Connections: The Sound of Speech and Classical Music
Background
Music is a fascinating art form and one that I have enjoyed for most of my life. I was raised listening to Classical Music and although at the time I didn’t understand all of the complexities and nuances of the form such as the theory behind it or all of the history associated with it, I was always certain that I loved listening to the pure sounds that were coming out of my parent’s record player. Anyone who observed me listening to Classical Music could easily see why I loved it so much. Whenever I listened, the music would evoke an emotional response in me.
This fact never changed even as I got older, but once I began studying music and learning the piano as an adult, one question always resounded in my mind: Why is it that music, particularly good, can evoke an emotional response from the listener? I thought about that for a time but then a possible theory emerged from my ponderings. What if there was a relationship between the sound of speech and music, in particular, Classical Music? This theory came to me as a result of what seemed to be my intuition. Sometimes when I listened to music, I felt as if I was hearing someone speak. I couldn’t derive individual words or even a possible meaning to an annunciation, but I felt that there was definitely something audibly, almost verbally, being expressed.
Examining this question and theory was on my mind when I came to ITP. Now, I finally felt I was in the perfect environment to explore the theory of the connection between the sound of speech and music. In Audio Art I learned about the many different ways sounds can be produced and manipulated and the many different programs that can be used to do this. Using my newfound knowledge, I began to put together the idea for project that could possibly test my theory.
Project
Step One – Recording Speech
It occurred to me that I would first need a recording of speech to use as a starting point. After thinking this over, I decided to enlist the help of my wife who at the time was taking an acting class at Hunter College. I asked her to choose any literary work and then read a passage from it that I would record. To my surprise, she chose my father’s autobiography, Noble Street, and out of that a passage detailing a humorous event from his childhood. I felt my wife’s choice was perfect because the work was something we were both very familiar with.
Using a tape recorder, I recorded my wife reading the following paragraphs from Noble Street.
We were still hiding under the truck, waiting for the opportunity to release Ducky from the freezer. But then we heard footsteps on the platform, and they were coming towards our location. We had to move further back under the truck. As the movement on the platform grew closer to us, we all grew silent.
But I could not keep myself from whispering, “All this for a hot dog. I ain’t never gonna eat another hot dog. Never.”
Then suddenly, there was the sound of someone jumping from the platform and landing next to the truck where we were hiding. We were about to be discovered, or so Norman, Frank and I thought, but Ducky came crawling under the truck wrapped in strings of hot dogs from his neck to his waist.
“Dammit,” Frank grabbed some of the hot dogs. “Why’d you scare of us like that? We thought you might be frozen stiff by now.”
“Naw.” Ducky laughed. “I wasn’t even all that cold. When that man rolled me in the freezer room, I just wrapped myself up with the hot dogs to keep warm, and after a while, I opened the door and came out.”
“You weren’t even cold, huh?” Norman slapped him gently upside the head. “And we out her were so worried about you.” Norman was relieved that Ducky was safe.
"Aw, lets just knock it off and get the hell outta here,” Frank said.
The passage describes an incident in the early 1940s when my father and three of his brothers were stealing food from warehouses near the docks in Philadelphia where my father was born. The food was needed as my father had eleven brothers and sisters and his own father was not around anymore. What is wonderful about the passage is that despite the desperate circumstances - stealing food to feed a starving African-American family - the story is funny and my wife reads it as such although she kept a straight face.
Step Two – Analysis of Speech
Now I had my starting material so the next step was to do something with it. I had already planned to dissect my wife’s reading of the passage to see if I could pull any sound fragments from it that might be transformed into a short musical fragment. From just listening to the recording, I felt that was definitely possible. One of the first clues I discovered in my exploration of a speech-music connection was the fact that human beings rarely speak in a monotonous way. It is easy to detect different pitches spoken among different syllables and words. The human voice rises and falls in many different ways when a person is speaking. The question is, though, what are those pitches?
To find out, I ran recording through the program, Audacity. The problem was able to analyze the different pitches in my wife’s voice, which gave me a starting point as I began to look at building a musical piece from the spoken passage. The first thing I did was to isolate the first syllable that my wife speaks, in this case for the word, “we.” Using Audacity I was able to determine that the pitch used for “we” was very close to the note A. With that fact now established I decided that I would now compose two melodies in the key of A that were based on two different sentences that were spoken from the passage.
Step Three – Composition of Melodies
I chose the sentences, “All this for a hot dog” and “You weren’t even cold, huh?” I chose these because I could definitely hear a rise and fall in my wife’s voice as she spoke them. As a result, several different pitches were clearly audible. At this point, I could have tried to isolate each pitch, but due to time constraints, I thought it was wiser to simply use my ear and guess. For me, what were equally important were not just pitches, but also the rhythms and the intervals that were used. These I could roughly figure out by ear.
The resulting melodies were each six notes long and each took up one measure in 4/4 time. I played them for my wife and she was able to determine which sentences I used to construct them. That confirmed my belief that I had done a good job with my estimations of pitch and rhythm. The next decision I had to make was what to do with these melodies.
Step Four – Choice of Instrument and Form
One danger that concerned me was that it would have been exceedingly easy to simply use the sentences as lyrics and add them to the melody. However, this was not my goal. Since my original theory concerned Classical Music, and more specifically, instrumental Classical Music, I determined my next step was to use the speech-born melodies as material for a piece composed for an instrument.
Automatically, I had many options that I could pursue. My musical listening background had begun with symphonies, most by Ludwig von Beethoven. Unfortunately, again due to time constraints, composing a symphony was out of the question. Consequently, I chose the piano as my instrument since I had learned to play the piano and the vast majority of my earlier compositions were for this instrument. But choosing the instrument was only the first step. I then had to choose a form for my composition and for that I chose Sonata Form.
Sonata Form was the standard form used in the overwhelming majority of pieces composed from the late Eighteenth century through to the beginning of the Twentieth century. The form was primarily used for the first movements of multi-movement works such as symphonies, piano sonatas and pieces for chamber music. Interestingly enough, the name “Sonata Form” was not invented until the late Nineteenth Century by musicologists. Composers such as Haydn, Mozart and Beethoven never used the name to describe the form even though the first two composers, Haydn and Mozart, were the ones who codified it.
The reason why I chose Sonata Form for my composition is that it is very easy to build an entire movement out of a small piece of musical material such as a melody or even a melodic fragment. The perfect example of this can be found in the first movement of Beethoven’s Fifth Symphony. The first four notes played by the strings and the clarinets, three Gs and an Eb, are the material that almost the entire movement and most of the symphony are built from.
Source: http://library.thinkquest.org/22673/forms1.html
Sonata Form is actually based around two melodies or themes and the form is divided into three main sections followed by a Coda section for the ending. In the first section, the Exposition, both melodies are played. However, the two melodies are composed in different but related keys. The relationship depends on whether the overall key of the movement is major or minor. If major, the first melody is in the tonic key, or the home key, and the second melody is in the dominant key, five notes away. If the movement is in a minor key, the first melody is again in the tonic but the second melody is in the relative a major key of the tonic, which simply means that both keys share the same key signature (numbers of flats or sharps).
The second section of Sonata Form is the Development, where the material from the Exposition is used and the material is often placed in new keys that have no relation to the first two. The third section is called the Recapitulation and here the melodies from the Exposition are repeated with one exception. The two melodies are now placed in the same key. Composers often vary the material in the Recapitulation just to make it sound more interesting. The movement is then brought to a close with the Coda, which can include old material or new material at the composer’s discretion.
Sonata Form was the logical choice since I already had two short melodies composed. I could have used each one as the first and second themes of the Exposition but instead I decided to combine them so that they both together represented the first theme. I did this for two reasons. The first was that the melodies were short, each one only a measure. The second was that often in Sonata Form the two melodies are meant to contrast one another. I could have listened through the passage again to find more material, but I decided that it would be wiser to simply compose the second theme from scratch, as I wanted to make sure of the theme’s contrasting nature.
My final act before beginning to compose was to decide that the resulting piece was going to become the first movement of my Fourth Piano Sonata in the key of A. The choice of the key was once again based on the pitch of the first syllable my wife spoke when she read the passage.
Step Five - Sonata No. 4 in A Major, First Movement
I was quite satisfied with my work once I completed it. The speech-born melodies began the piece right off, which I felt was a good touch since they are the material from which the whole movement was developed. I doubled the length of each melody by repeating it twice and then adding a closing phrase, also repeated twice, so that the entire first melody is six measures long. Following the form, the second melody is longer and more song-like. The Development section does take the composition through many unrelated keys and the Recapitulation does include variations on the melodies.
What I found interesting about the piece once I listened to it in its entirety was the fact that it sounds very punchy and happy. This raised an entirely new question. Was the composition punchy and happy because the two melodies it was created from sound that way? Or was it because in my mind I knew that the passage from which the melodies trace their origins to is a humorous one? In short, does the happiness (if one indeed determines the piece to be happy) come from phonological or psychological elements? As of right now, I cannot say.
Conclusion
Ultimately, I believe that this project is just the first step in my exploration of the speech-music connection. I definitely plan to do more research and perhaps compose more pieces as I dig deeper into this issue. I believe that this topic would be perfect for my thesis next year, although I have not yet determined what work I will actually conduct by that point. There are so many different directions I can take and I believe that my next step is to determine what avenue I want to pursue and what my goal should be.
What I find most interesting now is that I am not the only person pursuing answers to these questions about speech and music. After I began working on this project, I learned from Luke Dubois that one his former professors at Columbia, Fred Lerdahl, had co-written a book entitled A Generative Theory of Tonal Music. This book, first published in 1983, has spawned a whole community of musicians, musicologists, linguists and psychologists who are also working on finding answers to the music-speech connection. Not surprisingly, the origin of this new field of study, and the origin of my questions can each trace their paths back to one man and his work. In 1973, Leonard Bernstein presented the Charles Eliot Norton Lectures at Harvard University. His topic was the search for a link between transformational-generative linguistics and musical structure, which he termed “musical grammar.” I watched these lectures on tape when I began to study music and the authors of A Generative Theory of Tonal Music took a seminar that was inspired by Bernstein’s lecture. In a way, I feel my questions and theories have brought me full circle.