The dark side of agglomeration, sustained wealth and transposition of trading institutions—the case of Bordeaux in the 18th and 19th centuries

Abstract

The present paper explores a historical case of a location where a group of merchant families established powerful positions over many generations, involving both the international trade of fine wine and human beings sold as slaves. These families lived in Bordeaux, one of the largest trading centers for trade of enslaved humans in the world in the late 18th and early 19th centuries. These trading families were concentrated in a few quarters of the city, and most of the leading merchants were active in both lines of trade. More importantly, this paper shows that three of the key trading institutions, namely practices defining ‘product quality’, product classes and how prices were set, were remarkably similar in both fields of trade. This, in turn, facilitated a smooth and successful substitution of trade from enslaved humans into fine wine when abolition was imposed in the early 1800s. Overall, a story of the darker side of agglomerations is revealed. This study facilitates a new theoretical understanding of how wealth and prosperity can be secured over long time periods, through transposition of institutions between fields, and driven by forces of proximity.

1. Introduction

The current paper provides a historically informed theoretical narrative (MacLean et al., 2016) over a hitherto less known historical process of transposition of institutions, which enabled a limited number of merchant families in Bordeaux to secure prestige, power and wealth over many generations. In our case, transposition involved a very dark element, namely, the trade of human beings sold as slaves during the 18th century and into the 19th century. By transposing trading institutions in the form of business practices, sometimes codified into documented rules, a smooth and successful substitution of trade was facilitated. The rules in question regulated classes of ‘quality’ of a certain wine or enslaved human based on source of origin, a limited number of fixed ‘quality’ classes and the associated pricing mechanism.

During the early 19th century, at a time when France had lost its most important colony St. Domingue (currently Haiti) and the abolition movement grew in Europe, the traders of Bordeaux, being one of the larger trading depots for the slave trade in the world (Saugera, 1995; Geggus, 2001), faced immense pressure to shift from colonial commerce involving the trade of enslaved humans, into trade of something else, which turned out to be wine from the region. Through this shift these merchant families continued to thrive and accumulate wealth (see Figure 1). In fact, they were so skilled in this that now, centuries later, many of these merchant families that emerged in the 18th century still hold leading positions in fine wine and other trades, and many streets in the city are named after them. A report from the Bordeaux Chamber of Commerce in the late 19th century (Malvezin, 1892b), describing the period of 1814–1830, stated: ‘To recover its ancient splendor, Bordeaux was to seek to fill the immense void created by the loss of St. Domingue by seeking new outlets to enlarge its old European market for wines and eaux-de-vie’ (author’s translation).

Figure 1

Trade of enslaved humans and wine in Bordeaux (values by decades, 1730s–1860s, indexed 1730s = 1). The top part of graph is the trade of enslaved humans and the bottom part is the wine trade.

Notes: Despite great efforts, we have not been able to find complete statistics for the wine trade between 1780 and 1840. Dark gray columns in the wine graph: hypothesized development of the wine trade based on interpolation of accessible data and the fact that the French Revolution led to diminishing trade (O’Rourke, 2006). Gray top columns: Data for the number of expeditions leaving from Bordeaux to trade in enslaved humans, per decade. Black bottom columns: Value of wine exported from Bordeaux. Index set on 1730s value for the decade. For the 18th century, recalculated from livre tournois, via exchange value in grams of gold to euros (2006). Exchange value (based on value between 1720 and 1785): 1 livre tournois=0, 38 g of gold. For the 19th century, the value is recalculated from francs to euros (2006). Exchange values: 1850s: 1 franc=2, 53 euros (2006); 1860s: 1 franc=1,99 euros (2006).

Source: Top column trade of enslaved humans. (The Transatlantic Slave Trade Database, 2008). Bottom column trade of wine (Malvezin (1892a, 1892b).

In order to facilitate such an arduous transition, to secure family prestige, power and wealth, the merchant families of Bordeaux, of which 17 out of the 25 wealthiest families traded in both fields, possessed very strong reasons to ‘reuse’, or transpose, extant trading institutions, which had worked so well to facilitate international trade of ‘high value perishable products’ over more than a century. The Bordeaux standards defining the ‘quality’ of enslaved humans emphasized the source of origin (different regions in western Africa), and four basic ‘quality’ classes were established. The pricing mechanism used by the traders set up a baseline price for the highest ‘quality’, with lower percentages for each lower ‘quality’ class. As merchants put renewed emphasis on wine exports, initiatives were implemented to expand wine production in the region and to develop more codified practices regulating the international wine trade (Markham, 1998). Some traders began to acquire vineyards in the region, and built impressive châteaus on the premises (Joinville, 1914; Cavignac, 1985). In the early 19th century, a stream of quality classification documents for Bordeaux wines emerged, all of which emphasized the source of origin, the ‘terroir’, and a few quality classes of wine (Markham, 1998). These codified practices were remarkably similar to those of the trade of enslaved humans. The well-known classification system from 1855 defined wine quality among the top exporting wines (Markham, 1998; Croidieu, 2011). These rules stipulated five quality classes, rated from Premier Cru to Cinquième Cru, a classification system still in use today.

The empirical part of the paper is organized around seven themes. First, we present the historical context. Then the main actors in the case, the elite merchant families, are identified and their inter-connectedness and political power is discussed. This is followed by two sections describing how the two trades were organized and how the local agglomeration allowed for the development of a buzz in the city. This is followed by a discussion of the three trading institutions in focus of the study. Finally, we analyze why the high resemblance of the trading institutions in the two fields were most likely a result of transposition, enabled by the local buzz, and why alternative explanations are less likely. The section also discusses local buzz in relation to global pipelines.

2. Agglomeration and transposition of institutions

Geography literature on agglomeration offers an understanding of how spatial and social proximity plays a role in the emergence of institutions. Spatial proximity will prompt actors to develop similar attitudes, language and interpretative schemes (Lawson and Lorenz, 1999) and will enhance the establishment of trust among inside actors (Murphy, 2006; Zaheer and Harris, 2006). As Martin (2000) has argued, trust is not easily created, but is built in dense social relationships and networks (Powell, 1990) over long periods of time. Different societies tend to develop different and very specific market institutions and systems, including a range of economic coordination and control systems (Hess, 2004). Frequently, these systems are analyzed at the level of nation states (Whitley, 1992; Nelson, 1993; Lundvall, 2010), but regional and local institutional systems (or ‘regimes’) are equally important (Martin, 2000; Cooke, 2001; Asheim and Isaksen, 2002). In addition, Boschma and Frenken (2006) pointed to the importance of the territorial specificity of institutions.

Inspired by institutional theory (Granovetter, 1985) and social embeddedness (Bathelt et al., 2004; Hess, 2004), geographers have deepened our understanding of the local ‘buzz’ (Storper and Venables, 2004) in cities and clusters, forming norms, rules, customs and conventions (Storper and Walker, 1989). Such local institutions are characterized by inertia and durability (Martin, 2000), and they comprise historical social, cultural (DiMaggio, 1990, 1994; Gertler, 1997), political and economic processes (Dicken and Thrift, 1992). Institutions tend to be localized and ‘sticky’, including both collective tacit, as well as disembodied codified, knowledge (Asheim, 1999).

Another aspect of local buzz is the social dimension of spatial proximity and interaction, which has been referred to as a neighborhood effect (Malmberg and Maskell, 2006), or what Owen-Smith and Powell (2004) referred to as local broadcasting and Grabher (2002) as noise. Buzz consists of knowledge, intended and unanticipated meetings and shared cultural traditions and habits. Actors continuously add to, and benefit from, the dispersion of information, gossip and news by being there (Gertler, 1995). Thus, the emergence of institutions inside of an agglomeration is tightly linked to the local socio-institutional setting (Maskell and Malmberg, 1999), where stable social practices and trust-based interactions tend to emerge (Bathelt and Glückler, 2014). Institutions will also deepen through rapid imitation of business practices within the agglomeration (Marshall, 1920; Glaeser et al., 1992; Aharonson et al., 2007; Fujita and Thisse, 2013). By taking part in the buzz, the actors become involved in processes of local knowledge creation (Malmberg et al., 1996, Storper and Venables, 2004; Storper, 2013) and spillovers in the form of Marshall–Arrow–Romer externalities (Schoenberger and Walker, 2016). Storper and Salais (1997) described how conventions and practices developed in local milieus and enabled economic actors to build institutions, which in turn explained their success.

In order to develop the theory of local buzz and long-term success of actors, this paper is informed by theories of institutional change and institutional transposition. Early institutional theory tended to focus on exogenous factors when explaining change. Other drivers include strive for legitimacy (Parsons, 1960; Meyer and Rowan, 1977) and processes of isomorphism (DiMaggio and Powell, 1983,, 1991). Another line of research explained institutional change as an endogenous evolutionary process (Carroll and Hannan, 1989; Dobbin, 1994; Westphal et al., 1997). More recently, the discussion of institutional change has been expanded, and more intricate reasons behind institutional change have been proposed. Two major streams have emerged, involving institutional entrepreneurship (e.g., Eisenstadt, 1980; Rao, 1998; Garud et al., 2002; Rao et al., 2003) and institutional logics (e.g., Friedland and Alford, 1991; Thornton and Ocasio, 1999; Zilber, 2002; Suddaby and Greenwood, 2005; Lounsbury, 2007).

Between the notion of very stable institutions and institutional innovation and change, a discussion exists around institutional transposition, i.e. institutions new to a field but, in effect, transposed from another field. The concept of transposition originates from Bourdieu’s (1977) and Sewell’s (1992) discussions of transposability. The term ‘transposability’ was employed by Bourdieu (1977) to describe how an agent’s habitus can enable an analogical shift of schemes to solve different, yet similarly shaped, problems in different settings. Building on this, Sewell (1992) argued that the ability to transpose institutions should be understood as possessing knowledge of rules that can be applied in other settings. In the 2000s, the concept, now denoted as transposition, was taken up by organizational institutionalists, including Boxenbaum and Battilana (2005), Hwang and Powell (2005) and Schneiberg (2002, 2013), among others.

Scholars have demonstrated that transposition can lead to a heighted legitimacy for actors carrying out the transposition, as well as an increased legitimacy for the institutions that have been transposed. Boxenbaum and Battilana (2005), Padgett and McLean (2006) and Padgett and Powell (2012) all viewed transposition as a crucial component in larger processes of institutional innovation. Tracey and Phillips (2011) asserted that transposition could leverage the revenue of investments and make it easier to cope with uncertainties for entrepreneurs investing in uncertain institutional contexts. Haydu (2002) described how cultural transposition resulted in a successful class formation that led to increased legitimacy for employers in Cincinnati in the late 19th century. Moreover, Hwang and Powell (2005) pointed to the fact that transposed institutions gained legitimacy. Padgett and McLean (2006) showed that the transposition of the master and apprentice logic into the domain of international trade, in renaissance Florence, resulted in the birth of a new system of organizing business, i.e. the partnership system. This system gained in legitimacy and spread widely into new fields. They also demonstrated that actors carrying through transposition enhanced their social status (Padgett and McLean, 2006).

The literature on transposition describes structural aspects of both the home field (from where the institutions are transposed) and the destination field (to where the institutions are transposed) that would facilitate transposition. Some academics have highlighted how structural forces affecting both fields can lead to transposition. According to Hwang and Powell (2005), an important impulse for transposition is that certain endeavors are blocked or halted in the home field, and that these endeavors then find outlets in other fields. The destination fields for such endeavors are characterized by the fact that they are new and/or open (Hwang and Powell, 2005). Boxenbaum and Battilana (2005) described how the presence of an acute field problem in the destination field, as well as the partial deinstitutionalization of that field, were factors that led to transposition. Padgett and McLean (2006), on the other hand, described how an event (i.e. a political revolution and the consequent reconfiguration of political life) that affected both fields led to the transposition of institutions.

The literature provides different answers to the question of intentionality of actors. Therefore, it must still be considered to be an open question whether or not transposition must be a conscious action, or if it can be understood as a process that can take place without such calculation. In line with Padgett and McLean (2006), we do not take a definite stance on this question. We simply assume that transposition constitutes the result, whether intended or unintended.

Scholars have identified three different field affinities for actors pursuing transposition attempts (consciously or not): (1) actors bringing with them institutions from their home field to a destination field (e.g. Schneiberg, 2002; Padgett and McLean, 2006; Hwang and Powell, 2005; Powell and Sandholtz, 2012); (2) actors embedded in both the home field and the destination field (Boxenbaum and Battilana, 2005) and (3) actors embedded in the destination field that import institutions from other fields (Koleva et al., 2010). In the literature, the first affinity is by far the most common. Examples of this include Schneiberg (2002), describing how cultural carriers entering a new field can bring with them organizational forms that can be mobilized in the new setting, and Padgett and McLean (2006) describing how a certain type of banker, mostly engaged in domestic businesses, moved into the field of international trade. Similarly, Powell and Sandholtz (2012) claimed that what they call ‘trespassers’, or newcomers in a field, bring with them values, expectations and norms from their home field.

Furthermore, the transposition literature discusses embeddedness, incentives and knowledge as key attributes for actors carrying out transposition. In the case described by Boxenbaum and Battilana (2005), the actors that carried through the transposition were embedded in both the home and the destination fields. Motivation is also central to Hwang and Powell (2005). Haydu (2002) pointed to the importance of embeddedness in social networks as an important attribute for the actors carrying out the transposition. Tracey and Phillips (2011) argued that actors must possess deep knowledge of the institution to be transposed, as well as the destination field, to succeed with transposition.

The literature does not provide a coherent understanding of issues of proximity. In the case described by Boxenbaum and Battilana (2005), the distance between the home field and the destination field was relatively close in terms of institutional configuration, albeit the geographical distance was considerable. Schneiberg (2002) pointed out that immigrants originating from distant geographical and institutional fields carried with them institutional templates that they were able to transpose. In another case, Schneiberg (2013) showed how an institutional configuration connected to a specific location failed to be transposed outside of this location. Padgett and McLean (2006) described a case in which considerable institutional distance existed between the home field and the destination field, but where both fields were geographically proximal, which is similar to Powell and Sandholtz (2012) and Hwang and Powell (2005).

In line with Bathelt and Glückler (2014), institutions are understood as collectively recognized durable practices. Such practices are sometimes codified into documents and rules.

3. Research design, method and source materials

In this paper, we strive to establish empirical plausibility for a new theoretical construct, in which an agglomeration facilitates transposition of established institutions, which in turn leads to sustained wealth and prosperity. We follow a conceptionalizing approach to historical organization studies, in line with MacLean et al. (2016).

This paper is written with the aim of putting the concept of dual integrity into practice, meaning a thorough understanding of both a historical method and the theoretical lens of institutions (MacLean et al., 2016). This means that we have used a historical methodology when searching for and dealing with our data, including: (1) source-criticism of the historical remains in terms of realness, closeness and purpose (Lipartito, 2013); (2) empathic understanding of the historical context (Wadhwani and Bucheli, 2014) and (3) a self-reflexive approach to the effort. In terms of finding data, we have relied to a large extent on archival methods (Ventresca and Mohr, 2001).

Through the archival method (Ventresca and Mohr, 2001), we managed to collect the following kinds of data for our research: (1) national business directories, published with permission and approval from the state by private publishing houses; (2) decrees issued by heads of state that announce the founding of private enterprises; (3) stand-alone handwritten historical remains and (4) historical reports describing the long-term evolution of commerce in Bordeaux and the development of the Bordeaux Chamber of Commerce (issued by the same chamber).

Furthermore, we have used a digitalization of a claim report issued by the French state to compensate former French colonizers that lost their property on St. Domingue/Haiti in the Haitian Revolution.¹ These data contain, among other things, the name of the plantation owner, the place of residence of the owner, owners’ successors and the value of the estates.

We have also utilized two historical databases to access compiled historical source materials. The Trans-Atlantic Slave-Trade Database² (hosted by the Emory University Digital Library) contains information of the names of ships, owners and captains of ships, homeports of ships and dates of departure and arrivals at different locations. The database covers over 35,000 slave voyages over a period of more than 350 years. A second database that we used is ABC du Vin³ (curated by the wine author, Sylvain Torchet, 2004) containing data on historical owners of vineyards and châteaus. The main empirical part of this paper builds on a set of merchant families compiled by Cavignac (1985), including the 25 wealthiest families during the early 19th century (1830–1848). To map their addresses we used a business directory published in 1838 (Annuaire général du commerce, de l'industrie, de la magistrature et de l'administration: ou almanach des 500.000 adresses de Paris, des départements et des pays étrangers. See Henrichs et al., 1838).

In order to obtain a deeper understanding of the Bordeaux merchant families, we also collected an enlarged dataset of 87 merchant families (out of a total of approximately 250 families). We built our enlarged dataset from the following three sources: (1) leading wine traders in the Bordeaux section of the Almanac Générale des Marchants des Négocians and Commercans de la France et de le Europé Pour la Année 1772; (2) the 25 richest families in the mid-1800s (Kehrig, 1886) and (3) slave traders listed in the Trans-Atlantic Slave Trade Database—Voyages that outfitted three or more slave voyages. Of these 87 families, we identified 26 families that were involved in both the wine trade and the trade of enslaved humans, 18 specialized in the trade of enslaved humans and colonial goods and 43 specialized in the wine trade. We then compared the 87 families with the names of streets (smaller ‘passages’ to larger ‘cours’) and squares in the city of Bordeaux today, including a total of 2624 street and square names (source: Google Maps).

Data on merchant families, banks and insurance companies in Bordeaux, as well as the names of vineyards and châteaus, have been anonymized.

4. The case

4.1. Historical context

Bordeaux has a very long history of an ‘industrial atmosphere’ (Marshall, 1919) around international trading, where wine trade goes back to medieval times. Wine trade in the region (Graves and St. Emilion) traces back to these times, but was limited in scale. The British nobility had developed an affinity for red ‘Claret’ wine from the region. After British rule, Bordeaux became a part of France, but was under heavy Dutch influence during the 15th–17th centuries. The Dutch preferred lower quality wine from St. Emilion, with its better transportability, called ‘Black wine’ due to its dark color (Markham, 1998).

In the earliest periods (17th century), the trade of enslaved humans was limited in size and was organized through state-controlled monopolies, with the exclusive right to trade enslaved humans. In 1716, trade became free but restricted to a few port cities. Nantes, Bordeaux, Rouen and La Rochelle were home to the majority of French trade of enslaved humans (Stein, 1988). This partial deregulation of the trade in enslaved humans, and the growing plantation economy in the Caribbean, led to a significant increase in colonial trade. The opportunity to take part in this very profitable trade (Daudin, 2004) of enslaved humans attracted immigrating families, mostly from Ireland, England, The Netherlands, Spain and Portugal, to settle in Bordeaux as merchants. Most of these merchants established themselves in the 1730s or later, and settled in the neighborhoods of St. Pierre and Chartrons (Cavignac, 1985), forming a basis for local buzz (Storper and Walker, 1989; Storper and Venables, 2004) to emerge.

After Nantes, Bordeaux became the second most important port for the trade of enslaved humans in France during the 18th century (Clark, 1981), and the most important port in terms of colonial produce (Weber, 2010). As the bordelaise merchants mainly traded with St. Domingue, the boom in the plantation economy on that island led to a concurrent boom in the trade of enslaved humans among the Bordeaux merchants. During the late 1780s, Bordeaux became the greatest center for the trading of enslaved humans in France (Eltis, 1987; Geggus, 2001). In total, there are records of 443 ‘successful voyages’ outfitted by merchants based in Bordeaux (The Trans-Atlantic Slave Trade Database—Voyages). Following the political and economic turmoil of the French Revolution and, importantly, the Haitian Revolution in St. Domingue, colonial trade was significantly restricted (O’Rourke, 2006). This was especially the case for the trade of enslaved humans, which came to be officially abolished by the National Convention in 1794, but was re-legalized soon afterward (Saugera, 1995).

Napoleon re-legalized the trade of enslaved humans in 1805, but the subsequent trade did not come close to that of the 1770s and 1780s. The British abolition in 1807 also made trade much more difficult since the British would search for and intercept slave ships. It was now apparent to the bordelaise merchants that they had to substitute the trade of enslaved humans with another commodity, since that trade would soon cease (Hwang and Powell, 2005). The merchants would now have strong incentives to secure their positions in the new trade and control its structure (Fligstein, 1996).

As it turned out, the main substitute was wine. Now, the leading merchant families invested heavily into wine-making, purchasing old vineyards, draining sumps and planting new vines. Furthermore, based on accumulated wealth from the trade of enslaved humans, they built châteaus in medieval style, and made production and sales of wine more effective and at a larger scale (Cavignac, 1985; Saugera, 1995; Markham, 1998; Joinville, 1914). Most of the old large estate owners had their properties nationalized during the Revolution, and now, in the post-Revolution era, these vineyards and châteaus were up for sale, making it easy for wealthy merchants to enter this business. In other words, this destination field was open (Hwang and Powell, 2005) for new actors. The merchants also formed banks and insurance companies to organize and protect their trade. This process continued during the reign of Louis-Phillipe (1830–1848), a time under which a small elite of wealthy merchants from approximately 25 families emerged (Cavignac, 1985). A large majority of these 25 families had a background in the trade of enslaved humans (Table 1).

Table 1

The 25 leading merchant families in Bordeaux (first half of the 19th century) active in the wine trade and wine-making, trade of enslaved humans and plantations and banking and insurance organizations

	Wine trade		Colonial trade		Financial services
Family no.	Active in wine tradea	Ownership of domain(s) or château(s)b	Organization of slave trade journeysc	Ownership of colonial plantationsd	Co-founding of bankse	Co-founding of mutual insurance companiesf
1	1	1	1	1	1	0
2	1	1	1	1	0	0
3	1	1	1	1	0	0
4	1	0	0	0	1	0
5	1	1	1	1	1	1
6	1	1	0	0	1	0
7	1	0	1	0	0	1
8	1	0	0	0	0	0
9	1	1	1	0	1	1
10	1	1	0	0	0	0
11	1	0	1	1	1	1
12	1	1	1	0	1	1
13	1	1	1	0	0	0
14	1	1	1	1	1	0
15	1	0	1	0	1	1
16	1	1	1	0	1	0
17	1	1	1	0	0	0
18	1	1	1	1	1	1
19	1	1	1	0	0	0
20	1	1	0	1	0	1
21	1	0	1	1	1	1
22	1	1	1	0	1	1
23	1	1	0	0	1	0
24	1	1	0	0	0	0
25	1	0	0	0	0	0

a

Source: Cavignac (1985), Markham (1998), ABC du Vin Database (Torchet, 2004).

b

ABC du Vin Database (Torchet, 2004), Markham (1998), Cavignac (1985).

c

The Trans-Atlantic Slave Trade Database—Voyages, Saugera (1995).

d

Plantation owners of Saint Domingue (1789).

e

Founding member of the Banque de Bordeaux. Ordonnance du Roi (1818).

f

Founding member of one the following insurance companies: Compagnie d’assurance de Bordeaux (1818), Compagnie d’assurance maritimes de Bordeaux (1820), Compagnie d’assurances contre l'incendie (1820), Ordonnance du Roi (1819), Ordonnance du Roi (1820), Ordonnance du Roi (1820). 0 = not active, 1 = active.

4.2. Connected merchant families

The primary organizational form for the Bordeaux merchants, and French traders in general, was the family governing trade through kinship ties. The family constituted the natural base for business, sometimes complemented with partnerships with other families of a temporary or more stable nature. The bordelaise merchants preferred to do business with a close circle of other bordelaise merchants, and rarely invested jointly with merchants from other French towns (Butel, 1974). This would favor trust-based interactions (Bathelt and Glückler, 2014). In the enterprise, the head of the family played the leading role, whereas complementary positions on higher levels were assumed by the sons (Clark, 1981; Cavignac, 1985). Bordeaux hosted approximately 250 merchant families around the time of this case. Cavignac (1985) separates the merchants of Bordeaux during Louis Philippe (1830–1848) into three categories: (1) the modest merchants, constituting approximately 70% of the total merchants who traded mostly locally; (2) the middle merchants, constituting 20% who traded mostly regionally and (3) the elite merchants, trading internationally, constituting approximately 10% of the total number of merchants. They paid more than 2000 francs in taxes and possessed the largest fortunes (Cavignac, 1985). The group of elite merchant families, on which we focus our attention, was tightly networked and connected to a high degree through intermarriages, facilitating collaboration between the families (Philips et al., 2000; Lawrence et al., 2002). All of the 25 families were intermarried with at least one of the other 24 families. Intermarriages with three or more other families were not uncommon, and the family with the most marriage connections was intermarried with 11 of the other families (Cavignac, 1985). As the literature points out, building such ties and networks would facilitate the transposition process (Schneiberg, 2013).

In Table 1, we present data for the identified 25 wealthiest merchant families (category 3), who enjoyed the most prestigious and powerful positions in Bordeaux over many generations. All families were involved in the wine trade and 18 families had invested in wine-making. Seventeen families were also active in the trade of enslaved humans and, of these, nine had also invested in their own plantations in the West Indies. Fourteen families had been involved in the establishment of the leading merchant bank and 10 families had been involved in setting-up one or more of the three existing mutual insurance companies in order to secure their international trading operations.

4.3. Political power

Daudin (2004) pointed out that French slave traders did not suffer any social stigma in their homeports. Indeed, quite the opposite, the elite merchants, of whom almost all were involved in the trade of enslaved humans, held powerful political positions. An indication of this was that they acted as chairmen of the most important business organization, the Bordeaux Chamber of Commerce, already founded in 1705. Here, economic and political affairs were discussed and decided (Stein, 1988). The Chamber of Commerce was an important place for forming of both tacit and codified knowledge (Asheim, 1999). For example, the 1855 Committee (see discussion below) was commissioned by the Chamber (Markham 1998). During the period in which transposition took place, the 25 families held 15 out of 28 Chairman periods (1803–1860, marked with gray boxes in Figure 2). This is also the period in which codification of quality and pricing practices in the wine trade intensified (marked with small flags in Figure 2). Many of the elite merchant families also, at some point, held political positions in the regional Parliament of Bordeaux, in the district of Gironde and as Major of the City of Bordeaux. Banking and insurance in Bordeaux was also organized collectively among the leading merchants (Cavignac, 1985).

Figure 2

Key events in Bordeaux 1740–1860: Classification schemes and chairmen periods among the 25 families, political events and regimes in France.

Source: Markham (1998) and Malvezin (1892a, 1892b).

Another important form or organization involved freemasonry. Immigrant merchants from Ireland had introduced freemasonry in Bordeaux in the 18th century and, at the time of the French Revolution, there were as many as 15 lodges in Bordeaux. Interestingly, Bordeaux acted as the major conduit for the spread of freemasonry lodges into the Americas. These lodges counted many of the merchant families as members (Loiselle, 2009). Lodge membership naturally enhanced trust between the families (Murphy, 2006; Zaheer and Harris, 2006).

All of these circumstances point to the fact that the leading merchant families held substantial power in Bordeaux over generations. This implied that they had a dominant position and, therefore, a major impact on how institutions were formed and shaped (Bourdieu, 1977; Padgett and Ansell, 1993). The most powerful and prestigious families, which had traded in enslaved humans, became increasingly involved in the wine trade in the early 19th century. By successfully shifting trades, as we will discuss more thoroughly below, they were able to augment their prestige and legitimacy (Haydu, 2002; Padgett and McLean, 2006). To test this, we created an enlarged dataset of 87 merchant families consisting of: 26 families that were involved in both the wine trade and the trade of enslaved humans; 18 specialized in the trade of colonial goods, including the trade of enslaved humans and 43 families specialized in wine trade. We then compared these families with all names of streets and squares in Bordeaux today. It was determined that slave-trading families have a 6% probability of having a street named after them today, wine traders a 12% chance and combined wine- and slave-trading families a 69% probability. This may appear surprising, considering that over two centuries have passed since the peak of the trade of enslaved humans. As Oto-Peralías (2018) has shown, street names can be viewed as proxies for social and cultural characteristics of a city. The results of this test reveal, remarkably, how the prestige of the elite merchant families could be perpetuated.

4.4. Organization of the Bordeaux trade

Trading practices in the wine market in Bordeaux involved institutions that have survived more or less intact for centuries (Markham, 1998; Croidieu and Monin, 2010). One of the most discussed aspects of the Bordeaux wine market system is the specific value chain of three major steps (Figure 3). The first step is vine growing in the vineyards, a position that historically was occupied by both small self-owning farmers and estates (châteaus). The micro territory where the vine is grown, the ‘terroir’, has for centuries been considered to be of great importance for the quality of wine, and thus also a deciding factor for pricing of different qualities. The second step in the value chain is held by the courtiers, the local wholesalers/brokers that act as intermediaries and buy the produce from the owner of the domain. The courtiers are the wine experts in the region. They traditionally transported the produce to the city of Bordeaux where it was sold to the Négociants (referred to as ‘la Place de Bordeaux’), who held the third position in the value chain. The Négociant blended the wine and sold it under his or her own name in international markets. The Négociant, as a profession, dates back to the Middle Ages in Bordeaux (Markham, 1998; Brook and Latham, 2001).

Figure 3

Overview of colonial trade (including humans sold as slaves) and the wine trade.

In the colonial trade, the merchant family was referred to as ‘Armateur Colonial’ (Figure 3), and here the families organized international trade in the form of a triangle. The first leg involved the trade of barter goods to destinations in Africa, where they would be exchanged for enslaved humans. The enslaved humans would then, under truly horrific conditions, be transported across the Atlantic Ocean on the same ships to be subsequently sold to work on plantations in one of the French colonies. After this, the ships returned home to Bordeaux with colonial produce. Sugar was the overwhelmingly dominant produce from the French colonies, especially in the late 18th century. However, other forms of produce, such as indigo, tobacco and cotton, were also traded (Saugera, 1995). Certainly, society at the time considered the ‘business’ of buying and selling enslaved human beings, and related logistics and financing, as any other business. Thus, in order to illustrate similarities between this dreadful trade and the prestigious international wine trade, we have chosen to illustrate the two lines of trade side by side. It is worth pointing out that both lines of trade involved commerce over long distances, of what was considered very ‘perishable products’, from the perspective of traders at the time. This would naturally lead to a need for robust institutions regulating the trade in terms of quality and prices.

We now provide a more detailed description of the gruesome trade of enslaved human beings. The first step was the outfitting of a ship. Either a ship was built for the occasion or an old ship was reused. The ships used for trade from Bordeaux during this time were divided into three different categories: (1) Vaisseaux de long cours for Africa, Trans-Atlantic or the Pacific; (2) Le grand cabotage for the British islands, Holland, the Baltic, Scandinavia or the Mediterranean and (3) Le petit cabotage for the French mainland ports, except in the Mediterranean (Clark, 1981). The ships used for slave voyages belonged to the category of Vaisseaux de long cours, and the wine ships typically belonged to the category Le grand cabotage. However, it should be noted that since these two classes of ships had the same basic design, it was not uncommon that a ship that was assigned to the first category was reclassified into the second category and that a ship in the second category was reclassified into the first category (Clark, 1981). Thus, ships that had been equipped for slave voyages (which included large amounts of barrels for water) could be reequipped for wine transport (loaded as barrels), and vice versa. Outfitting also involved assembling of barter goods suitable for buying enslaved humans in ports on the African coast. The barter goods consisted of different kinds of fabric, iron, armaments, cowry shells (used as currency locally and typically brought from the Maldives), etc. (Richardson, 1991). Most of the barter goods were not on hand locally in Bordeaux, but had to be purchased from different locations in France, throughout Europe (such as iron bars from Sweden), and even further away (Clark, 1981; Stein, 1988).

The major destinations for French vessels included Whydah on the Bight of Benin, Malembo in West Central Africa and Cabinda and Loango, both ports located on the Loango coast. When landing in a port, the captain had to first secure permission of the local ruler to trade. After that, a negotiation began, usually involving bribes and gifts, and the captain unloaded his goods into a building in the port. African merchants brought enslaved humans to the port, where African brokers with exclusive rights resold them to the European captains. After being sold, the enslaved humans were branded on their shoulder or thigh, and taken aboard the ship (Geggus, 2001) and the vessel set sail to the French colonies in the West Indies. The most important destination by far was St. Domingue, accounting for more than 75% of French voyages. Other destinations included Martinique, Guadeloupe, Guyane and Grenada (Geggus, 2001).

When arriving at the colonial port, the captain established contact with one of the local colonial agents that facilitated the trade. In later periods, the bordelaise Armatures placed people that they could trust, preferable family members, in this position. The agent negotiated with plantation owners looking to buy enslaved humans. The enslaved human beings were typically sold in groups (from one or several regions), but could also be sold individually (Stein, 1988; Klein, 2010). The plantation owners generally paid for the slaves in colonial produce, and only in very rare cases did they pay in cash. Since enslaved humans were considered very expensive ‘goods’, it was not possible that the ship could bring back the whole value of the sold humans in colonial produce. Thus, ships were also sent directly from France to the colonies to collect produce. As a result, the plantation owners tended to become increasingly indebted to the Bordeaux merchants over time (Daudin, 2004). When the ship returned to Bordeaux, the colonial goods were either resold, or refined and then sold, at home or abroad (Stein, 1988).

Traditionally, traders were active in only one stage of the value chain in both fields. Now, in the 1770s and 1780s the elite merchants began expansion into the business of acting as colonial agents, managing plantations and investing into local industry refining colonial input in Bordeaux. Later, in the early 19th century, the same logic appeared in the wine value chain, when traders began to invest in vineyards and wine-making (Joinville, 1914; Cavignac, 1985).

With trade of enslaved humans coming to a halt, and leading families shifting over to wine, the necessary foundations for institutional transposition were at hand (Hwang and Powell, 2005), underlined by the agglomerated setting (Padgett and McLean, 2006; Powell and Sandholtz, 2012). In the next section we will expand on the prerequisites for collective practices to emerge in the local buzz (Storper and Venables, 2004).

4.5. Local buzz

In order to understand the depth of interaction over generations that would lead to collective durable business practices we take a tour of a part of Bordeaux of approximately one square kilometer. Here the merchant families lived, worked and socialized, forming an arena for a neighborhood effect to emerge (Malmberg and Maskell, 2006), and for quickly spread news and gossip (Gertler, 1995). Let us move into this district and we will start at Place de la Bourse. Already in 1705, local merchants had set up a Chamber of Commerce, one of the first in France. Moreover, during the period of 1730–1755, a majestic building complex was erected at 17 Place de la Bourse, and remains a centerpiece of the city to this day. Various carved symbols of trade enhanced the façade, including faces of black people, showing the significance of colonial trade. This building offered space for offices and meeting rooms, where trade issues were formally and informally discussed, and disputes were resolved.

Around the corner from the Place de la Bourse and onto Chapeau Rogue, a neighborhood exists with beautiful 18th and 19th century merchant homes. Within a few blocks, over half of the leading 25 merchant families (Table 1) lived during this period. Here, we also find the few restaurants of the time. The Grand Théâtre de Bordeaux is also here, built in 1780 by one of the masonic lodges. One could imagine how the merchants fraternized and discussed business in more relaxed terms at the restaurant tables, during freemasonry meetings and at dinner parties at home or at masquerades at the Grand Theater (Figure 4).

Figure 4

Neighborhoods of buzz.

Notes: Black dots represent addresses for the 25 families in 1838. Note that some families had more than one address. Black stars represent addresses of restaurants in 1838.

Source: Henrichs et al. (1838). Map data: Google Maps.

We now turn back toward the Garonne River, and pass the Place de Quinconces. The Place de Quinconces was laid out in the early 1820s (replacing an ancient fortress), was one of the largest squares in Europe and was a center for trade fairs, festivals and other large events in the city. A block away are the beautiful gardens, the ‘Jardin Public’, offering several hectares of grass, trees and a botanical garden. Here, families have strolled and played since 1746. Continuing toward the river, one reaches the quays in Chartrons, with its historical storage facilities, where sailing ships were once lined up.

Let us explore family number 7 (Table 1), a good example of a family involved in both trade of enslaved humans and wine and a family that was inter-married into other local merchant families, and with leading positions over generations in the Chamber of Commerce and other organizations. In 1730, Pierre (first) founded a trading company bearing his name. He had four children, two of whom married into family number 18, and the two families became tightly allied over generations and several of the descendants lived on Chapeau Rogue (Henrichs et al., 1838). Another son married into family number 3, one of the largest traders of enslaved humans. Two grandsons to Pierre took over the family business in 1770 (Cavignac, 1985). They outfitted at least six slave journeys. During this period, Pierre (third generation) established himself as a plantation owner on St. Domingue. His son, Pierre (fourth generation), in turn, took over the business in 1813 and 5 years later he co-founded the Compagnie d’Assurance de Bourdeaux. This insurance company was owned by 38 families, 10 of which belonged to the leading 25. During this period, family number 7, like many others, shifted their focus to the wine trade. Pierre (fourth) became an important actor in the reconstruction of the Bordeaux trade. He acted as Chairman of the Tribunal de Commerce, where disputes between merchants were settled, and he held the most prestigious position of them all, namely, Chairman of the Chamber of Commerce (for four periods during the 1820s and 1830s, see Cavignac, 1985). The head of family 18 at this time, a close ally, was considered the leading figure in the group of powerful merchants, organizing the reconstruction of the Bordeaux trade (Malvezin, 1892). The son of Pierre (fourth) served as Chairman of the Chamber during the 1860s, a period just after the 1855 classification had been commissioned and enacted.

What we have seen is a limited number of tightly knit merchant families who lived, worked and socialized in this proximate setting over many generations, forming the perfect arena for developing and reusing, i.e. transposing, well-functioning business practices across fields of trade.

4.6. Trading institutions

In order to meet the difficulties and high risks associated with international trade in the 18th century, institutions were formed to make the system governable and efficient. We will focus on three long-lived practices: (1) definition of quality as inherent in origin; (2) ranking of quality in fixed classes and (3) pricing based on fixed price ratios, including the use of a lead price. These institutions existed and functioned in both the gruesome trade of human beings and in the prestigious wine trade. Quality definitions and pricing were interconnected and shared the same purposes, i.e. to facilitate the exchange to the buying party, to make each step of the trade as transparent as possible (for trading parties and other investors) and to secure (and maximize) revenues. We begin with the practice that defined quality, bearing in mind that the comparison between humans beings sold as slaves and the wine trade is historically situated, i.e. the way that merchants at the time viewed it.

In both trades, quality was tightly connected to the place of origin. Humans originating from West Central Africa were defined as higher ‘quality’, whereas enslaved humans from Sierra Leone, Biafra and the Windward Coast were considered to be of lower ‘quality’. In practice, this meant that captains would always try to land in a ‘high-quality’ region. In many cases, however, this was difficult due to France’s weak position (compared with Britain) in these regions. Captains that did not manage to land in ‘high-quality’ regions would proceed to ‘lower quality’ regions. These locations were meticulously documented and, upon return in Bordeaux, presented to the merchants and others that had invested in the journey (Geggus, 2001).

Classification of wine quality has a history as long as the wine trade itself. The most elemental form of classification that emerged in the Middle Ages was classification according to larger regions: Bordeaux wines, Loire wines, Champagne, etc. Over time, a number of regional and sub-regional labels (down to terroir) emerged. These practices had in common that they defined which region, sub-region and domain produced the better-quality wines, and which produced the lesser-quality wines (Markham, 1998). The idea of a microclimate (soil, precipitation, etc.), connected with a particular small territory, was fundamental to the definition of quality (Malter, 2014). Thus, in both the trade of enslaved humans and the wine trade originating in Bordeaux, quality was understood as tied to a geographical origin. Contrary to our understanding of quality today based on numerous characteristics, what we have just described is an institution in which origin in fact is quality. In the case of humans being sold as slaves, this implied that people originating from West Africa were always and in themselves considered to be of ‘higher quality’ than people originating from, for example, the Windward Coast. In terms of wine, this meant that the wine produced from grapes from terroir X was always classified as being of better quality than wine produced from grapes from terroir Y.

The second institution that governed both lines of trade was the ranking of ‘quality’ in four or five classes. In the case of the trade of enslaved humans, this institution was used when the colonial agent sold the enslaved people in the colonies. As mentioned above, the first ‘quality measure’ would be the place of origin in Africa, but within a group of people from the same place, a class system defined ‘quality’ based on gender and age with four quality classes (Williamson and Cain, 2011), as shown in Table 2. As the merchant families shifted their attention over to wine, the business practice of defining products into different quality classes was formalized into documented rules for wines from the Médoc region. Several such documents, with four to five quality classes, were presented in the early 1800s (Figure 2) to enforce an institution that was already in place.

Table 2

Quality classes and standard price ratios for red wine compared with enslaved humans

Quality class	Red wine	Enslaved human
	Price level	Price level
1	First class	Man
	100%	100%
2	Second class
	75%
3	Third class	Woman
	58–62%	50–70%
4	Fourth class	Boy
	45–50%	40–60%
5	Fifth class	Girl
	33%	20–40%

Source: Red wine based on data from Markham (1998). Enslaved human column based on Saugera (1995) and Williamson and Cain (2011).

This was also the case for the well-known 1855 classification that divided the finest wines of the region into five different classes (Markham, 1998).

The third and final of the three institutions on which we have focused was the lead price institution. Highly connected with the two other institutions, this institution implied that a lead price was negotiated for the ‘first class’, and that prices for lower qualities would follow from this (without the need for negotiation) according to pre-defined ratios. In the trade of enslaved humans, this institution governed both the buying of the enslaved humans in Africa and the selling of the humans in the colonies. When buying enslaved human beings in Africa, a pricing practice was used called ‘pièce d’Inde’. This pricing practice received its name from the fact that a healthy man (the ‘first class’) should be exchanged for a piece of fine Indian cotton fabric of the length of 4 m. The other ‘classes’ would then follow with specific price ratios. As trade grew, other products were increasingly used in the barter trade, and cotton lost its special position. The system did however persist, and captains of slave ships would negotiate a lead price for a healthy man and then the other prices would follow in relation to that price, when buying humans from local African merchants (Geggus, 2001). After the Trans-Atlantic voyage, when the humans were sold in the colonies, they were priced according to certain price ratios defined by four classes (Williamson and Cain, 2011). In the wine trade in Bordeaux, the principle of the lead price took (and to some extent still takes) the following form. After the vine harvest each year, the courtiers negotiated with the châteaus and domains on a lead price for the first growths (first class); this price then determined the prices of the wines of the lower classes. If it was a good year’s harvest, the prices would be higher than those of a bad year, but the ratios between the different classes remained the same (Markham, 1998). Table 2 presents the quality classes and price levels for red wine next to those of enslaved humans and striking similarities are evident.

The three institutions were all tightly connected. As the trade of enslaved humans was coming to an end, work with defining wine qualities intensified (Figure 2).

4.7. Local institutional transposition

Our case has shown that three interrelated business practices emerged in one location and turned out to be remarkable similar in two highly separate fields of trade. This fact leads us to believe that we have found a case illustrating how old institutions were reused, and perpetuated through a process of institutional transposition, between the horrendous trade of human beings and the highly prestigious wine trade. Furthermore, we argue that by transposing these institutions the leading merchant families sustained their prosperity over many generations (Haydu, 2002; Hwang and Powell, 2005; Padgett and McLean, 2006). Let us advance this argument further.

The elite merchant families of Bordeaux were concentrated in a small part of the city and were highly interconnected, through for example business partnerships, inter-marriages and freemasonry lodges (Butel, 1974). In line with the theory presented above this would point to the importance of a local buzz, where local collective business practices could evolve over generations within the tight social setting (Boschma and Frenken, 2006).

During the reign of Louis Phillipe (1830–1848) almost all in the group of elite merchants (Table 1) who had traded in enslaved humans were now building a presence in wine trade and wine making; they directed their networks (Haydu, 2002), organization skills (Schneiberg, 2013), investments and knowledge (Tracey and Phillips, 2011) to the new field. Theory also supports the notion that, as we saw in our case, transposition of institutions is likely when trade in one field of business is coming to a halt, and established actors are looking for new opportunities (Hwang and Powell, 2005). Also, transposing institutions made it easier for them to deal with the high uncertainty associated with the new line of trade (Tracey and Phillips, 2011), thus making the collective merchant system and its practices and documented rules more stable.

The revolution had led to a state in which the wine trade was partially deinstitutionalized (Boxenbaum and Battilana, 2005) and open (Hwang and Powell, 2005), when the elite merchants (Padgett and McLean, 2006) decided to shift into that line of trade. We suggest that the starting point of the transposition process was around the turn of the century and in the early 1810s. At this time, the turmoil of the revolution had calmed significantly. The large wine estates that had been nationalized were now up for sale, and the important Chamber of Commerce was reestablished. The trade of enslaved humans, although relegalized under the first empire, was far from its apex in the 1780s, not the least because of the Haitian revolution, in which France lost its most important plantation colony. Britain had, at this time, outlawed the slave trade, and British vessels would patrol the seas to intercept suspected slaving ships. It must have been obvious for the collective (Storper and Venables, 2004; Storper, 2013) of Bordeaux merchants that a total abolition was also imminent in France, and legislation was in fact enacted just a couple of years later. Some individuals and families had, of course, a more prominent role, and during the process of the large turn-around from the trade of enslaved humans to the wine trade, the Chamber of Commerce referred to one family member (from family 18) as ‘the instigator of all these remarkable men’ (Malvezin, 1892b).

The paper also shows that the merchant families possessed substantial power and prestige, for example by acting as leaders in the powerful local Chamber of Commerce (Figure 2), and thus we can expect that they played a key role in setting business practices over time (Bourdieu, 1977; Padgett and Ansell, 1993). As an example, the merchants in the Chamber of Commerce commissioned the work leading to the important 1855 classification rules of fine wine. In summary, given the theoretical support and empirical evidence presented here, one could expect that as the merchant families shifted lines of trade, they would bring with them institutions (Schneiberg, 2002) that had served them well and earned their fortunes (cf. Padgett and McLean, 2006).

To challenge the reasoning above one could argue that these institutions would be more general in nature and used across many areas of trade or many locations, and that global pipelines would play a more important role than the local buzz. Some traits of business practices, such as the importance of terroir, can be found within the French cultural sphere, whereas other institutional aspects are tied to the location (Boschma and Frenken, 2006), in this case Bordeaux. As North (1991) pointed out, long distance economic activity poses problems of agency and contract enforcement, which are handled through various institutions, such as defined measures, weights and accounting measures. Another associated line of reasoning would be to suggest that the same institutions governed all goods traded in Bordeaux. However, this was not the case. On the contrary, when comparing, for example, the sugar business, we can conclude that the three particular business practices do not appear. In the sugar trade quality and pricing were based on the degree of refinement and the notion of terroir was absent (Stein, 1988). Further, our main point remains supported even if these institutions in Bordeaux were not exclusive for the trade of human beings and the wine trade, but also used in other trades. If that were the case institutional transposition would be a more central part of local buzz than the literature suggests.

Institutions are created within different regional and national cultural and legal contexts, and can travel across space. Global pipelines would play a more critical role if the Bordeaux merchants had ‘borrowed’ their institutions from afar, for example from Liverpool, the leading ‘star’ in the trade of enslaved humans. Let us take a look at the business practices that emerged in Liverpool. Liverpool was the largest homeport of slavers by far, sometimes referred to as the ‘slave capital of the world’ (Ingram and Silverman, 2016). Thus, one could expect Liverpool to be a leader in shaping trading institutions, to be copied by other ports. But Liverpool developed very different business practices compared with Bordeaux.

In stark contrast to the Bordeaux case, the English traders developed depersonalized market-oriented practices, and created formal joint stock companies. Another distinguishing feature of the business practices of the two trading systems involved pricing, where the bordelaise traders developed the system described above (stable ‘quality classes’ and price ratios) and the British relied on auctions and stock exchanges (Pearson and Richardson, 2008). The British system would emphasize the quality of each ‘product’, where prices ‘per unit’ were set at auctions and could vary substantially. The French, on the other hand, would rely on rules based on ‘terroir’, fixed and very stable quality classes and prices set as percentages of the top class (which would vary with demand and supply); institutions with deep roots in Bordeaux. While moving into the trade of enslaved humans in Liverpool was a socially dramatic move (Ingram and Silverman, 2016), and considered a ‘dirty business’ in the UK in the early 18th century (Brown, 2006), the slavers in Bordeaux constituted the elite with the highest political, social and economic status.

The case highlights the local buzz, but global pipelines played a complementary role. Many families, and particularly French, English, Irish and Dutch trading families, had moved in and out of different port cities in Europe and the new world in North America, a process driven by religious persecution and the search for new business opportunities (Bosher, 1995). In this effort, they established intricate international family networks.

In accordance with the ‘buzz-pipelines’ line of argument, emphasizing the importance of both local and global processes (Hägerstrand, 1968; Isaksen, 2003; Bathelt et al., 2004), this paper shows that global pipelines (Amin and Thrift, 2002) in the form of trading networks, markets and trusting relationships (with families spread around trading centers and colonies), were critical to the success of the Bordeaux merchant families. By the same token, however, local buzz in the neighborhood presented above, where the families lived and worked, was critical in shaping business practices affiliated with the Bordeaux case. The global pipelines were attached to market exchanges (as trade normally takes place outside of the cluster, see Malecki and Oinas 1999), while the buzz shaped the economic institutions, which, in turn, regulated those global market exchanges (cf. Polanyi’s notion of different exchanges, Polanyi, 1944).

So far we have only discussed geographical proximity between the merchant families. Geographers have also pointed to other forms of proximity, such as cognitive, organizational, social and institutional proximity (Boschma, 2005). The case identifies several proximity overlaps, where the geographical proximity stimulated social proximity, i.e. embedded, trust-based interactions between merchant families. The study also showed a case of organizational proximity, as trade was organized through similar family-based trade companies. We assert that social proximity helped to mitigate 18th century institutional weaknesses in the international trading system (definition of quality, setting of prices, delivery times, insurance, payments, etc.), where there was almost a total void of a reliable legal system (Boschma, 2005). Social proximity involved families at home, as well as expatriated family members in the colonies, who would reinforce the use of the Bordeaux trading practices; the local buzz was indeed an integral part of the global pipelines.

5. Discussion

By combining theories of agglomeration and institutions, this paper offers a novel understanding of how agglomerations facilitate institutional transposition. Such transposition of key institutions, in turn, creates the foundation for sustained prosperity among established actors, in our case wealth and power among a set of merchant families over many generations, and for the city of Bordeaux overall.

Awareness of the role of proximity in forming and facilitating processes of the transposition of institutions can assist us to better understand the root causes of institutional evolution and perpetuated prosperity. Cities, clusters and other agglomerations constitute a natural setting in which control over business and trade can be secured over long time periods, and when rules and practices in one field are reused, i.e. transposed, into other fields. In our case, the field of business changed over time, while actors and institutions remained.

The case presented above also involves a gruesome and immoral story, with merchant families shifting between atrocious and prestigious trade. Indeed, a very dark side of the perpetuation of wealth within agglomerations exists. In spite of the French merchant families building fortunes by sending enormous amounts of enslaved humans to the French colonies, very little of this history has been acknowledged. Only very recently has it surfaced, including a first memorial of abolition, which was set up in Nantes in 2012.⁴

As previously stated, we have found no literature connecting the two fields of trade in Bordeaux. Rather than avoiding research areas that could be considered sensitive, however, we call upon contemporary scholars in social science to confront and investigate precisely these dark sides. As this paper has shown, by investigating this darker side of agglomerations, which tends to be ‘forgotten’, a new theoretical understanding of perpetuated wealth and power can be unveiled. To purposefully avoid gruesome aspects when studying society is to limit not only oneself, but also the scholarly community and society at large. We mean that it is time for contemporary social scientists, including geographers, to acknowledge this and unpack the many untold dark stories that reside within tightly knit agglomerations.

Finally, we uphold that the eternal question of what makes certain actors or certain places survive and prosper over longer time periods is an important one. To answer that question, this paper has pointed to a need for more comparative studies of institutions and institutional change within agglomerations, such as cities and clusters. This also applies to comparisons across regions, as well as across nations. Future work must determine precisely how power and wealth among actors are sustained or lost, and under what specific conditions do agglomerations continue to prosper or fail. In answering these questions, deeply abhorrent facets may be revealed, and it is critical that they also be included so that a total understanding may be achieved.

Footnotes

1 Plantation owner of Saint Domingue 1789. Available online at: http://www.domingino.de/stdomin/index_colons_a_z_engl.html.

2 Available online at: http://slavevoyages.org/voyage/.

3 Available online at: http://www.abcduvin.com/.

4 https://www.nytimes.com/2018/10/30/world/europe/france-nantes-slave-trade-museum-memorial.html.

Acknowledgments

An early version of this paper was presented at a seminar at Uppsala University and at the International Markets Studies Workshop, IMSW 4, in St. Andrews in 2016. We in particular would like to thank Stefan Arora-Jonsson at Uppsala University for guidance early in the process. We would also extend a great thank you to the referees and the Editor-in-Chief of this journal Harald Bathelt for most inspiring and helpful comments to substantially improve the paper. Outstanding errors and omissions are solely our own.