Most readers should skip the bulk of this essay, which is dedicated to a thorough analysis of the piece in question. The “Putting it all together” section does not assume that the reader has read anything else, and importantly, provides hyperlinks back into the bulk of the essay, for the benefit of readers who want or need additional explanation, detail, &/or argumentation.

For readers already familiar with screamo &/or with Kidcrash, the “Historic background” section may be skipped, so that reading begins at “Putting it all together”.

The term screamo would seem to imply little more than a particular vocal style.^[1] But the maturation of screamo was a relatively long time in the works, and its patient fermentation led to an impressive variety of flavors, both commonplace and strange. After all, with apparently^[2] no pitch‐material to be had in the vocals, a fundamental function of one of the most basic instruments in popular music had been brazenly severed from it. Although this wouldn’t make screamo the first niche of popular music to do away with explicitly melodic vocals, it still meant that the vacuum left behind had to be addressed somehow.

Innovated in California in the early–mid 1990s, screamo would take on greater dissonance (in both pitch‐material & timbre), faster tempi, and more mercurial song structures than its hardcore & emocore forebears. By Portraits of Past’s 01010101 (1996)^[3], a style of elaborate forms, wide‐ranging dynamics, highly‐melodic material carried almost exclusively in guitars, and the palpable influence of post‐rock had already developed.

Some later acts of the more well‐known screamo era would elaborate upon the elements borrowed by Portraits of Past from post‐rock: Envy would bring the sounds of bands like Mogwai more explicitly into the genre, City of Caterpillar would create music that was as much a post‐rock masterpiece was it was a screamo one, and French bands like Mihai Edrisch, Daïtro, & Sed Non Satiata would refine the use of chords to incredible expressive precision.

Others would pave different paths. Bands like Jeromes Dream, Orchid, & Ampere would incorporate the extreme tempi and short, violent bursts of powerviolence to produce short songs (or even miniatures) with a frenzied, through‐compositional approach frequently drawing upon the rhythmic complexity and general “angularity” of mathrock. Saetia, & later Off Minor, would draw explicitly upon jazz for harmonic material, melodic material, and more.

But still others are yet further off the beaten path. Formed in 2000 somewhere in the U.S. (probably New Mexico^[4]) as The Kidcrash, this band began their career with New Ruins (2004)^[5], an album of a kind of pop‐punk–mathpop hybrid.

Although New Ruins contained some fruitful (if perhaps ungerminated) seed, it wasn’t until this band moved to Portland, Oregon and started playing screamo that they became the Kidcrash (not the The Kidcrash!) that we’re interested in here.

Kidcrash’s next major release would be Jokes (2007) — an album that, by all appearances, would be even less self‐serious than their previous efforts. With a name like Jokes, song titles like “Ron Ghousley’s Fucked Up Dream (Ron To The Hills, Ron For Your Life)” and “Life Was Real. Vital. Urgent. Important/Bum Guts”, and sketchy hand‐pencilled album artwork (see Fig. 1), the first‐time listener would be forgiven for expecting nine tracks’ worth of goofy mathrock or similar.

Most readers should skip ahead to “Putting it all together” at this point. Hyperlinks back into the bulk of the essay will be provided, for the benefit of readers who want or need additional explanation, detail, &/or argumentation.

The album begins with “Turtlelephant”, clocking in at 4:44. By carefully investigating this opening piece, I hope to answer at least a few questions about what makes this track, and the album as a whole, special:

• Is Kidcrash “painting an image” here? If so, then what is it like? How does such a “painting” compare to the Western classical notion of “tone‐poem” (Tondichtung)?
• If Kidcrash was (during this era) ostensibly a screamo band, then how does their music relate — or not relate — to other works in the genre?
• Those familiar with Kidcrash have wondered: how are they so “heavy” & yet so “melodic” at the same time? What would that even mean? Why does it seem like no other band writes music that comes anywhere close to Kidcrash’s style?
• Can we make sense of Kidcrash’s song structures? By drawing upon the experiment­al­ism of much post‐hardcore, Kidcrash’s music seems to venture quite far from the verse–chorus structures of much popular music — as well as its rhythmic structures. To what extent, if any, is Kidcrash’s music through‐composed?
• Are there any textural features that define Kidcrash’s sound on this album? Pitch‐material features? Timbral features?
• Kidcrash was also associated with mathrock. What about their use of drumkit, rhythm & meter, and perhaps texture contributes to this genre association?

The beginning of the album is a quiet, ominous theme:

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

Although the introductory gesture in Fig. 2 is played on an electric guitar, its sound is somewhat indistinct, and has a strange reverberation about it. The immediate impression is that the listener is hearing the guitar only from a great distance, presumably through some sort of barrier that’s gently muffling the sound. Indeed, analyzing a spectrogram reveals a strange hierarchy of harmonics, shown in Table 1.

There’s no particular preference for even or for odd harmonics; and yet, the harmonics are muffled seemingly at random, as if the fog that the sound is traveling through absorbs certain wavelengths better than others. The result is a strong emphasis on the M3−, and an unusual preference for the M2 over the less‐dissonant m7↓ two harmonics below it.

Fig. 2 includes tablature, due to the significance of the open notes (i.e. notes played on unstopped strings, denoted by “0” in tablature). Open notes naturally have a shinier, more chimelike quality to them.

Bells are known for their inharmonicity, and their bizarre distribution of amplitudes among their partials. Ultimately, the scene painted by this short introductory figure is one of bells being rung in a belfry in the distance:

• Timbral structure, including the clear attack, strange amplitude distribution of partials, and focus on M3− partials & the dissonant M2 partial.
• Notes are left to ring, which also causes clearly‐audible inharmonic beating effects.
• Slow pace (half‐notes).
• Use of only four distinct pitches (even without octave equivalence).
• The ritardando toward the end.

The pcs presented are {C♯, D♯, E♯, F♯}. Although played rather like chords (note the laissez vibrer indication), this is, pc‐wise, just a diatonic tetrachord — perhaps implying F♯ major. In particular, setclass ⁅0⁣2⁣4⁣5⁆_4‐11B,^[6] with the emphasis very much on the crunchy m2 interval between E♯ & F♯ in the bass.

Perhaps more important is the melodic contour of this figure:

• Each note is repeated twice in immediate succession.
• The melody repeats the following overall contour: go up twice, then down twice.
• When it reäppears, the “mobile” highest (read: most melodically prominent) note has descended by M2.

Some of this can be summarized by the Parsons code *rururdr(d). All of this will be relevant to the rest of the piece.

The intro opens up into ⟪A⟫ with a snare run (see: drum notation):

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

Full score for the first iteration of this section

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

At a tempo of roughly 𝅘𝅥 ≈ 147, this section is marginally faster than the intro. But the ritardando at the end of the intro obscures this relationship anyway. Nevertheless, owing also to the change from 𝄵 to 𝄴, we get the feeling of a sudden increase in pace.

The immediately striking aspect of ⟪A⟫ is its built‐in metric displacement: the drumkit — & to a lesser extent, the bass‐guitar — often follow the “true” meter (particularly, its downbeats); on the other hand, both guitars are a single eighth‐note “ahead”, meaning that their “downbeat” is on the “& of 4”. The drumkit & bass‐guitar also flirt with this displacement as well, with the drumkit consistently placing a kickdrum hit on both the “1” and the “& of 4”.

Interestingly, the drumkit & bass also flirt with the opposite displacement: the bass often comes down to its target bass note only just on the “& of 1”, and the drumkit consistently emphasizes both the “1” and the “& of 1” with kickdrum‐backed open hi‐hat hits. The overall effect is a subtly unstable meter that is, apparently, in three places at once.

In most measures, the drumkit places a snare hit on the “3” and the “a of 3”, effectively dividing the latter half of each measure into (3+3+2)⁄16 = ⟨⁠𝅘𝅥𝅮𝅭𝅘𝅥𝅮𝅭𝅘𝅥𝅮⁠⟩.

In the first guitar part (guitar 1), the first four measures of ⟪A⟫ very faithfully recast the basic thematic material outlined in the “Pitch‐material & contour” section above. The only arguable difference is that the c‐seg (rather than simply “contour” in the Parsons‐code sense) is slightly different, owing to the fact that there are now two “middle” pitches instead of just one: ⟨0⁣0⁣1⁣1⁣4⁣4⁣2⁣2⁣ 0⁣0⁣1⁣1⁣3⁣3⁣2⁣2⟩_c instead of ⟨0⁣0⁣1⁣1⁣3⁣3⁣1⁣1⁣ 0⁣0⁣1⁣1⁣2⁣2⁣1⁣1⟩_c.

The latter half (i.e. the following four measures) takes the basic ⟨0⁣0⁣1⁣1⁣2⁣2⁣1⁣1⟩_c segment and moves the entire thing up by (an interval of) a second, rather than only moving the highest pitch.

Thus, in addition to inheriting the *rururdr(d) contour of the intro, ⟪A⟫ also inherits the “‘mobile’ melodic prominence self‐differing by second” aspect, and quadruply so: in guitar 2, the high E is responded to by the high D; in guitar 1, we have the three periodic–hyper­metric relationships illustrated in Fig. 5 below.

The texture of ⟪A⟫ is of particular interest for basically two reasons:

• This general sort of texture — especially in its more elaborate forms, as we’ll later hear — is typical of Kidcrash, and as I want to argue, typical of much math­rock.
• This general sort of texture is also not well‐described by any of the traditional textural categories of Western music theory.

But first, we need to clarify the role of guitar 2. We can see in Fig. 3 that it mostly plays F♯₃, and for this reason, ⟪A⟫ is not a shining example of this texture. Nevertheless, Fig. 3 writes the F♯₃s in two distinct voices, because those with stems pointing down are played fretted (on the 5^th fret), and those pointing up are on the open string (this is reflected in the tablature included in the full score in Fig. 3). We can think of the part played here by guitar 2 as being a degenerate (in the mathematical sense) version of the basic ⟨0⁣0⁣1⁣1⁣2⁣2⁣1⁣1⟩_c segment, where the c‐pitches 0_c and 1_c have — in pitch‐space — been identified.

With this in mind, we can hear that the interplay between the three pitched instruments has the following general characteristics (bearing in mind that more complex & fleshed‐out examples exist, and will come later):

This texture is not well‐described by any traditional terminology, as can be demonstrated by process of elimination:

Counterpoint

Per (b.), voices are only quasi‐independent, frequently violating contrapuntal norms.

Moreover, per (c.), the lack of phrasal independence produces further distance from “counterpoint” in the usual sense.

Homorhythm

Per (c.), voices may have independent rhythms, in spite of phrase boundaries & tempi generally being dependent.

Moreover, home­ophony does not necessarily produce block‐chords — unlike homorhythm, which can be thought of as a sequence of block‐chords.

Homophony

Although some examples of home­ophony can technically be described as homophony (albeit not necessarily usefully so), others cannot, owing to the lack of distinction between “main melody” and “accompaniment”. This lack of distinction is even further heightened by (d.).

Heterophony

This is perhaps the most interesting comparison. However, (c.) ensures that there’s not enough independence in tempi or phrase boundaries to make the term heterophony useful.

Moreover, home­ophony does not necessitate that the voices be variations of a single melody, which is the main requirement for heterophony.

In the specific context of the ⟪A⟫ section of “Turtlelephant”:

1. The satisfaction of (a.) consists in the three guitars (bass‐guitar included) being independent enough that we hear three simultaneous melodies.
2. Then (b.) is satisfied given that, although the bass is largely independent of the other two voices — thus lending an aspect of true counterpoint — the two other guitars are not: the high‐pitch gestures of guitars 1 & 2 happen simultaneously, producing clear similar motion — and indeed, the two melodies have one & the same underlying contour, as discussed above. Still, they have plenty of oblique motion as well.
3. All voices have the same tempo and (again, especially in the non‐bass guitars) basically identical phrase boundaries, so (c.) is satisfied.
4. The freedom of (d.) is evinced by the melodic use of virtually every interval (at least, ignoring quality) from P1 through P8.

Traditional texture types do not observe unpitched material. It’s nevertheless worth noting that, owing to a rock‐music heritage, drumkit is frequently a meaningful part of the texture — as in this case. Perhaps this is relevant to (c.).

As a result of (L1d.), polyphonic melodic intervals should be treated slightly differently in music that makes use of home­ophony. In particular, we’ll be using normalized motion, for which you may refer to the relevant glossary entry.

In addition to its natural theoretic underpinnings, normalized motion makes voice‐leading relationships in home­ophon­ic textures clearer.

Throughout this essay, normalized motion will generally be the default. Un­normal­ized motions may also be provided for comparison, and in cases where individual motions are considered, those that are changed in any way by normalization will be prefixed with ⇕.

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

Because this section is primarily melodic (in particular, primarily home­ophon­ic), the pitch‐material is best described as simply “F♯ minor”. The centricity is pretty clear (making this not A major, nor any other mode), although there is just one nondiatonic pitch: A♯ in the bass. We might try to understand the A♯ by appeal to traditional harmonic analysis, but this turns out to be of limited utility:

As analyzed in Table 2, none of the harmonic progressions here are necessarily classically “invalid”, with the exception of a conspicuously “missing” V to connect the i⁴₃ at the end to the tonic at the beginning, which would produce a i⁶₄–V–i progression typical of classical minor harmony.

But none of this really explains the A♯, and more worryingly, doesn’t explain how the pitch‐material here still manages to feel like it has meaningful tonal movement, or pull, from start to end — and, importantly, from end back to start.

Although there’s no way to squeeze any kind of dominant chord out of the end of ⟪A⟫, I’d argue that what I’ve analyzed as a “♭13” acts as a semitonal upper neighbor to the 5^th factor of the tonic triad (or tonic seventh‐chord, in this case). This is particularly significant due to the prominence of C♯ in mm. 1–2. Moreover, the simple fact of the 5̂ being in the bass already has significant harmonic implications on its own — even though this is no classical cadence.

We hear the I⁶ in a place where we might classically expect i⁶, making this classifiable as “modal mixture” (in particular, from the parallel major). But given the apparently “dominant” functional intent, this is roughly as far as you could get from a tierce picarde — and I just generally don’t think this to be an adequate description. Instead, the effect that we hear comes from two sources:

• The A♯ is the only pitch that we hear that isn’t diatonic to F♯ minor, thus making it a source of some dissonance already.
• The intended harmonic progression is that of M3‐related major triads, whose efficient voice‐leading is ensured by the near‐evenness of major triads^[Tym11]: I–♭VI. The alteration from i to I is what gives this progression its “bite”: rather than there being a common tone of A♮, we have A♯ as a chromatic upper neighbor that resolves downward strongly by semitone.

So, perhaps unsurprisingly, the behavior of the pitch‐material in the home­ophon­ic ⟪A⟫ section is driven by voice‐leading. Let’s take a look at guitar 1 in particular:

It’s worth noting that none of the voices ever move by more than four semitones at a time; that is, they always move by two scale steps or less. More importantly, however, the sole instance of contrary motion (specifically within this part, that is) is at the boundary from m. 8 back to m. 1. This almost certainly lends much of the tonal effect of resolution back to the i⁷.

In fact, the “cadential” motion is ⟅c²⁣s⟆, which is the strongest possible motion between three voices^[10]. This contrasts strongly with the other motions noted in Table 3, which indeed get progressively weaker before arriving at the ⟅c²⁣s⟆.

I should remark that, although the traditional harmonic analysis is of limited use on its own, it does give us a taste of what is to come. We’ll later hear that Kidcrash’s pitch‐material language is unique enough that it requires nontraditional analytical frameworks, to be developed in later sections.

Mutually reïnforcing the harmonic movement are the dynamics of ⟪A⟫. We get a kind of crescendo during m. 4 & m. 8 (as indicated in Fig. 3), followed by a sudden drop back to the usual 𝆐𝆑.

Although we’ve yet to arrive at distorted guitars, Ciro Scotto’s notion of dist‐space is already relevant here:

[T]he pink noise spectrum of the cymbals underscores, supports, and imitates the sound of the higher harmonics of the distortion sounds found at the upper end of dist‐space.

In this case, it’s primarily the drummer’s switch to strong, continuous hits of the open hi‐hat — as opposed to the rest of the section’s soft hits of closed hi‐hat, interspersed with similarly soft & brief bursts of open hi‐hat — that transiently swell the song’s dist‐pitch, producing some intra­sectional dynamism.

The implication is a buildup to something — along with increased dissonance in want of resolution, further propelling the pitch‐material. But that “something” only really arrives in the next section.

The ⟪A⟫ section is a good example of a structure that we’ll hear throughout the piece: a period‐based bifurcating hyper­metric structure that I will refer to as periodic structure throughout this essay.

The use of bifurcating methods of timing organization is so common in music that it’s built into Western music notation: the whole‐note ⟨⁠𝅝⁠⟩ divides into two half‐notes ⟨⁠𝅗𝅥𝅗𝅥⁠⟩, which each divide into two quarter‐notes ⟨⁠𝅘𝅥𝅘𝅥𝅘𝅥𝅘𝅥⁠⟩, and so on. At the hyper­metric level, measures are commonly grouped into powers of two: groups of 4 measures, 8 measures, etc..

But the bifurcation found throughout “Turtlelephant” is particularly systematic, frequently having multiple nested layers of hyper­metric bifurcation (reminiscent of a perfect binary tree), with a clearly‐articulated relationship between the two “legs” of each bifurcation. This allows for a high degree of internal coherence within each section, even when the section is otherwise complex.

In the case of ⟪A⟫, its structure is outlined in Table 4:

Because of the typically deep nesting, I will be stretching the term hyper­beat to include groups of one or sometimes more measures, so that the hyper­measure (AKA hyper­bar) can sit at a higher level. Then, hyper­beats are numbered “h.b”, where h is the hyper­measure number, and b is the position within that hyper­measure.

This section is constructed via bisection (or conversely, doubling): the special case of bifurcation where the two legs are symmetric. Moreover, this bisection continues even at & below the metric level!:

1. each hyper­beat is bisected into two measures of 4⁄4,
2. each of which is bisected into an ascending half & a descending half,
3. each of which is bisected into two distinct notes,
4. each of which is bisected into a sequence of two identical eighth‐notes.

The result is thus extremely regular, and more straightforward than some of the more complicated examples that we’ll hear later in the piece.

The organization of ⟪A⟫ into hyper­beats of 4⁄4+4⁄4 is made obvious by the non‐bass guitar parts, where each chord takes up two measures each. The periodic structure being eight measures in extent is also obvious, because the whole thing repeats at that point. But it’s worth pointing out how the four‐measure hyper­measure is also clearly articulated:

• At every hyper­measure transition, we have strong voice‐leading that resolves some kind of dissonance: from m. 4 → m. 5, the I–♭VI discussed above; and from m. 8 → m. 1, the ⟅c²⁣s⟆ motion (i♭13⁴₃–i) also discussed above.
• The bass­line is clearly divided at the hyper­bar­line, as made explicit by the reduction in Fig. 4.
• The drumkit part is also clearly divided at the hyper­bar­line, anticipating each such line with the crescendo discussed in “Dynamics”.

Perhaps most notably, the repeated eight‐measure grouping is not just a period in name alone. The periodic structure in Table 4 is crucially reflected in the melodic contour — specifically, in responses that differ by second:

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

This section is very nearly a repeat of the previous section, but at a considerably higher dist‐pitch. Some of the dist‐pitch increase is from the non‐bass guitars now being massively distorted. The guitars in ⟪A⟫ were lightly overdriven — a bit of “dirt” on them, so to speak — but now they come in full force.

The distorted guitar timbre used throughout Jokes is perhaps jarring when contrasted with the (nearly‐)clean timbres used by the same instruments elsewhere in the same work. Kidcrash have no issue going from the shimmering, slightly‐jangly clean guitar tones that might be expected of, say, midwest emo, directly into screeching guitars that are fuzzy to the point of near‐inscrutability.

There is, however, a lot of nuance to be had here. Not only are the guitar timbres far from binary — as evidenced by, for example, the light distortion of ⟪A⟫ — but there’s frequently no direct correspondence between the dist‐pitch of the guitars and the dist‐pitch of the simultaneously‐playing drumkit. Furthermore, we might reasonably wonder: what’s up with the “screeching”?

Putting aside the complexity of the situation (e.g. how much “mids”^[11] did the amp have to begin with?) and the deeply subjective nature of what makes a timbre “good”, Kidcrash are not playing heavy metal, and the highly‐distorted guitar timbres on Jokes are not representative of “scooped tone” in the usual sense.

Nonetheless, taking a look at a spectrum visualization for ⟪A⟫ and for ⟪A′⟫ reveals that ⟪A′⟫ has a distinct bump in the 2 to 3⁢ kHz range, with a peak at ≈2.5⁢ kHz or so. This makes the range from roughly 500⁢ Hz to 2⁢ kHz sound comparatively “scooped”. This bump’s position is right around the nadir of the usual equal‐loudness contour, meaning that human auditory perception is particularly sensitive to this frequency range. This lends the “screeching” or “nasal” sound to the guitar distortion that is typical of the highest dist‐pitches of Jokes.

Another aspect of this same guitar timbre is its “fuzziness”. This is basically a way of saying that the distortion is so heavily clipped that the intelligibility of the result noticeably suffers. In some ways, this is an unfortunate artifact of the way that Jokes was produced. Still, the silver linings are that this album has a very distinctive sound, and that its highest dist‐pitches can have a special intensity to them.

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

The drumkit part does, however, differ from its ⟪A⟫ counterpart. The markèd use of a snare flam leads into m. 1 of ⟪A′⟫, which switches to a crash cymbal in place of the hi‐hat.

The intensity differences among cymbal hits leads to clearly‐audible gradations in dist‐pitch. As such, Fig. 6 notates three distinct basic levels: ride cymbal = 0_c, accented ride = 1_c, and crash = 2_c.

A somewhat interesting aural aspect of the cymbals is that the crash hits per se (at least, the 2_c ones) often fail to emerge from the mix, leaving only the wash of pink noise that trails them to inform the listener retrospectively that they were supposed to have just heard a crash hit.

Otherwise, this part is similar to that of ⟪A⟫, but with two exceptions. One is that the ride sometimes gets a sixteenth‐note gallop ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅯𝅘𝅥𝅯⁠⟩ on beat 3. More importantly, the final measure is quite different: the 3+3+2 division merely suggested by the snare placements of ⟪A⟫ is now elaborated into a fully‐fledged measure of (3+3+2)⁄8. By using snare flams, and by placing those flams on the onbeats (of (3+3+2)⁄8) and the crash‐&‐kickdrum hits on the offbeats, we get a fairly intense syncopation that forcefully propels the music to the next measure — and ultimately, the next section.

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

Like ⟪A⁣A′⟫, ⟪B⟫ is primarily home­ophon­ic. But befitting its position further into the piece, this texture is more fully developed. We now have four voices (sometimes five, when double‐stops occur) rather than three, enabled by guitar 2 producing a rhythmic pedal­point on F♯₃ via the strategic use of that open ③^rd string^[12]. We also have a greater variety of motion, although this is somewhat blurred by the fact that the highest three voices (i.e. guitars 1 & 2) are all in roughly the same register, leading to numerous voice‐crossings.

By briefly ignoring the pedal­point voice, we can statistically characterize the motion in ⟪B⟫:

Even considering classical motion types alone, there’s more than enough contrary & oblique motion to keep the three (or four, or five) voices clearly separate. The statistical method in Table 5 is consistent with, and relies solely upon, pairwise definitions of motion; but it tends to somewhat overrepresent oblique motion. If it helps, we can simplify further by dichotomizing motion into contrary & anti­contrary, ignoring oblique motion entirely; this gets a prevalence of 38.8% & 61.2% (a ratio of roughly 2∶3), respectively, which is an impressive showing indeed for a piece not written “as counterpoint” at all.

But at the same time, there’s likely more anti­contrary motion, and certainly more perfect parallel motion (6.6%), than we’d expect from a piece obeying classical contrapuntal principles. Moreover, the liberal use of voice‐crossing and large melodic intervals is also arguably inconsistent with classical principles.

The almost complete lack of phrasal independence among all voices further contributes to the lack of true contrapuntal texture. Nevertheless, the use of sometimes different rhythmic subdivisions in the different voices still produces a decent amount of rhythmic independence in general, and results in much of the observed oblique motion.

Just to be absolutely sure, we might want to make a direct comparison with an archetype of classical counterpoint. By picking a Bach chorale at random, we can do just that:

Bach is far less tolerant of perfect parallel motion, using only two occurrences (0.6%, or eleven­fold lower prevalence than in “Turtlelephant”’s ⟪B⟫; eight‐ or nine­fold lower when ignoring oblique motion) — both of which are parallel P4s. Bach also clearly prefers imperfect parallel motion, moreso than ⟪B⟫ — particularly in preference to similar motion. Almost all of these instances are parallel thirds or sixths, reflecting both Bach’s basis in tertian harmony, and — as we’ll later clearly hear — Kidcrash’s lack of exclusive basis in tertian harmony.

By making the same dichotomous simplification as before, we see that Bach also more strongly favors contrary motion: 51.4% contrary vs. 48.6% anti­contrary. This is a ratio of a little over 1∶1, which, when compared to “Turtlelephant”’s ⟪B⟫, gives an “odds ratio” (if you will) of 1.67.^[13]

In reality, appealing to statistics in this way is no more than a method of confirming one small part of what we already hear: the impression that this is not exactly counterpoint, in spite of what are ultimately simultaneous & texturally‐independent voices with no clear distinction between “main melody” and “accompaniment”.

This section represents the first real metric break from 𝄴 (ignoring my analysis of the intro as being 𝄵). In particular, we have a repeating three‐measure sequence of ⟨3⁄4, 3⁄4, 4⁄4⟩, for a distinctly irregular hyper­meter of (3+3+4)⁄4 = 10⁄4.

Although these are apparently three‐measure groupings of 3⁄4 with an “extra” beat at the end (= (3+3+3+1)⁄4), the end of ⟪B⟫ rereads this as actually being 3⁄4+3⁄4+2⁄4 with an “extra” 2⁄4 measure at the end to make (3+3+2+2)⁄4.

Whether the (hyper)meter is logically 3+3+3+1, or logically 3+3+2+2, is also disputed by the individual parts. In mm. 3 & 9, the drumkit uses snare flam & cymbal placements to suggest 3+3+3+1; yet in m. 6 (& naturally, m. 12 as well), these same elements are employed to suggest 3+3+2+2.

The division of each 3⁄4 is also conflicted, as the cymbals suggest 3⁄4, but the snare & guitars (bass included) suggest 6⁄8. And in the 4⁄4 measures, the F♯₃ pedal­point disagrees with the other voices by continuing the eighth‐note displacement, while the others play largely un­syncopated 𝄴.

Indeed, the continuation of the metric displacement by ±𝅘𝅥𝅮, as heard in ⟪A⁣A′⟫, is notable: throughout all of ⟪B⟫ — excepting the last three measures, unless you count the kick­drum hits — there’s always at least one voice that’s displaced in this way.

Although this section finally loses the *rururdr(d) contour of the intro & ⟪A⁣A′⟫, it basically retains both other important aspects of contour from previous sections:

• Guitars 1 & 2, at least during the 3⁄4 measures, continue with the “repeating each note twice in succession” part of the contour: they produce a ⟨⁠𝅘𝅥𝅘𝅥𝅘𝅥𝅮𝅘𝅥𝅮⁠⟩ sequence of notes of the same pitch. (Recall the metric displacement by eighth‐note, so the sequence starts on the “& of 1”.)

• We still get a “mobile” melodic prominence that self‐differs by second:

  • Guitar 1, m. 3 has B–C♯–B–C♯–E; m. 6 responds with A–B–A–B–D.^[14]
  • Guitar 1, beat 2 of m. 9 has E–F♯–G♯ (ignoring beat 3’s high {C♯, G♯} in the guitars, because it has no response), and m. 12 responds with D–E–F♯.
  • Guitar 2, m. 3 has a high C♯; m. 6 responds with a high D in the same voice.
  • (There presumably would be a fourth item in this list, but m. 12 is truncated.)

Given the self‐differing‐by‐second relationships, the hyper­meter discussed above, and the fact that guitar 2 plays a clearly different melody starting at m. 7, we can easily assemble the periodic structure of ⟪B⟫:

Indeed, if we consider mm. 7–12 to be a response to mm. 1–6, then we have something like a “hyper­period” spanning — hypothetically — a total duration of 40⁄4.

However, m. 12 is clearly truncated, leaving the second period of ⟪B⟫ slightly fragmentary. This tactic of truncating what is otherwise clearly a period, in order to gracefully lead into another section or transition, is used throughout the piece. By fraying the end of a period, it becomes less clear where the period’s material stops and the new material starts, so that sections are blurred together at the edges, rather than having a crisp & abrupt end‐to‐start junction (the musical equivalent of a visible seam).

From the very high dist‐pitch of ⟪A′⟫, we enter ⟪B⟫ with an abrupt 𝆍𝆏. However, the dynamics of this section’s parts are not all in unison. The four pitched voices vary significantly in their relative dynamic levels as they each fluctuate throughout the section.

Moreover, although the distortion of the guitars is all but removed, the drumkit continues at a surprisingly high dist‐pitch, maintaining a mostly constant wash of pink noise from the cymbals.

This section remains firmly in the key of F♯ minor, although this time, the pitch‐material is entirely restricted to the seven pcs of F♯ natural minor.

Nonetheless, there’s a surprising amount of vertical dissonance. A good chunk of this section involves voices harmonizing in seconds! There are also a few explicit vertical statements of the tri­tone {G♯, D}, which is also heard in ⟪A⁣A′⟫ in the form of guitar 1 arp­eg­gi­at­ing [C♯, D, G♯] ∈ ⁅0⁣5⁣6⁆_3‐5B.

Again, this section is melodic and focuses on its home­ophony. Thus, traditional harmonic analysis is not particularly useful, and would have to treat this section as “modal” anyway.

Indeed, similar to both the intro & the ⟪A⁣A′⟫ sections, we observe a penchant for dense diatonic fragments — diatonic clusters, if you will^[15]. Not only do we hear plenty of harmonization — sometimes at multiple levels — in seconds, but for instance, throughout beat 3 of m. 6, we hear all five members of ⁅0⁣2⁣3⁣5⁣7⁆_5‐23A simultaneously — an entire penta­chord, all at once!

The idea of “diatonic clusters” may sound preposterous to the reader who’s accustomed to using the word cluster exclusively for intervals that are fully chromatically “filled in” (i.e. every pitch within the interval’s bounds that’s also within the tuning system is sounded), most notably ⟦0⁣1⁣2⟧_3‐1.

But for almost any scale of sufficiently large card­i­nal­i­ty, there’s no particular reason why clusters cannot be relative to that scale^[7]; and indeed, the geometric notion of “evenness” in [Tym11] provides a way to understand generalized clusters: by minimizing evenness (i.e. by clustering), it’s possible to move from chord to chord (or sonority to sonority) with efficient voice‐leading, but without dramatically changing the shape of those chords.^[16]

And really, much of the effect is even simpler than that: use of generalized clusters turns voice‐crossing, & small overall harmonic movements, into perfectly commonplace features of the texture.

Thus, the use of diatonic clusters & harmonization in seconds — among other things, to be sure — arguably allows Kidcrash to write fundamentally melodic (read: horizontal) home­ophon­ic parts that, in spite of largely defying classical harmony and classical contrapuntal principles, manage to sound coherent & tonal.

We’ll hear plenty of intent use of diatonic clusters throughout the piece, but it’s worth emphasizing the degree to which they’re used as chords or sonorities in their own right, rather than as merely “passing”.

With just four (occasionally five) voices, we can require a cluster to have at least three distinct pcs, and call a sonority “cluster­like” if it’s either a cluster, or a cluster with one added pc. Then:

• Measures 1–3 are 40% cluster­like (65%, if a bare ⟦0⁣2⟧_2‐2 is permitted).
• Measures 4–6 are 40% cluster­like (65%, respectively).
• Measures 7–9 are 35% cluster­like.
• Measures 10–12 are 19% cluster­like.

The cluster­like sonorities are typically just plain diatonic clusters; and naturally, the others are not analyzable — at least, not so easily — as other types of sonorities. The cluster species heard in ⟪B⟫ are as follows (*the asterisked cluster only occurs with an added pc):

• *⁅0⁣2⁣3⁆_3‐2B.
• ⟦0⁣2⁣4⟧_3‐6.
• ⁅0⁣1⁣3⁣5⁆_4‐11A.
• ⟦0⁣2⁣3⁣5⟧_4‐10.
• ⟦0⁣2⁣4⁣6⟧_4‐21.
• ⁅0⁣2⁣3⁣5⁣7⁆_5‐23A.

And we should also note that most of the non‐cluster­like sonorities in mm. 1–12 are still non­tertian, with many of them being stacked fourths. This brings us to mm. 13–15.

The three chords in mm. 13–15 are all the same pc‐wise, but differ in the ascending voicings in guitar 1 that produce the melodic progression A–B–E. The pcs are {F♯, B, E, A} ∈ ⟦0⁣2⁣5⁣7⟧_4‐23, with E in the bass. This is plainly a quartal chord — in integer notation: [0, 5, 10, 15] — and is emphatically voiced as such (both non‐bass guitars play in P4s). We’ll hear this chord importantly resurface later on in the piece.

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

⟪C⟫ begins with changes in both tempo & time signature: an increase of roughly 10⁢ BPM from 𝅘𝅥 ≈ 147 to 𝅘𝅥 ≈ 157, and a switch to 6⁄8.

The switch to 𝄵 for ⟪D⟫ recalls the use of the same rhythm & meter in the intro (Fig. 2).

In keeping with ⟪A⁣A′⁣B⟫, ⟪C⟫ continues developing the home­ophon­ic texture, while retaining the F♯₃ pedal­point.

For textural purposes, we can think of the nigh‐uninterrupted stream of eighth‐notes in all three guitar parts as being like tremolos of (typically) longer notes. This makes it clear that what we have here is not really homo­rhythm at all, but instead, genuinely rhythmically‐distinct melodies in three voices (plus a fourth pedal­point voice). But is it really home­ophony, or is it simply counterpoint?

For direct comparison with ⟪B⟫, we can perform the same statistical analysis:

The contrary vs. anticontrary dichotomy comes out to 47.1% vs. 52.9% (a ratio of roughly 8∶9). Overall, the normalized statistics look quite favorable to hearing the texture as contrapuntal.

On the other hand, the stark contrast with the un­normal­ized statistics hints at a less‐than‐counter­point­like feature of the texture: resolution of dissonances via rather large leaps. Moreover, we might consider a 5.1% incidence of perfect parallel motion to be a bit high for true counterpoint.

Furthermore, statistical analysis of motion is only one small consideration. The relative lack of phrasal independence (phrasal boundaries are generally at the 6⁄8 bar­line in all voices) also nudges the texture closer to home­ophony. Normalization of motion is necessary to uncover the structurally‐important (as we’ll see) polyphonic motions, further suggesting that the texture is heard as home­ophon­ic.

Naturally, we must — as always — acknowledge that there are no hard boundaries between texture types, as they’re merely useful abstract categories that can (& should) sometimes overlap. In this case, the essential aspects of the texture are as follows:

The textural considerations for ⟪D⟫ are quite similar to those for ⟪C⟫, with one major exception: we have true homo­rhythm. This is important to note partly because it highlights the fact that homo­rhythm overlaps with other textural categories — especially homo­phony.

In this case, guitar 2 seems to be carrying the primary melody; however, given that guitar 1 begins in unison with guitar 2, the contrast between the paired melodies might be heard as home­ophon­ic or similar. Nonetheless, the texture is simple enough that it’s primarily heard as merely homo­rhythmic.

More than anything else, this use of homo­rhythm produces a dramatic effect, giving ⟪D⟫ a solemn character. This dramatic effect is intensified when the bass‐guitar & drumkit drop out, leaving only the two non‐bass guitars to form the texture.

In the “Pitch‐material” discussion for ⟪A⟫, I noted that within the eminent part of that section (that is, guitar 1’s), the sole instance of contrary motion was at the boundary between one period & the next. This essentially melodic (read: voice‐leading) gesture took on the largest part of what would more traditionally — that is, in most Western musical traditions — be the role of cadential harmonic relationships (dominant–tonic, etc.).

As we’ll see, like basically every other section of the piece, ⟪C⟫ has a fundamentally periodic structure of variations. As part of this structure, it features a regular pattern of contrary motions: always across the barline between the 1^st & 2^nd measures of each hyper­measure (see Table 14), and more importantly, always across the barline between hyper­measures.

These patterns of motion do at least as much work as the pitch‐material does to punctuate — basically caden­tial­ly — the structure of this section, analogous to the case of ⟪A⁣A′⟫. However, it’s also worth noting that the patterns require normalized motion, suggesting that the listener must hear ⟪C⟫ as home­ophon­ic on at least some level to get the intended effect.

Still more significantly, the ⟪C⟫ → ⟪D⟫ transition is of great structural import for two reasons:

• It’s the first time that we get something close to a reprise — and of the very beginning of the piece, at that.
• Said petite reprise anticipates the structural boundary between the initial intro–⟪A⁣A′⁣B⁣C⁣D⟫ sequence and the first time that we lose the F♯₃ pedal­point (which will be ⟪E⟫).

It’s thus particularly appropriate that exactly this transition evinces the strongest possible contrary motion between three voices^[10] (the same motion as at the end of ⟪A⁣A′⟫’s periods): ⟅c²⁣s⟆.^[9] (Moreover, the result is unison between the voices.)

This is perhaps part of the power of home­ophony: by plainly lacking general adherence to classical contrapuntal norms, this texture can make particularly strong polyphonic motion a markèd gesture, allowing it to buttress points of structural import.

A composer like Bach — and especially those after him — has relatively clear harmonic norms to uphold, and not infrequently uses quite weak polyphonic motion to achieve strong — & typically stereotyped — harmonic movement. Rather than necessarily being based upon voice‐leading as evinced at the surface level, harmonic norms can be based upon voice‐leading at a more abstract level.^[Tym11]

The home­ophon­ic approach, then, perhaps looks “primitive” by comparison: with more skeletal harmonic norms, the function of melody qua melody can be fore­front­ed, making the surface‐level voice‐leading — and of course, the melodies themselves! — the subjects of interest.

That being said, out of all sections so far, ⟪C⟫ is perhaps the most amenable to squeezing a traditional harmonic analysis out of it:

This gloss looks enticing: we start at i, oscillate between iv & i⁶ (“tonic prolongation”, minor plagal cadences, etc.), and then do a perfectly tonal (♭VI–)iv–ii°–V–i! Case closed…

There are some problems, however:

• We sometimes get quite clearly‐stated chord progressions, but when we do, they seem functionally aimless. For instance, mm. 4 & 12 give us an explicit iv–♭VI; but this seems like it’s going in reverse (we expect ♭VI–iv).

• Other times, we don’t get clearly‐stated chord progressions, and we have to squeeze a little harder:

  • The “G♯𝆩no3” in m. 8 is mysteriously missing its third factor.
  • The otherwise clearly‐stated Bsus2 at the end of mm. 4 & 12 doesn’t meaningfully resolve its suspension — nor does it really resolve in any way — in spite of the D sounding, in a different voice, in the next measure.
  • The “C♯5” at the end of m. 8 is a power‐chord at best, although even then, the F♯ sounding in the pedal­point confuses this analysis.
  • What’s going on in mm. 5 & 13? Hello‽

Most importantly, we’re missing the point (or at least half of it) with this gloss. What follow are some observations that are more interesting, and which can nonetheless be usefully complemented by the traditional harmonic gloss.

In addition to clearly being driven by (roughly) home­ophon­ic voice‐leading, the sonorities here are sensitive more to their raw intervallic content than they are to pre­ëxist­ing harmonic norms. (See also: the “Generalized clustering” section above.)

In this case, chords (in the generalized sense) whose ic‐vectors are 0 in both the first & last positions — in effect, the most dissonant positions — are used to represent “normal” (Ⓣ, Ⓢ, Ⓟ) functions, allowing the other chords to be used in more “dominant” (not necessarily in the classical sense) ways.

We typically have three (maybe four, pedal­point included) voices here. In such a case, Kidcrash is left with basically a half‐dozen‐ish “normal” chord types:

Indeed, the only chord in Table 10 that isn’t used here is the augmented triad, on account of it not existing within the diatonic scale. Already, simply by appealing to ic‐vectors, we can explain the things that traditional harmonic analysis cannot:

• This explains the bias in seventh‐chords (as opposed to triads) towards the minor variety: major seventh‐chords have ic 1 between 7^th factor & root, dominants ic 6 between 3^rd & 7^th factors, and (half‐)diminished ic 6 between root & 5^th factor.
• ⟦0⁣2⁣7⟧_3‐9 is symmetric & versatile (it has three distinct interpretations in Table 10), making it less surprising that it’s heard often in ⟪C⟫ — including in various instances elided from Table 9 for ease of analysis.
• Seemingly “aimless” chord progressions arise naturally as the result of these vertical constraints combined with the constraints of home­ophony.
• Measures 5 & 13 are significantly less of a mystery.

The other side of the same coin consists in those chords whose ic‐vectors have nonzero first &/or last positions — particularly the diatonic ones:

Once again, many of the chords here are now ascribed “functions” that they classically wouldn’t generally have: diatonic clusters become useful voice‐leading sonorities in their own right, quartal constructions (⟦0⁣2⁣7⟧_3‐9, ⁅0⁣1⁣6⁆_3‐5A, etc.) have clear implications, major & minor triads needn’t fulfil their classical functions, half‐diminished chords don’t actually need their otherwise‐crucial 3^rd factor, and so on.

When done carefully, as Kidcrash do in this piece, this does not have anything to do with “pan­dia­tonic­ism”, nor anything of the sort. Instead, judicious — but somewhat adventurous — voice‐leading, careful attention to interval­lic content, structural use of contrapuntal motion, & nods toward Western classical or popular‐music tonality, all combine to produce a distinctive sound that can reasonably be described as — & is heard by the casual listener as — a kind of tonality.

A slightly murkier — but still important — aspect of ic‐sensitivity in Kidcrash’s music is the “other” type of harmonic function: subdominant (AKA pre‐dominant), denoted Ⓢ. Often, use of this function borrows somewhat from traditional tonality, so that a chord analyzable as something like iv, IV, ii, etc. may be expected to function as Ⓢ. However, there are numerous important considerations:

In effect, ic 2 plays a quasi‐special role in ic‐sensitivity, because of its association with the clearly‐markèd ic 1: both typically have the same quantity, and differ only in quality. Nonetheless, the presence of ic 2 doesn’t always preclude Ⓣ function — the obvious example being i⁷. One consequence is that quartal constructions that lack both ic 1 & ic 6 generally have Ⓢ (or sometimes Ⓟ) function.

But we still have one crucial bit of ⟪C⟫ left to explain: m. 8 and its transition to m. 9. The traditional gloss in Table 9 analyzes this as a “ii°–V–i” of sorts; this is useful as a nod to traditional harmony, but the truth is in the voice‐leading.

Admittedly, we have to briefly cross the upper two voices to make this fully work, but said voices are playing in very similar registers anyway:

As illustrated in Table 12, the procedure is like so:

1. Move to the ic 6 between {D, G♯} (represented by “ii𝆩no3⁷” in the traditional analysis), invoking Ⓓ function.
2. Resolve this dissonance to the consonant ic 5 between {C♯, G♯}, by oblique motion (this produces the ⁅0⁣1⁣5⁣7⁆_4‐16A in passing, or *“♭VI∆♯11⁴₂”).
3. Treat this momentary consonance as a “dissonance” to be resolved by strong structural voice‐leading ⟅c²⁣p⟆ from a P5 to a M3.

Note that this “strongly resolves” C♯ to D, which would traditionally not make sense in the context of F♯ minor. This double chain of resolutions can be analyzed as “ii°–V–i”, but only if you squint quite hard.

In fact, this is fundamentally the same procedure as the one used to go from one period to the next in ⟪A⁣A′⟫:

1. ⟪A⁣A′⟫’s mm. 5–6 (specifically, the guitar 1 part) presents the {D, G♯} in the form of ⁅0⁣5⁣6⁆_3‐5B (a “Viennese trichord”); this is analogous to the former half of ⟪C⟫’s m. 8.
2. This ic 6 is resolved to ic 5, albeit a different one: {E, A}.
3. The resulting ⟦0⁣2⁣7⟧_3‐9 is then “really resolved” by strong structural voice‐leading back to the beginning of the period; this ⟦0⁣2⁣7⟧_3‐9 is [D, E, A] in the case of ⟪A⁣A′⟫, and [C♯, F♯, G♯] in the case of ⟪C⟫.

The fact that we observe the same harmonic & voice‐leading language — movement to ic 6 → half‐resolution to ⟦0⁣2⁣7⟧_3‐9 → nontraditional resolution powered by strong structural voice‐leading — as multiple occurrences in distinct sections indicates not only that said sections have shared thematic material, but also that Kidcrash are making use of a relatively unique, but genuine, tonal language.

Apart from the intro, the only pc that we’ve heard so far that’s not a member of F♯ natural minor is the A♯ (the major ♯3̂) in the bass‐guitar part of ⟪A⁣A′⟫. Moreover, although the intro did not contain any As (sharp or otherwise), it used a diatonic cluster (⁅0⁣2⁣4⁣5⁆_4‐11B) to imply an F♯ major mode.

In the ⟪D⟫ section (m. 16 ff. of Fig. 8), in addition to giving a petite reprise of the intro in various other ways (as discussed above & below), we finally get a very explicit statement of F♯ major — albeit not so much of the mode, as of the chord.

Indeed, much like in ⟪A⁣A′⟫, we only get A♯. Although this A♯ is clearly & consistently used to produce an F♯ major chord, the meticulous insistence upon this being the only mobile pitchclass very nearly suggests F♯ Aeolian dominant (the 5^th mode of B melodic minor, or 2^nd mode of E acoustic); or alternatively, F♯ melodic major descending.

In any case, the F♯ parallelism is clear — be it the “major”, the “melodic major”, or just the major chord. And speaking of chords…

Once again, the traditional harmonic analysis is of limited utility. We can interpret the first (but not so much second) I chord as a tierce picarde, making m. 15 → m. 16 a minor plagal cadence. Indeed, this is a fortuitous alignment between strong structural voice‐leading, and the stereotyped version of this voice‐leading that’s present in classical harmony.

But even putting aside the inverted power‐chord, the first problem is the ♭VI^∆7–I progression (mm. 19–21), both in root position. The converse of this basic progression is also heard in ⟪A⁣A′⟫, where I described it in terms of voice‐leading, rather than as a tierce picarde or “chromatic mediant”.

Once again, I think that such an analysis is the strongest. In this case, the pedal­point is the only thing supplying the 3^rd factor of the ♭VI^∆7, and the 5^th is omitted; this leaves only the 7^th & root — a m2 apart — as the real thrust of the ♭VI^∆7. In combination with ic‐sensitivity, this paints the major seventh‐chord as quite the dissonance — indeed, if we count the floor‐tom, this even produces a diatonic cluster ⁅0⁣2⁣3⁆_3‐2B. As a result, we have a dissonance followed by a consonance, with strong voice‐leading to connect them.

Then we have mm. 22–23. Although the 5^th factor of the ii⁴₂ is omitted, I think it’s fair to say that its position immediately following the I makes it likely that the listener hears it as a minor (rather than half‐diminished) seventh‐chord. And by m. 23, we seem to have the implication of another root‐&‐7^th‐factor combo — but with no 5^th factor, nor even 3^rd! The bare M2 simply moves directly as [E, F♯, E] $\underset{⟅{\mathrm{o}}^2\mathrm{s}⟆}{\overset{\mathrm{-2},0,\mathrm{-3}}{\to }}$ [D, F♯, C♯]^[17]. Analyzing this dyad as a shard of a seventh‐chord only produces the less‐than‐useful progression *“I–ii–i”.

I hear this motion as intentionally dissonant, and as evoking the pitch‐material of the intro. By moving directly & suddenly from {E, F♯} to {C♯, D}, we hear a parallel minor (or melodic major, if you prefer) version of the intro’s diatonic cluster tetrachord: {C♯, D, E, F♯} ∈ ⁅0⁣1⁣3⁣5⁆_4‐11A. Compare this to the intro’s {C♯, D♯, E♯, F♯} ∈ ⁅0⁣2⁣4⁣5⁆_4‐11B. The fact that they also happen to be inversion­al­ly equivalent is just icing on the cake.

This motion between two halves of what is really a larger diatonic cluster partially explains the somewhat awkward voice‐leading (although there is, at the very least, no parallel motion of any kind).

In the intro & ⟪A⁣A′⟫, the basic sub­phrasal unit was a pair of two identical notes: ⟨⁠𝅗𝅥𝅗𝅥⁠⟩ in the intro, and ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅮⁠⟩ in ⟪A⁣A′⟫. In ⟪B⟫, this became four notes ⟨⁠𝅘𝅥𝅘𝅥⁠⟩ + ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅮⁠⟩ = ⟨⁠𝅘𝅥𝅘𝅥𝅘𝅥𝅮𝅘𝅥𝅮⁠⟩, albeit not all of the same duration anymore. With ⟪C⟫, we’ve now basically ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅮⁠⟩ + ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅮⁠⟩ + ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅮⁠⟩ = ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅮𝅘𝅥𝅮𝅘𝅥𝅮𝅘𝅥𝅮𝅘𝅥𝅮⁠⟩.

The use of repeated notes — “repeated” in the sense of being the same pitch, but also the same duration in most cases — lends a certain coherence to these sections as they go by. But the gradual intensification of the aforesaid rhythmic units — in combination with the tempo increase at the ⟪B⁣C⟫ boundary — produces what Stephen Hudson calls structural acceleration (albeit not in the exact same way as in his example).^[Hud20]

This structural acceleration comes to a head not with a breakdown, nor with a chorus, but instead with a return to the intro’s rhythmic units in ⟪D⟫.

The reprise of the ⟨⁠𝅗𝅥𝅗𝅥⁠⟩ rhythmic unit also brings with it the innovation of ⟨⁠𝅗𝅥𝅘𝅥𝅘𝅥⁠⟩. These two rhythmic units alternate, but fundamentally, the latter is just the former with a bisected latter note.

Like ⟪B⟫, ⟪C⟫ apparently lacks the *rururdr(d) contour of the intro & ⟪A⁣A′⟫, fragmenting it to just *r; yet, also like ⟪B⟫, it manages to retain the “‘mobile’ melodic prominence that self‐differs by second” part of the contour:

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

This self‐differing by second is indeed periodic, thus making the full periodic structure of ⟪C⟫ evident:

Although the self‐differing by second isn’t present in ⟪D⟫, its periodic structure is still evident from how its latter half responds to its former half:

And as a contour‐related sidenote, the use of A♯ in ⟪D⟫ allows for tight chromatic melodic turns like A♮–B–A♯ ⟨0⁣2⁣1⟩ (guitar 2, mm. 15–16) and A♯–B–A♮ ⟨1⁣2⁣0⟩ (guitar 1, mm. 17–18).

⟪C⟫ presents a crescendo of dist‐pitch in the drumkit part, led by the hi‐hats, during mm. 5–8. This ultimately culminates in a sudden drop from 𝆑 to 𝆏 or so in the latter half of ⟪D⟫ (m. 21 ff.).

This marks the first time that we’ve been at such a soft dynamic level & dist‐pitch since the intro (partly due to the drumkit part in ⟪B⟫). It’s also a return to the intro’s guitar‐only instrumentation, as the bass‐guitar & drumkit fall completely silent.

I’ve been making claims that ⟪D⟫ is a petite reprise of the intro: not recapitulating the intro material exactly, but instead reproducing most of its important thematic material, thus giving the listener the impression — at the very least — of “return”:

• First return of 𝄵 since the intro; first re­üse of ⟨⁠𝅗𝅥𝅗𝅥⁠⟩ as the basic rhythmic unit.
• First particularly prominent use of A♯ to explicitly produce F♯ major (triad, or melodic major), akin to the intro’s implication of the F♯ major mode.
• The intro’s {C♯, D♯, E♯, F♯} cluster returns in its “descending” (in terms of melodic major) form {C♯, D, E, F♯}; both clusters are members of ⟦0⁣1⁣3⁣5⟧_4‐11.
• First return of a guitar‐only (bass‐guitar not included) instrumentation.
• First return to a 𝆏 or 𝆏𝆏 dynamic, and to overall dist‐pitch 0_c.

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

As expected, this section has a clear periodic structure. This structure is defined not only by melodic contour, but also by texture & register.

We begin (mm. 1–12) with alternation between contrasting halves, the first of which is relatively quieter. The quieter half is always a single measure of 6⁄8, but the louder half is either five quarter‐notes’ duration (written in Fig. 10 as ⟨6⁄8, 2⁄4⟩), or else six (written ⟨6⁄8, 6⁄8⟩).

The response given by the latter half of the period is somewhat subtle, but as expected, consists in melodic self‐differing by second: in mm. 5–6, guitar 2 plays G♯–E, whereas in mm. 11–12, it plays G♯–F♯. The resultant periodic structure is shown in Table 16:

This extremely irregular hyper­meter of 17⁄4 (note that 17 is prime) can be heard as 18⁄4 with a single beat “missing” mid‐hyper­measure, or equally well as 16⁄4 with an “extra” beat at the end.

The 6⁄8 measures are played as such, with the exception of the crash & snare in mm. 5, 6, 11, & 12. This allows (or perhaps even requires) Kidcrash to play the ⟨6⁄8, 2⁄4⟩ sequences as ambiguously (3+3+2+2)⁄8 or (3+3+3+1)⁄8. This is just the same ambiguity as in ⟪B⟫, discussed in “Rhythm & meter” above.

And naturally, the latter half of ⟪E⟫ has a similarly periodic structure, with the latter hyper­measure (mm. 17–20) contrasting with the former (mm. 13–16) in the pitches used by guitar 2 & the bass‐guitar.

The basic rhythmic unit of this section is clearly ⟨⁠𝅘𝅥𝅮𝅘𝅥⁠⟩. The main variants that we hear are augmenting the second note during 2⁄4 measures to ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅭⁠⟩, and bisecting the first note to produce ⟨⁠𝅘𝅥𝅯𝅘𝅥𝅯𝅘𝅥⁠⟩.

I hear this section as making a break from all previous sections in various ways (several discussed above & below). One of these ways is that any structural acceleration (or deceleration) due to “intensifying” rhythmic units is no longer relevant; instead, any acceleration in this section is due to the more obvious fact of the tempo increasing from 𝅘𝅥 ≈ 157 to 𝅘𝅥 ≈ 180.

This break is perhaps to be expected, given that we’ve just hit the point of structural importance that is the petite reprise ⟪D⟫.

Although the rhythmic unit ⟨⁠𝅘𝅥𝅮𝅘𝅥⁠⟩ (& its variants) sometimes represents two notes (or chords) of the same pitch, the participation in a *r‐based melodic contour is not evident — at least, not immediately.

In ⟪E⟫’s latter half (m. 13 ff.), however, the bass­line becomes arguably the most active part of the texture (excepting possibly guitar 2). In this capacity, it repeatedly presents ⟨1⁣1⁣0⁣0⁣2⁣2⁣0⁣0⟩_c, which is similar — save for the swapping of 0_c with 1_c — to the ⟨0⁣0⁣1⁣1⁣2⁣2⁣1⁣1⟩_c of the intro & ⟪A⁣A′⟫. If we normalize the motion, however, we instead hear ⟨2⁣2⁣1⁣1⁣0⁣0⁣1⁣1⟩_c, which is straightforwardly an inversion of the ⟨0⁣0⁣1⁣1⁣2⁣2⁣1⁣1⟩_c.

This section represents the second ascent to a very high dist‐pitch & general dynamic intensity (after ⟪A′⟫). But it goes even further in some ways: we get more dense crash hits — especially in mm. 2 & 8 — than ever previously heard, and we get full‐blown block‐chords from both guitar parts, thus producing even greater inter­modulation distortion.

The dynamics of ⟪E⟫ are tied to its texture: during the measures of counterpoint between guitars 1 & 2 (with no bass‐guitar present), we get a lower dist‐pitch & overall dynamic level in the drumkit part as well; during other measures, the bass‐guitar plays, the guitars produce chords, and we get crash cymbal activity.

The ability of ⟪E⟫ to oscillate rapidly between these two levels — rather than, for example, remaining at the highest dist‐pitch throughout — contributes to its higher perceived intensity & forward momentum.

The texture of this section is notable for being the first real break (since the intro) from being primarily home­ophony‐driven. In a way, the oscillating texture — and accompanying dynamics — are a kind of compromise between the home­ophony of past sections & the textures more standardly expected of post‐hardcore.

In particular, the entirety of ⟪E⟫ manages to retain the “duel­ling guitars” approach of the home­ophony in previous sections, even in cases where one or both non‐bass guitar parts are essentially just slamming block‐chords. The only genuine loss of independence is in mm. 2, 3, 8, & 9, during which the parts are identical.

The texture in mm. 1, 4, 7, & 10, while brief, is fairly contrapuntal, with all motion being strictly contrary. Although the texture in mm. 5, 6, 11, & 12 is similarly brief, and exhibits relatively little motion, it is virtually homo­rhythmic. The motion in guitar 2 (& in the bass‐guitar, for that matter) cannot be ignored, as it introduces pcs otherwise foreign to the region.

And finally, m. 13 ff. is still basically homo­rhythmic, but is more well‐developed to the point that we get something like homo­phony: melody & accompaniment. But by normalizing motion, we hear some home­ophony between guitar 2 & the bass‐guitar here: a decent amount of motion­al independence (contrary & oblique motions, plus some similar), but a total lack of phrasal independence, and liberal use of large melodic intervals.

What’s more interesting, however, is how material from these various textures is recombined in m. 13 ff.: the originally contrapuntal melody is synthesized with the chords from originally monophonic & homo­rhythmic textures to forge the “completed” (& texturally mixed) form of this section.

With one slight exception, this section (like most previous sections) adheres to F♯ natural minor.

The brief contrapuntal textures are notable for outlining a root‐position A∆7 ≘ ♭III^∆7. Thus, the progression from m. 1 → m. 2 (& m. 7 → m. 8) can be interpreted as a ♭III–i progression typical of rock music, but with the atypical use of a major seventh‐chord. As discussed in “Intervalclass sensitivity” above, the ic 1 of major seventh‐chords allows them to “function” as dissonances (more specifically “Ⓓ”), even when this doesn’t align with traditional function.

Also notable is that the only vertical ic 1 or ic 6 (or even ic 2, for that matter) in this phrase is the m9 ≘ ic 1 between C♯ & D, which occurs at the very end. This is a more straightforward & traditional use of ic‐sensitivity as I’ve defined it here.

More interestingly, we have the [A, D, G♯] chord that features prominently throughout this section; this is a return of the ⟦0⁣1⁣6⟧_3‐5 (“Viennese trichord”) from ⟪A⁣A′⟫, but this time in a quartal voicing [0⁣5⁣↋] ∈ ⁅0⁣1⁣6⁆_3‐5A, instead of a close inverted voicing [C♯, D, G♯] ≘ [0⁣1⁣7] ∈ ⁅0⁣5⁣6⁆_3‐5B. Here we once again hear ic‐sensitivity in action: although this trichord has no traditional function, its possession of ics 1 and 6 allow it to function as a clear Ⓓ that contrasts with the tonic F♯5.

But this skeletal form of the “dominant” chord is also heard with more muscle on its bones. For the most part, we seem to get what we’d expect: arbitrary diatonic fragments. When E is added, we can squeeze an E11 ≘ ♭VII^7,11 (in various inversions) out of it; when B is added, we can squeeze a B−13 ≘ iv^7,13 out of it (see especially the latter half of m. 12); and perhaps most interestingly, when solely F♯ is added (as in the former halves of mm. 12, 16, & 19), we get ⁅0⁣4⁣6⁣7⁆_4‐Z29B — an all‐ic tetrachord.

Although the F♯ of the tonic chord is “only” avoided 75% of the time, it is notable that the C♯ of the tonic chord never occurs as an addition to the basic [A, D, G♯] chord. This is discounting the guitar 2 part of m. 13 ff. as being the melody, and thus not part of the accompaniment; but note that guitar 2’s phrases always end on a long (𝅘𝅥) D or A, thus maintaining the dissonance.

I suspect that the C♮ ≘ ♭5̂ in the guitar 2 part of m. 17 is a mistake, especially considering that it’s consistently C♯ elsewhere. Nonetheless, it fortunately at least seems to fill the role of avoiding C♯, and suggests the usual alteration that forms the so‐called (F♯) blues “scale”…

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

This is the first verse. Perhaps unsurprisingly, the texture is the simplest yet: some block‐chords with effectively a single vocal melody on top.

What really drives this texture, however, is the way in which its rhythms interlock. Although this entire section is in 𝄴, its four‐measure hyper­bars are characterized by a compressed m. 1 that gradually decompresses into the relatively un­syncopated mm. 3–4:

In short: we start by losing a whole beat, then we get half of it back, then we get the other half back, and then rinse & repeat.

Against this backdrop, the main point of rhythmic interest is the vocals, which make free use of roughly one measure’s worth of material at a time, typically subdivided into eighth‐notes. The freeness is apparent from the roughly even split between starting 0, 1, or 2 eighth‐notes before the barline; the variety of sometimes inconsistent rhythms used; and even the intentionally “behind the beat” phrase of m. 10.

That being said, if there’s any general rhythmic principle at work in the vocals, then it might be the tendency toward (3+2+3)⁄8. Of the eight vocal phrases, four of them use this metric division, and another two are consistent with it without explicitly stating it in full — in total, 75% of all vocal phrases.

A particular outlier is mm. 11–12 (Now I’m grinding my teeth), which strikes me as the most syncopated phrase in this entire section, in addition to being the only vocal phrase that makes use of its particular register (roughly between C♯₄ & D♯₄). The suggested division is pretty clearly (2+3+3)⁄8 (preceded by ⟨⁠𝅘𝅥𝅮𝅻𝅘𝅥𝅮⁠⟩; in effect, 2+2+3+3).

But the phrase of greatest import is of course in the final measure (m. 16), where the two vocalists sing in unison. Not only is this the only vocal phrase that subdivides the eighth‐note pulse, but it also synthesizes the 3+2+3 division characteristic of this section with the variations on 3+3+2 heard previously: while the measure as a whole is clearly divided as (3+2+3)⁄8, its first half similarly clearly divides as (3+3+2)⁄16.

Use of the same “‘mobile’ melodic prominence self‐differing by second” trope of previous sections is evident in the bass­line, where the highest two pitches of each four‐measure hyper­bar differ by second from their counterparts in the previous hyper­bar.

Undoubtedly, this particular contour is difficult to pick out of the mix. Nonetheless, it is quite clearly there upon inspection. More importantly, however, the alternation between the second‐highest pitch (which lasts much longer than the highest) being D♯ and being E is made much more obvious by the crucial mm. 15–16, where E clearly dominates this dissonant chord, rather than being the D♯ of the less‐dissonant G♯−⁶₄ chords.

Moreover, this contour trope is arguably also present in the guitars, where each pair of chords has highest pitches differing by second.

In any case, with the bass contour in mind, we once again get a periodic structure:

Even in Kidcrash’s apparently most austere textures of pure block‐chords supporting screamed vocals, we can still hear spiritually horizontal (in this case, contour‐based) organization.

This section moves up from F♯ minor to G♯ minor. This represents the first real break from F♯‐centricity.

The chords here come in pairs, and those pairs themselves come in pairs. The first pair is always the same: C♯−7no3⁴₂–G♯− ≘ iv⁴₂–i. This functions as Ⓣ. The second pair more‐or‐less alternates between something like E–G♯−⁶₄ ≘ ♭VI–i⁶₄ and E–E∆7 ≘ ♭VI–♭VI^∆7. This alternation is possible because of the bass­line’s contour, as discussed in the “Melodic contour” section above.

Once again, we see a preponderance of “i”, “iv”, & “♭VI”, but no clear indications that this kind of analysis is directly useful. Observing in detail how the final chord of each hyper­measure evolves over the course of this section will help:

In mm. 3–4, any dissonance is at most mild, with merely the bass note of i⁶₄ suggesting some eventual return to root‐position i. Then, mm. 7–8 produces a real dissonance — namely, an ic 1.

The next period adds a little extra spice at first, by adding a C♯ (the perfect 4^th factor) to the i⁶₄, thus producing the gentle diatonic cluster {B, C♯, D♯} ∈ ⟦0⁣2⁣4⟧_3‐6, with Ⓢ function. The effect is subtle, to be sure, but much less subtle once we arrive at mm. 15–16. Here, we get a full (read: all factors, except the 9^th & 11^th) ♭VI^∆7,13, producing the much more significant (albeit more spaced‐out) cluster {B, C♯, D♯, E}, a member of our familiar ⁅0⁣2⁣4⁣5⁆_4‐11B.

This section generally lowers the intensity relative to the latter half of the previous section, which I’ve notated as a change from 𝆑𝆑 to 𝆑. However, it’s difficult to evaluate the overall dist‐pitch of this section: although the use of crashes & other noisy drum hits is considerably sparser, we can argue that this is compensated for by the addition of shouted vocals.

Given that the overall dist‐pitch is probably not significantly different from that of the previous section, the real change in intensity seems to come from the guitar parts. Because the individual block‐chords are held for upwards of a dotted whole‐note at a time, the extra intensity (& noise) of plectrum hitting string is far less present, and the gradual attenuation of the chords as they ring out further removes intensity.

This brings out an aspect of the texture: there’s a clear turn‐taking (or call‐&‐response) between the guitars & the vocals, so that the shouting only occurs when the guitars’ notes are left to ring, and conversely, the guitars only play new chords when a vocal phrase concludes. This produces a local kind of dynamic or dist‐pitch uniformity that wouldn’t be present if the vocals were written like a more straightforward verse.

It should be noted, of course, that the way that vocals are transcribed here — including, for example, in Fig. 11 — does little justice to the actual music. The pitches indicated are approximate, and any inflections distinctly present in the performance, any changes in technique & phonation, etc., are entirely absent.

Speaking of vocals, given that this album is widely understood as a “screamo” record, we might be surprised to hear vocals for the first time only a half‐dozen sections, and over two minutes, into the piece.

A vocals‐free first two minutes or so is not notable on its own. In reality, the screamo genre is noted for its dramatic song structures, quite frequently including lengthy introductions & interludes that serve to build tension and to provide contrast. What makes Kidcrash’s songwriting notable in this respect is much more closely related to the fact that we’ve had to go through some half‐dozen sections to get here.

Indeed, this piece already had a clear introductory section, which — although of great import — only actually lasted for 15 seconds or so. The intervening nearly two minutes — keeping in mind that the entire piece is less than five minutes in total — between the intro & the first instance of vocals is not meaningfully introductory nor interludic; it is “the meat” of the piece, that just so happens to not include any vocal parts.

The effect is that, even before the very first verse — ordinarily the real starting‐point of a piece of popular music — the listener has already been on a whole rollercoaster of emotions & motivic material. This directly relates to the text of the piece, as will be discussed later.

On this record, Kidcrash are like many — albeit not all — screamo bands in that their arsenal of vocal techniques does not include traditional “clean” singing.

In practice, although there are unclean vocal techniques that aren’t associated with screamo — particularly, those especially characteristic of certain kinds of heavy metal — the palette is still quite broad. Common descriptions may include highly‐inflected shrieks that feature violent changes in vocal register; harsh, noisy screams produced with the cuneiform cartilages; Sprechgesang‐esque styles that incorporate aspects of a spoken voice; so‐called “bird chirps” associated with Jeromes Dream’s Seeing Means More Than Safety (2000); etc..

On Jokes, the technique is perhaps best described as “shouted”. We can clearly hear the sonic striations produced by aryepiglottic manipulation (see below), but the phonation of the glottis is “clear” enough that the technique can sometimes dip into yelling or talking territory. Moreover, the technique carries a decent bit more pitch information than might be expected of a screamo vocal style; not Sprechgesang, but perhaps Schreigesang.

For these reasons, although I’ve used cross‐shaped noteheads to indicate shouting, the pitches transcribed in Fig. 11 really are there, and are fairly audible as well.

This vocal technique locates Kidcrash firmly within the realm of hardcore punk & post‐hardcore. Moreover, the duelling vocalists — especially when shouting together — evoke the “gang vocals” of some hardcore punk & of some other punk subgenres.

As befits the freeness of rhythmic patterns in the vocals, the poetic meter of the text here is irregular: ⟨6, 8, 4, 4; 5, 6, 5, 6⟩ syllables, for a mean of 5.5 syllables per line.

There is nevertheless some regularity even here, given the ⟨5, 6⟩ alternation of the 2^nd stanza, the exactly 22 (= 2 ⋅ 11) syllables of each stanza, and the fact that the latter two lines of the 1^st stanza are clearly paired — including each being a pair of iambs.

Moreover, if we split the text hy­per­met­ric­ally instead of periodically — that is, into four stanzas, rather than the two in Poem 1 — we observe that the text is clearly written for the turn‐taking style of vocals employed here, as a kind of call‐&‐response:

• Lines 1 & 2 form a single sentence, with the lines having matching present continuous constructions be sinking & [is] rising.
• Lines 3 & 4 are both NPs, both describing the same thing: physical attributes of Turtlelephant. Moreover, these lines are metrically paired.
• Lines 5 & 6 is call‐&‐response in a more literal sense: the speaker responds to Turtlelephant’s lumber‐shredding jaws with trepidation. These lines are also metrically paired, albeit in a different way.
• Lines 7 & 8 form a single sentence, and also have the same meter as lines 5 & 6.

On the one hand, we have the raggèd poetic meter, the lack of rhyme, and the irregular musical rhythms. On the other, we have some metric regularity in the text, a clear alignment with both hyper­metric levels of the music, and a close fitting of the vocal rhythms to the underlying pulse.

From this combination, the listener gets an impression somewhere in between that of a traditional “verse” (in the musical & poetic senses) and that of freely‐exchanged spoken word, especially — given the vocal technique — the confrontational shouted vocals typical of hardcore punk & its descendants.

English poetry is typically formalized in terms of accentual‐syllabic verse, plus possibly some rhyme scheme. Nevertheless, assonance & alliteration play a significant role, especially historically and in rap. It’s thus worth noting at least some use of assonance, whether intentional or not:

• Line 1, /s⁓ʃ/: The shore must be sinking.
• Line 3, /t; s⁓z/: The tides and tusks. Note the /ˈʌsP(s)/ (where /P/ is a voiceless plosive) slant rhyme with line 4.
• Line 6, /n⁓m/: Now I’m grindin’ my teeth.
• Line 7, /z/: Great legs of pillars.
• Line 8, /h/: Have somehow given out.

Although not of particular import, the pronunciation of the name Turtlelephant is revealed here. Perhaps unexpectedly, it’s neither */ˈtɜ(ɹ).tl̩ˌɛ.lɪ.fənt/ nor */tə(ɹ)tˈlɛ.lɪ.fənt/, as the spelling might suggest. Instead, we hear /tə(ɹ)ˈtɛ.lɪ.fənt/ [təʴˈtɛlɪˌfɘnt].

If taken literally, the Turtlelephant must be some kind of supernaturally gigantic sea monster that’s a hybrid of a sea‐turtle & an elephant, with the former presumably contributing its marine status, & the latter its size.

The largest known turtle is the extinct (dating from the Late Cretaceous, roughly 75 million years ago) marine turtle^[20] Archelon ischyros, with its largest specimen being about 4.6⁢ m from head to tail, and likely weighing somewhere around ≈2 750⁢ kg in life. Although this is undoubtedly an extraordinarily large turtle, it’s far from fanciful enough to be identified with Turtlelephant. Nonetheless, it is noted for its powerful jaws, well‐suited for both crushing and shearing (as in line 5).

A better chelonian comparandum is the aspidochelone^[21] of the Physiológos (ca. 2^nd–4^th c. CE). It was described as an island‐sized turtle — or in some cases, whale — known for tricking sailors into thinking that its shell was land. After disembarking, it wouldn’t be long before the aspidochelone dove back into the sea, thus drowning the sailors. This sense of scale, and confusion between turtle‐shell & land, is closely mirrored by lines 1–2.

In English‐language literature, the aspidochelone appears — with the somewhat corrupted name Fastitocalon — in a similar tale: “The Whale”^[22], an Old English poem that appears in the Exeter Book (late 10^th c.).

“The Whale” would then be retold in New English by J. R. R. Tolkien, in “Fastitocalon” (1962) (a revision of an earlier 1927 poem), which takes place in Middle‐earth.

In Tolkien’s retelling, the Fastitocalon is once again a turtle (rather than a whale), and this poem follows immediately after “Oliphaunt” in The Adventures of Tom Bombadil (1962). Taking the two poems together, we perhaps have *Fastitocal­oliphaunt; or rather, Turtlelephant.

It’s worth noting that Kidcrash were not the only mathrock band to (perhaps) take interest in colossal sea‐turtles & in Tolkien. The Californian band Ent — named for Tolkien’s Ent creatures, from OE ent “giant” — features the track “Archelon Was The Largest Sea Turtle” (2000) on their self‐titled & only full record^[23] (the preceding track is entitled “Underwater They Looked Like Spaceships”, suggesting a similar idea).

But the elephantine features of Turtlelephant are not merely implicit. Line 3 attributes tusks to the creature, and line 7 attributes to it the pillar‐like legs characteristic of elephants.

Turtlelephant is clearly depicted as terrifying. However, this wasn’t Kidcrash’s only work dealing with lurking horrors of the sea. For example, the final track on Snacks (2009) is “Sleeper Wave”, the lyrics of which are about exactly what it sounds like:

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

Following hot on the heels of the first verse introducing the titular sea monster, we have another 𝆍𝆏 that leaves behind an instrumentation solely of clean non‐bass guitars. Although the previous use of 𝄵 & of (melodic) major aren’t present this time, the connection is now clear: this is the deceptively peaceful lurking of Turtlelephant.

It’s worth noting that, although this is indeed not 𝄵, the overall pace of the notes — often upwards of a dotted whole‐note — is slow enough to evoke the slow pace also characteristic of previous uses of 𝄵. Naturally, this contributes to the overall soft dynamic as well.

This is followed by a jump back into the full band (minus vocals), with distorted guitars & all. But this jump isn’t as dramatic as the foregoing description might make it sound. Instead, there are a number of factors that make the latter half of ⟪G⟫ more like a middling 𝆐𝆑 or so:

• Relatively very sparse use of cymbals, limiting the overall dist‐pitch.
• Somewhat sparse notes overall, with the snare taking up most of the latter half of each measure.
• Each guitar (bass included) never sounds more than one note at a time, thus lowering the overall dynamic, in addition to mitigating the inter­modulation distortion that would otherwise be possible.

⟪G⟫ is essentially a development of ⟪F⟫. One of the two main elements that are transferred is the four‐measure rhythm carried by the guitars. In the first few measures, we get a straightforward restatement of the rhythm; however, this restatement is abruptly cut short from eight measures to just seven, directly launching into a condensed form of the same rhythm in mm. 8 ff..

This condensed form is not only two measures rather than four, but is also more “straightened out”, so to speak. The increased symmetry results from each measure being divided up in the same way, resulting in a consistent meter throughout mm. 8 ff..

The condensation itself is another good example of structural acceleration^[Hud20]. The result is a clear increase in rhythmic urgency, without the use of a literal increase in tempo.

In particular, the condensed basic pattern is something like ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅮𝄾𝅘𝅥𝅮𝅗𝅥⁠⟩, with a prominent but slight variation of ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅮𝄾𝅘𝅥𝅮𝅘𝅥𝅮𝅘𝅥𝅭⁠⟩. This somewhat ambiguously divides the meter up as (3+5)⁄8, although the drumkit more specifically suggests (3+3+2)⁄8 in mm. 8–9.

Briefly ignoring pitch & phrasal independence, we have considerable disagreement between voices:

• Free variation between the aforementioned two versions of the basic rhythmic unit.
• Guitar 2 frequently, but arbitrarily, bisects some of the eighth‐notes of the basic unit into sixteenth‐notes.
• The final quarter‐note’s worth of each measure may or may not be a rest.

These small internal rhythmic variations contribute an almost hetero­phonic feel to the texture, as the melodies played by the non‐bass guitars are heard as variations of the same melody.

Then again, this isn’t exactly the archetype of hetero­phony. We still retain several ho­me­o­phon­ic features, e.g. a variety of motion types — up to & including contrary motions.

Although this does result in a texture accurately described as ho­me­o­phon­ic, the textural reduction from the home­ophony and homophony heard previously to the texture heard here — as well as the reliance on small internal rhythmic variations — both go hand‐in‐hand with the rhythmic reduction discussed above.

It should be noted that this analysis relies on the non‐bass guitar voices having been notionally “un­crossed”. Like the vocals in the previous section, the guitars here employ a kind of turn‐taking whereby each phrase exchanges the voices between the two guitars. Nonetheless, I would argue that such exchanges are a kind of textural, or perhaps in­stru­ment­a­tion­al, feature that still has musical salience — especially given that the guitars are panned in opposite directions.

⟪G⟫ continues using exclusively pcs of G♯ natural minor. A♯ is absent throughout the section, in spite of all six other pcs being presented multiple times each.

Because — as discussed above — the pitch‐material fabric of the piece is intervalclass‐sensitive and based directly upon voice‐leading, even the mere two‐voice texture of the former half of ⟪G⟫ can be harmonically analyzed with relative ease:

Although we haven’t been seeing ♭III previously, we can think of it as a rootless i⁶, with Ⓣ function — and indeed, it corresponds exactly to i in the latter half of this section. The [E, D♯] of m. 4 clearly has an ic 1, and thus has Ⓓ “function”. And although the [G♯, F♯] of m. 3 has no traditional function either, here it clearly functions as Ⓢ (see List 3).

The result is a clear periodic structure, reïnforced by the regular pattern of motions: similar (anti­contrary) motion from Ⓣ $\stackrel{\mathrm{s}}{\to }$ Ⓢ, oblique motion to a resolutionary response gesture that internally moves by imperfect parallel (anti­contrary) motion, followed by a return to Ⓣ by oblique motion.

In the period’s former hyper­measure (mm. 1–4), we get an unsatisfactory “resolution” to ic 1; we expect the stronger (oblique) motion to resolve some kind of dissonance or suspension, but it does not, and only really “resolves” by moving back Ⓓ $\stackrel{\mathrm{o}}{\to }$ Ⓣ, again with the stronger motion.

In the period’s latter hyper­measure (mm. 5–7), we get the same structure, but instead arrive at a more satisfactory iv⁶ — notably not containing any ics 1 nor 6. However, this iv⁶ quickly (as a result of the “missing” m. 8) moves by contrary motion (see below) to an explicit i.

Thus, the iv⁶ has two functions: Ⓣ in its capacity as a consonance (a bare ic 3) that responds (periodwise) to the [E, D♯] of m. 4, and Ⓟ in its capacity as the first chord of a minor plagal cadence that resolves — by fairly strong motion — to i.

The contrary motion from m. 7 → m. 8 (in Table 19) is normalized from [E, C♯] $\stackrel{\mathrm{+7},\mathrm{+2}}{\to }$ [B, D♯] to [E, C♯] $\underset{\Updownarrow \mathrm{c}}{\overset{\mathrm{-5},\mathrm{+2}}{\to }}$ [B, D♯]. This com­men­su­rates the melodic intervals’ absolute sizes, and especially given that bass motion by P4 or P5 is so heavily stereotyped in tonal music, the listener quite likely hears this as strong motion.

But we should also note that this hearing is bolstered by the bass voice always & only descending B–G♯–E; that is, a root‐position E major triad. Especially because of the near‐evenness of major triads, this serves as its own kind of three‐note (tri­tonic) scale.^[Tym11] Thus, the bass voice is heard as simply continuously descending scale­wise throughout mm. 1–7 — and indeed, throughout the entirety of ⟪G⟫ (although it becomes the middle voice in m. 8 ff.).

In the “Thematic reprise of diatonic clustering & other pitch‐material” section above, I noted that a sudden movement from {E, F♯} to {C♯, D} by anti­contrary motion (actually ⟅o²⁣s⟆^[9]) produced the impression of the diatonic cluster {C♯, D, E, F♯} ≘ {5̂, 6̂, 7̂, 1̂} ∈ ⁅0⁣1⁣3⁣5⁆_4‐11A. We have the same gesture here in m. 3, where [G♯, F♯] suddenly moves to [E, D♯] by anti­contrary motion, producing the impression of {D♯, E, F♯, G♯} ≘ {5̂, 6̂, 7̂, 1̂}.

The latter half of ⟪G⟫ adds the bass‐guitar part, thus turning the bass voice into the middle voice. The basic harmonic pattern is as follows:

The pattern of motions here is clearly periodic — albeit on a smaller scale than in other sections, thanks to the structural acceleration. The first measure moves strongly away from the tonic as i $\stackrel{{\mathrm{c}}^2\mathrm{s}}{\to }$ iv⁵, but not so far as to make the iv⁵ $\stackrel{{\mathrm{o}}^2\mathrm{p}}{\to }$ ♭VI feel like a strong motion back to Ⓣ (read: cadence), and indeed that motion is considerably weaker. The second measure echoes the strong motion away from Ⓣ, but differs by moving to the next period with appropriately strong motion, thus effecting a genuine cadence ♭VIIno3⁷ $\stackrel{{\mathrm{c}}^2\mathrm{p}}{\to }$ i.

Note, however, that ♭VIIno3⁷ does not function as Ⓓ; it contains no ics 1 nor 6, thanks to its missing 3^rd factor. Per List 3, it suggests Ⓢ function; and given that it’s plainly used cadentially here, it functions as Ⓟ.

Still, there is some ambiguity in the fact that the ♭VIIno3⁷ is approached by very strong motion, and lasts longer than the previous & subsequent chords.

Measure 13 isn’t represented in Table 20, because it proceeds E ≘ ♭VI (as before) $\stackrel{{\mathrm{s}}^2\mathrm{i}}{\to }$ {C♯, D♯, E}. This weaker (but still decidedly polyphonic) motion, along with the explicit vertical ic 1, dissolves the ambiguity: this is simply Ⓓ function. Indeed, this is yet another explicit diatonic cluster: a member of ⁅0⁣2⁣3⁆_3‐2B.

The former half of this section of course retains the same basic *durdr contour (in the upper voice) of the previous section, along with its rhythm.

The latter half of this section still retains the basic shape, but simplified even further. However, it also allows for the true reäppearance of the *r‐based contour previously observed in the intro & in ⟪A⁣A′⁣B⁣C⁣D⟫ (& in ⟪E⟫, to some extent).

On top of the repeated B–G♯–E in the bass in the former half of this section, the other voice produces D♯–C♯–F♯–D♯ followed by D♯–C♯–E–C♯. Thus, the two halves of the period correspond by exactly the expected interval of a second.

In the latter half of this section, the upper voice instead holds just a C♯ constant, and we have D♯–C♯–E–C♯. Again, the change is by second. By reducing the constant fragment of the melody & the fragment that self‐differs by second, each from two pitches to one, the same structure is restated in a condensed form. This goes hand‐in‐hand with the rhythmic condensation.

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

Ordinarily, we think (& perhaps should think) of texture as being fundamentally momentary: at any given moment within a piece, the listener has some idea of what the current texture of the music is like.

Nonetheless, although the first 14 measures of ⟪H⟫ are broadly homophonic, the particular texture rapidly shifts, never staying in one place for more than two or three short measures at a time (bearing in mind the 𝅘𝅥 = 180 tempo). This causes some blurring of textural distinctions into a montage‐like effect, which goes hand‐in‐hand with the shifting dynamics, rhythms, & meters:

Although the second (mm. 3–4, 10–11) and third (mm. 5–7, 12–14) regions are clearly distinct in almost every way — including rhythmic concerns — the drumkit part nevertheless ties them together: both feature essentially the same kickdrum part, with hits on the “&” of each beat, plus one hit on the “1” of the initial measure. That being said, the other pieces of the kit are used differently, resulting in the blastbeat‐like drumkit part of the third region.

Measure 19 (the final measure of ⟪H⟫) is an intentionally false start, resulting in a lone measure of 3⁄8. Still, the next measure (m. 1 of ⟪I⟫) is 6⁄8, so this 3⁄8 measure doesn’t (at least, not initially) sound as orphaned as its written transcription might suggest.

This section presents the key of C♯ minor for the first time. Way back in Fig. 2, I noted that the guitars are tuned to drop C♯. The “natural” & typical key of pieces written for guitars tuned to drop P is P minor; so, in a way, we’re “finally here”. For this reason, and for some other significant uses of open strings, tablature has been included in Fig. 13.

Personally, I find it unlikely that a listener hears the large‐scale harmonic movement of the keys throughout the piece as being a progression in its own right, akin to that of an ordinary chord progression. Nevertheless, if we did make this supposition, then the listener would hear F♯ minor – G♯ minor – C♯ minor, which is a good ol’ (minor) iv–v–i. Charming!

With these things in mind, we’d expect that — at least, pitchwise — the commencement of ⟪H⟫ would represent a point of structural significance. In retrospect, this perhaps makes the seemingly slight detail of the two non‐bass guitars sliding out in opposite directions — up & down (or “doit” & “fall”) respectively — bear some significance, in the “structural voice‐leading” sense.

In any case, the key change is, naturally, not immediately obvious. The only diatonic change is A♯ ↦ A♮, but we don’t hear an “A” until m. 3’s “& of 2” — and then m. 5 hits us right in the face with it. Still, given the largely homo­phonic textures of most of this section, glossing the harmonic material shouldn’t be too difficult:

We hear v⁵ (AKA V⁵) at the beginning of mm. 3 & 10, but not in its traditional Ⓓ function; rather, it signals movement away from i and preparation for Ⓓ — that is, Ⓢ function (see List 3). Then, ♭VI∆⁴₂ (& later, ♭VI^∆7) provides the ic 1 that bears Ⓓ function. It should be noted that this chord is initially heard in very clearly third inversion (that is, with 5̂ in the bass), and that it would be ♭II relative to the key leading into this section — both of which are suggestive of Ⓓ in traditional tonal idiomata.

The mm. 5–7 (& mm. 12–14) sequence, then, returns to the i in a somewhat intriguing way. Some of the dissonance of ♭VI∆⁴₂ is neutralized by putting it in root position. Then, the next chord is ambiguous: although it contains a diatonic cluster {B, C♯, D♯} ∈ ⟦0⁣2⁣4⟧_3‐6 — & indeed, cannot be easily analyzed in a traditional way — it also doesn’t contain any ics 1 nor 6. The following ♭III⁶₄ is similarly ambiguous, but for a different reason: the traditional function of a mediant chord is ambiguous, being sometimes associated with Ⓣ, and sometimes with Ⓓ.^[24]

Given that the ♭III⁶₄ is in second inversion, we could imagine it being cadential, emphasizing the step 7̂ → 1̂; even though this is merely a subtonic, rather than a true leading‐tone. Moreover, m. 14 makes use of a subtle microtonal bend to heighten anticipation of the i.

However, given Kidcrash’s tonal language, I hear mm. 5–7 as a path back to i by way of gradual mitigation of the ♭VI∆⁴₂’s dissonance. In any case, the most salient aspect is the simple fact of ascent: not only do the pitches get higher & higher in register, but the chords trace out an ascending C♯–D♯–E melody.

This melodic perspective highlights the voice‐leading:

As can be clearly seen in Table 23, this gradual mitigation of dissonance also has excellent voice‐leading: no melodic intervals larger than a M2 (with the sole exception of the bass moving by M3 once), good contrary motions, plenty of oblique motion (read: common tones, voice‐leading parsimony), and no perfect parallel motion.

⟪H⟫ is the second section containing a vocal part, but unlike ⟪F⟫, we get something distinctly less verse‐like. They’re only really heard in the third of the four regions (mm. 5–7 & 12–14), taking advantage of the thick texture in the other instruments to produce perhaps the highest overall dist‐pitch in the entire piece. Moreover, we don’t have the vocal turn‐taking of ⟪F⟫ either.

This brief text continues where the previous text (in Poem 1) left off: describing Turtlelephant. The Thousands of birds are, presumably, lurkin’ on or underneath Turtlelephant’s colossal shell.

Although ⟪F⟫ includes the lyric Now I’m grindin’ my teeth, ⟪H⟫ is the first time that we get a distinct sense of narrative: the narrator is at sea, and has realized all too late that they’ve sailed into Turtlelephant’s territory.

This differs from the aspidochelone or Fastitocalon narrative, insofar as the sailor here is not fooled into believing that the monster’s shell (or back) is land, as evidenced by the first lyrics of the piece: The shore must be sinking, ‖ Or Turtlelephant’s rising up.

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

The most striking aspect of ⟪I⟫ is certainly its use of rhythms & meters. This is doubtless the most rhythmically‐disorienting section of the entire piece, such that upon casual listen, it’s not at all clear what’s even going on.

Upon close inspection, very nearly all notes played here are firmly within the basic grid of pulses, with only two or so exceptions, which are marked in Fig. 14. Given the rhythmic complexity, & the fact that this was presumably played live (in‐studio), this is pretty impressive.

The key here is that the basic meter is expressed by the non‐bass guitar parts — especially guitar 1. The drumkit part then elaborates the meter with often quite heavy syncopation. And finally, the bass‐guitar plays all its notes quite accurately in‐time (relative to the overall pulse), and synchronizes with the other parts at key points, but otherwise plays completely freely (rhythmically speaking).

In mm. 1–4, we mostly have just the non‐bass guitars playing, and m. 4 is divided ambiguously, hence the use of “(3+7)⁄8” in Fig. 14. Nevertheless, this sets up the basic structure, which is periodic as usual:

A slight simplification of List 5 can be written in table form, as in Table 24:

In addition to contributing rhythms not heard in the other parts, the drumkit part contributes alternative interpretations of the underlying metric structure.

In mm. 5–6, the drumkit plays 6⁄8+2⁄4+6⁄8, clearly poly­metric­ally (or at least poly­rhythmic­ally) against the meter of the non‐bass guitars. (Indeed, the hi‐hat pedal is even used to suggest that the 6⁄8+2⁄4 (= (3+3+2+2)⁄8) may be heard as 5⁄4.) This same phenomenon recurs in mm. 9–10.

Thus, the drumkit is actually reversing the hyper­measure by playing (3+3+2+2)⁄8 first, and then playing the 6⁄8.

Measures 7–8 are played in basically the same meter as the non‐bass guitars, except that the hi‐hat produces ambiguity by suggesting (3+2+2+3)⁄8. And finally, in m. 15, the drumkit defies the underlying meter somewhat by not playing anything on the “1”, and then accenting the “la of 1” (= the 2^nd eighth‐note of the measure).

Naturally, all this rhythmic & metric complexity, combined with the metric freedom of the bass‐guitar, produces nothing less than a dizzying effect — counteracted only by some of the non‐metric aspects discussed below. Thus, with the smooth transition to straight 3⁄8 in m. 17 ff., the listener gets a sense of metric “resolution”, where the smoothing‐out represents the metric equivalent of resolving a dissonance to a consonance.

In particular, m. 14 ff. starts by suggesting the beginning of another hyper­period. But its hyper­beat 2.2 (see Table 24) only gets past its first beat — that is, its first 3⁄8 — before all parts are brought into unison to play eight of the same dotted quarter‐notes together. This eight‐bar sequence is really its own hyper­measure, clearly separated from the rest of ⟪I⟫ up to that point by texture & rhythm (and indeed, tempo: the metric manoeuvring additionally serves to cleverly mask the jump from 𝅘𝅥 ≈ 170 to 𝅘𝅥 ≈ 185).

The result is that the last period is really truncated, but serves to seamlessly transition into the 𝅘𝅥𝅭 block‐chords by simply stubbornly repeating the 3⁄8 beat that the listener expected to be there anyway.

As mentioned, the bass­line is largely metrically free, even if nevertheless strictly adhering to the pulse. There are moments of synchronization, however, which tend to have a similar — but lesser — effect to that of the metric resolution discussed above:

• The single note of m. 4.
• The note at the end of m. 8.
• The end of the phrase in mm. 9 & 14. (The phrases are identical.)
• All of mm. 12–13.
• The two quarter‐notes of m. 15.
• Measure 18 ff..

Although the bass­line seems quite free at first, it actually makes use of the same handful of phrases. It even reüses some of the same metric placements (e.g. in m. 9 & m. 14), suggesting a kind of poly­metric feel that is not explicitly notated in Fig. 14.

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

As expected, the longest & most prominent notes of the two halves of the period (as shown in Fig. 15) differ by second (E vs. F♯). Both phrases have clear internal coherence, and it is largely their juxtaposition against the other parts — & their rhythmic variations — that produces the confusing texture. Ⓐ is just scale­wise descent from A to E, and Ⓑ is a C♯ minor pentatonic gesture.

Given the foundational nature of the non‐bass guitar parts, we might describe the primary component of ⟪I⟫’s texture as being basically ho­me­o­phon­ic: between these two parts is plenty of anti­contrary motion and a fairly close phrasal dependence; but at the same time, a good deal of oblique & contrary motion as well. Moreover, we get significantly different rhythmic choices at several points.

Once we also consider the bass‐guitar, however, we get something like a genuine — if perhaps wobbly — counterpoint, with an abundance of contrary motion, oblique motion, & phrasal independence.

This is perhaps a strange conclusion to come to, so it may be more effective to think of ⟪I⟫ as being primarily home­ophonic. But to that we should also add a reversal of rhythmic duties: the non‐bass guitars lay down the basic metric structure, while the drumkit & (especially) bass‐guitar are allowed to freely rhythmically innovate within & around that structure.

⟪I⟫ represents one of the quieter full‐band sections of the piece — at least, prior to m. 18. Nonetheless, the dist‐pitch can tell a different story in some ways: the drumkit part maintains a wash of pink noise with the cymbals, and although the guitars are on their low‐distortion setting, at moments where guitars 1 & 2 play double‐stops together, we get a palpable increase in distortion (“dirt”).

This results in a fairly timbrally‐stratified texture:

• The non‐bass guitars’ double‐stops provide distortion that momentarily swells the dynamic of the section.
• The non‐bass guitars’ single notes are clearly separate.
• The bass‐guitar occupies a space of its own, as usual; but here the stratification is clearer, as a result of its rhythmic independence.
• The cymbals maintain the stream of pink noise, while also playing often unique rhythms.
• The snare & kickdrum provide the main rhythmic force of the drumkit part.
• The hi‐hat is operated with the pedal, providing a sporadic but clearly independent rhythmic channel that easily sticks out whenever present.

This timbral–textural stratification perhaps contributes to the disorienting quality of this section, given that it makes the rhythmic complexity more evident: without a clearly overriding stratum, there is no obvious “anchor” upon which to focus the listener’s attention.

This section is twinned with ⟪B⟫ in several ways:

• Both feature a primarily home­ophon­ic texture between the guitars, with distortion at a minimum.
• Both juxtapose these clean intertwined guitars against the aforementioned constant wash of pink noise from the cymbals, resulting in an unexpectedly high dist‐pitch.
• Both have regularly‐shifting meters that culminate in a period whose final iteration is truncated to lead into a “resolution” of the metric dissonance (to straight 3⁄8 in this case, and to 𝄴 in the case of ⟪B⟫).
• Both metric resolutions lead to a texture purely of block‐chords, where all parts play the same rhythm.
• Both have non‐bass guitar parts largely in single notes, but with prominent double‐stops at the end of each hyper­measure.
• Both make clear use of quartal harmony; specifically the construction {0, 5, 10, 15} (see the “Pitch‐material” section below).

Even to a casual listen, I think that this structural relationship is reasonably evident. Thus, it provides a clear tie between the former & latter halves of the piece, but without actually requiring the ⟪B⟫ & ⟪I⟫ sections to share any “material” per se.

Perhaps the defining contour of this section is an ascending ⟨⁠𝅘𝅥𝅮𝅘𝅥𝅮𝅘𝅥𝅮⁠⟩. But given the prevalence of the “periodic self‐differing by second” in previous sections, we want to know to what degree — if any — that structure is present in ⟪I⟫.

Indeed, in the first statement of this section’s theme (mm. 1–4, the initial fledgling period), we do get this structure: m. 2 ends with a repeated high C♯, which is responded to in its latter counterpart by the high D♯ of m. 4. Later statements contain variations that always retain some trace of this C♯–D♯ response pattern, but the relationship becomes obscured.

The bass­line also contains this periodic ascent by second (but instead from E to F♯), as is made clear by Fig. 15.

Although the tone‐poetic significance of ⟪B⟫ was perhaps not immediately clear — partly because of its position near the beginning of the piece — the significance of its twin, ⟪I⟫, is much clearer: this is the ship at turbulent sea.

We’ve just heard the Shallow waters, sailed too far lyric of the previous section, and now we get:

• The most intensely disorienting (hyper)metric structures of the entire piece.
• A drumkit part that emphatically contradicts the basic metric structure, and a metrically free bass­line.
• A constant wash of pink noise from the cymbals, in spite of an otherwise low dist‐pitch.

The chaos of Turtlelephant emerging from the sea is clear in the texture, rhythms, meters, etc. of ⟪I⟫; and all this is juxtaposed against the pink noise of the crashing waves.

This increase in turmoil is the essential point of differentiation from ⟪B⟫, thus giving a sense of narrative progression felt at least as strongly — if not more — in the instrumental parts as in the lyrics.

Like the previous section, we are here diatonic to C♯ natural minor. By focusing on the basic hyper­periodic structure identified above, we can look past the predominantly home­ophon­ic & contrapuntal (read: melodic) texture to nevertheless obtain a reasonable harmonic gloss:

As expected by this point, we have a nearly complete avoidance of traditional functional harmony, in spite of the presence of a decent number of tertian constructions. However, as also expected, we get plenty of non‐tertian constructions, and the use of particular ics to suggest “function”.

Unlike previous sections, however, ⟪I⟫ leans a little more heavily on certain constructions previously heard only rarely:

• Extensive use of the ♭III & its many variants.
• Considerable use of chords that can be constructed quartally as a stack of three P4s, i.e. {0, 5, 10, 15} ∈ ⟦0⁣2⁣5⁣7⟧_4‐23. These are found rooted on v, as well as on iv.
• Very explicit sounding of the 2^nd (or sometimes 9^th) factor in the bass, so that we often have some variation of ♭III/4̂. This construction is effectively normalized by the time that we get to the former of the final block‐chords, at which point it’s played eight(!) times in succession.

Still, especially when keeping voice‐leading in mind, the chords in Table 25 are quite neatly classified by the frameworks developed so far:

Some comments on Table 26 are warranted:

• The 7^th factor of the ♭III^∆7 fairly consistently serves as a kind of “leading‐tone”, insofar as it resolves upward by m2 to an important pitch (e.g. the 3^rd factor of the i⁷).
• The iv: {0, 5, 10, 15} is used as Ⓟ, as is perhaps to be expected, given that its root is notionally 4̂; that is, we get something resembling a minor plagal cadence.
• Several diatonic clusters are used, never with Ⓣ function. Most notably, ♭III: {0⁣2⁣4⁣5} ∈ ⁅0⁣2⁣4⁣5⁆_4‐11B is used with clear Ⓓ function, corresponding to its possession of ic 1.
• The “♭III: {0⁣2⁣4⁣7}” is more like an ad hoc notation than it is a root assignment. Here, it’s quite evidently — including the 4̂ in the bass — not used as a ♭III chord in the usual sense, and is instead an extension of the cluster ⟦0⁣2⁣4⟧_3‐6 — which is also sounded, as “♭III: {0⁣2⁣4}” — both with clear Ⓢ function.
• The ♭VI^∆7 is never properly resolved until the beginning of the next section, thus increasing the tension of this section, and providing a bridge to the next.

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

This section is in relatively straightforward 𝄴 until m. 28 ff.. Hyper­metric­ally, however, there’s a rather interesting phenomenon that we might call telescopic halving (if you will):

The hyper­meter is thus based upon 7, resulting in a clearly irregular hyper­meter — in spite of the meter always being plain 𝄴.

In the drumkit, we find syncopation in the “downbeat” consistently landing on the “& of 1” of every other bar (in the first four bars of each hyperbar), resulting in a kind of (9+7)⁄8 feel. This displacement by 𝅘𝅥𝅮 recalls the same displacement as heard throughout ⟪A⁣A′⁣B⟫ and ⟪F⟫ (although the displacement in the latter is a result of what I called “decompression”).

With the addition of the tremolos in guitar 1, we get a distinct (3+5)⁄8 division of about half of the measures.

The vocals then provide the additional syncopation heard in ⟪J⟫. The first lyric introduces dotted eighth‐notes (m. 9), and in m. 18, we get a clear (3+2+3)⁄8 division.

This section is the only one to feature either of two techniques: sixteenth‐note hi‐hats, and tremolo picking (also in sixteenth‐notes). Given that the ways in which these techniques are used don’t challenge the prevailing pulse nor meter, they’re best thought of as timbral effects.

Especially in the first six measures, the sixteenth‐note hi‐hats contribute greatly to the texture by filling in the space between successive block‐chords. But the way in which they do this does not recall the washes of pink noise (from cymbals) heard in some previous sections. Although the hi‐hat is a cymbal — or rather, two of them sandwiched together — the sixteenth‐note bursts are relatively controlled (as the hi‐hat is firmly closed), resulting in a tight groove that brings attention to the drumkit itself, rather than producing an aural ambiance.

The addition of the tremolos in guitar 1 essentially mimic the hi‐hat. However, due to the high levels of distortion on the guitars, the resulting timbre is almost impossibly gritty, making the pitch‐material — including how many notes are even being played per chord — somewhat difficult to make out. In part, this is a simple result of the tremolo picking technique inherently resulting in a timbre dominated by attack.

This combines with the noise of the sixteenth‐note hi‐hats (& the vocals), to produce a high dist‐pitch — a wall of noise, really — unusually paired with a sparse low‐end^[11].

The texture of this section is essentially that of another verse — unlike that of ⟪H⟫, which was more of a quasi‐verse that only had two lines of text anyway. Thus, apart from the tremolos in guitar 1, we mostly have block‐chords accompanying single vocal lines, as in typical homo­phony. Moreover, we get a return of the turn‐taking between vocalists that was characteristic of ⟪F⟫.

Nonetheless, the texture here is notably “sparser” than that of ⟪F⟫ (the only other true verse in the entire piece), insofar as quite a bit of space is left in between successive block‐chords.

We must also note that, because the pitch‐material of this section relies on preëxisting harmonic norms — and indeed, does so more heavily than any other section in the entire piece — there is little or no need for convincing surface‐level voice‐leading; and indeed, the empty space between chords lessens any such necessity even further. This allows for the voicings to be dominated by parallel motion (so‐called planing), with the addition of only small amounts of similar & oblique motions.

As noted previously, Kidcrash are not playing metal(core) here. That being said, the tremolo‐picked [D♯, E] in mm. 13, 20, & 27 — and arguably the [C♯, D♯] in the previous measure, although it can be disqualified for being a major second rather than minor — can be compared to a panic chord.

At first blush, it might seem to tick all the boxes — but there are two problems. For one, the register isn’t high enough: its register is no higher than that of the block‐chord that immediately precedes it. Moreover, the dissonances of the [C♯, D♯] and then [D♯, E] are indeed intended to be treated; specifically, by resolution to ultimately something like an E major triad (but see below).

Indeed, the tremolo‐picked parts of ⟪J⟫ are just a particularly explicit example of the interval­class sensitivity already observed in previous sections: [E, G♯] ∈ ic 4 → [G♯, B] ∈ ic 3 → [C♯, D♯] ∈ ic 2 → [D♯, E] ∈ ic 1 → resolution.

Although the texture of ⟪J⟫ no doubt differs from those of previous sections, the basic pitch‐material régime needn’t differ quite so much — in spite of the new absence of surface‐level voice‐leading evidence. Thus, we get a sense of resolution already by the beginning of m. 14 (etc.), in spite of that chord clearly not being a traditional tonic chord (but again, see below).

As transcribed in Fig. 16, there are two suspicious‐looking voicings:

• The chord initially played in m. 1 consistently has a C♯ in guitar 1, instead of the perhaps more expected B that’s transcribed for guitar 2. This does indeed give an audible additional “crunchiness” to this chord, but the effect is subtle. As near as I can tell, the two guitars really are playing different block‐chords.
• The chord initially played in m. 6 fairly consistently has a low (but not bass) C♯ in both non‐bass guitars. This takes advantage of an open string tuned to C♯, and is quite audible (moreso than the above) in the recording.

The result is that, technically, there’s a C♯ present in absolutely every chord played within ⟪J⟫. This can be analyzed as a pedal­point, albeit only in the pitch‐material sense that it arguably shifts between NCT status and CT status, or at least between added‐tone status and not. But of course, it’s not a “pedal­point” in the textural sense.

Whereas the previous section was reasonably unambiguously in C♯ minor (with only some hints at E major), I analyze ⟪J⟫ as being essentially in E major, although it is ambiguous enough to allow analysis either way. In favor of E major, we have:

• IV^∆7–I6 is analyzable as a reasonably traditional tonal cadence — insofar as plagal cadences are commonly accepted, at the very least in popular music — whereas ♭VI^∆7–i⁶₅ as a “cadence” relies on some of the nontraditional frameworks proposed here.
• I pretty distinctly hear I–IV–V in mm. 1–6, which is a difficult association to shake, even if a “proper” resolution is consistently avoided.
• The C♯ in the tonic chord is somewhat inconspicuous: it’s in an inner voice, and at that, only in one part (guitar 1).
• E‐centricity implies that the tonic chord is in root position.
• The first tremolo‐picked phrase is of the chord [E, G♯], which more strongly suggests E than it does a rootless C♯−.

But we can marshal some evidence for C♯ minor, as well:

• There’s a C♯ pedal­point (again, specifically in the pitch‐material sense). A pedal­point on 6̂ would at least be unusual.
• Analyzing the first chord as i⁶₅ gives a tertian sonority with no added‐tone.
• Although I–IV–V is more convincing, approaching the tonic with ♭VI–♭VII (suggesting an Aeolian modal flavor) is not exactly unheard of in popular music.^[Eve04]
• The only non­diatonic pc in the entire section is A♯. This pc is — classically speaking — native to the key of C♯ minor, as a member of C♯ melodic minor.
• The previous section was in C♯ minor. Ceteris paribus, the listener is more likely to hear “no detectable difference” than anything else.

Unfortunately, the vocals are really no help, as they’re limited to just three pitches of the E major pentatonic scale, or equivalently, the C♯ minor pentatonic.

So, naturally, much of the problem comes down to tonic­ization. I’ve already noted that the ⟪I⟫ → ⟪J⟫ transition favors E major in this respect. But even more interesting is the turnaround (which we hear several times over):

• In C♯, we get ♭VII–IV–i⁶₅. This would be a double plagal cadence.^[Eve04]

  The main complication here is that the tonic chord is minor, meaning that we lose the characteristic downward resolution by semitone from root to 3^rd factor. Still, we do at least get upward “resolution” by semitone to the 7^th factor, which is of some utility.

• In E, we get V–II–I6. At first blush, this looks more difficult to explain, as *V–V/V–I6 is clearly nonsense.

  But given the evidence in favor of E‐centricity, it’s not difficult to hear this as analogous to the blues turnaround V–IV–I. Both II (as an altered ii, or as V/V) and IV traditionally have Ⓢ function, suggesting that the II is here used merely to mitigate what is ultimately a V–I.

  The fact that II–I has awkward voice‐leading (although we do get semitonal ascent from 3^rd factor to 5^th) further supports the idea that the II–I per se lacks functional &/or structural implications.

The idea that Kidcrash might be intentionally mitigating what would otherwise be a fairly ordinary V–I is quite credible, given what pitch‐material we’ve heard in previous sections.

Although Kidcrash pilfer more‐or‐less freely from conventional harmonic norms (of popular music, etc.), they are careful not to rely on such pilfered goods for extended segments of the piece; and when they do avail themselves, they do so only inconspicuously. Overt use of the most fundamental idiomata of Common Practice music is, put simply, too heavy‐handed. Stylistically, Kidcrash have no need for this level of unsubtlety, as they’ve already developed their own tonal language — as we’ve already heard.

This is the final section with vocals, and yet — perhaps frustratingly — has the least clear‐cut lyrics.

Assuming that this is still on the subject of Turtlelephant, then perhaps the message is that Turtlelephant is amphibious^[26] — as befits a cross between a marine turtle & an elephant — and that any attempt to really escape it is futile. From valleys, to deserts, to the deepest sea trenches, Turtlelephant will find you — and then it’s all over.

Playable MIDI rendering and LilyPond source

LilyPond sourceMIDI file (SMF)

Although not quite as complex as that of ⟪I⟫, ⟪K⟫ has its own periodic hyper­metric structure. Here, the foundational former half of each hyper­measure is 4⁄4+4⁄4; in the latter half, 4⁄4+2⁄4 is responded to by 3⁄4+3⁄4+2⁄4:

Once again we have the use of 3+3+2, and a prominent one at that. So prominent, in fact, that this section begins with a fragmentary “final quarter” of a period (a kind of “false start”), so that it can both begin & end with 3⁄4+3⁄4+2⁄4.

The 3⁄4+3⁄4+2⁄4 cannot be analyzed as 4⁄4+4⁄4. Similarly, the 4⁄4+2⁄4 cannot be analyzed as 6⁄4 (although a 3⁄2 interpretation is possible).

In any case, the former of the resulting hyper­measures is irregular (14⁄4), and so too is the overall period (30⁄4). Notice, however, that none of the meters are at all unusual: we have only 2⁄4, 3⁄4, & 4⁄4.

Just by glancing at the transcription in Fig. 17, there would seem to be little or no syncopation in this section. However, although the guitar 1 part is largely a constant stream of single eighth‐notes — never playing the same note twice in succession — the recurring gesture of ⟨0, 3, 0, −2⟩ anchors guitar 1’s rhythms.

The gesture begins with 0 ≘ C♯, but is transposed down (both chromatically & scale­wise) by P4 to 0 ≘ G♯ starting in m. 10. And indeed, this section essentially begins with two presentations of this motif, in mm. 1–2.

Articulation is important here: as notated in Fig. 17, some notes are slid to^[27] from the previous note (denoted by a glissando line), others are pulled‐off to (denoted by a descending slur), and the rest are plucked in the usual way. This gives the stream of eighth‐notes a distinct additional dimension that naturally connects & separates said notes. With the ⟨0, 3, 0, −2⟩ motif, we consistently get ⟨plucked, slid, plucked, slid⟩.

If we count the duration (in quarter‐notes) between each successive instance of the motif, we get the sequence in Fig. 18:

Clearly, these durations are restricted to either 3 (= 𝅗𝅥𝅭) or 4 (= 𝅝), with 3 being more common. Beyond that, however, there’s little that we can say about any potential patterns here. Both transpositions have exactly three instances of a 𝅝 duration; but given that the duration of a period is 30⁄4, we’d expect the number of 𝅝 instances to be restricted to a multiple of three anyway^[28].

I don’t hear, nor would I analyze, a pattern in Fig. 18. But given that there are so few variables here, the apparent randomness is justified by the structure of the music throughout this section: