Struga II

In this paper I would like to look at the relationship of music to words and, most particularly, to natural speech. n particular, I would like to share with you a study of the work of the American composer Steve Reich (born 1936), who - while acknowledging earlier composers' work in this area - has created what I believe to be pioneering work, always making use of technology as a tool to push musical boundaries rather than simply the gimmick that it can so easily become.

"In the beginning was the word" - but words in a musical context became stretched and squeezed as the primacy of melody overtook the natural cadences of speech, and with them the idiosyncracies of language were at best ignored, at worst, lost.

Reich's background was as a percussionist, and I share this training. Perhaps therefore it's not surprising that I start this study by looking at the rhythm of speech, and in looking for examples of where this has still been allowed to dictate the musical shape, I was reminded of my own musical background, even earlier than as a percussionist, as a choirboy at Jesus College, Cambridge, singing Anglican chants. To quote New Grove, these are:

"Harmonised formulæ used for the singing of psalms and canticles in the liturgy of the Church of England. A single chant comprises two sections, paralleling the bipartite psalm or canticle verse to which it is sung; the initial chord in each half is the 'reciting' chord to which a substantial part of the verse section is freely sung. The first half of the chant is concluded by a progression of between three and five chords, the second half by a progression of between five and nine chords. These are invariably measured out in semibreve, minim and crotchet values, the first comprising three bars, the second, four. There are many ways of 'pointing' or fitting the words to these chants, and various systems of symbols are used to indicate how this may be done."

In the following example you can see how it works in practice: the most important issue should always be to give the words their natural pattern, and every syllable should be heard:

Psalm 75
Unto thee O God do ' we give ' thanks : yea unto ' thee do ' we give ' thanks.
2 Thy Name also ' is so ' nigh : and that do thy ' wondrous ' works de'clare.
3 When I receive the ' congre'gation : I shall judge ac'cording ' unto ' right.
4 The earth is weak, and all the in'habiters there'of : I bear ' up the ' pillars ' of it.
5 I said unto the fools, Deal ' not so ' madly : and to the ungodly, ' Set not ' up your ' horn.
6 Set not up your ' horn on ' high : and ' speak not ' with a stiff ' neck.
7 For promotion cometh neither ' from the ' east nor ' from the ' west : nor ' yet from the ' south.
8 And why? ' God is the ' Judge : he putteth down one and ' setteth ' up an'other.
9 For in the hand of the Lord there is a cup and the ' wine is ' red : it is full mixed, and he ' poureth ' out of the ' same.
10 As for the ' dregs there'of : all the ungodly of the earth shall ' drink them and ' suck them ' out.
11 But I will talk of the ' God of ' Jacob : and ' praise ' him for ' ever.
12 All the horns of the ungodly also ' will I ' break : and the horns of the ' righteous shall ' be ex'alted.

On occasion the issue of the breaking down of a word into its syllabic constituents may be considered extreme, but it is no more than poetic licence. Here is a collector's item of a recording, made in 1966 by The Master Singers, a group that was to become the King's Singers. It deliberately sets out to show that the Anglican chant can be applied - to musical effect - to any piece of text, in this case the rules of the road as laid out by the British Government. It is, of course, deliberately extreme for comic effect!

A further situation where the text drives the music rather than the other way round is in the unique ancient Welsh art of cynghanedd, concisely described by New Grove as "a complex system of assonance, alliteration and internal rhyme" setting poetry in the Welsh language to well-known folk melodies or hymn tunes.

The antithesis of this naturalistic approach is found in such works as Schoenberg's Pierrot lunaire (1912), which while a remarkable and ground-breaking piece, introduces the form of sprechgesang to distort both text and melody. Of more interest as a precursor of Reich is Stravinsky, and in particular Les noces (1914-17), in which Stravinsky claimed in his 1935 memoirs:-

"I borrowed nothing from folk pieces... all the themes, motifs, and melodies [except one in the fourth tableau] are of my own composition... The recreation of a country wedding ritual (which, in any case, I never had the chance either to see or listen to) never entered my mind. Ethnographic questions occupied me very little."

As we will see, this claim is nothing if not economical with the truth, but Stravinsky's motives for such a claim have been convincingly argued as more to do with distancing himself from the Soviet socialist realism prevailing by the 1930s, in which folklore [and its implicit naturalism] was idealised as a model for 'new' music, than any attempt to claim personal credit for a work which is firmly rooted in the soil of Mother Russia. In a remarkable recording by the Pokrovsky Ensemble, it places a performance of the Stravinsky within a series of Russian village wedding songs collected by the group from south and west Russia. By this juxtaposition we can clearly hear the connections: here is a folk song, Cosmas and Demian, followed by Stravinsky, an extract from the Third Tableau:

Before we look at Reich's own work in this area, it is important to look at the work he himself acknowledges by earlier composers. In interview, he says:
Although I personally became aware of speech melody in my early tape pieces It's Gonna Rain (1965) and Come Out (1966), it is something I believe musicians have been aware of for centuries.

Not so long ago, the composer Leos Janácek wrote down the speech melodies of his fellow Czech citizens in his notebook and used them in his operas. Bartók noted that while collecting folk songs in (what used to be) Yugoslavia, the music would change when the language changed. It is no accident that rock and roll is strong in English and German and less so in French and Italian. It is also no accident that the bel canto voice arose in Italy. Some African languages are tonal - the speech melody is part of the meaning... In such tonal languages, the speech melody is not only a kind of emotional aura surrounding what is said but is also part of the literal meaning of the words. If the speech melody is different, the words have a different meaning... In our Western languages, speech melody hovers over all our conversations, giving them their fine emotional meaning - "It's not what she said, it's how she said it". We are, with speech melody, in an area of human behaviour where music, meaning, and feelings are completely fused...

Clearly, what language you speak will largely determine the rhythmic aspect of the melody of your speech. The question you ask is how does this effect the entire nation's music? I would answer that all vocal music using a particular language will in general be strongly affected by the rhythm and cadence of that language...

In his essay 'Hungarian Folk Music' (1933), Bartók gave a linguistic reason for a musical practice in Hungarian and English music:
"In the Hungarian language every word has the accent on the first syllable. The following syllabic sequence is impossible in German: short & accented + long & unaccented. (The English language is indeed more flexible, owing to the existence of the so-called Scottish rhythm, which corresponds exactly to the Hungarian dotted one.)"...

As already stated, Janáchek was the first to speak of speech melody and to use it in his operas (see ex. below). In 1918 he wrote:-

"I envision a Czech name in its Czech version next to its German one. That was how the guard called out the name of the railway station on 18 August 1917.

How the different spirit of both languages shone through here. Our version is ranged in notes of a warm triad, Db-F-Ab. The German version cut harshly and roughly in the same triad, with a dissonance of a seventh; it has crushed the third syllable and torn off the last one; it has ground into grumbling the sweetness of the first two. In the Czech version, you hear a song that winds along in equal lengths within a rainbow of colours; o-a-a-y ...

The melodic sweetness of the Czech word has disappeared in the German version, the musical union of speech melody has thinned down... It is what has not been translated from the Czech that has triumphed: speech melody, the seat of the emotional furnace..."

On the other hand, just in terms of the music, you may find the many speech melodies an unusual musical guide to personality. As Janá_ek said, "speech melodies are windows into people's souls... for dramatic music they are of great importance" - important because it's impossible to separate the music from the person speaking.

For me, Janá_ek shows this most vividly in his remarkable work The Diary of One who Disappeared (1917-21), which is directly influenced by the speech melody of his native Valachian dialect. Here are several extracts to illustrate:

So, how has Reich developed this awareness, described by J Neal on the All Classical Guide website as "[Reich's] fascination with conflating a word's acoustical parameters with its meaning and expressive qualities is a trait found in virtually all of his works involving the human voice." In a further interview, Reich explains:

My interest in using spoken language as a basis for music began as the indirect result of reading the poetry of William Carlos Williams in the 1950s. I tried to set his poetry to music and found I only 'froze' its flexible American speech-derived rhythms. Later, in the early 1960s, it occurred to me that using actual tape recordings of Americans speaking might serve as a basis for a musical piece that would utilise the same sources as Williams' poetry. This led to my two early speech tape pieces, It's Gonna Rain (1965) and Come Out (1966). Both of these tape pieces presented the voices of black Americans whose speaking voices were extremely melodic. Through the use of repetition and phase shifting, the speech melody of their voices became intensified.

Years later, in the 1980s, I became interested in constantly-changing meters to capture what I heard in setting the classical Hebrew of the Psalms in Tehillim (1981). This technique of constant meter change led to The Desert Music (1984) where I was finally able to successfully set Dr. Williams' poetry. Eventually all this came full circle in 1988 when I used recorded speaking voices together with a string quartet in Different Trains (1988). In this piece, the recorded speech melody formed the basis for melodic material to be played by the strings...

Since the early 1960s I have been interested in speech melody. That is, the melody that all of us unconsciously create while speaking. Sometimes this speech melody is quite pronounced (as in children) and sometimes it is almost non-existent (as in those speaking in a monotone)...

Using the voice of individual speakers is not like setting a text - it's setting a human being. A human being is personified by his or her voice. If you record me, my cadences, the way I speak are just as much me as any photograph of me...

The speech melody of each person really is a kind of musical portrait of that person. It's their melody and I begin by writing it down as dictation. I have to find out the exact notes, rhythm, and tempo of what they say...

Late in 1964, I recorded a black preacher in San Francisco, preaching about the Flood. I was extremely impressed with the melodic quality of his speech, which seemed to be on the verge of singing. Early in 1965, I began making tape loops of his voice, which made the musical quality of his speech emerge even more strongly. This is not to say that the meaning of his words on the loop, "it's gonna rain," were forgotten or obliterated. The incessant repetition intensified their meaning and their melody at one and the same time.

By using recorded speech as a source of electronic or tape music, speech-melody and meaning are presented as they naturally occur. It is quite different from setting words to music where one has to fit a number of syllables to a number of notes, and decide what their melodic relation will be. In speech, questions of how many notes to a syllable, or what their melody will be, do not arise; the speech just comes out. Instead of setting words to music, I simply chose the exact segments of recorded speech I was intuitively drawn to as musical material. My original interest in electronic music was the possibility of working with recorded speech...

Using actual recordings of speech for tape pieces was my solution, at that time, to the problem of how to make vocal music...

When people speak, the semantic, structural, and melodic issues from them in one breath. The phrase "It's gonna rain" as said by the black Pentecostal preacher is at once a common everyday American idiom, a reference to the biblical flood - and, therefore, the end of the world (which in the 1960s clearly had nuclear overtones) - and the pitches E-D-D-F#. How do you propose to separate these elements? The beauty (and terror) is that they are all one...

In his analysis of Reich's works, the English musicologist Keith Potter puts it thus:-

"Speech, after all, has a melodic profile of its own, which may be exploited in the act of composition. It's Gonna Rain begins with a pattern, based on a rising major third, which becomes important to the way the work unfolds.

Pitch most clearly becomes an issue in Part One, which uses only the single fragment of three pitches that gives It's Gonna Rain its title (reproduced above), suggesting the key of D major."

He goes on to analyse Come Out similarly:

"Reich achieves a more purely musical focus than he had managed in It's Gonna Rain with the aid of a more melodic phrase demonstrating greater pitch stability: its character determined by the interval of a minor third, suggesting C minor, and enhanced by clearer vowel sounds. (Ex. below gives the Basic Unit of Come Out.)"

So, back to Reich's own words:

Starting with Tehillim and then The Desert Music I found myself writing constantly shifting meters to accommodate the shifting stress of syllables.... There is no fixed meter or metric pattern in Tehillim as there is in my earlier music. The rhythm of the music here comes directly from the rhythm of the Hebrew text and is consequently in flexible changing meters....

Why is there no repetition of short patterns in Tehillim? The basic reason for avoiding repetition in Tehillim was the need to set the text in accordance with its rhythm and meaning. The Psalm texts set here not only determine the rhythm of the music (which is basically combinations of two or three beats throughout the piece combined so as to form constantly changing meters) but also demand appropriate setting of the meaning of the words...

While studying biblical Hebrew, I asked my teacher about the biblical accents or ta'amim and was informed that these were, among other things, the musical notation for the chanting of the Torah and Prophets and several other books of the Hebrew scriptures...

The ta'amim, or accents, serve three functions in the biblical Hebrew text. First, they show the accented syllable. For example, the first word in the Torah is Bereshíth and not Béreshith. Second, the ta'amim are the punctuation marks for the Hebrew text. All in all, there are 28 different ta'amim. Nineteen of these serve as disjunctive punctuation to show separations of varying degrees between parts of a sentence and eight are connectives. (To appreciate the degree of nuance such a large number of punctuation marks can create, bear in mind that there are only 12 punctuation marks on an English-language typewriter.) As with all punctuation, the biblical accents are used to clarify the meaning of the text. Third, the ta'amim serve as the musical notation for the chanting of the Hebrew biblical text...

So, on to Different Trains, and here - as stated before - Reich explains that the basic idea is that speech recordings generate the musical material for musical instruments...

In order to combine the taped speech with the string instruments, I selected small speech samples that are more or less clearly pitched, then transcribed them as accurately as possible into musical notation. For instance, see example below, from the first movement:

The strings then literally imitate that speech melody...

In the third movement, it is the tentative quality of voice and feeling that "Are you sure?" gives, along with its purely musical content (B, B, F#) that made me choose it...

The speech samples as well as the train sounds were transferred to tape with the use of sampling keyboards and a computer...

This idea is developed further in City Life (1995). [In this,] all the sounds I recorded in New York City were then fed into two sampling keyboards that are played live on stage along with the rest of the instrumental ensemble. When a street merchant says "Check it out", that pre-recorded voice is played back at precisely the right musical moment by the sampler players and at that same time the vibraphones and pianos and woodwinds all double the speaker's speech-melody with no use of tape recorders during the performance.

This allows all the musicians to make the small changes in tempo that naturally occur in performance but that are eliminated when following a pre-recorded tape. This kind of technology will undoubtedly find its way into other pieces of mine and many other composers in the future. [end of quotes]

I referred at the beginning to the use of technology as a means to an end and how it can easily be abused. Here is an early use of sampling keyboards, produced for the Christmas market in the UK in 1982. Of course, it's just a fun recording, but listen carefully:-

Even for something that seems as crude a sound as this, it can be detected when the sound has been treated electronically to raise or lower the pitch from its natural state, and when it is a naturally occurring variation.

Reich was quite right to predict his own increasing use of samplers: they have a prominent role in both The Cave (1990-93) and Three Tales (2001), written in collaboration with Beryl Korot. But perhaps more interestingly for this demonstration today, this recording technology has been picked up by a new generation, and Reich's own works have influenced the new breed of musicians, DJs, from whom yet another patina emerges.

It is my belief that a whole new musical language is being created in the latest work of Reich using such tools, and urge you to take any opportunity to follow his work in the years to come.

Reference sources:
New Grove Dictionary of Music and Musicians (second edition, Macmillan 2001)
Steve Reich, editor Paul Hillier: Writings on Music 1965-2000 (OUP 2002)
Keith Potter: Four Musical Minimalists (CUP 2000)