Connotations are, according to the Oxford-English Dictionary, “ideas or feelings a word invokes in addition to its literal or primary meaning.” These can be developed due to cultural beliefs or practices, associations with the first time you heard the word, or even their antonyms. Connotations can be helpful in understanding why a word is used within a sentence, and what emotions you are trying to exhibit when you say that word. Marking the verbs in these songs is particularly helpful, in that we can understand the relationship between the text and the music notation better when comparing the emotion and sentiment those verbs are trying to convey and how the music notation emphasizes them.
We have marked up the positivity, negativity, and neutrality of these verbs using the following key ideas:
Positivity: Verbs that bring up feelings of lightness, openness, and joy. Usually
words like this are associated with physical representations of happiness (i.e.
laughing, smiling).
Negativity: Verbs that bring up feelings of darkness, closed-offness, and anger or
sadness. Usually words like this are associated with physical representations of
anger, fear, or sadness (i.e. crying, screaming, cowering)
Neutrality: Verbs that are neither positive nor negative, but rather fall somewhere
in the middle. These verbs are typically broad, such as to be verbs or “know.”
Of course, assigning connotation values to verbs can be a bit tricky and the value might change depending on things like context and tone of voice. We realized this early on in our project and have tried to assign values based on general meaning of the word. Here are a few examples:
Want: We have marked "want" as a positive verb. On the one hand, we want more, we are greedy, and we should
be content with what we have, so want is negative. On the other hand, we want to grow and gain more, American Dream style,
making want positive. However, we looked specifically at what “want” seeks to accomplish, which, according to the Oxford-English Dictionary,
is to overcome the lack of something. This is a positive action, which is why we marked it positive.
Do: We have marked "do" as a neutral verb. the word “do” could possibly be seen as positive. We do things, we get things done,
we have achieved goals! However, it could also be negative. We do something that can cause harm, we think we are doing something to help but
instead it hurts. For this reason, we put “do” in the neutral category because it does not specifically fall on one end or the other.
Die: We have marked "die" as a negative verb. For this word we used a slightly different technique then the first two examples.
We looked at the verb's connotation in relation to another verb "live". “Live” in comparison to “die” is clearly positive, while “die” in
comparison to “live” is clearly negative. To "live" suggests being able to go on, to continue to do more in the world. To "die" suggests an ending,
a completion that might cut your plans short. Therefore we have marked “die” as negative and “live” as positive.
For this project, we utilized the Musical Encoding Initiative (MEI) for the notational markup. This markup allows you to keep track of and label music notation, including pitch name, note duration, note octave, stem direction, rest duration, dynamics, tempo, and much more. The schema for this markup was taken from the University of Virginia, which also served as our tutorial and go-to information hub when our many questions came up. Here is a basic methodology for how we utilized this language, but you can feel free to visit our Cheat Sheet (which is linked below) for a more in-depth explanation.
Getting Started: We start every song with the same header, which includes the namespace and the title of your document. Even if you don't think you'll use all of the tags, keep them! It is best, we've found, to stick with the schema and allow for empty elements. From there you will begin the next chunk of headers, which is specific to the song you are marking up. One MEI document can hold multiple songs, but each must be started with a specific header to determine time signature, key signature, number of staves, and any other characteristics unique to that song. See the Cheat Sheet for an example!
Marking up the Song: When we mark up each song, we read the sheet music left to right, and work top to bottom. That is, start with the uppermost staff and work your way down. In this instance, we had three staves, one for the vocal part and two for the piano part, so we started with the vocal, moved to the right hand of the piano, and then the left hand of the piano. Within these staves are layers, which follow the various lines of music. This suggests that a melody of notes that have stem directions going up is distinct from a melody on the same staff containing different notes and with stem directions going down. This was rather difficult to understand verbally when we first started using MEI, so we strongly recommend you visit the Cheat Sheet for a visual example of this.
Once you understand staves and layers, or, in other words, where to put your notes, you can begin to mark up your song! It is required that each note have at least the @pname (pitch name), @oct (octave), @dur (note duration), and @stem.dir (stem direction). Any other attributes, such as dots that indicate a longer duration, a tie, an accidental, etc. can be added after these four main attributes are present. We typically have these four, followed by @dots, followed by @tie, followed by @accid (accidentals). Keeping this consistency will be useful to you as you try to extract specific attributes using XPath, so we recommend keeping consistency in your own markup.
Caution! It is important to keep in mind that MEI is a completely different world from TEI, and you will be marking up essentially a different language. For that reason, we encourage you to read our Cheat Sheet and visit the University of Virginia's website before embarking on your own MEI XML markup, and always link to your schema!
For more information about MEI, click here!
What Are Dynamics? Dynamics in music are the markers of volume, or how loudly or softly a certain section oughtto be played. The two main dynamics, all of which are written in Italian, are piano and forte, which mean soft and loud. There are variations of these main dynamics throughout any piece, and we were curious about how the volume of verbs affect the emphasis on those verbs. By creating a sine wave graph, orsomething like one, we have a visual representation of how the volume changes over time throughout the song. We chose a sine wave graph rather than a line graph because the volume in these pieces are notrigid and staccato, but rather smoothe and gradual. The volume shifts from one standard to another smoothly, like a wave, and thus these waves help to better create a visual representation. By then placing the verbs on these waves, we can see where certain verbs are the loudest and where they are the quietest visually, and can compare these facts to their duration and connotation in order to better understand how they are emphasized.
Making the Graph: We begin this graph using SVG, creating the curves on which the verbs will sit. For this graph, we have three path elements created in order to create three semi-sine waves. We begin with the "M" as the initial point for the graph, and follow that using a series of "S" or "C" letters to create the curves. You must begin with an "S," but after that you can follow with either letter. The comma placement was suggested by the W3 schools website, and I have found that removing those commas returns an error in the SVG. For the coordinates, I began with an arbitrary starting point, noting that the first point needed to be the lowest as to mark the dynamic "piano," which means soft or quiet in volume. I then increased my x value alternately by 30 and 25 in order to create the sporadic ups and downs of volume. Doing this by hand meant I was simply had to guess and check, plugging in various numbers to see what worked and what did not in order to create these curves. We then placed the verbs on these curves by creating circles and doing more guess and check work to place them on the curves.
Hand Drawn vs. XSLT Creating this graph was no easy task. Typically this type of graph is written with some sort of equation, or fixed variable values in order to have consistency in how the graph moves. But it is impossible to write an equation for music, particularly for something like volume which is fluid and uncertain. Thus it seemed to be the best decision, initially, to create this graph by hand. Admitttedly this process took significantly longer than if we had some sort of equation, but it was the most accurate representation of our data, as it represents crescendos and decrescendos, or subtle rises and falls in volumes more smoothly than the XSLT produced data, and it creates a more condensed graph that is perhaps more aesthetically pleasing. However, the XSLT is a more uniform format of creating a graph to represent dynamics, and has a specific formula to plug into any song in order to accurately represent its dynamic changes. The downside to this graph is its rigid appearance, which does not allow for the crescendos and decrescendos the way the handdrawn graph appears. While both approaches to creating this graph have advantages and disadvantages, it is safe to conclude that the graph produced by XSLT is more practical and easier to use for future research.