Paweł Szymański (b. 1954) is currently one of the leading Polish composers, with works for orchestra, solo instrument and orchestra, chamber ensembles, solo instruments, choir, vocal-instrumental ensembles, an opera, film, theatre and radio music, as well as music for tape.
The basis for Szymański's creative work is what he describes as a two-level technique, composing not with the use of basic sound characteristics and simple relations between sounds, but with entire objects and complex gestures deriving from various musical traditions which, at subsequent stages of work, are subjected to transformations.Footnote 1 The first step of this technique is to create a complex and conventional structure (an arrangement of sounds). In compositions for solo instruments this often assumes the form of a tonal and recognisable melody or a tonal sequence of chords, while in works for more instruments it may take a tonal contrapuntal form, such as a tonal canon or a fugue. It is very important for the structure to have a clearly defined musical style, since – according to the composer – ‘in case of an entirely abstract model it would be impossible to distinguish the starting point from its transformation’.Footnote 2 In most cases the composer creates his structures in baroque and classic musical styles.
When an initial structure has been created, it is time to start transforming it in a variety of ways ranging from free to strictly algorithmic ones. In the opinion of the composer:
Music may be created by use of very algorithmic methods, as well as by ways of searching which are improvisational in their character. There is an enormous variety of means which can be found between those two extremes. When I write music, I sometimes move closer to one of them and sometimes to the other.Footnote 3
The number of ways in which a structure can be transformed is basically unlimited: a line can be split in two, with alternate sounds in the first and second voices; a melody can be presented in a vertical form, and so on. In the case of the algorithmic method the point is to construct ‘a mechanism which produces sounds and relations between them rather than only sounds and their configurations’.Footnote 4 Because in many cases, particularly in orchestral works, it is not enough to transform the concealed structure, Szymański adds further layers, each of which is a transformation of the initial structure. The technique presented above does not, however, appear in every work by Szymański, and certainly not to the same degree. Apart from compositions where this technique is clearly present, such as Partita III for harpsichord and orchestra, the composer's works also include such ones in which it recedes into the background as in, for example, Phylakterion for 16 voices and percussion instruments.
Since 1978 and Partita II for orchestra, Szymański has been moving more or less within the same world of music ideas and he has created a compositional idiom, recognisable from the first bars of each of his compositions. This idiom consists, among others, of short sounds, glissandi, vitreous sounds, numerous pauses and a two-movement form, usually resulting from the application of two different methods of working with the same material.Footnote 5 Although Paweł Szymański's music has been examined and interpreted in a variety of ways, its intertextual reading is the one which ‘compels attention, which imposes itself’.Footnote 6 While listening to this music, one cannot help feeling that it contains, alongside completely new elements, something already known, heard before, familiar. Irina Nikolskaya, Katarzyna Naliwajek and Anna Granat-Janki have all written on this aspect of Szymański's music and this article continues the discussion.Footnote 7
Paweł Szymański has never disclosed what algorithms he applies, but the discovery of these algorithms allows us to understand an aspect of his music and also opens the door for further research, including intertextual research. My research confirms that (1) algorithms constitute a part of Szymański's intertextual poetics; (2) the composer uses them in order to extend the conventional, initial structure; and (3) decryption of the algorithm allows for reconstruction of the conventional structure. Research into these algorithms has thus led to intertextual research, using existing intertextual theories to create a research method appropriate for Szymański's music.
Those aspects of Paweł Szymański's music that are openly algorithmic and intertextual may be described with the use of research categories elaborated by Michael Riffaterre, a French literature semiotician, and Ryszard Nycz, a Polish literary expert, whose theory of wide-range intertextuality is, in a way, an extension of Riffaterre's theory. I also believe that the same research method may be applied to other, freely composed works. Using Riffaterre and Nycz's theories and, in particular, their research categories – exponent of intertextuality, interpretant, and text characterised by intertextual play – I would first like to define how these categories should be understood in the case of Szymański's music and then consider two pieces, the first composed algorithmically, and the second more freely.
Categories of Intertextuality
The principal category of intertextual theory is the intertext, whose elements can be found in the text being studied. Riffaterre defines an intertext as
a corpus of texts, textual fragments, or textlike segments of the sociolect that shares a lexicon and, to a lesser extent, a syntax with the text we are reading (directly or indirectly) in the form of synonyms or, even conversely, in the form of antonyms.Footnote 8
Ryszard Nycz writes that relations between the text and the intertext can be divided into three kinds: text-text, text-genre or style (architext) and text-reality.Footnote 9 In most of his compositions Paweł Szymański refers to various intertexts, but references to well-known musical styles (architexts) are the dominant ones, a tendency confirmed by the composer himself.
Intertextuality Exponent
Riffaterre calls the fragments of intertexts discovered in the studied text ‘traces’ and divides them into grammatical and ungrammatical ones. Ryszard Nycz has considerably extended this aspect of Riffaterre's theory. In Nycz's theory Riffaterre's ‘traces’ are called ‘intertextuality exponents’ and can take one of the following forms: (1) logical-semantical, existential and pragmatical presuppositions; (2) grammatical, semantic and pragmatic anomalies; (3) attributions referring to ‘association of a given text or its fragment with certain contexts: other works and discussion areas, historically and functionally varied styles, genres and conventions which appear in the universe of an utterance’.Footnote 10
To date there has been no discussion of this aspect of Paweł Szymański's work, but it is widely known that he uses melodic-harmonic configurations as equivalents of intertextual exponents. The composer has also revealed, more than once, that his composing technique uses two levels: the first-level structure, also called the initial structure, which the listener does not hear, and the second-level structure, i.e. the first-level one after various transformations, which is what we hear in the composition. In order to differentiate between the non-deformed intertextual exponent and the one which is the result of transformation, I suggest talking about first- and second-level intertextual exponents, initial and ultimate ones. First-level intertextual exponents, representing various musical styles and specially composed by the author, are musical structures such as melodies, harmonic outlines or polyphonic constructions. It is interesting that second-level intertextual exponents do not operate locally, but appear in compositions from the beginning to the end, and are always clearly dispersed. What is more, such exponents are alternately grammatical and ungrammatical, which evokes in listeners a particular kind of response towards each composition.
Interpretant
The next category to consider in Szymański's music is the interpretant, or what Michael Riffaterre calls ‘the connection between what has already been said from the intertext and the new record, which is the text’.Footnote 11 The interpretant is a hidden metatextual element, which the recipient should be able to perceive and reconstruct, the reconstruction revealing the rules governing semantic play between the text and the intertext. Michał Głowiński explains that each interpretant consists of some ‘factors’.Footnote 12 In musicology the term ‘interpretant’ is rarely employed.Footnote 13 Despite the fact that the interpretants in Szymański's compositions are significantly different from one another, it is possible, looking at them from the most general perspective, to distinguish three main groups of factors: (1) the algorithm used to extend the first-level intertextual exponent; (2) non-obligatory transformations of the post-algorithmic second-level intertextual exponent, whose aim is to differentiate the course of the new, extended structure; and (3) aesthetic ornaments (articulation, dynamics, instrumental timbres) added to the transformed or untransformed second-level intertextual exponent. As each subsequent group of factors of the interpretant derives from the preceding one, this sequence is indicated as they are being discussed.
Text Characterised by Intertextual Play
According to Riffaterre the aim of literary research into a text is to capture its global meaning, taking into consideration the intertextual play which occurs in it.Footnote 14 Since music is a very different medium from literature, the issue of meaning requires some explanation. There are numerous definitions of musical meaning formulated within the areas of music theory, psychology, philosophy, semiotics and others. It is my opinion that in considering Paweł Szymański's music it is best to adopt a psychological approach, which takes into consideration the results of empirical research, as this is more universal and less complicated than other approaches. Using this approach researchers differentiate between two kinds of musical meanings: autonomous and designative ones. Autonomous meanings – defined as autotelic, absolute, syntagmatic and self-reflexive – are intrinsic to the music itself. The sound configurations that create the narration of a composition are related to each other and to the whole piece in a variety of ways, subordinated either to the rules of a convention or to the rules created by the composer. Such configurations ‘acquire meaning due to the fact that they constitute a part of a bigger structure or system; as a result, the whole (i.e. the musical work) also acquires its meaning’.Footnote 15 Other musical meanings – the designative ones – are those outside the music itself, and they depend to a great extent on guidelines and suggestions given by the title or the programme attached to the composition, or on the verbal text in a vocal piece. They can have either an intersubjective or subjective dimension, reflecting the so-called musical polysemy, i.e. the ambiguity of connections between the sign and the meaning. With regard to their origin, designative meanings are divided into icons, indexes and symbols.
In Paweł Szymański's music autonomous meanings are more important than designative ones, and they appear in compositions in a very characteristic way. First, there are usually two groups of different meanings: old and modern music. Secondly, the relations between old and modern meanings occur both vertically, simultaneously in time, and horizontally, linearly, within the passage of time. Because these two kinds of meanings continuously permeate Szymański's compositions, it is possible to discuss them as intertextual play, play of meanings, dialogue of meanings.
Two Studies for Piano (1986)
Paweł Szymański's Two Studies for piano were written in 1986, commissioned by Walter Lebhrart and performed for the first time by Szábolcs Esztényi during the Warsaw Autumn festival in 1990. The diptych has also been publishedFootnote 16 and recorded.Footnote 17 The Second Study is a single-voice melody, regularly flowing in semiquavers, played at a fast tempo in the upper register. It is, in a sense, mechanical, and it seems to refer to music that is already known, with fragments appearing every now and then of motives or phrases in a major-minor tonality. Given the uniform short note lengths, these tonal fragments evoke associations with Baroque sequential writing. To a listener familiar with classical music, the Second Study will be associated with an architext, the Baroque style.
An analysis of the score allows us to decode the algorithm, which, in turn, allows for reconstruction of the first-level intertextual exponent. The reconstruction consists in discarding the notes which are repeated and noting down the remaining ones in the same note lengths as in the composition, i.e. in semiquavers. Example 1 shows the opening fragment of the reconstructed first-level intertextual exponent. To make it more legible, the notes are shown in the middle register, in groups of four and equipped with orientation points, i.e. the Roman numerals marking the subsequent sections of the composition. Additionally, melodic progressions are marked by capital letters with numerals in the upper index.
Example 1: The author's reconstruction of the first-level intertextual exponent of the Second Study, fragment derived from staves 11–26 of the composition.
The reconstructed melody is in B minor, with fragments in closely related keys and has ‘melodic’ chords, mostly in fifth relations, together with insertion of dominants into the basic scale steps. The entire melody is modelled on Baroque sequential spinning, passing from one motive to the next and using repetitions of motives and melodic progressions. The features which depart from the Baroque style are the numerous diminished chords and two chromatic progressions. Irrespective of the degree to which the motives are rooted in the tradition, almost all the patterns of repetitions and progressions are of the following lengths: 4, 8, 12 and 16 semiquavers.
Algorithm and Aesthetic Factors of the Interpretant
Paweł Szymański transformed the Baroque melody into the Second Study using an interpretant composed of two groups of factors – extending and enriching ones. First he applied an algorithm multiplying the note material, using the following steps:Footnote 18
1. taking a motive of between 3 and 7 notes from the Baroque melody and combining it with an identical motive but without the last note; from a pair of such motives creating a single-voice desynchronising segment in which the notes of the longer motive are intertwined with the notes of the shorter motive; the extent of desynchronisation is determined by the Lowest Common Multiple of the two integers, where each integer denotes the number of semiquavers in the pair of motives;
2. adding an additional note whose pitch is the same as that of the last note of the longer motive and then taking from the Baroque melody the next motive containing from 1 to 14 notes and repeating each of the notes;
3. joining the desynchronising segment with the repetitive one;
4. treating both segments as one section, continue creating similar sections;
5. notating the resultant construction on single staves, mainly in treble clef, without any key signatures, without metre, without bar lines, in groups numbering from 2 to 15 semiquavers in such a way that the subsequent notes belong, alternately, to the first and the second performing part.
The second-level intertextual exponent, created by the algorithm, divides into 77 sections. Almost all of them are composed of two segments: a desynchronising one and a repetitive one. To create desynchronisation of two similar motives, Szymański used five pairs of motives with the respective note numbers 3 and 2, 4 and 3, 5 and 4, 6 and 5, 7 and 6, so that the desynchronising segments number successively 12, 24, 40, 60 and 84 notes. Irrespective of their length, all desynchronising segments have a predictably similar pattern. They begin with the alternate presentation of sounds from two similar motives, resulting in repetitions. The actual desynchronisation or intertwining of motives starts when the motives appear for the second time. Initially the heads of the motives gradually distance themselves from each other, and then gradually approach each other again. The length of desynchronisation depends on the length of the motives: the longer the motives, the longer the desynchronisation. The sequence of repetition of a few notes appears again at the end of the desynchronising segment. An added note, together with the last note of the segment, create an additional pair of notes, a natural transition to the next segment. The second segment – i.e. the repetitive one – is composed (together with an added note) of 3 to 29 notes. (Irrespective of the length of both segments, each section always contains an odd number of notes.) From the listener's point of view it is important that, after each fragment with actual desynchronisation, which appears to be an ungrammatical construction, there is a repetitive fragment (consisting of a repetitive segment and the ends of the neighbouring desynchronising segments), which is perceived as tonal and typically Baroque. This flow of syntactically varied music creates an impression of two musical times – one which seems to stop for a moment, and the second, which flies like an arrow in compliance with our deep-rooted expectations.
As has already been mentioned, Szymański uses the second group of interpretant factors to enrich the second-level intertextual exponent by: (1) introducing the poco marcato articulation; (2) placing the notes in the uppermost piano registers; (3) introducing the p-ppp dynamics and (4) the prestissimo tempo; (5) applying the pedal; and (6) the lascia vibrare effect; as well as (7) also possibly using mean-tone tuning, although equal temperament may be used.
The application of the poco marcato articulation produces untypical effects. This articulation, denoted in the score with the use of empty rhombi, is applied in desynchronising segments of the following 11 sections: VIII, XIX, XXIX, XXXVIII, XLVI, LIII, LIX, LXIV, LXVIII, LXXI and LXXIII, the distance between each of which constantly decreases and is, respectively, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2 sections long. The poco marcato articulation always appears at the beginning of the desynchronising segment and, most importantly, always in pairs on subsequent notes of the motive which is currently being elaborated. The number of notes in the section articulated in this way depends on the size of the elaborated motives, ranging from 2 to 10, while the length of the distances between the poco marcato note pairs is from 0 to 12 semiquavers. The articulation highlights the Baroque motives, creating grammatical figure against an ungrammatical background.
Intertextual Play Between Baroque and Modern Meanings
The Second Study is an unbroken intertextual play in which Baroque meanings combine with modern ones, characteristic of Szymański's idiom. The main feature of this play is the transition from repetition to desynchronisation – that is, from a tonal, typically Baroque structure to an ungrammatical, modern one. There is also a dialogue between short Baroque motives, highlighted by poco marcato, against the ungrammatical background. Another dialogue is created by the remaining parameters: on the one hand, we hear a typically Baroque rhythm and mean-tone tuning; on the other hand, an untypical piano timbre, continuously in the uppermost registers.
Dialogue is also present in the formal course of the composition. The Second Study is divided into five episodes, with the Baroque and modern elements in different proportions. The first episode (sections I–VII) is a kind of chaos made up of short segments of both kinds. In the second episode (VIII–XLVIII), characterised by numerous long desynchronising segments and relatively short repetitive ones, a listener can begin to recognise Baroque elements. The third episode (XLIX–LII) brings a brief balance between the two kinds of meaning. The fourth episode (LIII–LXXIII) is the opposite of the second – relatively short desynchronising segments are accompanied by increasingly longer repetitive segments, which creates here an impression that Baroque music dominates over the modern one. In the last episode only the longest desynchronising segments appear, and most of it has a decidedly modern character (LXXIV–LXXVII).
Example 2 presents an excerpt from the second episode of the Second Study, comprising sections XX–XXII (see analogical sections in Example 1). A rectangle comprises an actual desynchronisation, the numbers 59, 32, 26 denote the numbers of notes of the desynchronisation and repetitive fragments. The oval rim denotes an additional note. As can be observed, the actual desynchronisation is divided by two shorter repetitions. It is built on the longest motives with 7 and 6 notes. From a tonal perspective it is ungrammatical, as the VII degree in E-minor scale resolves onto degrees V, II and III more often than degree I, thus frustrating our expectations. Both repetition fragments run as in Baroque music. The large interval leap in the second repetition indicates that a new musical thought has begun in the first-level intertextual exponent.
Example 2: The second of the Two Studies for Piano, page 13, staves 22–24. © Copyright 1987 Chester Music Limited. All Rights Reserved. International Copyright Secured. Used by Permission of Chester Music Limited.
Compartment 2, Car 7 (2003)
Compartment 2, Car 7 for vibraphone, violin, viola and cello was commissioned by American choreographer Alonzo King and finished in 2003. The world premiere of the composition was in San Francisco in the same year, and it was first performed as an autonomous piece during the 2004 Warsaw Autumn festival. The score is available from the composer, and there is a CD recording by Krzysztof Jaguszewski and the members of the Silesian Quartet.Footnote 19 The work consists of two clearly diverse movements (attaca), of which the first was composed freely and the second algorithmically. To demonstrate that the proposed research method may also be applied to freely composed pieces, I will discuss the first movement.
Sarabande in J.S. Bach's Style as an Intertextual Exponent of the First Level
In both movements there are modern compositional devices and traces of Baroque music. On closer acquaintance, we might hear associations with instrumental music by J.S. Bach, including the Partitas for solo violin and Suites for solo cello. Familiarity with another composition by Szymański, Une suite de pièces de clavecin par Mr Szymański (2001), available from the composer and as a recording by Maciej Grzybowski,Footnote 20 brings us closer to revealing the secret. Szymański's Suite is composed of seven movements: 1. Ouverture, 2. Allemande, 3. Courante, 4. Sarabande, 5. Air, 6. Menuet I & Menuet II, 7. Gigue. Szymański himself says that these dances were composed using the fundamental principles of Bach's style, as well as small elements of his own style, and that they were created to become pre-compositions in new, original works.Footnote 21 The individual movements of the Suite are ‘clean’ structures, with no dynamic, articulation or even agogic markings. Comparing the first movement of Compartment 2, Car 7 with the individual dances of the Suite, it appears to be connected with the Sarabande.
Analysis of the first movement of Compartment 2, Car 7 confirms that the Sarabande from Une suite de pièces de clavecin par Mr Szymański acts here as an intertextual exponent of the first level. This particular Sarabande has 21 bars in 3/4 metre and consists of two movements, with the end of the second movement referring to the beginning of the composition. This stylised dance was written by Szymański in A minor, in polyphonic, five-part texture. The composition contains arpeggios and mordents on the second crotchet note of the bar, emphasising its dance character, as well as the bass, typical for a chaconne, descending chromatically from A to E. The large number of diminished chords indicates that this is the work of a contemporary composer. Example 3 shows the beginning of the Sarabande.
Example 3: Paweł Szymański, Sarabande from Une suite de pièces de clavecin par Mr Szymański, page 14, bars 1–4. Copyright 2001 Paweł Szymański.
Three Interpretants and Three Sections of the Dialogue Between Baroque and Modern Musical Meanings
The first movement of Compartment 2, Car 7 is in three sections, A-B-A1, and has a completely different character from the Sarabande. The interpretant which transforms the Sarabande consists of factors affecting both the entire composition and its individual sections. The material is transposed from A minor to G minor, the metre changed from 3/4 to 4/4 and rhythmic values are lengthened (approximately 7 times).
The opening section is based on the first nine bars of the Sarabande. The locally acting interpretant includes the following factors:
1. dividing the five parts between violin, viola and cello;
2. transformation of the texture from polyphonic to homophonic, employing 2 types of chords: the sounds performed simultaneously or with delay (‘supplemented’ chords);
3. replacing ornaments with anacruses;
4. introduction of fff dynamics with sforzato articulations and, here and there, ppp dynamics;
5. placing accents on all the single notes.
Section A has 51 bars, and from beginning to the end is a dialogue between two different kinds of musical meaning. Baroque elements such as concords and a major-minor harmonic outline are contrasted with modern elements such as complex and ‘supplemented’ chords, pre-chord anacrusis, strong dynamic contrasts and glissandi. The dialogue is developed both linearly and vertically, with the modern elements dominant. Because the play of autonomous meanings is in the foreground, this semantic play becomes the main subject of the first section. Example 4 shows a fragment of section A, the transformation of bars 2–3 of the Sarabande. Components of chords falling to the beginning of the subsequent crotchets in the Sarabande are joined by vertical lines, while the anacrusis on the G is marked with an oval rim.
Example 4: Paweł Szymański, Compartment 2, Car 7 for vibraphone, violin, viola and cello, page 3, bars 7–12. Copyright 2003 Paweł Szymański.
The pre-composition of the second section consists of bars 10–16 of the Sarabande, in which the music is broken up with pauses, and one of the voices always has melodic figurations. The interpretant transforming this excerpt of the Sarabande consists of:
1. free distribution of the Sarabande’s voices among the instrumental parts, including the vibraphone;
2. occasional use of figuration from the Sarabande, sometimes with imitation between parts;
3. using single notes and concords from the Sarabande in bigger rhythmic values;
4. glissandi;
5. various kinds of dynamics, from ppp to ff;
6. occasional accents and harmonics.
The final text, which in this case is the second section of the composition, is contained within 42 bars. Again the play of old and new meanings lasts from the beginning to the end. Baroque elements include melodic figurations and brief three-part imitations in third and sixth intervals; modern elements include the glissando endings of motives in the violin parts and the timbre of the vibraphone. This time, the old elements are dominant. As the second section is, in general, melodious, the semantic play is mostly linear, and the units of the dialogue gradually become longer.
As I said earlier, the third section – A1 − is similar to the first one but, because of its concluding character, offers a few new solutions. These include three figurations in the cello part, each subsequent longer than the previous one, the last one almost a virtuoso mini-cadenza. With regard to the dialogue between Baroque and modern means the A1 section may be regarded as the one which reconciles these different means. The intriguing title, a paraphrase of the title of a painting by Edward Hopper, relates mainly to the second, algorithmic movement of the composition.
Conclusions
Although Paweł Szymański's compositions do not use quotations from familiar works, they are clearly and consistently intertextual. The intertexts are consciously selected by the composer and the first-level intertextual indicators created by him are modelled on quite banal conventions. From this material the composer creates works characterised by a ceaseless intertextual play between the old and the modern. For both listeners and researchers these compositions constitute musical riddles that can nevertheless be solved, albeit at different levels of detailing. Although deeply rooted in tradition these compositions are, at the same time, modern and original, with the interpretants constituting the major source of their originality. Does this intertextual play have a wider significance? For Paweł SzymańskiFootnote 22 the use of stylistic dialogue is a metaphor for our human, imperfect learning about the world, our inability to reach the core of the matter, our fragmentary perception of reality.
