Abstract Recent work in economic geography has investigated how clusters evolve and change over time. Yet our understanding of such processes is still incomplete. Many accounts rest on a perspective that focuses on the development of solitary knowledge ecologies, while neglecting the fact that cluster formation may be triggered when different places are connected and begin to influence each other in mutual beneficial ways. This article argues that conceptualizations of cluster emergence need to understand the crucial ways in which this process is from the very beginning associated with external linkages and trans-local pipelines. A model of cluster formation is presented that suggests how buzz generation is driven by the connections between different localities in four stages: (i) pioneering, (ii) expansion, (iii) off-shoot, and (iv) fusion. We use the case of the global diamond industry, in both inductive and deductive ways, as an example to show that transnational communities and tight networks play a crucial role in forming cross-local connections that can trigger cluster emergence. 1. Introduction Following Porter’s (1990) seminal conceptualization of clusters, many efforts have been made to identify the factors and components underpinning the competitiveness of spatially intertwined industry agglomerations s (Isaksen, 1997; Gordon and McCann, 2000; Simmie, 2006) and their dynamics (Krugman, 1991; Avnimelech and Teubal, 2006; Ter Wal, 2013). When it comes to understanding the mechanisms that stimulate cluster development, most investigations have adopted a single-case study approach with a focus on internal factors, while the potential role of external linkages in triggering cluster emergence has not been well-understood. It is the goal of this article to tackle this shortcoming by presenting a conceptualization of how dynamic localized knowledge ecologies are from the very beginning (i.e. before an actual cluster comes into existence) generated or supported by trans-local knowledge linkages. In retrospect, three main perspectives can be identified that have dealt with the evolution and dynamics of regional clusters. First, cluster-life cycles have been conceptualized as regional versions of industry-life cycles (Pouder and St. John, 1996; Bergman, 2008; Menzel and Fornahl, 2010) in a way that emphasizes the role of technology and technological aging in their development. This view is characterized by elements of determinism in explaining the rise, growth, and decline of specific clusters. Even though external linkages have been considered as relevant factors in this context (Maskell and Malmberg, 2007; Menzel and Fornahl, 2010), they are typically not viewed as drivers of cluster emergence in this body of literature. Second, evolutionary approaches view clusters as path-dependent phenomena and suggest that new paths are likely to develop in regions characterized by multiple industry bundles that are related to one another (Frenken et al., 2007; Boschma and Iammarino, 2009). While there have been a number of studies that focus on cluster formation (Heebels and Boschma, 2011; Neffke et al., 2011), trans-local linkages are primarily considered as drivers of cluster dynamics in later development stages (MacKinnon, 2012), whereas early-stage development is typically connected with local factors (Braunerhjelm and Feldman, 2006; Fornahl et al., 2010). Sometimes, evolutionary studies describe how clusters co-evolve with industry-wide knowledge networks (Ter Wal and Boschma, 2011; Ter Wal, 2013) but do not investigate how these networks drive the generation of what has been termed buzz (Bathelt et al., 2004), i.e. the planned and unplanned knowledge flows associated with traded and untraded interdependencies between co-located firms (Storper and Venables, 2004). An approach often applied in life-cycle and evolutionary studies is a spin-off model that describes the dynamics of locally based start-up processes (Klepper, 2007). A third strand of literature adopts a relational perspective in treating cluster dynamics (Trippl et al., 2009; Fitjar and Rodriguez-Pose, 2011; Klein, 2011; Li et al., 2012). Based on the network tradition of cluster models (Cooke and Morgan, 1998; Gordon and McCann, 2000), the local-buzz-and-global-pipelines model (Bathelt et al., 2004) suggests that the growth of clusters is stimulated by reflexive relationships between vibrant local networks and knowledge pools and trans-local linkages or pipelines (Owen-Smith and Powell, 2004). Although these pipelines link the internal cluster mechanisms with external markets and technological opportunities, the underlying conceptualization remains somewhat static in that it assumes that industry agglomerations with corresponding knowledge flows already exist. Empirical analyses following this line of reasoning typically do not investigate how buzz is generated in the first place and presuppose that buzz is bound to certain localities or territories, which may not necessarily be the case. The above shortcomings will be taken as the starting point in this article. Building on case studies of cluster genesis and formation (Braunerhjelm and Feldman, 2006; Fornahl et al., 2010; Giblin and Ryan, 2012; Ter Wal, 2013), this article will extend our understanding of buzz beyond the context of solitary clusters. First, it explains the generation of dynamic buzz contexts in their initial stages through interaction processes between spatially separated nuclei of an industry. Second, the article conceptualizes the self-reinforcing fusion processes between different buzz contexts before an actual cluster develops. As such, the goal of this article is to connect the initial process of cluster formation with extra-regional linkages and knowledge exchanges (Bresnahan et al., 2001). Based on a methodology that is both inductive and deductive, a four-stage model of buzz generation is presented that explains how different nuclei of an industry become interconnected through trans-local linkages and how the intensification of such linkages generates mutually reinforcing knowledge ecologies. This model suggests that trans-local connections can be instrumental in providing access to knowledge, resources, and growth triggers across regional contexts. In the empirical part of the article, we use the emergence of Antwerp (Belgium) and Surat (India) as clusters in the global diamond industry to illustrate this model. Our argument develops in six parts. In Section 2, we present a four-stage model of how new buzz contexts can be created through trans-local linkages between spatially separated nuclei of an industry. Section 3 discusses the role of communities and buzz in the diamond sector, and Section 4 presents the methodology of our research. In Section 5, the emergence of the Antwerp and Surat diamond clusters is investigated. It is shown how cross-local knowledge fertilization between the two has triggered buzz generation and cluster formation at both ends. Section 6 summarizes the main arguments of the article and points at the wider implications. 2. Conceptualization: toward a model of buzz generation According to the buzz-and-pipelines approach (Bathelt et al., 2004), which has been applied in many empirical investigations (Moodysson, 2008; Fitjar and Rodriguez-Pose, 2011) and simulations (Morrison et al., 2013) over the past decade, the success of a regional cluster depends on reflexive relations between the internal and external knowledge networks of its firms. On the one hand, the firms benefit from a specialized knowledge ecology that develops from traded and untraded interdependencies between co-localized actors in the cluster (Storper, 1997; Storper and Venables, 2004). This knowledge ecology, or buzz context, provides firms with important feedbacks about innovation processes and the actions of competitors. And the co-localization of diverse actors contributes to the establishment of specific understandings and expectations, which allow firms to make sense of the respective knowledge flows (Bathelt et al., 2004). On the other hand, critical triggers for the generation of new knowledge are often not internally produced but associated with strategic partnerships to distant partners and corresponding knowledge flows from other regions or countries (Owen-Smith and Powell, 2004). Firms establish trans-local or global relationships especially for the purpose of obtaining knowledge about external markets and different technologies developed elsewhere (Giblin and Ryan, 2012). When comparing local and trans-local knowledge flows, it appears that these differ in important respects. While some local knowledge flows develop in generic and less structured form as a by-product of regular face-to-face contacts between the cluster actors, trans-local pipelines are more focused on specific objectives, are explicitly structured, and involve considerable uncertainties and costs (Bathelt et al., 2004; Maskell and Malmberg, 2007). To some degree, the nature of the interaction depends on the degree of trust between the partners involved and their shared understandings, interpretations, and visions. Considering the specific cultural and institutional contexts in which firms operate (Schoenberger, 1997), the establishment of trans-local relationships over sometimes large distances is a challenging and uncertain process (Bathelt et al., 2004). While this conception emphasizes that the competitive success of a cluster can only be understood in relation to both its internal and external linkages, the argument developed by Bathelt et al. (2004) focuses primarily on intra-cluster linkages and knowledge dynamics. Trans-local or global linkage patterns are not specifically conceptualized and the role of cluster networks is not explored in this approach. In other words, the buzz-and-pipelines approach does not fully overcome the solitary treatment of clusters that is characteristic of much of the literature. In reality, many firms are linked to and embedded in durable international linkage patterns which are part of global value and production chains (Gereffi, 1994; Humphrey and Schmitz, 2002; Coe et al., 2010). As a result of ongoing network development and investment activities, clusters become linked to other clusters both nationally and internationally (Enright, 2000; Bathelt and Li, 2014). Despite related debates about knowledge ecologies within and between clusters, little is known about the way how such knowledge ecologies emerge in the first place. The conceptualization developed in this article builds on two shortcomings of the cluster literature. First, much of this literature starts with the assumption that clusters already have well-developed local knowledge ecologies, rather than explaining how such contexts are initially created. Second, related knowledge ecologies are often treated as place-bound phenomena that exist within a cluster but not across its boundaries. In addressing these issues, this article develops a model of how two distant nuclei of an industry create ties that allow for systematic knowledge cross-fertilization and, over time, stimulate buzz generation and the emergence of knowledge ecologies at both ends.1 The model explores these processes in four ideal-type stages. These stages describe a possible sequence but should not be viewed as deterministic, since a specific industry or regional context may lead to deviations or open up opportunities for alternative development paths. While deviations from the proposed sequence may be possible, the empirical case studied here, as well as other cases of cross-local knowledge fertilization as indicated later in the article, suggest that the model may be a good starting point to conceptualize the formation of cluster ecologies. We use a two-region model to illustrate the process of cross-local knowledge fertilization and buzz generation along the stages of (i) pioneering, (ii) expansion, (iii) off-shoot, and (iv) fusion (Figure 1). While the first two stages precede the existence of a cluster, the latter two refer to the initial phases of cluster emergence.2 Figure 1. View largeDownload slide Stylized stage model of buzz generation and cross-local knowledge fertilization. Figure 1. View largeDownload slide Stylized stage model of buzz generation and cross-local knowledge fertilization. 1. Pioneering stage. Our model begins with a situation in which two small nuclei of an industry exist in two different localities and/or countries. The initial agglomerations are limited in terms of their size and internal networks. Their knowledge ecologies are highly localized during this stage and not linked to one another. This may be typical in situations of high local external effects with high spatial transaction costs, as described in the model of Scott (1998), or relate to a context of relative political and economic isolation. In fact, many initial industrial developments have a similar structure in their early stages, as suggested by Storper and Walker (1989) in their model about the evolution of new industries and their spatial structure. In an emerging new technology field, for instance, parallel developments in different regions can occur led by firms that are not aware of each other. The existence of separate initial industry concentrations may also be due to limited start-up processes around specific research facilities as in the biotechnology or nanotechnology industry (Feldman and Francis, 2004; Pisano, 2006), or around private incubators as in the ink jet printing industry (Garnsey et al., 2010). The important point is that such configurations initially lack the critical mass of firms and activities necessary to create strong linkages and further growth. They are disconnected from wider markets and other regions and do have not a well-developed knowledge ecology. 2. Expansion stage. In this stage, local networks in both localities begin to develop and spread further, as some knowledge about the advantages of the industry’s products makes its way around and local market legitimacy increases. Firms within each of the local networks are successful in approaching internal markets and begin to access external customers. As a consequence, the two small agglomerations begin to grow independently and, on the basis of internal and early external effects (Storper and Walker, 1989), to form elements of initial clustering. While firms still mainly rely on localized knowledge flows (Trippl et al., 2009), there is a high probability that successful networks expand further. A limited number of new pipelines are established, possibly initiated through existing civic associations and actors, such as local industry associations, post-secondary research institutes, and science parks (Wolfe and Nelles, 2008: 380; Lorenzen and Mudambi, 2013), or international trade fairs that operate as temporary clusters (Maskell et al., 2006). As a result, firms are increasingly attracted to both localities, which begin to form nodes of excellence and are characterized by increasingly widespread ties and a high quality of internal buzz (Andersen and Lorenzen, 2007). It is at this point that firms begin to create linkages with partners in other regions, especially to extend their markets. Within these slowly growing agglomerations, successful entrepreneurs become role models, start-up firms are established, more local networks develop, and the local knowledge ecologies begin to expand (Menzel and Fornahl, 2010). At some point, these regions and their internal knowledge flows may approach a critical mass for take-off; yet, development prospects and expansionary tendencies are still limited. Most importantly, the overall growth path still primarily focuses on regional resources and markets. 3. Off-shoot stage. It is in the off-shoot stage that vertical linkages with other regions are systematically triggered and become more widespread to integrate new trade partners and get access to the other region’s market. Early successes, processes of deregulation, and new market opportunities, combined with the foresight of individual entrepreneurs (Yeung, 2009), lead to an expansion of trans-local business linkages and the establishment of pipelines—a process that may even become a local norm (Bathelt et al., 2004). Firms in the two regions develop linkages that provide access to complementary assets (Niu et al., 2008; Saxenian and Sabel, 2009). The respective other region becomes an interesting target for firms because of its different market and technological specializations that have developed over time. By establishing cross-local linkages, for instance through partnerships and local representatives, firms can tap into these markets, resources, and the specialized labor market pools in related industries (Oliver et al., 2008; Bathelt and Li, 2014). It is during this stage that initial cluster development may take place and complementary support infrastructure becomes available. Further direct investments into local affiliates follow and generate additional linkages and knowledge flows (Ciravegna, 2011; Giblin and Ryan, 2012). Due to the different cultural and institutional backgrounds of the nuclei and not being familiar with the other location, the new entrants will not get immediate access to the localized knowledge ecology of the other region, but instead rely on their own knowledge ecology. Although first contacts with local entrepreneurs and businesses develop, the two knowledge ecologies remain largely separate, tied to their respective localized networks, and local businesses and new entrants from the other region, despite being co-located, do not yet engage in closer interaction. 4. Fusion stage. In the last stage of the model, successful migrant entrepreneurs, supported by their peers (Yeung, 2009), have learned to cope with and even adopted some of the structures of their new environment. They create more local linkages and networks while, at the same time, maintaining close relational ties to their original locations. Due to their success, they become role models and attract further entrepreneurs who connect the growing regions through pipelines with their respective home base (Yeung, 2007). In some industries, relational ties may rely on ethnic and/or family business networks (Saxenian, 2006; Henn and Laureys, 2010; Lorenzen and Mudambi, 2013). Overall, cross-local linkages develop further and become stronger. New entrants are now well established in the new locations and become embedded in the respective local environment. Their social division of labor becomes enmeshed with that of local firms. With increasing economic interaction, it is merely a matter of time before joint understandings and institutional contexts emerge. The new entrants now participate more effectively in the local knowledge ecology while, at the same time, intensifying, expanding, and modifying it. This also generates new opportunities for firms that did not engage in cross-local interaction before. As a corollary, the formerly separated and different local-buzz dynamics are being connected, the corresponding knowledge ecologies diversify and become more intertwined, and the regional development trajectories open up and link with each other. A complex interconnected knowledge ecology begins to take shape and creates new opportunities to generate linkages and develop ideas. This does not, of course, imply that the broader buzz context automatically leads to the development of new best-practice standards and immediate growth, as this process may be accompanied by some tensions between the communities involved. In fact, convergence may not be strong between the two emerging clusters, as the application and interpretation of the different knowledge components may still be highly localized, related to the specific requirements of the production systems (Storper, 1997). Yet, favorable conditions are in place at this point, in terms of rich knowledge ecologies and a critical mass, to trigger cluster growth and further spin-off processes in the future, potentially at both locations. This stylized model of cross-local knowledge fertilization provides an illustration and possible explanation of how new, vibrant, and fast-growing buzz contexts are generated in a bottom-up fashion through processes which combine and integrate distant knowledge ecologies that are originally small and locally confined. The model also indicates how the linkages between two low-growth contexts may spark entrepreneurship and innovation as they surpass a critical mass of economic activity (Klepper, 2007). Similar to the model of industrial evolution and spatial change by Storper and Walker (1989), our model of cross-local knowledge fertilization relates localized developments to external growth environments. 3. Methodology To investigate how dynamic local buzz contexts develop at different locations and explore the influence of cross-local linkages between the respective industries, this article draws on a typical case study approach (Seawright and Gerring, 2008; Tokatli, 2015). We chose this approach because it enables us to conduct an in-depth study of complex social phenomena and the processes of how they change over time (Yin 2009), while allowing us to elaborate the underlying causal mechanisms. Our approach is inductive and deductive at the same time. While the selection of the two cases and the original formulation of our model were inspired by earlier research on the diamond industry in Antwerp and Surat (Henn, 2010, 2012, 2013), we began to draw parallels with other empirical studies describing similar phenomena, such as the growing connections between Silicon Valley and Hsinchu in the high-technology industry (Hsu and Saxenian, 2000), between Vancouver and Hollywood in the film industry (Coe, 2000; Scott and Pope, 2007), between Castellon and Emilia in the ceramic tile industry (Oliver et al., 2008), and between Tuttlingen and Sialkot in surgical instruments (Nadvi and Halder, 2005). The suggested model itself was developed through decontextualization to conceptualize the process that we refer to as cross-local knowledge fertilization. It builds on the evolutionary model of regional industrial development by Storper and Walker (1989), who emphasize the role of external linkages as “growth peripheries.” The different stages of our model identify periods in the emergence of clusters when the relationships between local and non-local knowledge linkages undergo systematic changes. The case of the diamond industry and the relationships between Antwerp and Surat illustrate how linkages between spatially separate, small nuclei of an industry develop, and how the fusion of their knowledge ecologies through immigrant entrepreneurs and transnational investments can spawn cluster emergence. This study extends prior research into diamond manufacturing and trade, which shows how members of transnational communities contribute to changes in local cluster structures (Henn, 2010), how the evolution of knowledge flows in global value chains is based on transnational family ties (Henn, 2012), and how the growth of the Indian diamond cutting cluster depended on external knowledge flows (Henn, 2013).3 The article draws on semi-structured interviews with 52 workers and managers in the diamond industry (i.e. diamond traders, cutters, and polishers), 11 representatives of support organizations (e.g. banks) and 12 industry experts in Antwerp and Surat/Mumbai (Table 1), conducted between February 2008 and April 2010. Table 1. Interviews conducted in the diamond industry Place of interviews Number of interviews with Total (i) Diamond dealers and workers (ii) Representatives of support organizations (iii) Industry experts Antwerp 43 10 11 64 Surat/Mumbai 9 1 1 11 Total 52 11 12 75 Place of interviews Number of interviews with Total (i) Diamond dealers and workers (ii) Representatives of support organizations (iii) Industry experts Antwerp 43 10 11 64 Surat/Mumbai 9 1 1 11 Total 52 11 12 75 Table 1. Interviews conducted in the diamond industry Place of interviews Number of interviews with Total (i) Diamond dealers and workers (ii) Representatives of support organizations (iii) Industry experts Antwerp 43 10 11 64 Surat/Mumbai 9 1 1 11 Total 52 11 12 75 Place of interviews Number of interviews with Total (i) Diamond dealers and workers (ii) Representatives of support organizations (iii) Industry experts Antwerp 43 10 11 64 Surat/Mumbai 9 1 1 11 Total 52 11 12 75 To add some background information about the empirical approach, it should be noted that research in the diamond industry is a challenging endeavor for several reasons. First, because of tightly knit ethnic and family networks and practices of being careful about handing out information (Finkelman and Berkotiz, 2012), as well as for security reasons, the industry has always been very secretive, with little information being available in terms of exact data about employment, firm linkages, and commercial activities.4 Second, since the extremely high value of diamonds requires extraordinary safety standards, the dealers’ offices and potential interviewees’ work places are highly secured and cannot easily be accessed. Although it was possible to arrange for some interviews based on written requests, this context required the application of a snowball method to acquire interview partners. This approach was legitimate as it complies with established business practices in the industry of forwarding customer requests. The research process started out with a list of firms collected from regional business directories and was extended through personal contact information from prior research and personal networks. After identifying entrance points into the local networks of diamond manufacturers and traders in the two clusters, further interviewees were acquired through leads from early interviewees. The interviews, in most cases conducted in the interviewees’ offices, lasted between 30 and 90 minutes. The questions asked focused on the following themes: (i) the histories of the firms, (ii) the relevance of transnational networks for the transfer of knowledge, (iii) the position of the firms in the value chain, (iv) the role of kinship relations for carrying out business transactions, and (v) intercultural trading relations. The parts of the interviews most relevant to this article are related to the historical development of the two industry clusters and the processes, incentives, and roles of forging linkages between them. In almost every case, it was possible to record the interview. These recordings were later transcribed, triangulated by cross-referencing with other interviews and data sources, and organized for qualitative content analysis and systematic comparison using MaxQDA text analysis software. It should be mentioned that the evolution of the diamond industries in Antwerp and Surat was also affected by other factors that we do not focus on in this article. These factors include specific market dynamics as well as linkages to other clusters in the diamond industry. This article focuses on the linkages between the two primary regions Antwerp and Surat to elaborate those crucial processes that explain how the two originally separated, distant nuclei of the diamond industry became linked to one another and how dynamic buzz contexts were generated. 4. The role of local and relational buzz in the diamond sector The diamond sector has always been dominated by a small number of closely knit communities and their networks (Richman, 2006). In particular, Jews emerged as dominant players in the diamond trade in Europe since the Middle Ages (Richman, 2006), while in India, the so-called Palanpuri Jains from the town of Palanpur in today’s state of Gujarat became the most powerful group of actors controlling the industry. In comparing both communities, some striking similarities can be found. Both, the Palanpuri Jains in India and the Jews in Europe historically operated in environments in which their faith differed from that of most of their neighbors. This favored intermarriages between different family branches and the evolution of strong mutual and enforceable trust relations between these branches (Shor, 1989; Henn, 2012), spanning even across continents. Furthermore, both groups were active in the money lending business and took on typical middlemen roles (Cort, 2004; Shor, 2008). It is therefore not by accident that Weber (1958: 316) referred to the Banya—in particular the Gujarati Banya (to which some of the Palanpuris belong)—as “the Jews of India.” Supported by the fact that Jewish and Jain religious norms strictly forbid gossip and talking about others in a derogatory way (Finkelman and Berkotiz, 2012), communication and knowledge flows about new market opportunities, price developments, new technologies, and the different “tricks of the trade” developed primarily between peers inside these community networks. This “relational buzz,” made up of quick and reliable knowledge flows, provided the communities in the diamond industry with specific advantages. First, contracts in the diamond industry are generally not enforced by state courts. Rather, diamond merchants rely on trust-based handshake exchanges, which are secured by community organizations such as arbitration boards that can enforce these contracts (Richman, 2006). Gaining necessary information about the trustworthiness of potential transaction partners and sustaining a good reputation in order to carry out transactions, requires ongoing interaction and being “in the loop” of the respective knowledge circuits (Richman, 2006; Siegel, 2009). Such knowledge can be quickly exchanged in the above-mentioned networks, most easily in the local context. Second, while nowadays highly sophisticated technical solutions allow for the exact evaluation and certification of the individual characteristics of diamonds (in terms of cut, color, clarity, or carat) that enable a determination of prices on the basis of so-called price sheets, the diamond market in former times clearly lacked such transparency. As such, assessing the value of single stones almost exclusively depended on an individual dealer's experience and knowledge, leaving a large scope for opportunistic behavior. The potential of such opportunistic behavior was minimized when the business was organized in communities that stuck to the same norms and best practices. Third, stones often traveled back and forth between many different traders before reaching their final customer. It was unusual for a buyer to immediately find the one trader who would be able to offer exactly the stone she/he wanted to buy, and thus, it was necessary to acquire information about (i) the buyer’s specific wishes, (ii) the potential owner of such a stone and where to find the owner, and (iii) the price of the stone (Vedder, 1939; Laureys, 2005). Fourth, exact market information was needed about where and how to cut the rough diamonds. This was best accomplished within the local diamond industry in proximity to other specialists. These unique properties suggest that buzz in the sense of a well-developed local knowledge ecology and network relations have always been of crucial importance in the diamond industry, providing access to a complex mixture of contacts, market trends, cutting techniques, and rumors generated from local contacts. Even though market transparency is much greater today, the diamond market is still considered to be a “bazaar … rife with the buzz of peripheral, confusing, extraneous chatter, of obfuscation, prattle, banter, jokes, cajoling, and schmooze, all of which conceals or obscures the information the trader needs” (Shield, 2002: 4). 5. Case study: cross-local knowledge fertilization in the Antwerp and Surat diamond industries The global diamond business has historically been concentrated in a small number of highly specialized locations across the globe. A number of reasons explain this: (i) through concentration, transportation routes were minimized to reduce security risks; (ii) existing bonds between members of communities involved in the diamond trade had a regional base and catalyzed at the micro scale in everyday life, for example around synagogues; (iii) specialized regional labor pools were historically concentrated; and (iv) this, in turn, made possible the frequent short-notice meetings between diamond cutters and traders that were needed to inspect artifacts. The geographical concentration of actors and activities generated a strong incentive for industry members to locate within the local buzz context and participate in face-to-face contacts on a daily basis. Nowadays, the sector is dominated by five major global diamond clusters: Antwerp, Dubai, New York, Surat5, and Tel Aviv. While being firmly established and large in size today, historically, these industries were often small and did not exhibit strong growth dynamics. In this section, we use the proposed model of cross-local knowledge fertilization and buzz generation to illustrate how the formerly separated diamond industries in Antwerp and Surat emerged as dynamic industry clusters with strong local and cross-local knowledge ecologies, based on the establishment of linkages and networks between them. 5.1 Pioneering stage The pioneering stage in the case of Antwerp and Surat corresponds to the situation during the late 19th century, when diamond processing activities existed at both localities but were small and unconnected. While the roots of the Antwerp diamond industry date back to the 15th century, the local industry remained at a small scale for quite some time. Local linkages and associated knowledge exchanges, which developed in the city and its surroundings at that time, remained limited and infrequent and were confined to public places like specific pubs around the central train station, where diamond dealers used to exchange their products (Kockelbergh et al., 1992: 156). Since trade was almost exclusively controlled by one religious community, the Jews, relevant local knowledge flows not only contained business-related information but also included information about religious matters and private life. Such knowledge exchanges were difficult to grasp for outsiders. Of course, the local activities did not remain without external relations, especially since the industry was dependent on the supply of rough diamonds from abroad. Also, dealers used to sell polished stones to foreign buyers who used to come to Antwerp to acquire diamonds (Vedder, 1939). During this early stage of the industry, trading and polishing took place at the same site. This meant that crucial knowledge flows between the different production stages largely remained in-house. Indian businessmen did not establish representations in Antwerp at that time; neither were there any strategic linkages of Antwerp traders with Surat. The diamond history in the Surat region in Western India also goes back a long time. During the Middle Ages, diamonds were traded between India and Europe through the harbor of Surat (Ball, 1881). From this original nucleus, a small local diamond polishing industry formed, which was already known in the 16th century (Hunter, 1881; Bhatt, 2002). This industry still existed at the end of the 17th century (Fryer, 1698: 113), and there is evidence that a simple, small-scale cutting industry continued to exist until the 20th century that processed stones from Indian mines (Fischel, 1956). Nevertheless, between the 18th and the beginning of the 20th century, the diamond industry did not play a major role in India and intensive knowledge exchanges did not develop. The main reason for this was that the Indian diamond mines, upon which the industry historically depended, were exhausted. Aside from the diamond cutting industry, a jewelry business developed and trading went on. Parts of this business were in the hands of Jain traders from the city of Palanpur (Cort, 2004; Shor, 2008). Even though Jews from Europe reportedly were involved in the trade of diamonds from Surat to Antwerp, the regions developed independently from each other and were characterized by distinct, separate knowledge networks that remained relatively stagnant during this stage. In this respect, the situation resembled that of two localities with limited and unconnected knowledge ecologies and little growth as illustrated in Figure 1. 5.2 Expansion stage The situation of the two independent localities began to change around the turn of the 20th century. In the late 1800s, the diamond sector in Antwerp experienced significant growth due to large amounts of rough diamonds that found their way to Antwerp workshops in the aftermath of the discovery of rich South African diamond supplies (Kockelbergh et al., 1992). In fact, while in 1891, there were around 300 diamond workers in Belgium, their number increased to 16,000 in 1911. In the same period, the number of diamond cutting establishments rose from 5 to 300 (New York Times 1912). Associated with this was a strengthening of Antwerp’s institutional basis as illustrated in the foundation of specific labor unions, employer associations, and specialized banks that served the diamond business. Furthermore, so-called diamond bourses were established that aimed at facilitating trade by relocating deal-making from the streets or public cafés and concentrating it in large trading halls (Kockelbergh et al., 1992). These diamond bourses also provided the traders with opportunities to chat with one another, exchange experiences, and make business in front of each other. As a result, important local knowledge flows were stimulated and market transparency improved (Siegel, 2009).6 One interviewee emphasized that “the bourses also have another function, and that is a social function which I think is very important in our business. You need that. Many people do not even trade on the floor but they will come once a day or once a week or whatever. They talk to their colleagues, they will sit, have a cup of coffee, and they will talk and news gets around, you know: ‘Did you hear, Mr. X is at the moment buying ten per carat. And Mr. Y is always a bit of a trouble. Be careful, do not give him too much credit.’ And so forth and so forth. It is a whole social thing.” The growth of the Antwerp diamond industry went along with the strengthening of local networks. In this period, successful businesses began to develop initial vertical linkages to other regions as suggested in Figure 1. These became the first trans-local pipelines. For instance, linkages to the United States were aimed at tapping into new markets and at widening market access (New York Times, 1898). It was also in this context that initial strategic linkages between Antwerp and Surat were created. By 1919, 18 diamond traders from Europe had established offices in Bombay selling stones that had been polished in Belgium to the local rulers in what was then referred to as British India (Shor, 2008: 21). Such interactions across both localities were established because India developed into an important and growing market for jewelry during the colonial period. Only a little earlier, the first Palanpuri jewelers had made their way from Palanpur to Bombay (today’s Mumbai), where they wanted to make a living. Being very successful businesses, they strongly supported their peers from Surat in joining the trade. One of the early pioneers of the Palanpuri community was Shri Surajmal Lallubhai Mehta (1880–1932), who provided his friends and relatives from home with lodging in Bombay and introduced them to local business techniques. He is said to have encouraged around 200–300 Palanpuris to locate in the Zaveri Bazar, the main jewelry street in Bombay (Palanpur Online, n/a). As in Antwerp, the increasing number of industry actors went along with growing internal networks and the establishment of first linkages to other regions. Already in 1896, Surajmal, for example, developed relations with firms from London, Paris, and Antwerp, as well as Delhi and Rangoon. After his uncle—who was his business partner—died in Rangoon, Surajmal opened up a branch in Burma’s capital in 1902, where he sold diamonds, precious stones, and silver works. Additional similar investments in other regions followed. During this stage, selective expansion at both localities occurred as a consequence of market-driven growth triggers. This expansion generated a new institutional basis for the industry, a new organization and division of labor, growing networks between local firms, and the emergence of new businesses. As a consequence, distinct local knowledge ecologies began to develop but, as illustrated in Figure 1, were still limited, despite initial attempts of firms to create linkages with other regions to access new markets. Cross-regional knowledge flows at that time did not play a significant role. 5.3 Off-shoot stage The dynamics of the diamond industry in both localities changed fundamentally during the 20th century, when systematic linkages were created between both regions that went along with flows of goods, resources, firms, people, and knowledge, thereby entering the off-shoot stage.7 At this stage, initial cluster formation was clearly underway, especially in Antwerp, supported by a specialized institutional context. Having been in touch with the diamond traders from Antwerp, some Palanpuris began venturing to Antwerp in order to skip pricy middlemen and source precious stones directly from local traders. During their trips, they would stay in Antwerp for a limited period and later return to India to provide the Indian market with polished diamonds (Henn, 2010). In so doing, they developed more intensive linkages between the two locations, thus enabling important knowledge flows about the specifics of the respective diamond industry, which were beneficial to both localities since they triggered substantial expansion later on as indicated in Figure 1. Between 1921 and 1940, at least nine Indian businessmen visited the Antwerp diamond industry to conduct business. In these days, India developed into a key market for Antwerp firms, triggering further business linkages from Antwerp to India. Export statistics suggest that India, after the United States, had become the second most important market for Belgian firms by 1939 (Shor, 2008). As a consequence, some long-term cooperations between firms from both locations developed, which led to intensive knowledge exchanges that provided the two sides with crucial knowledge about the other region’s production and market structure, thereby intensifying existing relations between the two. One interviewee pointed out that a Belgian firm during the 1930s made contact with Indian firms that acquired polished diamonds in Antwerp. The firm developed an especially close partnership with one firm from Bombay, which had bought diamonds on a regular basis. The Indian partners helped the Antwerp firm not only to sell diamonds to the Far East but also organized a joint business trip to India in 1938. However, such developments were relatively rare at this time. Only about 20 diamond traders moved from India to Antwerp to establish small offices, where they stayed for a certain period until they were replaced by family members or friends (Henn, 2010). While this led to a regular movement of business representatives between the two locations, presence in the other cluster gradually became permanent. Yet at this early stage, actors were only weakly embedded in the local networks of their host locations. Antwerp-based diamond firms benefited from local networks and vertical linkages with their Indian representatives, while Indian firms in Antwerp relied on their own networks and corporate linkages. The knowledge ecologies of the two groups of diamond firms remained largely separate, and exchanges between them were limited. In 1940, World War II caused all Indian business people in the diamond industry to leave Belgium, and a little later, the diamond trade came to a complete halt (Laureys, 2005). After World War II, however, some Indian diamond traders returned to Antwerp. Since the Republic of India had forbidden the import of rough diamonds for political reasons, the Palanpuris now started sourcing such stones in the Flemish capital (Henn and Laureys, 2010) and processing them locally. When organizing this, they fell back to their pre-World War II contacts—as far as this was still possible after the Shoah—and started rebuilding their former trans-local pipelines that not only enabled the transport of artifacts but also generated listening posts to access remote knowledge in the sense of Maskell (2014) about both manufacturing techniques and market developments for polished stones. Especially the inflow of knowledge about production techniques was important in order to keep up with worldwide manufacturing standards, since the Indian industry in the 1960s applied obsolete production techniques (Visvanath, 1972: 21). Interviewees mentioned that already in the late 1940s, a master diamond cutter from Antwerp trained a number of Indian diamond cutters on how to use modern techniques of diamond manufacturing (Henn, 2013). In the other direction, business delegations were sent to Antwerp to source important production-related knowledge. One Belgian diamond trader remembered Indians coming to Antwerp “because we had a marvelous manufacturing business here, and they came to learn the skills … [that existed] in Antwerp at that moment.” Since this new manufacturing knowledge quickly spread to India through the social networks of the Indian diamond traders, these connections were crucial steps for the development of the Indian industry. In fact, the transfer of new technical and market information led to the establishment of many new diamond cutting firms as described in an unpublished study of the Indian diamond industry commissioned by De Beers (Gimpel and Santini, 1967: 3). By creating a new customer base in India, the Antwerp diamond dealers were able to sell certain diamonds at much higher prices than before, thereby also creating the foundations for downstream jewelry production in India. In Belgium, in contrast, the same segment of stones was under much pressure due to the high labor costs and stones were increasingly crushed mechanically to produce inexpensive diamond powder. This example shows how both emerging clusters mutually benefited from the presence of businesses from the respective other region. Yet up to the 1960s, still only a limited number of firms were involved in such linkages. While the Belgian diamond industry grew larger with almost 20,000 workers, the Indians were only able to catch the attention of relatively few Belgian traders8 (Henn, 2012). As the Indian traders continued to travel back and forth, there were only a handful of permanent Indian diamond offices in Antwerp. And since the Indian presence in Antwerp did not yet have a huge impact, two distinct and mostly separate knowledge ecologies developed in Antwerp as illustrated in Figure 1. Overall, however, the relations between both locations had intensified and each was developing into a more fully-fledged cluster with a large number of diamond manufacturers and traders and stronger institutional support and related service industries. 5.4 Fusion stage This drastically changed in the 1960s, when knowledge flows and linkages developed between the Indian and Belgian traders and broader buzz contexts evolved in a way that integrated local and non-local knowledge ecologies, as shown in Figure 1. This was the result of the fusion of networks and linkages between the now developing diamond clusters. By exploiting the advantages of low labor costs in their home country, the Palanpuris were able to develop a new market segment of small diamonds which, incidentally, met the demand of the growing US middle class. In addition, the Indian diamond industry benefited from developments on the supply side. The Australian Argyle deposits, for instance, which were being mined since the 1980s, carried exactly the type of stone that the Indians specialized in (Henn, 2012). As a consequence, Indian firms benefited from a growing mass market for small diamonds, providing existing firms with enormous growth opportunities, while leaving room for new firms to be established. Supported by their relatives and friends, many Palanpuris used this opportunity to enter the market and, by systematically sending family members abroad, developed new pipelines to Antwerp to source rough diamonds (Henn, 2013). Being the largest diamond trading and manufacturing center during this period, Antwerp became the most important global hot-spot for sourcing information about the markets for rough and polished diamonds and for related cutting techniques. This set the stage for the establishment of further permanent representations through offices and affiliates and much broader knowledge exchanges between the two clusters. While the Indian newcomers initially traveled between Antwerp and Mumbai, increasing experience in Antwerp led them to establish and expand their own permanent offices in the local cluster. Entrepreneurs realized that having a constant presence in Antwerp would provide them with opportunities to establish daily face-to-face contacts and strengthen their reputation through repeated interactions. This generated important business advantages in the diamond business and supported close and trustful business relations. This was reiterated by many interviewees and one trader emphasized that he had already been “coming to Antwerp for ten years, at least five to six times a year for a weekend, to buy rough diamonds and go back. … There was huge potential which I saw in moving to Antwerp … because when you are here … you are 24 hours … in this market. People know you, they come to you. You have a much better chance to get the diamonds you want.” As a result, bridges were increasingly established between the formerly distinct knowledge ecologies of the Belgian and Indian diamond traders. The consequences were more intensified business relations and new growth triggers on both ends. Moreover, participation in the local buzz context at both the home and the host location became a crucial factor that drove permanent relocations—first from India to Antwerp. The fact that the Palanpuris in Antwerp—although originating from different Indian firms and families—lived in the same neighborhood (in the area of Wilrijk), that they did not speak Flemish well, and that they tended to stick together during their leisure activities (Siegel, 2009; Henn, 2012) explains why market knowledge could rapidly spread amongst their peers in Antwerp. Although this indicates a certain reclusiveness of the Palanpuris community, their presence in the diamond industry stimulated broader local knowledge flows in manifold ways—a process which led to the adaptation of routines established by the other community and stimulated mutual learning processes. First, an increasing number of Indian dealers became bourse members due to the closer interactions between the communities. According to the membership lists of the bourses, about 10 percent of all bourse members were of Indian origin in 2008.9 As bourse members, the Indian diamond dealers quickly learned about new trends in markets and prices, trading techniques, and customer–supplier relations. Similar processes of learning also occurred in the manufacturing sector, related to the participation in local networks and observation of best practices of the other community (Henn, 2013). This did not just provide advantages to the Indian firms but also benefited the local Belgian firms, as Surat businesses provided Antwerp dealers with crucial information about the specifics of the developing Indian market and helped them to explore ways to get a foothold in that market. Second, interviewees confirmed that the growing cross-local knowledge ecologies led to changing trading rituals and interaction patterns, in terms of how business negotiations or conversations with government officials were conducted. While the Jains, for example, adopted the Jewish tradition of sealing business deals in the diamond sector by shaking hands and saying “mazal u’bracha” (Hebrew for “good luck and blessing”), the Belgian traders also adjusted their trading culture. One Jewish trader from Antwerp described the differences they coped with as follows: “They are different but they are very good businessmen. … We have to learn from them, they have patience for hours, for days – for one dollar more. … We get nervous; … they are always very quiet. They are different.” Even more challenging were extended credit lines used and introduced by the Palanpuri, which the Antwerp traders had to adopt in order to stay competitive. While the above examples primarily refer to the case of the Surat firms that became part of the Antwerp cluster and contributed to the development of a dynamic cross-community and cross-cluster knowledge ecology in Antwerp, a corresponding process also occurred in Surat, albeit a little later and not with the same intensity. In order to keep up with the worldwide development of production techniques in the diamond sector, some Indian firms started to hire Belgian master cutters (Gaitonde, 2013). The master cutters worked for a certain period in Indian firms where they taught specific cutting skills, thereby transferring knowledge developed in Belgium to the Indian production context. By taking advantage of relaxations in rules allowing foreigners to open up businesses, some agents from Antwerp established permanent footholds in Surat—a process that was supported by the raw diamond supplier De Beers in order to become more aligned with consumer markets and develop new jewelry brands (Muller, 2003). The Antwerp-based firm IDH Diamonds, for example, opened a new office in Surat in 2003. By doing so, the firm aimed to increase its Indian sales (Singer, 2003) while acquiring important knowledge about the latest developments in the Indian diamond and jewelry market. More recently, important trading associations from Antwerp also fostered linkages with India in the hope to further strengthen Indo-Belgian business ties. In 2010, for example, AWDC, an industry-established foundation with the mission of serving and supporting the Belgian diamond sector, organized the Antwerp Diamond Day in Mumbai. This networking event was aimed at introducing the most important players in the Indian diamond trade to their Belgian counterparts (Miller, 2010). Another example is the Antwerp-based firm HRD, one of the largest and most important diamond grading laboratories worldwide. HRD opened up two offices in Surat that offer quality products and services to the local diamond industry (HRD, 2014). These linkages indicate how knowledge was transferred both by agents from Antwerp to Surat and by Indians to the Belgian diamond industry. In this fusion stage, cross-local knowledge fertilization took off as depicted in Figure 1 since the number of direct permanent linkages between the two meanwhile well-developed clusters increased and complex trade, supplier, and knowledge networks developed. Cross-local knowledge flows intensified and corresponding knowledge was systematically applied to both contexts, thereby stimulating fusion processes and cluster growth in both regions. 6. Conclusion: the dynamics of cross-local knowledge fertilization and buzz generation This article addresses an important gap in the literature on clusters: a gap in our understanding of how clusters emerge and come into being. The contribution of our article goes beyond discussing a specific case in that we explore the very principles that impact the establishment of localized knowledge ecologies, or buzz contexts, and emphasize how their emergence can be triggered or supported by relations to other localities. We propose giving up a solitary view of emerging clusters in favor of a model that conceptualizes the process of buzz generation as one that is from the very beginning based on linkages with external markets and transaction partners in other regions and/or countries. In our relational approach, which draws on a research process involving a combination of inductive and deductive steps, we introduce a model that shows how linkages between initially unconnected nuclei of an industry stimulate the formation of dynamic knowledge ecologies and later trigger the emergence of clusters. The two-region model describes this process through a non-deterministic sequence of four stages. In the pre-cluster pioneering and expansion stages, the two regions are characterized by small industry agglomerations that are largely unconnected. This changes in the off-shoot and fusion stages when clusters begin to emerge and cross-local linkages trigger local growth and the formation of vibrant knowledge ecologies. The empirical example of diamond manufacturing and trade in Antwerp, Belgium, and Surat, India illustrates this model well. The case draws attention to the specific mechanisms that stimulate the expansion, cross-fertilization, and fusion of separate buzz contexts related to dynamic linkages between them. In our example, transnational entrepreneurs and family business networks were at the core of this process because they facilitated temporary and permanent presences in the two locations (Henn, 2012). They generated intensive cross-local linkages and knowledge flows and initiated the development of industry clusters at both ends. In this process, it was particularly important to access new market segments and to connect with related knowledge bases, production skills, and technologies. During the early stages of development, direct investments played an important role, particularly when no other reference points existed in the host region. All of this provided the architecture for durable cross-cluster knowledge networks and dynamic buzz contexts later on (Bathelt and Li, 2014). The case of the diamond industries in Antwerp and Surat also shows that the process of cross-local knowledge fertilization does not need to follow a linear pattern. Even though parts of the originally separated buzz contexts in Antwerp and Surat have merged and led to the development of two dynamic inter-linked industry clusters, the two foundational communities still remain separated in their private lives while conducting business with each other (Helmer, 2009; Siegel, 2009). This also creates tensions. For instance, Indian agents in Antwerp have been accused of pushing their own business interests and those of their community when electing peers to become members of local trade associations. This shows that processes of buzz fusion and cross-local knowledge fertilization are not necessarily harmonious. Rather, they may involve power struggles between the different communities as they aim to gain a competitive advantage. Resolving such conflicts requires time as well as ongoing efforts to preserve cross-community and cross-cluster communication. Of course, the conceptualization presented in this article cannot be viewed as a one-size-fits-all model that applies to every industry and/or region. In some cases, external relations may be less important or simply play a different role, and deviations from or different sequences in the emergence process are possible and should be expected when investigating different economic, cultural, political, or technological contexts of cluster formation. It should be clear that there is no “natural” pathway of cluster emergence and that the processes described in our model do not have a pre-determined outcome. Not every case will result in a tendency toward the development of what Scott (1998) calls a “super-cluster” and some initial cross-local interaction may even fail. We understand our model as an attempt to explain the development of an industry up to the stages where path development takes place. As suggested by Martin (2010), subsequent developments can take different directions and possible pathways may not always lead to the development of a fully fledged cluster. Despite these limitations, we believe that crucial elements of our model may be applicable to industries other than diamond manufacturing and trade. Some of the processes described in our model can be found in similar form in other cases of cross-cluster interaction and dynamics, be it in ceramic tile production (Oliver et al., 2008), surgical instruments manufacturing (Nadvi and Halder, 2005), or high-technology industries (Saxenian, 2006). The Taiwanese entrepreneurs in Silicon Valley, who developed close guanxi-based linkages to their home country and contributed to the establishment of a growing high-technology cluster in Hsinchu, Taiwan, provide strong evidence how cross-local linkages can trigger self-reinforcing growth at both ends (Hsu and Saxenian, 2000). And, finally, the case of the Hollywood and Vancouver motion picture industries shows how close production linkages and a division of labor can be beneficial to the development of both clusters (Coe, 2000; Scott and Pope, 2007). While the two latter cases deviate from the model presented in this article in that one region had already developed into a dominant cluster before systematic linkages were created, cross-local knowledge fertilization generated advantages on both ends and supported respective cluster dynamics. We believe that the existence of such parallels is not an anomaly, but suggests that cross-local knowledge fertilization plays a crucial role in triggering the formation of clusters in the first place. The powerful findings of He et al. (2017), who show that new development paths in China are fundamentally created through external linkages, support this conclusion. Our study helps to understand how extra-regional linkages help an infant development to grow into an industry that is organized around complex multi-scalar knowledge networks and therefore connects to the literature on infant industries and industrial path creation (Quitzow, 2015; Sengers and Raven, 2015). As a corollary, it becomes increasingly clear that cluster emergence cannot be understood as a solitary phenomenon, based on bounded localized networks of interaction and learning only. As illustrated in our case study, the formation of clusters is linked to the development of global value chains (Humphrey and Schmitz, 2002). The model presented in this article suggests that the formation of dynamic clusters may from the very beginning depend on trans-local linkages with similar, competing, or complementary developments in other regions and/or countries. This is also supported by empirical studies about the genesis and growth of information communication technology clusters (Bresnahan et al., 2001) which find that initial cluster success in these industries depended on wider international transaction networks, regional openness, and early access to major world markets. As such, our study can also be interpreted as a call to bring research on global value chains, market development, and industrial clusters closer together. In contrast to Bresnahan et al. (2001), who claim that cluster emergence cannot be triggered through regional policies, we suggest that important lessons for policy and further research can be drawn from our observations. A key implication of our analysis is that cluster formation can sometimes be better understood when investigating the external connections between regions (Parr, 2002). Rather than focusing primarily on internal linkages and knowledge networks, policymakers should pay attention to strategically generating connections with other regions and accessing complementary knowledge assets to facilitate the emergence of clusters, for example, by providing private, semi-private, or public support to migrants when they establish their businesses abroad. Such measures may consist of help in raising start-up capital or getting familiar with local markets and business environments, but may also include legal advice, counseling, and support in building networks with incumbent firms (Desiderio, 2014). Our case, in which two communities of agents have established the basis for economic development in different regions, further suggests that explorations of processes of cluster formation can benefit much from detailed qualitative and historical investigations of the agents and communities involved and their social networks. We believe that the processes discussed in this article are important drivers in the contemporary stage of intensified economic globalization. Future research should therefore put stronger emphasis on investigating the impacts of cross-local knowledge fertilization in comparative studies of different industry contexts. Such research may help to better understand the conditions under which spatially separated communities can co-create dynamic knowledge ecologies and stimulate the emergence of clusters. While our analysis has only focused on two interconnected localities, the evolution of related structures may involve interdependencies between multiple nuclei along and across global production chains. It is therefore necessary to conduct further studies that analyze how the emergence of complex systems of intertwined clusters depend on the manifold linkages between individual localities and their communities of firms and people. Footnotes 1 In doing this, we do not, of course, question the importance of internal processes and other determinants on cluster emergence. 2 In suggesting this model, we do not make specific assumptions regarding the duration of the different stages. While the stages may be relatively short in some industry contexts, they may extend over long periods in others, as for instance in our case study. 3 Of all important diamond clusters worldwide, the examples of Antwerp and Surat illustrate the model of cross-local knowledge fertilization best and without much interference by other important processes. The emergence of the diamond clusters in Tel Aviv and New York was also related to the Antwerp cluster and to Jewish refugees from this cluster (Henn and Laureys 2010). 4 Reservations in the sector have been reinforced by recent debates about moral issues associated with so-called “blood diamonds” (Campbell 2012), which has resulted in strong efforts by diamond firms to avoid any impressions that they may be somehow associated with illegal activities. 5 While all other centers are characterized by the co-localization of diamond manufacturing and trading, Surat acts primarily as a processing center. Trading activities in India are concentrated in Mumbai, a few hundred kilometers away. 6 Even today, the bourses fulfill this function. 7 The growth of Antwerp at that time was also supported by the fact that diamond businesses from Amsterdam started to expand and relocated to Antwerp because of its specialization in market segments that were in high demand, as well as due to lower production costs (Vedder 1939). 8 The first newspaper clip stored in the Antwerp city archive that explicitly refers to Indian diamond dealers in Belgium dates from 1968. Aside from the small number of Indians, this initial disregard can also be attributed to the fact that volume and value of diamonds were relatively small (Henn 2012). 9 In the case of the bourse “Diamantclub van Antwerpen,” the share of Indian members rose from about 3% to 8% between 1983 and 2008 (own calculations based on the registers of members). Acknowledgments Earlier versions of this article, to which both authors have contributed equally, were presented in 2015 at the Annual Meeting of the Association of American Geographers in Chicago and the APROS/EGOS Conference in Sydney. The authors wish to thank the three anonymous reviewers and Giulio Buciuni, Patrick Cohendet, Elisa Giuliani, Daniel Hutton Ferris, Sufyan Katariwala, Robert Panitz, and Lech Suwala for providing helpful and constructive comments on earlier versions of this article. Funding This research was funded by the German Research Foundation and the Canada Research Chair in Innovation & Governance at the University of Toronto. References Andersen K. V. , Lorenzen M. . ( 2007 ), ‘The stretching of weak ties. Clusters, pipelines, and the creation of small worlds,’ Paper presented at the DRUID Winter Conference, Copenhagen, http://www2.druid.dk/conferences/viewpaper.php?id=1101&cf=10 (November 11, 2015, date last accessed). Avnimelech G. , Teubal M. ( 2006 ), ‘ Creating venture capital industries that co-evolve with high tech: insights from an extended industry life cycle perspective of the Israeli experience ,’ Research Policy , 35 , 1477 – 1498 . Google Scholar CrossRef Search ADS Ball V. ( 1881 ), A Manual of the Geology of India. Part III. Economic Geology . Trübner & Co : London . Bathelt H. , Li P.-F. ( 2014 ), ‘ Global cluster networks – foreign direct investment flows from Canada to China ,’ Journal of Economic Geography , 14 , 45 – 71 . Google Scholar CrossRef Search ADS Bathelt H. , Malmberg A. , Maskell P. ( 2004 ), ‘ Clusters and knowledge: local buzz, global pipelines and the process of knowledge creation ,’ Progress in Human Geography , 28 , 31 – 56 . Google Scholar CrossRef Search ADS Bergman E. M. ( 2008 ), ‘Cluster life-cycles: an emerging synthesis,’ in Karlsson C. (ed.), Handbook of Research on Cluster Theory . Edward Elgar : Cheltenham , pp. 114 – 132 . Bhatt K. ( 2002 ), How did Surat become a centre for diamond polishing and diamond trading? 7Stars worldwide, GJEPC Export Awards Supplement, pp. 30–37. Boschma R. , Iammarino S. ( 2009 ), ‘ Related variety, trade linkages and regional growth in Italy ,’ Economic Geography , 85 , 289 – 311 . Google Scholar CrossRef Search ADS Braunerhjelm P. , Feldman M. P. (eds) ( 2006 ), Cluster Genesis: Technology-Based Industrial Development . Oxford University Press : Oxford . Google Scholar CrossRef Search ADS Bresnahan T. , Gambardella A. , Saxenian A. ( 2001 ), ‘ “Old economy” inputs for “new economy” outcomes: cluster formation in the new Silicon Valleys ,’ Industrial and Corporate Change , 10 , 835 – 860 . Google Scholar CrossRef Search ADS Campbell G. ( 2012 ), Blood Diamonds. Tracing the Deadly Path of the World’s Most Precious Stones . Basic Books : New York, NY . Ciravegna L. ( 2011 ), ‘ FDI, social ties and technological learning in new Silicon Valley clones: evidence from the Costa Rican ICT cluster ,’ Journal of Development Studies , 47 , 1178 – 1198 . Google Scholar CrossRef Search ADS Coe N. M. ( 2000 ), ‘ The view from out West: embeddedness, inter-personal relations and the development of an indigenous film industry in Vancouver ,’ Geoforum , 31 , 391 – 407 . Google Scholar CrossRef Search ADS Coe N. M. , Dicken P. , Hess M. , Yeung H. W.-C. ( 2010 ), ‘ Making connections: global production networks and world city networks ,’ Global Networks , 10 , 138 – 149 . Google Scholar CrossRef Search ADS Cooke P. , Morgan K. ( 1998 ), The Associational Economy . Oxford : Oxford University Press . Google Scholar CrossRef Search ADS Cort J. E. ( 2004 ), ‘ Jainas, caste and hierarchy in north Gujarat ,’ Contributions to Indian Sociology , 38 , 73 – 112 . Google Scholar CrossRef Search ADS Desiderio M. V. ( 2014 ), Policies to Support Immigrant Entrepreneurship . Transatlantic Council on Migration : Washington, DC . Enright M. ( 2000 ), ‘ Regional clusters and multinational enterprises ,’ International Studies of Management & Organization , 30 , 114 – 138 . Google Scholar CrossRef Search ADS Feldman M. , Francis J. ( 2004 ), ‘ Homegrown solutions: fostering cluster formation ,’ Economic Development Quarterly , 18 , 127 – 137 . Google Scholar CrossRef Search ADS Finkelman S. , Berkotiz Y. ( 2012 ), Chofetz Chaim. A Lesson a Day. The Concepts and Laws of Proper Speech Arranged for Daily Study. Based on his Works, Sefer Chofetz Chaim and Sefer Shmiras Halashon , Vols 1 and 2 . ArtScrol : New York . Fischel W. J. ( 1956 ), ‘ Leading Jews in the service of Portuguese India ,’ The Jewish Quarterly Review , 47 , 37 – 57 . Google Scholar CrossRef Search ADS Fitjar R. D. , Rodriguez-Pose A. ( 2011 ), ‘ Innovating in the periphery: firms, values and innovation in Southwest Norway ,’ European Planning Studies , 19 , 555 – 574 . Google Scholar CrossRef Search ADS Fornahl D. , Henn S. , Menzel M.-P. (eds) ( 2010 ), Emerging Clusters: Theoretical, Empirical and Political Perspectives in the Initial Stage of Cluster Evolution . Edward Elgar : Cheltenham . Frenken K. , van Oort F. G. , Verburg T. ( 2007 ), ‘ Related variety, unrelated variety and regional economic growth ,’ Regional Studies , 41 , 685 – 697 . Google Scholar CrossRef Search ADS Fryer J. ( 1698 ), A New Account of East-India and Persia, in Eight Letters. Being Nine Years Travels begun 1672. And finished 1681 . R. R. for Ri. Chiswell : London . Gaitonde A. ( 2013 ), ‘Indian diamond cutting and polishing sector,’ Rough & Polished, June 3, http://www.rough-polished.com/en/analytics/77889.html (November 11, 2015, date last accessed). Garnsey E. , Stam E. , Brychan T. ( 2010 ), ‘The emergence and development of the Cambridge ink jet industry,’ in Fornahl D. , Henn S. , Menzel M.-P. (eds), Emerging Clusters: Theoretical, Empirical and Political Perspectives in the Initial Stage of Cluster Evolution . Edward Elgar : Cheltenham , pp. 265 – 291 . Gereffi G. ( 1994 ), ‘The organization of buyer-driven global commodity chains: how US retailers shape overseas production networks,’ in Gereffi G. , Korzeniewicz M. (eds), Commodity Chains and Global Capitalism . Praeger : Westport , pp. 95 – 122 . Giblin M. , Ryan P. ( 2012 ), ‘ Tight clusters or loose networks? The critical role of inward foreign direct investment in cluster creation ,’ Regional Studies , 46 , 245 – 258 . Google Scholar CrossRef Search ADS Gimpel J. , Santini M. ( 1967 ), Changes in the Diamond Trade in India 1964–1967. (Unpublished report to De Beers). Gordon I. R. , McCann P. ( 2000 ), ‘ Industrial clusters: complexes, agglomeration and/or social networks ,’ Urban Studies , 37 , 513 – 532 . Google Scholar CrossRef Search ADS He C. , Zhu S. , Zhou Y. ( 2017 ), ‘ How to jump further and catch up? Path-breaking in an uneven industry space ,’ Journal of Economic Geography , 17 , 521 – 545 . Heebels B. , Boschma R. ( 2011 ), ‘ Performing in Dutch book publishing 1880-2008: the importance of entrepreneurial experience and the Amsterdam cluster ,’ Journal of Economic Geography , 11 , 1007 – 1029 . Google Scholar CrossRef Search ADS Henn S. ( 2010 ), ‘ Transnational communities and regional cluster dynamics. The case of the Palanpuris in the Antwerp diamond district ,’ Die Erde , 141 , 127 – 147 . Henn S. ( 2012 ), ‘ Transnational entrepreneurs, global pipelines and shifting production patterns: the example of the Palanpuris in the diamond sector ,’ Geoforum , 43 , 497 – 506 . Google Scholar CrossRef Search ADS Henn S. ( 2013 ), ‘ Transnational entrepreneurs and the emergence of clusters in peripheral regions: the case of the diamond cutting cluster in Gujarat (India) ,’ European Planning Studies , 21 , 1779 – 1795 . Google Scholar CrossRef Search ADS Henn S. , Laureys E. ( 2010 ), ‘Bridging ruptures: the re-emergence of the Antwerp diamond district after World War II and the role of strategic action,’ in Fornahl D. , Henn S. , Menzel M.-P. (eds), Emerging Clusters: Theoretical, Empirical and Political Perspectives in the Initial Stage of Cluster Evolution . Edward Elgar : Cheltenham , pp. 74 – 96 . Helmer G. ( 2009 ), Jaina in Antwerpen. Eine religionsgeschichtliche Studie . Akademische Verlagsgemeinschaft München : Munich . HRD ( 2014 ), HRD Antwerp Opens Second Indian Office in Surat, Antwerp, Belgium, http://www.hrdantwerp.com/en/news/hrd-antwerp-opens-second-indian-office-in-surat (November 11, 2015, date last accessed). Hsu J.-Y. , Saxenian A. ( 2000 ), ‘ The limits of Guanxi capitalism: transnational collaboration between Taiwan and the U.S ,’ Environment and Planning A , 32 , 1991 – 2005 . Google Scholar CrossRef Search ADS Humphrey J. , Schmitz H. ( 2002 ), ‘ How does insertion in global value chains affect upgrading in industrial clusters? ,’ Regional Studies , 36 , 1017 – 1027 . Google Scholar CrossRef Search ADS Hunter W. W. ( 1881 ), The Imperial Gazetteer of India , Vol. 8 . Trübner & Co : London . Isaksen A. ( 1997 ), ‘ Regional clusters and competitiveness: the Norwegian case ,’ European Planning Studies , 5 , 65 – 76 . Google Scholar CrossRef Search ADS Klein R. R. ( 2011 ), ‘ Where music and knowledge meet: a comparison of temporary events in Los Angeles and Columbus, Ohio ,’ Area , 43 , 320 – 326 . Google Scholar CrossRef Search ADS Klepper S. ( 2007 ), ‘ Disagreements, spinoffs, and the evolution of Detroit as the capital of the U.S. automobile industry ,’ Management Science , 53 , 616 – 631 . Google Scholar CrossRef Search ADS Kockelbergh I. , Vleeschdrager E. , Walgrave J. ( 1992 ), The Brilliant Story of Antwerp Diamonds . MIM : Antwerp, Belgium . Krugman P. ( 1991 ), ‘ Increasing returns and economic geography,' Journal of Political Economy , 99 , 483 – 499 . Google Scholar CrossRef Search ADS Laureys E. ( 2005 ), Meesters van het Diamant. De Belgische Diamantsector Tijdens het Nazibewind . Lannoo : Tielt, Belgium . Li P.-F. , Bathelt H. , Wang J. ( 2012 ), ‘ Network dynamics and cluster evolution: changing trajectories of the aluminum extrusion industry in Dali, China ,’ Journal of Economic Geography , 12 , 127 – 155 . Google Scholar CrossRef Search ADS Lorenzen M. , Mudambi R. ( 2013 ), ‘ Clusters, connectivity and catch-up: Bollywood and Bangalore in the global economy ,’ Journal of Economic Geography , 13 , 501 – 534 . Google Scholar CrossRef Search ADS MacKinnon D. ( 2012 ), ‘ Beyond strategic coupling: reassessing the firm-region nexus in global production networks ,’ Journal of Economic Geography , 12 , 227 – 245 . Google Scholar CrossRef Search ADS Martin R. ( 2010 ), ‘ Rethinking regional path dependence: beyond lock-in to evolution ,’ Economic Geography , 86 , 1 – 27 . Google Scholar CrossRef Search ADS Maskell P. , Bathelt H. , Malmberg A. ( 2006 ), ‘ Building global knowledge pipelines: the role of temporary clusters ,’ European Planning Studies , 14 , 997 – 1013 . Google Scholar CrossRef Search ADS Maskell P. ( 2014 ), ‘ Accessing remote knowledge—the roles of trade fairs, pipelines, crowdsourcing and listening posts ,’ Journal of Economic Geography , 14 , 883 – 902 . Google Scholar CrossRef Search ADS Maskell P. , Malmberg A. ( 2007 ), ‘ Myopia, knowledge development and cluster evolution ,’ Journal of Economic Geography , 7 , 603 – 618 . Google Scholar CrossRef Search ADS Menzel M.-P. , Fornahl D. ( 2010 ), ‘ Cluster life cycles – dimensions and rationales of cluster evolution ,’ Industrial and Corporate Change , 19 , 205 – 238 . Google Scholar CrossRef Search ADS Miller J. ( 2010 ), ‘AWDC organizes Antwerp Diamond Day in Mumbai,’ Rapaport Diamonds.Net, March 18, http://www.diamonds.net/News/NewsItem.aspx?ArticleID=30234 (November 11, 2015, date last accessed). Moodysson J. ( 2008 ), ‘ Principles and practices of knowledge creation: on the organization of “buzz” and “pipelines” in life science communities ,’ Economic Geography , 84 , 449 – 469 . Google Scholar CrossRef Search ADS Morrison A. , Rabelotti R. , Zirulia L. ( 2013 ), ‘ When do global pipelines enhance the diffusion of knowledge in clusters? ,’ Economic Geography , 89 , 77 – 96 . Google Scholar CrossRef Search ADS Muller E. ( 2003 ), ‘Diamond industry brings back sparkle to Surat,’ Polishedprices, March 17, http://polishedprices.com/go/market-news/news∼1179 (November 11, 2015, date last accessed). Nadvi K. , Halder G. ( 2005 ), ‘ Local clusters in global value chains: exploring dynamic linkages between Germany and Pakistan,' Entrepreneurship and Regional Development , 17 , 339 – 363 . Google Scholar CrossRef Search ADS Neffke F. , Henning M. , Boschma R. , Lundquist K.-J. , Olander L.-O. ( 2011 ), ‘ The dynamics of agglomeration externalities along the life cycle of industries ,’ Regional Studies , 45 , 49 – 65 . Google Scholar CrossRef Search ADS New York Times ( 1898 ), ‘Reports of our consuls. Interesting statistics concerning the commercial relations between Belgium and the United States,’ New York Times, August 14. New York Times ( 1912 ), ‘Diamond trade in 1911 had big slump,’ New York Times, March 10. Niu K.-H. , Miles G. , Lee C.-S. ( 2008 ), ‘ Strategic development of network clusters: a study of high technology regional development and global, ’ Competitiveness, Competitiveness Review , 18 , 176 – 191 . Google Scholar CrossRef Search ADS Oliver J. L. H. , Garrigós J. A. , Porta J. I. D. ( 2008 ), ‘ External ties and the reduction of knowledge asymmetries among clusters within global value chains: the case of the ceramic tile district of Castellon ,’ European Planning Studies , 16 , 507 – 520 . Google Scholar CrossRef Search ADS Owen-Smith J. , Powell W. W. ( 2004 ), ‘ Knowledge networks as channels and conduits: the effects of spillovers in the Boston biotechnology community ,’ Organization Science , 15 , 2 – 21 . Google Scholar CrossRef Search ADS Palanpur Online (2010) Surajmal Lallubhai Mehta (1880-1932), Mumbai, http://www.palanpuronline.com/Portal/Personalities1.aspx?ListingID=279&Pan=gvBusiness (December 8, 2015, date last accessed). Parr J. B. ( 2002 ), ‘ Missing elements in the analysis of agglomeration economies ,’ International Regional Science Review , 5, 151 – 168 . Pisano G. P. ( 2006 ), Science Business: The Promise, the Reality, and the Future of Biotech . Harvard Business School Press : Boston, MA . Porter M. E. ( 1990 ), The Competitive Advantage of Nations . Free Press : New York . Google Scholar CrossRef Search ADS Pouder R. , St. John C. ( 1996 ), ‘ Hot spots and blinds spots: geographical clusters of firms and innovation, ’ The Academy of Management Review , 21 , 1192 – 1225 . Google Scholar CrossRef Search ADS Quitzow R. ( 2015 ), ‘ Dynamics of a policy-driven market: the co-evolution of technological innovation systems for solar photovoltaics in China and Germany ,’ Environmental Innovation and Societal Transitions , 17 , 126 – 148 . Google Scholar CrossRef Search ADS Richman B. ( 2006 ), ‘ How community institutions create economic advantage: Jewish diamond merchants in New York ,’ Law & Social Inquiry , 31 , 383 – 420 . Google Scholar CrossRef Search ADS Saxenian A. ( 2006 ), The New Argonauts: Regional Advantage in a Global Economy . Harvard University Press : Cambridge . Saxenian A. , Sabel C. ( 2009 ), ‘ Roepke Lecture in Economic Geography. Venture capital in the “periphery”: the new Argonauts, global search, and local institution building,' Economic Geography , 84 , 379 – 394 . Google Scholar CrossRef Search ADS Schoenberger E. ( 1997 ), The Cultural Crisis of the Firm . Blackwell : Cambridge, Oxford . Scott A. J. ( 1998 ), Regions and the World Economy: The Coming Shape of Global Production, Competition, and Political Order . Oxford University Press : Oxford, New York . Scott A. J. , Pope N. E. ( 2007 ), ‘ Hollywood, Vancouver, and the world: employment relocation and the emergence of satellite production centers in the motion-picture industry ,’ Environment and Planning A , 39 , 1364 – 1381 . Google Scholar CrossRef Search ADS Seawright J. , Gerring J. ( 2008 ), ‘ Case selection techniques in case study research: a menu of qualitative and quantitative options ,’ Political Science Quarterly , 61 , 294 – 308 . Sengers F. , Raven R. ( 2015 ), ‘ Toward a spatial perspective on niche development: the case of bus rapid transit ,’ Environmental Innovation and Societal Transitions , 17 , 166 – 182 . Google Scholar CrossRef Search ADS Shield R. R. ( 2002 ), Diamond Stories: Enduring Change on 47th Street . Cornell University Press : Ithaca . Shor R. ( 1989 ), ‘ Family: diamond’s critical connection. A look inside the tradition-bound world of diamond trading through the eyes of six families ,’ Jewelers’ Circular-Keystone , 160 , 190 – 206 . Shor R. ( 2008 ), The New Moghuls: The Remarkable Story of India’s Diamond People, unpublished manuscript. Siegel D. ( 2009 ), The Mazzel Ritual . Springer : Dordrecht, Heidelberg . Google Scholar CrossRef Search ADS Simmie J. ( 2006 ), ‘Do clusters or innovation systems drive competitiveness?,’ in Asheim B. , Cooke P. , Martin R. (eds) Clusters and Regional Development: Critical Reflections and Explorations . Routledge : Abingdon , pp. 164 – 187 . Singer S. ( 2003 ), ‘Major Antwerp diamantaire opens office in Surat,’ Rapaport Diamonds.Net, May 26, http://www.diamonds.net/News/NewsItem.aspx?ArticleID=8089 (November 11, 2015, date last accessed). Storper M. ( 1997 ), The Regional World: Territorial Development in a Global Economy . Guilford Press : New York . Storper M. , Venables A. J. ( 2004 ), ‘ Buzz: face-to-face contact and the urban economy ,’ Journal of Economic Geography , 4 , 351 – 370 . Google Scholar CrossRef Search ADS Storper M. , Walker R. ( 1989 ), The Capitalist Imperative: Territory, Technology and Industrial Growth . Blackwell : Cambridge . Ter Wal A. L. J. ( 2013 ), ‘ Cluster emergence and network evolution: a longitudinal analysis of the inventor network in Sophia-Antipolis ,’ Regional Studies , 47 , 651 – 668 . Google Scholar CrossRef Search ADS Ter Wal A. L. J. , Boschma R. A. ( 2011 ), ‘ Co-evolution of firms, industries and networks in space ,’ Regional Studies , 45 , 919 – 933 . Google Scholar CrossRef Search ADS Tokatli N. ( 2015 ), ‘ Single-firm case studies in economic geography: some methodological reflections on the case of Zara ,’ Journal of Economic Geography , 15 , 631 – 647 . Google Scholar CrossRef Search ADS Trippl M. , Tödtling F. , Lengauer L. ( 2009 ), ‘ Knowledge sourcing beyond buzz and pipelines: evidence from the Vienna software sector ,’ Economic Geography , 85 , 443 – 462 . Google Scholar CrossRef Search ADS Vedder M. E. ( 1939 ), ‘ A visit to the diamond market ,’ Gems and Gemology , 3 , 34 – 36 . Visvanath N. ( 1972 ), ‘ Diamond cutting in India ,’ The Diamond News and S. A. Jeweller , 35 , 21. Yeung H. W. C. ( 2007 ), ‘Ethnic entrepreneurship and the internationalization of Chinese capitalism in Asia,’ in Dana L.-P. (ed.) Handbook of Research on Ethnic Minority Entrepreneurship: A Co-Evolutionary View on Resource Management . Edward Elgar : Cheltenham , pp. 757 – 798 . Yeung H. W. C. ( 2009 ), ‘ Transnationalizing entrepreneurship: a critical agenda for economic geography ,’ Progress in Human Geography , 11 , 61 – 85 . Weber M. ( 1958 ), The Religion of India: The Sociology of Hinduism and Buddhism [Translated and edited by Gerth H.H. , Martindale D. from the original 1921 German book]. Free Press : New York . Wolfe D. , Nelles J. ( 2008 ), ‘The role of civic capital and civic associations in cluster policy,’ in Karlsson C. (ed.) Handbook of Research on Innovation and Clusters . Edward Elgar : Cheltenham , pp. 374 – 392 . © The Author 2017. Published by Oxford University Press on behalf of Associazione ICC. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Industrial and Corporate Change – Oxford University Press
Published: Oct 10, 2017
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera