Traditional historiography on guilds has sketched a dichotomy between the regulated world of ‘corporatism’ and the unregulated world of free trade and the ‘invisible hand’. Entrance to a trade was regulated via apprenticeship, completion of master pieces and financial barriers to entry, and these led to a clear distinction between those ‘inside’ and those not.Footnote ¹ Journeymen had completed apprenticeships, masters had first worked as journeymen. In recent decades historical scholarship has qualified the impact of the guilds’ regulations and self-defined boundaries. From an economic perspective, it has been argued that craft guilds did not have a monopoly in the modern sense of the term.Footnote ² Guilds were not the sole sellers of products in a particular industry, nor did they hinder competition in export markets. Guild officials could only prevent artisans from making the same or similar products inside the same city. In order to compete against similar products coming from outside the city (from the countryside or other urban centres) they had to sell imperfect substitutes and invest in brand loyalty.Footnote ³

From the perspective of labour relations, the dichotomy between the regulated world of the guilds and the unregulated world of the ‘outside’ has also been revised. Historians studying the notorious faubourg Saint-Antoine in Paris have pointed to the interconnections between the formal economy intra muros and the informal economy on the outside. City-based masters could employ or subcontract to faux ouvriers in the faubourg; they could even work as subcontractors for, or in company with, an illegal entrepreneur.Footnote ⁴ Others have pointed to concentration trends and proto-industrialization even within city walls. Subcontracting was tacitly allowed so that masters could employ other masters (including masters from other trades) at their discretion, and masters could work in Kauf or Verlag systems of merchants.Footnote ⁵ This of course complicates the question of illicit labour. Exactly who in this context was legal or illegal? This question, despite the ‘return of the guilds’ to the historical agenda, remains largely unanswered.

Most historians connect illicit labour to false work at low wages. The artisans involved are termed chambrellans (in French) or Böhnhase and Störer (in German), alluding to a world of unskilled workers ‘without rank’ (‘sans état’) who make their produce surreptitiously (for instance at the back of the house).Footnote ⁶ Intuitively, clandestine work is related to a type of ‘guild decay’, particularly erosion from below, of the guilds’ rules.Footnote ⁷ In particular, guilds are seen as being ‘flooded by the market’, to paraphrase a metaphor of Steven Kaplan's.Footnote ⁸ The guilds’ eventual demise is typically linked to the removal of the guilds’ barriers to labour market entry and to the emergence of industries in the countryside (which was connected to proletarianization and proto-industrialization).Footnote ⁹ However, this is only one side of the coin, for these moonlighters must have been producing for either large entrepreneurs or merchants. How were the guilds’ boundaries blurred from this perspective? Regarding the Southern Netherlands, Catharina Lis and Hugo Soly have related the emergence of industrial capitalism to the initiatives of large masters, who not only employed more workers but also subcontracted to other masters; but at least in part this seems to have happened legally (as was subcontracting among masters).Footnote ¹⁰ In some sectors, it might have been merchants who engaged in production by employing licit or illicit workers instead of purchasing the finished product from masters.Footnote ¹¹ Regarding those workers, the question becomes one of distinguishing between, on the one hand, a master piece worker (perhaps working exclusively for one merchant and being paid in raw material) and, on the other, an illegal employee working (perhaps on his own premises as did a master) for a piece wage. However artificial a boundary this may seem, below we will see that for guild-based masters it was crucial.

The problem is that recent research has tended to focus somewhat one-sidedly on export industries. This bias has obstructed recognition of possible correlations between the undermining of master status and transformations in local distribution channels. Current research on industries working for local markets has highlighted the shift from working ‘on order’ to producing ‘to supply a market’ (prêt-à-porter or ready to wear) in the tailoring and shoemaking sectors. Such a shift may have been accompanied by wholesalers and retailers having products made or finished by myriad free and unfree workers.Footnote ¹² How were the guilds’ boundaries defined from this perspective? I argue that in early modern Antwerp the guilds’ rules were not gradually eroded or relaxed. Rather, they materialized and developed further in the early modern period, particularly in response to the clandestine activities of wholesalers and retailers. In the first section, we will see that standard ideas about deregulation of the labour market (e.g., workshop size gradually increasing in the context of emerging industrial capitalism and subcontracting among masters) should be complemented. While limits to workshop size tended to be relaxed from the sixteenth century on, other regulations – typically those defining master status – were strictly enforced and further developed. In the second section, attention shifts to the product market. It will be shown that guild officials were not principally concerned about possible proletarianization of masters or about the shift from a Kauf to a Verlag system. Rather, they were concerned about the connection between product quality and master status. In the third and final section the gradual demise of the guild system is reconsidered from the perspective of the so-called ‘retail (r)evolution’. The general thesis is that the guilds’ main problem was not that some masters grew larger but that mercers intruded into the field of production, thus blurring established differences between the retailing and production sectors.

For sources, I have used the broadest possible sample of guild ordinances. Virtually all craft guilds plus the mercers’ guild were included.Footnote ¹³ Additionally, an in-depth study of judicial litigations in a sample of guilds (the cabinet makers, the carpenters, the diamond cutters and gold and silversmiths, the pewterers and plumbers and the shoemakers and tanners) was analysed.Footnote ¹⁴ To compensate for the bias towards export industries found in most research and so as to concentrate on the changing relationship between producing and retailing, I have deliberately chosen crafts that were at least partially active on the local (and regional) market.Footnote ¹⁵

Distinguishing faux ouvriers from faux maîtres

Most literature on guilds and the problem of unfree labour does not clearly distinguish between unfree or un-privileged employees (faux ouvriers) and unfree or un-privileged employers (faux maîtres). For a clear insight, we should distinguish illicit ‘journeymen’ from illicit ‘masters’. Employees could be illegal when, for example, they exceeded the maximum numbers of journeymen or apprentices per master. In some guilds there was, moreover, a distinction between ‘free’ and ‘unfree’ journeymen so that journeymen, in order to be allowed to work in a master's workshop, had to complete an apprenticeship term and a master piece. The privilege of the so-called free journeymen (‘vrije gezellen’) consisted of a ‘right of preference’, which prevented masters from hiring outsiders (‘onvrije gezellen’), i.e., journeymen who had not finished an apprenticeship term (or a master piece).Footnote ¹⁶ To understand the rationale behind this, we should approach it from the perspective of journeymen. The right of preference forced masters to favour free journeymen when recruiting workers. Masters were allowed to recruit unfree journeymen only when no free journeymen were available or willing to work.Footnote ¹⁷ Not coincidentally, this right of preference was accompanied by a minimum wage, which masters tried to circumvent when hiring unfree journeymen (who had not finished an apprenticeship term). Similarly, the right of preference often included the obligation that the journeyman make a trial piece. This suggests that the practice of the ‘journeyman piece’ was established at the instigation of journeymen and with the aim of clearly distinguishing between ‘free’ and ‘unfree’ journeymen. A journeymen's trial, after all, only occurred in crafts with a right of preference.Footnote ¹⁸ As this right typically existed in industries with large numbers of journeymen, it is easy to imagine that their numerical strength enabled them to enforce the practice.

Remarkably, however, the distinction between free journeymen and unfree journeymen was not always present.Footnote ¹⁹ Gold and silversmiths, in their founding ordinance in 1454, prescribed an apprenticeship term, but article 2 explicitly states ‘that all good alien journeymen, wherever they come from, who want to earn their living in Antwerp, with a master for a daily wage, by contract or by piece, should be allowed to do so without further due’.Footnote ²⁰ The diamond cutters had a similar ruling, in that ‘every alien journeyman wanting to work here must do so with a free master and when asked have his name written down at the guild chamber within three days’.Footnote ²¹ The shoemakers stipulated that ‘no master shall employ an unfree journeyman outside his house nor will journeymen be allowed to work other than in a master workshop’, implying that they could work for a regular master.Footnote ²² In other words, these guilds established the obligation to work for a free master, not the obligation to be ‘free’ for workers. In most guilds there was no reference to distinction between free and unfree journeymen or to possible obligations for journeymen to have finished an apprenticeship term. Thus, apprenticeship should be considered an entry to mastership rather than to journeyman status.

Doubtlessly this has serious consequences for research on social mobility in guilds. Historians have generally assumed that apprentices typically rose from the status of apprentice to that of journeyman and then to mastership, but we must now distinguish between an apprentice who wanted to become a master directly (perhaps after having already worked as an ‘unfree’ journeymen for years) and apprentices who sought to become a journeyman.Footnote ²³ Secondly, and more important for this article, the question of the changing size of the firms should be reframed. What was the relationship between workshop size and the journeymen's right of preference? And how did workshop size evolve in the long term?

The absence of a right of preference did not thwart guilds from limiting workshop size, but in general the traditional narrative of gradual deregulation holds: rules concerning the maximum number of journeymen were typically relaxed, especially in the sixteenth and eighteenth centuries. As the economy expanded, concentration of production in the hands of large masters increased. This shift typically occurred in export industries – especially textiles – but even guilds as different as shoemakers and cabinet makers evolved from allowing one ‘workbench and one stall’ or ‘one journeyman and one apprentice’ at the end of the fifteenth century to allowing a maximum of six journeymen (in addition to their children) during the sixteenth century.Footnote ²⁴ In the mid-eighteenth century, the conflict between small masters striving to limit the number of employees and large masters wishing to deregulate the labour market seems to have turned in favour of the latter, suggesting that industrial capitalism came about as large masters prevailed and succeeded in concentrating production in their own hands.Footnote ²⁵

These findings correspond to Lis and Soly's thesis that the large entrepreneurs in the industrial sectors of Flanders and Brabant came from the ranks of the craftsmen rather than the merchants.Footnote ²⁶ In some sectors with a right of preference for journeymen, guilds adopted strategies to include unfree workers, most likely at the instigation of large masters. From the fifteenth and sixteenth centuries on, guild-based masters could often, within certain limits (mostly for a specific period of time), employ unfree journeymen when confronted with production peaks.Footnote ²⁷ The guilds of the carpenters and the masons (the Vier gekroonden) legalized unfree journeymen in the eighteenth century by establishing a bankgeld, a small tax related to how many days an unfree journeyman worked (typically one stuyver per day).Footnote ²⁸

At the level of the employers or masters, however, there was an opposite trend. Here regulations tended to become stricter, or at least more elaborate. At first glance the distinction between masters and illegal employers may seem far more straightforward than the distinction between free and unfree journeymen from the start. Anyone who finished an apprenticeship term, completed a master piece (if required) and paid the entry fees could call himself a master. On the face of it, moreover, it was clear what masters could and could not do. They could make all products on which the guilds had a monopoly, and have their apprentices and journeymen make the products. However, things become more complicated if we concentrate on labour relations at large. Prohibitions on subcontracting were usually absent, and this allowed masters to have their products made by other masters (regardless of the latter's trade) or to make them (or have them made) for other masters. Masters could even be explicitly permitted to have products made by non-masters outside their workshop.Footnote ²⁹ Nonetheless, guilds and guild officials continued to distinguish between masters and illegal entrepreneurs, and even devised new instruments for doing so. The question is why?

As we will see below, the guilds in our sample typically faced the problem of legal masters working directly for an illegal entrepreneur, not as a subcontractor or in a Kauf system (selling all his produce to one or several merchants) but as a type of employee.Footnote ³⁰ Guilds had few options in trying to counter this. In the absence of a master piece and regular workshop visits by guild officials, it was easy for an illegal entrepreneur to become a master. He simply registered as an apprentice with a free master, in return for financial or other compensation to the master in question.Footnote ³¹ Once the term prescribed by the guilds had expired, such an entrepreneur could become a master simply by paying the fees due and swearing an oath – even if he had not spent one day on the master's shop floor. One of the guilds’ responses was to obligate masters to make a master piece, which they appear to have done in particular in the long sixteenth century.Footnote ³² When this strategy failed – among other things because prospective masters did not always make their trial piece themselvesFootnote ³³ – guild boards forced apprentices to live on their master's premises, a measure they explicitly attributed to the need to block fictitious apprenticeships.Footnote ³⁴ In response to these measures, faux maîtres concluded fictitious apprentice contracts whereby fictious apprentices either moved in with their ‘master’ (allegedly to learn, while in reality the master worked for them) or the ‘master’ moved in with them.Footnote ³⁵ The gold and silversmiths considered setting a maximum age so as to prevent adult newcomers interested in settling in Antwerp from having other masters register them as fictitious apprentices to comply with guild regulations.Footnote ³⁶

But guilds still faced the problem of illegal masters using the trademark or the name of a legal master (in exchange for money or because they had founded a ‘company’) in order to legalize their products.Footnote ³⁷ As early as the third quarter of the sixteenth century, some gold and silversmiths founded companies with ‘unfree persons who do not live with their master and who are working at a place where the guild sign does not hang out’. Some were found ‘in a shop’ in the so-called Pand (building) of Our Lady, where they contracted work, in turn, to unfree workers.Footnote ³⁸ Thus, it was generally forbidden for masters to found companies, as such companies would enable them to regulate illegal entrepreneurs (and work for them at the same time).Footnote ³⁹ Similarly, free and unfree journeymen were sometimes allowed to work for a master only if they lived or at least worked directly on his premises.Footnote ⁴⁰ The reason for this appears to have been that such journeymen often worked on their own accounts, perhaps even employing the master rather than vice versa.

From proletarianization to discussions on product quality

When viewing the decline of the guilds, historians are accustomed to consider the major underlying factor to have been increasing outside pressure on a long-standing, fully fledged guild system. Yet in Antwerp, the reverse was true. The guild system actually matured in the sixteenth and seventeenth centuries. In the second decade of the seventeenth century, master diamond cutters protested against a Portuguese merchant who had several masters working directly for him. The guild repeatedly reaffirmed its bans on working in company and on ‘freeing’ unfree workers, but in the course of the eighteenth century the merchants prevailed and became true fabriquateurs.Footnote ⁴¹ The ribbon makers’ guild legalized the production activities of merchants in 1721. It was stipulated that merchants could become guild members if they paid an income fee (used to fund the poor box).Footnote ⁴² However, unlike the distinction between free and unfree journeymen, masters were not inclined to loosen the rules concerning master status until at least the end of the seventeenth century. Two of the guilds’ most typical regulations – the compulsory master piece and requiring apprentices to live on their master's premises – emerged as attempts to address the problems of entrepreneurs circumventing master status. Even a fixed term of service is likely to have been introduced for that reason. In the first half of the fifteenth century guild officials merely had to declare before the city magistrates who a person was when accepting him as a new member.Footnote ⁴³ At best, new masters were required to ‘demonstrate’ their competence, which presumably meant that master trials were required ad hoc for prospective masters not known in the city.Footnote ⁴⁴ From the second half of the century on, an apprenticeship period gradually became obligatory – followed (some decades later) by the installation of master pieces and backed by measures preventing fictitious apprentice contracts and the foundation of companies. The rationale was to define and defend master status. The difference between a ‘merchant’ illegally acting as a producer and a large ‘entrepreneur’ legally acting as a master was the obligatory apprenticeship term and master piece for masters.

This answers the question of what the difference was between an employee working for piece wages (perhaps on his own premises and at the same wage as a master's) and one selling his products by the piece to a merchant or a wholesaler (as a regular master, but perhaps being paid in raw material). An interesting judicial conflict between the ‘nation’ of the pewterers and plumbers (a guild subordinate to the mercers’ guild) and several large mercers is revealing in this context. About 1600 some ‘mercers’ (sometimes called ‘potiers’ (‘potverkoopers’)) were accused not of buying and selling pots but of producing them. The problem was that they hired pewterers, whom they paid in cash or with pewter, to make pewter lids that were then attached to earthen pots furnished by the mercers.Footnote ⁴⁵ The guild of the pewterers and plumbers claimed that the potiers had to be masters to do this, or, in other words, that they had to have finished an apprenticeship term and a master piece. The potiers claimed that they were merely selling pots, and that it was thus sufficient that they belong to the mercers’ guild. But the pewterers continued to differ. Indeed, it was allowed for tinsmiths to work exclusively for mercers and to sell lids to them by the dozens. The problem, however, was that the tinsmiths – in not only making the lids but attaching them to the pots – were thus working ‘as a servant’ (‘dienaer’) instead of ‘as a master. . .who sells his tin and earns a living in doing that’.Footnote ⁴⁶ Significantly, the case of the pewterers was based on an ordinance stating that tinsmiths could make tinwork only on their own account.Footnote ⁴⁷

Although the number of apprentices and journeymen working for the tinsmiths was mentioned in the judicial proceedings, the conflict revolved around the economic relationship between masters and mercers. In the end, the problem for the guilds was not so much that masters worked as servants or employees, but that some entrepreneurs, who were not masters and thus should not have been allowed to engage in production, were doing so nonetheless. This same problem underscored a conflict between the shoemakers and the tanners: the tanners contested the right of shoemakers to buy hides and have them tanned by the tanners (rather than buying tanned hides directly from them). A compromise was reached in the last decades of the seventeenth century. It stated that shoemakers had the right to buy hides and have them tanned by a tanner, provided the tanner was a master tanner. Conversely, tanners could have shoes made, provided they were made by a master shoemaker.Footnote ⁴⁸ Here as well, the craft guilds were determined to differentiate between a master working ‘on his own account’ and a master working as an employee to faux maîtres. Although the distinction between a master piece worker paid in tin and an employee working on his own premises and paid per piece may have seemed trivial, for the masters it was crucial.

The point, however, was not proletarianization of masters as we traditionally understand it. Indeed, in the conflict between the tinsmiths and the mercers, the question was not who possessed the means of production. Nor did these conflicts revolve around the difference between a Kauf and a Verlag system. The fact that the pewterers were sometimes paid in raw material was not seen as a problem. Instead, the issue was that large entrepreneurs who were not masters acted as producers rather than as merchants and that small masters served as employees instead of as subcontractors or as independent piece workers. These two problems should be considered as two sides of one and the same coin. For the craft guilds, the products had to be made (or at least the production process had to supervised) by a master.Footnote ⁴⁹ In 1652 a master silversmith with eye problems founded a company with his brother-in-law, whom he officially paid only food and board. The guild officials protested, arguing that the brother-in-law had a part in the company's profit, which was forbidden as he was not a master.Footnote ⁵⁰ Had the brother-in-law been able to prove that he received a wage, there would not have been a problem. For the same reason it was sometimes forbidden for masters to work or to open shops outside their own houses.Footnote ⁵¹

The ultimate argument of the guild-based masters was always product quality. In their conflict with the potiers, the pewterers argued that the lids were not made according to the standards applied by masters. For the tanners, the ultimate problem was that entrepreneurs integrating tanning and shoemaking could use tanned hides of lesser quality for shoes.Footnote ⁵² In a notorious judicial litigation from the first half of the seventeenth century between the gold and silversmiths and the mercers, the guild protested against the mercers’ habit of putting silver on leather belts. They argued that the mercers were not allowed to act as producers because they might use inferior silver instead of the alloy the guild prescribed.Footnote ⁵³

Of course, the importance of product quality as such does not explain the importance of defining master status. Guilds’ apprenticeship arrangements did not necessarily result in skills or products of higher quality. From the evidence of private apprentice contracts, apprentices often appear to have been under tutelage longer than the term the guilds prescribed;Footnote ⁵⁴ or they learned specialized skills or moved from workshop to workshop to learn a wider range of skills.Footnote ⁵⁵ As a result, the skills of guild-based artisans were not necessarily superior to the skills of outsiders. In the mid-eighteenth century, masters in the building sector discussing the use of unfree labour with free journeymen claimed that unfree journeymen, some of whom were aliens, were better trained and had skills more attuned to demand.Footnote ⁵⁶ Moreover, the range of wages within the group of free artisans could show huge differences among ‘qualified’ artisans, to the extent that differences within the group of free journeymen were greater than the differences between qualified and unqualified journeymen.Footnote ⁵⁷ By around 1600 unfree journeymen were sometimes paid more than free journeymen, suggesting that they were better trained.Footnote ⁵⁸

Still, as masters regarded themselves as the ultimate warrantors of product quality, any use of a trade mark by someone who had not finished an apprenticeship term and master piece – either by having the mark lent to him in exchange for money or because he worked in company with a legal master – was the ultimate offence. This is why, as a rule, only masters (about half the workforce at best) were required to make a master piece.

The reasons craft guilds were unable to safeguard master status are found in the product market rather than the labour market. Ultimately, the conflicts between the potiers and the pewterers and between the gold and silversmiths and the mercers were about products that crossed guild boundaries, e.g., earthenware with tin lids or leather belts with silver buckles. Similar problems surfaced in the production of luxury furniture. Luxury cabinets (cantoren, scribaenen and the like) were renowned for the myriad raw materials and skills used in making them; besides exotic timbers, these included silver, ivory, tortoiseshell, agate and lace. This was one of the sectors in which important art-dealers were accused of transgressing the thin line between trade and production. During the seventeenth century, art dealers (including Guilliam Forchondt) were accused of acting as large producer-entrepreneurs without being masters or being able to perform the job themselves.Footnote ⁵⁹

For sure, guilds had faced these problems for decades, but from the seventeenth century onward their arguments seem to have lost credibility. As I have shown elsewhere, beginning in the sixteenth century increasing product complexity and changing consumer preferences forced masters and merchants alike to adopt strategies aimed at reducing information asymmetries (i.e., the seller of a product having more or better information on product quality than the buyers, resulting in a lack of trust among the latter). The guilds adhered to their system of quality controls and trademarks, diversifying their marks as the product range increased. Merchants seem to have welcomed this initially, but as consumer preferences shifted towards less expensive and more fashionable products in the seventeenth and eighteenth centuries,Footnote ⁶⁰ the guilds’ arguments, based upon product quality and masters’ trustworthiness, became harder to sustain. Merchants and mercers progressively competed with low prices instead of high quality, and invested in novelty and design rather than brand loyalty.Footnote ⁶¹ In this context merchants and mercers encroached on the masters’ terrain, whereas guilds devised strategies (master pieces, boarding requirements for apprentices, etc.) to prevent them from doing so.

From an industrial to a retail revolution

Theoretically, of course, there would not necessarily have been conflict between merchants (or mercers) and masters over product quality. Both masters and merchants (or mercers) had an interest in reducing information asymmetries. Indeed, it is often difficult to distinguish between merchants or mercers who were delegating monitoring costs for guarding product quality to the guild and craft guilds referring to product quality in order to distinguish masters from faux-maîtres. The founding ordinances of new craft guilds established in the sixteenth century that relate deliberations preceding the establishment of the guild in question suggest that merchants and (some) masters had agreed on the necessity to erect a guild and issue rules.Footnote ⁶² Yet, in the seventeenth and eighteenth centuries tensions between masters and merchants (and mercers) appear to have increased.

On a cultural level, this might have been due to changing consumer preferences and concomitant ‘repertoires of evaluation’. Craft guilds, while establishing rules for product quality, inevitably objectified product quality.Footnote ⁶³ By establishing a distinction between ‘qualified’ and ‘unqualified’ products, they defined what quality was. Their ‘justification’ of quality was based largely on the quality of the raw material used. Trademarks typically referred to the quality of the wood used by cabinet makers, the quality of the tanned hides used by shoemakers and the alloy of tin and silverware used by pewterers, plumbers and gold and silversmiths. Within the same industry, trademarks tended to distinguish among the different sorts of raw materials used. The shoemakers’ trademarks, for example, distinguished between different parts of the hide, such as the parts of the animal skin that could be used for new shoes and those that could be used only to repair shoes or to make children's shoes.Footnote ⁶⁴ The trademarks used by cabinet makers referred to the quality of the wood used or to indicate that the right wood had been used (for example, genuine ebony wood rather than Spanish wood, which had a similar appearance). The pewterers and plumbers and the gold and silversmiths had different marks for different alloys and according to the quality and the amount of tin and silver in the alloy.Footnote ⁶⁵ In all these industries product quality was objectified by reducing it to an invisible feature of the product – not coincidentally, of course, the most expensive feature.

From this perspective, it is easy to imagine why the guilds’ claims lost credibility in the face of changing consumer preferences (stressing novelty and design). But how was this reflected in the local distribution channels? In the discussions between guilds and illicit entrepreneurs wholesalers and even retailers were involved.Footnote ⁶⁶ This is hardly surprising. As recent research has shown, Antwerp in the seventeenth and eighteenth centuries housed exceptionally high numbers of retailers. Apart from lace making, the retail sector appears to have been the most dynamic sector in the city after about 1650. The number of so-called faiseurs de rien, vendeurs de tout (that is, people who make nothing but sell everything) never dropped below 1,500 members, even in the so-called ‘age of crisis’ (1648–1748). Around 1690, the retail ratio in Antwerp is said to have been 1 mercer for every 26 persons. In 1773 it was 1 for every 16 persons.Footnote ⁶⁷ Clearly, the shift towards industrial capitalism was not the only transformation the guilds had to address. We should also be attentive to mercers and retailers and their relation with guild-based producers.

Unfortunately, current research on retailing tends to focus somewhat one-sidedly (and often implicitly) on changing demand and consumer preferences, thus leaving the relationship of retail and production structures obscured.Footnote ⁶⁸ What if we approach the so-called retail revolution from the perspective of masters losing independence and master status becoming precarious? Were the faux-maîtres in effect mercers who sold their products directly to customers? Or were they wholesalers supplying shops, perhaps shops in which other proletarianized masters (or their wives) earned their livings? An interesting path for future research would be to examine guild statutes and registration ledgers along with probate inventories and tax lists. A study that focused on Alost (a small city in south-east Flanders, on the banks of the Dender River) has revealed that crucial shifts may have occurred behind the masters’ doors. The emergence of large (illicit) entrepreneurs appears to have been related to the disappearance of the ‘traditional’ producer-retailers who sold their own products, either directly to consumers (in their shops) or to merchants. Whereas masters were accustomed to working in the front rooms of their houses, where they were visible to both the guild deans and the customers, the emergence of shops went hand in hand with either specialization within the sector (some masters specializing in retail, others producing exclusively for large entrepreneurs) or the gradual separation of work floor and shop within the house (the work floor moving towards the back of the house). In both cases, goods sold in the shop (perhaps by the master's wife) were not necessarily made in the back (by the master) but might be supplied by a large entrepreneur, while the ‘master’ in the back was perhaps producing not for his or his wife's shop but for a large entrepreneur.Footnote ⁶⁹

Further research should of course confirm this trend (the more so as our Alost sample was rather small) but crucial questions in any case arise about the relationship between transformations in local distribution channels and the disappearance of master status. Masters who lost their independence could choose between continuing their work in the same sector, working directly for a mercer or wholesaler (either in his own house or on his employer's premises), or he could become a retailer selling products a large ‘entrepreneur’ had manufactured or ordered. The existing literature offers examples of masters working for wages from another master in a putting-out system while still working on their own premises.Footnote ⁷⁰ Others went to work on the premises of a merchant, as did the master diamond cutters with the Portuguese dealer in diamonds. As to the manufacturing of cantooren, it is hard to imagine that a cabinet was moved between workshops, so there, too, masters appear to have stopped working on their own premises.Footnote ⁷¹ Other masters, although continuing to work in their own houses, may have stopped selling their own products. Some artisans, masters and others, began selling raw materials instead of concentrating on selling only finished products. While the masons (Vier gekroonden) in 1665 complained about unfree people who sold raw materials related to the building sector (stone, chalk, etc.),Footnote ⁷² some coopers sold wooden bands (‘repen’) rather than using them to make barrels.Footnote ⁷³ On the other hand, finished products were gradually sold to customers by people other than producers. In 1442 the Antwerp pewterers stipulated that master pewterers could sell only their own wares in stalls,Footnote ⁷⁴ yet during the seventeenth and eighteenth centuries this was far from straightforward. In their dispute against the mercers, the pewterers mentioned masters and masters’ widows selling items they had not produced.Footnote ⁷⁵ Cabinet makers, moreover, protested against masters who bought products outside the city and sold them in their ‘shops’ inside the city.Footnote ⁷⁶ The guild of locksmiths had to deal with apprentices and journeymen who moved to the countryside to make locks and other iron work that they sold in the city to ‘voorkopers’ (intermediaries).Footnote ⁷⁷

Of course, resistance to imported products was a time-honoured practice. For example, there were protests against hats made in England, France and Holland in the seventeenth century.Footnote ⁷⁸ There were also typical protests against unfree products from the countryside, such as by glove and purse makers in the first half of the eighteenth century.Footnote ⁷⁹ In these cases the distinction between guild-based products and non-guild-based products was clear. Complications ensued when either regular masters sold illicit products or illegal retailers or wholesalers sold legal products without it being clear if they had been made by masters. For the guild-based masters, production and retail – except on export markets – had to be done by one and the same person. Unfortunately for these masters, however, the straightforward bond between production and retailing – linked together by mastership – seems to have disappeared in the long run. The guild of glove and purse makers, for example, eventually allowed gloves and purses to be sold by people who had not finished an apprenticeship term or made a master piece.Footnote ⁸⁰ In the end, guild-based masters relinquished the visible connection between production and retail. Some gold and silversmiths even had two ‘workshops’.Footnote ⁸¹ One guild member was fined for not displaying the guild's sign (with the typical draughtboard pattern). The master argued that he did not work there, as his furnace was in the other house, and requested permission to separate the shop from the work floor, keeping only the balance, some small instruments and some drawings in the shop (which was situated in the city centre).Footnote ⁸² These masters, in other words, not only stopped working in their homes but stopped selling their products at the production site. They had stoves and bellows elsewhere and, more importantly, they ceased to display the guilds’ signs or marks.Footnote ⁸³ In short: they had relinquished the idea that the problem of product quality should be solved by tying together product quality, trade mark and master status in the front room of the masters’ house – where the product was made and sold under the sign or banner of the guild.

Whenever the link between production and retail disappeared, the guilds suffered a life blow. In the previously mentioned juridical proceedings, the link between product quality and master status (the latter embodied in the obligatory apprenticeship term, the master piece and the trade mark) was the ultimate motivation for their defence against unfree work. In the end, guilds could cope with non-guild products being sold at weekly and annual markets, which were free for all sorts of products. Trademarks distinguished their products from unfree wares and allowed them to continue to claim superior quality.Footnote ⁸⁴ But as retail activities expanded and reached the heart of the guilds, the link between product quality and master status was severed. In the eighteenth century Paris retailers eventually began applying their own marks on silver, placing them on items made by subcontractors, thereby overruling the marks of guild-based masters.Footnote ⁸⁵