The vast literature on transnationalism has primarily been concerned with people, practices, and social fields. Less attention has been devoted to the enigmatic relationship between transnationalism and place. A place itself cannot be transnational in the sense of operating or extending across national boundaries to places elsewhere. But if we understand place as a ‘meaningful location’ (Agnew, 1987), it is evident that transnational connections can be central to imbuing locations with meaning. In short, the transnational is simultaneously antithetical to place and constitutive of it. So, how can the transnational qualities of a place be empirically examined? This Featured Graphic explores the salient yet elusive transnationalism of a unique neighbourhood in Guangzhou, the third-largest city in China.
Guangzhou is the primary trading hub of the Pearl River Delta, which produces more than a quarter of Chinese exports. Some of these exports are destined for Africa. In fact, while most of the academic and policy interest in China–Africa connections has focused on Chinese demand for African materials, Sub-Saharan Africa has a large and growing trade deficit with China. The African market for Chinese manufactured goods has led thousands of African traders, brokers, logistics agents, and other service providers to come to Guangzhou.
Few places in China have been as marked by the African presence as Guangzhou’s Xiaobei neighbourhood (Li et al., 2009). Transnational connections and ethnic diversity are manifest in the urban landscape in a number of ways. There are directly trade-related shops and service providers and a multitude of trader-oriented businesses, such as hotels, beauty parlours, restaurants, and gift shops. Restaurant names and advertisements for shipping services contain jumbled references at various geographical scales, to Africa, Angola, and Luanda, for instance.
Diversity in Xiaobei defies established notions of immigrants and locals. Among the foreigners, there is a continuity of attachment, from first-time visitors via circulating traders to long-time residents. The latter group includes foreigners who provide services for traders on short-term stays, as well as people who export goods to business partners or customers overseas and only leave China occasionally. Some of these long-term residents have children who are born and raised in China.
After living in Xiaobei for a few months, I had a good sense of the neighbourhood’s transnational connections: they were primarily African, and remarkably diverse. But how could the transnational qualities be measured and mapped beyond impressionistic description? Sample surveys and official immigration statistics might both have their uses, but they are marred by methodological problems in this environment. What I pursue instead are auto-generated data from a pivot of transnational connections: international phone services.
Cheap calls have been described as ‘the social glue of migrant transnationalism’ (Vertovec, 2004) and telephone voice data have been used to examine transnational ties between country dyads (Perkins and Neumayer, 2013). The value of such data is sensitive to technological changes and country-to-country differences, for instance in the prevalence of VoIP traffic. In Xiaobei, the preferred mode for international voice calls is specialized public phone services. The area is dotted with providers of cheap calls, sometimes as dedicated single-function businesses and sometimes in the form of a row of telephones on the counter of a multi-purpose shop or office (Figure 1). Often, the most prominent signage is easily overlooked: lists of numbers such as 00243, 00244, and 00256, followed by numbers such as 1.0 or 0.8. These are dialling codes and prices. Even if other signs announce the service, dialling code price lists are usually also on display.
These displays are interesting data insofar as they reflect demand for calls to specific countries. Several features suggest that they do. First, they always constitute a selection of countries, never the full worldwide list. Second, no two lists are the same. Third, the turnover of businesses (and hence signage) is rapid, and even where businesses remain in place, the lists are often extended or amended by hand. Fourth, many service providers have a prominently displayed short list, plus a longer list for reference.
The lists make it possible to construct a ‘prominence index’ that gives each country in the world a score, a value that reflects the average number of other countries with which the country is listed, and the proportion of providers that have it listed.1
Figure 2 displays Xiaobei’s transnational connections as reflected by the prominence index for telephone services. As expected, ties to Central and West Africa dominate. Connections to India are also conspicuous, and resonate with the noticeable presence of Indians in the neighbourhood. More surprising are the connections to France and the United States, which appear unrelated to the demography of residents or visitors. These connections reflect the demand for Africa-oriented Chinese goods in the African diaspora in Europe and North America. For instance, a large part of the global trade in human hair is sourced and processed in China and destined for the African diaspora. Xiaobei’s connections to Thailand also seem puzzling, since there are few Thais in the neighbourhood. Again, trends in production and trade provide a cue: rising costs in China lead African traders to explore alternative production sites, including Thailand.
The methodology and the results have several implications for the challenge of conceptualizing and measuring transnational dimensions of place. First, the data on telephone services have not only pragmatic advantages, but also particular theoretical appeal as means to gauge the neighbourhood’s transnational qualities: they reflect dynamic social interaction as well as physical manifestations in the urban landscape. These two elements resonate with theoretical concepts of neighbourhood (Kallus and Law-Yone, 2000). Second, the data serve to caution against essentialist assumptions that equate transnational connections with national origins. The ties to France and North America exemplify inter-diasporic transnationalism, which coexists with diaspora–homeland connections (Sperling, 2013; Van Hear, 1998). Third, the data call attentional to the variable legibility of a landscape’s transnational qualities. Most people would not see a sign reading 00244 as a reference to Angola, but Angolans would. In other contexts, the legibility of transnational connections have more to do with interpretation and connotations, as Klaufus (2006) and Lopez (2010) have shown in their analyses of migrant houses and architectural styles.
1 For each provider where country a is listed, it is given a listing value of 100/n, where n is the number of countries on the list. If there are two lists (one large display and a longer list for reference), weights are calculated as 100/n for each list and countries are assigned their highest value if they are listed twice. The sum of listing values for country a is divided by the total number of providers, and the result is multiplied by the proportion of providers that list country a at all. The index thus has a theoretical maximum value of 100. If all providers had three countries on their lists, a country that appeared on half of the lists would have an index value of 100/3 × 0.5 = 16.7.