1. Antiquity
Lippman (1992: chapter 1) observes that Greek Antiquity is responsiblefor two ideas that shaped Western musical thought until at least theseventeenth century. The first amounts to a metaphysicalclaim. While details vary greatly, this is the claim that the laws andstructure of music correspond to those of the universe. This view isat times supplemented with the additional thought that celestialbodies themselves produce music as they follow their orbits. ThePythagorean concept of cosmic harmony, or harmony of the spheres, isthe first known example of the metaphysical conception of music. Thesecond idea is an ethical view, sometimes referred to asethos theory. According to this view, music has the capacity to altermood and mold personality, and thus may contribute to moral education(see Lippman 1964: chapter 2).
Some authors, especially those of Pythagorean persuasions, explicitlylink the ethical view to the metaphysical one: if the whole universeis musical, so is the human soul, hence music’s power over it.In other authors, such as Aristotle, the ethical and educational valueof music is construed as depending mainly on its capacity to imitateor arouse emotions.
1.1 The Universe as Harmony: Pythagoras and the Pythagoreans
No writings survive that are directly attributable to Pythagoras (c.570–c. 490 BCE). His view of music must therefore bereconstructed from the works of other Pythagoreans, or from more orless sympathetic accounts provided by other authors (a usefulcollection of texts is found in Barker 1990: chapter 1). Central toPythagorean philosophy is the notion of harmony as the unity ofvarious parts. In this sense, the concept has a much broaderapplication than to music, for harmony regulates the motion ofcelestial bodies as well as the human soul (on the concept of harmonyin ancient Greek philosophical thought, see Lippman 1964: chapter 1).In the Pythagorean tradition, mathematics is the discipline thatinvestigates harmony. Musical harmony is no exception to this, as theinterval of the octave may be expressed as a numerical ratio (2:1).The octave may in turn be considered as the sum of a fifth (3:2) and afourth (4:3). Pythagoras was generally credited with the discovery ofthe relations between these intervals. These are best illustratedusing a monochord, an instrument constituted by a single string, thevibrating sections of which are determined by the position of amovable bridge. However, it is unclear whether the monochord was infact used by Pythagoras (Barker 2014: 186), and it is also plausibleto assume that instrument makers had some understanding of theserelations before they became the object of intellectualspeculation.
The series of numbers involved in these ratios (1 to 4, thus adding upto 10) was considered sacred by the Pythagoreans, who named ittetraktys. This metaphysical view of music as a reflection ofcosmic order proved to be enormously influential in Western musicalthought. As the harmony found in music is essentially the sameprinciple as the one that regulates both the cosmos and the humansoul, at least some Pythagoreans thought that their view could explainthe effects of music on the listener’s mood and character.
The most notable opponent of the Pythagorean notion of celestialharmony was Aristotle. He offers two main reasons to reject this idea.First, he argues that physical bodies could not be composed ofnumbers, for the latter have no weight, whereas the former do(Metaphysics, 14.3). Second, he observes that loud soundsshatter glass and produce a physical effect upon us. The movement ofplanets in space, Aristotle argues, would produce an even greatereffect on us, the absence of which is proof that celestial bodiesproduce no sound (On the Heavens, 2.9). In addition torejecting the idea of a celestial harmony, Aristotle also rejects therelated notion that the human soul is a kind of harmony (On theSoul, 1.4).
1.2 Mathematical and Empirical Harmonics
‘Harmonics’ is the Ancient Greek discipline that studiedmusical entities, from intervals to scalar systems (Barker 1990: 3).The Pythagorean tradition marked the start of an influential approachto harmonics, focused on the definition of musical entities inmathematical terms. Early contributors to this approach were Philolausof Croton (c. 470–c. 385 BCE) and especially Archytas ofTarentum, a contemporary of Plato (Huffman 1993, 2005).
A mathematical treatment of harmonics informed by Pythagoreanprinciples is also given in the Sectio Canonis, a treatise ofdebated authorship, but often attributed to Euclid. Nicomachus (fl. c.100 CE) wrote an influential introduction to harmonics from thePythagorean standpoint (source texts with commentary are collected inBarker 1990; Barker 2007 reconstructs the development of Pythagoreanharmonics.)
Parallel to the Pythagorean tradition, an alternative approach soughtto ground the laws of music in the way it appears to a competentlistener, rather than in explanations from other domains of knowledge.This tradition attempted to provide descriptions of music inautonomously musical terms, and valued accordance with musicalpractice. This empirical approach to harmonics is in contrastto the mathematical approach adopted by the Pythagoreanschool (on the differences between the two schools, see Barker 2007and Barker 1990: 3–8). The contrast between these two schools isof philosophical interest, insofar as it concerns the appropriatelevel of explanation for aesthetic phenomena. Additionally,mathematical harmonics often were taken to support the metaphysicalworldview associated with the Pythagorean doctrine, whereas empiricalharmonics tended to follow from an Aristotelian framework (see Barker2007: chapter 4; Mathiesen 1999: 303).
By far the most important figure in the empirical tradition is that ofAristoxenus (see§1.4),although he relied on the work of previous theorists (Barker 2007:chapter 2). The mathematical and empirical traditions are oftenreferred to as Pythagorean and Aristoxenian, respectively.
Theophrastus (c. 371–287 BCE), who followed Aristotle as thehead of the Lyceum, is critical of the Pythagorean approach. Heobserves that we may find quantitative regularities in things otherthan musical intervals, yet we still consider these things asdifferent from musical entities. This, he argues, must be due to thefact that intervals and melodies are essentially something other thannumbers (Barker 1990: 111–114; Lippman 1964: 157–160;Barker 2007: chapter 15).
It is important to stress that the contrast between empirical andmathematical approaches was already evident to ancient theorists andcommentators, who often passionately took sides. Ptolemais of Cyrene(fl. between the 3rd century BCE and the 1st century CE), the onlyknown Greek woman music theorist, distinguishes betweenmousikoi and kanonikoi, a distinction that matchesthe one between Aristoxenians and Pythagoreans (Barker 2014: 185).
Some theorists developed hybrid views. Ptolemy (c. 100—c. 170CE) is a case in point, as he explicitly faults both Pythagorean andAristoxenian approaches (Harmonics, book I, 2). Hisphilosophical outlook is clearly Pythagorean. For instance, he defendsat length the notion of cosmic harmony, tracing parallels betweenmusical and cosmological entities (Harmonics, book III, 9).However, he attempts to develop mathematical constructs that arebetter adherent to musical practice than the ones proposed byPythagoreans, thus addressing a concern characteristic of theAristoxenian approach.
Aristides Quintilianus (3rd–4th century CE), an author knownuniquely for his treatise De Musica, also attempts a fusionof Aristoxenian and Pythagorean elements. While his view of harmonicsis indebted to Aristoxenus, his philosophy of music bears a strongPythagorean influence. Book III of his De Musica illustratesparallels between musical elements and the most disparate parts ofreality, from the senses, to virtues, all the way to celestial bodies.Aristides Quintilianus defends the ethical value of music, and indoing so he explicitly refers to Plato’s and Aristotle’sviews on the issue. He argues that music imitates moral actions in away that is superior to that of the other arts, because it achievesmimesis in more than one way, that is, through words andmelodic movement (De Musica, book II, chapter 4). AristidesQuintilianus also classifies musical elements, from modes toinstruments, as masculine, feminine, or a mixture of the two (modesare scales that divide up the octave in different sequences of tonesand semitones; though the names of Greek modes are identical to thelater Church modes, they name different scales). This classificationis supposed to help in determining the ethos of musical pieces andinstruments, feminine elements being associated with smoothness, andmasculine ones with roughness (De Musica, book II, chapter12).
1.3 Music, Emotions, and Society: Plato and Aristotle
A central concern in ancient Greek philosophy of music is theconnection between music’s capacity to imitate emotions and itssocial value. Plato (427/28–347 BCE) and Aristotle(384–322 BCE) both develop an account of this relation, and areparticularly interested in the pedagogical consequences of theirtheories.
When they talk about ‘music’ (mousikē),Plato and his contemporaries often have in mind a broader set of artforms than the ones we associate with music today. In this broadsense, mousikē denoted “a seamless complex ofinstrumental music, poetic word, and co-ordinated physicalmovement” (Murray & Wilson 2004: 7). Plato sometimes alsouses the term in a narrower sense to indicate melodic and rhythmicelements that we would consider today as properly musical (for a briefdiscussion of the broad and narrow sense of mousikē inPlato, see Schofield 2010: 230–231).
Plato believes that music may contribute to the education of theyouth, and more generally to the correct functioning of society, butalso holds that it may pose dangers. Underlying Plato’s concernsabout the musical education of citizens is the wider assumption thatchanges in musical taste must be avoided or at least closelyscrutinized, as they will produce changes in society(Republic, 4.424b–d, Laws 2.660a-b). In theLaws, Plato goes so far as to describe the changes in musicaltaste that followed the Persian wars as the trigger for subsequentrejection of authority and societal unrest (Laws,700a-701b).
According to Plato, the sensible world is an imperfect copy of perfectand immutable ideas. Plato’s general view of the arts, includingmusic, is that they are an imitation (mimesis) of objectsfound in the sensible world. This leads to his famous condemnation ofart in Book X of his Republic. For, if our world is already acopy, then the arts provide us with copies of a copy. Thus,Plato’s metaphysics motivates his dismissive view of art’sepistemic value.
However, Plato also holds that the musical imitation of some humanemotions may be ethically beneficial, especially at the stage whenchildren are too young to be responsive to ethical education thatrelies on a discursive and rational basis. Plato first describes theemotions in question indirectly, one as the emotion of a person who issteadfast and resolute in misfortune or while fighting, the other asthe state of “someone engaged in a peaceful, unforced, voluntaryaction” (Republic, 3.399b). Plato holds that theseemotions, which he then refers to explicitly as courage andmoderation, are best imitated through the Dorian and Phrygian modes,which
imitate the violent or voluntary tones of voice of those who aremoderate and courageous, whether in good fortune or in bad.(Republic, 3.399c)
Thus, Plato construes musical imitation of emotions as grounded on theresemblance of music to human expressive behavior, particularly vocal.At least on one occasion, he also seems to hold that music imitatesthe bodily movements associated with emotions (Laws,2.654e-655a). This could be considered the first defense of the ideathat musical expressiveness is due to the music’s resemblance tohuman expressive behavior, a position that remains popular to this day(see the entry on href="../hist-westphilmusicsince-1800/#AnalPhilMusi" history ofwestern philosophy of music: since 1800, section 2.6).
How does the musical imitation of courage and moderation achieve itseducational purpose? At least some passages suggest that Plato thoughtmusic to be capable of influencing human character in virtue of thecommon harmonic nature shared by the soul, music, and the universe.For instance, in the Republic, Socrates tells Glauconthat
rhythm and harmony permeate the inner part of the soul more thananything else, affecting it most strongly and bringing it grace, sothat if someone is properly educated in music and poetry, it makes himgraceful, but if not, then the opposite. (Republic, 3.401d;see also Republic, 3.412a, Timaeus, 47d-e, andLaches, 188d)
Young (forthcoming) observes that Plato seems elsewhere to reject thePythagorean view of the soul as harmony (Phaedo,89–95). On this basis, Young holds that Plato’s view ofthe ethical benefits of music is best interpreted as simply relying onthe music’s imitative capacity. Regardless of what one makes ofPlato’s view of the soul as harmony, it is clear that he wasinfluenced by the Pythagorean view of the universe as governed byharmony, especially in the Timaeus.
On the surface, Aristotle’s view may appear close toPlato’s. Both philosophers hold music’s value to reside inits imitation of emotions. However, Aristotle’s view isconsiderably more nuanced.
In the Poetics (1447a), Aristotle makes clear that all thearts are imitative, although they differ in the means, object, andmanner of imitation. Aristotle has a more positive view of imitationthan Plato, observing that humans are prone to it since childhood, andgranting that we may learn through imitation. This cognitive value ofimitations also explains why we feel pleasure when engaging with them,even when they imitate things we would find ugly or repulsive in reallife (Poetics, 1448b 5–20).
Discussing the object of musical imitations, Aristotle introduces aninteresting distinction between mere indications or signs(sēmeia) of emotional states and the states themselves.Signs of emotional states are the observable behavior that accompaniesthe occurrence of such states. Aristotle holds that the visual artsmerely imitate signs of emotions. For instance, a painter mayrepresent a man weeping or smiling. However, music is able to imitatethe emotional states themselves. A plausible reading ofAristotle’s view is that he regarded music as capable ofarousing the emotions it imitates, and that this capacity ultimatelyexplains how music may imitate emotional states themselves, as opposedto their manifestations. On this analysis, the object of musicalimitation cannot be specified separately from the emotional responsethe music arouses in the listener (Halliwell (2002: 248) stresses thispoint; Sörbom 1994 offers an alternative interpretation).
To understand the relation between Aristotle’s view of musicalimitation and its educational role, it is useful to examine thispassage:
Besides, when men hear imitations, even apart from the rhythms andtunes themselves, their feelings move in sympathy. Since then music isa pleasure, and excellence consists in rejoicing and loving and hatingrightly, there is clearly nothing which we are so much concerned toacquire and to cultivate as the power of forming right judgments, andof taking delight in good dispositions and noble actions.(Politics, 1340a)
Here Aristotle seems to be arguing the following: music imitatescharacter, that is, various dispositions and emotional states; themusical imitation of such states generates pleasure in the listener;so, if we restrict musical imitations to those of morally praiseworthystates, listeners will be pleased at such states in real life.Aristotle is relying on the assumption that the pleasure accompanyingour engagement with an imitation will extend to the experience of theimitated object in real life. As he expresses it:
The habit of feeling pleasure or pain at mere representations is notfar removed from the same feeling about realities. (Politics,1340a)
Therefore, the moral value of musical education depends on thecapacity of mimetic representations to induce pleasure in those whoappreciate them. While this capacity may be shared by both music andthe visual arts, it is only music, according to Aristotle, that isable represent states of character, as opposed to merely representingtheir manifestations.
An additional function of music, according to Aristotle, is catharsis.Music may help individuals who are in states of extreme enthusiasm,pity, or fear return to a more balanced state (Politics,1342a). The details of this process are hard to reconstruct, becausethey are tied to the debated concept of catharsis, introduced byAristotle in his definition of tragedy (Poetics, 1449b). Itseems clear that Aristotle thought of the cathartic process asrequiring music expressive of emotions akin to the ones to be purged.For example, a state of uncontrolled religious frenzy could be subduedby means of frenzied music.
Plato and Aristotle come apart also with regard to the role of thepleasure we take in listening to music. Plato’s approach is apragmatic one. The aspects of music he values are all and only thosethat contribute to its ethical role. Such a principle governsPlato’s observations on musical pleasure. He does not deny thatmusic may be able to induce pleasure. However, this pleasure is onlyvaluable insofar as it attracts listeners to music that is good fortheir moral education (Laws, 2.668a-b).
Plato also notes that pleasurable responses may become associated withmusic of the morally questionable sort, as such responses depend alsoon habituation (Laws, 7.802c-d). From this he concludes thatit is of great importance to control children’s musicalenvironment.
Aristotle’s view of musical pleasure is far more positive. It isclear from the above that pleasure plays a role in Aristotle’saccount of music’s educational value. But Aristotle also thinksthat pleasure generated by music is valuable for two other reasons.First, music may provide relaxation to those that are tired from work.In this sense, the pleasure it produces is similar to that produced byeating or sleeping (Politics, 1339a-b). Second, and perhapsmore interestingly, Aristotle holds that the pleasure of music shouldbe part of a cultivated life (Politics, 1339b; for adiscussion of the functions of music in Aristotle, see Lippman 1964:130–131.)
Both Plato and Aristotle comment on the value of instrumental music.Plato’s pragmatic attitude grounds his injunction againstcomplexity and subtlety in music (Republic, 3.399c-e). Heholds that music should have a vocal part, and argues that the musicalelements should conform to the emotional state described by thepiece’s lyrical content (Republic, 3.400d). Thisconcern with clear and unequivocal musical imitation also motivatesPlato’s stance on the value of instrumental music. While it isclear that Plato holds the purely musical elements of melody andrhythm to have an imitative potential, he attributes little value toinstrumental music, presumably because he believes that it is unableto provide an unequivocal imitation of courage and moderation(Laws 2.669 d-e).
Aristotle is again more nuanced. According to him, both vocal andinstrumental music produce pleasure (Politics, 1339b). Ingeneral, despite his reservations concerning the aulos (aninstrument comprising two, linked, reed pipes) (Politics,1341a), Aristotle’s attitude toward instrumental music is farless critical than Plato’s. For example, Aristotle is clear inconceding that even purely instrumental music can arouse distinctemotions, due to both its melodic and rhythmic elements(Politics, 1340a-b).
1.4 Aristotelian Empirical Harmonics: Aristoxenus
A disciple of Aristotle, Aristoxenus (fl. 4th century BCE) argues thatharmonics should not be concerned with an explanation of musicalphenomena in mathematical terms, but rather describe musicalstructures through the analysis of musical elements as objects ofperception. For example, Aristoxenus distinguishes between thespeaking and singing voice, the former being characterized by acontinuous movement across pitch space, the latter by intervallicmovement from one pitch to another (Elementa Harmonica, bookI, 9–10). On the basis of this distinction, he provides adefinition of note (phthongos) as any pitch on which thevoice rests and that has a position in a given melodic line(Elementa Harmonica, book I, 9–10; for a detaileddiscussion of Aristoxenus as a music theorist, see Mathiesen 1999:294–344).
Another striking element in Aristoxenus’ conception is his viewof musical understanding, which he thinks involves the faculties ofperception and memory,
for we have to perceive what is coming to be and remember what hascome to be. There is no other way of following the contents of music.(Elementa Harmonica, book II, 38–39)
Note that perception is not described as the mere awareness of what ispresent to our senses at a given time, but includes an anticipation of“what is coming to be”. Here, Aristoxenus seems toanticipate an account of musical understanding similar to the onerecently defended by Jerrold Levinson (1997).
Aristoxenus’ view of the moral effects of music is harder toreconstruct. While it is clear that he thought music to be a properobject of ethical evaluation, he likely had a less dogmatic conceptionof the link between given modes and their moral value than Aristotle(see Rocconi 2012 and Barker 2007: 245 ff.).
1.5 Music and Moral Education: Skeptical Views
Some ancient philosophers reject the view that music is valuablebeyond the pleasurable experience it affords. In doing so, theyquestion the educational value of music.
In his De Musica, the epicurean Philodemus (c. 110–c.30 BCE) rejects the widespread view that music may imitate thepassions, claiming that music is no better able to representpsychological states than the art of cooking (see Wilkinson 1938).Philodemus appeals to Democritus (c. 460–c. 370 BCE), butevidence regarding the latter’s views is scant. While it isclear that Democritus held music to stand lower than other arts interms of utility, it is probably not true that he thought thisdeprived it of educational value (see Brancacci 2007:193–95).
Sextus Empiricus (fl. 3rd century CE), in Against theMusicians, also denies that music could have any positiveeducational role. While he concedes that we may enjoy music, he deniesthat the study of music could contribute to such enjoyment. In supportof this, he also compares music to cookery, arguing that in both casesenjoyment is independent of our understanding of the thing enjoyed(Against the Musicians, 24–25).
2. Late Antiquity and the Middle Ages
Despite spanning about one thousand years, the period that goes fromlate Antiquity to the end of the Middle Ages brought little new toWestern philosophy of music.
A prominent feature of this period is the emergence of the Christianworldview. Christian thinking on music adapts some of the views it hadinherited from Greek Antiquity, chief among them the Pythagoreanharmony of the spheres and the idea that music may influencecharacter.
The medieval approach to music is biased toward the point of view ofthe theoretician. Music is typically grouped together with scientificdisciplines. The sensory pleasure of the listener is rarely remarkedupon, and at times considered with suspicion. The individuality andcreativity of the composer is also neglected, and the role ofperforming musicians almost entirely ignored or dismissed (Schuller1988, and Young forthcoming (chapter 2) are extensive introductions tomedieval musical aesthetics).
2.1 Early Christian Views of Music: Augustine and Boethius
Saint Augustine of Hippo (354–430), philosopher and Father ofthe Church, devoted to music his treatise De Musica. Althoughmostly technical in character, the book is an early statement of themedieval view of music as first and foremost a science, rather than apractical occupation. Augustine defines music as “the science ofmensurating well” (scientia bene modulandi—thislast word is at times rendered as ‘modulation’), that is,the discipline concerned with the attainment of measure and proportion(De Musica, book I, 2).
An almost identical definition of music is found in Cassiodorus(485–c. 585) who states that “Music indeed is theknowledge of apt modulation” (Institutiones, 5), aswell as in Isidore of Seville’s (560–636)Etymologiarum:
Music is an art of modulation consisting of tone and song called musicby derivation from the Muses. (Etymologiarum, book III,15)
Augustine’s attitude toward the sensory pleasure derived frommusic was conflicted. On the one hand, he recognizes music’spower in worship. On the other hand, he is aware that this could alsolead the faithful to stray away from the religious content of theexperience (Confessions, II).
The philosopher Anicius Manlius Severinus Boethius (c.480–525/26), normally referred to simply as Boethius, had afundamental role in shaping medieval musical thinking. Throughcommentated translations of Greek authors and original expositions oftheir views, Boethius allowed classical Greek philosophical ideas tosurvive into the Middle Ages. He did so in the field of music bywriting De Institutione Musica. In this work, Boethiussummarizes music theory as found in the Greek treatises he knew. Thebook also distinguishes between three types of music: musicamundana, humana, and instrumentalis. Musicamundana is the harmony of the cosmos, a Pythagorean concept thatfound widespread acceptance over the course of the Middle Ages, thoughit was recast in Christian terms. Musica humana is describedonly briefly by Boethius, though it is clear that it is the harmonybetween the various elements that compose the human body. Some modernaccounts of Boethius’s thought identify musica humanawith singing, but this is incorrect, though it is true that latermedieval thinkers used the term in such a way (see Dyer 2007: 59; 64;69). Finally, musica instrumentalis is sounding music, vocaland instrumental. While this is undoubtedly regarded today as the coreof music, Boethius considers it as music only in a derivative sense.Boethius’s tripartite division is repeated and accepted bynumerous medieval writers on music.
Boethius also takes from Greek Antiquity the Platonic view that musicmay arouse emotions and influence human character. He observes thatthis sets music apart from other mathematical disciplines (DeInstitutione Musica, book I, 1).
2.2 Christian Thought and Music
The impact of Christianity on medieval musical thinking is evident inthe religious interpretation given of two philosophical views fromAntiquity, the harmony of the spheres and the ethos theory, accordingto which music may shape the listener’s character. Both of theseideas were held by Boethius (see the preceding section) and foundwidespread acceptance among medieval writers.
The Christian take on the harmony of the spheres consists in the claimthat such harmony was imposed on the universe by the Creator. Thebeauty of music, including sounding music, is thus traced back tomusic’s partaking in that divine harmony. John Scottus Eriugena(c. 800–c. 877) expresses such a view in hisPeriphyseon, as evident from his very definition ofmusic:
Music is the art which by the light of reason studies the harmony ofall things that are in motion that is knowable by natural proportions.(Periphyseon, 475B)
That the harmony of the spheres could be interpreted in Christianterms did not prevent philosophers from rejecting it, especially onceAristotle’s works had become available. Albert the Great(Albertus Magnus) (c. 1200–1280) is an example of a prominentphilosopher who rejects the idea of a celestial harmony. Appealing toarguments based on the structure of the cosmos, he discards the ideathat the movement of celestial bodies could produce harmonious sounds(Liber de Causis Proprietatum Elementorum, book II, 2.1; onthe harmony of the spheres in the thirteenth century, see Mews2011).
The idea that music may influence character was widely accepted overthe course of the Middle Ages. It was supported by the exaggeratedaccounts of music’s powers that had survived from Antiquity. Themedieval Christian version of the ethos theory normally stresses thedevotional potential of music.
2.3 Music among the Liberal Arts
Following Martianus Capella’s (fl. c. 410) The Marriage ofPhilology and Mercury, medieval thinkers classified music as oneof the liberal arts. These were divided into the arts of thetrivium (grammar, logic and rhetoric) and thequadrivium (arithmetic, geometry, astronomy and music). Thearts of the trivium were referred to as artessermocinales, because they are related to discourse and language.The arts of the quadrivium were instead the artesreales, studying the mathematical and physical constitution ofthe universe. The classification of music as one of the artesreales is unsurprising in light of the Pythagorean-Boethian viewof music as primarily concerned with proportions and harmony.
While this classification remained mostly unchallenged, there weresome important dissenting voices (for an overview, see Dyer 2007).These disputes regarding the proper classification of music areimportant because they reflect underlying concerns with music’snature and purposes. From the thirteenth century onwards, theoristsand philosophers became more concerned with the practical and acousticsides of music-making, and less with its theological implications. Forinstance, the English grammarian John of Garland (1195–1272)distinguishes between plainsong, mensural music, and instrumentalmusic, and does so on the bases of historical and musicologicaldifferences between such repertoires (see Fubini 1991: 94).
The progressive assimilation of Aristotle’s work in medievalthought meant that the physical, sounding aspects of music began to beconsidered as important as its mathematical ones. Evidence of this isThomas Aquinas’s classification of music as one of thescientie medie, that is, disciplines that apply mathematicsto the study of natural objects (see Dyer 2007: 67–68).
2.4 Islamic Philosophy of Music: Al-Kindi, al-Farabi, and Ibn Sina
As observed by Young (forthcoming: chapter 2), it is appropriate toinclude medieval Islamic philosophy in a survey of Western musicalthinking, as Islamic philosophers relied largely on the same Greekauthors that were central to the European tradition. In fact,translations of Greek works were more widely available in the Islamicworld. Partly because of their earlier contact with Aristotle’sphilosophy, Islamic philosophers are more inclined to value thesensory character of music and to accept it as an object of aestheticappreciation.
The philosopher al-Kindi (c. 800–870) holds a broadlyPythagorean view of music (see Shehadi 1995: chapter 1, and Adamson2007: 172–180). According to him, a relationship of affinityholds between musical elements and parts of the cosmos. This affinityallows Kindi to trace correspondences between musical elements and theextra-musical world. Kindi’s acceptance of Pythagorean views isalso exemplified by his claim that music exercises a power of thehuman body in virtue of the affinity between its constitution and thatof various musical instruments.
Al-Farabi (872–950) is a notable representative of theAristotelian current in medieval Islamic philosophy of music (seeShehadi 1995: chapter 3). Farabi rejects the harmony of the spheres.Rather than speculating about music’s origin in the structure ofthe cosmos, he attempts to understand it as a human phenomenon,suggesting that vocal music may have developed out of necessities suchas the expression of pleasant and unpleasant states. Farabi alsodistinguishes three types of melody, according to the effect itproduces in the listener. Some melodies are pleasing to the ear,others evoke images, and yet others express psychological states. Itis possible for a melody to do more than one of these things, and infact Farabi holds that the best melodies tend to have all three sucheffects.
Ibn Sina, known in the West as Avicenna (980–1037) describesmusic as a mathematical science, but argues that a theory of musicshould be primarily concerned with the explanation of music as heard.This emphasis on the appropriate level of explanation is reminiscentof Aristoxenus (see§1.4),who exercised a clear influence on Islamic philosophy of music. IbnSina also revives the Aristotelian view that music may imitate humancharacter and psychological states (on Ibn Sina’s philosophy ofmusic, see Shehadi 1995: chapter 4).
3. Early Modern Period
3.1 Music and Sensory Pleasure: Tinctoris and Zarlino
From the fifteenth century, writers on music begin to elaborate on theaesthetic aspects of musical experience. The theorist and composerJohannes Tinctoris (c. 1435–1511) exemplifies various aspects ofthis new outlook on music. He defines harmony and consonance in termsof their appearance to the listener, rather than through mathematicalnotions. Harmony is thus described as the pleasing effect produced bysounds, whereas consonance and dissonance are respectivelycharacterized as combinations of sounds that bring sweetness to theear or hurt it (Liber de arte contrapuncti, chapter 2). Inhis Complexus effectuum musices, Tinctoris lists prominenteffects of music, most of which involve emotional arousal, while thededication of his Proportionale Musices refers back toAntiquity and the alleged emotional potency of its music.
Tinctoris thus prefigures two tendencies typical of early modernthinking about music: first, the focus on the subjective experience ofmusic, and on the feelings of pleasure and displeasure that accompanyit; second, the value attributed to music’s expressive power,and the related idea that ancient music’s capacity to arouseemotions was greater than that of the music of their time.
The first of these tendencies is also embodied in the work of GioseffoZarlino (1517–1590). A towering figure in Western music theory,Zarlino is normally considered as the theorist who initiated the moveaway from medieval modality and toward tonal harmony. According toZarlino, rules of composition are good insofar as they producebeautiful music, whose purpose is to “improve” and“delight” (Zarlino 1558: book III, 71). However, Zarlinodoes not reject the role of rationality, and especially ofmathematics, because its laws ultimately underlie the senses of sightand hearing (Zarlino 1588: 34–36).
3.2 Melody and Expression: The Florentine Camerata
In Book II of his Dodecachordon (1547), the humanist HeinrichGlarean (also known as Glareanus) (1488–1563) argues thatcomposers of polyphonic music are inferior to those who set a singlemelodic line to a text. He speculates that the latter was the wayGreeks, Romans, and early Christian communities composed music. Monodyis superior on the grounds that it requires an exercise in invention,whereas the polyphonist merely borrows a melodic line and then carriesout a task that is predominantly intellectual. According to Glareanus,monody has a natural character and an expressive immediacy that islost in subsequent development. Less than half a century later, boththe reference to a classical model and the vindication of monodybecame crucially important in Italian musical thinking and practice.Toward the end of the sixteenth century, a group of Florentineintellectuals and musicians gathered around the figure of Giovannide’ Bardi, a nobleman and amateur musician. This group isnormally referred to as the Florentine Camerata, or Camerata de’Bardi. Its goal was to recover the expressive potential of music, asdescribed in ancient Greek and Roman sources. Music’s power toarouse emotions was thought to have been lost during the Middle Ages,as polyphony developed. The envisaged solution was a return to monody,and more precisely to a melodic line that would follow and emphasizethe prosodic features of impassioned speech (Palisca 2006: chapter 7;Palisca 1989 is an influential collection of source readings). Thus,the main goal of the Camerata seems to have been expression, to beintended here as the arousal of emotion, while the imitation of speechprosody was considered a means to that end. This is philosophicallysignificant in that it indicates a revival of the classical Greek ideathat music is expressive in virtue of its resemblance to the way humanbeings express their emotions (see§1.3).
This resulted in the birth of the so-called stilerappresentativo, the vocal style characteristic of early opera,of which Claudio Monteverdi’s (1567–1643)L’Orfeo (1607) is considered the first masterpiece.Giulio Caccini (1550–1618) and Jacopo Peri (1561–1633)were active members of the Camerata, and have both left explicitdiscussion of the theories that informed their musical style (relevanttextual excerpts are collected in Strunk 1950: 370–392).
While Monteverdi may be considered the most representative musicianassociated with this turn to melody and expression, theCamerata’s most significant theoretical figure was VincenzoGalilei (c. 1520–1591), an accomplished composer and lutenist,father of the astronomer Galileo. In his Dialogue of Ancient andModern Music (1581), Galilei takes issue with contemporarypractices, particularly with counterpoint. He concedes that thepolyphonic music of his time is a source of aural pleasure, but faultsit for its incapacity to move the listener,
[f]or its sole aim is to delight the ear, while that of ancient musicis to induce in another the same passion that one feels oneself.(Galilei 1581 [1950: 317])
Both Monteverdi and Galilei faced the reaction of more conservativefigures. Giovanni Maria Artusi (c. 1540–1613), inL’Artusi or Of the Imperfections of Modern Music(1600), attacks the work of an unnamed composer, though it is apparentthat he has in mind composers such as Monteverdi and Carlo Gesualdo(1566–1613). Artusi concedes that the newly developed monodicstyle is expressive, but denies that expression is worth the sacrificeof compositional rules that are dictated by reason. The disputebetween Artusi and Monteverdi crystallized the clash between theso-called prima and seconda prattica, the formerindicating the old polyphonic style, the latter the new preference foraccompanied monody.
Galilei faced criticism from his former teacher Zarlino. In hisSopplimenti musicali (1588), Zarlino argues that theimitation of impassioned speech belongs to the art of rhetoric or todeclaimed poetry, not music. Musicians should confine themselves tothe imitation of the text’s content, rather than to its possibledelivery, and they should do so through the choice of appropriateharmonies (Zarlino 1588: 316–318). Zarlino’s reservationsabout expressive melody are of particular importance, as they may beseen as the first manifestation of a recurring contrast between twoopposite tendencies, epitomized by the clash between Rousseau andRameau (see§3.5).Whereas one strand locates music’s chief source of expressionin melody, the other defends the primacy of harmony, often downplayingthe role of possible resemblances between melodic contour and humanexpressive behavior (Young, forthcoming, names the two tendenciesempiricism and rationalism, respectively, followingtheir development and occasional overlap).
Despite their disagreements, both approaches valued expression inmusic set to a text. A full vindication of the expressive power ofinstrumental music was not among the programmatic goals of either theCamerata or of those who, like Zarlino, stress the importance ofharmonic elements. Caccini is critical of music that obscures thetext, because
such music and musicians gave no other delight than what harmony couldgive the ear, for, unless the words were understood, they could notmove the understanding. (Caccini 1600 [1950: 378])
Zarlino, following Plato, claims that melody and harmony should besuitable to the text, and cashes out this suitability in terms ofsimilarity:
we must also make a choice of a harmony and a rhythm similar to thenature of the matters contained in the speech in order that from thecombination of these things, put together with proportion, may resulta melody suited to the purpose. (Zarlino 1558: book IV, 32 [1950:256])
Moreover, these authors unanimously hold that it is the music that hasto be adapted to a text, rather than the other way around. In theforeword to Monteverdi’s Scherzi Musicali (1607), hisbrother Giulio Cesare claims that Claudio’s goal was “tomake the words the mistress of the harmony and not the servant”(Monteverdi 1607 [1950: 406]).
While this stress on the priority of the text was the product of arenewed interest in emotional expression in vocal music, Churchauthorities also became concerned with the confused polyphonic stylesand the way they obfuscated liturgical content, and thus they alsopromoted textural clarity in polyphonic music (see Fellerer 1953).
3.3 Sense and Rationality: Mersenne, Descartes, Leibniz
While it is important to note that the early modern period mitigatedthe medieval insistence on a mathematical treatment of music with anapproach that paid greater attention to its sensory appearance, itwould be incorrect to say that the search for mathematicalregularities was altogether abandoned. On the one hand, the buddingscientific outlook favored the extension of the experimental method tothe realm of sound, an operation that obviously included mathematicalmodeling. Thus, rather than giving up on mathematical descriptions ofmusic altogether, early modern theorists abandoned the abstractframework in which these descriptions had been formulated. On theother hand, accounts of music in mathematical terms continued tosurvive in the metaphysical speculations of music theorists andphilosophers.
Exemplary of these tendencies is the French mathematician MarinMersenne (1588–1648), who was interested in a number ofquestions connected to the mathematical nature of music. While heunderstood the importance of empirical verification, Mersenne was alsoconvinced of a fundamental harmony between the mathematical order ofmusic and that of the whole cosmos, as evident from the title of hismost important work, Harmonie Universelle (1636).
A particularly interesting theme in this sprawling treatise is thecontrast between the rational structure of music and its sensuousappearance, best illustrated through Mersenne’s discussion ofconsonance. Having defined consonance in terms of coincidence in airvibrations, Mersenne had to explain why consonance and sensoryagreeableness do not always go hand in hand. For instance, the majorthird is more agreeable than the fourth, although this is moreconsonant. Mersenne had an interesting correspondence with Descarteson the matter, in which the former manifests his wish to provide adefinition of agreeableness that is as objective as that ofconsonance, while the latter expresses skepticism about the project,and claims that agreeableness is dependent on variation in taste, aswell as in musical context (see Jorgensen 2012).
Early in his life, René Descartes (1596–1650) devoted atreatise to music, the Compendium Musicae (written in 1618and published posthumously in 1650). This work is dominated byscientific concerns rather than philosophical ones, but the very startof the treatise is characteristic of its age, as Descartes states thatmusic’s goal is “to please and to arouse variousAffections in us” (Descartes 1650). But Descartes’s mostimportant impact on philosophy of music is actually due to histreatise on emotions, The Passions of the Soul (1649). Inthis work, Descartes defines emotions as the result of the actionproduced by external objects on the animal spirits, a thin air-likesubstance that stirs our passions whenever it is set in motion.Descartes also distinguishes six basic emotions: wonder, love, hatred,desire, joy, and sadness.
This taxonomy of emotions and the mechanistic description of theirfunctioning will prove important to the so-calledAffektenlehre, or doctrine of the affections, the eighteenthcentury view that a piece of music should arouse a specific emotion inthe listener. Affektenlehre theorists believed that this goalcould be achieved through the use of specific musical devices, eachassociated with a given emotion. The German composer Johann Matthesonmay be considered the most prominent representative of this tradition.He explicitly endorses Descartes’ theory of emotions in histreatise Der vollkommene Capellmeister (Mattheson 1739: I,iii, §51).
Gottfried Wilhelm Leibniz (1646–1716), philosopher andmathematician, devoted to music only scattered observations. Yet, hisview of music’s value is of particular interest, as it isindicative of the attempt to reconcile rational and sensuous elementsin our appreciation of music (see Fubini 1991: 149). Leibniz acceptsthat the main goal of music is to move us. In fact, he notes that,while music is obviously so apt to do so, “this main purpose isnot usually sufficiently noticed or sought after” (Leibniz,On Wisdom). However, Leibniz also specifies that thissensuous pleasure is the result of the automatic and unconsciousapprehension through music of the rational structure of the universe.He writes:
Music charms us, although its beauty consists only in the agreement ofnumbers and in the counting, which we do not perceive but which thesoul nevertheless continues to carry out, of the beats or vibrationsof sounding bodies which coincide at certain intervals. (Leibniz 1714:614)
3.4 Imitation and Expression in the Eighteenth Century
Two related questions may be asked with regard to imitation in music.The first is whether music may imitate extra-musical objects, and whatvalue this imitation may have. The second concerns the role of suchimitation in the musical expression of emotions. Over the course ofthe eighteenth century, answers to both questions were characterizedby a growing skepticism toward imitation (see Lippman 1992: chapter6). The gradual rise of instrumental music was certainly partlyresponsible for this change in attitude. Its increasing formal,melodic, and harmonic complexity made it less plausible to argue thatit imitated the speaking voice, and yet this complexity onlyheightened music’s power to express and arouse the emotions.
A natural way out of this impasse is to explicitly define expressionin terms of emotional arousal, distinguishing it from the mere musicaldepiction of worldly objects, including the external manifestations ofhuman emotions. This is the option taken by Charles Avison(1709–1770) in his Essay on Musical Expression, firstpublished in 1752. Avison distinguishes between music resemblingextra-musical objects, including manifestations of emotions, such aslaughter, and music that is able to provoke an emotional response inthe listener. The former case is musical ‘imitation’,which Avison claims produces “a reflex act of theunderstanding” (Avison 1775: 50), while the latter is‘expression’. Avison argues that the goal of music is topursue expression, which is achieved by the competent use of melodyand harmony. While the idea that music is valuable because it movesthe listener was not new, Avison’s clarity in decoupling it fromimitation is noteworthy.
However, other theorists were not as explicit and careful as Avisonwith their terminology, nor as ready to jettison imitation, which theeighteenth century considered essential to the fine arts. Moreover, ithad been customary since the Camerata’s time to construeexpression as dependent on imitation (see Palisca 2006: chapter 10).Thus, a number of authors adopted conservative positions, defending aqualified concept of musical imitation, or simply accepting thatmusic’s imitative capacities are limited. In the second half ofthe century, imitation gradually took on a more peripheral role, andphilosophers started to focus on alternative sources of value formusic as an art, from formal structure to emotional arousal.
This gradual rejection of the idea of music as an imitative art isperhaps best manifested in France. This might seem puzzling, as theeighteenth century view of imitation as the hallmark of fine artsfound its most famous expression in The Fine Arts Reduced to aSingle Principle (1746), by Charles Batteux (1713–1780). Inthis work, Batteux defines art as the imitation of beautiful nature(belle nature). Different arts imitate different parts ofnature. Music imitates the human expression of emotions. It isinteresting to note that Batteux repeatedly states that musicexpresses the emotions, as opposed to imitating them,although the discrepancy seems to be merely terminological. Batteuxcan conceive of music that fails to be imitative, but does notconsider it valuable:
It would be like a chromatic harpsichord that offers us colours andseries of colours. It would amuse the eyes, perhaps, but bore themind. (Batteux 1746: part III, section III, chapter 2)
Earlier than Batteux, Jean-Baptiste Du Bos (1670–1742) hadsimilarly construed music as an imitative art in his CriticalReflections on Poetry and Painting (1719). Du Bos attempts toshow that, in vocal music, melody, harmony, and rhythm all contributeto the imitation of impassioned speech. Instrumental music imitatesnatural sounds instead. While he recognizes that music may charm thesenses, Du Bos thinks that the only valuable response to music is theone that comes from its appreciation as an imitation. Music may stillplease without imitating, but it is only when it imitates that itreally moves. Consequently, Du Bos condemns music that fails to beimitative.
The Encyclopedists Jean Le Rond D’Alembert (1717–1783) andDidier Diderot (1713–1784) also hold onto the idea that music isessentially imitative. D’Alembert states that
[t]he composers of instrumental music will make nothing but an emptynoise as long as they do not have in their heads […] an actionor an expression to be represented. (D’Alembert 1759:XXXVIII)
In the dialogue Rameau’s Nephew (1760), Diderot claimsthat
[s]ong is an imitation of physical noises or of the accents of thepassions, through the sounds of a scale invented by art or inspired bynature, whichever you please, either with the voice or with aninstrument. (Diderot 1760: 104ndash;5)
Musical imitation is also defended by Jean-Jacques Rousseau, but in animportantly qualified version (see the following section).
From the 1770s onwards, the view of music as an imitative art appearsincreasingly untenable, and the idea that music is expressive of theemotions because it imitates human expressive behavior also comesunder attack. This new trend is evident in works such as AndréMorellet’s (1727–1819) Of Expression in Music andImitation in the Arts (1771), as well as Musical ExpressionPlaced Among the Chimeras (1779). This latter work was written byBoyé, whose biographical details are unknown.Boyé’s treatise pairs the rejection of imitation with theclaim that melody and harmony are worthy of appreciation inthemselves:
The principal object of music is to please us physically, without themind putting itself to the trouble of searching for uselesscomparisons to it. (quoted in Lippman 1992: 95)
The composer and theorist Michel-Paul Guy de Chabanon(1730–1792) is arguably the most representative and radicalexample of the new skeptical mentality. In his Observations onMusic and Principally on the Metaphysics of Art (1779), Chabanonoffers a variety of arguments against the idea that the musicalexpression of emotions depends on the imitation of their vocalmanifestations. Children respond emotionally to music, he observes,yet they cannot appreciate musical imitation. The same is true ofindividuals who are unfamiliar with the Western musical tradition.Moreover, imitation is insufficient for expression. For example,laughter is naturally associated with gaiety, but a musical imitationof laughter would fail to result in gay music. Thus, music may expressemotions without imitating their external manifestations. Conversely,Chabanon observes that music may express emotions that do not have anytypical manifestation:
Many of our passions have no particular cry associated with them, andyet music may express them. (Chabanon 1779: VIII)
In addition to the already mentioned Avison, other English writersquestioned the role of imitation in music (see Schueller 1948).Francis Hutcheson (1694–1746), in An Inquiry concerningBeauty, Order, Harmony, Design (1725), distinguishes betweenoriginal and comparative beauty, the former being judged independentlyof any comparison, with the latter depending instead on imitation.According to Hutcheson, music is an example of original beauty,although its melodic element also allows for comparative beauty, inthat it may resemble impassioned speech. In detecting thisresemblance,
we shall be touched by it in a very sensible manner, and havemelancholy, joy, gravity, thoughtfulness excited in us by a sort ofsympathy or contagion. (Hutcheson 1725: section VI,XII)
This passage is significant because it gives a clear account of howmusic may express an emotion (through the imitation of itscharacteristic vocal expression), while at the same time arousing itin the listener.
In his Observations on the Correspondence between Poetry andMusic (1769), the Irish writer Daniel Webb (1719–1798)explains the emotional effects of music by appealing to aphysiological theory of emotions similar to Descartes’s.According to this view, music may give us impressions, thatis, it may put us in a state of mind, which may be further specifiedby accompanying words. In comparing music with painting and sculpture,Webb notes that the latter may act on our emotions only through theirimitations, that is, their represented content, whereas music may doso through both imitations and impressions (Webb 1769: 28). In thisway, Webb acknowledges that the expressive potential of music isgreater than that of the visual arts, and that this privilegedposition is not due to imitation.
James Beattie (1735–1803), a Scottish philosopher, voicesconcerns similar to those of Chabanon when he observes that there areemotions that music may express, that do not have a typical vocalmanifestation. On the basis of this, he rejects the view that music isof necessity imitative, though he concedes that it may be. Moresignificantly, Beattie claims that there is a crucial differencebetween music and the paradigmatically imitative art of painting.While good paintings are of necessity good imitations, and badpaintings are necessarily bad ones, good and bad works of music may beeither good or bad imitations (Beattie 1779: part I, chapter VI,section 1).
The Scottish philosopher Thomas Reid (1710–1796) argues thatperception presents us with sensations, which function as signs ofobjective qualities. Some sensations are artificial signs,instituted by human agreement. Such are words—the connectionbetween ‘cat’ and the animal sitting on my lap beingarbitrary. Other sensations are natural signs, and functionindependently of human agency—for instance, the connectionbetween smoke and fire. Natural signs typically require us to learntheir connection to the signified objective quality, as in the case ofsmoke and fire. However, in some cases a natural sign signifieswithout requiring prior experience—a child can detect the angerin an angry face and responds to it immediately. Musical expression isexplicitly described by Reid as an example of this category:
One kind of music inspires grief, another love, another rage or fury.These are all material representations of some affection of the mind.None of these are gained by experience. (Lectures on the FineArts, “Mind and Body”)
Thus, musical expression “is nothing but the fitness of certainsounds to produce certain sentiments in our minds” (Lectureson the Fine Arts, “Taste and the Fine Arts”). As forthe imitation of birdsong, battles, and the like, Reid maintains thatthese only marginally contribute to the beauty of music. Theimportance of Reid’s view resides in his philosophicallygrounded view of expression, which does not involve imitation atall.
The economist and philosopher Adam Smith (1723–90) traces asharp distinction between vocal and instrumental music. He holds thatthe greater the disparity between medium and imitated object, the morepleasure an imitative art will occasion (Of the Nature of thatImitation that Takes Place in what are called The Imitative Arts,part I, 6–7). Vocal music may thus only slightly resembleimpassioned speech, but precisely because of this it will delight uswhen it successfully imitates it (Of the Nature of thatImitation, part II, 12). Instrumental music, however, lacking atext and deprived of vocal articulation, cannot imitate any objectsuccessfully, as we can hardly ever recognize anything in it unlessprompted by a description (Of the Nature of that Imitation,part II, 17). Smith develops an alternative view of musicalexpression, which he illustrates with an interesting parallel. Just asa natural landscape may be gloomy without imitating anything, so maymusic. In both cases the description ‘gloomy’ ascribes adispositional property, that is, it refers to the capacity thelandscape and piece of music have of making us gloomy (Of theNature of that Imitation, part II, 23). Once more, expression andimitation go separate ways.
Although perhaps less radically than their French or Englishcounterparts, German intellectuals also increasingly questioned therole of imitation in music.
The philosopher and scientist Johann Georg Sulzer (1720–1779)still adheres to a rather orthodox view of imitation in music, holdingthat music imitates impassioned speech (Sulzer 1774,“Melody” [1792: III, 370–371; 1995: 91–92]).He also extends this account to instrumental music, and holds thatmusic that fails to express emotions is worthless (Sulzer 1774,“Instrumental music” [1792: II, 677–679; 1995:95–97]).
Johann Mattheson (1681–1764), the chief Affektenlehretheorist, embraces a Cartesian, physiological explanation of musicalexpression as the arousal of emotion produced by music’s actionon the animal spirits (this is the standard view of Mattheson’stheory—see Lippman 1992: 115–116; for an alternativeinterpretation, see Kivy 1984). Thus, the musician who wishes tocompose music expressive of an emotion will have to conform to themovements of the animal spirits associated with that emotion. Thisview alters the connection between emotional expression and imitation,as expression is no longer dependent on the imitation of impassionedspeech.
Johann Gottfried von Herder (1744–1803) retains the idea thatmusic is an imitative art, whose subjects are
all such things and incidents as are most eminently characterized bymotion and sound; these include all species of motions, sounds,voices, passions expressed through sounds, and so on. (Herder 1769: I,19)
This follows from the nature of music: whereas painting is the art ofspatial coexistence, music is that of succession. It is by respectingthese natural boundaries that the arts may achieve their effects.Thus,
music, which works wholly through the succession of time, must nevermake the depiction of objects in space its main object, asinexperienced bunglers do. (Herder 1769: I, 16)
There are similarities between Herder’s philosophy of music andRousseau’s. Herder also proposes that today’s music andlanguage may have a common ancestor, the main function of which wasexpressive (Herder 1772: part I, section III, 1; Herder 1769: IV,8).
Herder is skeptical with regard to the role of mathematical orphysical explanations in musical aesthetics, as these fail to explainour first-person hedonic and emotional responses. Much like Rousseau,he finds fault in Rameau on these grounds (Herder 1769: IV, 6).However, Herder does not abandon scientific explanations altogether,and in fact appeals to physiology to advance his own hypothesis(Herder 1769: IV, 6). While the details of this proposed explanationmight be uninteresting, it is noteworthy that, by appealing to acausal explanation, Herder is weakening the link between expressionand imitation.
3.5 The Primacy of Melody: Rousseau
Music was a lifelong concern for Jean-Jacques Rousseau(1712–1778), whose musical writings touch on an impressivevariety of topics, from music notation and theory to corephilosophical concerns such as musical value and expressiveness.Rousseau was also the author of a remarkably successful opera, LeDevin du Village (1752), performed around four hundred times overthe course of the fifty years following its premiere.
A contributor to Diderot and D’Alembert’sEncyclopédie (1751), Rousseau embarked on a systematicrejection of the views defended by Jean-Philippe Rameau(1683–1764), the leading music theorist at the time.Rousseau’s rejection of Rameau was not simply a rejection of hismusic theory, but rather of the entire musical practice it wassupposed to justify, that is, French music of the time.
Rousseau’s Letter on French Music (1753) may be seen asan episode in the so-called querelle des bouffons, a disputeconcerning the relative merits of Italian and French opera that wasprompted by the 1752 Paris performance of Giovanni BattistaPergolesi’s La serva padrona. In the Letter,Rousseau takes the side of Italian music, holding that its superiorityover French music is ultimately due to the Italian language, which ismore melodious and better articulated than French. French musicattempts to address the shortcoming of its language by adoptingcomplex harmonic structures, but Rousseau believes that this onlyfurther hinders its expressive potential. He observes how Italianmusic is more moving not in spite of its use of chords that are notcompletely filled out, but precisely because of this feature. In theLetter, Rousseau also develops the principle of “unityof melody” (unité de mélodie), accordingto which music ought to present the listener with one single salientmelody at any time, the other voices playing a subsidiary andsupportive role. It is from this melodic unity that music derives itscapacity to move the listener, and in this resides music’svalue. The concept of unity of melody is further elaborated in adedicated entry of Rousseau’s Dictionary of Music(1768), where he also states that his Devin du Village was anattempt to put that principle into practice (Rousseau 1768,“Unity of Melody”; on unity of melody, see Riley 2004:chapter 2, and Waeber 2009).
Rousseau’s polemical view of harmony as a hindrance to melodicdevelopment, as well as the consequent condemnation of French music,found a natural target in Rameau, who defended himself on variousoccasions (Rameau’s responses are collected in Scott 1998; onthe debate, see Verba 2016: chapters 2 and 3). In hisTraité de l’harmonie (1722), Rameau had alreadystated that melody is merely a part of harmony, and in his subsequentNouveau système de musique théorique, he hadattempted to ground the tonal system in the phenomenon of upperpartial harmonics, the faint overtones that are produced whenever anynote is struck. This means that there is strictly speaking no puremelody, as no note is ever heard in isolation from the harmonics itgenerates. The ascending series of upper harmonics, Rameau argues,contains the key to tonal harmony. In his Encyclopédiearticle “Dissonance”, Rousseau accepts that harmony may bepartly grounded in the structure of the harmonic series, but pressesRameau on the origin of the minor third, an interval that is ascrucial to tonal harmony as it is absent from the series. Moreimportant than these technical qualms is Rousseau’s fundamentalmethodological difference from Rameau. While Rameau’s musicalaesthetics depends on a theory of music including mathematical andphysical constructs inaccessible from the mere experience oflistening, Rousseau considers such experience the ultimate judge ofmusical value, and holds that it points to melody as the part of musicthat is responsible for its emotional impact on the listener (seeO’Dea 1995: 19–20). Insofar as it concerns the appropriatelevel of explanation of musical phenomena, the contrast betweenRousseau and Rameau is not unlike the one between Aristoxenian andPythagorean approaches to harmonics (see§1.2).
Rousseau’s views on the expressive power of music are furtherelaborated in his Essay on the Origin of Languages (1781), inwhich he argues that music and natural languages developed out of acommon source, a language used for the expression of feelings(Rousseau 1781: chapter XII, “Origin of music”; see Thomas1995: chapter 5). Following this common origin, both language andmusic underwent a process of decay. Languages developed in order tocommunicate increasingly complex thoughts, but lost expressive power,especially after the emergence of writing (Rousseau 1781: chaptersV–VI). Thus, Plato’s Greece, Rousseau states, had noshortage of philosophers, but lacked musicians (Rousseau 1781: chapterXIX). Regarding this remark, Martinelli (2012 [2019: 62])observes:
Anticipating the line of reasoning that would later be adopted byNietzsche, Rousseau openly contrasted music and philosophy, stronglyundermining the tradition that saw them as sisters.
According to Rousseau, music degenerated along with language andgradually lost its original expressive power. This happened as freemelodic phrases progressively turned into regimented musical systems,in which harmony and counterpoint were introduced in order tocompensate for the loss of natural melody (Rousseau 1781: chapterXIX).
It is tempting, though perhaps imprecise, to see in Rousseau’sview a precursor of the musilanguage hypothesis, recently proposed bySteven Brown (1999; on this, see Waeber 2013; Mithen 2005 advances aview similar to Brown’s).
As is apparent from his hypothesis concerning the origins of music,Rousseau thinks that the expression of emotions is music’s mainfunction. A qualified notion of imitation plays a crucial role inthis. While Rousseau accepts the eighteenth century view of art asimitation (Rousseau 1768, “Imitation”), he is also clearin stating that musical imitation is different from the kinds we findin other art forms. Music does not imitate the human passions or otherobjects by presenting us with their sonic equivalent, but rather byarousing in the listener the emotions that the object would give riseto. In Rousseau’s words,
It will not represent these things directly, but will arouse the samemovements in the soul that are experienced in seeing them. (Rousseau1768, “Imitation”)
It is because of this that musical imitation is not confined to onesense modality. If painting may represent only objects of sight,
[m]usic would seem to have the same limits with respect to hearing;nevertheless, it portrays everything, even objects which are onlyvisible: by an almost inconceivable magic trick it seems to put theeye in the ear, and the greatest marvel of an Art that acts only bymotion is to be able to form even the image of rest. Night, sleep,solitude, and silence are counted among Music’s great portraits.(Rousseau 1768, “Imitation”; similar claims on musicalimitation are made elsewhere in Rousseau’s oeuvre, see Rousseau1768, “Opera”, and Rousseau 1781: chapter XVI)
Effectively, the imitation that Rousseau considers central to music isthe arousal of an emotion that is specific enough to be associatedwith a given object. Thus, while Rousseau is not skeptical with regardto music’s capacity to imitate worldly objects, he is able todefend this view because he redefines imitation in terms of emotionalarousal. In this sense, he is still representative of a progressiveshift away from a view of music as imitation.
Despite its charm, Rousseau’s view of musical imitation does notseem to withstand critical scrutiny. It is unconvincing as anexplanation of the musical imitation of emotions. It may be thoughthat an emotion we perceive always corresponds to another emotion wefeel, and thus that we could depict the former by arousing the latter.But in fact, we may respond to, e.g., rage with a variety of emotions,from more rage, to fear, disappointment, schadenfreude, etc. Thetheory is only more implausible when it comes to the depiction ofobjects other than emotions, as the same object may arouse verydifferent emotions, depending on the subject’s disposition andbeliefs about it.
3.6 Music between Form and Content: Kant
While his comments on music are at times disparaged as lackingartistic sensitivity, Immanuel Kant (1724–1804) is arguably thefirst major modern philosopher to discuss instrumental music in thecontext of a systematic account of the arts. His influence onsubsequent musical thinking, and particularly on the formalist strand,is undeniable (see the entry onhistory of western philosophy of music: since 1800, section 1.6),but more controversial is whether he himself defends a variety offormalism. Kant’s main discussion of music is contained in thefirst section of the Critique of the Power of Judgment(1790), but some remarks are also found in his Anthropology from aPragmatic Point of View (1798).
Kant’s views on music follow from an account of beauty that mustbe briefly reconstructed. Kant contrasts the beautiful, whichaspires to universality, with the agreeable, which issubjective and idiosyncratic. According to Kant, when we ascribebeauty to objects we are making judgments of taste. These aresubjective judgments, and include a feeling of pleasure. In thissense, judgments of taste are quite like judgments of agreeableness.The beautiful, however, is different from the agreeable in that itrequires disinterested pleasure (interessenlosesWohlgefallen), that is, pleasure that does not presuppose theexistence of its object. For example, I may find beautiful an imaginedtune or painting, but I can’t find a refreshing drink agreeableif I merely think about it (Kant 1790: §2).
As they do not depend on concepts, and essentially include asubjective element of pleasure, judgments of taste are non-cognitive,and cannot therefore aspire to the same general validity as, say,mathematical proofs. However, judgments of taste are universal. Whenone judges an object to be beautiful, one is also at the same timeasserting that anyone ought to find it beautiful (Kant 1790:§6–7). This again stands in contrast with the agreeable.When I judge the taste of calvados to be pleasant, I am not requiringgeneral assent; in contrast, the pleasure associated with a work ofmusic is one I expect anyone to feel—it is a beautiful piece ofmusic.
The non-conceptual universality of judgments of taste is grounded inwhat Kant calls “the free play of the imagination and theunderstanding” (Kant 1790: §9). Ordinary perceptual andcognitive activities require the cooperation of imagination andunderstanding (here it is important to keep in mind that Kantconsiders imagination to be involved in perception) but in these casesimagination is guided by concepts. In the free play required byjudgments of taste, imagination and understanding interactunconstrained by concepts, but in a way that is common to all humanbeings (Kant 1790: §20 ff.).
Kant distinguishes between agreeable and beautiful arts. While bothinvolve pleasure, agreeable arts merely aim at arousing enjoyment,whereas beautiful arts involve cognition (Kant 1790: §44). Theapparent ambiguity in Kant’s remarks on music runs so deep thathe may seem to contradict himself even on the status of music as abeautiful art. He lists Tafelmusik (light music composed tobe a background accompaniment to dining) alongside the art of tellingamusing stories and jokes as an example of agreeable art, and addsthat this kind of music is supposed to encourage conversation“without anyone paying the least attention to itscomposition” (Kant 1790: §44). In the “Remark”that concludes the first section of the Critique, Kant oncemore lists music alongside the joke as an agreeable art. However,elsewhere in the same work Kant clearly counts music a beautiful art,ranking it alongside poetry and painting (§44, 191). Moreimportantly, Kant discusses music in §51, where he introduces hisdivision of the beautiful arts. Samantha Matherne (2014:135–138) has argued that Kant’s apparent hesitation isactually grounded in a different attitude we may take to music. If weattend to the bodily effect of the piece, music will be merelyagreeable, whereas a focus on the formal structure of the work willmake possible a judgment of taste, and thus music may be foundbeautiful.
A final point on musical beauty is worthy of mention. Kantdistinguishes between free and adherent (anhängend)beauty. Adherent beauty presupposes a concept of the object’spurpose. Thus, a work of architecture is judged beautiful or notaccording to its success or otherwise in housing people, institutions,or businesses. The resulting judgment of taste is characterized byKant as impure. Free beauty does not presuppose any concept. Asexamples of free beauty, Kant mentions abstract patterns, andinstrumental music, “indeed all music without a text”(Kant 1790: §16). In general, when Kant discusses music he haspure instrumental music in mind. If one considers that previousthinkers devoted almost exclusive attention to vocal music, this is asignificant shift in focus, and one that will be confirmed by thephilosophy of music following Kant.
A further ambiguity is represented by Kant’s apparent commitmentboth to a variety of musical formalism, according to which musicalbeauty uniquely resides in the relationship between its componentparts, and to the idea that music is expressive of emotions. This is amanifestation of a general tension in Kant’s aesthetics (seeGuyer 1977). A formalist tendency is evident when Kant claims that theproper object of the pure judgment of taste in music is composition,by which he means the arrangement of tones (Kant 1790: §14).Later on, Kant states that music “merely plays withsensations” (Kant 1790: §53), offering this in support ofhis claims regarding music’s low rank among the arts.
Despite these apparent professions of formalism, Kant also stressesmusic’s connection to the sphere of emotions. Particularly, heseems to accept a version of the Affektenlehre, according towhich works of music may arouse specific emotions in the listener. In§53, Kant affirms that music may convey an emotion because of itsanalogy with the tone of voice typical of impassioned speech (see alsoKant 1798: §18).
In the face of these textual ambiguities, an option is to adopt aformalist interpretation and hold that Kant was merely paying lipservice to the Affektenlehre (Kivy 2009: 50 ff.; Schueller1955: 224–25). However, other scholars deny that Kant was aformalist. Young (2020) attributes to him an imitation theory ofexpressiveness: music imitates the tones characteristic of impassionedspeech. Matherne (2014) attempts to reconcile formalist andexpressivist elements in Kant’s thought. Crucial to herinterpretation is the concept of an aesthetic idea, which Kant definesas
that representation of the imagination that occasions much thinkingthough without it being possible for any determinate thought, i.e.,concept, to be adequate to it. (Kant 1790: §49)
Aesthetic ideas are imaginative representations that cannot beexhausted by any given concept—think of an abstract painting andof the multitude of things it could look like or symbolize. Aestheticideas are crucially involved in our experience of the beautiful, andKant goes as far as to define beauty as the expression of aestheticideas (Kant 1790: §51). In his discussion of music, Kant statesthat the form of a composition—he explicitly mentions“harmony and melody”—expresses
the aesthetic ideas of a coherent whole of an unutterable fullness ofthought, corresponding to a certain theme, which constitutes thedominant affect in the piece. (Kant 1790: §53)
Here the expression of a piece’s overarching emotionalcharacter, a central concern for the Affektenlehre, isexplicitly linked to the piece’s formal structure. Matherne(2014: 134) concludes that Kant is defending what she terms“expressive formalism”, a view according to which music isable to express aesthetic ideas pertaining to the realm of emotions,and does so by means of its formal structure.
Kant’s view of music’s relative value as an art has alsobeen an object of critical attention (Weatherston 1996; Parret 1998).Kant ranks the arts according to two distinct criteria. When it comesto “charm and movement of the mind” (Kant 1790: §53),music is first among all of the arts, as it arouses emotions moredeeply than any other art form. It is clear that Kant does notconsider this criterion particularly important to artistic value.Additionally, music occupies the bottom of the hierarchy when oneranks the arts according to the second criterion, that is, “interms of the culture that they provide for the mind” (Kant 1790:§53). To provide culture for the mind is described by Kant as tocontribute to the enlargement of the faculties of imagination andunderstanding—think again of an abstract painting, and how itmay enrich our capacity to give visible form to some ideas.
While they may raise various difficult interpretive issues,Kant’s views on music have more than a purely historicalinterest. Some contemporary commentators and philosophers of musichave suggested that they might be brought to bear on current debates(see Bicknell 2002; Ginsborg 2011: 337; Matherne 2014:139–140).
Regardless of whether Kant’s formalism is or not a nineteenthcentury construction, his awareness of music’s purely formalaspect signals a change in Western philosophy of music. Vocal music isno longer the center of theoretical speculation. The philosophicalproblems that music poses are first and foremost those posed by pureinstrumental music, and it is its nature that must be clarified if oneis to understand music’s relation to language, concepts, andemotions.
