Post by Gypsy Picker on Nov 15, 2006 18:33:27 GMT -5
In the songwriting humor thread, Cornflake talks about his strict adherence to a predefined playlist. He develops his list with considerable thought about how an audience may percieve not just the song, but the entire string of songs as they unroll. I have been fascinated and puzzled by this very subject for some time, and have adopted a different approach. I'll get to that later (much later, in another post perhaps) -- first I want to postulate a conjecture about songs themelves.
We can classify songs in many ways by various properties they have, including their key, tempo, duration, genre, voicing, subject matter, and even expected emotional response, to name but a few. There are many other properties that every song has, but listing them all isn't my purpose here. Cornflake made the supposition that starting a set with a humorous song may cause some in the audience to then expect the next song (and perhaps the third song, etc.) to be humorous as well. In this way, he puts each song in some context with what happens before and after the song, thereby injecting some additional meaning to the song that only comes through by virtue of the context in which the song is presented.
We know that songs themselves also have multiple meanings, even if it is presented by itself with no context. The lyric alone has meaning (usually), but gains additional meaning in the context of the melody and harmony. One can even view key modulation, bridges, choruses, intros, outros, and various other musical parts as adding meaning (even if it's only puncuation of sorts). But the purpose of the song is not simply to tell the story contained in the lyrics, but to evoke an emotional response (or series of emotional responses) in the telling of the story lyrically AND musically combined. A song then comprises elements that themselves contain messages and meaning, but when taken as a whole contain a sort of "meta-meaning" evoked in the mind of the listener. That is to say that a song is, amongst other things, a "self-contained emotional meta-message" (SCEMM).
This brings us to my conjecture:
Premise: songs trigger emotional response(s)
If we accept the premise, then
1) Given a sufficiently adequate means to isomorphically (in a symbolically meaningful way) characterize emotional responses to songs, songs could then be described in terms (words) that symbolize their self-contained emotional meta-message (SCEMM).
Metaphorically, the phrase "walk through the tall grass" contains a message that evokes (triggers) a similar isomorphic meaning in the minds of all English speakers. We all envision in our minds essentially the same thing. Note that taken separately, the words "through", "grass", "walk", "the", and "tall", each have little or no inherent meaning. Strung together, the context of the words trigger their respective individual meanings, and moreover, present the collective meta-meaning that we all can envision.
If we accept (1), then
2) Once given "words" (the isomorphic symbology), SCEMMs can be combined to form strings of SCEMMs (or SCEMM phrases) and strings of SCEMM phrases, such that some will contain additional meaning (meta-meta-meaning). That is to say that presenting songs in succession can trigger new emotional responses that are not available without the context of the whole SCEMM string. Note that this does not guarantee that all such SCEMM strings will have additional meaning. Just as stringing together random English words can make non-sensical phrases (where no meaningful isomorphic symbol is triggered), the same must hold true for SCEMM strings.
Metaphorically, books can be made from chapters that are collections of paragraphs strung together from sentences strung together from words. Just as we can have emotional responses to the concept of an entire book that we've read, so too can we have emotional responses to a collection of songs we've heard in succession.
If we accept (2), then
3) There must be rules of grammar that operate on the words that symbolize SCEMMs such that well-formed SCEMM strings can be produced. This does not guarantee meaningfulness, just as a grammatically correct English phrases can be constructed that conveys no isomorphic meaning: "the quick fence became noticebly absorbed in peacock tremblings".
Because human emotional responses are not isomorphic to any "real-world" construct - that is we cannot map the concept of "humility", for example, to any specific shared cultural or physical phenomenon. Hence, where we all have a similar symbolic representation and understanding of "walk through the tall grass", we do not all share a common representation and understanding of complex emotional states. We may be able to conceptualize "humility" or "grief-stricken" as an emotional property belonging to someone, but only in the sense that it is symbolized internally in each of our own minds. The only commonality that allows us to communicate these concepts is that we ourselves have experienced them to some extent, but often in very different ways from others. In this way, however, we get around the sticky problem of IF a complex emotional brain-state CAN be isomorphized onto a symbol - this example proves that it can be done. By virtue of being able to communicate the concept of "empathy" without a direct "real-world" counterpart, we effectively show that given adequate language tools we can meaningfully symbolize (isomorphize) an emotional state.
Note also that some people have different emotional responses (collectively called "brain-states") to different triggers. The loss of the family pet, for example, may cause one Mrs. Johnson considerable grief and Mr. Johnson considerable relief (the cold-hearted bastard). This not-withstanding, the intended emotional response of "his dog up and died, his dog up and died, after twenty years he still grieves" used in a lyric evokes the same emotion in both Mr. and Mrs. Johnson simply because the intent is made clear within the context of the lyric and musical presentation of the concept. Hence, even though Mr. Johnson feels relief at the loss of his own dog, he can feel empathy for Mr. Bojangles as intended.
This all brings us to the big question: Does there exist (and if not, CAN there exist) an adequate set of language tools for us to characterize SCEMMs and build meaningful SCEMM phrases? To do this, we would not only need words that are already isomorphically mapped to emotions and complex brain-states triggered by songs, but phrases that convey additional isomorphic meaning when these words (and song counterpoints they symbolize) are strung together in meaningful ways.
We obviously already have a vocabulary for communicating emotional concepts, and to some degree can use this vocabulary to charactize a song (or parts thereof) in very general terms. However, we do not in everyday parlance describe a song in terms of all of its emotional triggers and how they relate to each other, nest within each other, or collectively string together to produce complex brain-states. And how would we accomplish this? First, we would have to determine if songs are even isomorphic to existing language constructs. In other words, can a song's component parts be mapped to words, phrases, sentences, paragraphs and the like? We would have to agree what components of a song could be mapped to which symbols (English words), and which symbol is most isomorphic to the emotion or brain state it represents. Then there is a matter of scale - we do not percieve a melody by contemplating the meaning of it's individual component notes, but rather as the string or "sentence" (or even paragraph) the notes combine to make when taken as a whole, complete with punctuation.
Moreover, we know that there do exist words that are commonly used to describe the component parts of a song: Intro, verse, chorus, bridge, etc. We also have many other isomorphic English (and adopted foreign language) words to describe and understand song structure: melody, harmony, bass, rythym, key, time signature, tempo, bar, measure, arpeggio, crescendo, cadence and many others. But which of these constructs (or combinations thereof) and what specific qualitative property or properties do they posess that are isomorphically mapable to emotional triggers?
If we accept (1) and (2), then
3) There exists sufficiently adequate language tools to isomorphically describe a song as a self-contained emotional meta-message, and further, to describe a string of songs as a meaningful string of SCEMMs.
Hence, what is lacking is not so much the language tools, but a set of "song rules" that map the song parts and there given properties to specific emotional responses and brain-states.
Now I shall leave this discussion to your contemplation.
We can classify songs in many ways by various properties they have, including their key, tempo, duration, genre, voicing, subject matter, and even expected emotional response, to name but a few. There are many other properties that every song has, but listing them all isn't my purpose here. Cornflake made the supposition that starting a set with a humorous song may cause some in the audience to then expect the next song (and perhaps the third song, etc.) to be humorous as well. In this way, he puts each song in some context with what happens before and after the song, thereby injecting some additional meaning to the song that only comes through by virtue of the context in which the song is presented.
We know that songs themselves also have multiple meanings, even if it is presented by itself with no context. The lyric alone has meaning (usually), but gains additional meaning in the context of the melody and harmony. One can even view key modulation, bridges, choruses, intros, outros, and various other musical parts as adding meaning (even if it's only puncuation of sorts). But the purpose of the song is not simply to tell the story contained in the lyrics, but to evoke an emotional response (or series of emotional responses) in the telling of the story lyrically AND musically combined. A song then comprises elements that themselves contain messages and meaning, but when taken as a whole contain a sort of "meta-meaning" evoked in the mind of the listener. That is to say that a song is, amongst other things, a "self-contained emotional meta-message" (SCEMM).
This brings us to my conjecture:
Premise: songs trigger emotional response(s)
If we accept the premise, then
1) Given a sufficiently adequate means to isomorphically (in a symbolically meaningful way) characterize emotional responses to songs, songs could then be described in terms (words) that symbolize their self-contained emotional meta-message (SCEMM).
Metaphorically, the phrase "walk through the tall grass" contains a message that evokes (triggers) a similar isomorphic meaning in the minds of all English speakers. We all envision in our minds essentially the same thing. Note that taken separately, the words "through", "grass", "walk", "the", and "tall", each have little or no inherent meaning. Strung together, the context of the words trigger their respective individual meanings, and moreover, present the collective meta-meaning that we all can envision.
If we accept (1), then
2) Once given "words" (the isomorphic symbology), SCEMMs can be combined to form strings of SCEMMs (or SCEMM phrases) and strings of SCEMM phrases, such that some will contain additional meaning (meta-meta-meaning). That is to say that presenting songs in succession can trigger new emotional responses that are not available without the context of the whole SCEMM string. Note that this does not guarantee that all such SCEMM strings will have additional meaning. Just as stringing together random English words can make non-sensical phrases (where no meaningful isomorphic symbol is triggered), the same must hold true for SCEMM strings.
Metaphorically, books can be made from chapters that are collections of paragraphs strung together from sentences strung together from words. Just as we can have emotional responses to the concept of an entire book that we've read, so too can we have emotional responses to a collection of songs we've heard in succession.
If we accept (2), then
3) There must be rules of grammar that operate on the words that symbolize SCEMMs such that well-formed SCEMM strings can be produced. This does not guarantee meaningfulness, just as a grammatically correct English phrases can be constructed that conveys no isomorphic meaning: "the quick fence became noticebly absorbed in peacock tremblings".
Because human emotional responses are not isomorphic to any "real-world" construct - that is we cannot map the concept of "humility", for example, to any specific shared cultural or physical phenomenon. Hence, where we all have a similar symbolic representation and understanding of "walk through the tall grass", we do not all share a common representation and understanding of complex emotional states. We may be able to conceptualize "humility" or "grief-stricken" as an emotional property belonging to someone, but only in the sense that it is symbolized internally in each of our own minds. The only commonality that allows us to communicate these concepts is that we ourselves have experienced them to some extent, but often in very different ways from others. In this way, however, we get around the sticky problem of IF a complex emotional brain-state CAN be isomorphized onto a symbol - this example proves that it can be done. By virtue of being able to communicate the concept of "empathy" without a direct "real-world" counterpart, we effectively show that given adequate language tools we can meaningfully symbolize (isomorphize) an emotional state.
Note also that some people have different emotional responses (collectively called "brain-states") to different triggers. The loss of the family pet, for example, may cause one Mrs. Johnson considerable grief and Mr. Johnson considerable relief (the cold-hearted bastard). This not-withstanding, the intended emotional response of "his dog up and died, his dog up and died, after twenty years he still grieves" used in a lyric evokes the same emotion in both Mr. and Mrs. Johnson simply because the intent is made clear within the context of the lyric and musical presentation of the concept. Hence, even though Mr. Johnson feels relief at the loss of his own dog, he can feel empathy for Mr. Bojangles as intended.
This all brings us to the big question: Does there exist (and if not, CAN there exist) an adequate set of language tools for us to characterize SCEMMs and build meaningful SCEMM phrases? To do this, we would not only need words that are already isomorphically mapped to emotions and complex brain-states triggered by songs, but phrases that convey additional isomorphic meaning when these words (and song counterpoints they symbolize) are strung together in meaningful ways.
We obviously already have a vocabulary for communicating emotional concepts, and to some degree can use this vocabulary to charactize a song (or parts thereof) in very general terms. However, we do not in everyday parlance describe a song in terms of all of its emotional triggers and how they relate to each other, nest within each other, or collectively string together to produce complex brain-states. And how would we accomplish this? First, we would have to determine if songs are even isomorphic to existing language constructs. In other words, can a song's component parts be mapped to words, phrases, sentences, paragraphs and the like? We would have to agree what components of a song could be mapped to which symbols (English words), and which symbol is most isomorphic to the emotion or brain state it represents. Then there is a matter of scale - we do not percieve a melody by contemplating the meaning of it's individual component notes, but rather as the string or "sentence" (or even paragraph) the notes combine to make when taken as a whole, complete with punctuation.
Moreover, we know that there do exist words that are commonly used to describe the component parts of a song: Intro, verse, chorus, bridge, etc. We also have many other isomorphic English (and adopted foreign language) words to describe and understand song structure: melody, harmony, bass, rythym, key, time signature, tempo, bar, measure, arpeggio, crescendo, cadence and many others. But which of these constructs (or combinations thereof) and what specific qualitative property or properties do they posess that are isomorphically mapable to emotional triggers?
If we accept (1) and (2), then
3) There exists sufficiently adequate language tools to isomorphically describe a song as a self-contained emotional meta-message, and further, to describe a string of songs as a meaningful string of SCEMMs.
Hence, what is lacking is not so much the language tools, but a set of "song rules" that map the song parts and there given properties to specific emotional responses and brain-states.
Now I shall leave this discussion to your contemplation.
