keys are as follows [12]:
6
[ Venting for the six main holes will be indicated:
Thus, the multiphonic fingering shown below would be executed: left hand: B and A keys depressed, G key depressed only on rim leaving cent hole open, low B key depressed. right hand: F# and E keys depressed, D key open, C key depressed 2. Slurred Multiphonics
Only some multiphonics can be slurred. Inability to slur many multiphonics results from the fact that most players cannot gradually slide one finger off one key and onto another. As we can see in Fig. 5, the fifth finger of the left hand cannot readily slide to
7 another of the five keys it operates. [13] Similarly, the fifth finger of the right hand usually cannot slide among the three keys it controls. [14] While a few unusually skilled oboists can overcome some of these difficulties, they are exceptional. Half-hole fingerings
as well as those in which only the rims of the keys are
depressed do not jeopardize the ability to slur. A passage by Takemitsu, going from a G#4 to a multiphonic and then back to the G#4, requires that the second finger of the left hand slide to uncover the center hole of the key (Ex. 11). Given the complexities of slurred multiphonics, interested composers would be well advised to consult an oboist. 3. Reliability
The reliability of the multiphonics listed in the following pages will vary somewhat for each player depending upon differences in, among other things, embouchure, reed materials and instruments used, individual reed making techniques, and national reed making styles. From the acoustical point of view, the effect of just the reed alone is substantial (Benade, 1976, p. 566): The internally measured spectrum can he changed drastically when the player makes small changes in the size of the reed cavity (this is particularly true among the conical instruments). The reason for this is that cavity changes shift the frequency relationships among the air-column resonance peaks, and so oases the subtle interplay between them and the generated components. The relationship between the internal and the external spectrum of a multiphonic sound is not of the simple sort we find for more normal woodwind tones. While these acoustical and instrumental factors are too large a topic to consider here, their effect on any attempt to standardize the reliability (and response) of different fingerings must be taken into account. In general, multiphonic production requires more embouchure pressure and/or more reed in the mouth than conventional playing. Thus, problems encountered in producing these sounds will result most likely from either the reed or an inappropriate embouchure. Players should realize that reed flexibility is the critical prerequisite to the production of multiphonic sounds; only a flexible reed will respond to the embouchure adjustments needed to avoid the pitfalls of multiphonics (Benade, 1976, p. 565): On the pianissimo side, the instrument may lapse into producing an essentially pure tone based on the tallest resonance peak, and on the loud-playing side the reed may simply choke up and snap closed if too many of the generated components lie at dips in the resonance curve. Because even the most reliable multiphonics vary in their response characteristics, each multiphonic will be coded by number to indicate: (1) dynamic range, (2) reliability/stability, and (3) ease of response. Dynamic range will be indicated by the standard symbols. Ease of response and stability/reliability will be given letter namesA, B, or C, to signify the following:
8
Reliability/stability •
•
•
A = highly stable/reliable B = moderately stable/reliable C = usually stable/reliable, but somewhat difficult to produce
Ease of response •
•
•
A = highly responsive B = moderately responsive C = moderately responsive, but somewhat difficult to produce
Thus, a typical multiphonic entry might read: 1. pp-ff (dynamic range) 2. A (reliability/stability) 3. A (ease of response) Multiphonics are listed according to their highest pitch, since this is often the pitch with the most reliable intonation. Where the player can slur either into or out of a multiphonic via one of its constituent pitches, that pitch is indicated by a fermata. A slur marking shows whether the single note preceeds or follows the multiphonic. Within a multiphonic, pitches marked by parentheses indicate notes which are either difficult to hear or which occasionally may not sound at all. Beating multiphonics are listed by the predominant pitch, with the beating note in parentheses. Frequently, the main pitch can be produced alone, and then the beating effect added. This usually entails a very low dynamic level, and the results vary from player to player. 4. Intonation
Multiphonic fingerings below utilize the following system to indicate microtonal intonation:
= 1/4 tone sharp
= 1/4 tone flat
9
= slightly sharp
= slightly flat
10 MULTIPHONIC FINGERING CHART
11
12
13
