Philip Tagg (for J-J Nattiez, September 2015)


This document is not formatted correctly.
It contains no images, diagrams, quote indents,
music examples, symbols, font changes, etc.
It exists solely for site-indexing purposes.

To view this document correctly click here

Why was I sent here and not direct to the PDF document?

The Urgent Reform of Music Theory

Contribution to Festschrift for Jean-Jacques Nattiez, September 2015*

by Philip Tagg


In this text I argue for an urgent overhaul of music theory terminology. I try to explain the inadequacy of some widely used concepts and to suggest alternatives to the misleading and ethnocentric confusion they can cause. I will also highlight some aspects of structural denotation sidelined or absent in conventional music theory.

The main reason for my insistence on reform in this area of study is that from 1971 until 2009 I taught the history and analysis of popular music (including music for moving images), and that I could not explain the workings of such musics using solely the concepts of the music theory I had been taught. Since retiring in 2010, I’ve had time to reflect on, and become increasingly troubled by, the inability or reluctance of conventional music theory to address these issues.

Since my ‘campaign’ for reform has so far focused mainly on problems of tonal terminology, I’ll give here no more than a brief overview of some of those. Then I’ll summarise a few issues relating to timbre and form, omitting, for reasons of space, matters of metre, rhythm, duration and aural staging. I’ll end with some reflexions on the progress of my efforts to bring music theory out of the nineteenth century. But first I need to clarify a few basic issues.

Basic definitions and abbreviations

Music theory is itself a strange term. It’s part of musicology but distinct from anthropological, political, semiotic, social or any other theory of music involving a dynamic between music as sonic structure and its uses in human society, rather than the identification and systematisation of musical structures per se. In fact, the meaning of ‘music theory’ may be even more restricted.

‘Most commonly, the term describes the academic study and analysis of fundamental elements of music such as pitch, rhythm, harmony and form’.

Defined in this way, ‘music theory’ shares a lot in common with what the French used to call solfège in that both subjects focus on notatable musical structures (‘pitch, rhythm, harmony and form’) detached from their cultural meanings and uses. Now, while it’s clear that theories of sonic structure are essential to the identification of elements in a tripartite model of music semiotics (Nattiez, 1990), it’s obviously unsatisfactory if such identification is based on false premises. It’s those false premises that I try to expose in this text whose aim is to advocate not the abandonment but the radical reform of ‘music theory’. That’s why I have to distinguish between music theory in general, reformed or unreformed, and, with apologies for the clumsy expression, conventional [music] theory that is in acute need of reform.

To save space and avoid confusion, I will use the abbreviations of tonal designation presented in tables 1 and 2.

Table 1. Abbreviations for note, chord and key names (e.g. ‘E’)

Denotation type Symbol Typography Example

note e lower-case sans-serif e is a major third above c

lead-sheet chord E upper-case sans-serif From B7 to E (major triad)

key (Tonart) E upper-case serif constitutes a V-I cadence in E.

Lead-sheet and roman-numeral chord abbreviations follow the conventions set out in Everyday Tonality II, pp. 223, 231-233.

Space-saving reference symbols (b, 0, t, G, E, etc.) are explained at the start of the Reference Appendix. Please note that all source types are included in that single appendix.

Table 2. Common scale degree abbreviations

Scale degree note in C Popular verbalisation

 c one

$Ê | ^Ê | #Ê d$ | d@ | d# flat or minor 2 | [major] two | sharp 2

$Î | ^Î e$ | e@ flat or minor 3 |major 3

Ô | #Ô f | f# 4 (four) | sharp 4

$Û | Û | #Û g$ | g@ | g# flat 5 | 5 (five) | sharp 5

$â | ^â a$ | a@ flat or minor 6 | major 6

$ê | ^ê b$ | b@ flat or minor 7 | major 7

I also have to posit axiomatic definitions of some very basic terms.

• Note (n.): a single, discrete sound of finite duration.

• Tone (n.): a note with audible fundamental pitch.

• Tonal (adj.): having the characteristics of a tone or of tones.

• Tonic (n.): reference tone, keynote, tone of central importance.

• Tonality (n.): system or pattern, explicitly codified or not, according to which tones are configured.

• Mode (n.) tonal vocabulary (often abstracted and arranged in scalar form for theoretical purposes) of a piece or extract of music.

• Modal (adj.): having the characteristics of a mode.

• Euroclassical (adj.): relating to or having the characteristics of European classical music.


The central problem with tonal terminology in conventional theory is encapsulated in the two false and mutually contradictory pairs of opposites tonal v. atonal and tonal v. modal.

Tonal v. atonal (?!)

Here I’m assuming, in line with binaries like historical/ahistorical, typical/atypical, etc., that the initial A in atonal is an alpha privative, i.e. that atonal means without tones, non-tonal. If so, it’s small wonder that Schönberg disliked the label ‘atonal’ applied to his music because most of it followed the tonal rules of twelve-tone composition. Besides, neither he, nor Berg nor Webern were famous for their use of hi-hat, snare drum, sampled traffic or other non-tonal (= atonal) elements in their œuvre. Indeed, it’s hard to understand how euroclassical music theorists managed to confuse the notion of music full of tones but with no intended tonic, as in the work of twelve-tone composers or in music for the shower scene in Psycho (Herrmann, 1960), with music containing no tones, as in, say, taiko drumming (e.g. Kodō 1985) or the cue ‘Crows attack the students’ from The Birds (Herrmann, 1963). Surely music theory can distinguish between tone and tonic, can’t it? I mean, just as clinical things occur in clinics, just as musical things happen in music, and just as rhetorical devices, like this ‘just as’, are used in rhetoric, music featuring a tonic should be called tonical and tonal music with no intended tonic should be called either atonical or non-tonical.

Tonal v. modal (?!)

Tonal v. atonal may be a preposterous binary but it’s certainly no sillier than tonal v. modal, as explained next.

Asked to comment on the song cited in example 1, my students would often say ‘it’s modal’. ‘Which mode, with what as tonic?’, I would retort. ‘Dorian in C’ was a good guess, I thought, but still wrong, because the dorian consists of scale degrees Â Ê $Î Ô Û ^â $ê (c d e$ f g a b$ in C) whereas The Female Drummer contains no third, neither $Î (e$) nor ^Î (e@). ‘What mode are we in, then?’ was perhaps an unfair question because the music theory my students had learnt didn’t have a name for it. Nor did I at the time. Unable to find any theory of even the most common tonical hexatonic modes I had to invent a system to classify tunes like It’s Not Unusual (Tom Jones, 1964: doh-hexatonic), Jolene (Dolly Parton, 1973: la-hexatonic) and The Female Drummer (ré-hexatonic). Moreover, euroclassical music theory’s notion of ‘pentatonicism’ was limited to just two anhemitonic specimens, its ‘heptatonicism’ to no more than the seven ‘church modes’.

Ex. 1. Steeleye Span: The Female Drummer (Eng. trad., rec. 1971)

Now, if it can be agreed that mode names like ‘ionian’, ‘dorian’, ‘ré-hexatonic’ and ‘Hijaz’ denote configurations of particular tonal vocabularies, then modes must be intrinsically tonal, No one mode can therefore be qualified as more or less tonal (or modal) than any other. However, that’s exactly what conventional music theory seems to assume: the ionian (Â Ê ^Î Ô Û ^â ^ê) and the ionianised minor heptatonic modes (Â Ê $Î Ô Û ^â ^ê and Â Ê $Î Ô Û $â ^ê) are thought of as ‘tonal’ and as constituting ‘tonality’, while others —for instance Â Ê $Î Ô Û ^â $ê (dorian) or Â Ê Ô Û $ê (ré-pentatonic)— are called ‘modal’ and categorised as exemplifying ‘modality’.

It’s hard to explain such bizarre categorisation in other terms than either ignorance of, or unwillingness to recognise equal importance in, several different types of tonal configuration (tonality). Circumstantial evidence of this blind spot abounds in institutions of musical learning and in the culturally restricted repertoire[s] from which their terminologies derive. It’s a conceptual blind spot that is to all intents and purposes ethnocentric because it effectively relegates all types of tonality other than that of the euroclassical and classical jazz repertoires to the ‘non-tonal’ rag-bag of ‘modality’. Little wonder, then, that my students could do no better than identify the entirely ré-hexatonic Female Drummer as ‘modal’ (not ‘tonal’) while at the same time labelling the entirely ionian first eight bars of The Blue Danube as ‘tonal’ (not ‘modal’). Students should not have to put up with such nonsense.

Conversely, failure to conceptualise the ionian as just one mode among many others has at least two serious consequences. [1] If the ionian is not considered a mode, it cannot be readily compared with others in efforts to discover how its particular characteristics became so central to European culture in the age of Enlightenment during the rise of the bourgeoisie. [2] The more the ionian and the ionianised minor-key modes are accepted as the tonal normality from which other types of tonality are thought to diverge, the more difficult it becomes to hear and understand the dynamic of those other tonalities. Here are seven examples of that problem.

1. Notions of dominant, subdominant and perfect cadence become irrelevant or misleading, as with IV, V and I in Guantanamera, La bamba (I-IV-V), etc., or in V-IV(-I) blues progressions, or mixolydian chord loops ((I-)$VII-IV-I), etc.

2. Other harmonic cadence labels become absurd; for example, ‘interrupted’ and ‘half’ cadences are often final or complete.11

3. Bimodality, shuttling between two tonal centres, modal counterpoise etc. are ignored as key factors of tonal construction.

4. Full, final phrygian cadences are not heard as final.

5. Many common modes are either misnomers (e.g. ‘phrygian’ and ‘lydian dominant’ scales with no dominant function, ‘frigio mayorizado’ instead of Hijaz) or not named at all (e.g. ‘ré-hexatonic’, ‘doh-hexatonic’, ‘la-hexatonic’).

6. Given the existence of tonal traditions other than the tertial euroclassical before, during and after the historical period covered by conventional music theory’s ‘tonality’ (e.g. pentatonicism), the concepts ‘pretonal’ and ‘post-tonal’ make no sense.

7. Quartal harmony is radically undertheorised.

The seventh point demands some clarification.

Quartal tonality

Put simply, quartal qualifies tonality featuring fourths and their octave complement, fifths, rather than thirds (and sixths). The first conceptual problem here is that quartal sounds, like the non-ionian[ised] modes discussed earlier, have in conventional music theory largely been seen as divergent from a culturally restrictive notion of ‘tonality’. They have required identification as an ‘Other’ (‘quartal’, ‘modal’), whereas euroclassical tonality has unilaterally been assumed to need no such categorisation. However, as soon as the concept of tonality had to become field-dependent, the previously absolute norm also needed a label to distinguish it from the ‘abnormal’. At least four bizarre labels for third-based, euroclassical harmony resulted from this process: tonal, diatonic, functional and triadic. The label tonal harmony is useless because it merely reiterates the old absolute norm. Diatonic is absurd because it erroneously implies that no quartal polyphony ever visited all notes in a diatonic heptatonic mode. Functional harmony is no better because it implies that quartal harmony has no functions. Finally, triadic harmony is, as shown next, a seriously flawed concept.

Ex. 2. Four tertial and five quartal chords

Example 2 shows nine chords, the first four based on stacked thirds, the last five on stacked fourths. Chord numbers 1, 2, 4, 5, 6 and 7 are triads because each contains three differently named tones, chords 3 and 8 are tetrads (four differently named tones) and chord 9 is a pentad (five). Chords 5, 6 and 7 are quartal triads, chord 8 a quartal tetrad and chord 9 a quartal pentad. Since chords 1, 2 and 4 are neither more nor less triads than chords 5-7, and since harmony based on stacked fourths is called quartal, harmony characterised by the stacking of thirds should be called tertial. In short, the binary triadic v. quartal is false because it confuses two distinct criteria of designation —the number of notes in a chord (‘triadic’, ‘tetradic’, etc.) and the principle of interval stacking in a chord —tertial for thirds, quartal for fourths. Chords 1, 2 and 4 are therefore tertial triads, chord 3 a tertial tetrad. As such, chords 1-4 belong to the sphere of tertial harmony and are indicative of tertial rather than quartal tonality.

Quartal ubiquity and structural theory

Quartal sounds abound in everyday tonality. Not only are they a feature of vernacular chorality in rural Russia, of Central Asian komuz accompaniment, of Appalachian banjo playing, of various styles of folk and prog rock, of news jingles, of library music for corporate modernity, and of signals for digital devices; they also feature prominently in recordings by jazz artists like McCoy Tyner, as well as in the work of Stravinsky, Copland, Hindemith and Bartók. As with its apparent disregard for tonical hexatonicism, conventional music theory has also, it seems, failed to deal systematically with the widespread practices of quartal tonality. Unfortunately, there’s no room here to explore border areas between quartal and tertial harmony, but it is worth drawing attention to a few key aspects of quartal tonality.

The most obviously false label applied to quartal tonality is ‘sus’ (suspension) because no quartal chord can be a suspension in a quartal context any more than Î can be dissonant in a tertial chord. In quartal harmony, no Ô needs resolving to Î and any Ê, ô or $ê will almost certainly be intrinsic to the quartal consonance of which it is part. The notion of an ‘omitted’ third is also meaningless in quartal harmony. These basic truths affect quartal chord designation so that, for example: [1] the ‘open fifth’ dyad a-e should be abbreviated as A5, not AT3, because nothing is omitted; [2] the triad a-d-e should be A4, not AS4, and [3] the triad d-e-a should be D2, not DS9 because nothing is suspended.

Another cardinal difference between tertial and quartal chords is that the former can preserve their tonical identity in inversion: both Czq (e-g-c, first inversion) and Czs (g-c-e, second) are heard as variants of C (c-e-g, root position), so to speak. Quartal triads don’t work that way because a triad like c-f-g (C4) inverted as f-g-c will sound like F2 and, inverted to g-c-f, it becomes GÁ. The idea that the tonal centre of quartal chords can float up to two positions round the key clock without upsetting their basic sense of tonal rootedness can also be understood by comparing the progress of tertial and quartal triads round the circle of fifths (example 3).

Ex. 3. Tertial (1) and quartal (2) triads flatwards round the key clock

Five key points can be gathered from example 3.

1. While each tertial progression (1) involves holding one and changing two of the triad’s constituent notes, proportions are inverse for each quartal change (2).

2. One of the note changes in the movement of tertial triads involves a semitone, e.g. from e to f, a minimal pitch distance but all of five steps away on the key clock. That’s what gives ^Î its strong leading-note directionality towards Ô (equivalent to ^7-8 in the target triad). In quartal triad progressions, the single replacement note is only three key-clock steps away at the minor third (e.g. g replaced by b$).

3. Quartal harmony has to shift tonal centre by three key-clock steps to sound like a ‘new key’. A tonal area covering three key-clock positions functions as a tonical neighbourhood.

4. All notes in a tertial triad have disappeared just two ‘hours’ later round the key clock. Each quartal triad note lasts for three.

5. Quartal triads contain notes related to each other by a fourth and/or fifth. Notions of V or Û as ‘dominant’ and IV or Ô as ‘subdominant’, as well as of modulation between I and IV or V, are therefore meaningless. ‘Major’ and ‘minor’ keys, as well as ‘perfect’, ‘plagal’, ‘interrupted’ and ‘half’ cadences are also largely pointless.

These are just a few of the key theoretical points that can help explain the workings of widely used tonal idioms qualifiable in general terms as ‘quartal’.


Timbre is probably the parameter of musical expression most frequently singled out by popular music scholars as conventional music theory’s most serious blind spot. This observation derives from such evidence as: [1] timbre’s conventional relegation to the status of a secondary parameter; [2] conventional theory’s lack of systematic vocabulary descriptive of timbre; [3] the paucity of timbral data in the notation that is conventional theory’s preeminent form of musical reference, storage and retrieval. These three interrelated symptoms are part of a vast problem complex involving such issues as unawareness of the relative importance of different parameters of expression in different musics and insufficient problematisation of the notion of a musical work and of its commodity form. There’s no room here to discuss this problem complex any further, but I will suggest two lines of inquiry —one more structural, the other relating to the binary poïesis v. aesthesis— that may shed some light on the matter.

Power chords

Example 4 illustrates a more structural approach. It shows the full pitch content of a power chord, in this case the dyad A5 played on distorted electric guitar. Although only two notes —a (110 Hz) and e (165 Hz)— are actually played by the guitarist, the full sound produced consists not only of those two fundamentals and their overtones but also of an acoustically quantifiable difference tone at a=55 Hz —the distortion fundamental— plus all of its overtones. Analysis of this sound suggests that power chords are extremely stable tonal entities thanks to the harmonic series consisting of partials defining them and by the actual acoustic presence of a difference tone. This tonal rootedness makes the power chord an essential constituent of heavy metal, industrial and grunge music because it allows for the unequivocal statement of substantial ‘tonal elsewheres’ in relation to the tonic (I5). The $II5 of many phrygian heavy metal pieces illustrate this phenomenon, as do the bold harmonic steps in Nirvana’s Lithium and Teen Spirit (1991).

Ex. 4. Basic power chord harmonics for A5 (a2 110 Hz, e3 165 Hz)

It should be noted that there are no clear-cut borders between tone and timbre in these heavy metal and grunge contexts. Timbre cannot be considered as a secondary parameter to tone; it should instead be understood as a co-determinant in the production of tonality in the repertoire of which it is part.

Poïetic and aesthesic descriptors

As noted above, conventional music theory lacks a systematic vocabulary descriptive of timbre. It also relies on notation as its primary mode of storage, which, as already stated, encodes little or no timbral data. Moreover, since notation must normally be performed for it to be heard, it can be considered an intrinsically poïetic medium. These observations seem to suggest that structural descriptors of timbre may, at least at the time of writing, be easier to create on an aesthesic rather than poïetic basis. Indeed, I’ve found that students with little or no knowledge of music theory —’non-musos’— have generally been less inhibited than their ‘muso’ peers when it comes to characterising aspects of non-notatable sound. This tendency has been particularly striking in discussions of vocal timbre where persona descriptors, be they psycho-somatic/emotional (e.g. over-the-top, edgy, cheeky) or demographic and archetypal (e.g. young, working-class, Celtic folk virgin, suicidal student), occur more frequently than do more directly sound-descriptive concepts (e.g. high-pitched, rasping, indistinct). It’s also worth underlining that aesthesic descriptors can also connect with notatable parameters of expression, for example the final minor major nine chord at the end of the James Bond theme (Em^9, a poïetic descriptor) referred to as a ‘spy chord’, ‘detective chord’, ‘danger chord’, etc. (aesthesic descriptors).

The most obvious problem with aesthesic descriptors is that they inevitably rely on linguistically and culturally specific patterns of metonymy. For example, ‘wet echo’ as an aesthesic descriptor of reverberation lasting, say, 3½ seconds with in- and output signals set to the ratio 65:35 (poïetic description) makes no sense translated literally into Italian as un eco bagnato, just the culturally more adequate rendering of a ‘wet echo’ as un eco della Madonna would be meaningless if translated back into English as ‘an echo of Our Lady’. However, not only can aesthesic descriptors of this type work well inside local areas of cultural reference: they can also be compared with each other. Their commonalities of spatial, gestural, tactile and kinetic connotation could be investigated and, together with studies of lexical patterns used in the naming of loops, sound effects templates and synthesiser presets, could form the basis of a more transculturally viable vocabulary of aesthesic descriptors.


According to the Oxford Concise English Dictionary (1995), form means the shape or pattern into which different parts or elements are arranged, ordered, or otherwise combined into a whole. Con-template, if you would, this bush.

The bush’s constituent elements (stem, branches, twigs, etc.) have been intentionally cut and trimmed (arranged) into a form resembling that of a mushroom. It may have taken a long time to produce that form but its perception as a form is virtually instantaneous. Now please consider example 5. It lasts for four seconds (8 beats at =120) and is heard several times in succession, with a few minor variations, before being replaced by a slightly different groove. It contains several constituent elements —the jil L on guitar and bass, jl on violins and flute, zl il_z on trombones plus hi-hat and kick drum patterns (not shown). All these elements have been combined into an identifiable whole (Gestalt) that repeats as a single ongoing unit. It has, in short, a form.

Ex. 5. Museme stack (excl. drumkit) in Shaft (Isaac Hayes, 1971)

The mushroom bush and example 5 illustrate another conceptual conundrum in conventional music theory which, despite the dictionary definition, is unlikely to consider example 5 as form. Instead it reserves the term to mean the way in which episodes or sections are ordered ‘horizontally’ along the unidimensional axis of passing time to create longer patterns of musical change and recurrence. ‘Sonata form’, ‘rondo form’, ‘da capo aria’, ‘32-bar jazz standard’ and so on exemplify that use of the term. Now, of course, the narrative or episodic aspect of arranging, ordering or otherwise combining parts into a whole is an essential aspect of form in music, but it is certainly not the only one. Indeed, if scholars seriously wish to investigate form as form rather than as ‘form’, they will need to consider at least two equally important aspects of the phenomenon: [1] its extensional, episodic, narrative, diachronic, ’horizontal’ properties —its diataxis— and [2] its intensional, synchronic, ‘vertical’ properties —its syncrisis. Syncritic form is, as illustrated in the four seconds of example 5, perceptible within the limits of the extended present, and, in many types of popular music, both intended and heard as a composite of aurally staged, simultaneously sounding motifs, riffs, chords, instruments, voices, timbres, pitches, rhythms, etc. in a particular metre at a particular speed and dB level, etc. Syncrisis is in other words a batch of ‘now sound’ that involves more the shape and form of a dynamic state than of a process or narrative, of a ‘being’ rather than of a ‘becoming’. As such, it constitutes the ideal formal unit for examining museme stacks that may, through repetition, occupy entire episodes but which can also function as building blocks in larger formal units (periods, episodes, sections, entire pieces, etc.).

Conventional music theory’s conceptual monopolisation of ‘form’ is in other words similar to its hijacking of ‘tonality’. The main difference between the two is that while the conceptual falsification of tonality gives rise, as explained earlier, to a seriously confused and flawed terminology —’tonal’ as opposed not only to ‘atonal’ but also to ‘modal’, ‘pretonal’ and ‘post-tonal’—, its inability or reluctance to deal with form other than diataxis constitutes an error of conceptual omission. There is in other words no valid reason why narrative processes in music (diataxis) should be prioritised at the expense of music’s dynamic states (syncrisis), especially in relation to musics where the latter is of primary interest.

Que faire?

Since October 2011 I’ve conducted a mini-campaign aimed at convincing colleagues and students that something needs to be done about the sort of conceptual chaos described in this text. Progress has been slow and I’ve had to resort to several tricks of persuasion to further the cause. I have, for example, distilled some of the issues down to simple requests, like ‘please distinguish between tone/tonal and tonic/tonical’, ‘please distinguish between triad/triadic and third/tertial’, ‘please treat all types of form as form’, ‘please help create a viable vocabulary for timbre and aural staging’, etc. Then I’ve delivered different versions of my ‘Troubles with Tonal Terminology’ presentation in numerous places and tried to reach a wider audience by producing a video called What (the Hell) is Tonality? I’ve also added other, equally problematic, factors to the mix, such as conventional theory’s inadequate conceptualisation of metre, rhythm, syncopation, periodicity, polyphony, counterpoint and duration (including the extended present). I’ve even tried appealing to a collegial sense of pride in musical scholarship in the hope of minimising the intellectual and ethical embarrassment that music theory’s terminological disorder can cause in interdisciplinary contexts. For example, I’ve had to point out that no self-respecting linguist would ever conclude that, say, English or Spanish were grammatical languages and that, say, Chinese or Russian were not, just because definite and indefinite articles are key features of syntax in the former but not in the latter, whereas many music theorists seem quite happy to think in that way about tonality: one type of tonality with its particular traits is called ‘tonal’ but others, with different traits, are not. I’ve repeatedly expressed astonishment at the absence of any outcry in my discipline against such foolishness.

Moreover, if musics other than those in the euroclassical and classical jazz traditions remain uncodified, the terminology of conventional euroclassical music theory will stay unchallenged and continue to marginalise, trivialise or falsify any type of music exhibiting important traits for which that theory has either flawed concepts or no concepts at all. Not only would such neglect prolong the undemocratic disrespect and ethnocentric ignorance it seems to show towards so many musics used by a majority of the world’s population; it would also, as argued earlier, obstruct efforts to understand what made the musical tradition on which it based that same terminology so influential and unique.

Final reflexions

Reactions to my little ‘campaign’ have generally been more positive than negative. Here are just two of them.

‘I’ve taught analysis in my university for 25 years and I’ve met the same problems as you… We even tried to set up a musical terminology commission ten years ago but’...

‘I was raised in a wide range of musical traditions. Then I recently studied musicology, so had to confront all those [strange terms]… Well, with the [local/regional/national] musical traditions I know... Wow! A lot of those terms just don’t work at all.’

Some reactions have been more guarded, for example:

‘Of course you’re right and what you say is perfectly logical but there’s not a hope in hell that anything will come of it!’

I fear that this opinion, expressed by a senior professor of composition after one of my presentations on the topic, may be a realistic assessment of the situation. Two closely related types of reaction have prompted me to think along such lines. One is the compact silence I met on several occasions when presenting the sort of issues discussed here. ‘They think you’re causing unnecessary problems’, a young researcher told me outside the venue at a particularly soul-destroying music analysis event in Italy. ‘They just want you to disappear so they can get back to “business as usual” in their ivory towers’.

As with the shut up and go away strategy just described, the other type of negative reaction is also characterised by a complete lack of will to enter into any sort of dialogue: I’m not even presented with any counterarguments refuting my line of reasoning. This second type of strategy is, however, more aggressive than the first. I interpret it as a rearguard action mounted by those who feel threatened not just by budget cuts to the arts and humanities but also by those of us who advocate real reform. The problem is that instead of safeguarding the future of their discipline by addressing the multicultural internet reality in which students actually live, these fellow scholars of mine regress into the illusory safety of an institutional past with its single time-honoured repertoire and a storage technology dating back to the fifteenth century. These knights of the musicological rearguard, keen to protect their notions of music, including its structural theory (not to mention their jobs) from attack, then pull up the drawbridge and shore up the defences of their fiefdoms, ensuring that only those who defer to their authority can enter. They form alliances with others in similar fiefdoms and banish from their realms anyone who commits the heresy of questioning their legitimacy. These acts of excommunication are often accompanied by derogatory rumours circulated about the heretic and his/her followers, who are then willfully kept out of the institutional loop of academic exchange and employment. This sort of behaviour is exacerbated by the managerialisation of university life, by barrages of Kafkaesque audits and assessment exercises, as well as by subjection to ‘league tables’ that can logically function as a competitive comparison mechanism only if there’s a consensus about ‘what everybody does’ or ‘always has done’. Those bizarre mechanisms favour seats of musical learning which focus on traditions that have to be socially dead, or at least moribund, in order for them to become fixed as canons —for example, the euroclassical canon, the jazz canon, the ‘academic safari’ canon and, more recently, the rock canon. Such fixation of repertoire, of aesthetics and structural theory, is more often than not understood as a necessity in institutions that have to repeat courses from one year to the next in the name of course content consistency or cost cutting. All these mechanisms encourage stagnation and generate a climate of fear detrimental to reform and innovation. To survive short-term in the system you need to toe the line and keep a low profile. To secure funding, you and your institution have to chase brownie points, tick the right boxes and submit to the academic publishing protection racket. If you don’t feed the management monster in these ways you’ll incur its wrath and be visited by its inquisitors.

Heresy, inquisition, excommunication? Who knows how far things can go in a university system under the pall of market fundamentalism? Maybe, now that students are ‘customers’ and teachers ‘service providers’, we’ll soon have to sell degrees like indulgences. When that happens and enough of us are excommunicated, our universities can be left entirely in the hands of accountants. Then, like Luther and the German peasant rebels of the 1520s, we can have a full-blown reformation of university life. In the meantime, we cannot do much more than expose flaws in the current system and pin our theses to the door of musical learning. This text is intended as just one such thesis.


* This article, accepted by the editors for inclusion, will/did not appear in the Festschrift because I was regrettably unable to sign the publishing contract which required me to relinquish rights to use my text as I saw fit and to not make it publicly available before its appearance in the volume. It was too late: I’d already put it on line! Besides, several other conditions in the contract were contrary to principles of accessibility to knowledge that I’ve explained in ‘The Academic Publishing Protection Racket’ (G and ‘Pay for knowledge: why?’ (G [both 160418].

1. More reasons are given in Tagg (2014: 1-6).

2. See instead Tagg (2013: 281-303) and Lacasse (2000, passim).

3. ‘Music theory’ entry in Wikipedia, 2015-09-07.

4. These pages can be viewed on line without downloading the whole book: G

5. For full explanation, see ‘Producing a Reference Appendix for Studies of Music in the 21st Century’ at G [150919].

6. Please note that in neo-Latin languages tonalité, tonalità, tonalidad etc. mean Tonart (German) or key (English). I used to advise my francophone students to keep tonalité in the sense of key and to use an expression like idiome tonal to cover Tonalität and tonality.

7. The euro prefix avoids confusion with non-European classical traditions, e.g. Tunisian nouba, Indian rāga, the yăyuè of imperial China; see Tagg (2014: 488).

8. Of course, the whole-tone scale is hexatonic, but it is not tonical (no , no ). For a systematic theory of tonical hexatonic modes, see Tagg (2014: 165-173).

9.Only the doh/major and la/minor pentatonic modes were at all familiar to my students. The thirdless ré- and sol-pentatonic modes were totally unknown, as were hemitonic pentatonic modes. Even the ‘church modes’ were a mystery to many of my students (except heavy metal adepts), as were other common heptatonic modes like Hijaz and Niavent, both included in the applet Λαϊκοί Δρόµοι ( which lists scale degrees for nineteen heptatonic modes commonly used in Greek popular music.

10. C5 is played as an ‘open’ 5th dyad c-g, B$5 as b$-f, etc. The tune, as sung and as played on fiddle, varies slightly from verse to verse.

11. Bimodality: see Tagg (2014: 433-442, 453-455); counterpoise: see Tagg (2014: 336-339, 396-398).

12. Final phrygian: see Tagg (2009: 15:05-19:00; 2014: 436‐442).

13. See Tagg (2014: 421-450); see also Tagg (2009). Lydian and phrygian ‘dominants’ are misnomers used mainly in jazz theory.

14. i.e. when it became necessary to compare one type of tonality wish another on an equal footing. Field-dependent argumentation: see Toulmin (1972).

15. It’s as if Béla Bartók, McCoy Tyner, Clarence Ashley, Joni Mitchell or Sokratis Málamas has never made music.

16. It could also be argued that quartal elements are present in music by Mussorgsky, Debussy and Dvořák. For more on this topic, see Tagg (2014: 306-349).

17. For example, the contribution of chords of the eleventh (from Dvořák’s New World Symphony and Burleigh’s arrangements of spirituals via Copland to gospel harmony and film/TV underscores) to a sense of US-American tonal identity in the twentieth century; see Tagg (2014: 306-315).

18. For more, see Tagg (2014: 293-294). N.B. The notions presented here, including the chord abbreviations, are not part of standard music theory teaching.

19. For explanation of quartal chord symbols, see Tagg (2014: 293-296).

20. Many important features of quartal tonality are absent from this account, e.g. [1] the confusion, in jazz theory, between quartal harmony and quartal voicing; [2] the quartal character of much ‘droned’ music using open tunings and its tonal kinship with anhemitonic pentatonicism, especially the quartal (thirdless) ré and sol modes; [3]

21. See, for example, Meyer (1989: 14-22).

22. Phrygian heavy metal examples: Powerslave (Iron Maiden, 1984); Wherever I May Roam (Metallica, 1991). Nirvana sequences: Lithium: D5 F#5 B5 G5 B$5 C5 A5 C5 in D; Smells Like Teen Spirit: F5 B$5 A$5 D$5 in F.

23. Figure 4 is a based on Lilja (2009: 104‐114, esp. p. 113), in its turn based on research conducted at Helsinki University of Technology’s Department of Signal Processing and Acoustics.

24. Poïesis and aesthesis are concepts deriving from Molino via Nattiez (1976). Their meanings are adapted here as follows. Poïetic descriptors are those that designate terms denoting a structural element of music primarily from the viewpoint of its construction (e.g. con sordino, glissando, major minor‐nine chord, analogue string pad). Aesthesic descriptors are terms denoting structural elements primarily from the viewpoint of how they’re perceived (e.g. allegro, spy chord, Scotch snap, cavernous reverb).

25. See ‘Vocal persona’, chapter 10 in Tagg (2013: 343-382).

26. The original name of the Dr No title theme is actually 007 (Norman, 1962).

26. The extract starts at 1:44, transcription by Bob Davis (2005: 299).

28. Diataxis (διάταξις) means ‘order of service’ in Byzantine Orthodox liturgy. As a chronological ordering of constituent episodes into an overall form, diataxis can be used to designate what concepts like ‘sonata form’ and the ‘32-bar jazz standard’ have in common. Syncrisis (σύγκρισις) literally means a putting together, an aggregate, combination, etc. Syntax can be considered to constitute a third and equally essential type of musical form. For further explanation of points in this footnote, see Tagg (2013: 383-385; 417; 586, 603).

29. The extended present (also misleadingly called the ‘specious present’) lasts no longer than a musical phrase, or an exhalation, or a few footsteps, or a short gestural pattern. It’s a duration experienced as a single unit (Gestalt) in present time, as ‘now’ rather than as an extended sequence of ideas. The extended present can also be imagined as the human brain’s equivalent to a computer’s RAM where information is processed immediately, rather than as its hard drive (longer-term memory) where access and retrieval times are longer. For further explanations, see Tagg (2013: 19-20, 272-273, 588).

30. Aural staging: see Tagg (2013: 299-303, 583).

31. For explanations of durational hierarchies in music, see Tagg (2013: 281-288).

32. E 9ZE-DpkRYb0 [150919].

33. I received these comments by email in late 2013 from a senior music professor in Venezuela and from a musician and musicology student in Colombia.

34. I regret not having noted the student’s name. I just wrote down his comments.

35. Gutenberg (1398-1468) introduced Europe to the printing press with movable type. Petrucci (1466-1539) is often cited as the father of music printing.

36. The situation is particularly serious in Italian institutions of musicology (see ‘Popular Music Studies in Italian Universities —a petition’ G [150918], especially about the baronie academiche and settori disciplinarie). We are currently preparing a new report on irregularities in the field of Italian musicology.

37. See, for example: [1] ‘Academic publishers make Murdoch look like a socialist’ by George Monbiot. The Guardian, 110829. G [150918]; [2] ‘How researchers can protect themselves from academic journal racket’ by Coen van Laer, Online Library, Maastricht University, 150917 G [150918]; [3] ‘Aaron Swartz Was Right’ by Peter Ludlow, The Chronicle of Higher Education, 130225 G [150919].

38. I was subjected to a management inquisition at Liverpool University; see Audititis: a rampant contagion G [180919].

39. See George Soros: The Crisis of Global Capitalism as reviewed by Beams (1998).

40. Thanks to Goffredo Plastino (Newcastle) for the indulgences analogy and to Kaire Maimets (Tartu) for reminding me of the perversion of the student-teacher relationship into that of buyer and seller.

41. Luther is said to have nailed 95 theses to the door of the Schlosskirche Allerheiligen in Wittenberg in 1517, thereby igniting the Protestant reformation. If only Luther had supported the peasant rebels instead of condemning their actions!


This appendix contains all types of reference. To save space, the following symbols are used: b bibliographical source (written word); t audiovisual source; 0 audio recording; G internet reference; E YouTube file. Please note that E replaces so that, for example, ‘E rWlt9Is1nms’ means the URL .

b Beams, Nick (1998) ‘Soros warns of “market fundamentalism”’.

World Socialist Web Site, 1998-12-22

G [180918].

b Davis, Bob (2005) Who Got da Funk? An Etymphony of Funk from the 1950s to 1979. PhD, Faculté de musique, Université de Montréal

G [2012-11-15].

0 Herrmann, Bernard (1960) Psycho (Colonna sonora originale).

RCA Cinematre NL 33224 (1975) EXd_OolN34eA [131229].

t —(1963) ‘Crows attack the students’ from The Birds

E hplpQt424Ls (0:42, ff.) [131229].

0 Jones, Tom (1965). It’s Not Unusual. Decca F 12062.

0 Kodō (1985) ‘Miyake’. Kodō – Heartbeat Drummers Of Japan

0 Sheffield Lab – cd-kodo E juT0drDIcvw [111128].

b Lacasse, Serge (2000). Listen to My Voice: The Evocative Power of Vocal Staging in Recorded Rock Music and Other Forms of Vocal Expression. PhD diss., Institute of Popular Music, University of Liverpool.| [101021].

b Meyer, Leonard B. (1989). Style and Music: Theory, History, and Ideology. Philadelphia: University of Pennsylvania Press.

b Nattiez, Jean‐Jacques (1976) Fondements d’une sémiologie de la musique. b Paris: Ugé.

b — (1990). Music and Discourse: Toward a Semiology of Music. Princeton University Press, 1990

0 Nirvana (1991) ‘Smells Like Teen Spirit’ and ‘Lithium’. Never Mind, Geffen 434425‐2-S.

0 Norman, Monty (1962) ‘Theme from Dr No’ (a.k.a. ‘James Bond Theme’, a.k.a. ‘007’). The Best of Bond, United Artists UAS 29021 (1975).

0 Parton, Dolly (1973). Jolene (single); also on Jolene; RCA Victor AFL10473 (1974).

0 Steeleye Span (1971) ‘The Female Drummer’ (Eng. trad. via The Watersons). Please to See the King. Crest 8.

t Tagg, Philip (2009) Dominants and Dominance E rWlt9Is1nms [110723].

b —(2013) Music’s Meanings: a modern musicology for non-musos. New York & Huddersfield: Mass Media Music Scholars Press

G [130215].

b —(2014) Everyday Tonality II: towards a tonal theory of what most people hear. New York & Huddersfield: Mass Media Music Scholars’ Press.

b Toulmin, Stephen (1972). Human Understanding: The Collective Use and Evolution of Concepts. 1972. Princeton University Press. 2015-09-19, 15:58