The international film actor network played a key role in the development of what Latapy, Magnien and Del Vecchio (2008:33) refer to as a “post-1998 network analysis”. This network was one of the example networks used by both Watts and Strogatz (1998) and Barabási and Albert (1999) in developing their network models, and these two publications inspired a multitude of studies that analysed the topological properties of a variety of real-world, complex networks – from neural networks to airline networks. Numerous studies were done after these publications that also used the international film actor network, usually relying on the data provided by the Internet Movie Database (www.imdb.com). There are, however, two problems with previous studies of the international film actor network. First, while the Internet Movie Database is by far the most comprehensive database available on the international film industry, it is not complete or entirely accurate – especially when non-Hollywood industries such as the Afrikaans film industry are considered. Secondly, the international film actor network only forms part of the film industry as a whole, and cannot be considered a complete social network.
To address these issues the current study gathered data on the Afrikaans film industry directly from the end credits of the films themselves, taking into account 55 Afrikaans films that were released from 1994 to 2014 (these films are listed in the article). Secondly, all role players listed in the credits were entered into a database, including role players such as directors, assistants, make-up artists, power generator operators and extras. This allows for an analysis of the Afrikaans film industry as a whole, rather than just the actor component. Nevertheless, the Afrikaans film actor network was also analysed in order to facilitate a comparison with the international film actor network. The article then focuses on the small-world features of this network, namely average path length and transitivity or clustering, against the background of studies of the international film actor network and other complex networks. By taking into account the 805 103 ties between 6 274 people in this industry, it was found that the Afrikaans film actor network and film industry as a whole have a shorter average path length (2,353 and 2,167 respectively) and higher transitivity (0,937 and 0,92 respectively) than what was found in studies of the international film actor network, suggesting that this is a very close-knit industry. It is also shown that non-actor role players in this industry are important contributors to the industry’s small average path length, but also that the high number of people who contribute to only one film (such as extras) result in the slightly lower clustering coefficient for the industry as a whole when compared with the film actor network.
Apart from studying these macrolevel topological features of the Afrikaans film industry, the article also conducts a node-level analysis using degree, betweenness, closeness and Eigenvector centrality in order to identify the key role players in the Afrikaans film actor network and the Afrikaans film industry as a whole. In doing these calculations, both the bipartite and single-mode renditions of these networks are taken into account, since this addresses Newman’s (2010:125) concern that the projection of the bipartite network on to a single-mode network “discards a lot of the information present in the structure of the original bipartite network and hence it is, in a sense, a less powerful representation of our data”. It is shown which actors, directors and people on the production side played important roles in this industry over the past two decades, and their positions in the network are also indicated visually using the force-directed layout developed by Fruchterman and Reingold (1991). In the actor network Lelia Etsebeth, Andrew Thompson, Altus Theart, Ian Roberts and Richard van der Westhuizen are shown to have played important roles in this industry. Some examples of key figures on the production side are also highlighted, including C.A. van Aswegen, who contributed to a large number of films in his role as colour specialist, the sound engineer Basiami Segola, the boom operator Doctor Ntlama, and the make-up artist Julia Rubinstein. Using the example of Remano de Beer, who made the film Spoofie,it is also shown how valuable it is to use these centrality measures together and in conjunction with the Fruchterman and Reingold layout algorithm, since closeness centrality taken by itself overestimated Remano de Beer’s importance in this industry.
While this article constitutes the largest analysis of an Afrikaans cultural network to date, suggestions are also made for further research. This includes mapping the entire Afrikaans cultural industry that could look at how the film industry ties in with music, literature and theatre – a suggestion found by identifying the important structural role played by the Afrikaans musician Richard van der Westhuizen in the Afrikaans film industry.
Keywords: Afrikaans film industry; Afrikaans culture; Social Network Analysis (SNA); complex networks
Read the article in Afrikaans: ’n Veelvlakkige netwerkontleding van die Afrikaanse filmbedryf en filmakteurnetwerk (1994–2014)