How Hooker found his boogie: a rhythmic analysis of a classic groove


Popular Music ( 2009 ) Volume 28/1. Copyright © 2009 Cambridge University Press, pp. 19–32
doi:10.1017/S0261143008001578 Printed in the United Kingdom

How Hooker found his boogie:
a rhythmic analysis of a classic
groove

F E R N A N D O B E N A D O N † a n d T E D G I O I A ‡
†American University, 4400 Massachusetts Ave NW, Washington DC 20016, USA
E-mail: fernando@american.edu
‡Plano, TX, USA
E-mail: tedgioia@hotmail.com

Abstract
This article closely analyses the rhythmic components in John Lee Hooker’s boogie. We show how
Hooker recasts a signature riff from a ternary to a binary beat subdivision, paving the way for the
triple-to-duple shift that characterised mid-century American popular music. Further, we
attribute the boogie’s ‘hypnotic’ feel to two psychoacoustic phenomena: stream segregation and
temporal order misjudgement. Stream segregation occurs when the musical surface is divided by
the listener into two or more auditory entities ( streams ), usually as a result of timbral and
registral contrasts. In Hooker’s case, these contrasts occur between the guitar groove’s downbeats
and upbeats, whose extreme proximity also blurs their temporal order. These expressive effects are
complemented by global and gradual accelerandos that envelop Hooker’s early performances.

Introduction
John Lee Hooker’s recording of ‘Boogie Chillen’ in November 1948 stands out as one
of the greatest, and most unlikely, hits in the history of blues music. Composers of
popular songs often claim that a ‘hook’ is essential to a successful recording, so if
‘Boogie Chillen’ had a hook, surely it was a rhythmic one.1 The song’s melodic and
harmonic content is practically nil. It is mostly an up-tempo talking blues, spoken or
semi-spoken in a type of blues recitative. Nor did Hooker rely on clever arrangements
and slick studio effects – so common during these post-war years – to secure a hit
song. He performed solo, with only his guitar and stomping foot for accompaniment.
Yet this stark, idiosyncratic song proved to be a runaway hit. ‘Boogie Chillen’ climbed
to the top spot in Billboard’s R&B charts, and propelled Hooker from obscurity to fame
almost overnight. It is hard to imagine the lyrics as the source of the song’s appeal;
they are simple, almost simple-minded, little more than a series of short declamatory
sentences such as:
I heard everybody talking about . . . the Henry Swing Club . . .
I decided I’d drop in there that night . . . and when I got there . . .
[Sung] I said yes, people . . . yes, they was really having a ball . . .
[Spoken] Yes, I know . . . [Spoken rapidly] Boogie Chillen!

The song, in fact, is more a loose improvisation, a stream-of-consciousness
monologue let loose over a repetitive guitar groove. This groove, or what Hooker

19



called his ‘boogie’, proved to be so irresistible that it eventually formed the basis of
more than one third of his early recorded output. Why did the boogie hold such
appeal for audiences, who purchased the recordings of ‘Boogie Chillen’ and its many
knock-offs and imitations?

Blues historians have described Hooker’s boogie as a ‘rhythmic drone, twisting
and turning the sound’ ( Oakley 1997, p. 220 ), a ‘jumping, polyrhythmic groove’
( Murray 2002, p. 127 ) that can ‘rattle your bones’ ( Davis 1995, p. 218 ). Equally
colourful is the rhetoric of popular magazine writers. According to Downbeat maga-
zine’s obituary, Hooker’s boogie had a ‘soulful, grinding edge. Once Hooker locked
into a boogie . . . the most powerful hurricane could not knock him out of his groove’
( Koransky 2001, p. 16 ). To sound like him, advises Guitar Magazine, try ‘dispersing
with everything but the upbeats [to] get unadulterated, driving syncopation’ ( Ellis
2005, p. 110 ).

Hooker himself had difficulty in describing the origins of his boogie. ‘It’s just
there. I can’t explain it. And it just comes out’ ( O’Neal and van Singel 2002, p. 204 ).
When pressed, he would sometimes cite his step-father Will Moore, a guitarist who
played in the Clarksdale, Mississippi area before World War II, as the originator of this
distinctive boogie rhythm. But Moore is a puzzling figure who does not figure in any
of the oral histories related to the Delta blues, making it difficult to trace specific
linkages between Hooker’s own style of playing and the Delta tradition.2

Although the early history of Hooker’s sound may elude us, this paper hopes to
unravel some of his boogie’s visceral and hypnotic aspects by focusing on its rhythmic
properties. The blues literature has allocated limited space to the subject of rhythm,
but a salient feature that has received some attention is the irregularity of metre and
hypermetre in the music of bluesmen Blind Lemon Jefferson ( Evans 2000 ) and Robert
Johnson ( Ford 1998 ). Titon’s ( 1994, pp. 144–52 ) insightful rhythmic analyses of
‘downhome’ blues cover some of the topics addressed in this paper, including tempo
fluctuation and the interplay of duple and triple rhythms. This last topic is central to
Stewart ( 2000 ), who lays out the rhythmic transformation of popular music in
America as a transition from shuffle ( ‘swing’ ) to duple ( ‘even’ ) rhythms. Our aim
here will be to focus on only a fraction of the rhythmic components that make up
Hooker’s boogie, all the while recognising that his boogie is greater than the sum of
its parts. We shall undoubtedly overlook important elements, unintentionally in
unforeseen ways which we hope other writers will address, and intentionally in
several ways, one of which requires mention. Our analysis is restricted to the first four
years of Hooker’s recorded output ( Detroit 1948 to 1951 ), mainly because this period
finds his boogie in its most original and bare form.

Two tempo regions
Prior to delving into the boogie’s rhythmic detail, it will be helpful to gain a bird’s eye
view of Hooker’s overall rhythmic landscape by examining his choice of song tempos.
These span a wide spectrum of metronome markings ranging roughly from 60 to 220
beats per minute – a testament to the blues’ ‘enormous potential for tempo transfor-
mation’ ( Kubik 1999, p. 86 ). In the middle of this range, however, there is a narrow
strip of unused tempo markings: of the 180-plus songs that Hooker recorded between
1948 and 1951, none is faster than 110 and slower than 130 beats per minute.3 In other
words, there is a gap between these two metronome values that provides a clear-cut
separation between the slow and up-tempo songs. This divide allows us to categorise

20 Fernando Benadon and Ted Gioia



Hooker’s early oeuvre according to two tempo regions, which we will call blues
( 60–110 bpm ) and boogie ( 130–220 bpm ). The sizeable buffer zone between the two
regions underscores their stylistic differences. A blues may be as fast as 110 bpm;
speed it up further and its inherent rhythmic makeup becomes – to Hooker’s ears, we
must presume – unacceptably agitated. Likewise, a boogie begins to sound languid
once it dips below 130 bpm, the tempo of Hooker’s slowest unaccompanied boogie,
‘Talkin’ Boogie’. ( This hypothesis can be tested by artificially slowing the song’s
tempo to 120 bpm with a time-expansion algorithm that retains the original pitch
level, an operation that disintegrates the boogie’s rhythmic feel. )

That Hooker’s boogie should comprise the fast-tempo category is obvious but
not trivial. Fast tempos are generally associated with two characteristics not readily
attributable to the slower paced blues: dancing and physical restlessness – which may
explain why younger people prefer fast tempos ( LeBlanc et al. 1988 ) and why, in a
1968 Downbeat interview, Hooker explained that his ‘jump’ style was intended ‘for the
kids’ ( quoted in Evans 1982, p. 84 ).4 Besides raw speed, a distinguishing feature
between the blues and boogie tempos relates to the ‘free’ speech-like rhythms. In
Hooker’s blues, these brief passages usually consist of a unison ( almost heterophonic )
melody line shared by the guitar and voice, during which the otherwise unambiguous
beat often evaporates momentarily. This rhythmic device occurs only within the blues
tempo region ( although not in every song ). Of course, it makes sense that these free
rhythms should be absent from the boogie, since their unpredictability would jar the
groove- and dance-inducing regularity of up-tempo music. Even though the spoken
and sung passages in boogies do exhibit considerable rhythmic freedom, they float
above a steady pulse without impinging upon it.

Duplets, triplets, and the mutable isoriff
With the exception of the free rhythms mentioned above, the principal organising
force in Hooker’s blues and boogies is the triple subdivision of the beat, sometimes
completely filled in and sometimes with a tacit middle subdivision that produces a
long-short rhythm. 12/8 is generally recognised as the de facto metre of blues music,5

although many exceptions can be cited. For instance, Charley Patton’s first country
blues recordings from 1929 omit the triplet entirely ( Stewart 2000 ); 12/8 metre does
not figure prominently until his final recordings in 1934. Blind Lemon Jefferson’s
introduction to his ( 1927 ) ‘Matchbox Blues’ makes explicit use of even quavers;
the mostly-triplet feel that ensues in the accompaniment is frequently warped into
moments of duple subdivisions. Robert Johnson’s ( 1936 ) ‘Cross Road Blues’ also
shifts back and forth between two’s and three’s, although the governing metre is
clearly compound ( as in all his other slow blues from that decade ). Sometimes,
Johnson’s rhythm undergoes a gradual microrhythmic morphing from triple to duple
and back, as in the alternate take of ‘Ramblin’ On My Mind’ ( Benadon 2007 ).6 And
the first-ever recorded vocal blues, Mamie Smith’s ( 1920 ) ‘Crazy Blues’, inhabits
the metrically liquid zone of duple-triple ambiguity typical of jazz. Thus, we are
reminded that within the ( nominally ) duple/triple divide, there exists a rich
amalgam of beat subdivisions.

We also find exceptions to ternary subdivision in Hooker’s music. Within the
blues tempo region, instances of duple subdivision are rare, and they tend to be
microrhythmically derived rather than metrical, as in Robert Johnson’s example
above. In other words, Hooker sometimes tones down the 2-to-1 ratio of the ternary

How Hooker found his boogie 21



long-short pattern by evening out the durations of the two notes, a feature of jazz
rhythm ( Friberg and Sundström 2002; Benadon 2006 ) and possibly of rhythm
production in general ( Collier and Wright 1995 ).

The boogie also exhibits the expected microrhythmic variance typical of long-
short pairs. For instance, in ‘Boogie Awhile’ ( Example 1 ), scarcely any beats break
down cleanly into a triplet – the long-short ratio being frequently skewed in favour of
the second note in the pair – so much so, that at times the durations are unambigu-
ously even. More significantly, Hooker challenges many assumptions about musical
logic by producing a rhythm that sounds long-short no matter how it is approached –
upbeats to beats or vice versa. Normally, if a rhythm’s beat/upbeat pattern is long-
short, then its upbeat/beat pattern should be the reverse, short-long. But not here. The
duration ratios for Example 1 are shown in Table 1. Values near 2.0 represent a ternary
long-short 2-to-1 ratio between two adjacent notes, values near 1.0 represent an even
1-to-1 ratio ( two equidurational quavers ), and intermediate values ( such as 1.5 )
denote evened-out long-short patterns. Two concurrent relationships are at play:
beat/upbeat ( x:y ) and upbeat/beat ( a:b ) pairs. The foot acts as a metronome that
provides the beats; they are subdivided by the guitar’s upbeats. These x:y ratios are
usually tripleted ( shown in bold type ), but enough of them are not to preclude a tidy
triplet feel from forming. Still, the overall feel produced by foot beats and guitar
upbeats is long-short ( x greater than y ), as expected in compound time. But the guitar
line in isolation reveals a puzzling piece of evidence. The a:b ratio measures the
guitar’s upbeats ( a ) against its – rather than the foot’s – beat attacks ( b ). To agree with
the long-short x:y ratio, the a:b ratio should be short-long and therefore less than 1.0 ( a
short upbeat followed by a longer duration on the beat ). In other words, the value of
a should be smaller than the value of b. However, in this case the guitar follows the
opposite route: a is almost always longer than b, yielding a:b ratio values above 1.0.
This occurs because the downward guitar slides lead to constant delays with respect
to the foot. The result is a kaleidoscopic rhythm consisting predominantly of long-
short patterns regardless of whether one listens to pairs of beats/upbeats ( x:y ) or
upbeats/beats ( a:b ).

Beyond microrhythm, duple subdivision often plays a central and explicit role,
as with the transformed version of the motive shown in Example 2( a ). Commonly
referred to as the Elmore James riff ( because he featured it on ‘Dust My Broom’ in

Example 1. ‘Boogie Awhile’ ( 0:58–1:03 ).

22 Fernando Benadon and Ted Gioia



1951 ), this motive was in fact used rather extensively more than a decade earlier by
Robert Johnson.7 This riff, which we will refer to as the isoriff because it is isochronous
and ‘isophonic’ ( a single note or chord comprised of pitches from the tonic triad or
tonic dominant ), appears in almost all of Hooker’s Detroit recordings. Even though
the isoriff is inherently a triplet, Hooker’s boogie often converts it into a duple
construct, as shown in Example 2( b ). The reason for this might be purely practical,
since maintaining the speed of the triplet at a boogie tempo may have translated into
a technical feat beyond Hooker’s reach.8 Instead of abandoning the isoriff during the
fast songs, Hooker essentially retains the motive’s basic physical speed, allowing it to
fit comfortably into a binary rather than ternary subdivision of the beat. In a sense, the
slower note value compensates for the faster tempo. As mentioned previously, the
spontaneous alternation of two- and three-way beat subdivisions is hardly exclusive
to Hooker’s boogie. But the importance of his duple isoriff should not be understated,
for it lies at the heart of what Stewart ( 2000, p. 296 ) calls a ‘sweeping metric shift in
popular music that began in the 1950s’. He notes that as the decades passed, ‘the
underlying rhythms of American popular music underwent a basic, yet generally
unacknowledged transition from triplet or shuffle feel ( 12/8 ) to even or straight
eighth notes ( 8/8 )’ ( ibid., p. 293 ).

Figure 1 illustrates the changing note values that comprise the isoriff as it
migrates across tempos.9 Notice how the isoriff operates within well-defined tem-
poral boundaries. Its notes are about 300 milliseconds at their slowest and 150 ms at
their fastest – regardless of their actual music notation values. No boogie contains

Table 1. Durational ratios for ‘Boogie Awhile’ ( see Example 1 ). The x:y ratio
measures beats/upbeats between foot and guitar; the a:b ratio measures
upbeats/beats in the guitar. Values above 1.0 denote long-short patterns.

Beat Measure 1 Measure 2 Measure 3 Measure 4

x:y a:b x:y a:b x:y a:b x:y a:b

1 1.6 1.2 1.1 1.5
2 2.0 1.4 1.9 0.9 1.1 1.0 1.9 1.1
3 1.7 1.3 0.8 1.9
4 2.0 1.2 1.7 1.3 2.0 1.3 2.1 1.2

Example 2( a ). Triple isoriff.

Example 2( b ). Duple isoriff.

How Hooker found his boogie 23



triplet-quaver isoriffs because they would be too fast ( less than 150 ms per note ).
Conversely, quaver isoriffs are too slow ( more than 300 ms ) for the blues and are
therefore absent from that tempo region. Hence, the isoriff can take on either form
provided it lies within the rather narrow temporal band of 150–300 ms. It is interesting
to note how the quaver (! in the graph ) dovetails roughly where the trendline of the
triplet-quaver (") trails off, providing a link across the blues/boogie tempo gap
mentioned earlier. The graph also shows that at sufficiently fast tempos, the isoriff can
take the form of crotchet triplets (d), or even crotchets ( × ) if the tempo is faster still.
This metrical fluidity suggests that the isoriff’s character is defined more by the raw
physicality of successive strokes than by the actual note values, which result from
accommodating the riff’s technical demands to the given tempo. Thus, a 194 ms-per-
note isoriff can comprise quavers at 156 bpm ( as in ‘Good Business’ ) or crotchet
triplets at 206 bpm ( as in ‘Devil’s Jump’, not shown ). Some tempos make it possible
for more than one type of isoriff to appear within the same song, as in ‘My Baby’s Got
Something’, ‘Talkin’ Boogie’, and ‘Goin’ Mad Blues’. At times, the boogie begins with
a fanfare-like isoriff that is so metrically nondescript that the tempo remains a mystery
until the isoriff subsides and the foot stomping joins in. For instance, the opening
isoriff of ‘She Left Me By Myself’ can be heard as being made up of triplets ( if this were
a 92 bpm blues ) or quavers ( a ‘slow’ 140 bpm boogie ); instead, the entrance of
Hooker’s foot, which will tap at a rate of 184 times per minute, reveals the thumb
downstrokes to have been crotchet triplets. A similar guessing game opens ‘Boogie
Chillen 2 ( Jump Chillen )’, where the isoriff accelerates as much as it disorients. In
‘Boogie Chillen 2 ( I Gotta Be Comin’ Baby )’, the metric relationship between foot-
stomp and isoriff is so tenuous – likely a result of Hooker changing his mind as to the
song’s appropriate tempo – that one must wait another minute-and-half for the return
of the isoriff to confirm the crotchet triplet guess as correct. The isoriff, then, is a
metrically malleable figure par excellence ( London 2004 ). Metric malleability occurs
with melodic or rhythmic figures that can be interpreted in more than one metric
context, although performers usually disambiguate these kinds of patterns with such
cues as expressive timing and dynamics ( London 2004, p. 79 ). In the isoriff’s case, the
disambiguating detail consists of a grace note-like slide or hammer-on that Hooker
sometimes injects every three or six strokes if the isoriff is a triplet or every eight if it
is a duplet.

Even though a detailed analysis of blues guitar technique is beyond the scope of
this article, our discussion of the isoriff leads naturally to some brief observations
about Hooker’s right hand approach. There are several versions of the isoriff, ranging
from single-string picking to multiple-string chord bashing. Two of the isoriff types
employ the thumb only, either to pick a single-note rhythm in the low or middle
register, or to brush full chords with thumb downstrokes. In another version of the
isoriff, the thumb and index finger work together isorhythmically, with the thumb
providing the chord’s bass note and the index finger brushing two or three treble
strings upstroke to complete the chord. Sometimes the thumb rests, leaving only the
index or middle finger to brush upstroke dyads or triads in the treble register.
According to Cohen’s ( 1996 ) classification of blues picking styles and hand postures,
Hooker’s mixed approach to timekeeping places him in the ‘utility-thumb’ category,
in contrast to the ‘dead-thumb’ and ‘alternating-thumb’ players, who confined the
thumb to basic beat-keeping either on a single bass string or alternating between two
low strings.10 More specifically, Cohen sees Hooker’s boogie as a hybrid of utility- and
dead-thumb approaches, ‘a younger style’ of the post-War era ( ibid., p. 472 ).

24 Fernando Benadon and Ted Gioia



Hearing double
The underlying rhythmic engine of Hooker’s boogie is the perpetual cycle of alternat-
ing long beats and short upbeats. This pattern appears in many composite guises that
include strumming, foot taps, muffled strings, picked monophonic lines, staccato
chords, and even ‘loud’ rests, a term devised by London ( 2004, p. 87 ) to describe
unexpected silences on strong beats. Iyer ( 2002 ) believes that long-short patterns are
conducive to stream segregation, the phenomenon whereby the mind separates the
auditory stimulus into two or more ‘auditory objects’ ( Bregman 1990 ). Such rhythmic
unevenness is also present in jazz, where the ‘swung’ eighth-note

facilitates the perception of higher level rhythmic structure. An immediate consequence of the
swing feel is that it suggests the next level of hierarchical organization. In conventional terms,
the swung eighth-note pairs are perceptually grouped into the larger regular interval, that is,
the quarter note. If all subdivisions were performed with exactly the same duration, it would be
more difficult to perceive the main beat. The lengthening of the first of two swung notes in a pair
amounts to a durational accentuation of the beat. . . . The observation that swing enhances the
perception of the tactus is no surprise, given its primary function in dance contexts. ( Iyer 2002,
pp. 404–5 )

In Hooker’s boogie, the syncopation generated by the omnipresent upbeats is so
prominently featured that one is led to wonder whether they, and not the beats, carry
the groove’s timekeeping function.11 Hence, the boogie can be seen as a dichotomous
framework in which beats and upbeats vie for control of the listener’s internal clock
via stream segregation. Moore and Gockel ( 2002 ) have suggested that stream segre-
gation is dependent on the degree of perceptual difference between successive
sounds. According to this paradigm, the more two adjacent sounds differ in duration,
pitch height, timbre, and/or loudness, the more likely it will be for the sounds to
segregate. Such is usually the case in Hooker’s boogie, whose beat/upbeat cycle
features all of the above-mentioned perceptual differences. Added together, this

Figure 1. Isoriff note values at different tempos.

How Hooker found his boogie 25



creates the ear-boggling illusion of two parallel, quasi-simultaneous pulse keepers
( Table 2 ).

Moreover, Bregman and Campbell ( 1971 ) have shown that listeners can have a
difficult time determining the order of tones within a fast repeated cycle of highs and
lows. This is because a further consequence of stream segregation is that it hinders
temporal order judgements, especially given a close temporal proximity between
events – as in the boogie’s upbeats and beats ( see Moore 2003, pp. 288–9 for a review ).
Consider what happens to the long-short pattern as the tempo is increased, say from
a slow blues to a fast boogie: the temporal interval between the upbeat ( short note )
and the subsequent downbeat ( long ) decreases. In theory, we could increase the
tempo as much as we please to bring the upbeat closer and closer to the downbeat. At
a sufficiently fast tempo, the two notes’ extreme proximity renders their temporal
ordering ambiguous – that is, the listener becomes unsure of which of the two notes is
played first and which second. This ‘temporal order threshold’ has been determined
to be as large as 200 ms ( Warren et al. 1969 ) and as little as 50 ms ( Winckel 1967 ),
depending on how the experiment is constructed. Within the range of boogie tempos,
the duration of the upbeat triplet-quaver ranges from 100 to 142 ms, which suggests
that at least some temporal ambiguity is likely to arise. Examples of this beat/upbeat
diffusion abound in Hooker’s boogies, as in ‘Momma Poppa Boogie’ ( Example 3 ).
Like a dog chasing its own tail, the upbeat guitar note precedes Hooker’s foot taps by
such a narrow margin ( about one tenth of a second ) that one is dizzied by the blur of
their cyclical alternation. Indeed, at times it appears as if the foot is providing the
upbeats and the guitar the downbeats.

A potent mix
Although other blues guitarists employed boogie rhythms, they typically imitated the
stylistic devices of piano players. This was a popular recipe with blues fans. During a
seven-year period, for example, Jimmy Reed enjoyed no fewer than eighteen hits that
reached the Billboard R&B charts, building most of these performances off simple
piano-type boogie riffs translated to the guitar. Other blues artists – Muddy Waters,
Howlin’ Wolf, Elmore James – may have earned more respect from serious blues
fans and music writers, but none of them could match Reed’s sheer popularity in

Table 2. Stream segregation as a result of beat/upbeat perceptual contrast.

BEAT UPBEAT

Duration Longer
Approximately in a 2:1 ratio to
the upbeat

Shorter
Approximately in a 1:2 ratio to the beat

Pitch Lower
Usually a drone-like bass string

Higher
Usually a treble chord or single-pitched note

Timbre Darker
Thumb on low string or percussive
foot stomp

Brighter
Brushed chord or single-pitched note

Loudness Softer
Sometimes silent

Louder
Usually accented

26 Fernando Benadon and Ted Gioia



the late 1950s. Many garage bands and aspiring blues guitarists imitated Reed’s
performances, which employed simple figures, easy to learn and play. Hooker’s
boogie rhythms were also popular but, in contrast to Reed’s, proved far more difficult
to imitate. He had transformed the boogie from a simple jam session beat to a taut
system of feints and counter-feints that not only played the beat, but played with it as
well.

To be sure, the rhythmic features mentioned above – upbeat/downbeat
proximity, up-tempo vamping, long-short rhythmic patterns – can be found in other
musical styles during the decades leading to Hooker’s first recordings. But Hooker’s
( or Will Moore’s ) contribution lay in the innovative rhythmic concoction of these
elements. More specifically, Hooker’s boogie conflated features from two dance-
oriented black styles of music from the preceding decades: piano boogie-woogie and
big band swing. Piano boogie-woogie is characterised by a relentless left-hand motto
perpetuo within an up-tempo 12/8 feel. This two measure cycle – sometimes referred
to as ‘the sixteen’ because it contains that many quavers – constitutes the style’s
principal driving force. As with Hooker’s boogie, the accented triplet upbeat is a
recurring element. But there is a crucial difference between the two styles. As dis-
cussed earlier, Hooker’s boogie tends to use upper-register ( usually chordal ) accents
on the upbeat, providing a stark timbral contrast to the lower-register thumb notes
and/or foot taps that mark the beat. Boogie-woogie lacks this perceptual differentia-
tion. A style that is replete with accented triplet-upbeats is big band swing. But those
upbeats which do form part of the essential driving force in the rhythm section, such
as the ride cymbal or piano’s chordal comping, are not incessantly cyclical and
therefore not an integral ingredient of the overall groove. And they are not loud
enough, either. The truly salient upbeat accents belong to the brass and saxophone
sections, which, though frequent, are even less systematically cyclical. Hence,
Hooker’s boogie can be said to combine the best of both worlds because its main
rhythmic cell, the tripleted long-short module, employs the pronounced perceptual
distinction of upbeats found in big band swing and the non-stop cyclicality of piano
boogie-woogie. Moreover, the static, drone-like harmony of Hooker’s boogie acts as a
foil to its rhythmic syncopation, a quality that is absent from the harmonic mobility of
the other two styles.

Example 3. Blurred temporal order as a result of beat/upbeat proximity in ‘Momma Poppa Boogie’
( 2:19–2:25 ).

How Hooker found his boogie 27



When Hooker’s boogie came face to face with the piano boogie-woogie in ‘Do
the Boogie’ ( recorded in 1949 and originally unreleased ), his electric guitar avoided
the stylistic contradictions by relinquishing the leading role. This first accompanied
boogie finds him mainly as vocalist and bandleader, with pianist James Watkin and
drummer Curtis Foster carrying most of the rhythmic momentum. Later that year,
Hooker let his groove mingle with that of his two accompanists in different versions of
the instrumental ‘609 Boogie’. What ensued can be described as ( i ) a harmonic
free-for-all of mismatched tonics, subdominants, and dominants; and ( ii ) a rhythmic
tug-of-war that pitted Watkin’s formulaic handwork against Hooker’s personalised
approach. Things would improve slightly over time as, in the words of Neil Slaven
( 2001, p. 6 ), ‘[Watkin] more or less adapted himself to the vagaries of Hooker’s
performance’. The later recordings show that Hooker also made efforts to reach a
stylistic compromise with his bandmates, even if this meant distilling some of the
complexities of his groove.

Accelerating tempos
Without exception, Hooker’s tempos are always on the rise. The thrill infused by
the boogie’s relentless beat is further heightened by a pervasive accelerando that
gradually ups the tempo by anywhere from 6 to 20 metronome points. Even when
the recklessness of the opening tempo seems immune to quickening, Hooker still
insists on cranking up the pace, as in the 200-to-220 bpm boost of ‘Goin’ Mad
Blues’.12 These unidirectional tempo transformations are not unique to his boogie;
they are a universal trait of Hooker’s early recordings, blues included. Admittedly,
many performers and styles of music allow for some degree of tempo drift. Perfectly
stable tempos can only be the outcome of machine-led renditions, such as those
structured around the dictates of a metronome click or drum-machine track. None-
theless, tempo stability is generally positively valued, since being accused of ‘rush-
ing’ or ‘dragging’ the tempo is as ego-damaging as being told that one’s playing is
uninspired or out of tune – the stuff of amateurs. What is particularly striking about
Hooker’s accelerandos is their ever-presence, which suggests that they are as funda-
mental to his musical persona as his stationary harmonies and original strumming
patterns.

Tempo acceleration appears to be an inherent feature of much blues music.
David Evans’ blues transcriptions reveal widespread ‘acceleration of the tempo
across the course of the performance’ ( Evans 1982, p. 45 ), as do Titon’s ( 1994 )
transcriptions of forty-four ‘downhome’ blues. Titon’s sample, which includes
recordings made between 1926 and 1930, includes tempos that speed up in magni-
tudes comparable to John Lee Hooker’s. Other writers have also pointed to specific
instances of accelerando in the performances of Robert Johnson: Ford ( 1998, p. 73 ) in
the second take of ‘Ramblin’ on My Mind’ ( ‘a tempo that accelerates by a proportion
of 5:4 in the third verse’ ), Boone ( 2002, p. 64 ) in ‘Cross Road Blues’ ( ‘Johnson’s
rhythm is similar to [Fred] McDowell’s in its intensity, its regular, swinging beat, and
its gradually quickening tempo’ ), and Backer ( 2002, p. 120 ) in Johnson’s overall
output ( ‘The playing is assured; the time is solid although the tempo increases. This
has the effect of building excitement as the acceleration is gradual and consistent’ ).
Evans ( 2000, p. 94 ) also noticed it in Blind Lemon Jefferson’s ‘Tin Cup Blues’, ‘with
the metronome reading fluctuating from qn = 98 and qn = 106, tending overall to
accelerate from the slower to the faster tempo’. And Keil ( 1966, p. 54 ) alluded

28 Fernando Benadon and Ted Gioia



indirectly to this trend, pointing out that as blues music evolved over time, ‘a broader
spectrum of tempos is found, and the tempo selected is more rigidly maintained’. To
expand this list, we add, in chronological order, four classic recordings that also
undergo a tempo increase: Mamie Smith’s ‘Crazy Blues’ ( 1920, 110–114 bpm ), Pine
Top Smith’s ‘Pine Top’s Boogie Woogie’ ( 1928, 166–176 bpm ),13 Muddy Waters’ ‘I
Can’t Be Satisfied’ ( 1948, 196–218 bpm ),14 and Bo Diddley’s ‘I’m A Man’ ( 1955,
80–86 bpm ).

Accelerandos have also been noted beyond the realm of the blues. ‘Most black
song’, writes Alan Lomax ( 1993, p. 232 ), ‘both post- and antebellum, even when its
mood is somber and serious when the song begins, usually picks up tempo and is
transformed into music that can be danced to before it has been sung to a con-
clusion’. A 1940 report by the Georgia Writing Project’s Savannah Unit describes
drum music at a Daddy Grace religious service in the Georgia sea islands: ‘At first
the procession is orderly and fairly quiet, but as time passes the music becomes
increasingly loud [and] the pulsating rhythm of the instruments increases in
tempo’ ( quoted in Bastin 1995, p. 158 ). Another employee of the Federal Writers
Project, folklorist Zora Neale Hurston, describes a prayer ritual at a southern black
church:
There is in the body of the prayer an accelerando passage where the audience takes no part. It
would be like applauding in the middle of a solo at the Metropolitan. It is here that the
[preacher] artist comes forth. He adorns the prayer with every sparkle of earth, water and sky,
and nobody wants to miss a syllable. He comes down from this height to a slower tempo and is
borne up again. ( Hurston 1934/1995, pp. 873–4 )

Given the foregoing quotes, one is tempted to suspect the accelerando phenom-
enon to have been a prominent feature of early American black musical culture. But
accelerating tempos are also prevalent in numerous musical cultures from around the
world, including Native American powwow ( Goertzen 2005, p. 289 ), Hindustani rag
( Clayton 2000, p. 88 ),15 and the chanted bargaining procedure of Venezuela’s
Yanomami Indians ( Epstein 1995, p. 344 ). Why accelerate in the first place? In music,
gradual increases seem to be preferable to decreases, as may be observed in other
musical parameters besides tempo. For instance, Bach fugues add voices gradually
but remove them suddenly ( Huron 1990a ), and Beethoven’s piano sonatas contain
more crescendos than decrescendos, with the former usually lasting longer than the
latter ( Huron 1990b ). In his 1804 Treatise on the Art of Teaching and Practicing the Piano
Forte, composer/theorist Daniel Gottlob Türk wrote that ‘in pieces of a fiery, violent,
and furious character, the strongest passages must be hastened a little, or played
accelerando’ ( plate shown in Hudson 1994, p. 123 ).16

As hinted by Türk’s didacticism, musical tempo correlates with cardiovascular
and respiratory rates ( Bernardi et al. 2006 ). Gilbert Rouget, in his book Music and
Trance, has noted that trance-inducing music is frequently marked by accelerando and
crescendo, and that these processes are linked to the intensification and dramatisation
of sound as music reaches a climactic moment ( Rouget 1985, p. 84 ). Comparable
situations are not unknown in Western concert music. For example, musicologist
Reinhard Kopiez and his colleagues ( Kopiez et al. 2003, p. 249 ) noticed a slight tempo
increase halfway through a twenty-eight-hour piano performance of Erik Satie’s
‘Vexations’, at which point the pianist reported being in a state of trance. A suggestive
body of work studying the process of entrainment, the tendency of brainwave pat-
terns to imitate external rhythms – a drum beat, a flashing light – still offers us more
questions than answers ( Gioia 2006 ), but the process itself is well known, measurable

How Hooker found his boogie 29



and repeatable. The literature surrounding the ecstatic trances of shamans is
especially rich with examples, and it is perhaps revealing that Hooker’s biographer
Charles Shaar Murray ( 2002, p. 371 ) likens Hooker to a shaman, although Murray
seems unaware of the systematic tendency toward tempo acceleration in the blues
musician’s work.

Housed in these global accelerandos are the three signature rhythmic techniques
described above: metric mutations of the isoriff, perceptually distinct upbeats, and
blurred temporal order of attacks. The fast tempos, stationary harmonies, and fluid
triplet vamps further personalise John Lee Hooker’s now classic groove. These kinds
of rhythmic features – which are so significant in many blues-influenced styles – have
been conspicuously neglected in popular music studies. In this light, Hooker’s boogie
of the late 1940s presents a rich case study of rhythmic expressivity during a pivotal
moment in American popular music.

Endnotes
1. Definitions of what constitutes a ‘hook’ vary.

The nature of rhythmic hooks is examined in
Traut ( 2005 ).

2. David Evans and Gayle Dean Wardlow ( per-
sonal communication ) confirmed that Will
Moore was virtually unknown in the Clarksdale
scene. Evans notes, however, that Hooker’s
style bears some resemblance to the little-
known work of Country Jim Bledsoe and
Clarence London, guitarists who were recorded
in Shreveport, Louisiana in the early 1950s.
Since we know that Moore hailed originally
from Shreveport, we can speculate that
Hooker’s approach may reflect a regional style
centred in Shreveport that his step-father
brought with him to the Clarksdale area. How-
ever, our present knowledge of the early
Shreveport blues tradition is limited, although
this could present an interesting area of research
for later scholars.

3. The exceptions are all post-1951. The tempo of
‘That’s Alright’ is 114–118 bpm; it is the result of
a duo session led not by Hooker but by his
musical sidekick Eddie Kirkland in 1952. ‘Too
Much Boogie’ ( 1953 ) drags in reminiscing slow
motion at 112 bpm. By 1954, tempos in the
120s would not be uncommon, especially in
accompanied numbers such as ‘I’m Ready’
( 126–129 bpm ) and ‘Hug and Squeeze’
( 120–123 bpm ).

4. Some songs, such as ‘Don’t Go Baby’, have a
boogie tempo and a stomping foot, but their
character is more contemplative than rambunc-
tious. A couple of others, such as ‘Low Down
Boogie’, are really slow blues despite their
title.

5. We agree with ‘the apparent four-ness of 12/8’
( Agawu 2006, p. 21, note 42 ). Hence, when dis-
cussing ternary subdivisions of the beat, we
consciously interchange the concepts of 4/4
triplet-quavers and 12/8 quavers.

6. Clearly, neither the duple nor triple subdivi-
sions receive metronomic treatment.

7. Examples include ‘I Believe I’ll Dust My
Broom’, ‘Honeymoon Blues’, ‘Kind Hearted
Woman Blues’, and ‘Ramblin’ On My Mind’.

8. Pianists did not face same the same challenge as
guitarists. For an example of triplet-quaver iso-
riffs at a boogie tempo, listen to ‘Do the Boogie’
( 1:33–1:41 ). At 200 bpm, James Watkin manages
to articulate the isoriff at a tremolo-like rate of
ten attacks per second.

9. Millisecond values do not correspond to exact
measurements of the audio signal. Rather, they
are calculated arithmetically based on the
song’s approximate tempo. For example, a qua-
ver is 200 ms long at a tempo of 150 bpm,
although the actual durations as performed will
be naturally higher and lower.

10. Cohen ( 1996, p. 465 ) also describes the ‘t-1 run’,
a technique whereby the thumb and index
finger alternate attacks on a single string, like a
plectrum, to play scalar passages. This tech-
nique was more prevalent in the East than in the
Delta. Although Cohen is careful to note that
players are likely to employ more than one
approach to hand posture and finger picking, he
maintains that these physical aspects are closely
intertwined with regional styles.

11. A milder form of this effect was noted by Titon
( 1994, p. 146 ): ‘When the syncopation is con-
tinuous, the natural result is a conflict of pulse
patterns ( polyrhythm ) between the vocal and
the accompaniment’. But such instances last
no longer ‘than two and a half beats’ – a stark
contrast to Hooker’s perennial syncopation.

12. In some instances, the poor quality of the acetate
recordings results in a gradual fall in pitch – and
a( n uncharacteristically ) steady tempo. For
instance, the ‘Stomp Boogie’ tempo of 164 bpm
undergoes a negligible increase of two points,
while the pitch drops by about 75 cents. Restor-
ing the ending pitch ( by ear ) to match that of
the beginning reveals that the song originally
accelerated to 172 bpm. Similar re-enactments
of otherwise tempo-static performances of

30 Fernando Benadon and Ted Gioia



‘Miss Pearl Boogie’ and ‘Boogie Awhile’ reveal
comparable tempo changes ( 170–178 and
190–202 bpm, respectively ).

13. This is not a gradual accelerando; the tempo
changes rather abruptly towards the end of the
song.

14. But on the other side of that hit record, ‘I Feel
Like Going Home’ maintains the tempo
constant at 78–79 bpm.

15. Hooker’s accelerando resembles two of the four
types distinguished by Clayton ( 2000, pp. 88–9 )
in Hindustani music: ‘( a ) Gradual and slight,

and perhaps unintentional’ and ‘( b ) Gradual
but significant; resulting in increased tension
and excitement’. ( The other two are ‘( c ) Step-
wise; a conscious acceleration at a particular
point in a performance’ and ‘( d ) Temporary; for
example to serve the needs of a tabla solo’. )

16. More recently, composers as diverse as Conlon
Nancarrow ( Gann 1995, pp. 146–72 ) and John
Adams ( Adams et al., pp. 96–7 ) have often used
gradual accelerandos to animate the musical
surface.

References
Adams, J., Jemian, R., and de Zeeuw, A.M. 1996. ‘An interview with John Adams’, Perspectives of New Music,

34/2, pp. 88–104
Agawu, K. 2006. ‘Structural analysis or cultural analysis? Competing perspectives on the ‘‘standard

pattern’’ of West African rhythm’, Journal of the American Musicological Society, 59/1, pp. 1–46
Backer, M. 2002. ‘The Guitar’, in The Cambridge Companion to Blues and Gospel Music, ed. A. Moore

( Cambridge, Cambridge University Press ), pp. 116–29
Bastin, B. 1995. Red River Blues: The Blues Tradition in the Southeast ( Urbana, IL, Illini Books )
Benadon, F. 2006. ‘Slicing the beat: jazz eighth-notes as expressive microrhythm’, Ethnomusicology, 50/1,

pp. 73–98
2007. ‘A circular plot for rhythm visualization and analysis’, Music Theory Online, 13/3

Bernardi, L., Porta, C., and Sleight, P. 2006. ‘Cardiovascular, cerebrovascular, and respiratory changes
induced by different types of music in musicians and non-musicians: the importance of silence’, Heart, 92,
pp. 445–52

Boone, G.M. 2002. ‘Twelve Key Recordings’, in The Cambridge Companion to Blues and Gospel Music, ed. A.
Moore ( Cambridge, Cambridge University Press ), pp. 61–88

Bregman, A. 1990. Auditory Scene Analysis ( Cambridge, MA, MIT Press )
Bregman, A., and Campbell, J. 1971. ‘Primary auditory stream segregation and perception of order in rapid

sequences of tones’, Journal of Experimental Psychology, 89, pp. 244–9
Clayton, M. 2000. Time in Indian Music: Rhythm, Metre, and Form in North Indian Rag Performance ( Oxford,

Oxford University Press )
Cohen, A.M. 1996. ‘The hands of blues guitarists’, American Music, 14/4, pp. 455–79
Collier, G., and Wright, C.E. 1995. ‘Temporal rescaling of simple and complex ratios in rhythmic tapping’,

Journal of Experimental Psychology, Human Perception & Performance, 21/3, pp. 602–27
Davis, F. 1995. The History of the Blues ( New York, NY, Da Capo Press )
Ellis, A. 2005. ‘How to play like John Lee Hooker’, Guitar Player, January, p. 110
Epstein, D. 1995. Shaping Time: Music, the Brain, and Performance ( New York, NY, Wadsworth Publishing )
Evans, D. 1982. Big Road Blues: Tradition and Creativity in the Folk Blues ( Berkeley, CA, Da Capo Press )

2000. ‘Musical innovation in the blues of Blind Lemon Jefferson’, Black Music Research Journal, 20/1,
pp. 83–116

Ford, C. 1998. ‘Robert Johnson’s rhythms’, Popular Music, 17/1, pp. 71–93
Friberg, A., and Sundström, A. 2002. ‘Swing ratios and ensemble timing in jazz performance: evidence for

a common rhythmic pattern’, Music Perception, 19/3, pp. 333–49
Gann, K. 1995. The Music of Conlon Nancarrow ( Cambridge, Cambridge University Press )
Gioia, T. 2006. Healing Songs ( Durham, NC, Duke University Press )
Goertzen, C. 2005. ‘Purposes of North Carolina Powwows’, in Powwow, ed. C. Ellis, L.E. Lassiter and G.H.

Dunham ( Lincoln, NE, Bison Books ), pp. 275–302
Hirsh, I.J., and Sherrick, C.E. 1961. ‘Perceived order in different sense modalities’, Journal of Experimental

Psychology, 62, pp. 423–32
Hudson, R. 1994. Stolen Time: The History of Tempo Rubato ( Oxford, Oxford University Press )
Huron, D. 1990a. ‘Increment/decrement asymmetries in polyphonic sonorities’, Music Perception, 7/4,

pp. 385–94
1990b. ‘Crescendo/diminuendo asymmetries in Beethoven’s piano sonatas’, Music Perception, 7/4,
pp. 395–402

Hurston, Z.N. 1934/1995. ‘Spirituals and Neo-Spirituals’, in Folklore, Memoirs, and Other Writings, ed. C.
Wall ( New York, NY, Library of America ), pp. 869–874

Iyer, V. 2002. ‘Embodied mind, situated cognition, and expressive microtiming in African-American
music’, Music Perception, 19/3, pp. 387–414

Keil, C. 1966. Urban Blues ( Chicago, IL, University of Chicago Press )

How Hooker found his boogie 31



Kopiez, R., Bangert, M., and Altenmüller, E. 2003. ‘Tempo and loudness analysis of a continuous 28-hour
performance of Erik Satie’s composition ‘‘Vexations’’ ’, Journal of New Music Research, 32/3, pp. 243–58

Koransky, J. 2001. ‘Boogie endin’: John Lee Hooker: 1917–2001’, Downbeat, 68/9, pp. 16–17
Kubik, G. 1999. Africa and the Blues ( Jackson, MI, University Press of Mississippi )
LeBlanc, A., Colman, J., McCrary, J., Sherrill, C., and Malin, S. 1988. ‘Tempo preferences of different age

music listeners’, Journal of Research in Music Education, 36/3, pp. 156–68
Lomax, A. 1993. The Land Where the Blues Began ( New York, NY, Pantheon Books )
London, J. 2004. Hearing in Time: Psychological Aspects of Musical Meter ( Oxford, Oxford University Press )
Moore, B. 2003. An Introduction to the Psychology of Hearing ( San Diego, CA, Academic Press )
Moore, B., and Gockel, H. 2002. ‘Factors influencing sequential stream segregation’, Acta Acustica –

Acustica, 88, pp. 320–32
Murray, C.S. 2002. Boogie Man: The Adventures of John Lee Hooker in the American Twentieth Century

( New York, NY, St. Martin’s Griffin )
Oakley, G. 1997. The Devil’s Music: A History of the Blues ( New York, NY, Da Capo Press )
O’Neal, J., and van Singel, A. 2002. The Voice of the Blues: Classic Interviews from Living Blues Magazine

( New York, NY, Routledge )
Rouget, G. 1985. Music and Trance: A Theory of the Relations Between Music and Possession, trans. B. Biebuyck

( Chicago, IL, University of Chicago Press )
Slaven, N. 2001. Liner notes to The complete John Lee Hooker, Detroit 1949–1950, Body & Soul 3067872
Stewart, A. 2000. ‘ ‘‘Funky drummer’’: New Orleans, James Brown and the rhythmic transformation of

American popular music’, Popular Music, 19/3, pp. 293–318
Titon, J.T. 1994. Early Downhome Blues ( Chapel Hill, NC, University of North Carolina Press )
Traut, D. 2005. ‘ ‘‘Simply irresistible’’: recurring accent patterns as hooks in mainstream 1980s music’,

Popular Music, 24/1, pp. 57–77

Discography
John Lee Hooker, The Complete John Lee Hooker. Body & Soul; vol. 1 Detroit 1948–49, 3057012, 2000; vol. 2:

Detroit 1949, 3063142, 2000; vol. 3: Detroit 1949–50, 3067872, 2001; vol. 4: Detroit 1950–51, 3074242, 2003;
vol. 5: Detroit 1951–53, BS 2500, 2004; vol. 6: Detroit/Miami 1953–54 ( BS 2653, 2005 )

32 Fernando Benadon and Ted Gioia