This paper was originally delivered to the delegates at the ICE-Z2 conference in Rome, 2006. During the conference knees-up in the evening, baby manufacturer and comedy poet Ben Watson berated my paper for its inadequate coverage of Jimmy Carl Black, Arthur Dyer Tripp III and Billy Mundi. Certainly, they don't get a mentioning but that has nothing to do with their drumming ability. It's merely that I am interested in stylistic aspects of drumming that developed later. Since the conference I've tweezed it slightly and (since this is the internet and nothing is sacred anymore) will continue to work on it...

Ever wonder why your daughter looked so sad?

Non-danceable beats: getting to grips with rhythmical unpredictability in Project/Object

I became a drummer on Christmas day 1985 when my uncle, a drummer himself, sent me some sticks and a copy of Rhythm magazine. It was probably about this time I was planning on world domination with my school friend Steven Maloney. Steve was an amazing comic book artist even at the age of twelve but we both wanted to be rich and famous rock stars so we started a band called TNT. We would write fairly puerile songs, usually influenced by a mixture of the Young Ones, jokes about shit, AIDS and other sexually transmitted diseases, and in Steve’s case, a love of Marvel Comics. What was particularly bizarre about TNT was the way we divided the role of the drummer so that I would play the rolls and Steve would play the beats. We didn’t have a drum kit or know anything about drumming so this made perfect sense… plus (in theory) we would both get to play the drums. In retrospect, this is quite interesting because without realising it, we’d identified two fundamental properties of popular music. Drumbeats embody repetition and drum rolls, i.e., the breaks in repetition, embody difference. We understood we needed them both for our songs to be popular, and they had to operate in a certain way. They had to be predictable.

TNT never evolved past the plotting stage and like all great bands it eventually split with a degree of acrimony between its members. In the years that followed Steve became hooked on the Sex Pistols and I became obsessed with Zappa (or, while he minded the bollocks, I wallowed in lumpy gravy). In the back of the Rhythm magazine my uncle had given me was an advert for Orchard Percussion Studios. With no-one to play the drumbeats anymore it was out of necessity that I should learn to play the drums properly. For the next six years I had lessons with Paul Francis, an experienced player who had landed his first professional gig in 1962 at the age of fourteen with Rolf Harris’s band, The Didgeridoos. Paul owned LPs of Hot Rats and The Rite of Spring, and had a collie dog called Jake that would invariably try to maul me as I walked up the driveway to his house for lessons. Paul had attended the Rainbow Theatre gig in 1971 when Zappa was knocked off stage and he and some friends had telephoned the hospital afterwards to check with staff that Zappa was still alive. It was Paul who first suggested I listen to Zappa’s music because of the wide-ranging types of rhythm and great drumming. The first record I heard was Tinsel Town Rebellion on which I was not only able to identify some of the things I was being taught (I could pick out various styles and time signatures) but I was also exposed to progressive uses of rhythm, such as polyrhythms, metric modulations and unusual beat subdivisions. Eventually I ordered a copy of The Black Page drum set score from Barfko Swill and pondered its statistical density. It looked hard. As I strove to understand Zappa from a rhythmical perspective, I quickly got bored with everything else the popular music industry was offering at the time. It all seemed too predictable; too intro-verse-chorus-verse-chorus-middle-eight-double chorus-fade out (outro).

There are two fundamental categories of musical unpredictability. Horizontal unpredictability refers to a series of musical events which are unforeseen at the outset, this could be almost any piece of music containing a surprising twist or unexpected moment. Likewise, vertical unpredictability takes place at a specific moment, e.g., an improvised orchestral stab where there are no specific instructions as to which pitches or sounds should be performed. Zappa’s uses both across his body of work. From album to album, song to song, bar to bar, the listener’s ability to make judgments about the course of events is put to the test. It happens lyrically as well as musically. Often the unpredictability is designed to expose the cheepis of standardised, traditional popular music forms. A joke is made by setting up something predictable, only to have it fail to meet with the listener’s expectations. To achieve compositional unpredictability Zappa used many techniques, drawing upon the best of what the twentieth century had to offer: serialism, improvisation, chance, indeterminacy, musique concrete and conduction all contribute different kinds of unpredictability to the mix.

Perception plays an important role in how unpredictable something sounds. Mari Riess Jones’s theory of Dynamic Attending is especially useful when considering the perception of Zappa’s rhythmical language. It is comparable with the “good continuation” principle of Gestalt theory which refers to how predictions are made over time. Her theory states that when listeners experience a piece of music, they automatically try to deduce when the events of the piece will occur. At a basic level, the time interval between the first two events of a piece may be used to predict when the third event might take place. If the time interval between the second and third event is less than or greater than the time interval between the first and the second, then the prediction is wrong. If, however, both time intervals are the same, then it is proved correct. Either way, the subject is able to use this information to predict future events. New predictions are based upon the success of previous predictions, where “success” is measured by way of a comparison with what actually happens. The whole process takes place unconsciously and within an instant. It also occurs on different temporal levels; if specific rhythm patterns are repeated, they are grouped together, allowing a higher-level temporal framework to develop. This idea relates to the “proximity” principle of Gestalt theory, the grouping together of elements based on their closeness.

For lovers of critical theory, Dynamic Attending can even be used to interpret Adorno’s concept of “pre-listening” discussed in On Popular Music, which then becomes not so much a case of popular music “listening for the listener”, but of being predictable for the listener. When perceiving standardised forms, the need for the listener to conduct pattern prediction is somewhat reduced because the probable outcomes are, to an extent, already mapped out. In popular music, listeners have grown to expect that certain things will usually happen and this expectation feeds back into the process of making popular music, since it is in the interests of those making pop records to meet with the demands of consumers. Here lies a link between perception and reception. To employ unpredictability is to work against the listener’s instinctual ability to find patterns. This undermines the pleasures associated with pattern recognition (especially repetition) on which the success of chart-orientated popular music is largely dependent. However, it should be remembered that pleasure is derived from the experience of change as well as sameness.

As a young composer inspired by the intricate works of Webern, Zappa discovered the limitations of systems approaches. He once stated in an interview, “If the intrinsic value of the music depends on your serial pedigree, then who in the fuck is going to know whether it's any good or not?” This criticism aimed at serial music benefits from a little unravelling. The process of composing a serial work can be somewhat laborious, depending on how many musical parameters one wishes to derive from the series. Creative decisions normally made by the composer are replaced by a series of rules governed by the system. This restrictive, less than engaging procedure was born out of a desire to free music of its composer’s personality, each step of the process further removing the composer from the final result. The danger is that the net result sounds like a sequence of pitches pulled out of a hat. So why not just do that? This issue eventually divided Boulez and Cage, with Boulez favouring the rigor of total serialism to Cage’s incorporation of I Ching and leaving things to chance.

As a consequence of his early experiences with serialism, Zappa adopted a more ‘haphazard’ approach comprising whatever sounded good to him, for whatever reason. Haphazardness allowed rule breaking to take place if it sounded good. In applying this strategy, Zappa bought himself the freedom to take ideas from whichever source he liked in so far as they proved useful to his own aesthetic, and this eclecticism contributes to the aural experience of unpredictability. It often seems as if Zappa was in a permanent state of restlessness, to prevent an idea from becoming stale, interest is maintained by testing the limits; the desire is to pervert it, turn it into something else, something unexpected. Haphazdness summarises Anything, Anytime, Any Place… It connects Zappa’s works like a unifying constant where the one thing you can always be sure of is that you can’t be sure of anything.

As a drummer myself, the one thing that really stands out to me, it is the advancement of rhythmical ideas from album to album as Zappa’s knowledge grew and the musicians he was able to hire became more technically able to meet with the challenges presented to them. In mathematics, high levels of complexity are used to model randomness. Similarly, the more complicated the rhythmical language of Zappa’s music becomes, the more inclined the listener may be to interpret it as being unpredictable. Zappa was able to incorporate absurdly difficult rhythm patterns in rock music whilst retaining the groove. Take for example the ‘Black Page drum solo part 1’; it’s in 4/4, but this has relatively little impact on how the music is phrased. There are no obvious cues in the music to emphasis four beats in the bar since it is more or less through composed across the bar lines. Some bars contain multiple beat subdivisions whereas others are accented irregularly. And yet it still retains a pulse; you can tap your foot to it at a crotchet equals sixty.

To my ears at least, Aynsley Dunbar was probably the first of Zappa’s drummers to effectively counterpoint Zappa’s lead guitar playing (the crucial word here is 'counterpoint'; I view JCB's drumming as superb accompaniment whereas Dunbar was fighting for an equal say - actually, Les Papp might be the REAL hero here!) He was also the first of Zappa’s drummers capable of producing a hard rock sound. But there is another aspect to Dunbar’s playing which is quite striking, like Vinnie Colauita’s “optometric abandon”, Dunbar had a style which at times approached high levels of haphazardness, e.g. the his drum solo on ‘Waka/Jawaka’. Other prominent drummers of the day, such as Ginger Baker, Mitch Mitchell and Keith Moon, all had their trademarks. Baker was famous for somewhat laborious drum solos consisting largely of military drum rudiments played around the kit. Although skilful, they could get a little tedious. Mitchell’s improv-jazz-rock style perfectly suited Hendrix’s playing, no arguments here, although it could have been Dunbar backing him. Apparently, both drummers auditioned for Hendrix, with Mitchell winning the gig on Hendrix’s coin flip. Keith Moon’s trademark absurd drum fills and general craziness probably entertained Zappa, but his proneness to unpredictable behaviour (detonating explosives within his bass drum, taking elephant tranquilisers, etc.), suggest he was too much of a free spirit to ever be a Mother. It was a miracle he lasted until 1977 in The Who.

Dunbar’s playing perfectly suited Zappa’s guitar. He once stated that when Zappa took off, he took off. Zappa even likened him to two drummers playing at once. Such journalistic hyperbole seems the only reasonable. Without question, the explosive force of Dunbar’s playing helped establish Zappa as a serious force in rock music. Upon moving to LA and setting up his drum kit in Zappa’s basement Dunbar was immediately put to the test: “OK, now remind me why I hired you”, said Zappa. The first thing they played together was the piece which became “Chunga’s Revenge”. This provides a good starting point when considering the freedom that Zappa gave drummers. It also makes the point that in Dunbar, Zappa had found a musician able to provide an unpredictable drumbeat. The best example comes not from the album Chunga’s Revenge but the recent release Quaudiophiliac. “Chunga’s Basement” is an astonishingly intimate piece. Whereas the track “Chunga’s Revenge” is a driving rock instrumental with a ballsy sound, “Basement” really does sound like it was recorded in a room under the stairs. The surround sound recording possibly aids this; with the listener squatting in the middle of the band, it almost feels voyeuristic. The difference between the two tracks does not end there. Musically, ‘Revenge’ is far more regimented than ‘Basement’ and its drum pattern distinctly more predictable.

In ‘Chunga’s Basement’, Dunbar enters after the bass guitar and electric piano. Immediately it becomes clear that he is not playing an entirely repetitive pattern. Each bar of drumming can be divided into two halves. Beats one and two in each bar tend to be performed on bass drum and snare respectively. This establishes a pattern at the start of each bar and gives the listener something to latch onto, reinforcing the sense of 4/4 already provided by the bass guitar. The drumming over the second half of each bar is less predictable with Dunbar improvising over beats three and four differently each time. This technique is reminiscent of a modular approach existing in some African music, whereby certain melodic events can be filled with a small number of pitches in any order but the first and final ones are fixed. In ‘Chunga’s Basement’ though, the unpredictability operates in several ways, for example, note placement, note value, choice of drum and accenting. Some of the figures played are based upon standard rudiments, such as the two consecutive 5-stroke rolls at the end of bars 15 and 16. At other points Dunbar sounds random. One can speculate Zappa giving Dunbar an instruction to play it randomly. In Baby Snakes we see Zappa wiggling his fingers up and down an invisible air keyboard. It’s one of his conduction signals interpreted by Tommy Mars as “play lots of random notes quickly”.

A small amount of cognative research has been carried out into how humans imitate random behaviour. In 1960, Paul Baken asked a class of seventy students to produce a determinant sequence of heads (H) and tails (T) as would occur if an unbiased coin was flipped three hundred times. Ninety percent of the students wrote sequences with too many alternations, having an average repetition probability of 0.42. A truly random sequence of coin flips would have an average repetition probability of 0.5. In a later experiment, William Wagenaar discovered that this effect became even more pronounced when respondents were asked to alternate randomly between more than two choices. His experiment, which involved pushing six buttons randomly, produced a greater degree of alternation than is actually required. It seems logical that a musical analogy to this experiment could happen to musicians trying to play “randomly”. Waganaar’s results imply that they might instead head towards a pattern of alternation.

It is testament to Dunbar’s skill as a drummer that he can turn such unpredictability into a solid base. For all its irregularity, it’s identifiably in 4/4. Dunbar is playing within the meter and it pulsates. There is no sense that he needs to think about the choices he makes either; it all seems to happen automatically, instinctively. In ‘Revenge’ this unpredictability has been almost totally replaced by a fixed pattern with a tom-tom consistently played on the 8th quaver. It has become a more commercial product, with a big sound and much of the spontaneity removed. There remains some unpredictability, though; the cowbell pattern across beats 2 and 3 is pretty tricky to guess.

In ‘Basement’ it would seem that Dunbar had more freedom to try out ideas. Chad Wackerman once commented that the drummers had the most freedom in Zappa’s band. It’s probably true for most bands. From a compositional standpoint, drumming is concerned with rhythm, dynamic contrast and timbre – playing it the right key is not usually a concern (unless your Terry Bozzio). It’s also about knowing how and when to fill. In jazz, skeleton scores are often used in preference to full notation, leaving room for the musicians to improvise. The drum part may consist of nothing more than an indication of the accented beats and where to fill, or the drummer will use the lead sheet and work it out from that. How inventive the drummer is within these parameters is a mark of their musicianship. When Dunbar left The Mothers it was partly because Zappa was producing increasingly notated drum parts which allowed little room for interpretation.

Terry Bozzio was arguably Zappa’s next rock drummer after Dunbar. This is not to belittle the drumming of Ralph Humphrey or Chester Thompson who were both freakishly good, but in terms of raw, animal instinct and showmanship – Bozzio pounded the shit out of his kit. Surprisingly, a couple of days before his audition for Zappa he hadn’t heard any of his music. Bozzio is amazingly open about feeling sick with nerves every time Zappa stuck a piece of music in front of him and is very much under the impression that he would write difficult music just to torment him, a feeling shared to some extent by Colaiuta. Bozzio claims that he never played a creative role in Zappa’s band. He followed orders.

Today Bozzio is well known for two things: having an enormous drum kit, part of which comprises fifteen tom-toms tuned to the diatonic scale, and performing the ‘Black Page’, the drum solo Zappa wrote especially for him. The main challenges of the piece are the irregular beat subdivisions, the complexity of which is a source of unpredictability. The piece is a long, ridiculous fanfare; a rhythmically sophisticated expansion of ta-ta ta-taaaa or maybe its evil twin from a parallel universe… ta…ta-ta-ta-ta-ta…..ta-ta ta….taaa…etc. Listeners may also notice a similarity between the opening melody of The Rite of Spring (played on the bassoon) and first few notes of the Black Page which are an approximate inversion; similar in melodic shape but upside down. They are not altogether different rhythmically either. Bar 15 would appear the most absurd because it contains a broken septuplet and a quintuplet within a triplet. It’s the sort of thing that requires amazing skill to perform accurately.

With the theory of Dynamic Attending in mind, the presence of note groupings in the Black Page is evident, yet their details are unpredictable. This is especially the case in the closing bars (27 and 29) where a barrage of tom-tom fire confronts the listener. To understand this section better, it is worth considering the probability of a random sequence, such as one hundred coin flips. Statistically, a head followed by a tail has the same probability as a head followed by a head, and a tail followed by a tail. The implication is that a truly random sequence of any length consisting of two states (heads and tails) should contain 50% repetition and 50% alternation. If a six-sided die is used to generate a sequence the probability of alternation is 83% (i.e., 5/6) and of repetition is 17% (i.e., 1/6), so the likelihood of repetition diminishes as the number of outcomes increases. Since bars 27 and 29 of The Black Page are performed over six drums we would expect them to have similarly weighted probabilities were they to be random. In fact, by counting the repetitions and alternations of these two bars it is revealed that bar 27 comprises 94% alternations and 6% repetitions, and bar 29 comprises 70% alternations and 30% repetitions. In other words, both bars contain note sequences that are statistically similar to patterns which might result from rolling a dice but whereas bar 27 underestimates the amount of repetition, bar 29 overestimates it. If we take an average for both bars, though, we arrive at 82% alternation and 18% repetition, which compares much more favourably to the dice model. Obviously this method of analysis has its limitations. It does not take into account the unpredictability of rhythm, merely the unpredictability of sequences; however, it does add another level of statistical density and help explain why some things sound more unpredictable than others. Whether Zappa did use a dice to generate bars 27 and 29, we will never know. Against this notion is the fact that the patterns are organised in groupings which make sense musically, and are consistent with a style of writing used elsewhere in the piece; however, it would be wrong to discount the possibly that he was attempting to write a sequence which might be perceived as random.

Chad Wackerman later put his own spin on this kind of random-sounding drum onslaught with his ‘random/hold’ technique. His approach was based on a setting commonly found on synthesisers which when activated produces a continuous flow of random pitches or data used to control other settings. The technique differs from Dunbar’s unpredictable beat placement on “Chunga’s Basement” in that it is less about rhythm, more about sequences. Whereas Dunbar achieves a groove primarily between the snare and bass drum, Wackerman expands the field of possibilities by performing ‘random hold’ fills on the entire kit. When Steve Vai transcribed Zappa’s solos for The Frank Zappa Guitar Book, he included several drum parts improvised by Colaiuta. The scores are full of irregular note sub-divisions, some of which (Zappa admits) are the result of inconsistencies in tempo (notes being rushed) and are a little misleading; however, Vai’s transcriptions are for the most part an accurate representation of what is heard. Sometimes unusual note groupings are written across the bar line. One might imagine that such complexity could only exist in Zappa’s music. In fact, it represents the kind of rhythmical notation taught to Berklee students.

Leaving the band only to be replaced by Colaiuta left Bozzio feeling depressed. He thought he’d set a precedent. He had; however Colaiuta was unbelievably good. Unlike Bozzio, Colaiuta was a Zappa fan before he joined the band. He even transcribed the Black Page and played it at his audition. As a boy he would practise at every available opportunity, a habit which continued on tour, whether in a hotel lobby or an airport waiting room. At school he would sit at the back of the English class with a practise pad and practise double-stroke rolls. Invariably he would get kicked out of class. Such delinquent behaviour is reminiscent of the adolescent FZ bringing his guitar to school.

Things became a bit more interesting when Colauita enrolled at Berklee to study drums with Gary Chaffee, the then Head of Percussion. Chaffee is well known for his four volume series of ‘Technique’ books. They represent to drummers what Nicholas Slonimsky’s Thesaurus of Scales and Melodic Patterns represents to players of pitched instruments. Virtually every possible sticking combination is examined, every possible beat subdivision, subdivided. In volume one, he deals with odd-rhythms, polyrhythms and mixed time signatures. What’s interesting is that even the early exercises have a frightening look about them. He mixes irregular note groupings and inserts rests in ways that appear counter-intuitive. As the book progresses the exercises get progressively difficult; the hapless drummer expected to switch between subdivisions whilst timekeeping foot ostinatos chug along. Rhythms are stretched and distorted in ways that make the ‘Black Page’ seem relatively easy, although it never gets as perverse as bar 15.

It’s no surprise then that Colaiuta’s treatment of the unguessable beat is derived polyrhythmically. On pieces such as Stucco Homes, his playing is so intricate, so dazzlingly entwined with Zappa’s solo it’s as if he is actually predicting Zappa’s every move. Vinnie, it would seem, can play anything. Not only this, Vinnie does play anything. These days he’s a top session musician with a substantial list of credits including the Jeff Beck, The Pussy Cat Dolls, Megadeath, Destiny’s Child, the Kids from Fame...

...and then of course….Sting.

Throughout the 1990s he toured with Sting. In an interview from around the time of Sting’s Seven Summoner’s Tales, Colaiuta declared that Sting’s ‘smoothing out’ approach to irregular rhythm patterns so that they would feel as natural as 4/4 was closer to his own than the ‘angular’ approach favoured by Zappa. Colaiuta played the Sting gig with all the musicality one would expect from a world-class drummer but anyone wishing to hear astonishing outbursts of optometric abandon as featured on SUAPYG and Joe’s Garage would ultimately find Colaiuta’s playing on the leash, his combustible vapours exSTINGuished. Zappa liked to chop from one meter to another, an idea he probably picked up from examining The Rite of Spring score but one also that proved to be a practical solution when recreating musique concrete-style edits with a live band. Sting’s easy listening sensibilities, however, are more inclined to mask the edges, smoothing out the changes from one time to another to produce radio friendly songs in 7/8.

It is quite possible to listen to ‘Chunga’s Basement’ without even realising the drum pattern is unpredictable. In some ways it’s a different take on the ‘smoothing out’ concept favoured by Sting. Dunbar does not disrupt the flow; he maintains it. Although his rhythmical choices are unpredictable they are not unpredictable like, for example, waiting for the last few bits of popcorn to pop. He remains in 4/4. This contrasts to ‘The Black Page’ which is written in 4/4 but doesn’t sound like it is. The popcorn comparison is more relevant. It’s not only hard to predict when a group of notes will occur, it’s also hard to predict the sequence of notes that will occur within each group.