Seite - (000170) - in Knowledge and Networks

163 71 % of recording sessions in 1960, falling to 54 % by 1969. By contrast, the large European cities cited in Fig. 8.3 had 19 % of recording sessions in 1960, but rising to 27 % towards 1969. Jazz scholars identify this period as strongly fragmented because of the adoption of distinct styles. Figure 8.4 shows that the average betweenness centrality during this period is higher than the average degree centrality, indicating higher reliance on brokers. With greater stylistic differentiation emerging, musicians can be expected to play increasingly among peers sharing the same style. The shift from the more cohesive configuration seen in the1945–1960 period to the less cohesive configura- tion of the 1945–1960 period is also accompanied by a stark change in the pattern of relationships (see Table 8.5 for the statistically non-significant correlation between the 1955–1959 and 1960–1964 periods). Figure 8.1g shows the division of the field’s core into two, nonetheless intercon- nected, shared cores. In other words, although musicians were going in different directions, the process cannot be termed fragmentation, because the field’s blocks remain connected. Elite musicians (from both generations) are represented occupy- ing the shared Core blocks together with Ivy League musicians. Figure 8.2g reveals a rearticulation of Elite-2 from 1960 to 1964, with the cluster occupying a more central role, articulated with Shooting Star-2 and Elite-1. In the period from 1965 to 1969, the network was relatively stable in comparison to other periods (see Table 8.5, 0.333 correlation across periods), while the tendency for further prominence of brokerage over cohesion was still operating. Table 8.1 indicates that the average betweenness centrality is higher than the average degree centrality. In comparison to the previous period, a clearer core-periphery configura- tion (Fig. 8.1h) is seen. During this period, several Elite-2 musicians were present at the core (Fig. 8.1h). As a group, though, Elite-2 became less cohesive (Fig. 8.2h). New groups emerged as central and cohesive, like Cluster 6, the Up-Starters (Fig. 8.2h). It is worth noting that Elite-2 was neither able to entirely displace older groups nor become a peren- nial cohesive group. In addition, Fig. 8.2h shows a return of Elite-1, although the expectation would be for the core to be held by the newly established generation. This evidence suggests that, given a stylistic crisis in the field, previous Elite gen- erations were able to fill the void left by the competing new generation. Discussion The above analysis reveals a rearrangement in the field structure. While the field was relatively centralized (both geographically and socially) during the 1930–1934 period, it became more cohesive in the following periods, regaining a higher cen- tralization toward the 1965–1969 period (Fig. 8.4). In broad strokes, this is observ- able throughout the different blockmodeling configurations (Fig. 8.1a–h). A clear core-periphery structure is discernible during the 1930–1934 period, with the peripheral blocks becoming more connected throughout the coming periods, and 8 Trajectory Types Across Network Positions: Jazz Evolution from 1930 to 1969