What is Music? The Ontological Status of Musical Works

The ontological status of musical works is a controversial topic among those in the field. This paper aims to argue for a nominalist and non-reductive approach to musical works that differentiates musical works from the sound structures commonly equated with them. The difference between sound structures and musical works is heavily emphasized and I conclude that musical works have emergent attributes sound structures do not. These attributes being creatability, fine individuation, and the inclusion of performance means. After establishing this I begin to build an argument for a nonreductive approach to musical works by rejecting the extreme view held by compositional nihilists. Once I establish that musical works are entities in our ontology I argue against them being abstract. I end by defending the nominalist approach. What is Music? The Ontological Status of Musical Works What precisely is a musical work? What sort of thing is music, if it even exists? These questions seem rather silly at first glance. Music is what we hear, just another form of artistic expression, a thing we listen to and derive enjoyment from. Yet from a metaphysical standpoint these answers are not satisfying; we do not come any closer to the truth from the folk view of music. When the metaphysician is willing to write away the tangible world as not truly real we must ask where that leaves the intangible, including things like music. Philosophers are heavily divided on the ontological status of musical works, disagreeing not only on how musical works exist, but whether they exist at all. Many of the opposing options offered by thinkers prove unsatisfactory. Asserting that musical works are a higher-level kind that exist as non-abstract entities intuitively seems to be a tenable approach to the ontological status of musical works. What is a sound? Prior to addressing the ontological status of musical works one must first consider what sort of entity a sound is, for one surely cannot have music without sound. Perhaps sounds are objects? This seems unlikely, we would be asserting that sounds are an abstract particular that exists independent yet somehow related to some source object. This approach raises more questions than answers. Perhaps, then, sounds Res Cogitans (2017) 8 Pipko | 91 2155-4838 | commons.pacificu.edu/rescogitans are properties of objects, of a sounding body? This is the view supported by Locke, now referred to as the Object-Property View of sounds. On this approach sounds are like colors, they exist as either a stable or transient property of an object. This explanation of the relationship between source and sound is a major advantage for Object-Property View, however it is not without its problems (Leddington 2, 29). Consider the sound created when dribbling a basketball. Each time the ball makes contact with the floor it creates a dull thud. By what means is the basketball instantiating a sound? Is the sound not a property of the floor when hit by a ball-like object? Treating sounds as properties of an object does not offer any explanation as to what is actually creating the sound, this is a major disadvantage of the Object-Property View. Leddington offers a far more tenable view of sounds as properties; he posits that a sound has a temporal and spatial relationship to an event from which it originates. A change in a sound relates to a corresponding change in the source event, for example "when the sound of a plucked guitar string decays, this corresponds to the decay in the vibration of the guitar string" (2). In the case of the basketball, the sound is instantiated by the ball hitting the floor. This approach is known as the Event-Property View. Attributing sounds as properties of events does not seem to be enough; we can extrapolate an even more informative conception of sounds. Events are intangible, they are composed of the interaction of objects; perhaps we should not separate the two. In the example above, the properties of the ball and floor are changed during the event in such a way that they acquire the property of sound. Sounds are therefore constituent properties of events. Viewing sounds as constituent properties of events offers us the most plausible conception of what a sound is. Before moving on I'd like to address an objection to approaching sounds as properties: Instead of making sounds properties, why can we not simply equate the sounds we hear with sound waves? This seems reasonable, however it is counterintuitive. Sound waves do not do "justice to the apparent locatedness of sounds" (2). Waves travel through a medium from a source object, we are unable to attribute the sound to the object, or event, from which they appear to originate. It might also be a hasty move to equate the sounds we interpret from sound waves with the waves themselves. Sound waves seem to be able to exist independently of sounds themselves. Consider the old conundrum of a tree falling in the woods with no observer, sound waves are certainly created -but is there a sound? Most importantly, this approach offers us little from a metaphysical standpoint. Identifying sounds with sound waves does not offer any greater explanatory power over identifying sounds with the constituent properties. A guitarist strums a chord, this causes an event that produces a sound that is a property of the particular strings strummed on the chord. This guitarist continues to strum different chords, creating a structured sequence of sounds. Is this music? Could we say that music is simply sound organized in a particular fashion? The answer seems to be Res Cogitans (2017) 8 Pipko | 92 2155-4838 | commons.pacificu.edu/rescogitans no, there are a number of kinds of things that are structured sounds that are not music, such as human speech. We also cannot say that musical works are merely structured sounds with some kind of aesthetic qualities, as poetry is also an aesthetic experience that is capable of being heard (Kania). All musical works are certainly aesthetic sound structures, but it would be erroneous to assume that there is nothing else to a piece of music over and above these two qualities. Even if we posit that musical works possess the necessary condition of being created we are still met with resistance from poetry, which is also created. What then differentiates a piece of music from a work of poetry? We could suppose it is 'tonality', features such as pitch or rhythm, but what of pieces that do away with musical convention, like harsh noise? The best way to approach music is perhaps as a sound structure created with aesthetic attributes that are not exclusively linguistic, and that will usually have some kind of 'tonality' (Kania). Even these attributes are debated and we come no closer to a satisfactory understanding of the ontological status of musical works. Let us further consider the requirements for what we think of as a musical work. A musical work possesses some features above and beyond that of a sound structure. However, is there a parsimonious method of integrating the features we ascribe to musical works into our ontology? Levinson offers three important criteria worth considering: creatability, fine individuation, and the inclusion of performance means. These requirements will allow us to better understand how to approach musical works from an ontological standpoint. If sound structures were equivalent to musical works then musical works would not have the capacity of being created. I assume, in agreement with others, that sound structures are abstract entities. It is important to note that "purely sound structures are in effect mathematical objects -they are sequences of sets of sonic elements" (Levinson 7). The abstract nature of sound structures make them unable to be identical with musical works. Abstract entities are atemporal, therefore capable of being instantiated at any point in time. It is ridiculous to assert that a piece of music could have existed prior to it being created by whoever wrote or composed it; can we truthfully say that David Bowie's "Ziggy Stardust" existed prior to 1972? If "Ziggy" were nothing more than an abstract sound structure it could have been instantiated during the Paleozoic Era, existing even before the instruments that are featured on it. The main reason for holding onto creatability is simply "that it is one of our most firmly entrenched ideas concerning art. There is probably no idea more central to thought about art than that it is an activity in which participants create things -these things being art works ... that it is a godlike activity in which the artist brings into being what did not exist beforehand ... that artists truly add to the world." (8). The creatability requirement is an essential attribute of musical works. Res Cogitans (2017) 8 Pipko | 93 2155-4838 | commons.pacificu.edu/rescogitans Another issue we run into when we equate abstract sound structures with musical works is that it seems possible that two distinct composers could determine the same sound structure, thus composing the same work. However, distinct composers create differing works, even if they determine the same sound structure. It is essential to consider the musico-historical context in which a work is composed (10). Consider "The Fish Cheer", a psych rock song released in 1967 that, adhering to the Zeitgeist of the sixties, protested the war in Vietnam through fuzzy, distorted riffs. Such a work written today, assuming it had not yet been created, would be laughable not only in subject matter but also in style. Context is essential to musical works. Imagine if a similar sound structure to the Ramones "Blitzkrieg Bop" had first been instantiated by an aesthetically different kind of punk band, Crass. Taken with their other works, we would have a song that was darker, angrier, and more raw and visceral (11). Levinson provides his second requirement, that of fine individuation, which states that "musical wo


What is Music? The Ontological Status of Musical Works
What precisely is a musical work?What sort of thing is music, if it even exists?These questions seem rather silly at first glance.Music is what we hear, just another form of artistic expression, a thing we listen to and derive enjoyment from.Yet from a metaphysical standpoint these answers are not satisfying; we do not come any closer to the truth from the folk view of music.When the metaphysician is willing to write away the tangible world as not truly real we must ask where that leaves the intangible, including things like music.Philosophers are heavily divided on the ontological status of musical works, disagreeing not only on how musical works exist, but whether they exist at all.Many of the opposing options offered by thinkers prove unsatisfactory.Asserting that musical works are a higher-level kind that exist as non-abstract entities intuitively seems to be a tenable approach to the ontological status of musical works.
What is a sound?Prior to addressing the ontological status of musical works one must first consider what sort of entity a sound is, for one surely cannot have music without sound.Perhaps sounds are objects?This seems unlikely, we would be asserting that sounds are an abstract particular that exists independent yet somehow related to some source object.This approach raises more questions than answers.Perhaps, then, sounds 2155-4838 | commons.pacificu.edu/rescogitansare properties of objects, of a sounding body?This is the view supported by Locke, now referred to as the Object-Property View of sounds.On this approach sounds are like colors, they exist as either a stable or transient property of an object.This explanation of the relationship between source and sound is a major advantage for Object-Property View, however it is not without its problems (Leddington 2,29).
Consider the sound created when dribbling a basketball.Each time the ball makes contact with the floor it creates a dull thud.By what means is the basketball instantiating a sound?Is the sound not a property of the floor when hit by a ball-like object?Treating sounds as properties of an object does not offer any explanation as to what is actually creating the sound, this is a major disadvantage of the Object-Property View.Leddington offers a far more tenable view of sounds as properties; he posits that a sound has a temporal and spatial relationship to an event from which it originates.A change in a sound relates to a corresponding change in the source event, for example "when the sound of a plucked guitar string decays, this corresponds to the decay in the vibration of the guitar string" (2).In the case of the basketball, the sound is instantiated by the ball hitting the floor.This approach is known as the Event-Property View.Attributing sounds as properties of events does not seem to be enough; we can extrapolate an even more informative conception of sounds.Events are intangible, they are composed of the interaction of objects; perhaps we should not separate the two.In the example above, the properties of the ball and floor are changed during the event in such a way that they acquire the property of sound.Sounds are therefore constituent properties of events.Viewing sounds as constituent properties of events offers us the most plausible conception of what a sound is.
Before moving on I'd like to address an objection to approaching sounds as properties: Instead of making sounds properties, why can we not simply equate the sounds we hear with sound waves?This seems reasonable, however it is counterintuitive.Sound waves do not do "justice to the apparent locatedness of sounds" (2).Waves travel through a medium from a source object, we are unable to attribute the sound to the object, or event, from which they appear to originate.It might also be a hasty move to equate the sounds we interpret from sound waves with the waves themselves.Sound waves seem to be able to exist independently of sounds themselves.Consider the old conundrum of a tree falling in the woods with no observer, sound waves are certainly created --but is there a sound?Most importantly, this approach offers us little from a metaphysical standpoint.Identifying sounds with sound waves does not offer any greater explanatory power over identifying sounds with the constituent properties.
A guitarist strums a chord, this causes an event that produces a sound that is a property of the particular strings strummed on the chord.This guitarist continues to strum different chords, creating a structured sequence of sounds.Is this music?Could we say that music is simply sound organized in a particular fashion?The answer seems to be 2155-4838 | commons.pacificu.edu/rescogitansno, there are a number of kinds of things that are structured sounds that are not music, such as human speech.We also cannot say that musical works are merely structured sounds with some kind of aesthetic qualities, as poetry is also an aesthetic experience that is capable of being heard (Kania).All musical works are certainly aesthetic sound structures, but it would be erroneous to assume that there is nothing else to a piece of music over and above these two qualities.Even if we posit that musical works possess the necessary condition of being created we are still met with resistance from poetry, which is also created.What then differentiates a piece of music from a work of poetry?We could suppose it is 'tonality', features such as pitch or rhythm, but what of pieces that do away with musical convention, like harsh noise?The best way to approach music is perhaps as a sound structure created with aesthetic attributes that are not exclusively linguistic, and that will usually have some kind of 'tonality' (Kania).Even these attributes are debated and we come no closer to a satisfactory understanding of the ontological status of musical works.
Let us further consider the requirements for what we think of as a musical work.A musical work possesses some features above and beyond that of a sound structure.However, is there a parsimonious method of integrating the features we ascribe to musical works into our ontology?Levinson offers three important criteria worth considering: creatability, fine individuation, and the inclusion of performance means.These requirements will allow us to better understand how to approach musical works from an ontological standpoint.
If sound structures were equivalent to musical works then musical works would not have the capacity of being created.I assume, in agreement with others, that sound structures are abstract entities.It is important to note that "purely sound structures are in effect mathematical objects --they are sequences of sets of sonic elements" (Levinson 7).The abstract nature of sound structures make them unable to be identical with musical works.Abstract entities are atemporal, therefore capable of being instantiated at any point in time.It is ridiculous to assert that a piece of music could have existed prior to it being created by whoever wrote or composed it; can we truthfully say that David Bowie's "Ziggy Stardust" existed prior to 1972?If "Ziggy" were nothing more than an abstract sound structure it could have been instantiated during the Paleozoic Era, existing even before the instruments that are featured on it.The main reason for holding onto creatability is simply "that it is one of our most firmly entrenched ideas concerning art.There is probably no idea more central to thought about art than that it is an activity in which participants create things --these things being art works … that it is a godlike activity in which the artist brings into being what did not exist beforehand … that artists truly add to the world."(8).The creatability requirement is an essential attribute of musical works.
2155-4838 | commons.pacificu.edu/rescogitansAnother issue we run into when we equate abstract sound structures with musical works is that it seems possible that two distinct composers could determine the same sound structure, thus composing the same work.However, distinct composers create differing works, even if they determine the same sound structure.It is essential to consider the musico-historical context in which a work is composed (10).Consider "The Fish Cheer", a psych rock song released in 1967 that, adhering to the Zeitgeist of the sixties, protested the war in Vietnam through fuzzy, distorted riffs.Such a work written today, assuming it had not yet been created, would be laughable not only in subject matter but also in style.Context is essential to musical works.Imagine if a similar sound structure to the Ramones "Blitzkrieg Bop" had first been instantiated by an aesthetically different kind of punk band, Crass.Taken with their other works, we would have a song that was darker, angrier, and more raw and visceral (11).Levinson provides his second requirement, that of fine individuation, which states that "musical works must be such that composers composing in differing musico-historical contexts who determine identical sound structures invariably compose distinct musical works" ( 14).
The final element that helps us distinguish between sound structures and musical works is that of performance, which is an essential quality of music.Musicians are familiar with the sounds they intend to create; they intend for their work to be performed in a particular way or with a particular instrument.Essentially, "musical works are such that specific means of performance or sound production are integral to them" (19).Consider divorcing the works of Louis Armstrong from his trumpet, or that of Giorgio Moroder from the synthesizer.Musical works are intrinsically linked with their method of performance.We can imagine there is some leniency we can allow for, especially when we consider covers that use different but similar instruments, ¬"All Along the Watchtower" for instance.These are in some sense differing pieces of music, so generally this is a safe requirement to place upon musical works.
These three criteria offer a convincing case for musical works not being identical with the abstract sound structures we associate with them.Does this then entail that musical works are not abstract entities?These criteria certainly help our argument, but they don't offer an entirely convincing case.We still have not pinpointed exactly what kind of thing a musical work is.Moreover, we must still contend with the arguments coming from the anti-realist camp.It seems that even after we differentiate musical works from abstract sound structures some will still hold that musical works are abstract entities.Others, like the compositional nihilist, will entirely reject musical works as a part of our ontology entirely.
The compositional nihilist holds essentially that there are no things as we know them, only the conctreta that compose the entities we interact with on a daily basis.This would entail that the things we interact with in our daily lives, objects like tables, chairs, and so forth, are not real objects that exist in their own right.This view goes 2155-4838 | commons.pacificu.edu/rescogitansagainst our intuitive, common sense approach to everyday objects.What do we actually gain from writing away the everyday world, besides perhaps a more parsimonious ontology?Contra the compositional nihilist consider the view that constituted objects have fundamentally "different essential properties from their lower-level constituters.
[For example], my socks and the pieces of cloth that constitute them have different persistence conditions: The piece of cloth could survive being cut into a flat piece; my socks could not."(Baker 318).The properties of some higher level objects are not possessed by their constituents.This accords better with common sense.
Let us examine the case of a statue.What is a statue?We cannot equate it with the lump of clay that makes it up due to the fact that the clay had existed long before the sculpture was created.An appealing move would be to state that the statue is not identical with the lump of clay, but rather that the two bear the very intimate relation of constitution.This view is quickly dismissed by Cameron because it seems to allow "two wholly distinct objects [to] occupy exactly the same region of spacetime" (297-298).He argues that we instead approach the statue from the viewpoint of a compositional nihilist, suggesting that there is actually no statue.Instead, we have entities arranged in a statue shape that provide a truthmaker for the English sentence 'there are statues' (299).On his view there are only simples.However, we find that the statue will have attributes that its constituents do not.A statue can be admired, it may have an off white color, and so on.An atom will not have any of these properties.The circumstances that are statue-favorable are intentional; a sculptor intends for a statue to come into existence from what was lump of clay.The clay constitutes the statue and has different persistence conditions from what it composes (Baker 318).
Dismissing statues as not truly real goes against many of our intuitions toward statues.Stating that something isn't ontologically real, but that there are truthmakers that allow us to say that it is real in everyday English is a roundabout way of saying that a given thing is 'real enough' for us to talk about.If something is 'real enough' for us to not only talk about, but to base our nomological claims around, should we say that it isn't real?Do we want to sacrifice common sense for parsimony?Cameron's argument hinges on the semantics we use when we approach ontology.His approach seems excessive.
Cameron uses the analogy of the statue, among other examples, to show that there are no musical works --entirely rejecting them as a part of our ontology.To Cameron a musical work is neither abstract nor concrete.He holds that the truthmaker for English sentences like '"Maybellene" is a popular Chuck Berry song' exist and make the sentence true, but that we do not accept "Maybellene" as a real entity (305).In other words, Cameron is not ontologically committed to musical works like "Maybellene", all that needs to exist is a sound structure indicated by Chuck Berry that resembles "Maybellene".This is a meta-ontological position that manages to still maintain the criteria of creatability and performance, but has trouble with the principle of fine 2155-4838 | commons.pacificu.edu/rescogitansindividuation.Cameron seems to suggest that an abstract sound structure will possess the qualities the composer imparts into their musical work.Let us return to the two variations of "Blitzkrieg Bop", one being indicated by the Ramones, while the other, "Blitzkrieg Bop*", indicated by Crass.Cameron's argument suggests that the sound structure that is "Blitzkrieg Bop*" is angrier than the sound structure that is "Blitzkrieg Bop", even though we have established that these two sound structures are identical.One abstract sound structure cannot be both "Blitzkrieg Bop*" and "Blitzkrieg Bop", but by claiming the non-existence of musical works there is no musical work for either band to create.Cameron's sound structures can take on different attributes without being instantiated as musical works, which, when we recall what a sound structure is, seems implausible.
The purpose of Cameron's argument that musical works are not real entities is to show that the following inconsistent triad holds.(1) Musical works are created.(2) Musical works are abstract objects.(3) Abstract objects cannot be created.We have already discussed why (1) holds, creatability is an integral element of musical works.We also know (3) to be true intuitively; attempt to create an abstract object like a number --you can't.Cameron suggests that ( 2) is also self-evident, however we find that this is not the case.Once we show (2) to be unnecessary we are left wondering why we actually need an approach like Cameron's.He shows that his approach allows for this triad to hold.We see that (1) and (3) hold regardless of the truth of (2), and we can show (2) to be false --so then what is the need for Cameron's roundabout meta-ontology?
Once we allow for musical works as real entities in our ontology we must decide whether they are concrete, or if they number among the abstracta.Cameron assumes that they are abstract entities.However, we quickly find that the Platonist approach is unsatisfactory when considering musical works.A major issue lies with that of creation, if musical works are abstract objects then when a composer creates a work they are actually discovering an already existing work (Predelli 2).The musical Platonist must violate the creatability criteria, which as I've shown is on good footing.This approach also seems to be equating abstract sound structures with musical works; all the arguments that discredit abstract sound structures being the same as musical works apply to musical works as abstract entities.Recall fine individuation and the necessity of performance means.Moreover, we are still met with an ontological problem that is a major issue of Platonist theories.This problem being that these abstract entities must exist somewhere, complicating our ontology.
Once we reject a Platonist approach toward musical works we must accept that these entities have concrete manifestations in our world, the nominalist view.In what way could music manifest itself in the everyday world?The answer is simple and intuitive, clearly music is manifested by performance.This explanation coincides best with what we already know about sounds and works of music.We have already established that 2155-4838 | commons.pacificu.edu/rescogitanssounds are constituent properties of events.It follows that sound structures are merely a structured sequence of sound producing events that, when indicated by a composer in some musico-historical context for some specific sound producing bodies, can become a musical work.It is not necessary to give musical works any kind of special priority in our ontology because they exist as a constituent property of an event whose goal is to produce a sound structure resembling some kind of indicated musical work.A musical work is identical with the sound structure that emerges from a correct performance of the musical work.Some may be tempted to claim that this doesn't account for recordings of musical works, or that a recording creates a separation between the performance and the actual musical work (Leddington 17).We can dismiss such an objection rather easily, a recording is a reproduction of the sounds produced during a performance, similar to how a photo reproduces the colors of a landscape (18).A much more pressing objection is that of inheritance.
The nominalist approach to musical works struggles in responding to the inheritance problem.If the nominalist position is correct then a musical work will "inherit certain problematic features from their concrete manifestations" (Tillman and Spencer 252).If every performance of "Raw Power" has a wrong note, does it follow that "Raw Power" has incorrect notes?Clearly this is not the case, Iggy has some idea of what he intends "Raw Power" to sound like.Does this mean that a nominalist or materialist approach toward musical works is incorrect?One response would be that a whole does not does not necessarily have all the features of its parts.For example, while every part of a cat may weigh less than nine pounds the whole does not weigh less than nine pounds (252).One could further argue that musical works enjoy priority over their parts, that musical works could have "certain strongly emergent properties, like having a nature that is incompatible with including wrong notes, that is not shared by their parts" (256).This could be justified by stating that an artist's intention comes prior to the actual work being performed (257).
Moreover, if a performance of "Raw Power" has some wrong notes this does not mean that it is identical with the "Raw Power" indicated by Iggy Pop."Since the predicates shared by work and performance denote correct properties that a correct performance of a work must have, it is clearly possible that there be a performance of it which lacks those properties" (Predelli 2).If works are concrete then they can have different features than those in their performances.Recall that composition is not equal to identity.Composition is a relation that unites items at various metaphysical levels into wholes that are not always reducible to the simples that comprise them; a credit card has properties above and beyond the plastic that comprises it (Baker 318).In the same vein, we cannot necessarily say that if an indicated sound structure, like "Raw Power", has the property of having wrong notes in one performance that "Raw Power" has wrong notes across all performances.
2155-4838 | commons.pacificu.edu/rescogitansAnother, perhaps better, move could be, instead of rejecting inheritance as a problem of musical nominalism, to accept it as a feature.If every performance of "Raw Power" has a wrong note, and Iggy did not write his work down, what actually makes this a wrong note?The fact that Iggy did not intend for that note to be there?It may seem like an unnecessary move to place the ideal musical work into the artist's head and to suggest that all instances of the work are incorrect.If all instances of a work have a 'wrong' note then perhaps that work simply contains a different note that may or may not have been intended by the artist.We need more leniency in what we accept as a correct performance of a musical work.Accepting this also accounts better for improvisation in works.When we consider a work like John Cage's "4'33"" there are no 'correct' notes, only whatever sounds happen to occur without Cage's direct influence.In either approach, we find that the inheritance problem does not entirely discredit a nominalist view of musical works.
The ontological status of a musical work is debated by metaphysicians, I have shown that many of these views prove to be unsatisfactory in one regard or another.The most tenable view is that of nominalism, which accounts for the Event-Property nature of sounds while still doing justice to the criteria we ascribe to works of music.However, we cannot leave the definition of music up to the metaphysicians entirely.I have addressed what a musical work could be from an ontological standpoint, but this may not give a very satisfactory answer to the aesthetician.Perhaps its best we leave answering the rest of the question, 'what is music?', to them.