CHAPTER 1 - Additional Notes

The Role of Syntax in Music

What is surprising about tonal music is that a scale made up of such a few notes should be capable of such a large number, possibly infinite variety, of melodies, harmonic structures, styles and varieties of music.

This reminds us of the similar problem in language where the syntax enables us to generate an infinite number of sentences from a finite number of words (see previous section on linguistics). The reason such a large variety of melodic, rhythmic and harmonic structures can be generated is because we hear music not just as a series of individual notes but because we are capable of subconsciously assembling these notes into logical structures. A common understanding of such patterns between composers and listeners is dependent upon there being a commonly understood system of syntax (for a particular style of music). Just as there is for language. This book (and website) attempts to explain the nature of this syntax for the 18th and 19th century western art music and some of the music of the 20th century (in particular popular music). For simplicity, I will describe this as tonal music. The evidence for the existence of this syntax is demonstrated in a full thesis currently in preparation.

Some people would deny the existence of rules in music. They would claim that rules merely restrict and that music is an art form and therefore is completely free in its construction.

Against this argument are the following considerations:

1. As we can see from language (see previous section on linguistics).

a) Rules don't restrict possibilities - they increase them.

b) Rules don't have to be explicitly understood.

A native speaker of a language subconsciously follows the rules of the language with accuracy without necessarily being able to say what those rules are or without even having the means to describe those rules. The absence of an explicit grammar for music does not necessarily indicate the absence of an implicitly understood grammar.

2. We know that in tonal music chords have to be constricted in certain ways. Discords are prepared and resolved according to given rules. Voices have to move in certain ways to avoid bad musical effects. The basic teaching of musical theory at all musical institutions acknowledges the way certain rules govern whether combinations of notes sound tonal or don't sound tonal. Why then, when discussing chord progressions should we deny that there may be rules governing what progressions produce correct tonally sounding music? Anyone who has tried to harmonise a melody will have found that some successions of harmonies sound tonal but others do not. Just following rules concerning chord structuring and voice leading is not sufficient to create a harmonisation that sounds right.

3. We know there are areas of music where rules are essential. An example of this is the rules governing the system of tuning known as equal temperament. This is has been, for over two hundred years, the generally accepted method of tuning western instruments. Previous systems of tuning only allowed modulation within a small range of closely related keys but a system of equal tuning allows modulation to all 12 possible keys equally effectively. Rather than restricting the possibilities, these equal temperament rules substantially increased the number of usable keys and the possible extent of modulation. The music of Wagner and Richard Strauss would not have been possible without the imposition of equal temperament. This is a clear example of how rules are utilised in music and how they can extend the possibilities rather than restrict them.

This is not to say that tonal music is just a series of rules. Of course this is not the case. Linguists are quite happy to acknowledge that language has creative components and systematic components. Each component has to be studied in its own terms. Music, like language, has components which are creative and components which are systematic. The creative components are governed by artistic concepts such as: balance, form, beauty and expression. The systematic components are governed by rules which determine the structures of chords, the voice leading and the progression of chords.

The form of Syntax in music

Just as the grammar of a language has several components, the grammar of tonal music is made up of components. These can be described as:

a) The structure of chords:
b) The rules for voice leading (part writing)
c) The syntax of chord progressions

In the discussion of the similarity between the syntactic structures in music and those in language it should not be assumed that language structures are replicated exactly in music. There is no equivalent in music of the noun and the verb etc. Different languages have different syntaxes so it would be strange if music exactly followed the syntax of a particular language. The structures in music parallel those in language by deploying the same devices as each language does in creating its own syntactic structures. For the purpose of this discussion I will refer to this as meta syntax as the connection is at a level beyond the surface syntax. This should be reasonably straight forward to understand as follows:

1. Language and Music grammars both involve a system of classifying their components

Systems of classification are used in many ways because they are a good way of organising complexity. Objects classified together share some properties whilst being independent in other ways.

In language, words can be classified as: noun, verb, adjective etc. Thus all words described as nouns have the shared property that they describe the name of something but individual nouns describe different objects. Phrases can be described as a Noun Phrase; Verb Phrase; Adjectival Phrase etc. Sentences can be understood as simple or complex. This grouping simplifies the rules of syntax because rules can apply to the whole group rather than to the individual components.

Music involves similar classifications: Chords can be structural, auxiliary or passing etc. Syntactic Elements can be static or dynamic and phrases can be complete or incomplete. These classifications will be fully explained later. However, you may, in the mean time, follow the links to the glossary.

2. The complexity of language requires structures which exist at three levels

The syntax of the sentence structure is made simpler by its organisation into three levels as follows:

The sentence, The phrase, The word.

By grouping words into phrases the number of rules governing the syntax is less than would be the case if the relationship was directly between the word and the sentence.

Tonal music also utilises a three level structure to construct its syntax. These levels are:

The Phrase, The Syntactic Element, The Chord.

Some previous attempts at the syntax of music depend on a direct relationship between the phrase and the chord but satisfactory results can not be achieved with this lack of sophistication. A successful theory of musical syntax requires the identification of the intermediate level of structure between the musical phrase and the single chord. This intermediate level is the syntactic element which is a component identifiable by the change of state between static harmony and dynamic harmony.

3. In both language and music there is one basic structure which can be extended by combining complete and incomplete structures in various ways

We have seen from the discussion on language structures that additional phrases can be embedded within sentences and sentences can be combined to produce conjoined sentences. This increases the possible degrees of complexity whilst introducing a minimal number of additional rules.

It will be shown that similar processes exist in music whereby complete and incomplete phrases can be combined in ways that produce larger complete syntactic structures.

It is at this meta-syntax level that the similarity exists between music and language.

It is important to note that the theory discussed in this book is not an attempt to impose a structure on music or to attempt to describe music by the use of a metaphor. Some writers have done this, (see history chapter). The problems of starting off with a preconceived metaphor are clear. Metaphors work well where the patterns are well known. A description in a more familiar area can aid understanding if the metaphor represents a model similar to the concept being described. But to start off with a metaphor and then try and find correlations is dangerous. How do we know that the metaphor chosen is an inappropriate one? What if the metaphor is close, but has aspects that are different to what is being described? The arbitrary application of an inappropriate metaphor can hide aspects of the true structures being described.

The patterns described here are totally the result of the analysis of musical data which uncovers structures that are observable in the music itself. I will include more on this in the full thesis being prepared.

It is also important to note that the above comparisons of syntax in language and music should not be taken as a justification in itself for the ideas presented. It is made merely to help the reader more easily to understand the ideas. Justification is made purely on the basis of the analyses of data from musical scores.

Return to Chapter 1

CONTENTS

Ver. 2.8