Annotated Bibliography Assignment

 

By Shaun Sanders, Textones Shakespearean Sonnet Team

 

 1.

 

Torres, David 1, Turnbull, Douglas 1, Barrington , Luke 2, Lanckriet , Gert 2. “Identifying

Words That Are Musically Meaningful.” Dept. of Computer Science and Engineering1, Dept. of Electrical and Computer Engineering2, University of California, San Diego 2007 http://cosmal.ucsd.edu/~gert/papers/MusicVocab_ISMIR07.pdf

 

This paper from the halls of UC San Diego concerns itself primarily with identification of words within a piece of music, where the word is to be recognized by a computer, not as text, but as an audio phonic signal. It is interesting for this reason because it seeks to perform a function that is in some ways the opposite of the function we will attempt to produce, which is the application of musical values to words. However, the paper is not interesting for that reason alone; it’s primary importance to our work concerns a section where the concept of Human Agreement is discussed.

Using data from 500 songs, undergrad students were asked to categorize them by genre using a limited vocabulary of 173 words. After analyzing the results, the researcher concluded: “Using this data, we can calculate a statistic we refer to as human agreement” (2). They then came up with a mathematical equation which determined “word-song pairs.” Use of this equation allows the researchers to come up with a scientifically accurate method to determine how the computers parameters should behave, and which search word should be assigned to what songs.  

This concept of agreement is important to us because we have some debate as to how to assign tonal values to words. While we are in agreement that parts of speech are good units of assignation, confusion arises when we consider which tone should represent which part of speech. For example, should a noun be assigned the root note in our scale, or should a different tonal value be used? If so, why? We have discussed two tonal concepts so far, and may settle on our second one, which emphasizes commonly used western scale intervals rather than simple chromatic assignations. However, until we experiment, we have little idea how our results will sound, and it is very likely at some stage that we will have to apply a type of mathematical equation to a surveyed group of our peers in order to determine the “human agreement” value of the tonal values we assign.

This paper goes on to discuss acoustical correlations and semantic annotations within the Canon, however, this area does not intersect our own line of inquiry.

* * * * *

 

---

 

2. 

Scher, Steven Paul, Lodato, Suzanne M, Bernhart ,Walter, Aspden, Suzanne. Word and Music

Studies: “Essays in Honor of Steven Paul Scher and on Cultural Identity and the Musical Stage.”  Editions Rodopi B.V., Amsterdam-New York 2002 http://books.google.com/books?id=ZWOmyQemqMYC&pg=PR7&dq=Essays+in+Honor+of+Steven+Paul+Scher+and+on+Cultural+Identity+and+the+Musical+Stage&sig=JdRXqch2rRARiy6WgyIouQhyjdE#PPR8,M1

 

This work deals with the relationship between words and music, especially in terms of contextuality. In the introduction, several researchers are quoted, among them, Eric Prieto, who is, I believe, the same French-Italian Professor I am currently taking a class with at UCSB. This essay states that Prieto deals with the concept of contingency, “specifically in metaphorical meanings assigned to music” (Scher viii). Scher himself maps “ ‘intracompositional’ word/music relationships.”

            Just as the researchers in my first citation have a different focus than we do, these researchers are not applying tonal values to text. However, they are closely examining the relationship between text and music, questioning the metaphorical application of music to text, and examining the level of contingency upon excerpts of text. While this might be considered a traditional “close reading” approach, the application of which we choose to avoid, their findings have a direct bearing on our results in a comparative sense. For example, will we find their theories have any relevance to the sonic soundscapes we may be able to produce? Will our tonal applications share the same metaphorical world as Professor Prieto’s, and, therefore, invoke a new interest in the relationship between words and music?

            Acting as a yardstick of literary interpretation, this source allows us to address questions that will surely be lurking in the back of our collective mind as we experiment with our tonal values: How close to making “normal” sounding music should we attempt to come? At what point is the attempt at making music interfering with a simple rendering of tonal values that, though they might sound unmusical, reveal something of interest concerning literary structures? And, finally, will we find that we can marry these seemingly divergent ideas into a soundscape that might be considered entirely contingent on the message of the text, yet sounds pleasing to the human ear?

 

* * * * *

---

 

3. 

 

Protools (PC) 2005 Digital Design

Reason (PC) 2005 Propellerhead Audio Sequencing software.

 

The ultimate goal of our project is to find or create a simple music sequencing application that can be used by any internet surfer (Nerfer?) who wants to do an audio analysis of a literary work. For that reason, both these software applications are unsuitable for our purposes. This software is not available in the public domain, and, while it is not difficult to learn with time, it is probably too complex for the average literary buff to use with ease; convenience is important if we want our system to be used by such people. However, when considering our allotted five week time span, our goals seem somewhat ambitious. To that end, we have a fallback plan, which is to create audio maps of Shakespearean sonnets ourselves within the realm of Propellerhead/Reason, a fairly sophisticated audio sequencing software that is run in conjunction with Protools, a system of which I happen to be the proud owner. With this software, we can experiment with a wide range of audio sounds in the hope that we may find a few sounds that are most suited to our endeavors. Their use will also allow us to consider the questions of tempo. For example, Reason will allow us to adjust tempo without raising the pitch of a tone, while Protools will give us a visual representation of note placement on a scale. Both systems allow for the juggling of meter, and this will allow us to further consider the ramifications of syllable use, phrasing, and syntax as they apply to our end result. We will probably have to settle on an appropriate bar/ time signature for iambic pentameter and then possible compare them to results gained from using other types of literature, if we find we have the time.

            The results of our experiments with these software tools can be loaded as audio files for close listening by those who are interested, and they can then help determine if our line of thinking regarding Textones has any validity. However, our primary goal is to provide a service that, while it may be more simple to use than these software applications, will allow end users to apply our Textones system to texts of their choice.

 

* * * * *

---

4. 

Introduction to Sound and Meter.  Purdue UniversityOnline Writing Lab (OWL).

http://owl.english.purdue.edu/handouts/print/general/gl_soundmeter.html

 

Purdue University is well known for its Online Writing Lab (OWL), so well known, in fact, that their terminologies and analytical tools might be considered definitive in educational circles. To that end, their examination of meter and sound in text serves as an important guideline in our decision making regarding applications of meter and time-count to our audio representations of the Shakespearean sonnet. An example of one problem we have already encountered is the allotment of beats given to each line of a sonnet. For example, iambic pentameter is not often perfectly regular; irregularities sometimes appear in the written work that are addressed by syntax and rhyme of a following line. Sometimes a line may have ten, perfectly spaced beats, but sometimes not. Therefore, we are presented with a choice: do we run every line back-to-back with no gaps in between, or do we leave a gap in order to reflect the textual irregularity within our soundscape? The problem arises that, if we don’t make some compensation for an irregularity, the listener has no idea where a line starts and ends, and the soundscape then loses its purpose. We feel the text has to adhere to a regular time-count, much as a song might adhere to a four-beat bar to keep pace with a certain rhythm, and this is where the principles of literary meter come into play. We must decide that, if a sonnet is mostly in iambic pentameter, should we create a ten-beat bar per line? Or, as pentameter feels rather truncated at the end of each line, should we allow a sixteen beat bar? Imagine it this way: When reading iambic pentameter, even when it is very evenly composed, we instinctively want to leave a “rest” at the end of each line—we cannot rush into the next line. However, the opposite is true; if we pause at the end of each line, we create a “rest” that seems quite natural. Typically, I believe we truncate the last few bars of an iambic pentameter line purely for the sake of time conservation. 

To this end, we will experiment with different bar lengths and different meters to see what gives the most interesting results. In any event, we will have to research concepts of meter thoroughly in order to apply them to our Textones.

 

* * * * *

---

5. 

 

 

Shakespeare, William. Sonnets. http://www.shakespeares-sonnets.com/

            Oxquarry Books Ltd,  Oxford, OX1 4LF, UK.

 

This website is owned by Oxquarry Books Ltd, a publishing company specializing in Shakespearean literature. They say that “Sonnet 60, gives the flavor of many,” so this may make a good starting point for our line of inquiry.

 

LX

 

Like as the waves make towards the pebbled shore,

 

So do our minutes hasten to their end,

 

Each changing place with that which goes before

 

In sequent toil all forwards do contend.

 

Nativity, once in the main of light,

 

Crawls to maturity, wherewith, being crowned,

 

Crooked eclipses 'gainst his glory fight

 

And Time that gave, doth now his gift confound.

 

Time doth transfix the flourish set on youth,

 

And delves the parallels in beauty's brow,

 

Feeds on the rarities of natures truth,

 

And nothing stands but for his scythe to mow;

 

 And yet, to times, in hope, my verse shall stand,

 

 Praising thy worth, despite his cruel hand.

 

 

Shakespearean sonnets are the perfect guinea pig for our teams attempts at applying tonal values to words. This is because they are, for the most part, strict in their adherence to meter. The benefit of this is that, when we apply tones to parts of speech, we will be able to compare his sonnets side by side at an audio listening in order to detect similarities or contrasts in any emergent patterns. It is possible that we may also detect some kind of correlation between the sonnet form and any emergent audio patterns. Next, we might compare Shakespearean sonnets with sonnets by other writers to see if similar patterns appear between the two, and whether a writer has as distinct a tone within the soundscape as he or she does in the textual arena.

            If we can establish that there is some predictable soundscape presenting itself from the sonnet form, we may then be able to go on and, by applying tones to other poetic forms, see if there is then a correlation existing in those forms also. Ultimately, we may be able to compare various poetic forms and, maybe, even short stories or narratives.

            Beyond the sonnet form, by using an acknowledged genius like Shakespeare as a starting point, we might optimistically predict the most pleasing results.

 

 * * * * *

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

6.

TAPor, Crawdad, other text analysis tools.

 

    While these tools are remarkable for their computational power, they are limited by their parameters. That is to say, they do not think. In a typical text analysis search, a computer will come up with a word because it is looking for the codes that are assigned to certain letters. For example, the word “think” might appear to the computer as “110 010 101 100 111” (although it is probably more complex than this example). It simply seeks out words that match the order of the letter codes as they are typed into the search panel.

            Our imaginings run such that we think it should be possible to have a computer search for, not just words, but parts of speech. This sounds simple enough; the computer should be able to find a verb if I type in a verb. But how about if I request it finds me all the verbs in a literary work by the description, “find verbs”? It would have to compute, not only the binary codes for most alphabetical letters, but it would also have to understand the relationships of those letters to each other in their formation of verbs, thus allowing the search for verbs to commence.

            While this might be a bit futuristic, it is a logical step in computer development. The advantages of such a text analyzer is that, instead of our team working with short literary works—assigning tonal values physically to each part of speech—a researcher might be able to feed an entire novel into an analyzer and thereby apply 12 tones to the entire work. I emailed the people at Crawdad regarding this possibility, but have not heard back. They probably thought I was mad. Our examination of text analyzers like Crawdad and TAPor may simply result in a description of their limitations. However, until we find a text analyzer that fulfills our requirements, we might still attempt to use them on one or two slightly larger works in order to determine if our Textones system can be applied to bigger bodies of work successfully.

 

* * * * *

 

---