Jazz Field Evolution, 1930-1969: from Centralized to Decentralized Field
Abstract
By reconstructing the social networks among Jazz musicians, from 1930 to 1969, I obtain a
topography of band leaders and their respective styles. The field of Jazz makes a transition
from a centralized to a more decentralized one. The findings suggest one might explain the
locus of new style emergence by the field’s structure evolution. When the field is centralized,
innovations are born in the core of the field. When it becomes decentralized, one observes
innovation at the fringes. Also, the uncoupling of structural and capital variables helps to
explore the dynamics of social change. Finally, conflict might precede the transition between
centralized and decentralized, suggesting an intermediary transitional stage.
1. Introduction
The discussion on legitimacy in the production of art has been present at least since
Plato’s writings. As a matter of fact, the question whether an art work is appropriate in
relation to a given ethical, religious or philosophical authority has been the core of art studies.
From this perspective, the investigation on the development of artistic styles has been
traditionally coupled with theological or philosophical accounts.
Although Danto was not the first scholar to criticize this perspective, he was probably
one of the first contemporary philosophers to formalize the separation of philosophy and art
(DANTO, 1964). Danto establishes the idea of “Art World”, which encompasses artists,
critics, curators, gatekeepers and audience. In this article, Danto claims that only accepted
members of art worlds are entitled to establish aesthetical judgments on art pieces produced
by this community. Although members of an art world might borrow philosophical arguments
in order to legitimize and rank art pieces, the production of art is not ultimately subordinated
to philosophical claims. In other words, it is the art world’s member’s responsibility to set the
boundary between art and non-art1. For Danto, members in an art world produce aesthetical
judgments based on assertions grounded on theories of beauty. These assertions are not
necessarily arrayed in a coherent way. Moreover, states Danto, the richness of an art world is
given by the plurality of assertions that coexist.
1
Danto explores Duchamp’s art as examples of pieces that would be conventionally classified as mundane
objects. They are considered art pieces due to the reception and judgment of members in the XX century Art
World.
1
Becker (1982) argues that Danto inaugurated the possibility of institutional analysis of
artistic production. From this perspective, the evolution of styles would not be considered as
subsidiary to the societal development, but gains the status of an autonomous object for
investigation. To be sure, art worlds are not insulated from society. Technological,
demographic, legal, political, economic and other factors do impact the evolution of art
worlds (PETERSON ; ANAND, 2004). However, art world members have to interpret and
enact exogenous impacts in order to take action (SCOTT, 2001). As a result, the response to
external stimuli is also affected by institutional constrains. Moreover, actors might interpret
these stimuli in different ways, which might produce conflicting views on the community’s
future.
The relative buffering of artistic spheres from the other social spaces has been object
of study of classic sociology. Weber, for instance, claimed that the constitution of an
autonomous artistic sphere occurs in tandem with the process of rationalization and
modernization. The uncoupling of economic, religious and aesthetical values permits, from
Weber’s perspective, the emergence of autonomous spheres in a society. As a result, the
source of legitimacy for the artistic production is increasingly found inside artistic
communities (WEBER, 1995). Bourdieu (1993a,1993b) takes both Weber’s and Danto’s
intuition in order to elaborate his conception of artistic field. For Bourdieu, actors in a field
engage in a plurality of purposes. These purposes might be contradictory, which leads actors
to clash in order to establish legitimacy to their claims.
Frith (1996) identifies in Bourdieu’s writings three sources of legitimacy that
commonly clash inside an artistic field. First, art is considered as an intellectual production
and correlates with the expression of social distinction (BOURDIEU, 1984a). From this
perspective, art judgment is backed by critics who are usually linked to academic institutions
(GREENFELD, 1989). Conversely, the elites enact and support the legitimacy of academic
institutions in order to establish the boundaries between low and high-brow art forms
(DIMAGGIO, 1982). In music, the classical music is the best example of genre linked to the
academic source of legitimacy.
The second source of legitimacy is the consumption market. Sheer success in sales is
found to motivate many artists to follow fads and fashions. Although aiming at commercial
success is considered as deconsecrating the artistic sphere, several artists insist that their
purpose is not to please transcendental philosophical claims, but to provide entertainment to
their audiences. From this perspective, insists Frith, the live experience and relationship
2
between artist and audience justifies the commercial appeal. Country music is probably the
best example of genre grounded on the commercial source of legitimacy (PETERSON, 1997).
Finally, Frith identifies the folk roots as a source of legitimacy. Folk artists claim that
they produce pieces that follow a consolidated tradition. Attempts to include foreign elements
to this tradition are seen with suspicion. Also, commercial success is regarded as deleterious
to the continuity of folkloric heritage. Finally, Frith depicts folk art as belonging to the dayby-day habits of a community. As a consequence, the boundaries between audience and artist
are murky, as all individuals involved share the same cultural background. The Blues is one
example of genre close to folkloric roots.
Peterson (1997) believes that all styles are grounded in one or another source of
legitimacy. Although not all styles migrate across sources of legitimacy, Peterson (1972)
developed a model where Jazz would have crossed these three institutional boundaries along
its history. In its inception, in the dawn of the twentieth century, Jazz was closer to its
folkloric roots. As it achieved commercial appeal, it was transformed to entertain larger
audiences. Finally, its artists promoted the transformation of Jazz in a high art form, and
institutionalized it in academic courses. The evolution of Jazz from folk to high art form
should not be seen as a general law. On the contrary, Peterson takes pains in explaining that at
each stage several exogenous and endogenous factors led Jazz musicians to pursue this path.
He is explicit in arguing that if more advanced recording technology existed in the forties,
jazz might have never crossed the high art boundary.
Although Peterson is careful in avoiding a teleological account of Jazz evolution, his
description of a macro process tends to overshadow the inner conflicts within the Jazz field
along this path. Also, it may lead us to believe that once the jazz field crossed an institutional
boundary, all its members were converted to a new ethos. In contrast, I will show that the
evolution of the Jazz field is not explained by the substitution of legitimacy sources, but by
the increasing plurality of legitimacy sources. Once this is accepted, then the challenge is to
understand how the field evolved internally towards this stage and what new field logics
emerged in tandem with this plurality of styles.
2. An Institutional Account of Jazz Evolution
The history of an art field might be told from several perspectives. A musicologist
would choose those records that were landmarks in the evolution of a style. Such history
would depict how innovations like dissonant scale changes, atonal chords and instrumental
3
development were created and diffused in the field. Such effort would necessarily need the
support of musicologists and critics in order to codify several albums under these music
categories.
A historian of music would be interested in the macro societal trends that occurred in
tandem with and might have influences the evolution of a music. Such factors include the
technological changes, demographic and taste shifts, economic development or slowing down,
and the industrial organization of the phonographic and entertainment sectors. Although these
elements are crucial for understanding the macro surrounding of the development of a style,
they do not suffice for exogenous impacts are translated to fields through specific endogenous
mechanisms. Although my focus will be on the internal mechanisms, I will not overlook those
crucial societal changes, neither the more important musical innovations.
I want to pinpoint those processes that constituted the internal sources of legitimacy
and mechanisms in the functioning of the field. I argue that these processes interact to each
other and explain the emergence and evolution of styles in the field. These elements are: (1)
the emergence of a professional body of musicians, (2) formalization and rationalization of
Jazz, (3) the emergence of a professional body of critics, and (4) the consolidation of
academic institutions that supported directly or indirectly the education of jazz musicians.
2.1. Emergence of a Professional Body of Musicians
New Orleans is celebrated as the geographic place where Jazz was born (GIOIA,
1997; BERENDT ; HUESMANN, 1998; LOPES, 2002; GRIDLEY, 2003). It is there where
prominent African-American musicians like Louis Armstrong and Sidney Bechet started their
careers. Most musicians assembled around schools bands, temples and entertaining venues.
From this perspective, Jazz was present in several spheres of a musician’s life: at his
community and at his job place. The continuity of day-by-day life and professional
performance is typical of music styles linked to folk roots (FRITH, 1996).
Trained black musicians moved away from concert music (DEVEAUX, 1997, p. 55).
In comparison to this body of African-American musicians, we observe an emergent group of
White musicians educated in conservatories. A high-brow/low brow relationship is
established, where the “formally educated” musicians are regarded as superior in artistry to
their self-educated counterparts. As Swing emerges in the thirties, white and conservatoryeducated musicians and black and self-educated musicians started to play together.
4
A young generation of white musicians like Benny Goodman, Bix Biederbeck, Bunny
Berigan, Frank Trumbauer, Artie Shaw, Red Nichols, Red Norvo, Tommy and Jimmy
Dorsey, were trained by sweet bands like Paul Whiteman, influenced by the original “hot”
New Orleans’s vernacular. As romantic or deviants, they clashed with black musicians who
struggled to establish a serious professional ethos (LOPES, 2002, p. 143).
The commercial success of Swing led several African-American musicians to join the
jazz market at large cities like New York and Chicago. An army of leaders and sidemen start
to interact in jam sessions, where they could exchange information, learn and get gigs. The
jam session served as a rank mechanism, where musicians disputed for status among their
peers. A musician’s rank was based mostly on his capacity of improvising (LOPES, 2002, p.
96). This contrasts sharply from the sweet band’s ethos (LOPES, 2002, p.147). Jam sessions
were also important to allow musicians to develop new styles and buffer them from
commercial pressures. Finally, musicians looked for sheer pleasure with their peers
(DEVEAUX, 1997, p. 30; LOPES, 2002, p. 151-153).
The practice of jamming – groups of musicians improvising
outside regular commercial jobs – represented the extent to which
this practice defined these musicians’ artistry (…) It was a
performance space in which musicians could articulate an ethos
independent of commercial popular performance conventions.
This separate cultural space as ritual became the locus for various
transformations in the ethos of professional musicians. The first
transformation, of course, was improvisation becoming the
defining skill of this professional class of musician. Eventually,
however, this autonomous cultural space was where professional
musicians developed the ethos of improvisational jazz as a distinct
‘art’ separate from commercial popular music. (LOPES, 2002, p.
153)
Following Parsons (1959), we observe an institutionalized way socializing new
members and dealing with competition. Musicians frequently played different songs and
styles at jam sessions and at Swing concerts. At the former venues, they played for
themselves. While at the later venues, they followed the band leader’s request. I suggest that a
mechanism similar to decoupling (MEYER ; ROWAN, 1991) occurred with this separation:
musicians ritualistically behaved in one way in front of a white middle class audience, while
behaved in another way at jam sessions. As Scott (2001) suggests, decoupling mechanisms
should not be understood as dichotomist phenomena, as if decoupling entailed in complete
buffering against external influences. On the contrary, Swing orchestra leaders frequently did
absorb innovations originated at jam sessions.
5
The Bop revolution (DEVEAUX, 1997) promoted the music played at the jam
sessions as the “real” Jazz, while Swing would be a commercial distortion. I suggest that Bop
was a revolution not only in musical terms, but as an evidence of the affirmation of the body
of professional Jazz musicians as hegemonic in the field. As a result, whatever was produced
at jam sessions was now of interest of most jazz fans. To be sure, the jazz audience mingled,
as it turned more intellectualized. However, those who attended to Bop performances were
loyal to what they believed to be the true Jazz.
As the jam session mechanism became more tightly coupled with the commercial side
of the music industry, social attainment became closer to achieved rather than ascribed
features (PARSONS, 1959). The consequence was a crystallization of social hierarchies: a
musician’s career was linked to a ladder grounded in the jazz community. Moreover, the
higher the importance of this hierarchy, the higher was the centralization of musicians in the
field. In that sense, the Bop revolution helped to reinforce and strength the earlier hierarchical
ladders. Jam sessions phased away in the fifties for several reasons: emergence of real state
speculation in mid-Manhattan, watering of the disputes at jam session, and sheer ritualization
of the improvisation.
The rearticulation of musicians took place around associations during the sixties.
Many of these associations were responsible for training and launching many Free Jazz
musicians (GIOIA, 1997).
2.2. Formalization and Rationalization of Jazz
Weber (1995) describes how the Western music evolved from its folk roots to the
classical vernacular. Within this process, Weber sheds light on the process of rationalization
and formalization of music. The adoption of music sheets, the creation of standards and
rational division of notes occurred in tandem with the process of modernization in the
Western world. To be sure, this process was supported by different social groups. In medieval
times, monks were in the forefront in this process. During the eighteenth century, in contrast,
a professional body of musicians pushed this process forward, with the inception of tempered
scales (ISACOFF, 2003).
The formalization and rationalization in music supports the normative constitution of a
professional body. Those who are skilled in a particular set of rules are accepted by its
professional community. While the formalization is important in order to create and diffuse
common language and norms among musicians, the process of rationalization occurs in
6
tandem with the establishment of professional musicians as an autonomous source of
legitimacy. We may compare the Weberian account with Bourdieu’s description on how art
fields become autonomous (BOURDIEU, 2002). As art fields become autonomous, its
members will generate and share cultural capital, which becomes a distinctive feature of its
participants. Nevertheless, Bourdieu doesn’t expect this process to follow a linear path.
A similar process may be observed in Jazz, but its evolution follows an oscillating
movement between improvisation and formalization. In its origins, the New Orleans and
Classic Jazz had little formalization. Musicians used to improvise barely relied on any written
records. In contrast, the emergence of the white bands (either in New Orleans, Sweet Bands,
or Big Bands) was marked from its inception with the introduction of musical sheets. The
formalization of several Jazz elements helped its absorption in the emergent American Classic
music, exemplified in the work of George Gershwin, Rhapsody in Blue. The diffusion of
music sheets led many African American musicians to learn how to read and write them.
Ironically, during the thirties, when the Swing bands were in their apex, and AfricanAmerican bands played to white audiences, the former had to memorize music sheets, for
white audiences could not bear the idea that African-American musicians could read. In spite
of these contradictions, Duke Ellington is probably is considered one of the most important
composers of the Swing era (WILLIAMS, 1983).
Berliner (1994) describes the evolution of Jazz as a constant tension between the
freedom of improvisation and the boundaries imposed by the composition. When Charlie
Parker and Dizzy Gillespie started their experiments with Bop, a new wave of improvisation
took place in Jazz – mostly based on soloist improvisation. Although the improvisational
feature had regained a foothold in commercial venues, it didn’t take long until bebop started
to be codified in music sheets. As with Swing, the formalization of Bop helped to its
diffusion. By the late fifties, musicologists already observed exhaustion in the Bop paradigm.
Several pundits described many improvised solos as previously rehearsed and thus
unauthentic.
When Ornette Coleman introduced his version of Free Jazz style, one of the most
important features was the return to the collective improvisation that existed in the New
Orleans and Classic Jazz. It is not surprising, for that reason, that so many European
musicians took a leap from Trad Jazz (New Orleans and Classic Jazz) directly to Free Jazz –
the collective improvisation basis was a springboard in this transition.
7
2.3. Emergence of a professional body of critics
I suggest that the evolution of the body of critics in Jazz followed three stages: (a)
ideal consumer, (b) educator and (c) narrator of the Jazz evolution.
The first issues of Metronome and Down Beat, traditional jazz magazines, already
brought critical appraisals on Sweet and Swing bands. “Since the new jazz art world was to be
built on connoisseurship, jazz criticism as the domain of special connoisseurs was a regular
refrain among early jazz critics.” (LOPES, 2002, p. 178). Most of these accounts were
concerned in portraying the experience these critics had at the concert. As a consequence, the
reader was led to become identified with the critic as his representative (WHITE, 1993;
CAVES, 2000). Musicians were not always in agreement with the critics’ assessment of their
work. Duke Ellington’s complained on how his work was analyzed became a historical
landmark in the Jazz field, for it called both for the legitimacy of a body of critics, but at the
same time, the formalization of criteria (LOPES, 2002, p.199).
In tandem with the increasing success of Swing, many fans and musicians grew
concerned with how new listeners were introduced to the field. Their fear was that many new
listeners would take the commercial side of Swing as the true Jazz. As a consequence, many
of these acknowledged fans became critics themselves, and took up the role of educators of
the masses of new listeners. The emergence of these body of critics led to what Gendron
(1995) called the first war in Jazz. One group of critics, in the early forties, defended that
“Swing” was harmful for “Jazz”, and the true jazz was that played in New Orleans style. In
contrast to standardized “riffs”2, widely used in Swing, these critics defended that musicians
should rely on collective improvisation. In contrast with European influences, they should
stress the African roots. A dialogue between critics and musicians was established (LOPES,
2002, p.171), where critics at small magazines like “Record Changer” defended the return to
the New Orleans tradition, while critics at Metronome and Down Beat defended the modernist
push of Swing (LOPES, 2002, p. 201). Charged with the attack that Swing was too
commercialist, the critics at Down Beat and Metronome started to develop criteria to
distinguish between “art” and “commerce” in Swing. Duke Ellington would be in the former
group, while Glenn Miller would be in the latter.
2
"[A riff] is a short melodic ostinato, usually two or four bars long, which may either be repeated intact (strict
riff) or varied to accommodate an underlying harmonic pattern. The riff is thought to derive from the repetitive
call-and-response patterns of West African music, and appeared prominently in black-American music from the
earliest times." The New Grove Dictionary of Jazz.
8
Critics who supported Traditional Jazz triggered more profound studies on Jazz, which
spurred the quest to define Jazz’s boundaries. At the same time, they intended to impose
obstacles to inner development in Jazz. As bop players started to introduce new elements to
Jazz, these critics were mostly concerned in maintaining the traditional Jazz intact.
Intentionally or not, these critics defended the buffering of early Black culture from the
European (and white) culture, for only the protected black culture created genuine Jazz (i.e.
New Orleans Jazz, in contrast with the “decadent” bop, LOPES, 2002, p. 189). The end result
was a patronizing position towards Jazz musicians; yet paradoxically, it also represented leftwing intellectuals identified with New Deal policies and against commercialism (LOPES,
2002, p. 194). Nevertheless, the introduction of the discussion around “how should Jazz be
judged” opened the way for the introduction of Bop as a legitimate and modernist style.
Regardless of the conflict between jazz traditionalists and jazz
modernists, early jazz enthusiasts did create the foundations of a
jazz art world that made in many ways the modernist revolt
possible. Ironically, the high art appreciation promoted by jazz
enthusiasts for genuine jazz was easily co-opted by professional
musicians as they occupied the emerging art world. (LOPES,
2002, p. 216)
The clash between Bop (modernists) and New Orleans and Dixieland (traditionalists)
constituted the second war in jazz (GENDRON, 1995). The clash between traditionalists and
modernists was favorable to the latter. As Jazz faced a decline in late forties, with the decline
of big bands, critics in Down Beat and Metronome embraced the Bop trend as an alternative
to revitalize the Jazz world. As a consequence, the role of criticism was not anymore of the
priest, who sanctions deviations from the established tradition, but it was much closer to the
prophet, who saw in new talents the revelations of the Jazz to come:
The critic and musician are fellow acolytes of the first church of
jazz. (…) The magical mysteries of the universe and the teleology
of jazz are decoded by the keepers of the holy secrets of intuition.
(ELWORTH, 1995)3
3
It is striking that Elworth compares the role of a critic to a religious role. Following a sociological approach,
Greenfeld (1989) analyzed the Israeli art market using Weber’s accounts on prophets and priests. The key
distinction proposed by Weber (1952) is between the legal and rational authority of the priest in contrast with the
charismatic authority of the prophet. While the priest would be institutionally linked to the temple, prophets were
seen as isolated individuals, and as such, had greater freedom to challenge the established institutions. Berger
(1963) revised this model and proposed that prophets were themselves linked to local sects; hence the distinctive
feature between prophet and priest is not on whether prophets were institutionally coupled, but to which
institutional sphere he was linked. Also, as a member of a sect, the prophet counted not only with his charisma,
but with reinterpretation of the Divine Law. As a consequence, the prophetic movement offered the elements of
alternative path of rationalization.
9
See for instance how the critic Martin Williams (1983) defends Ornette Coleman as at
the same time the heir of Charlie Parker and the initiator of a new avenue in Jazz. If critics are
entitled and expected to infer an evolutionary pattern from new talents and new albums, it is
unavoidable the clash among these multiple conceptions. During the sixties, Nat Hentoff and
Martin Williams confronted each other through the pages of Down Beat on whether Jazz
albums were to be judged on aesthetical values only, or whether political affiliation was an
important factor. Both had in mind quite different conceptions on how Jazz had and should
evolve. Conversely, if during the Bop era we observe an increasing collusion between critics
and musicians, when Jazz fragments in several styles and tendencies during the late fifties and
sixties, we observe an alignment of factions of critics and musicians (KOFSKY, 1971).
As we will see below, the emergence of academic institutions and courses was crucial
for housing and legitimating the jazz critics (LOPES, 2002, p. 264).
2.4. Consolidation of academic institutions in Jazz
In the early jazz, musicians learned one from each other its “chops” (required skills for
the job). Among these skills, we find ways of making a horn sound like human voice, chord
progressions and collective improvisations. The tacit knowledge, entailed in signals, standards
and etiquette at the stage are still today largely diffused at informal settings (BERLINER,
1994). From this perspective, the academic institutions might seem irrelevant for the first Jazz
musicians. As a matter of fact, musicians like Duke Ellington even avoided the formal
education in conservatories. Because the racial segregation made slim the odds of AfricanAmerican musicians ascend to orchestras, these musicians looked at these settings as more
attractive places for learning. In comparison, most white musicians went to conservatories to
gain formal musical education during the early Jazz and Swing era. As Lopes (2002) points
out, when Swing reaches a larger audience in the U.S., it is the feature of high education that
gave legitimacy to white leaders to conduct the most successful big bands.
This situation was almost unchanged until the inception of Bop. Several biographies
on Charlie Parker showed that he yearned for a formal education. The movie “Bird” on his
life shows the character standing at the gate of Stravinsky’s house, as he hoped for a contact
with the maestro. Around the same period, Miles Davis was already attending the Julliard
School in New York (SZWED, 2002). After few months he gave up his classes, favoring the
jam sessions and informal settings. At some point he claimed that he learned more with
10
Charlie Parker and Dizzy Gillespie rather than at Julliard’s classes. In spite of Miles’ disdain
to his Julliard background, his formal learning helped him to develop the Modal style he
would introduce to Jazz years later.
By the late forties, several white musicians who attended conservatories would blend
Bop elements with classic elements, generating the predecessors and the seeds of the Cool
style (GIOIA, 1992). In contrast, many African-American musicians, in opposition to a
“European” influence into Jazz looked for reintroducing the Blues and other African elements
in the music. This movement towards African elements was later known as Hard Bop, in
opposition to the “Coolness” of white, European influenced musicians (BERENDT ;
HUESMANN, 1998).
During the fifties several new Jazz schools were established, as well as high school
and college Jazz classes are introduced in order to educate new jazz musicians. It was also
during the late fifties and sixties that musicians create associations to develop new styles. This
effort towards formal institutionalization of Jazz (LOPES, 2002) resulted in the consecration
of Jazz as a high-brow art, and the placement of academic institutions as a substitute (or at
least complementary) to the earlier jam sessions4. In addition, the consolidation of academic
courses and research programs on Jazz is important to house critics in academic positions5.
3. A History of Jazz
Chart 1 depicts the evolution of Jazz sessions from 1930 to 19696. It confirms the
insights collected in secondary bibliography on Jazz history. From 1930 to 1934 we observe a
sharp decline in sessions, due to the Great Depression. This is a result of a waning
phonographic industry, which saw its sales decreasing from $106 million in 1921 to $ 5
million in 1933 (DEVEAUX, 1997, p. 121). The decline in sales led to a wave of mergers and
acquisitions in the industry. As a result, Victor, Columbia and Decca emerged as the survivals
(DEVEAUX, 1997, p. 121). Few recording companies controlled a hierarchical and vertically
concentrated industry: “The industry in the early 1930s moved towards a more centralized
organization in radio, recording, publishing, film and live performance, which included a
greater concentration in booking and management of musicians.” (LOPES, 2002, p.100). The
4
Whoever visits nowadays Jazz classes at the New School in New York will observe that improvisation has
become a regular University discipline.
5
For instance, Dan Morgenstern, former jazz producer and editor of the magazine Down Beat, heads today the
Jazz Institute at Rutgers.
6
See in the attachments the section “Jazz Recording Sessions Data”, for the description on the source and
methodology on recording session data.
11
phonographic and entertainment industry became highly concentrated and tightly controlled
by few companies. This was expressed through a “tightly integrated networks of production
and dissemination of cultural content [which] matured by 1930 (DEVEAUX, 1997, p. 118).
Because these companies favored styles with better commercial fit (like Sweet Bands), we
may conclude that the New Orleans and Classic Jazz styles suffered with the Great
Depression in the U.S. and with the contraction in the Music Industry.
The low wages during the depression allowed the assembling of large and hierarchical
big bands (DEVEAUX, 1997, p. 117):
Life in the big bands was always difficult and never as glamorous
as the tuxedos, stage lights, and mirrored balls in the dance halls
would suggest. Bands were loosely strung together groups of men,
most of whom shared little more than music, and on long and
arduous trips their personal habits and problems, their differences
in ages and backgrounds, could make for chaotic life. The bands
were paternalistic at best, despotic at worst, and could make
demands on individuals that were brutally unrealistic. (SZWED,
2000, p. 56).
In 1935, the United States started to present signals of economic recovery. This
recovery is expressed on the increase in recording sessions, which coincided with the increase
in the Big Band and Swing popularity. Benny Goodman’s famous concert in the Carnegie
Hall marked the dawn of Swing’s golden age, as the most popular style in the U.S. Lines of
teenagers packed the streets of New York City and Chicago in order to have the opportunity
to dance with their friends7. The introduction of the Juke Box in the late 1930s helped to
revive the recording industry (DEVEAUX, 1997, p. 126).
7
As Swing penetrated the recording industry, it was possible the emergence of soloist improvisers who would be
precursors to Bop (DEVEAUX, 1997, p. 11)
12
3000
2500
2000
1500
1000
500
19
68
19
66
19
64
19
62
19
60
19
58
19
56
19
54
19
52
19
50
19
48
19
46
19
44
19
42
19
40
19
38
19
36
19
34
19
32
19
30
0
Chart 1: Evolution of Sessions
Source: Summary of data extracted from Jazz Discography
The whole organizational field was marked by a strong centralization. There were few
intermediary agents (William Morris Agency, Music Corporation of America (MCA),
General Amusements Corporation) who played the broker role in the distribution of content
(DEVEAUX, 1997, p. 122; LOPES, 2002, p. 101). White managers booked black musicians
engagements, which led to the emergence of white gatekeepers (LOPES, 2002, p. 113-114).
The forties experienced two important drawbacks in recording session volume. In
1942, just before U.S.’s joined the World War II, musicians went on strike against recording
companies. The ban on recording started on August 1st, 1942 (DEVEAUX, 1997, p.7)8.
James Petrillo, president of the American Federation of Musicians (AFM) was leading the
ban. The American Guild of Musical Artists and the Boston Orchestra followed Petrillo’s
leadership. The motivation for the strike was the fear that recordings would eventually
substitute live performances. As a consequence, the struggle was not for control of recordings,
but for a fixed fee from every sale (DEVEAUX, 1997, p. 295).
The ban led to the creation of several small independent record companies who
complied with the musicians’ requirements (DEVEAUX, 1997, p. 30). Small labels (although
more flexible and willing to decrease margins) could not take musicians bound by contractual
obligations with larger labels. Hence, they looked for new niches (gospel, R&B, etc.)
8
Another ban against the recording companies was called in 1948.
13
(DEVEAUX, 1997, p. 303). One of the unexplored niches was the small-combo jazz. Labels
that followed this lead included Commodore, Blue Note, Signature, Keynote and Savoy.
When the big companies finally folded under the musicians’ pressure9, they followed the
small companies in recording small combo jazz (DEVEAUX, 1997, p. 297).
During the war, recording companies faced important obstacles to produce and
distribute records. First, the material used in the records was imported from Eastern Asia,
which was under Japan’s control. The war in the Pacific Ocean threatened the supply of
shellac, an important ingredient of 78 rpm records (DEVEAUX, 1997, p. 297); hence,
recording companies stopped channeling investments to slow selling and low margin LPs in
the Jazz genre. (DEVEAUX, 1997, p. 297)
Second, fuel and train usage was channeled to war efforts, which made distribution
costly. As a result, the U.S. government imposed harsh restrictions on ‘nonessential’ driving,
from January 1943 until September 1944. This restriction impacted black bands, for they were
more dependent on road for income (DEVEAUX, 1997, p. 244)10. As traveling went down,
the Jazz world became increasingly concentrated in the 52nd street in New York City, where
the majority of black combos played (DEVEAUX, 1997, p. 285)11.
The post-war was a watershed for the Jazz history. The original Swing fans were
growing older, and less prone to attend all-night concerts. In parallel, the black youth was
shifting to Rhythm-and-Blues (closer to African roots and more accessible to the black
workers). Also, the post-war recession helped to depress the music industry. These effects
combined led to a fast decline among Swing bands (LOPES, 2002, p. 217) and the choosing
of Bop as the scapegoat for the relative decline of Jazz in comparison with other genres
(SZWED, 2000, p. 86).
Swing faced a sharp decline after the war for reasons related with the field’s dynamics.
First, several Swing musicians were drafted during the war (DEVEAUX, 1997, p.7). Second,
the increased competition among new big bands forced prices down (DEVEAUX, 1997, p.
143). Third, because the quantity of talent was constant, musicians didn’t have many
constraints in staying at a single band. As a consequence, recognized talent created their own
bands, diluting talent and increasing competition (DEVEAUX, 1997, p. 144). Fourth, as
9
Decca folded in September 1943, Columbia and Victor folded in 1944.
Since white bands had better access to venues, black bands were forced to travel, in order to find gigs. These
trips could include incursions into the U.S. “deep south”.
11
The emergence of the 52nd street as a Jazz hub is due to the availability of venues. As Prohibition ceased to
rule, Jazz was no longer constrained to underground venues (DEVEAUX, 1997, p. 126). Also, in the Spring of
1944 the cabaret tax increased to 30% on venues that promoted public performance. But this tax was not
applicable to instrumental music, like Jazz, which led to the flourishing of 52nd street.
10
14
venues were crowed by white Swing bands in the large urban centers in the north, Black
musicians had to go to Southern states for available audiences. But discrimination was even
stronger there, what made such traveling unattractive (DEVEAUX, 1997, p. 150). As a
consequence, the several dissolutions of black bands led to the over supply of well qualified
musicians, that migrated to the small combo format.
The early fifties staged a fast recovery in the U.S., which may be observed in the
increasing figures in recording sessions. This increase should be qualified for two reasons.
First, due to the shift from 78 rpm to 331/3 rpm technology, more songs could be compiled in
a single album, and the cost of minutes per album decreased. As a consequence, recording
sessions were longer, in order to record more songs. It was not unusual to consume several
days in order to fill up an album. Second, Jazz increased less than other genres, like R&B. As
a result, while the overall music industry volume increased, Jazz was not able to retain its
share. Nonetheless, the apex in the late fifties also coincides with the period when some of the
best known Jazz albums were released: Miles Davis’s “Kind of Blue”, Charles Mingus’s
“Mingus Ah Um”, Dave Brubeck’s “Time Out”, Ornette Coleman’s “The Shape of Jazz to
Come” and John Coltrane’s “Giant Steps” (BARBER, 2004). Szwed (2002, p. 127) and Lopes
(2002) interpret that the surge in quality was an indirect effect of the overall recovery in the
phonographic industry.
The late fifties and during the sixties the Jazz community observed a steady decline in
absolute volume of work, upon the surge of Rock-n-Roll (PETERSON, 1990) and the
strengthening of R&B. A recover in the late sixties is probably due to external influences like
Bossa Nova and the Rock-n-Roll fusion.
Chart 2 depicts the evolution of the number of leaders in my sample. It contrasts with
the evolution of recording sessions in two significant ways. First, from 1930 to the late fifties
we generally observe a steady increase in the total number of active leaders, which expresses
an increase in total resources available for this population. Second, the decline in number of
sessions in the early sixties is much sharper than the decline in number of leaders, which
shows a declining rate of sessions per leader. In other words, leaders were willing to stay in
the industry, even at lower average work volume.
15
700
600
500
400
300
200
100
66
64
62
60
58
56
54
68
19
19
19
19
19
19
19
50
48
46
44
42
40
38
36
34
52
19
19
19
19
19
19
19
19
19
19
32
19
19
19
30
0
Chart 2: Evolution of Active Leaders
Source: Summary of data extracted from Jazz Dischography
3.1. The Styles of Jazz
The AMG’s classification of albums includes 63 different styles in the Jazz genre.
These styles are grouped in “style families”: New Orleans and Classic Jazz, Swing and Big
Band, Bop, Cool, Hard Bop, Soul, Fusion, Free Jazz and Latin/World. I use these families of
Jazz styles along this chapter, for I believe they borrow legitimacy from distinct sources.
3.1.1. New Orleans/Classic Jazz:
The “New Orleans/Classic Jazz” comprises in Chart 3 the styles “Hot Jazz”, “New
Orleans/Dixieland” and “Revival, Trad.”. It is mostly associated with the roots of Jazz and for
many it was considered the true Jazz (PANASSIE, 1973).
As many accounts of the history of Jazz go, Jazz was originated in New Orleans in the
beginning of the twentieth century. As the African-descendent masses of laborers started to
immigrate to the northern industrial cities, the style was spread across the country. Important
names like Louis Armstrong, King Oliver were associated with “hot jazz” in the twenties and
thirties.
16
1920
1930
1940
1950
1960
1970
FOLK
Hot Jazz
Revival, Trad
New Orleans/
Dixieland
Swing,
Big Band
POP
Progressive
Big-Band
AfroCuban
Bop
FINE ART
Easy Listening
Cool, West
Coast
Hard Bop,
Post-bop
Third
Stream
Fusion
AvantGarde, Free
Jazz
Adapted from Peterson (1972)
Chart 3: Jazz Styles’ Time Line
Several contemporary scholars show that this style was considered low-brow for
several reasons (LOPES, 2002; APPELROUTH, 2003). It resembled African music for its
“call-and-response” dynamics, didn’t follow the Western canon of notes, included
improvisation and imitated the human voice (APPELROUTH, 2003, p. 119). With the
inception of Swing, a way of blending earlier Jazz elements with orchestral features, New
Orleans and Classic Jazz was overshadowed by its successful cousin. For years, the word
“Jazz” was replaced by the word “Swing”. In contrast with the Europe-oriented Swing, “hot”
Jazz was stereotyped as black (and hence libidinal) music, generating a monopoly for black
musicians on New Orleans and Classic Jazz (DEVEAUX, 1997, p. 125).
A strong reaction came in the forties, as Jazz collectors looked for the original roots of
Jazz in order to resist the commercial trends in Swing. Lopes (2002, p. 160) depicts them as
“self-proclaimed” Jazz experts whose mission was to educate the audience. This education
could encompass a dismissal of musicians’ desires and ethos. Hugues Panassié, for instance,
claimed that jazz had to be protected from the phonographic industry’s commercialism.
17
DeVeaux (1997) labels such position as “primitivism”, for it aims at keeping the community
of Jazz musicians apart from any musical change.
This group also saw in the emergence of Bop in the forties a threat to their conception
of real Jazz12. From their perspective, Bop represented just another attempt to turn Jazz in
merchandise. Although Williams (1983) doesn’t draw a sharp boundary between these styles,
Peterson (1972) associates this family with the folk roots of Jazz.
If in its beginning mostly African-American musicians played this style, later White
and non-American musicians adopted, diffused and transformed this vernacular.
3.1.2. Swing/Big Band
Early versions of Swing existed at least since 1916 (LOPES, 2002, p. 108). It became
hegemonic during the thirties, as a synthesis of sweet music and hot jazz.
As pointed above, Swing combined both orchestral and original Jazz elements, making
extensive use of big bands. However, we should make a distinction between Swing and its
predecessor, the Sweet bands. Sweet bands were also constituted by large sets of musicians,
and might incorporate some Jazz elements. One famous example was Paul Whitman’s band.
Ironically, he was called the “King of Jazz”, while in reality his music contained very little
Jazz. Nonetheless, it was Paul Whitman who ordered from George Gershwin the piece
“Rhapsody in Blue”, in 1923.
Some of Whitman’s sidemen were able to incorporate some “hot” elements from the
original Jazz in their music. Among them, Benny Goodman was one of the most famous and
popularized Swing among teenagers in 1935. Besides its undisputable commercial success, it
brought together black and white musicians, blurring the established “racial” boundaries. One
of the reasons for this integration was the white musicians’ sheer lack of repertoire. For
instance, Benny Goodman was one of the first white musicians who played with black
musicians (LOPES, 2002, p. 123).
3.1.3. Bop
Bop was born in the jam sessions at the Harlem. In contrast with Swing’s riffs, Bop
playing was unpredictable (DEVEAUX, 1997, p. 425).
12
Louis Armstrong was believed to scorn Bop (PORTER, 2002).
18
The new bop melodies were more angular than pop songs and
older jazz tunes, and the intervals between notes were wider. (…)
[B]op melodic phrases were longer and less repetitious, but at the
same time unevenly structured and irregularly placed. (SZWED,
2000, p. 35).
It is filled with drum accents, erratic “bombs” (SZWED, 2000, p. 36). DeVeaux
(1997) considers Bop as a major revolutionary watershed in the history of Jazz. If before bop
Jazz was strongly associated with folk roots or commercial appropriation, Bop opened the
way for an intellectual appropriation of the Jazz idiom.
Szwed (2000) and DeVeaux (1997) point that Bop was not born out of nothing. The
discontinuity and ambiguity in chord progression was already applied, in a lesser scale, by
Lester Young (DEVEAUX, 1997, p. 113). It was actually a creative blending of elements
already present in the playing of the following precursors: Roy Eldridge (trumpet), Charlie
Christian (electric guitar), Art Tatum (piano), Jimmy Blanton (bass), Lester Young (tenor sax)
(SZWED, 2000), Coleman Hawkins (DEVEAUX, 1997).
DeVeaux (1997) goes beyond the musical elements in Bop. For this researcher,
precursors like Coleman Hawkins introduced the ethos of the Jazz musician as a professional
musician. This ethos favored the development of Bop as closer to an art form, rather than
another fad. Bop musicians, like Charlie Parker and Dizzy Gillespie, pushed the borders in
terms of musical sophistication, while recovering elements from “hot jazz”. The dissonances
and scale changes introduced in Bop made it increasingly harder to dance and less attractive
for the large audience13. Conversely, several Bop musicians developed hostility towards the
dancing audience (DEVEAUX, 1997, p. 24)14.
In contrast, musicians like Dizzy Gillespie attempted to combine bop and big band
elements, in order to gain legitimacy with a commercial side of industry (DEVEAUX, 1997,
p. 338)15. Although attempts to keep Bop closer to the commercial side of Jazz existed, as a
sub-style of swing, it was doomed to fail. Benny Goodman didn’t acknowledge it as late as
1949.
This ambiguous position was hardly sustainable. At some point, Bop musicians gave
up promoting Bop as a dancing style. It eventually became appreciated by smaller groups of
13
Several established Swing musicians (among them, Louis Armstrong, Fletcher Henderson, Tommy Dorsey)
reacted negatively towards Bop. Many argued that Bop destroyed the possibility of dancing with “weird” chord
progression, hurting jazz in general (LOPES, 1997, p. 222).
14
Bop’s image’s association with drugs, spurred by the mass media, reinforced this reciprocal estrangement.
Gillespie was among those who struggle to disentangle Bop from drugs (DEVEAUX, 1997, p. 398).
15
This gives us an idea of Swing institution’s robustness, for Gillespie tried to combine bop and swing for a long
period (DEVEAUX, 1997, p. 439).
19
fans, who attended small clubs, instead of dance halls and carved a niche in the periphery of
the industry in 1945 (DEVEAUX, 1997, p. 364).
While at jam sessions Bop’s harmonic obstacles served as a strategy for winning
cutting sessions, in recording sessions it became codified (DEVEAUX, 1997, p. 311), which
allowed the diffusion and institutionalization of the new style16.
And still, in its pure form it was short-lived, as many small clubs disappeared with the
real-estate speculation of the mid-fifties (DEVEAUX, 1997, p. 442)17.
3.1.4. Cool
The style “cool” is regarded as a natural development from Bop. It is closer to
contemporary classical music. Miles Davis is known as one of the first musicians to play in
this idiom, when he recorded the album The Birth of the Cool in 194918. Yet, many scholars
consider cool mostly associated with White musicians. First, the incorporation of classical
music elements took the idiom away from its African blues roots19. Second, it spread rapidly
among White musicians who attended classic music conservatories. The very word “cool” is
controversial among Jazz scholars. Among some, cool is related to “intellectual”, but also
“European”, in contrast with “hot”, found in Hot Jazz a later in Bop. In contrast, many
scholars see in the word “cool” a link with the African’s value of proud and self-possession
(SZWED, 2000, p. 199).
A derivative of cool is the “West Coast Jazz”, played by musicians based in the West
Coast. Many critics, however, do not consider West Coast as a distinct style.
3.1.5. Hard Bop
Hard Bop emerged in the fifties, mostly as a reaction from African-American
musicians to the presence of white musicians in the recording industry with Cool (LOPES,
2002, p. 252, SZWED, 2000, p. 253). As Cool became strongly associated with European
16
One may relate the formalization effort to the social construction of reality (BERGER; LUCKMANN, 1966).
As Bop became more formalized, it became also more objectified. The neo-institutionalist tradition links the
formalization and codification to the diffusion of social forms in a field (see for instance, LOUNSBURY; RAO,
2004). Bourdieu (1990b), suggests that the codification and formalization are the final outcomes of the process
of fixing forms to phenomena previously not classified. Bourdieu proposes that classificatory systems are stable
because they gain a status of “natural” to social actors. Bourdieu suggests that one of the most important tasks
for the analyst is to describe the “naturalization” of classificatory and code systems.
17
Becker (2002) suggests that a sociology of places should complement a sociology of social actors. He shows
that the Jazz styles evolved in tandem with the different types of venues available to musicians.
18
As Bop became less accessible to the large white audience, Miles perceived that some changes in the music
were required. These changes led to Cool (SZWED, 2002, p. 70)
19
Szwed (2002) suggests that it recovers certain sweet elements from Swing developed by Lester Young.
20
elements and White musicians, many African-American musicians turned to African and
Blues elements in order to express their ethnic identity:
Stronger, more interactive drumming, funky and soulful melodies,
and a reassertion of the primacy of the blues. It is a music that
rejected the reserve of cool jazz and reclaimed the principles of
bebop in a more recognizable and accessible African American
form. (SZWED, 2000, p. 114).
Thus, their playing contrasted with the “coolness” and “softness” of Cool. As a
consequence, they adopted the adjective “hard” in order to contrast their playing to Cool.
Some scholars believe that Hard Bop is nowadays the mainstream idiom in Jazz. In
comparison to Bop, it softened the musical sophistication and made the music more accessible
to the black audience lost to R&B. Also, many claim that Hard Bop is more assertive in
incorporating blues elements. Yet, some scholars find it difficult to define exactly what is the
“blackness” feature in the Hard Bop playing. Berendt and Huesmann (1998) find that this
distinction is arbitrary and is mostly used to distinguish African-American from White
musicians.
Moreover, several scholars suggest that African-American musicians promoted the
label Hard Bop in order to segregate white musicians and revert the “Jim Crow” relationship
(DEVEAUX, 1997, p. 18; LOPES, 2002, p. 254): “Some white critics, especially Nat Hentoff
and Leonard Feather, did point to how the experience of Jim Crow created Crow Jim.”
(LOPES, 2002, p. 255)
One way or another, hard bop was also very associated with the Civil Rights
movement in the fifties and sixties. Max Roach was among the most politicized side of hard
bop (SZWED, 2000, p. 230).
3.1.6. Soul
If Hard Bop attempted to recover African roots and conquer a black youth lost to
R&B, Soul was bolder towards this direction. Musicians like Julan Adderley and Brother Jack
McDuff included soul and funk elements in Jazz, towards a fusion with these genres. To be
sure, it is not just a development from Hard Bop (SZWED, 2000). The further detachment
from the Bop paradigm was not well received by all Hard Bop musicians (GIOIA, 1997).
21
3.1.7. Free Jazz and Avant-Garde
Retrospectively, scholars have found elements of Free jazz in early pieces as in Lennie
Tristano’s in the forties and fifties. But it was only in the late fifties with Ornette Coleman’s
recordings that the Free Style emerged as a distinct idiom.
The free jazz broke with several standards consolidated in jazz since the inception of
Bop. Paradoxically, its defenders were both accused of disrupting the Jazz tradition (LOPES,
2002, p. 264), and at the same time, they were celebrated for resuming the revolutionary
efforts initiated by Charlie Parker (WILLIAMS, 1983). Besides Ornette Coleman, John
Coltrane, Charles Mingus were among the most influent musicians in this style (JOST, 1994).
Most of the resistance against Free Jazz came from mainstream musicians, who claimed that it
“was not based on harmonic progression, it did not rely on modes or scales, it was not
tempered in pitch, and it did not always follow a strict rhythm” (RADANO, 1985, p. 72).
In comparison to other styles, the Free Jazz was well received by critics (RADANO,
1985). The intellectual interpretation of its evolution, and the effort to associate it to the
classic music avant garde movement made Free jazz strongly associated with high art form.
Ironically, however, “avant-garde musicians were unconcerned” with the criticism’s support
(RADANO, 1985, p. 74). In contrast, free jazz musicians disdained the criticism as another
attempt to freeze the Jazz musicians’ creativity in a new label. Yet, it was not as popular as
other styles:
While a more accessible version of avant-garde jazz would later
reach the popular ear, it was, and still is in its ‘purest’ form,
inaccessible to the masses. Musical norms operate like all other
social norms – they change slowly. And the acceptance of the
abstract music of jazz avant-garde could not be forced; only
gradually could it reach popular tastes. (RADANO, 1985, p.77).
3.1.8. Fusion
It is not impossible to find the word “fusion” applied to any blend of Jazz and foreign
idioms20. Nevertheless, this style has been usually associated with the blend of Jazz and Rockn-Roll.
A word of caution should be made here. It is true that Rock was usually associated
with pop and commercial vernacular, especially by actors within the Jazz field:
20
From this perspective, albums like “Sketches of Spain”, by Miles Davis, would be considered fusion. Blends
with other styles, but Rock, were classified under the “Latin and World Fusion” category.
22
From their position on the fringes of the commercial music world,
jazz musicians had always assuaged themselves with the belief
that they were superior musicians, that they had chosen the more
difficult road with their art. (…) As they watched the current
boom in pop music, many harbored the faith that if they wanted to,
they too could sell out and make best-selling pop albums. Yet the
truth is that they seldom had anything to sell a pop market and no
idea about how to make a pop record, and when they were offered
the chance, the results were often comic or dreadful. (SZWED,
2000, p. 283).
However, Rock itself was dramatically transformed during the fifties and sixties. Also,
many of the Jazz elements blended were borrowed from Free Jazz. For that reason, Williams
(1983) considers that Davis actually attempted to cross two boundaries at once. First, Jazz’s
internal boundaries, by embracing free jazz elements, and second, Jazz external boundaries,
by blending it with Rock and introducing electronic instruments.
3.1.9. Latin and World Fusion
As the Bop and its offspring achieved maturity by mid-fifties, Jazz musicians
increasingly incorporated elements external to Jazz into their music. These elements ranged
from Indian music (John Coltrane’s “My Favorite Things”, in the album “My Favorite
Things”, released in 1960) to Brazilian Bossa Nova (see Stan Getz, “Desafinado”, 1962).
Chart 4 depicts the evolution of styles, associated with recording sessions.
100%
Soul
90%
New Orleans and Classic
Jazz
80%
70%
Latin
Hard Bop
Bop
60%
Cool
50%
Fusion
40%
Free
30%
Swing and Big Band
20%
10%
0%
I
II
III
IV
V
VI
VII
VIII
Chart 4: Evolution of Sessions per Style Family
23
Source: Author’s analysis
Note: vertical axis depicts total number of sessions. Leaders classified under multiple styles had their sessions
split proportionally. Horizontal axis depicts periods.
Following, I collected from the secondary bibliography reviewed some of the most
important albums that established new styles and innovations. This list will help me to locate
in the social network those musicians who introduced novelties in the Jazz field.
Style
Musician or Band
New Orleans/Classic Jazz • ODBJ
(and Revival)
• King Oliver
•
Swing/Big Band
•
•
•
•
•
•
•
•
Bop
Cool (and West Coast)
•
•
•
•
•
Multiple
1959
Hard Bop
Revolution
•
in •
•
•
•
•
•
•
Record or Song (Year)
• “I’ve lost my heart in
Dixieland” (1920)
• “Dippermouth Blues”
(1923)
• “West
End
Blues”
Louis Armstrong
(1928); “Satchmo at
Symphony Hall” (1947)
• “Copenhagen” (1924)
Fletcher Henderson
• “Carnegie Hall Concert”
Benny Goodman
(1938)
• “White Jazz” (1931)
Casa Loma
• “Cotton Tail” (1940)
Duke Ellington
• “Jumpin'
at
the
Count Basie
Woodside” (1937)
• “1937-1938” (1938)
Louis Armstrong
Charlie Parker
• “Koko” (1945)
Dizzy Gillespie
• “Cubano
Be/Cubano
Bop” (1947)
Bud Powell
• “Un Poco Loco” (1951)
Thelonious Monk (p)
• “Misterioso” (1948)
Miles Davis (t)
• “Birth of the Cool”
(1949)
Stan
Getz,
Kenny • “People Time” (194X)
Barron
Modern Jazz Quarter • “Django” (1954)
(MJQ)
Gerry Mulligan
• “Bernie’s Tune” (1952)
• “Giant Steps”
John Coltrane
• “Kind of Blue”
Miles Davis
• “Time Out”
Dave Brubeck
• “Mingus Ah Hum”
Charles Mingus
• “The Shape of Jazz to
Come”
Ornette Coleman
Clifford Brown, Max • “Daahoud” (1954)
Roach Quintet
• “Hard Bop” (1957)
Art Blakey
24
•
•
Soul
•
•
•
Free Jazz
Gard
and
(Rock) Fusion
Avant- •
•
•
Horace Silver
•
Julian
“Cannonball” •
Adderley
•
Brother Jack McDuff
Richard
"Groove" •
Holmes
•
Stanley Turrentine
Cecil Taylor
Ornette Coleman
Albert Ayler
•
•
•
•
•
“Senor Blues” (1956)
“Mercy, Mercy, Mercy”
(1966)
“Brother Jack McDuff
Live!” (1963)
“Blue Groove” (1966)
“That's Where It's At”
(1962)
“Unit Structures” (1966)
“Free Jazz” (1960)
“Spiritual Unity” (1964)
“Ascension”
“Jazz in Silhouette”
(1958)
•
•
John Coltrane
Sun Ra
•
•
Archie Schepp
Miles Davis
•
•
“Four for Trane” (1964)
“Bitches Brew” (1969)
Stan Getz
Stan Getz/João Gilberto
•
•
“Desafinado” (1962)
“Getz/Gilberto” (1964)
Latin and World Fusion – •
Bossa Nova
•
Frame 1: Major Art Works by Style and Musician
Sources: Gioia (1997), Szwed (2000), AllMusic.com.
4. Theoretical Development: Field, Institutional Change, and Circulation of Elites
A central concern for this investigation is the understanding of how groups attain
better positions in their social milieu in tandem with the creation or transformation of
institutions. This concern touches on two theoretical bodies: on one hand, a theory of fields,
institutionalization processes, and institutional change and on another the theories on how
elites attain their prominent position. This effort requires the creation of hypotheses that
bridge these two theoretical bodies.
Fields are understood in the neo-institutional traditional as those social spaces where
the involved actors share standards, norms and meanings (DIMAGGIO ; POWELL, 1983). It
is also a locus for struggle, where the involved actors fight for resources and symbolic
hegemony (BOURDIEU, 1993b). Several scholars developed the idea that fields differ in
internal institutional logics, structure and relative autonomy to other fields. Following, I
propose two ideal types for fields: “centralized” and “decentralized”.
25
4.1. Centralized Fields
A centralized field resembles the “normative” field proposed by Anand (2000), and
Peterson and Anand (2002)21. Institutionally, centralized fields are organized mainly under
macro-coercive or normative rules. In the former case, a heteronymous power imposes
common practices. In the latter case, a cohesive professional body enacts common norms.
Structurally22, a centralized field presents a distinctive core (a set of central actors that
hold central positions in the social network) that dominates its periphery by controlling the
majority of resources. Its structure resembles a “star”, as a consequence, the central player
will have a disproportionate betweenness centrality23:
F
G
E
A
B
D
C
“Star” structure
Schema 1: Star Structure
Source: Author’s adaptation of Kadushin (2004a)
The central actors control the field’s schemata. Thus, it is not surprisingly that the
introduction of innovations is usually top-down, held by central entrepreneurs, whose
position is relatively buffered from market pressures. By this account, a classical example of a
normative field is given by DiMaggio’s research on U.S. art museums (1991), where a
handful of central player’s action modified the field’s logic. Nonetheless, it doesn’t mean that
innovations are not created at the periphery. In centralized fields, innovations from the
periphery are gradually absorbed into the core, as incumbent actors are able to incorporate
them while avoiding threatens to their position. This behavior is also supported by
Schumpeter’s (1934) studies on innovation. For him, only oligopolies would be able to
accumulate enough resources in order to introduce innovations.
21
I don’t the term “normative”, for any field, by definition, entails a diffusion of norms and practices.
Here “structure” refers to the social network configuration. It is not directly related to Giddens’s concept of
structure. See “Glossário” for a full discussion on the term.
23
The concept of “betweenness centrality” is present throughout the thesis. See “Glossário” for an explanation
and discussion.
22
26
As Faulkner (1983) shows, central players tend to be at the same time actors who have
a long track record, and concentrate most opportunities in the field. Bourdieu’s reading on
incumbent players goes along the same lines. Players who are “old” in the field tend to
become more dominant and concentrate most market share in an industry (BOURDIEU,
1993b, 2005).
Norms will be stable as long as not perturbed by a sudden change in the market’s
logics or a break in the organizational core coalition (MARCH, 1962). As long as the central
coalition of power is stable (FLIGSTEIN, 2001b), the field institutions should be stable. In
tandem, in an oligopoly, we expect higher homogeneity of art forms, due to a high coercive
dominance of core over peripheral players (PETERSON ; BERGER, 1975). In our example,
this should be reflected in stability of musical styles and tastes. Hence, in centralized fields,
we should find high homogeneity and stability of styles.
4.2. Decentralized Fields
Decentralized fields are similar to Peterson and Anand’s “competitive fields”. They
are much less autonomous in relation to market trends. Competitive forces mark a balance
between innovation and imitation; as a consequence, they are marked by the predominance of
mimetic isomorphism.
Structurally, the field is still articulated by a core, but a more differentiated periphery
emerges. Kadushin (1976) suggests that networks in “circle” shape occur in tandem with the
introduction of innovations in a field. He offers the example of avant-garde painters in the
apex of their creativity: instead of being organized under a hierarchy, they form circles of
innovators. Although one has little power to impose its dominance over others, reputation and
status are central in order to gain access to resources and opportunities (JONES, 2002;
PODOLNY, 2001). The association with prominent players provides legitimacy to those
associated with them. Hence, novice players will be driven to develop joint projects with
already established players in the field. As a consequence, a core will emerge around those
more prominent players, even in the absence of formal sanctioning rules. The proposed
“decentralized structure” takes a mid-ground between the “star” structure and the “circle”
structure. A core is still recognizable, but the periphery is connected as if in a circle24:
24
Future analysis could compare the “star” structure with the “small world” structure (WATTS, 1999). In
comparison with the former, the later requires unconnected “pockets” of cohesive groups in the periphery. In
comparison with the “decentralized structure proposed”, it might provide a better explanation for differentiated
spaces in the field (UZZI ; SPIRO, 2005).
27
F
F
E
G
D
A
B
E
G
C
“Circle” structure
Core
A
B
D
C
“Decentralized” structure
Schema 2: Circle and Decentralized Forms
Source: author’s adaptation of Kadushin (2004a)
This structure occurs in tandem with the emergence of higher agency discretion for
bottom-up innovations25. Leblibici, Salancick, Copay and Kinget (1991) provide the
example of the transformation of the radio broadcasting industry in the U.S, showing that
most innovations were introduced by peripheral actors. Powell suggests that “In the absence
of incentives embedded in existing institutional arrangements, innovation and diversity will
be more likely to come from the periphery of organizational fields or from outside sources”
(POWELL, 1991, p. 198).
Bourdieu’s (1993b, 2002) analyses on art fields show that younger and less powerful
actors introduce avant-garde styles. As these styles become consecrated, the artists who
introduced them become themselves incumbent actors in the field. A Paretian account of
“elite circulation” (PARETO, 1935) would expect that older incumbent artists be expelled by
younger and up-coming artists. In contrast, old and new styles co-exist (SIMMEL, 2004). For
that reason, old and new generations of elite co-exist in the field. Because decentralized fields
tend to promote innovations in a faster way, we should expect that newer artists will achieve
success (market share in number of sessions) faster. If that is true, it follows that age in the
field and market share might not coincide: older artists might present lower success than
their younger competitors.
As suggested above, the higher periphery differentiation and the co-existence of
several generations might lead to a higher heterogeneity of styles. As a result, decentralized
fields present a “fragmented” picture of styles. Also, in a decentralized field, struggle for
25
One must be careful when linking a social network structure with its actors’ underlying dispositions. Swedberg
(2005) shows that a univocal association between these two categories was exactly what Weber harshly
criticized in Simmel’s work. For Weber, sociology could not be wholly developed only on the study of social
network structures. A similar criticism might be found at DiMaggio’s (1993) criticism on research based only o
social network analysis.
28
survival is fiercer for all players. Products have shorter span of life and competitors gain and
loose field’s dominance more quickly. Christensen’s (1997) example of the hard-disk industry
is illustrative of this point. Innovations were hardly introduced by incumbent firms, but by
peripheral players. In addition, as the technology of hard disks evolved, the average time
length of a company’s leadership became increasingly shorter. In other words, although
novice players revolve around older players, the former efforts to displace the latter are more
frequent and more successful than in centralized fields. Hence, styles will remain less time in
a
dominant
position,
given
the
higher
motility
of
its
musicians
and
faster
desinstitutionalization. In other words, there will be lower stability of styles. Please find on
Schema 1 a summary of features in Centralized and Decentralized fields.
Features
Centralized
Decentralized
Sources of isomorphism
Coercitive and normative
Mimetic
Social Network Structure
“Star”
“Decentralized Structure”
Source of Innovations
Central actors, “top-down”
Peripheral actors, “bottomup”
Old actors concentrate most Young actors might achieve
Dominance
market share
High
Styles
homogeneity
stability of styles
high market share
and High heterogeneity and low
stability of styles
Frame 2: Centralized and Decentralized Fields
Source: Author’s proposal
4.3. Control Variables
Many studies on Jazz consulted approached the theme of “race” (some of the most
important references include JONES, 1963, KOFSKY, 1971, DEVEAUX, 1997). In most of
these studies, African-American musicians are depicted as segregated from the American
society. At the same time, their innovations are believed to have been exploited by white
musicians, better positioned at the music industry.
The theme of race is a very controversial one. As far as I know, nobody would defend
that human beings are inherently different, as if a genetic analysis would provide us a
palpable distinction between races. However, we should identify two opposing positions. A
first claims that once “race” is a driver for social segregation, analysts should consider “race”
29
whatever is embedded in the “racist” interpretative schemata. The other position claims that
because “race” is socially constructed, we should avoid this category and focus only on other
(objective) social relations, like class, social strata, etc.
Although this paper doesn’t give priority to “race”, I chose to include it as a control
variable. It is large the amount of bibliography that points that “race” and ethnicity is one of
the most relevant explanatory variables on social segregation. For instance, Tilly (1999, 2003)
defends that social boundaries around ethnicity are created in order to maintain the protection
of privileges (see also LAMONT ; MOLNAR, 2002). Lopes (2000) makes a direct criticism
to Bourdieu’s conception of field for not including the category of “race”. In his opinion, an
“ethnic capital” could help to supplement other types of capitals.
The effect of “race” might be confounded with cultural capital (for more educated
white leaders are compared with non-educated African-descendents). But the causality might
be reversed at this point: low mobility prospects didn’t encourage the accumulation of formal
cultural capital26. As a consequence, African-Americans were led to avoid conservatories and
formal education. It is not surprising that Bebop revolution in the forties helped to reverse this
trend. As African-American musicians started to claim rights, they started to enroll in
academic institutions for a formal education. This came also in tandem with the reinforcement
of the normative strength originated at the professional musician sphere in the late forties. In
that sense, as the internal dynamics in the jazz field changed to benefit achievement rather
than assigned characteristics, we should observe the weight of variable decrease in strength.
I included also “gender” and whether the leader was from the U.S. Both variables
receive less attention in the secondary literature, although recent studies started to flash
important aspects in relation to gender and national origin in the history of Jazz. For instance,
Tucker (2000) analyses the role of women in the development of Swing. Although women
have been present throughout the whole history of jazz27, it is the Swing era when they
achieved the higher prominence. As a result, we should observe a high correlation between
the presence of female leaders and the Swing style.
Berendt and Huesmann, German writers, argue that Jazz has been in its beginning
mostly an American invention. They support such claim by depicting the European Jazz as a
copy of the American Jazz (BERENDT ; HUESMANN, 1998). After the Second World War,
the flow of American musicians to Europe helped to increase the proficiency of European
26
From a different theoretical approach, Fernandes (1969) suggested that the racial segregation in the Brazilian
society could not be reduced to class segregation.
27
As we observe later, on Table 3.1, the percentage of female leaders is always around 10%.
30
musicians. However, claim the authors, it is only when European musicians develop local free
jazz style that authentic European jazz emerges. Conversely, we are able to see an influx of
Japanese, Brazilian and other nationalities to the Jazz world. In parallel, as Americans shied
away from jazz during the fifties and sixties, the foreign audiences increased in relevance to
jazz musicians. As a consequence, we should observe an increase in prominence of non-U.S.
musicians, as well as an increasingly association with avant-garde styles.
5. Data Sources
5.1. Jazz Recording Sessions Data
The source on the recording sessions is Tom Lord’s CD-ROM (LORD, 2005). I used
both the 4.0 and the 5.0 versions for this purpose. This database includes only a subset of the
recording sessions stored in these CD-ROMs. The criteria applied for extracting sessions from
this database was the following. First, the band leader had to have at least twenty sessions
recorded. This set of band leaders amassed over 73,000 recording sessions, or 50.5% of all
recording sessions included in the CD-ROM. In contrast, they represent only 6% of all 26,147
leaders present in the CD-ROM.
Next, I selected only those sessions that fell within the time frame, from 1930 to 1969.
As a result, many leaders were not included, for their sessions were either after 1969 or before
1930. In personal communication, Mr. Lord expressed that many non-jazz artists were
included due to the pressure from CD-ROM buyers. Nevertheless, Mr. Lord believes that in
the coming versions he will be able to exclude non-Jazz musicians. I also asked to Mr. Lord
what was his criteria for including artists in his database. My concern was that he could be
excluding important artists by setting arbitrary boundaries on what is Jazz. In general, Mr.
Lord tended to err in the side of inclusion rather than exclusion of artists.
As a consequence, I triangulated my database with Scott Yanow’s style classification.
As a result, I classified all leaders in spells of styles28. Many artists played other genres, like
R&B and Rock-n-Roll. All those leaders who were never under the Jazz umbrella were
excluded from my database. I ended up with 388,204 sessions29 distributed across 1,112
leaders30.
Some bands didn’t have a single leader. They either had shared leadership or just changed
leaders along their existence. As a consequence, I produced an N by N database matching
28
See below
A leader’s sessions spread out in a single day were grouped in a single session.
30
Actual individual musicians, in contrast with band name.
29
31
sessions to leaders. The major sources of information for this table was the New Grove
Dictionary of Jazz (KERNFELD, 2001), and biographies at the web sites referenced above.
5.2. Style Assignment Methodology
Ideally, each session would be classified under the existing styles by its contemporary
critics. Moreover, I should able to combine the opinions of several experts on each session
and extract an Intercode Reliability Index on each session. Alternatively, as DiMaggio (1987)
suggests, I should be able to locate each recording session in a web of style classification,
which would include not only the musicians’ self assessment, but critics’, producers’ and
distributors’ classification.
Although such material would be certainly rich for analyzing the classification
methods across different groups in the jazz field, that strategy was not pursued for two
reasons. First, the major question in this article is not on the sense-making processes and
schemata around style assignment. On the contrary, I want to understand how the styles
evolved during Jazz’s history. Hence, I was compelled to accept some degree of reification of
these categories. As a result, I take for granted that a current (2005) labeling crystallizes a
consensus on a classification system. The database consulted was the All Media Guide, under
the direction of Scott Yanow. I triangulated several of his and his associates’ classifications
with The New Grove Dictionary of Jazz classifications. There was a high match in
classification, with the exception of the “trad” and “modern” labels, which are usually used in
Europe. For the former, I used AMG’s “big band/swing” and “New Orleans/Classic Jazz”
labels, and for the former I used the “Bop” label.
Second, the large number of recording sessions included in my database would make
it quite difficult to classify them all by interviewing critics, musicians, producers or
distributors. Moreover, many if not the majority of these actors are not alive any more. Even
if I tried to classify most of these sessions using the AMG database my effort would be
frustrated. Many of these recording sessions were never released, or aired for few just minutes
in a radio station. Many of them are lost or buried in collectors’ shelves.
I also considered the strategy of text recognition of record criticism. By scanning
contemporary criticism, I should be able to extract from this material references to styles. This
effort would be also inappropriate for my purposes. First, many references to styles were
negative, rather positive association. Second, many references were unrelated to the record
under review. Finally, much criticism doesn’t make any reference to style, which would
32
make it difficult to assess the soundness of my sample. As a consequence, I would be forced
to read and code each criticism, in order to extract the record classification.
I adopted a more pragmatic approach. Rather than classifying each recording session, I
identified for each leader in my database those records that marked any change in AMG’s
classification. For instance, if a leader’s records were always classified under “Swing” and
“Bop”, I stored as a turning-point event the first record classified under “bop” only. If the
leader returned to Swing, I stored again only the record session correspondent to this change.
As a result, I ended up with a database of turning-point events for each leader, where each
record depicts a change in style classification. To be sure, each event may be classified under
multiple styles.
Next, I assigned all sessions between two turning point events under the first event
classification. In several cases, AMG didn’t present an artist’s early albums. For these cases, I
supplemented the AMG information with other data sources (see above). I also triangulated
several AMG’s classification with these alternative sources. Nevertheless, some left and right
censoring was unavoidable. I treated left-censored cases as missing cases. Right missing cases
were treated as missing cases whenever the gap without classification surpassed a period of
five years.
Below I depict an illustration of the style assignment mechanism:
Left
Censored
Swing
Bop
Swing, Bop
AMG:
Swing
AMG:
Swing,
Bop
Bop, Hard
Bop
AMG:
Bop
AMG:
Bop, Hard
Bop
Styles are assigned to turning points, or “events”. Sessions will receive the styles
assigned to their precedent event. If a session is not preceded by any event, it falls in the “leftcensored” region, and no style is assigned. Conversely, if a session is not preceded by an
event at least five years old, it falls in the “right-censored” region and no style is assigned.
5.3. Musicians’ Demographic Data
The most important source for demographic data came from the All Music database.
At this site I was able to extract the musician’s date of birth, perform a pictorial analysis to
33
identify whether the musician was African-Descendent31, gender, and check the musician’s
nationality32. Alternative sources included the The New Grove Dictionary of Jazz, The Big
bands database plus site, Jazz associations and musicians themselves, cited in the
“Agradecimentos” section.
The number of sessions was obtained from the number of sessions the leader was
present. Years at the field counted difference between the average year in the period and the
year when the leader entered the field (both come from Lord’s CD-ROM). The average
betweenness centrality was obtained with UCINET’s analysis of each period’s social network.
6. Methodology
The analyses performed in this article have two purposes. First, connect the evolution
of Jazz to the proposed ideal types. Second, explore the process throughout the Jazz field
migrated from one type to another.
6.1. MRQAP
It is not always possible to grasp directly a field’s institutional logics (THORNTON ;
OCASIO, 1999). In such cases, this task has to be performed indirectly, through the analysis
of the actors’ behavior and supplemented with qualitative research. In our case, the overall
field’s logic of interest is the association patterns among musicians.
In order to analyze how different factors explain the association among leaders, I let
the square matrix a that relates leaders to be the dependent variable. As independent variable
I used four sets of matrices.
“Power matrices” comprise those privileged variables under Bourdieu’s theory of
field: age at the field (Field Age) and market share (Session number). Because of the
distribution of these variables are strongly skewed towards one, I used the natural logarithm.
The two matrices used as independent variables had for each pair the difference between
leaders. Positive and significant parameters denote a preference for leaders with higher (or
lower) number of sessions or years at the field.
The second set (“categorical variables”) comprises those ascribed characteristics that
impacted the success and association patterns among musicians and a measure of style
31
The challenge here was to apply a categorization that scholars on racial boundaries could reasonably accept. I
would like to make this classification available for third-party cross-checking, as far as necessary to improve it.
32
Tom Lord’s database provides an identification of nationality.
34
dissimilarity. These variables are: style dissimilarity, gender, “race” and whether the
musicians are from the U.S. The dissimilarity measure was calculated in the following way:
first, I created an affiliation matrix of all leaders by all styles (an “n” by “s” rectangular
matrix). Next, I transformed this rectangular matrix in a square matrix f , containing the
Euclidean distance between each pair of leaders. A positive and significant parameter denotes
a preference for leaders who don’t play the same style. The gender matrix contained
information on whether both leaders were of the same gender. If both leaders were women,
for instance, the pair received “1”. In contrast, if one was a male and the other female, the pair
received “0”. Positive and significant parameter denotes a preference for same-gender leader.
I applied a stricter approach for national origin: if both leaders were from the U.S., the pair
received “1”, otherwise, it received “0”. With this approach, I focused on the weight of U.S.
leaders’ endogamy. I used a similar approach to “race”. I created two matrices: Both AfricanDescendents and Neither African-Descents. In the “Both African-Descendents” matrix, if
both leaders were African-descendents, the pair received “1”, otherwise “0”. The same logic
was used to construct the matrix “Neither African-Descents”, but now I coded “1” only for
those pairs where both leaders were not African-Descendents33. The construction of these two
matrices helps us to isolate endogamous tendencies among African-descendents and NonAfrican-descendents.
The third set, “relational variable” comprises a comparison in betweenness centrality.
The resulting matrix denotes the difference in betweenness centrality for each pair of leader.
Hence, a positive and significant parameter is interpreted as a preference for leaders with
higher or lower betweenness centrality.
Finally, the “Styles” set comprises matrices for each style. For each style, I coded “1”
whether both musicians played that particular style, and “0” if any of them didn’t play it. As a
consequence, for each style I obtained a square matrix where the “1” denotes “both leaders
play this style” and “0” for all other cases. A positive and significant parameter is interpreted
as an endogamous preference among leaders who play that style.
The first model relates the first four sets of independent variables to the dependent
variables and its results are summarized in Table 2:
aP Matrix (Diff. Field Age (ln)) + Matrix (Diff. Number of Sessions (ln)) +
33
I also tested a matrix with “same ‘race’” coding. The separation of this matrix in the two matrices mentioned
above decreased the model’ error.
35
Matrix(Style Dissimilarity) + Matrix (Same Gender) + Matrix(Both African-Descendents) +
Matrix(Neither African-Descendents) + Matrix (Both U.S. Musicians)
The second model adds to the model above the matrices by style. Its results are
summarized on Table 3.
I estimated the odds-ratio of a by performing a MRQAP analysis (Multiple
Regression Quadratic Assignement Procedure). This algorithm developed by Dekker,
Krackhardt and Snijders (2005) estimates the odds-ratio of a dichotomic and square matrix.
The advantage of this procedure over other available algorithms is its better sensitivity
analysis to collinearity among the independent variables.
6.2. Blockmodeling
One of my concerns is to reconstruct the evolving macro structure of the field, in order
to understand the constraints imposed to musicians’ action, but also to explore how this very
action affected the field’s structure. In order to obtain a topology of the Jazz field structure
over time, we applied a blockmodeling approach from the Social Networks Analysis tradition.
As noted in the chapter “Fields and Networks”, the methodology of blockmodeling
was developed by White, Boorman and Breiger (1976) in order to extract positions from a
network. Faukner (1983) used a similar approach in order to recognize central and peripheral
blocks among networks of film composers.
The use of blockmodeling has become widespread in organizational research.
DiMaggio (1986) was one of its first proponents by using it to study the relation among
theaters’ managers in the U.S. Mohr’s (1994) study of non-profit organizations in the
beginning of the twentieth century, and Padget and Ansell (1993) investigation of the
relationships among Florentine families in order to understand the rise of the Medici are other
seminal contributions that used blockmodeling. The idea of blockmodeling is to group
together those individuals who share similar patterns of relationships with all individuals in
network (see BREIGER, 2004 for a review on the methodology). As a result, musicians
placed in the same block will be structurally similar, which means that they will be likely to
bear the same pattern of ties to other actors in the network. Nevertheless, structurally similar
actors are not necessarily connected among themselves.
I used the Tabu Search algorithm in order to obtain the networks’ partitions. The user
sets the number of partitions desired. Next, the Tabu Search algorithm builds a matrix c, a p
36
by p ideal matrix, where p is the number of partitions originally set. Next, the algortithm
permutates the orginal a matrix (composed by the n individuals) and modifies c until it finds
the partitions that maximize the correlation between c and a. As Hanneman (2001)
indicates, there is no “good” or “bad” number of blocks. I chose eight blocks, as it would
yield a number of blocks still feasible to analyze, increase the correspondence between the P
and N.
I used UCINET 6.71 in order to obtain the permutated matrix c and its respective
image matrix (Table 4). The image matrices were generated following Breiger, Boorman, and
Arabie (1975). I obtained the image matrix by calculating the density of ties in each block.
Next, I dichotomized density matrix of each period using the following procedure: if the
density of a given block was above the average density of the image matrix, I considered as
having a tie between a pair of blocks. I used the period’s average density as a threshold for
dichotomization of ties among blocks. Table 6 shows correlations for 2, 4 and 8 blocks for
each period. As expected, by using 8 blocks we obtain substantially higher correlations. Table
5 shows the distribution of densities for each period, for the 8-block approach. Periods V and
VIII presented the lowest correlations. It is expected that the distribution be bi-modal, in order
to make the analysis meaningful. Periods II and VIII had the farthest distributions from the bimodal ideal. Chart 5 reproduces the image matrices on table 4.
7. Results
7.1. MRQAP Analyses
Table 2 summarizes the results for the MRQAP analysis for all periods, without the
style variables. As expected, differences in Field Age and Number of Sessions were
negatively related with the likeliness of tie between two leaders. In other words, the stronger
the difference between two leaders’ profiles of age and share of markets, the lower was the
probability of sharing sidemen. Nevertheless, the weight of these factors is not constant. In
period V and VII Age Field comes closer to zero, while in periods III and VII Number of
Sessions becomes less negative.
37
Table 1
Period
Evolution of Leaders, per major Categorical Group
(II)
(III)
(IV)
(V)
(I)
1930 to 1934
1935 to 1939
1940 to 1944
1945 to 1949
1950 to 1954
(VI)
(VII)
(VIII)
1955 to 1959
1960 to 1964
1965 to 1969
Total Number of Leaders
158
213
277
435
537
669
703
695
Non African Descendent
African Descendent
103
55
144
69
184
93
255
180
318
219
396
273
407
296
410
285
Non African Descendent %
65%
68%
66%
59%
59%
59%
58%
59%
African Descendent %
35%
32%
34%
41%
41%
41%
42%
41%
Male
Female
140
18
194
19
250
27
391
44
479
58
596
73
635
68
634
61
Male %
89%
91%
90%
90%
89%
89%
90%
91%
Female %
11%
9%
10%
10%
11%
11%
10%
9%
U.S.
Non U.S.
134
24
162
51
205
72
344
91
413
124
502
167
507
196
472
223
U.S. %
85%
76%
74%
79%
77%
75%
72%
68%
Non-U.S. %
15%
24%
26%
21%
23%
25%
28%
32%
Source: Author’s analysis
38
0.179
0.089
0
25,440
Adjusted R2
0.107
Probability
0
Number of Observations
12,432
Note: Observation refer to ties valued "1"
* P<0.05
** P<0.01
**
**
0.019
-0.022
0.029
0.145
0.201
0.01
-0.039
0.083
0.289
0.218
**
-0.077
-0.073
-0.013
0.014
1935 to 1939
**
**
**
**
**
0.076
0
39,402
0.126
0.001
0.021
0.09
0.111
0.217
-0.073
-0.043
1940 to 1944
**
**
**
**
**
**
0.077
0
78,680
0.187
0.028
-0.01
0.085
0.077
0.156
-0.088
-0.064
1945 to 1949
**
**
**
**
*
**
**
0.052
0
145,542
0.176
0.003
0.009
0.033
0.045
0.116
-0.053
-0.064
1950 to 1954
**
**
**
*
**
**
0.094
0
253,512
0.186
0.011
-0.026
0.068
0.033
0.187
-0.078
-0.12
1955 to 1959
MRQAP Parameters per Period, on Power, Categorical and Relational Variables
(II)
(III)
(IV)
(V)
(VI)
1930 to 1934
(I)
0.031
Power Variables
Diff. Field Age (ln)
Diff. Number of Sessions (ln)
Categorical Variables
Style Dissimilarity
Same Gender
Both African-Descendents
Neither African-Descendents
Both U.S. musicians
Relational Variable
Diff. Betweeness Centrality
Period
**
**
*
**
**
**
(VII)
0.083
0
270,920
0.083
0
-0.009
0.089
0.005
0.107
-0.037
-0.031
1960 to 1964
**
**
**
**
**
(VIII)
0.049
0
295,392
0.148
-0.028
0.009
0.093
-0.031
0.066
-0.043
-0.049
1965 to 1969
**
**
**
**
**
**
**
Table 2
39
Style Dissimilarity was usually not significant, which we may interpret that the overall
style categorization was not an impediment for sidemen sharing. Exceptions are observed in
period IV and period VIII. While in period IV Style Dissimilarity becomes positive, in period
VIII it becomes negative. In the former period, leaders were actively sharing sidemen
associated with leaders affiliated to different styles to their own. In contrast, in the later
period, leaders were shying away from sharing sidemen with leaders associated with different
styles.
As expected, leader’s gender was weakly associated with the formation of ties. In
contrast, it is enlightening to observe how ethnic preferences evolved in time. In period I,
when both leaders were non-African descendent their likeliness of association through
sidemen was much higher than African-American leaders’. This might be interpreted as an
effect of the Jim Crow policies, which segregated African-American musicians from nonAfrican-American communities. While African-Americans were willing to associate with
White musicians, the later presented much of the resistance. This relationship changed
overtime. “Neither African-Descendents” factor approached zero in period VII to become
negative in period VIII. In contrast, the factor “Both African-Descendents” swung around
positive and zero over time. Periods IV, VI and VIII show an increased tendency for
endogamy among African-Descendent leaders.
Not surprisingly, the tendency that American band leaders were connected was always
positive. To be sure, this is mostly due to the fact that for most of its history, Jazz was
concentrated in the United States. Nevertheless, this tendency weakens, as Jazz becomes
shared by communities of musicians outside U.S.
The Betweenness centrality effect is usually positive, as expected. On periods I and
VII it becomes non significant.
At Table 3, I present the MRQAP analyses with the Style Classification factors. We
promptly observe that the Field Age and Number of Sessions factors were barely affected
with this inclusion.
Style dissimilarity was also barely affected, although while before the period VIII
factor was negative, at this analysis it became positive. As we directly account for style
endogamy, the search for diversity becomes again positive and significant.
40
0.093
0
25,440
-0.005
0.064
-0.013
0.005
0.062
0.036
*
0.171
0.03
**
**
0.021
-0.015
0.031
0.141
0.208
0.011
-0.033
0.076
0.3
0.223
**
-0.077
-0.068
-0.015
0.013
1935 to 1939
*
*
**
**
**
*
**
**
0.096
0
39,402
0.053
0.151
0.03
0.018
0.005
-0.004
0.11
0.015
0.036
0.085
0.116
0.234
-0.042
-0.036
1940 to 1944
*
**
**
**
**
**
**
*
*
0.09
0
78,680
0.003
0.053
0.096
0.029
0.014
-0.005
-0.008
0.011
0
0.185
0.028
-0.009
0.08
0.08
0.153
-0.078
-0.059
1945 to 1949
*
**
*
**
**
**
**
**
**
**
*
**
**
0.066
0
145,542
-0.002
0.074
0.077
0.025
0.043
0.014
0.005
0.012
0.023
0.172
0.005
0.011
0.025
0.044
0.113
-0.032
-0.063
1950 to 1954
**
*
*
**
*
**
**
**
**
**
*
**
**
0.098
0
253,512
-0.005
0.031
0.024
0.013
0.024
0.028
-0.004
0.002
-0.001
0.187
0.008
-0.027
0.065
0.034
0.204
-0.071
-0.119
1955 to 1959
MRQAP Parameters per Period, on Power, Categorical, Relational Variables and Styles
(II)
(III)
(IV)
(V)
(VI)
1930 to 1934
(I)
Adjusted R2
0.11
Probability
0
Number of Observations
12,432
Note: Observation refer to ties valued "1"
* P<0.05
** P<0.01
Power Variables
Diff. Field Age (ln)
Diff. Number of Sessions (ln)
Categorical Variables
Style Dissimilarity
Same Gender
Both African-Descendents
Neither African-Descendents
Both U.S. musicians
Relational Variable
Diff. Betweeness Centrality
Styles
Both Leaders play New Orleans
…………………….Swing
…………………….Bop
…………………….Hard Bop
…………………….Cool
…………………….Soul
…………………….Fusion
…………………….Free Jazz
…………………….Latin/World
Period
**
**
*
**
**
**
**
**
**
**
(VII)
0.066
0
270,920
-0.015
0.019
0.022
0.033
0.015
-0.019
0
0
-0.004
0.137
-0.007
-0.022
0.148
-0.019
0.059
-0.086
-0.082
1960 to 1964
*
*
**
**
**
**
**
(VIII)
0.071
0
295,392
0.007
0.03
0.037
0.091
0.011
0.096
0.021
0.018
0.032
0.142
0.045
-0.004
0.067
-0.01
0.06
-0.029
-0.049
1965 to 1969
**
*
**
**
**
**
*
**
**
**
**
**
*
Table 3
41
Surprisingly, little effect is observed on “Both African-descendents” factor. That is to
say: regardless of the style adopted, African-Descendent endogamy remained steady. In
contrast, the non-African-Descendent exogamy observed in period VIII turned to be not
significant, which could mean that this trend was mostly driven by style preference.
With the inclusion of the “Style” matrices, the betweenness centrality difference is
significant and positive in all periods, with exception of Period I.
When we turn to the analysis of Style endogamy, it is worthy to comment on each
one’s evolution and in tandem with their counterparts. New Orleans endogamy is positive
significant at period I and III. While in the former period it is mostly associated with Jim
Crow policies and the low-brow stigma, in the later period in depicts the revival reaction
towards other groups. We observe the highest parameter associated with Swing at Period III,
declining to become non-significant at Period VII.
It is interesting that by the inception of Bop, we observe a negative trend towards
endogamy. Bop leaders were strongly associated with the Swing world: they were actually
former sidemen of prominent Swing leaders. At this early stage, bop leaders attempted to
develop their music in the interstices of an industry dominated by Swing. This factor became
positive and significant, achieving its maximum value at period IV, while declining there
after. Compare the Hard Bop factors with Bop’s. They evolve as if both styles were tightly
linked. Hard Bop achieves a positive and significant parameter at period IV (one later than
Bebop), but declines right after. A recovery is observed on Period VIII.
The trends seem to suggest that endogamy is highest at the moment of inception34:
Cool at Period V, Soul at Period VI, Fusion at Period VIII, and Latin at Period V.
Nonetheless, Period VIII shows a recovery in all parameters (with exception of Cool and New
Orleans/Classic Jazz).
34
I offer two complementary reasons to this fact, which is epiphenomenal to this article’s scope. First, as Weber
(1950) suggests, new social groups who are eager to establish their boundaries become more sectarian than the
mainstream group (Weber contrasts the new Protestant groups to the Catholic group). We find the same
argument in Coser (1966), where closure comes in tandem with social identity construction. A second reason is
incidentally explored in the “Careers in the Right Beat” article. As a leader attempts to establish a new style, he
has to rely on specialized resources. The maintenance of this resources together might occur in tandem with a
higher endogamous tendency, at least until the style is ready to be codified and diffused.
42
7.2. Analyses of Blocks
7.2.1. Period I (1930-1934)
A first inspection of the reduced graph gives us the picture of a shared core: Blocks
I.1, I.7, I.3 and I.6 form a connected square. I.6 serves as a broker to I.4 - I.2 side, while I.5
and I.8 are isolated. Nonetheless, a deeper analysis reveals that I.1 is clearly the dominant
block in this period. Only 13 leaders concentrate 23% of all sessions. Among them, we find
Benny Goodman, Tommy Dorsey and Red Nichols. They tend to be older in the field, with an
average field age of 10.08, and have the highest average number of sessions, 32 (Table 7.1).
In the U.S. circuit, it is the block with highest betweenness centrality (2.08, lower than I.4,
constituted with European musicians, and I.6, which is occupied by a single musician who
performs a bridge between the U.S. and the European worlds). Interestingly, its leaders tend
to play Swing more than New Orleans, which points to the direction of adoption of the
commercial style. Only 8% are African-Descendents (much below the field’s average of
29%). They are all American and mostly White.
Contrast this block with blocks I.5 and I.8. These blocks concentrate most AfricanDescendent leaders. They are not connected among themselves, and have low average market
share (7 and 12 respectively), although their field age are closer to average (Block I.5’s
average is 8.22, close to the field average of 8.28). Block I.5 resembles a lot Block I.1 when
we consider the styles played. Both tend to play Swing, while retaining New Orleans roots.
Louis Armstrong, King Oliver, Sidney Bechet, in spite of being among the most important
innovators in Jazz, are located in the isolated I.5. To be sure, their innovations will be
absorbed by core actors (KOFSKY, 1971). It is also interesting to observe white Swing
leaders like Glen Gray (from the Casa Loma Orchestra) in this block. This suggests that even
those early innovators among white leaders were in close contact with African-descendent
musicians while creating the new style. Duke Ellington and Fletcher Henderson had better
luck than their African-American counterparts. They were located at I.7, closer and connected
to the power center in the field.
Consider now Blocks I.4 and I.2. They are mostly comprised by non-U.S. leaders.
Between them, Block I.2 is the dominant: higher average number of sessions and field age.
Also, Block I.2 is gearing towards Swing, while I.4 is strongly associated with New Orleans.
It is of special interest to find musicians who occupy privileged positions in a network.
During Period I, Jimmy Dorsey occupied a brokerage position between the Non-U.S.
(European and mostly British), and the U.S. jazz worlds (Block I.6).
43
Block III.6
Block III.7
Block III.5
Isolated
Period III – 1940 to 1944
Block I.7
Block III.4
Block III.1
Block I.3
Block I.1
Period I – 1930 to 1934
Block III.8
Block III.2
Block III.3
Block I.6
Block I.5
Isolated
Block I.2
Block I.4
Block I.8
Block II.4
Block II.1
Block IV.6
Block IV.3
Block IV.7
Isolated
Period IV – 1945 to 1949
Block II.7
Block II.8
Period II – 1935 to 1939
Block IV.1
Block IV.4
Block II.2
Block II.6
Block II.5
Block IV.8
Block IV.2
Block IV.5
Block II.3
Chart 5: Tabu Search Blockmodeling - Reduced Graphs
44
Table 4
Tabu Search Blockmodeling: Density Tables and Image Matrices per period
Density Tables
Image Matrices
1
2
3
4
5
6
7
8
1
1.00
1.00
0.25
0.06
1.00
0.86
0.21
2
0.02
1.00
0.90
0.07
0.20
0.03
0.03
3
1.00
1.00
0.25
0.04
1.00
0.68
0.07
4
0.06
0.95
0.13
1.00
0.15
1.00
0.06
0.20
Period I - 1930 to 1934
5
6
7
8
0.05
1.00
0.84
0.10
0.04
0.01
0.03
0.03
1.00
0.69
0.07
0.09
0.50
0.09
0.20
0.04
0.02
0.05
0.07
0.04
0.92
0.20
0.03
0.97
0.36
0.06
0.06
0.27
0.10
0.11
Average Density
0.35
Blocks
1
2
3
4
5
6
7
8
1
1
0
1
0
0
1
1
0
2
0
1
0
1
0
0
0
0
3
1
0
1
0
0
1
1
0
4
0
1
0
1
0
1
0
0
5
0
0
0
0
0
0
0
0
6
1
0
1
1
0
1
1
0
7
1
0
1
0
0
0
1
0
8
0
0
0
0
0
0
0
0
1
2
3
4
5
6
7
8
1
0.98
0.01
0.04
0.15
0.24
0.20
0.33
0.26
2
0.02
0.63
0.03
0.04
0.08
0.40
0.01
3
0.03
0.01
0.77
0.22
-
4
0.18
0.04
0.14
0.02
0.21
1.00
0.06
Period II - 1935 to 1939
5
6
7
8
0.17
0.25
0.28
0.27
0.06
0.70
0.22
0.02
0.15
1.00
0.06
0.04
0.08
0.03
0.29
1.00
0.07
1.00
0.09
0.05
0.06
0.02
0.07
Average Density
0.19
Blocks
1
2
3
4
5
6
7
8
1
1
0
0
0
1
1
1
1
2
0
1
0
0
0
0
1
0
3
0
0
1
0
1
0
0
0
4
0
0
0
0
0
1
1
0
5
0
0
1
0
0
0
0
0
6
1
0
0
0
0
1
1
0
7
1
1
0
1
0
1
0
0
8
1
0
0
0
0
0
0
0
2
3
1
2
3
4
5
6
7
8
1
0.99
0.98
0.02
0.16
0.16
0.37
1.00
0.15
0.03
-
0.25
0.79
0.01
-
4
0.94
1.00
0.01
0.25
0.27
0.80
Period III - 1940 to 1944
5
6
7
8
0.03
0.15
0.20
0.25
0.03
0.01
0.05
0.26
0.33
0.78
0.17
0.01
0.02
0.01
0.19
0.04
0.10
0.01
0.03
0.05
0.09
0.04
0.13
0.13
0.73
Average Density
0.19
Blocks
1
2
3
4
5
6
7
8
1
1
0
0
1
0
0
0
1
2
0
1
0
0
0
0
0
0
3
0
1
1
0
0
0
0
0
4
1
0
0
1
0
1
1
1
5
0
0
0
0
0
0
0
0
6
0
0
0
1
0
1
0
0
7
1
0
0
1
0
0
0
0
8
1
0
0
1
0
0
0
1
1
2
3
4
5
6
7
8
1
0.92
0.05
0.03
0.82
0.03
0.22
0.10
0.95
2
0.10
0.90
0.02
0.13
0.77
0.02
0.03
-
3
0.03
0.04
0.28
0.04
0.02
0.02
0.02
-
4
0.83
0.09
0.01
0.72
0.03
0.17
0.08
0.79
Period IV - 1945 to 1949
5
6
7
8
0.06
0.27
0.20
0.98
0.87
0.01
0.05
0.01
0.02
0.02
0.19
0.10
0.68
0.37
0.01
0.04
0.01
0.04
0.03
0.10
0.04
0.02
0.06
0.05
0.02
0.14
0.11
0.98
Average Density
0.21
Blocks
1
2
3
4
5
6
7
8
1
1
0
0
1
0
1
0
1
2
0
1
0
0
1
0
0
0
3
0
0
1
0
0
0
0
0
4
1
0
0
1
0
0
0
1
5
0
1
0
0
1
0
0
0
6
1
0
0
0
0
0
0
0
7
0
0
0
0
0
0
0
0
8
1
0
0
1
0
0
0
1
1
2
3
4
5
6
7
8
1
1.00
0.81
0.15
0.04
0.21
0.32
0.14
0.96
2
0.96
1.00
0.90
0.04
0.38
0.25
0.24
0.91
3
0.23
0.85
0.85
0.13
0.04
0.07
0.18
4
0.06
0.04
0.19
0.01
0.05
0.01
0.06
Period V - 1950 to 1954
5
6
7
8
0.25
0.32
0.17
0.90
0.41
0.21
0.26
0.78
0.17
0.04
0.09
0.12
0.01
0.05
0.01
0.05
0.08
0.05
0.04
0.19
0.06
0.21
0.05
0.34
0.04
0.05
0.03
0.14
0.23
0.36
0.18
0.94
Average Density
0.28
Blocks
1
2
3
4
5
6
7
8
1
1
1
0
0
0
1
0
1
2
1
1
1
0
1
0
0
1
3
0
1
1
0
0
0
0
0
4
0
0
0
0
0
0
0
0
5
0
1
0
0
0
0
0
0
6
1
0
0
0
0
0
0
1
7
0
0
0
0
0
0
0
0
8
1
1
0
0
0
1
0
1
1
2
3
4
5
6
7
8
1
0.92
1.00
0.10
0.85
0.02
0.35
0.04
0.99
2
1.00
0.56
0.61
0.97
1.00
0.66
0.10
1.00
3
0.08
0.99
0.62
0.06
0.74
0.08
0.01
0.16
4
0.89
0.80
0.07
0.81
0.04
0.31
0.05
0.98
Period VI - 1955 to 1959
5
6
7
8
0.02
0.40
0.07
1.00
0.81
0.17
1.00
0.82
0.10
0.02
0.18
0.02
0.36
0.08
0.97
0.90
0.03
0.02
0.04
0.01
0.25
0.05
0.72
0.01
0.04
0.02
0.09
0.02
0.76
0.14
1.00
Average Density
0.43
Blocks
1
2
3
4
5
6
7
8
1
1
1
0
1
0
0
0
1
2
1
1
1
1
1
1
0
1
3
0
1
1
0
1
0
0
0
4
1
1
0
1
0
0
0
1
5
0
1
1
0
1
0
0
0
6
0
0
0
0
0
0
0
1
7
0
1
0
0
0
0
0
0
8
1
0
0
1
0
1
0
1
1
2
3
4
5
6
7
8
1
0.96
0.96
0.96
0.26
0.17
0.97
0.18
0.97
2
0.96
1.00
1.00
0.84
0.16
1.00
0.76
0.98
3
0.96
1.00
1.00
0.94
0.17
0.98
0.93
0.96
4
0.32
0.78
0.97
0.96
0.09
0.73
0.94
0.33
Period VII - 1960 to 1964
5
6
7
8
0.20
0.95
0.16
0.97
0.18
1.00
0.38
1.00
0.21
0.96
0.88
0.96
0.10
0.78
0.91
0.25
0.03
0.17
0.04
0.16
0.22
0.98
0.14
0.97
0.06
0.20
0.86
0.08
0.20
0.93
0.07
0.98
Average Density
0.63
Blocks
1
2
3
4
5
6
7
8
1
1
1
1
0
0
1
0
1
2
1
1
1
1
0
1
1
1
3
1
1
1
1
0
1
1
1
4
0
1
1
1
0
1
1
0
5
0
0
0
0
0
0
0
0
6
1
1
1
1
0
1
0
1
7
0
0
1
1
0
0
1
0
8
1
1
1
0
0
1
0
1
1
2
3
4
5
6
7
8
1
1.00
0.18
0.02
0.23
0.27
1.00
0.25
2
0.22
0.20
0.04
0.01
0.02
0.04
0.18
0.02
3
0.03
0.01
0.01
0.01
0.01
4
0.02
0.02
0.01
0.01
0.02
0.01
0.02
0.03
Period VIII - 1965 to 1969
5
6
7
8
0.36
0.36
1.00
0.31
0.03
0.05
0.19
0.01
0.02
0.01
0.01
0.01
0.01
0.01
0.03
0.75
0.05
0.18
0.06
0.05
0.22
0.04
0.25
0.26
1.00
0.05
0.05
0.04
1.00
Average Density
0.16
Blocks
1
2
3
4
5
6
7
8
1
1
1
0
0
1
1
1
1
2
1
1
0
0
0
0
1
0
3
0
0
0
0
0
0
0
0
4
0
0
0
0
0
0
0
0
5
1
0
0
0
1
0
1
0
6
1
0
0
0
0
0
1
0
7
1
1
0
0
1
1
1
0
8
1
0
0
0
0
0
0
1
Blocks
Blocks
Blocks
Blocks
Blocks
Blocks
Blocks
Blocks
45
Table 5
Tabu Search Results: R2 per Period and Number of Blocks
Number of Blocks
2
4
8
I
0.114
0.262
0.455
II
0.170
0.165
0.185
Period
IV
0.070
0.113
0.162
III
0.097
0.140
0.217
V
0.047
0.072
0.121
VI
0.063
0.294
0.351
VII
0.045
0.236
0.225
VIII
0.056
0.089
0.131
Table 6
Block Robustness: Distribution of Densities
Period
I
II
III
IV
V
VI
VII
VIII
0% ≤ D ≤ 25%
66%
78%
80%
77%
70%
51%
34%
83%
25% < D ≤ 50%
5%
10%
6%
5%
11%
6%
6%
8%
50% < D ≤ 75%
3%
3%
2%
3%
0%
10%
2%
2%
75% < D ≤ 100%
27%
10%
13%
16%
19%
33%
58%
8%
Table 7.1
Period I: Selected Statistics per Block (Number of Leaders and Discrete Variables)
Block
I.1
Total
Number % Leaders number of
% of
Avg. Num.
Avg.
of Leaders on Total
Sessions
Sessions
Sessions Field Age
10%
23%
13
420
32
10.08
I.2
4
3%
113
6%
I.3
2
2%
41
2%
I.4
I.5
I.6
I.7
I.8
Total
4
50
1
39
15
128
3%
39%
1%
30%
12%
100%
65
325
21
632
183
1,799
4%
18%
1%
35%
10%
100%
(5.38)
28
(8.2)
21
(2.5)
16
(6.34)
7
(1.02)
21
N/A
16
(2.29)
12
(3.73)
14.05
(1.3)
(0.83)
7.25
(2.72)
8.00
(0)
6.25
(0.95)
8.22
(0.72)
10.00
N/A
8.74
(0.71)
6.47
(1.02)
8.28
(0.4)
Avg.
Betweeness
Centrality
2.08
(0.44)
0.34
(0.18)
0.94
(0.07)
3.34
(3.06)
0.28
(0.09)
5.85
N/A
0.49
(0.12)
0.55
(0.14)
0.71
(0.13)
46
Table 7.2
Period I: Number of Leaders Classified under selected Dichotomic Variables
AfricanDesc.
Block
I.1
I.2
I.3
I.4
I.5
I.6
I.7
I.8
Total
Pearson Chi-Square
Asymp. Sig. (2-sided)
Female
1
0
0
0
24
0
5
7
37
0.001
Non U.S.
2
0
1
0
4
0
2
1
10
0.412
0
3
0
3
9
0
1
4
20
0.000
New Orleans
Swing
6
1
1
2
26
0
17
4
57
0.831
8
2
1
2
30
1
18
10
72
0.695
Percentage on total number of leaders in each block (Yellow, Bold Italic: above average)
AfricanDesc.
Block
Female
Non U.S.
New Orleans
Swing
I.1
8%
15%
0%
55%
73%
I.2
0%
0%
75%
33%
67%
I.3
0%
0%
50%
50%
50%
I.4
0%
0%
75%
50%
50%
I.5
48%
8%
18%
57%
65%
I.6
0%
0%
0%
0%
100%
I.7
13%
5%
3%
52%
55%
I.8
47%
7%
27%
33%
83%
Avg. Total (in Blocks)
29%
8%
16%
51%
64%
47
7.2.2. Period II (1935-1939)
The square-like core structure seems to remain in the field (II.1, II.7, II.4 and II.6),
while less connected than before (II.1 and II.4 are not connected). In contrast, we don’t
observe isolated blocks as before. But analyze Table 8.1: the disparity among blocks is
striking. Eighteen leaders in II.1 have in average 52 sessions and concentrate 31% of the
field’s production. Their average field age is higher than the field’s (10.67 versus 9.22), and
yet, slightly lower than II.7’s age (12 years). Leaders at II.1 play mostly Swing. Benny
Goodman, Artie Shaw, Red Norvo are among the prestigious white leaders in this block. But
now the integration of African-American popular leaders permits the presence, inter alia, of
Billy Holiday and Louis Armstrong who climbed from less prominent blocks in the previous
period.
Right after II.1 we observe that leaders at II.6 produced in average 23 sessions, and
concentrated 13% of the field’s production. Nevertheless, their field age is almost identical to
II.1. They are connected to II.1 and are playing mostly Swing. In comparison to II.1, II.6 has a
higher rate of African-descendents. Duke Ellington and Fletcher Henderson are among them.
These “field-young” leaders tend to play Swing and their success brought them to an
advantaged position in the field. In contrast, Benny Carter at block II.7 is one of the oldest
leaders in the field and has the highest betweenness centrality among all blocks.
Blocks II.5, II.3 and II.2 tend to be non-U.S. leaders, especially the later. They tend to
be younger than the average leader. II.2 has also the lowest average session in the field.
Among its leaders, we find Coleman Hawkins.
48
Table 8.1
Period II: Selected Statistics per Block (Number of Leaders and Discrete Variables)
Block
II.1
Total
Number % Leaders number of
of Leaders on Total
Sessions
II.2
II.3
II.4
II.5
18
10
8
12
4
10%
% of
Sessions
936
6%
79
4%
112
7%
214
2%
96
31%
3%
4%
7%
3%
II.6
17
9%
386
13%
II.7
1
1%
17
1%
II.8
111
61%
1,165
39%
181
100%
3,004
100%
Total
Avg. Num.
Avg.
Sessions Field Age
52
(14.28)
8
(1.87)
14
(3.41)
18
10.67
(1.11)
8.90
(2.02)
2.63
(0.5)
8.42
(5.23)
(1.82)
24
(14.29)
23
(4.79)
17
N/A
10
(1.14)
3.00
(1.08)
10.18
(1.03)
12.00
N/A
9.62
16.60
(1.92)
(0.64)
9.22
(0.47)
Avg.
Betweeness
Centrality
2.29
(0.5)
1.12
(0.59)
1.16
(0.79)
0.85
(0.46)
0.17
(0.11)
0.86
(0.26)
13.92
N/A
0.21
(0.04)
0.69
(0.12)
Table 8.2
Block
Total
Pearson Chi-Square
Asymp. Sig. (2-sided)
Period II: Number of Leaders Classified under selected Dichotomic Variables
AfricanDesc.
Female
Non U.S.
New Orleans
Swing
Bop
II.1
4
3
0
4
14
II.2
3
0
6
4
9
II.3
0
0
8
0
7
II.4
4
2
4
3
10
II.5
1
0
2
3
3
II.6
6
3
9
7
15
II.7
1
0
0
0
1
II.8
38
5
12
46
63
57
13
41
67
122
0.411
0.199
0.000
0.119
0.018
Percentage on total number of leaders in each block (Yellow, Bold Italic: above average)
AfricanDesc.
Female
Non U.S.
New Orleans
Swing
Bop
Block
II.1
22%
17%
0%
25%
88%
60%
44%
100%
II.2
30%
0%
100%
100%
II.3
0%
0%
0%
33%
17%
33%
100%
II.4
30%
50%
75%
II.5
25%
0%
75%
35%
18%
53%
88%
II.6
41%
100%
100%
II.7
0%
0%
0%
34%
48%
II.8
5%
11%
66%
31%
7%
23%
42%
76%
Avg. Total (in Blocks)
1
1
0
1
0
1
0
2
6
0.765
6%
11%
0%
10%
0%
6%
0%
2%
4%
Hard Bop
2
0
0
0
0
0
0
8
10
0.663
Hard Bop
13%
0%
0%
0%
0%
0%
0%
8%
6%
49
7.2.3. Period III (1940-1944)
Period III, from 1940 to 1944 was marked by the World War II. Not surprisingly,
Blocks III.2, III.3 and III.5, mostly non U.S. leaders are not connected to the U.S. circuit. In
contrast, we found III.1, III.7, III.4 and III.8 forming again an almost entirely connected
square. III.4 and III.8 are by far the dominant blocks. The former’s leaders have in average
54 sessions, and the highest field age, 16.6 years. Their betweenness centrality was also the
field highest, 3.82. Leaders at III.8 had in average 52 sessions, and 13.5 years in the field. In
contrast to III.4, they presented an average betweenness below Block III.3’s average (2.18).
Together, III.4 and III.8 amassed 45% of all field’s production, with only 14% of leaders.
They tended to play Swing, but is worth noting the presence of Bop leaders. At Block III.4,
we find Benny Goodman, Coleman Hawkins and Eddie Condon. At III.8 we find Artie Shaw,
Bing Crosby, Gene Miller, Tommy and Jimmy Dorsey and Woody Herman.
Famous African-American Swing leaders like Louis Armstrong, Fletcher Henderson,
Benny Carter, Ella Fitzgerald, Duke Ellington are all in III.7, as well as the Bop innovators
like Charlie Parker and Dizzy Gillespie. This is the largest block in the field, with 137 leaders,
48% of them are African-descendents (lower than Block III.1 80%). When we inspect the
percentage distributions of styles at Table 9.2, we find this block’s leaders following closely
the field’s average. III.1 also includes famous African-American leaders: Billy Holiday and
Hot Lips Page.
It is interesting to note the inception of Bop at Blocks III.1 and III.7. These are the
blocks that most concentrate African-descendent leaders. They are both connected to the
dominant blocks III.4 and III.8. Also, worth noting that many of these African-American
leaders mentioned above were at dominant blocks in earlier years. Perhaps the subsequent
segregation of successful African-American leaders had created the context for the emergence
of Bop.
50
Table 9.1
Period III: Selected Statistics per Block (Number of Leaders and Discrete Variables)
Block
III.1
III.2
III.3
III.4
III.5
III.6
III.7
III.8
Total
Total
Number % Leaders number of
of Leaders on Total
Sessions
10
2
10
5
15
20
137
26
225
4%
1%
4%
2%
7%
9%
61%
12%
100%
175
22
137
268
181
228
1,220
1,358
3,589
% of
Sessions
5%
1%
4%
7%
5%
6%
34%
38%
100%
Avg. Num.
Avg.
Sessions Field Age
18
(2.42)
11
(4)
14
(3.08)
54
9.90
(1.53)
12.00
(2)
7.00
(0.45)
16.60
(21.06)
(2.11)
12
(2.05)
11
(2.26)
9
(1.14)
52
(12.69)
9.73
(1.57)
12.10
(1.49)
11.36
(0.63)
13.50
15.95
(1.95)
(1.03)
11.43
(0.45)
Avg.
Betweeness
Centrality
1.02
(0.25)
0.69
(0.54)
2.18
(1.42)
3.82
(0.72)
0.02
(0.01)
0.41
(0.11)
0.23
(0.04)
1.49
(0.61)
0.59
(0.11)
51
III.1
III.2
III.3
III.4
III.5
III.6
III.7
III.8
III.1
III.2
III.3
III.4
III.5
III.6
III.7
III.8
Avg. Total (in Blocks)
Block
Total
Pearson Chi-Square
Asymp. Sig. (2-sided)
Block
0.403
0.000
0.000
0
2
10
0
11
10
12
0
45
Non U.S.
2
1
1
2
3
5
55
6
75
0.185
New Orleans
Swing
0.012
8
2
9
5
11
15
79
21
150
Bop
0.060
3
0
1
1
1
0
8
1
15
0.681
Hard Bop Cool
0
0
0
1
0
1
13
2
17
0.297
1
0
0
0
0
0
2
2
5
0
0
0
0
0
0
3
0
3
0.962
Soul
0
0
0
0
0
0
1
0
1
0.999
Free Jazz
Percentage on total number of leaders in each block (Yellow, Bold Italic: above average)
AfricanDesc.
Female
Non U.S.
New Orleans
Swing
Bop
Hard Bop Cool
Soul
Free Jazz
0%
25%
0%
0%
0%
80%
20%
100%
38%
13%
0%
0%
0%
0%
0%
0%
0%
100%
50%
100%
0%
0%
10%
0%
0%
0%
0%
100%
90%
10%
20%
0%
0%
0%
0%
20%
40%
100%
20%
20%
7%
7%
23%
0%
0%
0%
0%
73%
85%
8%
25%
0%
29%
0%
6%
0%
0%
0%
50%
88%
9%
65%
7%
2%
48%
12%
45%
11%
2%
1%
0%
4%
0%
26%
4%
0%
0%
91%
9%
9%
36%
9%
20%
33%
67%
7%
8%
2%
1%
0%
2
0
0
1
1
0
16
1
21
Female
Period III: Number of Leaders Classified under selected Dichotomic Variables
8
0
0
1
1
5
66
0
81
AfricanDesc.
Table 9.2
52
7.2.4. Period IV (1945-1949)
The block structure on period IV brings a novelty. Instead of the square structure, we
observe a triangle, formed by IV.1, IV.4 and IV.8. Block IV.1 is clearly the dominant in this
period. Its leaders produced in average 65 recording sessions (Table 10.1). They presented
also the highest betweenness centrality and their average field age (15.08 years) is statistically
equivalent to IV.4 (15.65). Five of these leaders are African-Descendents (Table 10.2; in
percentage terms, similar to the field’s average). They play mostly Swing and Bop, while
already including one leader who plays Cool. They are all Americans. This block brings
together famous white leaders of Swing bands like Benny Goodman, Tommy Dorsey and
Woody Herman. At the same time, Louis Armstrong is also in this block. In contrast, Charlie
Parker, Dizzy Gillespie and Sarah Vaughan, reknown leaders in the aggressive and emergent
Bop movement are also in this block. It is ironic that these two conflicting groups share the
same position at this period. While Benny Goodman claimed in an interview that he didn’t
know Bop and Louis Armstrong mocked publicly the new style while reintroducing his New
Orleans style, Parker and Gillespie struggled to advance Bop.
Adjacent to IV.1, in a clearly peripheral position, we observe IV.6. This is the largest
block in number of leaders (272 musicians, or 80% of leaders). Nonetheless, they concentrate
only 57% of the recorded sessions. Thelonious Monk, one of the exponents in the Bop
movement, is in this block, as well as Stan Getz, who was developing his own approach
which would influence the Cool style.
IV.4, also in the dominant axis, concentrates old leaders in the field, and their
betweenness centrality is second only to IV.1’s. Yet, their average session production was 21,
much below IV.1’s average. Similarly to IV.1, its leaders are predominantly AfricanAmericans. In contrast, they present more leaders playing New Orleans/Classic Jazz. As a
matter of fact, we observe in this block all styles represented, at higher rates in comparison
with the field’s average. This is a position where established African-American Swing leaders
share common sidemen. Coleman Hawkins, Lester Young, Count Basie, Lionel Hampton,
Billy Holiday and Ella Fitzgerald are in this block. It is worth noting the first two in this list
were predecessors to Bop. It is also of interest the presence of the young Miles Davis in this
position.
Although leaders at IV.8 presented a lower field age in comparison to IV.4, they
produced more than IV.4’s leaders, achieving an average of 41 sessions. They are all white
53
and Americans. They tend to play Swing and Bebop, shying away from New Orleans/Classic
Jazz. Traditional white leaders are here: Artie Shaw, Gene Krupa and Harry James.
Compare this axis with Block IV.7. Its leaders have the lowest average production of
sessions, and, at the same time, still closer to New Orleans/Classic Jazz style. In comparison,
Block IV.6 lump together 272 leaders (almost 80% of the field’s leaders) produce only 57%
of the field’s sessions. And yet, their average field age is 11.57, only 3.5 years below the
elite’s average at IV.1. Their demographics resemble a lot the field’s average (concentration
of African-descendents, women and non-Americans) as well as the styles played.
Blocks IV.2, IV.5 and IV.3 constitute the burgeoning jazz world outside U.S., mostly
in Europe. While block IV.5’s leaders tend to play Swing and Bop, block IV.2 geared towards
Bop. IV.3, in contrast, tend to play only Swing and Bop, shying away from New
Orleans/Classic jazz. These are very field young leaders (as a matter of fact, the youngest
block in the field), and yet their average production of sessions surpasses the field’s average.
Table 10.1
Period IV: Selected Statistics per Block (Number of Leaders and Discrete Variables)
Block
IV.1
IV.2
IV.3
IV.4
IV.5
IV.6
IV.7
IV.8
Total
Total
Number % Leaders number of
of Leaders on Total
Sessions
12
5
9
20
6
272
12
7
343
3%
1%
3%
6%
2%
79%
3%
2%
100%
785
48
135
488
75
2,529
74
290
4,424
% of
Sessions
18%
1%
3%
11%
2%
57%
2%
7%
100%
Avg. Num.
Avg.
Sessions Field Age
Avg.
Betweeness
Centrality
65
(13.33)
10
(2.04)
15
(4.38)
24
15.08
(2.33)
7.60
(1.54)
5.00
(2.59)
15.65
(0.46)
(3.48)
(1.44)
(0.29)
13
(5.49)
9
(0.87)
6
(2.19)
41
(14.21)
10.17
(2.33)
11.57
(0.49)
11.83
(1.78)
13.86
12.90
(1.09)
(2.1)
11.73
(0.43)
1.97
(0.46)
1.21
1.17
(0.72)
1.30
1.01
(0.83)
0.20
(0.03)
0.23
(0.13)
0.52
(0.08)
0.38
(0.05)
54
IV.1
IV.2
IV.3
IV.4
IV.5
IV.6
IV.7
IV.8
IV.1
IV.2
IV.3
IV.4
IV.5
IV.6
IV.7
IV.8
Avg. Total (in Blocks)
Block
Total
Pearson Chi-Square
Asymp. Sig. (2-sided)
Block
0.704
0.020
0.000
0
4
8
0
6
44
3
0
65
Non U.S.
3
1
7
6
0
72
6
1
96
0.004
New Orleans
Swing
0.001
10
2
2
17
5
144
2
7
189
Bop
0.300
4
2
0
5
2
45
0
2
60
0.311
Hard Bop Cool
1
0
0
5
0
27
2
0
35
1
0
0
1
0
9
0
0
11
0.955
Period IV: Number of Leaders Classified under selected Dichotomic Variables
0
0
0
1
0
14
1
0
16
0.870
Soul
0
0
0
1
0
1
1
0
3
0.047
Free Jazz
0
0
0
1
0
1
0
1
3
0.026
Fusion
0
0
0
0
0
3
0
0
3
0.996
Latin/World
Percentage on total number of leaders in each block (Yellow, Bold Italic: above average)
AfricanDesc.
Female
Non U.S.
New Orleans
Swing
Bop
Hard Bop Cool
Soul
Free Jazz Fusion
Latin/World
42%
8%
0%
25%
83%
33%
8%
8%
0%
0%
0%
0%
20%
80%
50%
20%
25%
50%
0%
0%
0%
0%
0%
0%
89%
88%
22%
0%
25%
0%
0%
0%
0%
0%
0%
0%
55%
15%
32%
89%
26%
26%
5%
5%
5%
5%
0%
0%
100%
83%
33%
0%
0%
0%
0%
0%
0%
0%
0%
0%
44%
9%
33%
66%
21%
12%
4%
6%
1%
16%
0%
0%
25%
75%
25%
13%
13%
17%
0%
25%
0%
0%
0%
0%
14%
100%
29%
14%
0%
0%
14%
0%
0%
0%
0%
0%
41%
9%
19%
28%
55%
17%
10%
3%
5%
1%
1%
1%
1
1
0
3
0
24
0
1
30
Female
5
1
2
11
0
119
2
0
140
AfricanDesc.
Table 10.2
55
Block VII.5
Isolated
Block VII.8
Block VII.7
Block VII.2
Block VII.4
Block VII.3
Block VII.1
Block V.1
Block V.8
Block V.4
Period VII – 1960 to 1964
Block V.6
Block V.7
Isolated
Period V – 1950 to 1954
Block VII.6
Block V.3
Block V.2
Block V.5
Block VIII.6
Block VIII.2
Period VIII – 1965 to 1969
Block VI.5
Block VI.7
Block VI.6
Period VI – 1955 to 1959
Block VIII.7
Block VIII.1
Block VIII.8
Block VI.2
Block VI.8
Block VIII.5
Block VIII.4
Block VIII.3
Isolated
Block VI.3
Block VI.1
Block VI.4
Chart 6: Tabu Search Blockmodeling - Reduced Graphs
56
7.2.5. Period V (1950-1954)
The structure of Period V brings us back to the quasi-connected square we have
observed on earlier periods. Again, we are able to identify a dominant block: V.2. The three
leaders in this block have the highest average of production (48 sessions). Their field age
average (17.67) is the second, preceded by (22.33). They have also the highest average
betweenness centrality (3.74), which we may also observe on the reduced graph: V.2 connects
the square to V.5 and V.3. The leaders in this block are all African-American men, who play
mostly Swing, Bop and Latin. It is worth noting that their rate of Bop and Latin playing
surpasses by far the field’s average (67% versus 26%, and 33% versus 2%). These leaders are
Charlie Parker, Dizzy Gillespie and Roy Eldridge.
Block V.8 is very similar to V.2: its leaders have a high average number of sessions
(41) and field age (17.07), although their average betweenness is lower (1.21 versus 3.74).
Five of the fifteen leaders are African-descendents. There are mostly American musicians,
with higher presence of women. They tend to play Swing (much above the field average), Bop
and Hard Bop. These are in fact traditional white and African-American Swing leaders: Stan
Kenton, Artie Shaw, Billy Holiday, Louis Armstrong, Harry James, and Count Basie.
Block V.1, connected to V.2, counts with the highest average field age in the period,
although their average betweenness centrality is lower than V.2’s and equivalent to V.8.
These established musicians tend to play Swing, are all American men and have a lower
presence of African-descendents that the field’s average. This is a declassé block: once
hegemonic leaders like Benny Goodman, Red Nichols and Tommy Dorsey used to be in the
dominant block in earlier periods.
If V.1 and V.8 have lower presence of African-Descendent leaders, V.3 and V.5
(again, connected to the former by V.2) have higher presence of African-descendent leaders.
These two blocks are shying away from New Orleans/Classic (especially V.3, with Bing
Crosby). V.3 is also shying away from Swing. Their tendency is to play Bop and related styles
(Hard Bop, Cool and Soul). It is in this block that Clifford Brown introduces his style,
predecessor of hard bop. Observe their average session number and field age: both blocks
count with field young leaders. Leaders at V.5 (which includes established swing leaders like
Louis Armstrong, Coleman Hawkins, but also the bop innovator Bud Powell and cool
innovators like John Lewis and Gerry Mulligan) produced fewer sessions than the field’s
average, while V.3 produced somewhat above (13 versus 10.17 sessions). This figure was
57
much below their neighbors at V.2. It is worth noting that the once well connected Louis
Armstrong is now brokered by Bop leaders.
Isolated blocks confirm that victorious musicians are also well connected and play
trendy styles. Look at Block V.4. Fourteen out of eighteen leaders play New Orleans/Classic
Jazz. They are unconnected among themselves, and have lower average number of sessions
produced (11). The situation at Block V.7 is even more dramatic. It concentrates 68% of the
field’s leaders, while controlling only 48% of its session production. Their average session
production is the lowest in the field (7). Finally, observe block V.6. Although connected to the
square, leaders in this block is not connected among themselves. This is a typical situation of
a fragmented periphery. Altough its leaders are traditional in the field (average field age of
17.18 surpasses Block V.8’s 17.07) they were not able to keep up with new times. They play
New Orleans/Classic more than the field average, and have low acceptance of new styles.
Table 11.1
Period V: Selected Statistics per Block (Number of Leaders and Discrete Variables)
Block
V.1
V.2
V.3
Total
Number % Leaders number of
of Leaders on Total
Sessions
9
3
12
2%
1%
3%
258
144
158
% of
Sessions
6%
3%
4%
V.4
18
4%
192
4%
V.5
48
11%
454
10%
V.6
V.7
V.8
Total
33
289
15
427
8%
68%
4%
100%
440
2,078
620
4,344
10%
48%
14%
100%
Avg. Num.
Avg.
Sessions Field Age
29
(6.44)
48
(17.32)
13
(2.75)
11
22.33
(2.11)
17.67
(3.18)
11.58
(2.67)
14.17
(2.66)
(2.52)
9
(1.73)
13
(2)
7
(0.46)
41
(8.42)
12.10
(1.02)
17.18
(1.36)
12.26
(0.47)
17.07
10.17
(0.65)
(1.8)
13.11
(0.4)
Avg.
Betweeness
Centrality
1.28
(0.26)
3.74
(0.21)
0.56
(0.13)
0.68
(0.25)
0.23
(0.06)
0.40
(0.11)
0.19
(0.03)
1.21
(0.21)
0.33
(0.03)
58
Avg. Total (in Blocks)
Block
Total
Pearson Chi-Square
Asymp. Sig. (2-sided)
Block
V.1
V.2
V.3
V.4
V.5
V.6
V.7
V.8
V.1
V.2
V.3
V.4
V.5
V.6
V.7
V.8
0.189
0.004
0.007
0
0
2
2
10
1
81
1
97
Non U.S.
2
1
1
14
8
20
77
1
124
0.000
New Orleans
Swing
0.004
9
2
4
3
20
17
133
12
200
Bop
0.001
1
2
7
0
17
4
76
3
110
0.070
Hard Bop Cool
0
0
4
0
10
2
49
3
68
1
0
2
1
7
2
40
0
53
0.495
Period V: Number of Leaders Classified under selected Dichotomic Variables
0
0
4
0
10
0
28
0
42
0.001
Soul
0
0
0
0
1
0
7
0
8
0.938
Free Jazz
0
0
0
1
0
0
6
0
7
0.737
Fusion
0
1
0
0
0
0
8
1
10
0.034
Latin/World
Percentage on total number of leaders in each block (Yellow, Bold Italic: above average)
AfricanDesc.
Female
Non U.S.
New Orleans
Swing
Bop
Hard Bop Cool
Soul
Free Jazz Fusion
Latin/World
22%
0%
0%
22%
100%
11%
0%
11%
0%
0%
0%
0%
100%
0%
0%
33%
67%
67%
0%
0%
0%
0%
0%
33%
58%
0%
17%
9%
36%
64%
36%
18%
36%
0%
0%
0%
22%
6%
11%
20%
0%
0%
7%
0%
0%
0%
93%
7%
54%
15%
21%
20%
50%
43%
25%
18%
25%
3%
0%
0%
18%
3%
3%
63%
53%
13%
6%
6%
0%
0%
0%
0%
37%
7%
28%
30%
52%
30%
19%
16%
11%
3%
2%
3%
33%
20%
7%
7%
80%
20%
20%
0%
0%
0%
0%
7%
37%
7%
23%
29%
47%
26%
16%
12%
10%
2%
2%
2%
0
0
0
1
7
1
20
3
32
Female
2
3
7
4
26
6
106
5
159
AfricanDesc.
Table 11.2
59
7.2.6. Period VI (1955-1959)
At the center of the reduced graph on Period VI we observe a prominent position:
Block VI.2. It connects three separate groups: the first is composed by VI.6, VI.8, VI.4 and
VI.1, the second is formed by VI.5 and VI.3 and finally VI.7. Surprisingly, this block is
occupied by only one leader, Woody Herman, an established white American Swing musician
who is now playing Bop and Cool as well. He is one of the oldest musicians in the field (27
years), has a high number of sessions (35) and the highest betweenness centrality. What
different worlds does he connect?
Blocks VI.8, VI.6 and VI.4 tend to combine higher rates of African-Descendent
leaders. VI.4 achieves 58%, in comparison to the field average of 40%. Leaders in these
positions play for Swing, Bop, Hard Bop. In contrast, VI.4 and VI.6 tend to play Cool, Free
Jazz and Soul, while leaders at VI.8 gear towards Fusion and Latin/World.
At Block VI.4 we observe important names in Hard Bop: Art Blakey, Julian
“Cannonball” Adderley, Miles Davis, as well as Bop/Latin leaders like Dizzy Gillespie,
Thelonious Monk, or older Swing leaders like Count Basie. In average, their field age (16.77
versus 18.45) is close to VI.8, although their average session produced is much lower (18
versus 32). It is at this period that Miles Davis records “Kind of Blue”, one of the most
influential records ever released.
VI.6 counts with traditional leaders like Duke Ellington and Louis Armstrong.
Younger and successful leaders pushing the music towards Cool and new experiments are
also present (see above, the “Multiple Revolution in 1959”): Charles Mingus (strongly
influenced by Ellington), John Coltrane , Art Pepper, Abbey Lincoln, and Chet Baker. This is
a block where “West Coast” musicians are strongly represented. It is interesting to observe
how this block resembles VI.4 in field age (16.30 versus 16.77) and average session number
(15 versus 18).
Block VI.1 presents unexpressive average session number (15, just above the field’s
average), and its leaders’ average field age is not much higher than the average (16.71 versus
15.21). These are mostly white American men who play Swing, Bop and Cool.
Between these two positions, Block VI.8 represents a bridge. It brings together famous
names in Swing like Benny Goodman, Coleman Hawkins, Gene Krupa and Billy Holiday.
VI.8 would be the dominant block in the whole structure, if VI.2 didn’t play such role. Its
eleven leaders have in average 32 sessions produced, and their average field age (18.45) is
second only to VI.2. Also, their average betweenness centrality is the second highest (1.29).
60
These results are quite surprising for two reasons. First, we would expect that these
old leaders at blocks VI.8 and VI.2 be completely displaced by younger generations. Second,
we would expect younger generations to be cohesive in establishing a dominant core. The
reverse occurs: the young leaders build up separate Jazz worlds, not necessarily based on style
differentiation, but on geographic distance (East and West Coast). What is on stake here is the
hegemony of New York City as the center of Jazz. Also, old leaders are able to provide the
bridge between these two worlds: their prestige allows them to participate in both worlds.
Block VI.7 is the largest block in terms of quantity of leaders. Although it
concentrates 277 leaders (51%), it represents only 31% of all field’s sessions. Moreover, its
leaders presented the lowest betweenness centrality. As a peripheral block, it is surprising that
it is internally cohesive. We find some important names in this Block: Dave Brubeck (Cool
innovator), Ornette Coleman, Sun Ra (Free Jazz innovators), and Horace Silver (Hard Bop
innovator). Brubeck released “Time Out”, a still popular record among Jazz fans, and
Coleman was introducing his first free jazz album (“The Shape of Jazz to Come”).
Most of the leaders who compose blocks VI.3 and VI.5 are not Americans. In terms of
styles played, they tend to converge to the field’s average. VI.5 presented a higher rate of
leaders playing New Orleans/Classic Jazz than the field’s average. In contrast, we observe on
block VI.3 a higher rate of leaders playing Latin/World styles.
In general, it is interesting how well balanced all blocks are in terms of age field and
session number. Especially on the former dimension, the blocks are not too disparate from
each other.
61
Table 12.1
Period VI: Selected Statistics per Block (Number of Leaders and Discrete Variables)
Block
VI.1
VI.2
VI.3
VI.4
Total
Number % Leaders number of
of Leaders on Total
Sessions
% of
Sessions
17
3%
258
4%
1
0%
35
1%
18
77
3%
14%
321
1,384
5%
22%
VI.5
5
1%
82
1%
VI.6
134
25%
1,946
31%
VI.7
VI.8
Total
277
11
540
51%
2%
100%
1,968
357
6,351
31%
6%
100%
Avg. Num.
Avg.
Sessions Field Age
15
(1.54)
35
N/A
18
(3.54)
18
16.71
(2.13)
27.00
N/A
12.78
(1.82)
16.77
(1.53)
(0.83)
16
(3.23)
15
(1.29)
7
(0.45)
32
(3.42)
10.40
(1.25)
16.30
(0.77)
14.24
(0.57)
18.45
11.76
(0.53)
(3.53)
15.21
(0.39)
Avg.
Betweeness
Centrality
0.32
(0.04)
3.04
N/A
0.64
(0.21)
0.39
(0.05)
0.34
(0.13)
0.21
(0.03)
0.07
(0.01)
1.29
(0.17)
0.21
(0.02)
62
VI.1
VI.2
VI.3
VI.4
VI.5
VI.6
VI.7
VI.8
VI.1
VI.2
VI.3
VI.4
VI.5
VI.6
VI.7
VI.8
Avg. Total (in Blocks)
Block
Total
Pearson Chi-Square
Asymp. Sig. (2-sided)
Block
0.010
0.000
0.000
2
0
16
3
5
5
96
1
128
Non U.S.
3
0
3
15
2
29
88
2
142
0.035
New Orleans
Swing
0.259
9
1
7
42
2
68
100
6
235
Bop
0.049
7
1
6
28
1
38
52
4
137
0.693
Hard Bop Cool
3
0
6
27
1
40
65
3
145
5
1
3
14
1
29
24
1
78
0.007
Period VI: Number of Leaders Classified under selected Dichotomic Variables
3
0
3
15
0
25
52
2
100
0.965
Soul
0.958
0
0
0
3
0
3
6
0
12
Free Jazz
2
0
0
1
1
4
5
3
16
0.000
Fusion
2
0
2
2
0
4
9
1
20
0.379
Latin/World
Percentage on total number of leaders in each block (Yellow, Bold Italic: above average)
AfricanDesc.
Female
Non U.S.
New Orleans
Swing
Bop
Hard Bop Cool
Soul
Free Jazz Fusion
Latin/World
24%
0%
12%
18%
53%
41%
18%
29%
18%
0%
12%
12%
0%
0%
0%
0%
100%
100%
0%
100%
0%
0%
0%
0%
11%
6%
89%
19%
44%
38%
38%
19%
19%
0%
0%
13%
58%
14%
4%
20%
55%
37%
36%
18%
20%
4%
1%
3%
0%
0%
40%
20%
20%
0%
0%
0%
100%
40%
20%
20%
46%
8%
4%
22%
52%
29%
30%
22%
19%
2%
3%
3%
35%
6%
35%
36%
41%
21%
26%
10%
21%
2%
2%
4%
45%
36%
9%
18%
55%
36%
27%
9%
18%
0%
27%
9%
40%
8%
24%
26%
44%
25%
27%
14%
19%
2%
3%
4%
0
0
1
11
0
11
18
4
45
Female
4
0
2
45
0
62
96
5
214
AfricanDesc.
Table 12.2
63
7.2.7. Period VII (1960-1964)
Period VII probably presents the most strongly connected structure. At a first glace,
we observe a fully connected (Blocks VII.2, VII.7, VII.4 and VII.3) square and a fully
connected pentagram (VII.2, VII.3, VII.6, VII.8 and VII.1).
Block VII.1 is one of the most important positions in the field, and yet, it is not as
hegemonic as were the dominant blocks in earlier periods. To be sure, its eight leaders have
the highest average session number (38 sessions) and the highest field age (26.38 years). Yet,
their average betweenness is dwarfed by the more central leaders at Block VII.3. They are all
African-American leaders: Budd Johnson, Coleman Hawkins, Duke Ellington, Count Basie,
Etta Jones, J.J. Johnson, Donald Byrd, Nat Adderley. Isolated, but powerful, they play a mix
of already “old” styles like Swing and Bop, with Soul, Free Jazz and Fusion.
In contrast, the six leaders at Block VII.3 occupy a very privileged position. Although
they amass in average a lower number of recording sessions (28 versus 38), their field age
(21.33 years) is very close to VII.1’s and their average betweenness centrality is the highest in
the field. Benny Goodman, Dizzy Gillespie, Ray Charles and Sammy Davis Jr. are among the
leaders in this field35. They are all American leaders, and four out of six are AfricanAmericans. They tend to play Swing, Bop, Soul and Latin. Together with leaders in Block
VII.2, the leaders in this block connect the square with the pentagram.
Compare these blocks with Block VII.8. It is the most numerous block in the field (39
leaders) after the isolated and fragmented VII.5. Most of them are African-American leaders,
and they tend to play Bop, Hard Bop, Soul, Cool and Fusion, shying away from New Orleans
and Swing. They achieved the third place in average sessions number produced, while their
field age is statistically equivalent to the field’s average. At the heath of the Civil Rights
Movement, we find in these block conscious band leaders like Charles Mingus, as innovators
in Soul like Julian “Cannonball” Adderley, Brother Jack McDuff and Stanley Turrentine. It is
surprising to find Stan Getz at this same block, for this is the period when Getz is introducing
Bossa Nova to the Jazz field, and won the Grammy for “Desafinado” in 1963. A possible
explanation would describe how Charles Mingus (and other West Coast Hardboppers) and
Stan Getz shared the same resources (west coast musicians) and were led to differentiation in
style later on. In contrast with the ecological argument that differentiation is preceded by
35
If in earlier periods the collocation of Swing and Bop musicians could spur (or occur in tandem with) public
conflict, now these two social groups seem to be integrated under the wider umbrella of “Jazz”. At this period
Gillespie is no longer seen as rebellious. As a matter of fact, he was chosen by the U.S. government to represent
his country abroad. In contrast, the spot light is shifted on the conflict between mainstream jazz and “free-jazz”.
As a consequence, I focus my investigation on where new conflicts emerge vis-à-vis structural changes.
64
crowding out a niche, we may observe a different movement: Charles Mingus later shift to
free-jazz might have been a move taken in order to differentiate himself from the more
commercial Stan Getz.
Still in the pentagram, we observe Block VII.6, which is also connected to VII.4, in
the connected square. Fifty percent of the eight American leaders in this block are AfricanAmericans. But in comparison to VII.8, they still tend to play Swing more than the field’s
average. In many aspects, but one, they are quite similar to VII.8. They have a relatively high
average session number (19 versus 20), and betweenness centrality (0.75 versus 0.62).
However, their field age is much higher (20.25 versus 16.46). This block includes leaders like
BenWebster, Gerry Mulligan, and SarahVaughan.
Block VII.2 groups only three leaders, Gene Krupa, Lalo Schifrin and Mark Murphy.
These leaders do not have an expressive average number of sessions, nor field age. Yet, their
betweenness centrality (0.95) is second only to Block VII.3. Block VII.2 is connected to
Block VII.7, whose seven leaders together have the second oldest average field age (24.43
years). They are all men, mostly white and American. They present the lowest average
betweenness centrality (0.12) in the field. Nonetheless, their average number of sessions is
higher than the average field. Laurindo de Almeida, Frank Sinatra e Charlie Barnet are in this
position, strongly associated with Swing and Latin/World styles.
There are fifteen leaders at Block VII.4, who present the third highest field age (21.33
years). Like Block VII.7’s, this block’s average number of sessions is relatively higher than
the field’s average (17 versus 9.95), but in contrast, its betweenness centrality is higher than
the field’s average (0.39 versus 0.25). There is a high presence of women, as well as leaders
who play Swing, Cool and Latin. Among them we observe Stan Kenton, who was a Swing
leader. He was famous and controversial for his experiments.
65
Table 13.1
Period VII: Selected Statistics per Block (Number of Leaders and Discrete Variables)
Block
VII.1
VII.2
VII.3
VII.4
VII.5
VII.6
VII.7
VII.8
Total
Total
Number % Leaders number of
of Leaders on Total
Sessions
8
3
6
15
461
8
7
39
547
1%
1%
1%
3%
84%
1%
1%
7%
100%
302
38
169
262
3,629
151
104
786
5,441
% of
Sessions
6%
1%
3%
5%
67%
3%
2%
14%
100%
Avg. Num.
Avg.
Sessions Field Age
Avg.
Betweeness
Centrality
38
(15.03)
13
(0.33)
28
(3.28)
17
26.38
(4.66)
18.00
(9.5)
21.33
(3.64)
21.47
(0.04)
(3.78)
(1.53)
(0.07)
8
(0.41)
19
(5.22)
15
(4.82)
20
(1.95)
16.49
(0.46)
20.25
(3.68)
24.43
(1.67)
16.46
9.95
(11.78)
(1.11)
16.99
(9.7)
0.79
(0.23)
0.95
1.43
(0.32)
0.39
0.18
(0.02)
0.75
(0.16)
0.12
(0.04)
0.62
(0.08)
0.25
(0.46)
66
VII.1
VII.2
VII.3
VII.4
VII.5
VII.6
VII.7
VII.8
VII.1
VII.2
VII.3
VII.4
VII.5
VII.6
VII.7
VII.8
Avg. Total (in Blocks)
Block
Total
Pearson Chi-Square
Asymp. Sig. (2-sided)
Block
0.365
0.000
0.001
0
1
0
0
129
0
1
2
133
Non U.S.
1
1
1
0
126
1
1
5
136
0.069
New Orleans
Swing
0.051
4
1
3
9
156
3
6
10
192
Bop
0.066
3
1
3
3
101
2
0
17
130
0.005
Hard Bop Cool
5
1
1
3
152
3
0
24
189
0
0
1
4
55
2
0
6
68
0.485
Period VII: Number of Leaders Classified under selected Dichotomic Variables
3
0
2
0
107
1
0
20
133
0.001
Soul
0.760
1
0
0
0
28
1
0
1
31
Free Jazz
1
0
0
1
16
0
0
4
22
0.487
Fusion
0
0
2
2
20
0
1
1
26
0.033
Latin/World
Percentage on total number of leaders in each block (Yellow, Bold Italic: above average)
AfricanDesc.
Female
Non U.S.
New Orleans
Swing
Bop
Hard Bop Cool
Soul
Free Jazz Fusion
Latin/World
100%
13%
0%
13%
50%
38%
63%
0%
38%
13%
13%
0%
0%
0%
33%
33%
33%
33%
33%
0%
0%
0%
0%
0%
67%
0%
0%
17%
50%
50%
17%
17%
33%
0%
0%
33%
27%
20%
0%
0%
60%
20%
20%
27%
0%
0%
7%
13%
39%
7%
23%
4%
5%
28%
29%
36%
35%
13%
25%
6%
50%
13%
0%
13%
38%
25%
38%
25%
13%
13%
0%
0%
29%
0%
14%
14%
86%
0%
0%
0%
0%
0%
0%
14%
79%
13%
5%
13%
26%
44%
62%
15%
51%
3%
10%
3%
43%
7%
24%
25%
35%
24%
35%
12%
24%
6%
4%
5%
1
0
0
3
30
1
0
5
40
Female
8
0
4
4
181
4
2
31
234
AfricanDesc.
Table 13.2
67
7.2.8. Period VIII (1965-1969)
It is worth recalling that the blockmodeling analysis found the weakest correlation in
this period. That means, on one hand, that our technique for grouping around positions was
not acceptable for this period. On the other hand, it also means that the specialization of
resources was not strong enough to generate distinct blocks. In other words, musicians were
widely shared by leaders. When this occurs, I suggest that the allocation of musicians to
leaders gets closer to an ideal decentralized type, where resource specificity decreases and a
commoditization permits fluidity of personnel. Bearing that in mind, let us analyze its
structure.
The structure at Period VIII brings us back to the centralization around a dominant
block. The four leaders at Block VIII.1 present both the highest average session number (33)
and highest average betweenness centrality (1.40). Its average field age is second only to
VIII.2 (20.75 versus 21.29 years). They are all American men, and three out of four leaders
are African-Americans. Bop, Hard Bop, Soul, Fusion and Latin are the prevalent styles in this
block. We do not observe Swing nor New Orleans played anymore. Among them, Herbie
Mann and Wes Montgomery were in the spotlight for advancing the Fusion-Rock
experiments.
Compare this block to VIII.2. Although leaders in this block present the highest
average field age (21.29 years), their average session number is just above the field average (8
versus 7.65). Their average betweenness is also not much higher than the field’s average (0.43
versus 0.25, take in account the respective standard deviations). This block is occupied mostly
by American white men, who play Swing and New Orleans in much higher rates than the
field’s average. It is surprising that it has any connection at all with the modernist VIII.1.
Nevertheless, leaders like Stan Getz (who at this period won a Grammy award for his “Girl of
Ipanema”) was able to connect modern and traditional styles.
Better challenge to VIII.1 would be expected to come from VIII.7. Its 27 leaders have
the second highest average session number (14), and yet, it is far below VIII.1’s average. It is
also the second highest in average betweenness (0.86 versus 1.40), although their field age is
just equal to the field average (18.44 years). Most of its members are African-Descendents.
Although the styles played are much closer to the field’s averages when comparing to VIII.1,
we still observe a bias towards Hard Bop and Soul. Dizzy Gillespie, Herbie Hancock and
Clark Terry are among the leaders in this block.
68
Block VIII.8 presents the second highest average field age (20.75), the same of its
neighbor VIII.1. And yet, its average session number and betweenness centrality (respectively
11 and 0.54) are about a third of the numbers achieved by leaders at VIII.1. The eight leaders
at Block VIII.8 tend to be Non-Americans. But in comparison to their isolated and noncohesive counterparts at Blocks VIII.3 and VIII.4, they are cohesive and connected to the U.S.
Jazz world. Nevertheless, there is a clear tendency to play older styles like New
Orleans/Classic Jazz and Swing.
The non-cohesive Block VIII.6, connected to VIII.2, VIII.1 and VIII.7 is the largest
block in the period. It contains 392 leaders (71% of the total), but they produced in average
only 7 sessions. Their average betweenness is just below the field’s average (0.2 versus 0.25)
and their average field age is just above the field’s average (18.54 versus 18.44 years). Due to
the high number of leaders, in almost all styles and demographics it resembles the field’s
averages: above 40% are African-descendents, and they tend to play modern styles. Although
this block could be considered to lump together less successful musicians, we find leaders as
diverse as Miles Davis, Max Roach, Astrud Gilberto, Art Blakey, Benny Goodman, Ornette
Coleman among many other ones. Richard “Groove” Holmes, Julian “Cannonball” Adderley Soul.
In average, Block VIII.5 contains the youngest leaders in the field (8.82 versus
18.44). Although its average session production is above average (13 versus 7.65), it is much
lower than VIII.1’s figure. They tend to be African-descendents, and play Soul and Free Jazz.
In spite of the avant-gardist preference, these leaders are not sectarian, as we observe an
average betweenness centrality of 0.63 (higher than the field average of 0.25). Among the
eleven leaders in this position we will find Archie Shepp, Cecil Taylor, John Coltrane and
Albert Ayler, exponents of Free Jazz.
The isolated and non-cohesive Blocks VIII.3 and VIII.4 present the lowest average
session number in the field (respectively 4 and 5), and the lowest average betweenness
centrality. Yet, leaders at Block VIII.3 present the fourth highest average field age (20.45).
Both VIII.3 and VIII.4 present a high percentage of Non-American leaders who tend to play
older styles like Swing and New Orleans/Classic Jazz. Neverthless, it is remarkable the
higher rate of Free Jazz leaders at VIII.4. Although European leaders were struggling to create
their own version of Free jazz, they did not share sidemen with their American counterparts.
69
Table 14.1
Period VIII: Selected Statistics per Block (Number of Leaders and Discrete Variables)
Block
VIII.1
VIII.2
VIII.3
VIII.4
VIII.5
VIII.6
VIII.7
VIII.8
Total
Total
Number % Leaders number of
of Leaders on Total
Sessions
4
17
29
67
11
392
27
8
555
1%
3%
5%
12%
2%
71%
5%
1%
100%
133
140
112
317
146
2,937
369
91
4,245
% of
Sessions
3%
3%
3%
7%
3%
69%
9%
2%
100%
Avg. Num.
Avg.
Sessions Field Age
Avg.
Betweeness
Centrality
33
(6.09)
8
(1.77)
4
(0.63)
5
20.75
(2.32)
21.29
(2.96)
20.45
(2.54)
17.40
(0.12)
(0.63)
(1.42)
(0.02)
13
(3.3)
7
(0.48)
14
(1.18)
11
(2.19)
8.82
(1.52)
18.54
(0.53)
18.44
(1.7)
20.75
7.65
(9.2)
(1.78)
18.44
(10.71)
1.40
(0.45)
0.43
0.13
(0.09)
0.10
0.63
(0.29)
0.20
(0.02)
0.86
(0.19)
0.54
(0.09)
0.25
(0.51)
70
VIII.1
VIII.2
VIII.3
VIII.4
VIII.5
VIII.6
VIII.7
VIII.8
VIII.1
VIII.2
VIII.3
VIII.4
VIII.5
VIII.6
VIII.7
VIII.8
Avg. Total (in Blocks)
Block
Total
Pearson Chi-Square
Asymp. Sig. (2-sided)
Block
0.844
0.000
0.000
0
3
13
35
1
105
1
6
164
Non U.S.
0
14
11
19
0
71
3
3
121
0.000
New Orleans
Swing
0.154
0
5
10
17
0
101
10
4
147
Bop
0.012
2
2
2
11
0
88
10
4
119
0.001
Hard Bop Cool
2
3
4
18
4
161
18
3
213
0
1
2
4
0
48
3
0
58
0.492
Period VIII: Number of Leaders Classified under selected Dichotomic Variables
3
0
3
7
3
110
16
1
143
0.000
Soul
0.000
0
0
3
11
11
40
1
0
66
Free Jazz
2
0
2
5
0
35
4
0
48
0.063
Fusion
1
1
1
1
2
18
4
0
28
0.039
Latin/World
Percentage on total number of leaders in each block (Yellow, Bold Italic: above average)
AfricanDesc.
Female
Non U.S.
New Orleans
Swing
Bop
Hard Bop Cool
Soul
Free Jazz Fusion
Latin/World
75%
0%
0%
0%
0%
50%
50%
0%
75%
0%
50%
25%
24%
6%
18%
82%
29%
12%
18%
6%
0%
0%
0%
6%
28%
7%
45%
41%
37%
7%
15%
7%
11%
11%
7%
4%
28%
4%
52%
28%
25%
16%
27%
6%
10%
16%
7%
1%
0%
9%
0%
0%
0%
36%
0%
0%
73%
27%
100%
18%
43%
8%
27%
19%
26%
23%
42%
13%
29%
10%
9%
5%
81%
4%
4%
11%
37%
37%
67%
11%
59%
4%
15%
15%
13%
0%
75%
38%
50%
50%
38%
0%
13%
0%
0%
0%
42%
7%
30%
22%
26%
21%
38%
10%
26%
12%
9%
5%
0
1
2
3
0
31
1
0
38
Female
3
4
8
19
8
167
22
1
232
AfricanDesc.
Table 14.2
71
8. Discussion
The MRQAP analyses give us a sense of how the field changed. It is against these
trends that we compare the results in the blocks.
As a general trend, the preference for leaders with similar field age and number of
sessions decrease in time, which may be interpreted as a lower level of centralization among
top leaders. Nevertheless, the preference for partners with “equal power” never vanishes.
Instead, the element of “race” changes significantly within this period. If during the
“Jim Crow” era success was linked to ascribed characteristics, the Bop revolution reinforced
two phenomena: mobility and endogamous association of African-decedent leaders, which
was frequently, interpreted as a “Crow Jim” phenomena (LOPES, 2002). Nonetheless, as we
observed at Table 3, this effect loses much of its strength when we added the “style” matrices,
which denotes that “racial” preference was intermediated by style preference.
It is also worth noting the periods when the field experiences turning points. Period IV
announces a change: as style dissimilarity becomes slightly positive and significant, we
interpret that leaders seek leaders who are not originally affiliated with their own styles. But
at the same time, the African-American endogamy increases. We observe at this period the
Bop revolution that occurred in the inner core of the Jazz world. Young African-American
sidemen who were tightly connected to Swing leaders become leaders themselves and
introduce the new style in the core of the network. This revolution is felt on Period V: leaders
seek leaders who have different field age and lower (or higher) number of sessions. The
example of Coleman Hawkins is illustrative of this change: some old and established Swing
leaders seek to adapt themselves to the new times.
A new point of inflection comes at the Period VII. Again leaders associate with
leaders with different field age and level of sessions produced, which denotes an interchange
between prominent and up-coming leaders. Also, this is the period when endogamy among
white leaders finally approaches zero. But most important is the fact that the difference in
betweenness centrality lost its significance. This does not mean that the actors are equally
connected and have similar betweenness centrality. Instead, that means that the association is
not ruled by “star” structure logic as it was in earlier periods.
At Period VIII the significance of the betweenness centrality difference becomes again
significant and positive, which may characterize a return to equilibrium. But we observe a
quite different picture in comparison with earlier periods. The Neither-African Descendent
factor becomes negative (which is strongly intermediated by Style affiliation, as we observe
72
on Table 3). Style dissimilarity becomes negative in period VIII (Table 2, although becomes
positive when we account the styles individually at Table 3). The affiliation to styles becomes
an important explanatory factor for association.
In summary:
•
The preference for association with leaders of same age at the field and same
level of production (or market share) decreases but never disappears
•
The “Jim Crow” effect becomes less significant for association. Instead, we
observe a rising endogamy among African-Descendent leaders and exogamy
among non-African-Descendent leaders.
•
These effects become highly intermediated by style endogamy.
It is in contrast with these trends that we now discuss the results observed among and
within blocks.
A centralized structure remains apparent from period I to period IV. At Period I, block
I.1 is dominant in field age, number of sessions produced and betweenness centrality. It is at
that block that Benny Goodman introduces innovations in Swing that led him to the famous
Carnegie Hall concert in 1935. Period II follows closely this model with some interesting
deviations. II.1 is the dominant block, but it has not the highest field age. Nevertheless, the
position with the highest field age is II.7, occupied by only one individual. Also, we see now
Louis Armstrong in the dominant block (II.1), which shows some social mobility, especially
crossing racial boundaries. Nevertheless, a Paretian follower would not be surprised with this
finding: it is important for the incumbent elites to absorb some of the most talented
individuals from the lower strata, in order to contain conflicts. Yet, the structure is clearly
centralized and the “racial” segregation is observable within blocks. Period III is again clearly
centralized. III.4 has the highest average session production, field age and betweenness
centrality. However, III.7 shows a motility effect of leaders, like Louis Armstrong, and
competition with same resources with up-coming boppers. The meeting at the same social
space of individuals who descended from upper strata, with individuals who are ascending to
better positions promises future conflicts but at the same time interchange of ideas36. Finally,
Period IV has still a dominant structure. Block IV.1 is the dominant position: the average
sessions number is much higher than other blocks, statistically its leaders are as old as the
oldest (IV.4) and its average betweenness centrality is the highest (1.97). It combines leaders
36
It is believed that Jazz was born in similar conditions: as the Creole (free men, descendents of AfricanAmericans and French descendents in Louisiana) lost their social status and were forced to live with former
slaves. The cultural interchange between the two groups mingled together Blues roots (from the slaves) with
orchestral elements (from the Creole) to create the predecessors of jazz (Rag Time, Boogie-Woogie).
73
from III.7 and III.4, and now the conflict between boppers and traditional leaders is open in
the air. At the same time that Armstrong performs a return to his New Orleans roots by
innovating in that style, Gillespie and Parker introduce Bop innovations at the core. As
Collins (2000) suggests, frequently many relevant innovations are not introduced by new
comers, but by incumbents afraid of losing their position. In contrast with Kadushin (2004a),
conflict is not preceded by fragmentation among blocks. On the contrary: conflict is stronger
exactly among those leaders who share the same resources and because of that are placed in
the same block37. This confirms Blau’s insight that conflict stems from interaction (BLAU,
1977). This is a period when conflict reveals some core latent values within the Jazz
community: both Trad and Bop musicians agree on reinforcing the worth of a professional
career in Jazz. Both look for Jazz’s original roots: while Trad musicians recover the collective
improvisation and typical voicing from New Orleans, Bop musicians stress the solo
(individual) improvisation and dissonant scales. Nevertheless, Charlie Parker is deeply
influenced by the Blues, and that affects his music and his followers’. As a new moral order
emerges, new deviants are regarded with suspicion. Miles Davis’s association with Mulligan
at the “Birth of the Cool” yields a sound too close to Harry James’s: a vibrato too far from the
African tradition. He introduces what will become a predecessor of cool in a peripheral block.
It is also, nevertheless, an attempt of a young leader to dissociate himself from his former
leader: Charlie Parker.
In the Period V, Bop is no longer a fad. V.2 (where we observe Gillespie, Parker and
Eldridge) is the dominant block in average number of sessions and betweenness centrality.
But the leaders are significantly younger than leaders at other blocks. One may suspect that
the emergence of Bop as a new paradigm (DEVENAUX, 1997) places the Bop masters where
the Swing leaders were before them. Thus, although the style changed, the field’s structure is
still centralized. Nevertheless, the centralization occurred during the war (DEVENAUX,
1997) is counterbalanced by a post-war migration to West Coast cities. Period V also shows
the emergence of the Cool style with the Modern Jazz Quarter and Gerry Mulligan at a
37
The temptation at this point is to give in to a functional analysis of social differentiation. A functionalist
(durkheimian, and later ecologists’) interpretation would follow this rationale: crowding out leads to competition
for the same resources, which leads to social differentiation, as actors look for new niches. However, these
phenomena might be described from the creation of discourses and the underlying legitimacy of style affiliation.
From this perspective, I rephrase my interpretation along the lines: a hegemonic style constraints differentiation,
which leads to crowding out of central positions. Conflict emerges as social actors’ habitus are no longer able to
interpret their new social reality (scarcity of resources previously abundant). New niches are created as
rebellious actors generate new discourses that at the same time create novelty and articulate with the core values
in the field. These processes occur almost in parallel, robbing the causality explanation from the ecologists: new
niches are above all new discursive places.
74
peripheral block (V.5). In another peripheral block (V.3) Clifford Brown develops his Hard
Bop style. If at Period IV both conflicting styles (Trad and Bop) where in the same block,
now the conflicting styles (Hard Bop and Cool) are in separate blocks, but intermediated by
the dominant V.2. This intermediation is also a “linguistic” intermediation: in a genealogical
analysis, Bop antecedes both Cool and Hard Bop. Thus, it is not surprising that V.3 and V.5
are peripheral to V.2. These are stylistic differentiations from the same root: Bop.
Period VI brings signals of change: Woody Herman is central to the field, at Block
VI.2. He has the highest filed age, highest average number of sessions, and the highest
betweenness centrality. To be sure, this finding is counter-intuitive: Woody Herman was at
that period an established Swing leader who incorporate Bop elements in his music. How was
he able to conquer this position? As DeVeaux (1997) shows, Herman was able to record Bop
before Gillespie38, which gave him a better position in the field. In addition, the increasing
tensions among younger leaders created a void in the core of the field. I offer the
interpretation that when a conflict becomes disruptive, the empty space among the contending
parties is occupied by a third party, who is able to bridge them. Herman occupied it. This
interpretation would be supported by Simmel’s intuituion on the tertius gaudens (SIMMEL,
1950), and Burt’s (1992) concept of structural hole. But my question goes beyond the social
network intermediation: I am concerned in asking how an actor is able to perform a brokerage
between the two opposite sides, when both exert pressures on the third party to take sides. In
other words, I am concerned in identifying how the third party is able to sustain the legitimacy
of that position. I propose that as young leaders depleted their legitimacy through predatory
struggle, older leaders (like Herman) who have their reputation intact are able to bridge the
contending parties. But who are the contending parties at this period? Gioia (1997) refers to a
“fragmentation” in the Jazz world. Szwed gives us the picture of “multiple” revolutions.
Throughout this turbulent period we observe leaders developing Jazz in very different
directions: Miles Davis advanced his modal style with “Kind of Blue”, Coltrane departed
from Davis’s band to develop his own style, exemplified in “Giant Steps”. It was also the
moment when Ornette Coleman recorded “The Shape of Jazz to Come”, in the forefront of
Free Jazz. Charles Mingus was going in a very counterintuitive direction: he recovered Bop
elements and combined them to a novel Swing orchestration.
All these leaders were
positioned at peripheral blocks. Also, in the forefront of Hard Bop we observe Art Blakey and
Horace Silver at a peripheral block: VI.7. (it is uncohesive block, whose link to VI.2 is
38
See Chapter 4 on Trajectory for an analysis on Herman’s trajectory as distinct from Gillespie’s.
75
unidirectional – see Table 4). Below VI.2, all blocks have average session production much
lower and closer to the field’s average.
The reorganization in the field continues through Period VII. VII.1 is the most
prominent block, but we should suspect that it is not dominant. To be sure, it presents the
highest session number (38) and highest field age (26.38), but is not the central in terms of
betweenness centrality (0.79). Compare this figures to Period IV’s block IV.1’s statistics. The
latter presented in average 65 sessions produced. It is not only the sheer absolute number, but
the relative number in comparison to the second in the list. While VII.7 presents an average
session number of 24.43 (8% lower than VII.1), IV.8 presents an average session of 41 (35%
below IV.1!). Now, compare the field average in both periods: Period IV had in average 12.9
sessions, while Period VII presented 16.99. I hope these are enough evidences the Period IV
was more centralized (in terms of sessions) than Period VII. Another important insight here is
that the most central block in the network is not necessarily the one that yields the higher
number of sessions anymore. To be sure, Period VII is also a period of absolute decline in
number of sessions (see chart 2). As a consequence, boundary-spanning actors at the fringe of
the Jazz world might be able to bring fresher ideas and new influences. The grammy recipient
Stan Getz, who was one of the Jazz leaders who brought Bossa Nova to the Jazz world was
not in the center, but in the periphery - Block VII.8. In the same Block (VII.8), we observe
Brother Jack McDuff and Stanley Turrentine developing a Soul version of Jazz.
Before proceeding to the discussion of Period VIII, it is worth noting a word of
caution. Table 5 and Table 6 show very little fit between the block model results and the
expected image matrix. To be sure, VIII.1 is a candidate for a dominant block. It presents a
high average number of sessions (33) and relatively high betweenness centrality (1.40). Also,
its average field age is high (20.75 years) but it is not the highest. It suggests that at Period
VIII the field is rearticulated, in comparison to Period VII. But this articulation is far from the
concentration in the first periods of this analysis. The features of a decentralized field are still
present. Miles Davis recorded “Bitches Brew”, an innovative Fusion album, at a non-cohesive
peripheral position - VIII.6. In the same block, Soul leaders like Richard "Groove" Holmes
and Julian “Cannonball” Adderley are located. Leaders engaged in the Free style are located
at a cohesive block, but still peripheral VIII.5.
When we return to the ideal types proposed above, and compare the evolution of the
jazz field, we are able to assess (1) the models seem to fit our empirical findings and (2) to
understand the process underlying a change between these models. Following, the confirming
evidences that support the proposed ideal-types.
76
Social network structure: while in the first four periods the field presented a
centralized structure, with one or few dominant blocks, later periods (especially on Period VII
and onwards) presented a decentralized structure, where blocks were highly interconnected.
Source of Innovations: During Periods I to IV, innovations (Swing and early Bop) are
mostly introduced by central actors. This trend is slowly reversed in later Periods: Cool and
Hard Bop are introduced in blocks connected to dominant players. When we arrive to Period
VIII, we observe Fusion at the fringes of the social network.
Profile of dominant leaders:
in early periods, we generally observe a coupling
between field age, number of sessions and betweenness centrality. Dominant leaders tended to
combine these characteristics. It doesn’t mean, however, that old musicians were always
dominant. King Oliver, one of the innovators in the Early jazz style, is placed in the peripheral
strata in the Jazz world, as Swing achieves success. Yet, as we advance in the history of Jazz,
we observe that dominant players become younger. The most important example is the
emergence of Gillespie and Parker. The decoupling of variables shows interesting
phenomena: old players may introduce innovations at the peripheral positions of the social
network.
Homogeneity and Stability of Styles: the sheer fragmentation of styles depicted at
Chart 2 shows the decline of the Swing paradigm. In its place, Bop seemed to be an
alternative paradigm, but it was quickly overshadowed by its heirs (Cool and Hard Bop). The
same cycle is observed again, with the emergence of Free jazz, Soul and Fusion. The style
fragmentation becomes more accentuated, and the permanence shorter.
Types of Isomorphism. DiMaggio and Powell (1983) suggested three types of
isomorphism (coercive, normative, and mimetic) as competing mechanisms of diffusion
within fields. The way I explored this three mechanisms was indirect: I first linked the styles
to their legitimate sources and then, by analyzing both the evolution of the institutional
spheres and the evolution of styles I inferred the prominence of one or another mechanism.
We may observe that the shift from Swing to Bop and finally to Fusion marks a change in the
fields meta logics, which favor first coercive mechanisms from the recording industry, then
the normative body of musicians and finally the mimetic processes of absorbing external
influences.
When we turn to the transition process between these two models, some interesting
findings propose it is possible further development in the original model.
First, we observe that “declassed” Swing leaders at Period III meet the up-coming Bop
leaders in the same social network position. This interaction might have spurred competition
77
for the same resources in the same block, as old and young leaders compete for the same
sidemen. This seems to contradict Burt (1987), for leaders structurally equivalent are not
pursuing the same style, but antagonist styles. I interpret this in the following way. First, we
observe an ecological competition for resources, as old and young leaders compete for the
same resources. This competition among leaders is transformed in a competition among forms
(Bop and New Orleans jazz). The legitimacy clash around these two styles brings, in a first
moment, the emergence of common values: jazz is linked to African roots, it is characterized
by improvisation (collective or solo) and is linked to the construction of a solid professional
career. As a result, the similarity between Boppers and New Orleans leaders was not at the
form level, but on the underlying value level. This increase in the legitimacy of both styles is
due to the emergence of values that serve both purposes. In that sense, the conflict between
them generated positive outcomes for both. At Period IV, both bop (Gillespie and Parker) and
traditional leaders (like Louis Armstrong) are located at the core. As Baum and Oliver suggest
(1996), the increased legitimacy of forms might compensate for the increasingly crowding
effect.
At Period VI, the clash among leaders pushing to different directions was not
concurrent on finding latent values in the Jazz paradigm. On the contrary, the emphasis was
on the emergence of individuality and pluralism. At first, this new ethos emerged in
opposition to the establishment of the Jazz canon. Critics would ask “where is Jazz going”. At
this transition point, older and well positioned leaders like Woody Herman were able to
bridge contending parties. But unless the central actor is able to seize power and establish his
own paradigm as dominant39, his position will be hardly sustainable in the long run. Elworth
(1995) shows that the sixties and seventies were the decades when “teleological”
interpretations on Jazz were dropped and the value of pluralism emerged. No longer only one
paradigm would rule the field, as internal dissent and external influences flourished across the
Jazz world.
At this point, it might be worthwhile to discuss whether the plurality value emerged
before or in tandem with the field decentralization40. A facile answer in the Jazz evolution
would be to assume that both, decentralization and emergence of new values occur in tandem.
However, we should ask whether the unfolding in different styles would be possible without
39
We observe this shift from a advantaged social network position to an advantaged political and coercive
position in the emergence of the Medici family (see PADGET; ANSELL, 1993).
40
This discussion touches a central debate between Simmel and Weber (SWEDBERG, 2005). While Weber
understood the emergence of values as a major driver of social change (WEBER, 1950), Simmel stressed that
changes in social forms (e.g. changes in a social network structure) led to changes in values. On the first page of
“Economy and Society” (WEBER, 1978) we observe a criticism to Simmel’s approach.
78
the conflict observed on Period IV. We may speculate whether Jazz musicians would remain
in the same world if the discussion on “what is jazz” never happened.
9. Conclusions
When we observe the transition between a centralized to a decentralized field, an array
of questions emerges. First, what did we learn with the transition path? Second, what are the
limits of the decentralization trend?
Becker (1998) proposes that we deal with the ideal-types’ unbundled features when
we describe the underlying processes within the transition between them. The features that I
chose to perform this analysis are the homogeneity/heterogeneity of styles and the
centralization and decentralization of the network. Below, I depict a schematic framework of
Homogeneous
Heterogeneous
these two dimensions:
Open-conflict
field:
• Normative
isomorphism
Decentralized
field:
• Mimetic
isomorphism
Centralized
field:
• Coercitive
isomorphism
Centralized Network
Decentralized
Network
Schema 3.3: Transition Path between Types of Fields
Source: Author’s proposal
When we observe Periods I to III, we observe Swing overcoming New Orleans and
Classic Jazz. This ecological process depicts a substitution of one dominant paradigm by
another. At Period IV, we observe both the recovery of New Orleans and Classic Jazz under
the revivalist auspices and the emergence of Bop.
To be sure, the Jazz field is still centralized, but the increasingly heterogeneity of
styles leads the field to a conflictive competition and discussion of legitimacy. Periods V and
79
onwards (especially VII and VIII) depict a field where decentralized structure occurs in
tandem with a fragmentation of styles.
The centralization and decentralization of the network occurs in tandem with the
evolution of the Jazz musicians’ perspective. During the centralized period musicians are
focused inward the field: success comes with alignment with the industry’s requirement or
with the professional body’s imposition. The institutionalization of an industry requires the
institutionalization of a professional body. As this professional body strives for autonomy,
internal sources of legitimacy emerges (BOURDIEU, 1993b). The increase in autonomy
reinforces the inward tendencies in the field. We might face here an important depart from
Peterson’s model (1972), where he depicts styles shifting from folk, to commercial and finally
to high art. In the example exposed, both revivalists and boppers seeked to reinforce the
field’s autonomy by promoting the professional body. This is intrinsic to their sources of
legitimacy: traditions and progressive norms, respectively. Paradoxically, professional
socialization and ethos might evolve in conflictive ways, which leads to the polarization and
weakening of the professional authority. As the (professional) normative authority is relaxed,
a more outward perspective emerges. It is possible to introduce influences from outside the
field without threatening the survival of Jazz. To be sure, this transition is not smooth. There
is no assurance that crises will lead to emergence of more cosmopolitan values, and that the
fragmentation of forms will not lead to the fragmentation of the social tissue41. But as Coser
(1966) suggested, the absence of conflicts might be indicative of a moribund community.
References
ANAND, N. Defocalizing the organization: Richard A. Peterson's sociology of organizations.
Poetics, v.28, p.173-184 2000.
ANSELL, C. K. Symbolic Networks: The Realignment of the French Working Class, 18871894. The American Journal of Sociology, v.103, n.2, p.359-390. 1997.
APPELROUTH, S. Constructing the meaning of early jazz, 1917–1930. Poetics, v.31, n.2,
p.117-131 2003.
BARBER, G. 1959: A Great Year in Jazz. All About Jazz 2004. Available at:
http://www.allaboutjazz.com/php/article.php?id=15310 Accessed in Feb 12, 2006
41
Again we face a classic discussion. Toennies perceives individualism as a threat to a community, due to
intestinal conflicts. In contrast, Weber and Simmel stress that conflict doesn’t lead to fragmentation, but
reinforces values (SEGRE, 1998).
80
BAUM, J. A. C.;OLIVER, C. Toward an Institutional Ecology of Organizational Founding.
The Academy of Management Journal, v.39, n.5, p.1378-1427. 1996.
BECKER, H. S. Art Worlds. Berkeley, CA: Univerisity of California Press. 1982
BECKER, H. S. Tricks of the Trade: How to Think about Your Research While You're
Doing It. Chicago: University of Chicago Press. 1998. 244 p.
BECKER, H. S. Jazz Places. GALLERY, A.-K. A. Buffalo, NY 2002.
BECKER, H. S. A New Art Form: Hypertext Fiction. Em: BERNSTEIN, M. (Ed.). How to
Read a Hypertext. Cambridge: Eastgate, 2006.
BERENDT, J. E.;HUESMANN, G. El Jazz: de Nueva Orleans a los años ochenta. México,
D.F.: Fondo de Cultura Económica. 1998. 901 p.
BERGER, P. L. Charisma and Religious Innovation: The Social Location of Israelite
Prophecy American Sociological Review, v.28, n.6, p.940-950. 1963.
BERGER, P.;LUCKMANN, T. The Social Construction of Reality: A Treatise in the
Sociology of Knowledge. Garden City, NY: Doubleday. 1966
BERLINER, P. F. Thinking in Jazz: the infinite art of improvisation. Chicago: The
University of Chicago Press. 1994. 883 p.
BLAU, P. M. Inequality and Heterogeneity: A Primitive Theory of Social Structure. New
York: Free Press. 1977
BORGATTI, S. P.;EVERETT, M. G.;FREEMAN, L. C. Ucinet for Windows: Software for
Social Network Analysis. Harvard, MA: Analytic Technologies 2002.
BORGATTI, S. P.;JONES, C.;EVERETT, M. G. Network Measures of Social Capital.
Connections, v.21, n.2, p.36. 1998.
BOURDIEU, P. Distinction: a social critique of the judgement of taste. Cambridge, MA:
Harvard University Press. 1984a
BOURDIEU, P. Structures, habitus, practices. Em: (Ed.). The Logic of Practice. Stanford:
Stanford University Press, 1990a. p.52-65.
81
BOURDIEU, P. Coisas Ditas. São Paulo: Brasiliense. 1990b. 234 p.
BOURDIEU, P. The Production of Belief: contribution to an economy of symbolic goods.
Em: (Ed.). The Field of Cultural Production. N.Y.: Columbia Univ. Press, 1993a. p.74-111.
BOURDIEU, P. Some Properties of Fields. Em: BOURDIEU, P. (Ed.). Sociology in
Question. London: Sage, 1993b. p.72-76.
BOURDIEU, P. But Who Created the Creators? Em: BOURDIEU, P. (Ed.). Sociology in
Question. London: Sage, 1993d. p.139-140.
BOURDIEU, P. As Regras da Arte. São Paulo: Companhia das Letras. 2002.
BOURDIEU, P. Principles of an Economic Anthropology. Em: SMELSER, N. J. e
SWEDBERG, R. (Ed.). The Handbook of Economic Sociology. Princeton: Princeton
University Press, 2005. p.75-89.
BURT, R. S. Social Contagion and Innovation: Cohesion Versus Structural Equivalence. The
American Journal of Sociology, v.92, n.6, p.1287-1335. 1987.
BURT, R. S. Structural Holes. Cambridge: Harvard University Press. 1992
BURT, R. S. Structural Holes versus Network Closure as Social Capital. Em: LIN, N.;COOK,
K., et al (Ed.). Social Capital: Theory and Research. New York: Aldine De Gruyter, 2001.
BREIGER, R. L. The Analysis of Social Networks. Em: HARDY, M. e BRYMAN, A. (Ed.).
Handbook of Data Analysis. London: Sage, 2004. p.505-526.
BREIGER, R. L.;BOORMAN, S. A.;ARABIE, P. An algorithm for clustering relational data
with applications to social network analysis and comparison with multidimensional scaling.
Journal of Mathematical Psychology, v.12, p.328-383. 1975.
CAVES, R. E. Creative Industries: contracts between art and commerce. Cambridge,
MA: Harvard University Press. 2000
CHRISTENSEN, C. M. The Innovator's Dilemma. Cambridge, MA: Harvard Business
School Press. 1997
COLLINS, R. The Sociology of Philosophies: A Précis. Philosophy of the Social Sciences,
v.30, n.2, p.157-201. 2000.
82
COSER, L. A. The functions of social conflict. New York: Free Press. 1966
DANTO, A. C. The Artworld. The Journal of Philosophy, v.61, n.19, p.571-584. 1964.
DEKKER, D.;KRACKHARDT, D.;SNIJDERS, T. A. B. Sensitivity of MRQAP Tests to
Collinearity and Autocorrelation Conditions. Nijmegen School of Management. - Working
paper series on research in relationship management, RRM-2005-11-MKT 2005.
DEVEAUX, S. The Birth of Bebop: a social and musical history. Berkeley, CA: University
of California Press. 1997. 572 p.
DIMAGGIO, P. J. Cultural entrepreneurship in nineteenth-century Boston: the creation of an
organizational base for high culture in America. Media, Culture and Society, v.4, p.33-50.
1982.
DIMAGGIO, P. J. Structural Analysis of Organizational Fields: a blockmodel approach. Em:
STAW, B. e CUMMINGS, L. (Ed.). Research in Organizational Behavior Greenwich, CT
JAI Press, v.8, 1986. p.335-70.
DIMAGGIO, P. J. Classification in Art. American Sociological Review, v.52, n.4, August,
1987, p.440-455. 1987.
DIMAGGIO, P. J. Constructing an Organizational Field as a Professional Project: U.S. art
museums. Em: POWELL, W. W. e DIMAGGIO, P. J. (Ed.). The New Institutionalism in
Organizational Analysis. Chicago: The University of Chicago Press, 1991. p.267-292.
DIMAGGIO, P. J. Nadel’s Paradox revisited: Relational and cultural aspects of organizational
structures. Em: NOHRIA, N. e ECCLES, R. (Ed.). Networks and organizations : structure,
form, and action Boston: Harvard Business School Press, 1993.
DIMAGGIO, P. J. The New Institutionalisms: avenues of collaboration. Journal of
Institutional and Theoretical Economics, v.154, p.696-705. 1998.
DIMAGGIO, P. J.;POWELL, W. W. The Iron Cage Revisited: Institutional Isomorphism and
Collective Rationality. American Sociological Review, v.48, n.2, p.147-160. 1983.
DURKHEIM, E. Da divisão do trabalho social. São Paulo: Martins Fontes. 1999. 483 p.
ELWORTH, S. B. Jazz in Crisis, 1948-1958: ideology and representation. Em: GABBARD,
K. (Ed.). Jazz among the discourses. Durham: Duke University Pres, 1995. p.57-75.
83
FAULKNER, R. R. Music on Demand: composers and careers in the Hollywood film
industry. New Brunswick: Transaction Books. 1983. 281 p.
FERNANDES, F. The Negro in Brazilian society. New York: Columbia University Press.
1969. 489 p.
FLIGSTEIN, N. The Architecture of Markets: an economic sociology of twenty-firstcentury capitalist societies. Princeton: Princeton University Press. 2001b
FREEMAN, L. C. Centrality in Social Networks: Conceptual Clarification. Social Networks,
v.1, p.215-239. 1979.
FREEMAN, L. C. The Development of Social Network Analysis: A Study in the
Sociology of Science. Vancouver: Empirical Press. 2004. 218 p.
FRITH, S. Performing Rites: on the value of popular music. Cambridge, MA: Harvard
University Press. 1996. 352 p.
GENDRON, B. "Moldy Figs" and Modernists: Jazz at War (1942-1946). Em: GABBARD, K.
(Ed.). Jazz among the discourses. Durham: Duke University Press, 1995. p.31-56.
GIDDENS, A. The constitution of society : outline of the theory of structuration.
Berkeley: University of California Press. 1986
GIOIA, T. The History of Jazz. New York: Oxford University Press. 1997
GREENFELD, L. Different Worlds: a sociological study of taste, choice and success in
art. Cambridge, UK: Cambridge University Press. 1989. 206 p.
GRIDLEY, M. C. Jazz Styles: history and analysis. Upper Saddle River, NJ: Prentice Hall.
2003. 442 p.
HANNEMAN, R. A. Introduction to Social Network Methods. Riverside: Department of
Sociology, UC-Riverside. 2001
ISACOFF, S. Temperament: How Music Became a Battleground for the Great Minds of
Western Civilization. New York: Vintage. 2003. 288 p.
JONES, C. Signaling Expertise: how signals shape careers in creative industries. Em:
PEIPERL, M. A.;ARTHUR, M. B., et al (Ed.). Career Creativity: explorations in the
remaking of work. Oxford: Oxford University Press, 2002. p.209-228.
84
JONES, L. Swing - from verb to noun. Em: JONES, L. (Ed.). Blues People: Negro Music in
White America. New York: Quill William Morrow, 1963. p.142-165.
JOST, E. Free Jazz: The Roots of Jazz. Cambridge, MA Da Capo Press. 1994. 220 p.
KADUSHIN, C. Networks and Circles in the Production of Culture. Em: PETERSON, R. A.
(Ed.). The Production of Culture. Beverly Hills: Sage, 1976. p.107-122.
KADUSHIN, C. Introduction to Social Network Theory. Waltham, MA Brandeis University
2004a.
KOFSKY, F. The Jazz Tradition: Black Music and Its White Critics. Journal of Black
Studies, v.1, n.4, p.403-433. 1971.
LAMONT, M.;MOLNÁR, V. The study of boundaries in the social sciences. Annual Review
of Sociology, v.28, p.167-195. 2002.
LEBLEBICI, H.;SALANCIK, G. R., et al. Institutional Change and the Transformation of
Interorganizational Fields: An Organizational History of the U.S. Radio Broadcasting
Industry. Administrative Science Quarterly, v.36, n.3, p.333-363. 1991.
LOPES, P. D. Pierre Bourdieu's Fields of Cultural Production: a case study of Modern Jazz.
Em: BROWN, N. e SZEMAN, I. (Ed.). Pierre Bourdieu: Fieldwork in Culture. Lanham:
Rowman & Littlefield Publishers, 2000. p.165-185.
LOPES, P. D. The Rise of a Jazz Art World. Cambridge: Cambridge University Press. 2002
LOUNSBURY, M.; RAO, H. Sources of Durability and Change in Market Classifications: A
Study of the Reconstitution of Product Categories in the American Mutual Fund Industry,
1944-1985. Social Forces, v.82, n.3, p.969-999. 2004.
MARCH, J. G. The Business Firm as a Political Coalition. The Journal of Politics, v.24, n.4,
p.662-678. 1962.
MERTON, R. K. Social theory and social structure. New York: Free Press. 1968. 702 p.
MEYER, J. W.;ROWAN, B. Institutionalized Organizations: Formal Structure as Myth and
Ceremony. Em: DIMAGGIO, P. J. e POWELL, W. W. (Ed.). The New Institutionalism in
Organizational Analysis. Chicago: The University of Chicago Press, 1991.
85
MISCHE, A. Partisan Publics: Contention and Mediation across Brazilian Youth
Activist Networks. Princeton: Princeton University Press. 2006
MOHR, J. W. Soldiers, mothers, tramps and others: Discourse roles in the 1907 New York
City charity directory. Poetics, v.22, p.327-57. 1994.
NADEL, S. F. The Theory of Social Structure. London: Cohen & West. 1957. 159 p.
NORTH, D. C. Institutions and Economic Theory. The American Economist, v.36, n.1, p.36. 1992.
PADGETT, J. F.;ANSELL, C. K. Robust Action and the Rise of the Medici. American
Journal of Sociology, v.98, p.1259-1319. 1993.
PANASSIE, H. The Real Jazz. Greenwood Press: Westport, CT. 1973. 284 p.
PARETO, V. The mind and society (Trattato di sociologia generale). New York: Brace
and Company. 1935
PARSONS, T. The Structure of Social Action: A Study in Social Theory with Special
Reference to a Group of Recent European Writers. New York: Free Press, 1949. 817p.
PARSONS, T. The School Class as a Social System. Harvard Educational Review, v.29,
p.297-318. 1959.
PETERSON, R. A. A process model of the Folk, Pop and Fine Art phases of Jazz. Em:
NANRY, C. (Ed.). American Music: from Storyville to Woodstock. New Brunswick, NJ:
Transaction Books, 1972. p.135-151.
PETERSON, R. A. Why 1955? Explaining the Advent of Rock Music. Popular Music, v.9,
n.1, p.97-116. 1990.
PETERSON, R. A. Creating Country Music: fabricating authenticity. Chicago: The
University of Chicago Press. 1997. 306 p.
PETERSON, R. A.;ANAND, N. How Chaotic Careers Create Orderly Fields. Em: PEIPERL,
M.;ARTHUR, M., et al (Ed.). Career Creativity: Explorations in the Remaking of Work
Oxford: Oxford University Press, 2002. p.257-279.
PETERSON, R. A.;ANAND, N. The production of culture perspective. Annual Review of
Sociology, v.30, p.311-334. 2004.
86
PETERSON, R. A.;BERGER, D. G. Cycles in Symbol Production: The Case of Popular
Music. American Sociological Review, v.40, n.2, April, p.158-173. 1975.
PODOLNY, J. M. Networks as the Pipes and Prisms of the Market. American Journal of
Sociology, v.107, n.1, p.33-60. 2001.
PORTER, E. What is this thing called Jazz? African American musicians as artists,
critics, and activists. Berkeley: University of California Press. 2002
POWELL, W. W. Expanding the Scope of Institutional Analysis. Em: POWELL, W. W. e
DIMAGGIO, P. J. (Ed.). The New Institutionalism in Organizational Analysis. Chicago:
The University of Chicago Press, 1991. p.267-292.
RADANO, R. M. The Jazz Avant-Garde and the Jazz Community: action and reaction.
Annual Review of Jazz Studies, v.3, p.71-79. 1985.
SALANCIK, G. R. WANTED: A good network theory of organization. Administrative
Science Quarterly, v.45, n.1, p.345-349. 1995.
SCHUMPETER, J. The Theory of Economic Development: An Inquiry into Profits,
Capital, Credit, Interest, and the Business Cycle. Cambridge, MA: Harvard University
Press. 1934
SCOTT, J. Social Network Analysis: a handbook. London: Sage. 2000
SCOTT, W. R. Institutions and organizations. Thousand Oaks: Sage. 2001. 255 p.
(Foundations for organizational science)
SCOTT, W. R. Institutional Theory: Contributing to a Theoretical Research Program. Em:
SMITH, K. G. e HITT, M. A. (Ed.). Great Minds in Management: The Process of Theory
Development. Oxford, UK: Oxford University Press, 2005.
SEWELL, W. H., JR. A Theory of Structure: Duality, Agency, and Transformation. The
American Journal of Sociology, v.98, n.1, p.1-29. 1992.
SIMMEL, G. The Sociology of Georg Simmel. New York: Free Press. 1950
SIMMEL, G. The Philosophy of Money. New York: Routledge. 2004
SWEDBERG, R. Can there be a sociological concept of interest? Theory and Society v.34,
n.4, p.359 - 390. 2005.
87
SZWED, J. F. Jazz 101: a complete guide to learning and loving jazz. New York:
Hyperion. 2000. 354 p.
SZWED, J. So What: the life of Miles Davis. New York: Simon & Schuster. 2002. 488 p.
THORNTON, P. H.;OCASIO, W. Institutional Logics and the Historical Contingency of
Power in Organizations: Executive Succession in the Higher Education Publishing Industry,
1958-1990. The American Journal of Sociology, v.105, n.3, p.801-843. 1999.
TILLY, C. Durable Inequality: University of California Press. 1999
TILLY, C. Social Boundary Mechanisms. Philosophy of the Social Sciences, v.34, n.2,
p.211-236. 2003.
TUCKER, S. Swing Shift: 'All-Girl' Bands of the 1940s. Durham, N.C.: Duke University
Press. 2000. 413 p.
UZZI, B.;SPIRO, J. Collaboration and Creativity: The Small World Problem. American
Journal of Sociology, v.Forthcoming. 2005.
WASSERMAN, S.;FAUST, K. Social Network Analysis: Methods and Applications.
Cambridge: Cambridge University Press. 1997
WATTS, D. J. Networks, Dynamics, and the Small-World Phenomenon. American Journal
of Sociology, v.105, p.493-527. 1999.
WEBER, M. The Protestant Ethic and the Spirit of Capitalism. New York: Scribner's
Press. 1950
WEBER, M. Ancient Judaism. Glencoe, IL: Free Press. 1952. 484 p.
WEBER, M. Economy and Society: An Outline of Interpretative Sociology. Berkeley:
University of California Press. 1978
WEBER, M. Os fundamentos racionais e sociológicos da música. São Paulo: Edusp. 1995.
159 p.
WEICK, K. Educational organizations as loosely coupled systems. Administrative Science
Quarterly, v.21, n.1, p.1-19. 1976.
88
WHITE, H. C. Careers and Creativity: social forces in the arts. Boulder: Westview Press.
1993
WHITE, H. C.;BOORMAN, S. A.;BREIGER, R. L. Social Structure from Multiple
Networks. I. Blockmodels of Roles and Positions. The American Journal of Sociology,
v.81, n.4, p.730-780. 1976.
WILLIAMS, M. The Jazz Tradition. Oxford: Oxford University Press. 1983. 287 p.
Sources of Primary Data (all web sites were consulted during the year of 2005 and 2006)
ALEXANDER, S. The Red Hot Jazz Archive. Available at: www.redhotjazz.com.
KERNFELD, B. The New Grove Dictionary of Jazz. Oxford: Oxford University Press.
2001. 3000 p.
LORD, T. The Jazz Discography. Taylor Way, West Vancouver, BC, Canada Lord Music
Reference Inc. 2005.
PFEFFER, M. L. Big Bands Database Plus. Available at: http://nfo.net/.
YANOW, S. All Media Guide. Available at: www.allmusic.com
89
Glossary
Art World: Becker (1982), inspired by Danto (1964), defines an Art World as
[E]veryone whose activity has anything to do with the end result.
That includes the people who make materials, instruments, and
tools; the people who create the financial arrangements that make
the work possible; the people who see to distributing the works
that are made; the people who produced the tradition of forms,
genres, and styles the artist works with and against; and the
audience. For symphonic music, the list of cooperating people
might include composers, players, conductors, instrument makers
and repairers, copyists, managers and fundraisers, designers of
symphony halls, music publishers, booking agents, and audiences
of various kinds. For contemporary painting, an equivalent list
would include painters, makers and purveryors of canvases,
paints, and similar materials, collectors, art historians, critics,
curators, dealers, managers and agents, such auxiliary personnel
as, say, lithographic printers, and so on. (BECKER, 2006).
This definition of Art World is close to DiMaggio and Powell’s definition of organizational
field, which may be rephrased as the set of relevant actors in the creation of art works. In
contrast with an organizational field, Becker focuses on the trajectories of actual pieces of
work. As an art world becomes more developed, a criticism body emerges, which is able to
define the objects that will be considered art. Compare this conception with Bourdieu’s
(1993d). In comparison with the concept of “field”, “Art World” lacks the idea of struggle and
power developed in the idea of field. Still, because it is a more neutral term, I deploy it
whenever I aim at a concept wider than field.
Betweenness Centrality. See Centrality.
Centrality. The idea of centrality in SNA is present since Bavelas’s early studies
(FREEMAN, 1979, 2004). The main promise is to deliver a set of measures that yield relevant
characteristics about each node (and more recently, edges). In spite of this promise, Scott
(2000) points that there still exists much confusion around these concepts. For the purpose of
this thesis, I explain three common measures. Freeman (1979, p. 218), offers the following
schematic network in order to explain centrality measures:
90
P3
P2
P4
P1
P5
Each point is a “node” and each segment uniting two points is an “edge”. When two nodes
are connected, they are “adjacent”. The Degree Centrality is the number of nodes to which a
node j is adjacent. Hence, in our example, P3’s Degree Centrality is 2 (the Freeman Degree
standardizes this measure taking in account the size of the network).
In order to go from one point to another, we have to follow a “path”. From P1 to P5 I point
two paths: P1ÆP2ÆP4ÆP5, and P1ÆP2ÆP3ÆP4ÆP5. The first path is the shortest one
between P1 and P5, hence it is the “geodesic” among P1 and P5. The Betweenness Centrality
of j is number of geodesics that pass through this node, adjusted for the probability of using
that path, whenever more than one geodesic exists between two nodes. In our example, no
geodesics go through P1, P3 or P5, hence their betweenness centrality is zero. In contrast, P4
and P2 concentrate each one three geodesics, hence each one has degree “3”. This is the most
used centrality degree used in this thesis. As Freeman states: “Betweenness is useful as an
index of the potential of a point for control of communication” (FREEMAN, 1979, p. 224,
italics in the original; WASSERMAN; FAUST, 1997, for a review)). Scholars have associated
to the concept of “social capital” one’s capacity to access resources in a given social space
(see BURT, 2001, for a review). Burt suggests a measure of Structural Holes in order to
obtain a proxy for social capital. A structural hole is the property of one’s ego network of
concentrating two non-redundant contacts. An actor with high number of structural holes will
have better chances of obtaining fresh information in comparison with those actors with few
structural holes. Burt’s suggestion advances a long genealogy of measures based on the
simmelian concept of brokerage. Yet, Borgatti, Jones, and Everett (1998) argue that the new
measure doesn’t displace older ones. The “betweenness centrality” measure might be more
adequate and sophisticated than the “structural hole” measure when data on the whole
network is available.
Freeman explains also the concept of closeness. Closeness is the measure of independence a
node has in relation to other nodes. For instance, in the example above, P1 always depends on
P2 to reach other nodes. Hence, the higher the number of nodes one has to access in order to
access other nodes, the lower its closeness centrality. As a consequence, we first measure a
point’s Farness, and then calculate its centrality (in inverse relation). In our example, P1 and
91
P5 have to access the highest number of nodes. P1’s Farness is 8 (for every node in the
network it must reach, count the number of steps; P2’s is only 5) which yields a closeness of
50, lower than P2’s 80.
As a concluding remark, it is worthy to recall Salancik’s (1995) warnings on an exaggerated
focus on point centrality. Although centrality sheds light to individual action, it might lead the
researcher to loose perspective of the whole structure.
Closeness Centrality. See Centrality.
Coupled (Systems): Parsons (1959) provides a good example of a coupled system: schools
exist in order to socialize children in the adult world, as well as select the brightest ones to
college. Also, Parsons suggests that women pursue college education in order to become good
role models to their children. Hence, we state that in a coupled system, all parts and
functions are connected to each other and the feedback loops are predictable and
analyzable. On the school system example, one may ask what would happen if women did
want to pursue their careers, instead of simply pursuing high education in order to become
good mothers. Already in the functionalist tradition we observe sharp criticism to this
conception. Merton (1968) analyzed the unpredicted effects of social action. Weick (1976)
provides a full rupture, suggesting the existence of loose coupling systems. Loose coupling
systems are present when several means can produce the same result, there is loose
coordination, weaker regulation, and slow feedback time. If feedback times are slower,
organizations will not adapt immediately to environment changes. In contrast, it will probably
buffer its internal operations from minor changes, while providing autonomy to boundaryspanning individuals to deal with environmental turbulence. As a result, the only sub-systems
will adapt to environmental changes, preserving the overall system’s consistency. Meyer and
Rowan (1991) advanced Weick’s proposition, by suggesting that boundary spanning
individuals might be able to decouple the signals to the environment from the internal
organizational mechanisms. As a result, the signals sent to the environment would not be
necessarily representative of the organization’s actual processes. Scott (2001) suggests
that although the idea of decoupling was posited as opposing to the idea of coupling, one
should consider that organizations fall at a continuum between these two poles. In other
words, it is impossible complete coupling or decoupling.
Decoupled (Systems): see Coupling.
92
Degree Centrality. See Centrality
Field: In the “Introduction” and in the “Fields and Networks” sections I provide two
definitions of field. The first is the neo-institutionalist’s most common intuition that a field
comprises all involved actors who are involved or impact directly a sector or industry. The
second comes from Bourdieu, where a field is a social space where actors struggle in two
different levels: they compete to amass capital (especially symbolic) and they struggle
for the right of defining what is valuable for the field’s members. I rarely use
“organizational field”, for it is too close to the “art world” concept, and it requires a focus on
the “organization” as an underlying and well defined sub-unit. Rather, I prefer to use
Bourdieu’s “field”, for it refers directly to the set of individual musicians engaged in the
creation of music and struggling for recognition, and to other actors, like critics, producers,
etc.
Institution: Scott (2001; 2005) postulates that institutions are “variously comprised of
‘cultural-cognitive, normative and regulative elements that, together with associated activities
and resources, provide stability and meaning to social life”. (SCOTT, 2001, p. 48). This
definition of institution might be too broad for an economist. North (1992), for instance,
defines “institutions” only those formal mechanisms that regulate social and economic life.
DiMaggio (1998) points out that the economists’ definition of institution fits well a family of
theories closer to rational choice. Sociologists, in comparison, find “institutions” everywhere,
from a handshake to a classification system. I espouse the sociologist definition of institution.
Highly institutionalized fields are marked by a intense sharing of cognitive schemata, strong
professional ethos (in both musicians and critics bodies) and selection mechanisms, and
emergence of formal institutions like academic programs.
Institutionalized: see Institution.
Organizational Field: see Field.
Social Network: see Structure.
Social Structure: see Structure.
Symbolic Network: Ansell (1997) defines a symbolic network as the set of relations among
diverse discourses. In contrast with a social network, it doesn’t entail an actual intearaction
among social actors. The links between discourses might be defined as opposition, consent,
articulation, etc. Ansell recovers the history of insurgence of the working class in France,
93
from 1887 to 1894 in order to show how collective action was possible as a collective
discourse emerged from previously conflictive ridden discourses. Mische (2006) performs a
similar analysis by describing how the Collor impeachment was possible as conflicting
disourses converged around the impeachment goal.
Structure: The term “structure” is a powerful metaphor found in many all sociological
schools. Whenever I refer to this broadly defined metaphor, I use the term “social structure”.
As Sewell (1992) states, the word “structure” hasn’t received yet a clear-cut definition.
Maryanski and Turner (2000) approach the subject tracing the use of this term to Spencer and
later to Durkheim. Durkheim (1999) describes a society as increasingly structured as it
becomes more differentiated. Parsons (1949) borrows from Durkheim the elements for his
functionalism: differentiation has to be followed by integration and the generation of a
“conscience collective”. As a result, Parsons offers a sophisticated theory on how the
personality, cultural, relational and institutional dimensions are entangled together in a selfordering system. Nadel (1957) performed a critical analysis on the Parsonian system,
disentangling the actual social interactions from the roles held by social actors. Nadel defines
a social Structure simply as “an overall system, network or pattern of relations” (p. 12).
By “Network” (sometimes I call it a Social Network Structure), Nadel meant “the
interlocking of relationships whereby the interactions implicit in one determine those
occurring in others” (p. 16). As Sewell (1992, p. 6) shows, while structures refer to the
principles underlying action, the social system refers to the pattern of relations. As a
consequence, what Nadel chose to call “structure” is closer to Giddens’ “social system”.
DiMaggio and Powell classic text on organizational fields (1983) depicts the level of
structuration as a key variable in a field. Scott (2005) suggests that a social structure
encompasses “schemas, rules, norms, routines”. By including these categories, closer to social
practices, Scott suggests a concept closer to Gidden’s suggestion of structure: “[Structure is
constituted by] rules and resources, recursively implicated in the reproduction of social
systems. Structure exists only as memory traces, the organic basis of human
knowledgeability, and as instantiated in action” (GIDDENS, 1986, p. 377). Hence,
“structuration” in the neo-institutional tradition follows closely Giddens’ proposition. It is
both the set of schemas, beliefs and rules, as well as the hierarchy among them, or the field’s
meta-logics. As far as neo-institutionalists are concerned with how clashing institutional
spheres are organized through a hierarchy, I’ll favor the use of “meta-logics” term for the
neo-institutional understanding of structure.
94
As shown on section 2.3, Bourdieu’s concept of structure is grounded on the objective
difference of amount of capitals among social actors. As a consequence, Bourdieu grounds his
idea of structure on an objectivistic approach. Nevertheless, this structure is enacted by
individual actors and represented by one’s habitus. As far as the habitus is at the same time
structured and structuring of social relations, Bourdieu attempts to avoid the objectivist and
subjectivist chasm. Yet, he seems to never give up the idea of an objective social structure
which is empirically given by the global distribution of capitals. Although the chapter “Fields
and Networks” explores the intersection between Bourdieu’s and the social network analysis,
its aim is not a synthesis between the two concepts of structure. I chose to call “structure” the
pattern of relations given in social interactions (hence, in agreement with Nadel). In
contrast, section 2.4 and 2.5 show that Bourdieu borrows from Simmel the same elements that
fed the social network concept of structural equivalence. Because the “capital” and the
“interactional” dimensions are not always coupled to each other, I claim that it is legitimate to
reserve the word “Structure” for the former approach, while keeping the latter as an actor’s
attribute. To be sure, the term structure is used in different ways other than Nadel’s
proposition whenever I aim at discussing the other theoretical propositions.
Uncoupled (Systems): see Coupling.
95