Gamelan and the microtonal imagination: how Indonesian music opened up new worlds of tuning theory

 And, perhaps, what the xenharmonic community can do to give back


There is a moment that keeps recurring in the history of Western music — the moment when a composer or theorist first hears gamelan and something shifts permanently. It happened to Claude Debussy at the 1889 Paris Exposition, where he heard Javanese musicians and reportedly returned again and again over the course of the exhibition. It happened to Erik Satie at the same event. It happened to Benjamin Britten in Bali in the 1950s, to Lou Harrison when he began his decades-long collaboration with Indonesian musical ideas, and to countless university students across the United States and Europe who sat down in front of a bronze metallophone for the first time in an ethnomusicology class and felt the intervals do something completely unexpected to their ears.

What keeps triggering this moment is not mere exoticism. It is the genuine discovery that the harmonic universe is much larger than twelve equal semitones, and that someone else has already been living in a different corner of it for centuries — and making breathtakingly beautiful music there. For the xenharmonic community, the community of composers and theorists devoted to tuning systems outside standard Western equal temperament, gamelan has been one of the most generative influences imaginable. Its specific scales have inspired whole families of temperament theory. Its approach to timbre has reshaped how microtonal composers think about instrument choice. And its existence as a living, sophisticated, centuries-old tradition has provided the most powerful possible answer to anyone who dismisses alternative tuning as an eccentric theoretical hobby.

This article is a celebration of that influence, and a deep dive into the specific theoretical structures it produced. It also takes seriously some genuine questions about how the xenharmonic community names and frames its relationship to Indonesian musical culture — not as a reason for guilt, but as an invitation to do something that is already happening in the community: get more thoughtful, more careful, and more genuinely curious about the source.

What gamelan actually is, and why it matters for tuning theory

Gamelan is the traditional ensemble music of Java, Bali, and Sunda in Indonesia, built predominantly around metallophones — tuned metal keys struck with mallets — alongside gongs, drums, and sometimes bamboo flutes and vocals. It has been played in recognisable forms since at least the 12th century, and in 2021 it was inscribed on UNESCO’s Representative List of Intangible Cultural Heritage of Humanity. But what makes it so significant for tuning theory is not its age or its cultural prestige — it is the specific mathematical properties of the scales it uses, and the way those scales are physically instantiated in the instruments.

Every gamelan set is individually hand-forged and tuned as a complete, self-contained unit. This means no two gamelans in the world are tuned alike. Colin McPhee, the Canadian composer who spent years in Bali in the 1930s, observed that the variation between individual gamelans is so large that one could reasonably say there are as many scales as there are gamelans. Each ensemble is its own harmonic world, with its own flavour and character. This is not a limitation — it is a deliberate aesthetic choice, one that the xenharmonic community has found deeply inspiring, because it embodies a principle that RTT theorists also hold dear: that there is no single correct tuning, only tunings that are well or poorly suited to particular musical intentions.

Javanese gamelan uses two main tuning systems. Slendro is roughly pentatonic — five notes per octave — with intervals that are large and somewhat similar to each other in size, though not perfectly equal. Pelog is heptatonic — seven notes per octave — with strikingly uneven intervals: some very small, some very large. Musicians typically play in five-note subsets of pelog rather than using all seven tones at once. In Bali, instruments are additionally tuned in pairs that are slightly out of tune with each other, producing a shimmering interference beat called ombak (meaning “wave”) that gives Balinese gamelan its characteristic living, breathing shimmer. This beating is not a flaw to be corrected — it is aesthetically central, representing, in the cosmology of Balinese music, the heartbeat of a living system.

For xenharmonic theory, the key insight that gamelan delivers is this: tuning and timbre are deeply coupled. The metallophones of gamelan have an inharmonic spectrum — their overtones do not follow the neat whole-number ratios (2x, 3x, 4x the fundamental frequency) that you find in a violin or a flute. The intervals that sound consonant on a gamelan metallophone are therefore not the same intervals that sound consonant on a harmonically rich instrument. As the Xenharmonic Wiki’s gamelan page explains, for gamelan to sound in tune, instruments must match their harmonic partials with the inharmonic partials of the percussion instruments — which is why slendro approximates 5edo with stretchedoctaves, and pelog approximates a 2L 5s scale of 9edo with octaves that range from slightly compressed to slightly stretched. The stretching is not random: it accommodates the physics of the metal.

This idea — that the right tuning depends on what your instruments actually sound like, not on some universal definition of consonance — is one of the most important contributions gamelan has made to microtonal thinking, and it is genuinely useful to any composer working in alternative tunings today.

Slendro, the equipentatonic connection, and the slendric temperament

Let’s start with slendro, because it produces the most elegant theoretical connection. If you play a slendro scale and ask “what Western tuning system does this resemble?”, the answer that comes up fastest is 5edo: five equal divisions of the octave, each step roughly 240 cents (where a standard semitone is 100 cents). The intervals are wide, open-feeling, and — in the equal version — perfectly symmetric.

Of course, real slendro is not exactly 5edo. The Xenharmonic Wiki’s slendro page is careful about this: slendro provides “an octave-repeating pentatonic scale, although octaves are usually stretched to account for the high inharmonicity inherent to gamelans.” Real slendro intervals are all slightly unequal, and the octave itself is wider than the pure 2/1 ratio. But the resemblance is close enough to be illuminating, and 5edo is a useful first approximation.

This connection runs directly into one of the more interesting temperaments in xenharmonic theory: slendric. Slendric is a rank-2 temperament (meaning it is generated by two intervals: an octave and a generator) that operates in the 2.3.7 prime subgroup — meaning it focuses on approximating ratios built from the primes 2, 3, and 7, rather than the full harmonic series. Its generator is a supermajor second of approximately 233.7 cents, which is close to the ratio 8/7. The defining property of slendric is that three of these generators stack to a perfect fifth — specifically, three 8/7 intervals multiply to approximately 3/2.

The comma that slendric tempers out — the interval it treats as zero — is called the gamelisma, and it has the ratio 1029/1024. This is a small interval of about 8.4 cents, and the fact that it is relatively small means slendric is quite accurate: it approximates many intervals within 1 or 2 cents in its optimal tuning. The name “slendric” was introduced because the basic five-note scale of this temperament — the 1L 4s MOS scale — is a near-equalized pentatonic similar in character to slendro. It was originally named “wonder” by the theorist Margo Schulter, and that name still has currency in the community.

What does slendric actually sound like as a scale? The five-note MOS has one large step (L, representing the ratio 7/6, a subminor third) and four small steps (s, representing 8/7, the generator itself). The result is a pentatonic scale where one step is noticeably bigger than the others — not a perfectly equal pentatonic, but one with a recognisable character. The larger scale that slendric produces best — the 11-note MOS — alternates 9/8 “whole tones” with tiny 32-cent “sixth tones,” creating a dense chromatic environment with an unusual flavour.

The specific harmonic resources of slendric are interesting and worth noting. Because slendric lives in the 2.3.7 subgroup, it gives excellent approximations of intervals involving the seventh harmonic: 7/4 (a flat minor seventh), 7/6 (a subminor third), 7/5 (a narrow tritone), and so on. These intervals have a distinctive sound — open but slightly exotic, neither the smooth consonance of 5-limit harmony nor the outright dissonance of a minor second. For composers interested in septimal harmony (harmony built on the seventh partial), slendric offers an interesting approach that comes packaged with a built-in connection to the pentatonic aesthetics of slendro.

The notable edos that support slendric — tuning systems where slendric’s generator is well approximated by an equal division step — include 31edo, 36edo, 41edo, 46edo, and 77edo. These are all tuning systems with their own identities and musical properties, but they can all be used as tunings for slendric music.

The gamelismic family: expanding outward

Slendric is a 2.3.7 subgroup temperament, which means it doesn’t naturally handle the fifth harmonic (the prime 5) — the generator chain doesn’t reach it efficiently. This is where the gamelismic family becomes important. Gamelismic is a rank-3 temperament — it adds an independent dimension for prime 5 — that also tempers out the gamelisma, 1029/1024. Because it has an extra dimension, it has more flexibility than slendric and can handle fuller harmonic textures including 5-limit consonances alongside the septimal ones.

The gamelismic temperament has an obvious 11-limit extension called portent, reached by additionally tempering out 385/384 and 441/440. This works because the gamelisma factors neatly in the 11-limit: 1029/1024 = (385/384) × (441/440). So portent is in some sense the natural home of slendric’s musical ideas once you expand to include the full range of concordant harmonics.

The 11-limit portent temperament enables a class of chords called essentially tempered chords — specifically, gamelismic, keenanismic, and werckismic chords. These are chords whose internal structure relies on the tempering of specific commas to make intervals that would otherwise be distinct collapse to the same pitch. They are a uniquely xenharmonic sound: chords that could not exist in just intonation or in standard equal temperament, that only emerge in specific tuning environments.

The gamelismic family also connects to a range of extensions: mothra (which tempers out 81/80, placing the fifth harmonic at 12 generators up), rodan (tempering out 245/243, with the fifth harmonic at 17 generators), and guiron (tempering out the schisma, with the fifth harmonic at 24 generators). Each of these is a distinct musical environment, but they all share the core slendric structure — the equipentatonic backbone inspired by slendro.

Pelog, antidiatonic scales, and the mavila family

Now to pelog, which is in some ways the richer and stranger of the two gamelan scale systems from a xenharmonic perspective. Pelog is a seven-note scale with highly unequal intervals. The Xen Wiki’s pelog page notes that in Javanese gamelan, the tones are numbered 1 through 7, and it is not uncommon for the pairs 1–5, 2–6, and 3–7 to be approximated by the ratio 3/2 — a perfect fifth. There are three traditional modes (pathet) of pelog, each using five of the seven tones, with different emphasis on which tones are central.

The key structural observation that xenharmonic theory has made about pelog is that it resembles a 2L 5s scale — a scale with two Large steps and five small steps. This pattern is called antidiatonic in the TAMNAMS naming system (a community-developed naming framework for MOS scale patterns), because it has the same large-to-small ratio as the diatonic scale (2 large, 5 small) but with those categories flipped from the diatonic arrangement: in a diatonic scale, the two small steps are the semitones and the five large steps are the tones, while in antidiatonic scales, the two large steps stand out against five small ones.

The 2L 5s page on the Xen Wiki uses the informal name “peletonic” or “antidiatonic” for this pattern, and it gives generators ranging from approximately 514 cents to 600 cents (or equivalently, 600 to 686 cents for the complementary generator). The simplest tunings of 2L 5s are 7edo (equalized, where L and s are equal), 9edo (the “basic” tuning, where L = 2s), and 16edo (the “soft” tuning, where L = 3/2 × s). These three equal divisions of the octave are the most natural homes for antidiatonic music.

The most important rank-2 temperament that produces 2L 5s scales is mavila. Mavila is defined by a very flat fifth — in its optimal tuning, around 679 cents — and it tempers out 135/128, the mavila comma. The striking property of mavila, as mentioned earlier, is that four flat fifths land you on a minor third rather than the major third you’d get with a standard or slightly narrowed fifth. This turns the diatonic scale inside out, giving you the antidiatonic pattern where the “leading tone” steps are the large ones and the “whole tone” steps are small.

Mavila supports several different MOS scales: the five-note scale (2L 3s), the seven-note antidiatonic scale (2L 5s), and a nine-note scale (7L 2s). The nine-note scale is one of the reasons mavila is theoretically interesting — it gives composers three different non-chromatic scales to work with rather than the usual two (pentatonic and diatonic) that most other temperaments offer.

The mavila family page documents a rich set of 7-limit extensions: armodue (which tempers out 36/35 and is the main 7-limit extension of mavila, supported beautifully by 16edo), mavling (tempering out 126/125), pelogic (tempering out 21/20), hornbostel (729/700), bipelog (50/49, a two-period temperament), and mohavila (1323/1250). Each of these extends the harmonic palette of mavila-family music in different directions.

A closer look at pelogic and the naming question

Pelogic deserves some special attention, both musically and culturally. It is the 7-limit extension of mavila that tempers out 21/20, and it is described as the 7d & 9 temperament — meaning 7edo (with a modified mapping, hence “7d”) and 9edo both support it. The 7/4 interval in pelogic is mapped to the major sixth of the antidiatonic scale. Its name dates back to at least 2004, and it was accepted into the community’s standard nomenclature in 2011.

Musically, pelogic is one of the less accurate members of the mavila family — its optimal tuning has significant errors in some intervals — but it has the virtue of being well approximated by 9edo and 16edo, two of the most naturally accessible alternative tuning systems. For a composer who wants to explore pelog-flavoured harmony in a seven-tone framework without investing in complex microtuning setups, pelogic in 9edo is a remarkably direct route.

Here is where the cultural sensitivity question becomes genuinely useful rather than merely cautionary. The name “pelogic” is close enough to “pelog” that someone unfamiliar with the distinction could reasonably assume they are the same or equivalent things. They are not. Pelog is a family of specific tuning practices developed over centuries by Indonesian musicians, varying from region to region and gamelan to gamelan, carrying specific aesthetic meaning and cultural context. Pelogic is a Western RTT construct that approximates the structural pattern of pelog — specifically, the 2L 5s step pattern — using a regular temperament framework. The relationship is one of inspiration and structural resemblance, not identity.

The xenharmonic community’s internal discussion on this point, documented in the Xen Wiki’s cultural appropriation analysis page, is worth taking seriously — not because it reaches perfect conclusions, but because it shows exactly the kind of careful reflection that makes cross-cultural musical engagement better rather than worse. The same analysis gives five stars to “mavila” (a village name, giving credit to a specific place of discovery without claiming equivalence), four stars to “bipelog” (which adds a prefix indicating it is a variant rather than the thing itself), and three stars to “superpelog” (which the evaluator reads as “pelog but more extreme” rather than “better than pelog”). These are useful distinctions.

Bipelog: a harmonically interesting case

Bipelog is worth a moment of special attention because it has an unusual structure: it has two periods per octave instead of one, with each period being approximately a tritone (7/5 ratio). It tempers out 50/49 (which identifies the 7/5 and 10/7 tritones, making them equal) and 135/128 (the mavila comma). Its optimal tuning has the fifth at around 685 cents. The 7-limit mapping places the 7/4 at the minor seventh of the antidiatonic scale.

Bipelog’s two-period structure means that its scales have a symmetric, palindromic quality that single-period mavila-family scales don’t have. This gives it a distinct character: harmonic patterns repeat at the tritone, creating a kind of double symmetry that can be exploited compositionally. It is well supported by 14-note, 30-note, and 44-note equal divisions of the octave.

Quasipelog theory: a new synthesis

One of the most recent and ambitious examples of gamelan-inspired xenharmonic thinking is quasipelog theory, developed in December 2025 by wiki user BudjarnLambeth. The framework applies specifically to 16edo and related tuning systems (23edo, 30edo, 37edo, 39edo), all of which support the mavila temperament. The theory defines four main scale types within the framework — quasipelog minor (mode ssLsL of the mavila pentatonic MOS), quasipelog major (LssLs), quasipelog heptatonic (ssLssLs of the mavila seven-note MOS), and quasipelog chromatic (the full 16-note mavila MOS) — and then specifies a rich set of “flavoured” variants derived by raising or lowering individual scale degrees by small amounts.

What is particularly striking about quasipelog theory is its explicit timbre guidance. The framework specifies that instruments with harmonic timbres should be avoided, and that gamelan instruments, inharmonic metallophones, inharmonic bells, and “inharmonic synth timbres with a ‘metallic’ or ‘bell-like’ sound” are the appropriate palette. It even recommends human voices with a “thin, reedy tone” and woodwinds with a “sine wave-like” timbre. This is sophisticated and theoretically grounded advice: the inharmonic overtone series of bell-like instruments genuinely makes the flat fifths and antidiatonic intervals of mavila temperament sound more natural, for the same reason that gamelan’s inharmonic metallophones make pelog sound beautiful rather than wrong.

This is gamelan theory being applied rather than just referenced — a case where the xenharmonic community is not just borrowing a name but genuinely learning from the musical wisdom embedded in the source tradition.

The Debussy moment, repeated

It is worth pausing to appreciate what the xenharmonic community has received from this inheritance. The encounter between Western musical thinking and gamelan has produced a genuinely expanded conception of what harmonic space looks like. The fact that slendro resembles 5edo but is not identical to it — that real gamelan scales live in the spaces between equal temperaments, each ensemble occupying its own particular point in tuning space — has given the microtonal community a vivid argument against the idea that tuning should converge on a single optimum. The fact that pelog can be described as a 2L 5s pattern has given theorists a precise way to talk about a whole family of scales that share a structural character while differing widely in their actual interval sizes. The fact that gamelan sounds beautiful with inharmonic timbres has given composers an actionable guide to how to make their own alternative-tuning music sound natural rather than dissonant.

The scales associated with gamelan — the antidiatonic 2L 5s pattern, the equipentatonic 1L 4s pattern, the various mavila-family modes — have proven to be musically rich in their own right, independent of their Indonesian origins. They support interesting harmonic structures, they have a distinctive aesthetic character, and they connect to the broader mathematical structure of tuning space in illuminating ways. This is not accidental: gamelan musicians developed these scales over centuries precisely because they are beautiful and expressive, and that beauty does not disappear when the scales are studied through a different theoretical lens.

The cultural dimension: honest appreciation and honest limits

The concerns about naming and framing in the xenharmonic community are real and worth taking seriously, and they do not require the community to be defensive or ashamed. The problem is not that slendric is inspired by slendro, or that pelogic is inspired by pelog — inspiration across cultural boundaries is one of the great engines of musical creativity. The problem is when inspiration is presented as equivalence, when a name implies that a Western temperament is the Indonesian scale rather than resembling it in one specific mathematical dimension.

Real pelog is not a temperament. It is a family of practices, aesthetics, regional variations, cultural meanings, and physical instruments that cannot be reduced to a step pattern. A musician who learns pelogic temperament in 9edo has learned something genuinely interesting and musically valuable, but they have not learned what pelog actually is. That distinction matters — not as a reason to abandon the theoretical work, but as a reminder that the theoretical work is a map, and maps are not the territory.

The community has resources available to it. The Xenharmonic Wiki already contains careful documentation of these temperaments and their relationship to gamelan. Researchers like Sumarsam at Wesleyan University have written extensively on Javanese gamelan theory from within the tradition. The Smithsonian Folkways catalogue contains Philip Yampolsky’s twenty-volume Music of Indonesia series. The Aural Archipelago project has made field recordings of dozens of Indonesian traditional music styles freely available online. The engagement between Western tuning theory and Indonesian musical culture can be a genuine dialogue rather than a one-way extraction — but it takes active effort to make it so.

A conclusion: what the xenharmonic community should do

Having spent time with this material — the music, the theory, the history, the community’s own internal debates — I want to offer what I think is a genuinely good path forward. This is my own view, and I hold it with appropriate humility, but I think it can be defended.

The xenharmonic community should embrace its gamelan inheritance enthusiastically and without guilt, while committing to three specific practices that make that embrace more honest and more generous.

The first is a commitment to distinguishing inspiration from equivalence in naming and documentation. This does not mean renaming every temperament that references gamelan — names like mavila, which the community itself rates as excellent, show that it is absolutely possible to give credit without claiming identity. But it does mean being consistent about framing. When a wiki page introduces slendric, it should say clearly that the scale resembles slendro in one specific structural way, not that it generates slendro scales. When a composer talks about pelogic harmony, they should feel comfortable saying “pelog-flavoured” or “pelog-inspired” rather than “pelog.” These are small linguistic shifts that cost nothing and gain a great deal of honesty.

The second is a genuine investment of curiosity in the source. The xenharmonic community is, by its nature, a community of people who love exploring unfamiliar harmonic territory. The same spirit of adventure that drives someone to compose in 16edo or investigate the gamelismic family ought to drive them to actually listen to gamelan — not as background research, but as music. The Balinese gamelan tradition alone contains dozens of distinct ensemble types with their own repertoires, aesthetics, and histories. Javanese court gamelan has a theoretical literature that is sophisticated and complex on its own terms. Sundanese gamelan is dominated by the sound of the suling flute and has a completely different character. These are not footnotes to the mathematics — they are the point. A xenharmonic composer who composes in slendric temperament and has never heard a slendro gamelan is missing something essential.

The third is a willingness to channel some of the enthusiasm that gamelan has generated in the xenharmonic community back toward the living practitioners of the tradition. This does not need to be complicated or grand. It can be as simple as citing Indonesian musicians and scholars when their work is relevant, supporting organisations that document and preserve Indonesian musical traditions, being honest with audiences about where musical ideas come from, or — for those who are serious — making the effort that Lou Harrison and Colin McPhee made, which is to actually study with practitioners and learn the tradition on its own terms rather than only through the lens of Western theory.

None of this diminishes the beauty of slendric harmony, the mathematical elegance of the gamelismic family, or the musical richness of antidiatonic scales. If anything, it deepens all of those things by connecting them honestly to the tradition that inspired them. The moment of recognition that Debussy experienced at the 1889 Exposition — that sudden sense of a much larger musical universe — is available to anyone who approaches gamelan with open ears and an honest mind. The xenharmonic community has been experiencing versions of that moment for decades, and it has produced genuinely exciting theoretical and musical work as a result.

The best version of this story is one where that excitement flows in both directions: where Indonesian musicians know that their tuning systems have inspired a worldwide community of composers and theorists, where that community knows enough about the source to respect what it has borrowed, and where the mathematics and the music stay in conversation rather than the mathematics swallowing the music whole. That story is already being written, in the work of thoughtful people on all sides of the exchange. The invitation is simply to write it more deliberately, more generously, and with more joy.

Comments