- Visual Analytics of an Eighteenth-Century Business Network
A social network consists of a finite set or sets of actors and the relation or relations defined upon them. The presence of relational information is a critical and defining feature of a social network.1
Historians have become increasingly interested in networks as an analytical tool for eighteenth-century commerce. In much of the historiography these networks are treated as inherently beneficial for the wider economy and the actors themselves. Recently, however, historians have started to problematize networks and to complicate our understanding of them. Indeed, the quote above stresses that a network is not simply the actors within it, but the relationships between them. Realizing this facilitates an understanding of how such networks function. [End Page 1]
The purpose of this paper is to disseminate the results of an interdisciplinary experiment in which social network analysis methodologies are applied to a historical case study. It visualizes the network and then measures the relationships between its actors while mitigating the source-centricity problem often found in historical network analysis. In this way, historians can better understand the dynamics of a network under analysis.
This paper first outlines the major themes in the related literature on networks in history and places our contribution in context. It then outlines the eighteenth-century case study and methodology for its analysis. This is followed by the quantitative analysis and a discussion of its results. Finally, we make our conclusions.
Historians and Networks
Historians have been using networks as an analytical tool for some time. When historians discuss networks, they usually mean networks of people. However, networks may also describe patterns of distribution, credit, or information. In this paper we discuss people, and in particular, a "pattern of relationships" or the "connectedness between either individuals or organizations."2 As Paul Duguid has recently pointed out, the increasing discussion on networks in history seems to imply that there is something new about them.3 This is clearly not so. Eighteenth-century merchants were well aware of the usefulness and agency of their networks, even if they used the term "association," rather than network.4 These networks, or associations, are often ascribed positive attributes by historians; that is, they are interpreted as being inherently good both for the economy and the actors within it. Quaker networks are the epitome of this approach, even if they are somewhat unrepresentative.5 Their constant monitoring of one another produced a reputation for good business practice, which spread outside their own community. However, while there is no doubt that the Quakers provide us with an example of good business practice, such behavior was not confined to their community. This ascribed or characteristic-based trust can also be seen in [End Page 2] non-Quaker networks.6 Anyone of the same religion or sect, such as Anglican, Jewish, Catholic, or Unitarian, for example, might be considered as more trustworthy by those within that group. Ethnicity and nationality could be another reason for entry into a network, such as the Scots in the Caribbean.7 The most obvious ascribed trust was that between family members. Peter Mathias has highlighted the importance of the "kinship nexus," especially for business start up in a low-institutional environment.8 Indeed, such networks should be able to provide emotional support as well as capital and training.9 They should also work to reduce information and transaction costs and to avoid the costs of vertical integration. We can see this in the eighteenth-century Atlantic in terms of merchants purchasing sugar through their networks of factors in the West Indies, rather than purchasing their own plantations.10
There is therefore a sense that networks are essentially benign, but this is not necessarily true. Linda Renzulli et al. have described how kinship ties can be restrictive because they do not open up new opportunities and contacts.11 Charles Tilly has also shown that networks based on ethnicity can encourage members to be inward looking when other contacts available through weak ties may have provided better opportunities.12 Nor were family members always reliable. Francis Bright, the Jamaica factor for the Bright–Meyler...