Hostname: page-component-7dd5485656-g8tfn Total loading time: 0 Render date: 2025-10-27T17:36:53.477Z Has data issue: false hasContentIssue false

Seasonality and the Practices of Illegal Trade in the Faroe Islands and the Northern and Western Isles, 1775–1785

Published online by Cambridge University Press:  27 October 2025

Anna M. C. Knutsson*
Affiliation:
Department of History, Uppsala University, Uppsala, Sweden Clare Hall, Cambridge, UK
Rights & Permissions [Opens in a new window]

Abstract

By investigating 321 smuggling vessels that travelled between the Faroe Isles and the British Isles between 1775 and 1785, this article explores the regionality and seasonality of illegal trade in the north Atlantic. The article centres on a neglected oceanic region, thereby inverting historical geographies, shifting the focus from port centres like London and Copenhagen to the Scottish Northern and Western Isles, as well as the Faroe Islands. Epitomizing the ‘dichotomy of insularity’, in the eighteenth century these areas were located along maritime highways of global trade, and illegal trade flourished there, despite their challenging location. The article demonstrates that this illegal trade had different seasonal cycles from those of legal trade. While legal trade halted in the winter season, illegal trade continued throughout. Linking illegal maritime activities to more lawful ones, like fishing, the article suggests that the recurring nature of the contraband trade ought to be understood in relation to the broader coastal culture, in which maritime knowledge was circulated and the local relationship to the sea defined. The study suggests that illegal trade should be viewed not merely as an extension of globalization but as a rooted phenomenon, which developed in unison with the local environment.

Information

Type
Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press.

‘It cannot be well exprest with a pen’, the Danish priest Lucas Debes wrote in 1673, ‘how fierce the Sea is, not [sic] to what height it raiseth it self, when the wind and the stream are against one another.’Footnote 1 Appearing in his Færoæ et Færoa reserata, one of the earliest books ever written about the Faroe Islands, these words of Debes capture the brutality of conditions in the north Atlantic. The volcanic Faroe Islands’ location, surrounded by three seas – the Norwegian Sea to the north-east, the North Sea to the south-east, and the Atlantic Ocean to the west – exposes them to rough winds and a rapidly shifting, harsh climate, becoming most ferocious during the winter months. Since 1578, the trade to the islands, which were a Danish possession, had been controlled through a royal monopoly, and vessels of no other nationalities were allowed to trade there. Traditional accounts tell us that royal supply ships sailed to the Faroes four times a year, supplying the islands’ only domestic shop with all of the vital necessities that had to be imported, such as grain, wood, iron, hemp, and salt.Footnote 2 Due to the weather conditions, the first voyage usually took place in April or May and the last in September or October. Crucially, this left a six-month period when the islands were essentially cut off from the outside world. At least, that is the official story.

Despite the challenging conditions of the Faroe Islands, a Danish state-sponsored contraband entrepôt, stocked with West Indian rum, Chinese tea, American tobacco, and European spirits, opened up just outside Torshavn on the Faroe Islands in the 1760s, catering to Scottish, and to a lesser degree Irish, smugglers. As this article will show, this contraband trade persisted throughout the seasons. By exploring the relationship between seasonality, regionality, and inter-island connectivity, as observed in the illegal trade of global goods between the Faroe Islands and the Northern and Western Isles of Scotland, this article examines why and how this trade thrived when other trade hibernated.

Drawing on David Armitage’s idea of ‘cis-Atlantic history’, a history based on unique locations within a larger network of Atlantic connections, the article considers the interactions between three north Atlantic island regions: the Faroe Islands and the Northern and Western Isles of Scotland.Footnote 3 Entanglements between the Faroes and northern Scotland in the eighteenth century remain poorly understood. Indeed, a prevalent idea seems to be that there were no interactions between them in this period, as states turned inward and centralized control in pursuit of a mercantilist ideal.Footnote 4 However, we now know that the early modern economic world was far more multi-faceted and dynamic than traditional interpretations of mercantilism have allowed for.Footnote 5 To understand insular connectivity we need to move beyond national and even imperial narratives and adopt a trans-territorial perspective on the island regions.Footnote 6

The Faroe Islands and the Northern and Western Isles were all considered peripheries of the Danish and British empires, but they were also linked to imperial and global trade networks, frequently through illegal trade channels.Footnote 7 Research on illegal trade has shown that fringe areas like these, especially islands, often played a crucial role in the illegal economy, and some have argued that we ought to take these ‘off-grid’ exchanges seriously as part of the early modern processes of globalization in order to challenge our understanding of ‘peripheries’ as ‘peripheral’.Footnote 8 As Julian Hoppit points out, the black economy was more likely to thrive in areas where revenue officers were few and far between, legal opportunities limited, and poor aid and charity hard to come by – just such places as the Scottish Northern and Western Isles.Footnote 9

The illegal trade to Scotland has received quite a bit of scholarly attention in recent years. This research has to a significant extent focused on particular families and mercantile ties, as well as on a specific good: tea.Footnote 10 It has shown that families, often dispersed across Europe as a consequence of several failed Jacobite risings, played a crucial role in the creation of a global underground in Scotland, turning it into one of the main markets for contraband tea in Europe.Footnote 11 The purpose of this article is different; here, the focus is on the practices of illegal trade, especially on how they were affected by the specific regionality and seasonality of these northern waters. Earlier research on overland illegal trade in the north of Europe has shown that winter increased the autonomy of local populations, while undercutting control of remote possessions, turning smuggling into a wintertime activity in many places.Footnote 12 Hitherto, no comparable studies have been made of maritime smuggling.

Meanwhile, accounts of maritime transport in the northern hemisphere, often focused on high commerce, have suggested that there was a winter hiatus in the traffic, similar to what we see in the monopoly trade to the Faroe Islands. This hiatus would have been increasingly pronounced the further north one went, being particularly noticeable in areas such as the Baltic Sea, which partially froze over during the winter.Footnote 13 However, Horden and Purcell have questioned this hiatus in relation to the Mediterranean, suggesting that low-intensity, short-distance transport was probably fairly constant all year, although it would have peaked during the more favourable seasons.Footnote 14 Similar observations have been made for the Nordic countries, where trade, particularly short-distance trade, appears to have continued all year round, ice allowing.Footnote 15 Indeed, winter trading may have been an important tool in the competition for international trade. As Dagomar Degroot has shown, the Dutch ability to harness seasonality and to continue trading during cold spells gave them a competitive advantage and played a crucial role in the prosperity of the Dutch Republic during its golden age.Footnote 16 Yet, while extending the trading season may have been desirable, it was not easy.

Before the codification of maritime knowledge and the development of new technologies to aid sea travel, such as sea charts and lighthouses, the ability to navigate relied heavily on local, inherited, and learned knowledge of landmarks and memorized coastlines, sometimes referred to as mental mapping.Footnote 17 This type of knowledge, often embodied in the pilot, was also a crucial tool of small-scale skippers and fishermen.Footnote 18 It offers a clear reminder that, before the codification of maritime knowledge, ocean travel was not restricted by blank spaces on maps, but that these individuals possessed vital agency that connected the world, patch by patch. Consequently, Sara Caputo refers to maps and the tracks made across the oceans as the ‘colonial appropriation’ of local knowledge, rather than as the invention of new knowledge.Footnote 19 This local maritime knowledge was particularly important during the winter, when storms caused challenging and swiftly altering travel conditions, as they did in the high north Atlantic.

No studies to date have considered how illegality and seasonality interacted in the north Atlantic, despite the fact that winter had a fundamental effect on all life in the area. It has therefore not been clear whether legal and illegal ocean-going traffic were affected differently by seasonal change. In seeking to address this issue, the article investigates how the natural environment, in particular the dynamic ocean, shaped trade in the region. It engages with the ocean as an active setting, conditioned by relatively stable seasons, currents, and prevailing winds, interrupted by specific temporary factors such as weather and volcanic eruptions.Footnote 20 The main focus is therefore on the effects of normal damaging environmental factors rather than those of extraordinary catastrophes.Footnote 21 While I am not explicitly exploring climate change, ‘gradual climate change’ is still an underlying historic reality. This article therefore looks at how climate, particularly seasonality, and human activity interacted ‘across short timescales in local environments’, in order to gain new insights into how human history has developed in union with the environment.Footnote 22 Drawing on other local maritime activities, particularly fishing, it asks how an inclusive approach to local marine practices can help us understand illegal trade.

In exploring these topics, the article draws on four types of primary sources. The first body of records consists of the contraband entrepôt’s own tax records. The entrepôt, initiated by the prominent Danish merchant Niels Ryberg, had received a state licence in 1766. While Ryberg owned the entrepôt, the daily running of the business was administered by the company Rosenmeyer & Flores, composed of Christian Rosenmeyer and Otto Urbanus Flores. As part of the licence, the entrepôt’s transactions had to be filed annually with the Danish authorities. This included records of all incoming and outgoing ships, as well as all the wares they brought with them, registered by volume or quantity, along with the duty paid. These records provide the core of the current article. As there were some changes in how goods were registered over time, discussion mainly focuses on the period from 1775 to 1785, when the records were consistent and comparable. The second type of source material is commercial correspondence, namely the letters to and from the entrepôt, as well as between merchants and sailors. In addition to these sources, I refer to Scottish court cases and to newspapers.

I

The Faroe Islands are quite distant from any other landmass, the nearest inhabited neighbours being the Northern Isles of Scotland, comprising Shetland and Orkney, which are located 228 nautical miles from the Faroes, and the Western Isles, including the Isle of Lewis, situated 180 nautical miles away. In the eighteenth century, the distance between the Faroes and Orkney equated to roughly two and a half days’ sailing in fair weather, and between the Faroes and the Isle of Lewis, in the Outer Hebrides, to around two days.Footnote 23 Under more taxing conditions, the journey could take significantly longer. The stretch of water between the Faroe Islands and northern Scotland is perilous, and navigating it was challenging in all types of weather. The dangers inherent in travelling across the northern waters led locals to associate the ocean with something evil, mysterious, and threatening, while knowledge of how to safely traverse the waters was highly valued.Footnote 24

Winter exacerbated these dangers. The meeting of the North Sea and the Atlantic Ocean produces dangerous riptides, while the Atlantic creates cyclonic storms during the winter months, in addition to the frequent gale-force winds of the North Sea. Due to the location of the Faroe Islands, in the middle of the western wind belt, the islands suffer from extreme winds, particularly during the winter. Even when the weather is calm, danger may lurk under the surface. As Lucas Debes noted, ‘when there will come a strong Frost and Snow in Winter: the Currents grow then stronger, The Sea beats, the Bossves arise, though it be very still weather’.Footnote 25 In addition to all of this, there is the all-enveloping darkness. The winter months supply only a few hours of daylight to navigate safely through the many unmapped skerries and banks that litter the coastlines, not to mention the whirlpools (sometimes referred to as sea abysses) that lie between the islands.Footnote 26 The extreme conditions and the complicated geography set the Faroe Islands apart even in a north Atlantic context.

Meanwhile, by the eighteenth century the intensification of north European involvement in global trade had led to an increase in long-distance traffic through northern European waters. Shipping highways from the Americas and the East Indies to northern Europe criss-crossed the area.Footnote 27 Located on the threshold of Europe, the Faroes, Shetland, and Orkney were often the first inhabited landmasses that cross-Atlantic vessels encountered after a long journey.Footnote 28 Nordic East India ships, as well as vessels returning from the Danish colonies – particularly St Croix in the Danish West Indies – regularly passed by the islands, carried there by the north Atlantic current of the Gulf Stream.

The critical location of the Faroe Islands within the Danish transport network was a significant contributing factor in determining why this location was chosen to house a contraband entrepôt. The islands provided a convenient stopping point for incoming Danish ships stocked with global wares, and the entrepôt received deliveries straight from the Danish West Indies on average once or twice a year, as well as from the Danish East India Company’s warehouse in Copenhagen. Over time, there were also increasing numbers of deliveries from Amsterdam, Rotterdam, and Barcelona.Footnote 29 By the middle of the eighteenth century, these islands could therefore be considered both inaccessible insular peripheries, from an imperial political perspective, and nodes in the increasingly tight web of global trade. They epitomized Louis Sicking’s idea of ‘the dichotomy of insularity’: islands that were defined both by their connectivity and by their distinct and isolated nature.Footnote 30

While the Faroes were chosen, at least in part, for their strategic position within the Danish transport network, the trade also relied on a pre-existing and highly regional type of connectivity between these north Atlantic islands. The Faroes and the Northern Isles had once been part of the medieval Norwegian thalassocracy, but over time the border between Scotland and the Danish-Norwegian crown had shifted from the Pentland Firth to north of Shetland. This area had long also been home to strong regional-insular identities that transcended new and old national borders and engaged in trans-territorial cultural, mercantile, and political exchange.Footnote 31 Seafaring in these largely uncharted waters built on inherited local knowledge and interpersonal exchange.Footnote 32 Information would have been passed down orally as well as through shared experience, as generations were trained in memorizing the seascape and reading the wind, water, stars, and sun.Footnote 33 Knowledge of this type could prove remarkably durable, and there are indications that sailing know-how and practices adopted from Old Norse seafarers continued to be in use until the nineteenth century.Footnote 34 By the eighteenth century, this oceanic territory had therefore long been entangled, tied together by local maritime tradition and knowledge.

Seasonality played an intricate role in this northern naval world. Ancient Norse sailors advised against ‘sailing out of season’ – that is, during the autumn or winter.Footnote 35 In the early modern period, most captains still attempted to avoid sailing through the area during the winter period, usually defined as between October and March; those who did, did so with trepidation.Footnote 36 This is reflected both in the royal monopoly ships sailing to the Faroe Islands and in the ships supplying the entrepôt with contraband, which were sailed by a handful of Danish sailors.Footnote 37 Just like the royal monopoly trade, the entrepôt’s licence specified that only Danish vessels were allowed to bring contraband to the islands.

II

A very different movement pattern emerges if we look at the ships bringing contraband out of the Faroe Islands to the British Isles. Of the 321 vessels that left the entrepôt during the period from 1775 to 1785, approximately 34 per cent, or 111 ships, departed between October and March, and 66 per cent, or 210 ships, left from April to September.Footnote 38 While there was a reduction in traffic during the winter months, it certainly did not cease altogether (Figure 1). Intriguingly, by breaking this data down by year we can see that there was also an increasing tendency to engage in winter smuggling after 1776, peaking in 1782, when 44 per cent of all contraband left the entrepôt during the winter (see Figure 2). The discrepancy between summer and winter thus became less pronounced over time. This is particularly noteworthy when we align the shipping movement with contemporary weather data.Footnote 39 By comparing the departures from the contraband entrepôt with historical weather accounts, we find that years with particularly bad weather, such as 1782–4, did not see a reduction in traffic: if anything, there was a slight increase. The reasons for the increased turnover are complex and related to the ongoing American Wars of Independence.Footnote 40 However, here it is enough to note that winter smuggling gradually became more common despite the increasingly inclement weather. It is also worth pointing out that the smugglers were not engaged exclusively in a one-way exchange; rather, the Scottish and Irish vessels frequenting the entrepôt are known to have brought passengers as well as food to the Faroes.Footnote 41 The persistence of illegal trade forces us to reconsider the idea that winter was a period of isolation and disconnect in the Faroe Islands.

Figure 1. Smuggling vessels leaving the entrepôt by month of departure, 1775–85. Source: DNA, Toldregnskaber, Reviderede Regnskaber, Færøske Toldregnskaber, vols. 14 (1775–85).

Figure 2. Percentage of vessels each year that left the entrepôt during the winter months (October to March), 1775–85. Source: DNA, Toldregnskaber, Reviderede Regnskaber, Færøske Toldregnskaber, vols. 1–4 (1775–85).

Naturally, the continued contraband traffic during the winter also meant that the smuggling vessels – mainly smaller, single-masted boats such as sloops – were particularly exposed to sea hazards. It is no exaggeration to say that the winter was an extremely dangerous period for travelling between the Faroes and northern Scotland. By some estimates, 80 per cent of all shipwrecks in the waters around the Northern Isles occurred between September and March.Footnote 42 As one scholar notes, the crews that headed out into these waters, particularly in smaller vessels, took their lives in their hands.Footnote 43 Reflecting on the dangers of winter, the Orkney-based mariner and smuggler George Eunson noted: ‘What a melancholy scene must it be, to behold mens [sic] bodies floating on the water, or dashed to pieces on the rocks daily, in the winter season, when tempestuous gales of wind, and weather, drive them on shore.’Footnote 44 While Eunson’s claim that this was a daily occurrence is certainly an exaggeration, his account acutely underlines the connection between winter sea travel and danger.

In addition to the conditions at sea, the winter brought darkness. During the darkest months of November, December, and January, the sun rises over the horizon for just five and a half hours a day in the Faroes, and six in Orkney. While the remainder of the day would not have been spent in complete darkness, the winter months would have meant significantly more cloudy than starry skies. In 1771, Niels Ryberg complained to the Danish government about the particular difficulty of navigating into the entrepôt’s harbour, Frederiksvaag, in the winter. Ryberg suggested that the entrepôt should be moved to the sheltered fjord Tronnewaag (now Trongisvágur) on Su∂uroy, the most southern of the Faroe Islands.Footnote 45 However, the government was not interested in moving the entrepôt further away from its own power centre in Torshavn. Instead, a lighthouse was erected on Nólsoy in 1782, to guide vessels safely into Frederiksvaag through the darkness.Footnote 46

Along the Scottish north coast, things were not much better. From 1786 there were increasing efforts to illuminate the coastline, but the expansion of the lighthouse network took time. The first lighthouse to be built on the northern Scottish coast was on North Ronaldsay, Orkney, but only after many vessels had been lost in the preceding years. This was one of the first four lighthouses built on the initiative of the newly established Northern Lighthouses Trust.Footnote 47 It was finally lit in October 1789.Footnote 48 Writing about the waters around North Ronaldsay and Sanday just before the erection of the lighthouse, George Eunson noted that: ‘These are the most dangerous of the northern isles, and often [cause] the destruction of a number of merchant ships, and their cargoes; and what is more dismal, the loss of numbers of poor seamen; particularly North Ronaldshay [sic], and Sanday.’Footnote 49 The coast of Shetland suffered an even slower evolution: the first lighthouse, Sumburgh Head Lighthouse, was not lit until 1821.Footnote 50 Thus, while we can see the emergence of new strategies to attempt to offset some of the risks involved in winter travel in the region, technical advancements such as the erection of lighthouses were sporadic and long in coming. For most of the entrepôt’s active period, there was only a single light in this entire island region to guide the contraband traffic through the winter.

It is difficult to assess just how dangerous this maritime trade could be in winter. Newspapers and Lloyd’s List reveal that many smuggling vessels ran aground or sank in winter storms.Footnote 51 However, it has not been possible to satisfactorily calculate the accident rates, as most lost ships were never recorded in official sources because of the clandestine nature of their trade; official sources can therefore at best only give a partial image of the situation. Another approach is to trace departing vessels and investigate what happened to them. This is also challenging, as it is often necessary to rely on patchy archival evidence.

To illustrate this, let us look at the winter of 1783. In the summer of that year, the Laki volcano on Iceland erupted and it is believed that this contributed to the particularly harsh winter of 1783–4, which around Christmas and New Year culminated in one of the biggest storms to hit the area during the active years of the entrepôt.Footnote 52 Due to some irregularities related to the logging of departures, it is unfortunately not possible to match departures with the exact weather on the days of leaving the islands.Footnote 53 However, we can still make some observations. During the 1783 storm period, three vessels departed during the most critical time, mid-December to early January. The smuggling veteran Allen Morrison’s Bella is recorded as having left the entrepôt on 15 December 1783, David Boag’s Hope departed on 23 December, and Andrew Lusk’s Nancy on 8 January 1784, just as the storm was abating.Footnote 54 We know from an ensuing court case that the Bella was wrecked in the process of getting back to Scotland.Footnote 55 Circumstantial evidence suggests that Boag, who left Fredericksvaag for Orkney shortly before the storm subsided on 25 and 26 December, arrived safely with his cargo in Orkney: the shipping slip that Boag turned over to the Kirkwall-based merchant John Urquhart upon the safe delivery of his cargo of gin and porcelain has been identified among Urquhart’s commercial papers.Footnote 56 Intriguingly, the last ship, the Nancy, captained by Andrew Lusk, actually arrived in the Faroes on 31 December; due to the travel times involved, Lusk must have set out either around the time of the storm or just before it, but there is currently no definitive evidence either way.

Although it must be assumed that most shipmasters waited for good weather and a favourable wind direction, the conditions could alter swiftly.Footnote 57 The dangerous nature of winter smuggling meant that accident rates were probably high, but it is unlikely that we will ever know just how high the loss rate actually was. Despite this, neither winter in general nor any particular storm periods appear to have halted the contraband traffic, which weathered them all.

Just as with legal shipping, where insurance premiums for the North Sea were often much higher during the winter, the dangers involved in the wintertime contraband trade were part of a calculated risk.Footnote 58 At least some of the smuggling sloops appear to have been insured for their winter journeys, but it has not been possible to establish how common this practice was.Footnote 59 When the contraband cargo was insured, the policyholder was usually the purchaser of the goods rather than the entrepôt itself. A crucial reason for this was that it was not necessary for Rosenmeyer & Flores to insure their deliveries. When an order was sent from the entrepôt, the company issued a slip of paper to the captain which he had to sign. It read: ‘I promise to Deliver unto Mr … or Order … Sea Hazard and Seizure only Excepted.’Footnote 60 The captain was responsible for his cargo until it was handed over either to the buyer, to customs, or to the sea floor. In that last case, the responsibility for the cargo was transferred to the buyer, who had to pay the entrepôt. This was contentious, and over the years Rosenmeyer & Flores brought various merchants to court for failing to reimburse them for a lost cargo.

Five such cases have so far been identified in the Scottish Court of Sessions.Footnote 61 We have already encountered one of them, the Bella, captained by Allan Morrison, which sank during a storm on its way back from the Faroes to Sutherland, Ross-shire, and Cromartyshire.Footnote 62 Donald Ross from Culaig, Sutherland, was one of the merchants who had ordered the lost goods. He protested against Rosenmeyer & Flores’s attempts to seek reimbursement. In the court proceedings, we can see that Ross’s representative argued that he had ordered the goods in the spring, but that Rosenmeyer & Flores had not shipped the order until December 1783, a delay that had likely contributed to the disaster: ‘by the delay in shiping [sic] the articles, they were altogether lost, which would not in all probability have happened if they had been shipt [sic] early in the season when the commission was given’.Footnote 63 In most of these cases, Rosenmeyer & Flores eventually received their outstanding payments. The financial risk was therefore not taken by the entrepôt itself but was transferred to the purchaser, who in turn could choose to insure the cargo. This helps to explain why Rosenmeyer & Flores were still trading in the winter. But if we are to understand how this situation could continue, we must look more closely at the people who took the physical risks: the captain and his crew.

III

So, who were these mariners who were willing to risk their lives? We have already met a few of them in passing, but it is worthwhile considering their demographics in more depth. During the period from 1775 to 1785, 78 out of 111 ships (74 per cent) of the winter traffic to leave the Faroes had arrived from Scotland. Three main locations emerge from the data: 12 vessels came from Orkney (11 per cent of the total traffic in winter), 11 from Shetland (10 per cent), and 12 from the Outer Hebrides (11 per cent), together making up 32 per cent of the total winter trade. By comparison, Inverness, Aberdeen, and Glasgow together made up just 6 per cent of the winter traffic. Unfortunately, over time there was an increased tendency to record ‘Scotland’ for these vessels rather than their specific ports of origin, so these numbers are therefore minimum estimates. During the summer season, the dominance of sailors from northern Scotland is slightly less pronounced, as Scotland accounted for a slightly smaller share of the total trade during the summer. Out of 210 vessels that arrived at the entrepôt during the summers between 1775 and 1785, Scotland was the origin for 134 (64 per cent). The comparative dominance, particularly during the winter, of these northern insular regions with small populations becomes even more pronounced if we add the north-facing coastlines of Moray, Caithness, and Sutherland. For a breakdown of all Scottish traffic in winter and summer, see Figure 3.

Figure 3. Originating ports of 217 Scottish ships that arrived at the entrepôt, 1775–85. Source: DNA, Toldregnskaber, Reviderede Regnskaber, Færøske Toldregnskaber, vols. 1–4 (1775–85).

In order to understand the predominance of captains from these insular regions in the Faroese trade, as well as the relentlessness of illegal trade, it is worth having a look at the local fishing industry. During the eighteenth century, dangerous maritime winter activities had become a normalized, although not uncontroversial, facet of insular life in the area, as evidenced by the extremely hazardous practice of open-boat haaf (deep-sea) fishing in Shetland.Footnote 64 Dangerous fishing habits might not only have contributed to a generally increased acceptance of the dangers of the winter sea; we can also see a direct overlap between fishing and the contraband trade. In an anonymous letter from the Isle of Lewis dated 21 March 1786, the author mentions that ‘under the cover of fishing [contraband vessels] supply themselves & neighbours on the mainland coast from Cape Wrath to Ardnamurchan with Spirits and liquors imported or of their own distilling; likewise with vast quantities of Tobacco, tea, Coffee &ct &ct from Farro [Faroes] Belfast &ct &ct’.Footnote 65

This practice is also discernible in the customs office letter books from Stornoway, in which we can see that the same ships and shipmasters that were frequenting the Faroese entrepôt were also involved in fishing and in the related cabotage of salt. This includes one of the most prolific Stornoway smugglers, Norman Morrison, and his sloop Bella, who visited the entrepôt ten times between 1776 and 1785.Footnote 66 Indeed, nearly all vessels that were involved in the contraband trade between the Faroe Islands and the Isle of Lewis were ‘herring buses’ engaged in the local herring fishing business and in transporting salt.Footnote 67 Just like the Shetlandic haaf fishing, the herring fishing business was to a large extent conducted during the winter months and usually in the middle of the night, due to the movement patterns of the fish.Footnote 68

While there were concerns that navigation was ‘so uncertain and dangerous at this Season of the Year’, it is clear that this did not stop the fishing, just as it did not put an end to the contraband trade.Footnote 69 It is likely that, as a result of the close and sometimes overlapping relationship between the contraband and fishing trades, the normalization of sea hazard that derived from the fishing trade spilled over into contraband shipping too. The reason why the local population engaged in such dangerous economic activities was largely related to poverty and, crucially, a lack of alternative employment.Footnote 70 Poverty motivated fishermen to go to sea in the darkness of winter, and it was most likely also a contributing factor to the continuation of the local contraband trade.

The prospective gains would certainly have been enticing. Contemporary letters and account books reveal that the Faroese contraband trade could be highly profitable for the sailors engaged in it.Footnote 71 The renumeration corresponds fairly well to the estimates of the commander of the customs cutter Royal George, who calculated that a crew of seven would receive £63 for a successful run, and a crew of twelve would earn £108.Footnote 72 The customs officers in Glasgow had intelligence indicating that the renumeration for a successful run was £5 for a regular sailor and £10 for the master or captain.Footnote 73 This would have been a considerable sum for a fairly short journey. By comparison, a sailor in the Norway trade setting out from London or Hull would receive between £1 10s. and £1 15s. per journey, depending on the season.Footnote 74

Nor did these shipmasters lack agency. At least some of the captains were actively involved in proposing trade. In late September 1781, Captain Andrew Irving, a kelp transporter and habitual smuggler, wrote a letter from Stornoway to the merchant Alexander Bain in Wick, Caithness. Irving suggested that Bain might partake in a smuggling endeavour ‘if you thought proper your Self to Rusque [sic] the Sloop to N[orth] F[aro]’. ‘Fishing’ for a vessel, Irving proposed that Bain could join several other merchants in Tain and Cromarty to make up the freight from the Faroe Islands to Loch Laxford, in a remote part of north-west Sutherland.Footnote 75 Furthermore, there is evidence to suggest that the contraband freight rates, and hence the skipper’s potential profit, increased during the winter months. In correspondence with the Lerwick merchant William Watt Junior, the merchant Robert Sandison of Stromness complained that the season was so advanced that his delivery was bound to become very expensive.Footnote 76 Shortly thereafter, on 7 November 1776, Sandison received his cargo, noting with resignation: ‘There was no helping for it with regard the rude night coming on them but it makes the freight come dear.’Footnote 77 Winter deliveries could then be as expensive for the merchants as they were lucrative for the risk-taking mariners.

Looking at mariners in the East India trade, Beverly Lemire has suggested that the dangers involved in shipping encouraged mariners through a specific type of moral economy, whereby the risks they took at sea entitled them to reap the benefits of their extra-legal trade.Footnote 78 Considering the multiple engagements of these northern sailors, it is possible that a similar type of attitude developed among the coastal population in the Northern Isles, but more research remains to be done to gain better insight into this local economy. Nevertheless, it seems likely that the renumeration for running contraband might well have played a significant role in the makeshift economy of the coastal population.

IV

The local expertise that these insular sailors brought with them doubtlessly played a critical role in enabling the contraband trade in global wares to continue throughout the seasons. This becomes clear if we look more closely at some winter smuggling endeavours in the area. To get a better understanding of the specifics of such cases, Murdoch Mackenzie’s 1750 hydrological charts of Orkney and the Isle of Lewis will serve as a guide.Footnote 79 Mackenzie’s maps were inspired by the inherent dangers involved in travelling through these treacherous waters. While the charts unlocked covert local knowledge for a larger audience beyond the islands, a sea chart alone was not enough to guarantee a safe journey across this oceanic space, and the dangers were still acute, particularly for outsiders who were not used to navigating these waters during winter.

On Monday, 30 October 1780, a sloop captained by a man called Paterson, who does not appear to have been a regular visitor to the entrepôt, was wrecked in the Bay of Cleat on Westray, in north-west Orkney.Footnote 80 A cargo of 462 ankers of brandy was later salvaged from the rocky shore, but the crew, consisting of seven men, were all lost. The only survivor was a passenger named Taylor, who had been taken ashore just before the storm broke.Footnote 81 While the Bay of Cleat is marked as a safe anchorage on Mackenzie’s map, he specifies in his description of the bay that this was a safe place to anchor in the summer; in the middle of a winter storm, even such safe havens could quickly turn treacherous.Footnote 82

Safe anchorages could also turn into traps in themselves. The Friendship of Kintyre, captained by John Brown, sailed from the Faroes in the middle of December 1781. A few days after departing, Brown found himself forced to seek shelter at Pan Hope on the island of Flotta in southern Orkney, due to contrary winds.Footnote 83 On Mackenzie’s map this bay is marked with an anchor cross, demarcating a particularly safe place to anchor, and he notes: ‘Pan-hope in Flota [sic] is clean Ground and abundantly safe.’Footnote 84 However, it also lies just at the entrance to the dangerous Pentland Firth, which should only be traversed in fair weather. Trapped by contrary winds and bad weather, Brown was stuck at Pan Hope for several weeks before the customs cutter Experiment, captained by Alexander Cooke, eventually seized the Friendship of Kintyre on 24 January 1782, bringing Brown and his crew back to Aberdeen to stand trial.Footnote 85

As these examples show, dynamic maritime skills were desirable when it came to navigating the winter waters. Locally rooted knowledge could provide a crucial edge in identifying and choosing appropriate routes and anchoring spots. We can also see this when it comes to the offloading of the ships. Due to the clandestine nature of the trade, smuggling vessels could not use regular harbours with defences reinforced against the elements to offload their goods, but instead had to hover near the coastline, risking being smashed against rocky shores or wrecked on invisible skerries.Footnote 86 Here, too, we find that familiarity with the local geography was important. By comparing recorded cases of contraband landing ashore to Mackenzie’s map, we can see that the sites were strategically chosen. When Captain David Boag, a native of Orkney, was to deliver goods to Kirkwall, the buyer Alexander Logie asked Boag to offload his cargo in one of the sandy coves near Carness just outside of Kirkwall, or leave them with a farmer in Eva (Evie) on the north side of the island of Pomona, from where they could later be brought into Kirkwall at a time when no ‘king’s vessels’ were around.Footnote 87 The subsequent offloading and transporting overland were a similarly local affair, where knowledge of the land was crucial to success.Footnote 88

Even when disaster was unavoidable, a familiarity with the coastline was an advantage. Wrecking was not uncommon, and several of the shipmasters sailing to the Faroes, including George Eunson himself, are known to have foundered at some point. More often than not, they returned to the contraband trade with new vessels.Footnote 89 It is very possible that these seasoned mariners strategically engaged in the practice of safe-wrecking – wrecking their ships in relatively ‘safe’ locations where they could get ashore unharmed and later retrieve the cargo – but it is difficult to prove this with any certainty. It is nevertheless likely that when Captain William Gray, one of the most active shipmasters in the Faroese contraband trade, wrecked his ship on a beach in Sutherland during a storm in December 1781, he did so strategically. Gray managed to salvage most of the cargo and later attempted to sell it for private profit, although this ultimately proved unsuccessful.Footnote 90 In situations of acute distress at sea, being able to navigate into a ‘safe’ wreckage location could prove the difference between life and death. With enough skill, knowledge, and luck, a sinking might therefore not be too disastrous. Veteran mariners accustomed to these challenging conditions would have been well aware of the risks they were taking, but they would also have had tools to deal with them.

Local knowledge was flexible and multi-layered. It could help mariners identify the best naval routes to cope with various conditions as well as suitable offloading grounds. And when their vessels were in distress, such knowledge could also help them to navigate to appropriate shelter and, in a crisis, aid them in recognizing ‘safe’ wrecking sites.

In addition to the higher renumeration, there were definite benefits to be had for those able to navigate the winter sea. One significant one was the reduced risk of being seized by customs cutters, which, just like much other regular traffic, were not set up for winter sailing. Writing to the Scottish Customs Board in 1774, William Ferguson of Peterhead argued for the need to invest in two ‘Harwich smacks that can keep the Sea [in] the Strongest gale of Winter’ to hamper the illegal trade from the Faroes.Footnote 91 Ferguson’s request reflects a warranted concern. The relentless continuance of the Faroese trade, which would only intensify in the subsequent years, was not matched by any sustained customs surveillance throughout all of the seasons. Customs cutters do not appear to have been a prioritized expenditure, particularly along the northern Scottish coast. By 1779, there were two cutters patrolling the west coast: the Prince of Wales and the Combrais, and three on the east coast.Footnote 92 Nine years later, the situation looked roughly the same: four official customs cutters reported to the Scottish Customs in 1788: the Royal George, the Prince Edward, the Prince Augustus Frederick, and the Princess Royal.Footnote 93 Needless to say, this fleet was not able to patrol the entirety of the Scottish coast, and the north coast, along with the Northern Isles, appears to have been especially neglected. Instead, areas with particularly rife smuggling were targeted in temporary operations. One such took place in February 1783, when the Combrais cutter was ordered ‘to cruize for three months, between the Island of Mull and Stromness, & to rendezvous at Stornoway for the Suppression of Smuggling’.Footnote 94 While there was a constant risk of being seized by a customs cutter, the risk was less prevalent during the winter months as there were few customs cutters that could handle the winter seas.

Harsh and contrary winds could also be used to explain away suspicious activity in the Northern Isles. Captains who were seized could argue that they were actually going somewhere completely different, usually Bergen or Madeira, when they were caught in a storm or trapped by bad winds just off the Scottish coast. The very dangers of winter could then be used as a defence in a court of law to deflect the crime. In one such instance, the shipmaster Peter Copland argued that he was sailing from Rotterdam to the Faroe Islands but had been trapped in Orkney after a storm damaged the hull of his vessel, causing it to take in water. Copland, who had been apprehended by the customs cutter Osnabrugh for breach of the Hovering Acts, was subsequently imprisoned for smuggling in the tolbooth in Kirkwall, from where he petitioned for his release.Footnote 95 It is not evident how convincing the Court of Exchequer found these claims about alternative destinations, but they would certainly have been a complication. To this day, we cannot establish with certainty the destination of certain vessels nor their intention to smuggle, due to the use of such dummy destinations.

Finally, we have the darkness, which may have made sea travel more challenging but would also have provided shelter for illegal activity. Writing in June 1775, the Stromness-based merchant David Rich complained that he could not possibly help with a proposed contraband mission as ‘[the revenue officers] are Such a Great Look out & the nights So Short’.Footnote 96 By contrast, the long nights of winter offered opportunity to keep both unloading and overland transport out of sight.Footnote 97 Even though the winter was, and remains, a perilous period to sail between the Faroe Islands and northern Scotland, there were certainly benefits to be had for the sailors willing to head out onto the winter ocean, and plenty of them evidently chose to do so.

V

This article has explored a part of Europe – the Faroes and the northern reaches of Scotland – that has often been sidelined in discussions about the process of globalization, but which during the eighteenth century fed on the jugular vein of global commerce. While their very insularity helps to explain why islands like the Faroes could emerge as centres for illegal trade, being able to harness local skill and knowledge was key to overcoming the seasonal obstacles to this trade. The trade continued all year round, not merely because it had to contend with other concerns than legal trade, such as the desire to keep out of sight of the customs cutters, but precisely because it was conducted by locals who neither could, nor desired, to disengage from their own lived environments for months on end. Expertise in handling the shifting seasonal conditions was an integral part of seamanship in the north Atlantic. The seasonality of maritime illegal trade in this area bears similarities to illegal trade in other northern European areas, such as parts of Scandinavia, where we find that local cross-border autonomy increased during the winter, while outsider interference and top-down control mechanisms generally weakened.

In uncovering the illegal interconnectivity between the Faroes and Scotland, we must also reconsider the state of insular entanglement within this area. More than a barely remembered feature of the old Norwegian thalassocracy, this northern insular connectivity was, during the second half of the eighteenth century, vibrant and much more frequent than the ‘internal’ connections between the Faroes and Denmark. What has emerged from this article is a dynamic, trans-territorial island region, where mariners generated a thriving perennial contraband trade, drawing on local skill and knowledge. By harnessing these forms of knowledge, the Danish contraband trade was able to expand and continue throughout the seasons, benefiting the Faroese entrepôt and, arguably, the local coastal economy.

To understand how illegal trade could flourish in the eighteenth century, it is necessary to consider illegal trade from the margins. This perspective demonstrates that trade was a highly rooted phenomenon, often shaped by regional specificities. Local agency, knowledge, and environment were all important, yet often overlooked, features of illegal trade, offering crucial insights into the dynamics of the black market and its role in the early modern world.

Acknowledgements

This article has been through several reincarnations on its road to completion. A heartfelt thank you needs to be expressed to all the readers who helped me realize the final version, in particular Paul Warde, Fredrik Charpentier Ljungqvist, Hanna Hodacs, Leos Müller, Margaret Hunt, Gustav Ängby, and the anonymous reviewers who have given me such insightful and helpful comments along the way and helped me to reimagine this article. I also want to extend a thank you to Jeremy Lowe for his painstaking work with proofreading the final text. I take full responsibility for the scientific content of the article and any errors that may occur. And, finally, a very special thank you to my old friend Jimmy Chadda, who helped me get hold of some precious archival documents during the pandemic when many doors appeared to be closed to me.

Funding statement

The work on this article has been generously funded by the Swedish Research Council (2020-00756), as well as the Åke Wiberg Foundation (H23-0280), the Lars Hierta Memorial Foundation, and the Sven and Dagmar Salén Foundation.

Competing interests

The author declares none.

References

1 Lucas Debes, Færoœ, et Fœroa reserata: that is a description of the islands & inhabitants of Foeroe: being seventeen islands subject to the king of Denmark (London, 1676; originally published in Danish in 1673), p. 37. The book contained the first illustrated map of the Faroe Islands.

2 N. Andersen, Færøerne, 1600–1709 (The Faroe Islands, 1600–1709) (Copenhagen, 1895), pp. 14; Peter Korsgaard, ‘Den kongelige enehandel på Færøerne 17091776’ (‘The royal monopoly in the Faroe Islands, 1709–1776’), Erhvervshistorisk Årbog (Business History Yearbook), 31 (Aarhus, 1982), pp. 7–51, at p. 7.

3 David Armitage, ‘Three concepts of Atlantic history’, in D. Armitage and M. Braddick, eds., The British Atlantic world, 1500–1800 (New York, NY, 2002), pp. 212.

4 By contrast, entanglements between the islands have featured much more in research on the middle ages and the early modern period: see, for example, Louis Sicking and D. Abreu-Ferreira, eds., Beyond the catch (London, 2008); P. Holm et al., ‘The North Atlantic fish revolution (ca. ad 1500)’, Quaternary Research, 108 (2019), pp. 92–106; as well as the ongoing ‘Looking in from the Edge’ project (LIFTE), run by the University of the Highlands and Islands Archaeology Institute, the University of Lincoln, and the German Maritime Museum.

5 See, for example, Philip J. Stern and Carl Wennerlind, eds., Mercantilism reimagined: political economy in early modern Britain and its empire (Oxford, 2013).

6 Gunvor Simonsen and R. Christensen, ‘Together in a small boat: slavery’s fugitives in the Lesser Antilles’, William and Mary Quarterly, 80 (2023), pp. 611–46.

7 On these areas as peripheries, see Edvard Holm, Danmark-Norges indre historie under enevælden fra 1660 til 1720 (The domestic history of Denmark-Norway during the absolute monarchy, 1660–1720) (Copenhagen, 18856), p. 426; Korsgaard, ‘Den kongelige enehandel på Færøerne’, p. 18; Fernand Braudel, Civilization and capitalism, 15th–18th century, trans. Siân Reynolds (3 vols., London, 1984), iii, p. 365; Julian Hoppit, The dreadful monster and its poor relations (London, 2021), pp. 801.

8 Beverly Lemire, Global trade and the transformation of consumer cultures (Cambridge, 2018), particularly ch. 4. For some European examples, see, for example, Gregory Stevens Cox, St Peter Port, 1680–1830: the history of an international entrepôt (London, 1999); David Chan Smith, ‘Creative friction, legal pluralism and the eighteenth-century smuggling economy in the Channel Islands’, in Lyndsay Campbell and Shaunnagh Dorsett, eds., Legal histories of empire: navigating legalities (London, 2025), pp. 213–39.

9 Hoppit, Dreadful monster, pp. 801.

10 A selection of research about tea smuggling to Scotland: Philipp Robinson Rössner, Scottish trade in the wake of the Union, 1700 1760: the rise of a warehouse economy (Stuttgart, 2008), ch. 2.2.4; Derek Janes, ‘Fine Gottenburgh teas: the import and distribution of smuggled tea in Scotland and the north of England c. 17501780’, History of Retailing and Consumption, 2 (2016), pp. 223–38, at pp. 2245; Hanna Hodacs, ‘Keeping it in the family: the Swedish East India Company and the Irvine family, 1731 to 1770’, Journal of World History, 31, special issue, Crossing companies: mobility and cooperation between early modern national monopoly trading companies, ed. Philip Stern and Felicia Gottmann (2020), pp. 567–96.

11 Janes, ‘Fine Gottenburgh teas’, pp. 2245; Klas Rönnbäck and Leos Müller, ‘Swedish East India trade in a value-added analysis, c. 1730–1800’, Scandinavian Economic History Review, 70 (2020), pp. 118; Hodacs, ‘Keeping it in the family’; Meike von Brescius, Private enterprise and the China trade: merchants and markets in Europe, 1700–1750 (Leiden, 2022), ch. 2.

12 Anna Knutsson, Shadow economies in a globalizing world (Abingdon, 2023), ch. 3.

13 Leos Müller, ‘Under svensk flagg i Medelhavet: algeriska sjöpass 17701800’ (‘Under the Swedish flag in the Mediterranean: Algerian sea passports, 1770–1800’), in Marie Lennersand and Leos Müller, eds., Från Afrikakompaniet till Tokyo (From the Africa Company to Tokyo) (Knivsta, 2017); Milja van Tielhof and Jan Luiten van Zanden, ‘Productivity changes in shipping in the Dutch Republic: the evidence from freight rates, 15501800’, in Richard W. Unger, ed., Shipping and economic growth, 1350–1850 (Leiden, 2011), p. 54; Yrjö Kaukiainen, ‘At the far end of oceanic seaways: St. Petersburg shipping in the eighteenth century’, International Journal of Maritime History, 33 (2021), pp. 489–508, at p. 494.

14 Peregrine Horden and Nicholas Purcell, The corrupting sea: a study of Mediterranean history (Oxford, 2000), pp. 1423; also David Abulafia, ‘The problem of the kingdom of Majorca (1229/76–1343) 2: economic identity’, Mediterranean Historical Review, 6, (1991), pp. 3561, at p. 46.

15 Müller, ‘Under svensk flagg i Medelhavet’; see also Dagomar Degroot, The frigid golden age: climate change, the little ice age, and the Dutch Republic, 1560–1720 (Cambridge, 2018), pp. 11130.

16 Degroot, Frigid golden age, pp. 3034.

17 Alexandra Sanmark and Shane McLeod, ‘Norse navigation in the Northern Isles’, Journal of the North Atlantic, 44 (2024), pp. 1–26, at p. 5; Renaud Morieux, The Channel: England, France and the construction of a maritime border in the eighteenth century (Cambridge, 2016), pp. 98100.

18 Morieux, Channel, pp. 98104; Sara Caputo, Tracks on the ocean: a history of trailblazing, maps and maritime travel (London, 2024), pp. 389.

19 Caputo, Tracks on the ocean, p. 215.

20 Emmanuel Le Roy Ladurie, Histoire du climat depuis l’an mil (Paris, 1967).

21 Horden and Purcell, Corrupting sea, pp. 33940.

22 Degroot, Frigid golden age, p. 302.

23 Ship speed in fair weather would have been roughly 4 knots at that time. For ship speeds in the eighteenth century, see Morgan Kelly and Cormac Ó Gráda, ‘Speed under sail, 1750–1850’, Centre for Economic Research Working Paper Series, no. WP14/10, School of Economics, University College Dublin, 2014, available at https://www.econstor.eu/bitstream/10419/109720/1/786607629.pdf.

24 Philip E. Steinberg, The social construction of the ocean (Cambridge, 2001), p. 70.

25 Debes, Færoœ, p. 34.

26 Ibid., p. 46.

27 Erik Gøbel, ‘Asiatisk Kompagnis Kinafarter 17321772: sejlruter og sejltider’ (‘The Asiatic Company’s China voyages, 1732–1772: sailing routes and sailing times’), M/S Museet for Søfarts årbog (Yearbook of the M/S Maritime Museum), 37 (1978), pp. 7–46, at pp. 1617.

28 Murdoch Mackenzie, Orcades; or a geographic and hydrographic survey of the Orkney and Lewis islands, in eight maps (London, 1750), pp. 13.

29 Incoming ships are listed in Danish National Archives (DNA), Toldregnskaber (Customs accounts), Reviderede Regnskaber (Audited accounts), Færøske Toldregnskaber (Faroese customs accounts), 571, vols. 14 (177585).

30 Louis Sicking, ‘The dichotomy of insularity: islands between isolation and connectivity in medieval and early modern Europe, and beyond’, International Journal of Maritime History, 26 (2014), pp. 494–511, at pp. 4945.

31 Barbara E. Crawford, The northern earldoms: Orkney and Caithness from ad 870 to 1470 (Edinburgh, 2013), pp. 215; Steinar Imsen, ‘Introduction’, in Steinar Imsen, ed., Rex insularum: the king of Norway and his ‘Skattlands’ as a political system c. 1260–c. 1450 (Bergen, 2014), pp. 1314.

32 Sanmark and McLeod, ‘Norse navigation in the Northern Isles’.

33 Ibid., pp. 1518, 20; Jeemsie Laurenson, ‘Part VI’, in Charlie Simpson, ed., Shetland’s open boat days (Shetland, 2017), pp. 713.

34 Laurenson, ‘Part VI’, pp. 71–3. On the survival of Norse nautical terminology, see, for example, Ian Fraser, ‘Norse and Gaelic costal terminology in the Western Isles’, Northern Studies, 11 (1978), pp. 3–16.

35 Fraser, ‘Norse and Gaelic coastal terminology’, p. 6.

36 David M. Ferguson, Shipwrecks of Orkney, Shetland and the Pentland Firth (Newton Abbot, 1988), pp. 2943.

37 Four individuals appear to have been involved in the deliveries to the entrepôt: Christian Jörgensen Beck (active 17708), Peter Hansen (active 177489), Peter Paulsen (active 17825), and Jens Hansen Wennersted (active 17824). See DNA, Færøske Toldregnskaber, vols. 1–4.

38 This numerical data comes from the entrepôt’s own tax records, DNA, Færøske Toldregnskaber, vols. 24 (1775–85). It is possible to also extract some information from the Skansenprotokollir (Skansen Protocols), 17829, Faroese National Archive (FNA), but, as this is not recorded in the same manner, it is not well suited for this type of statistical comparison.

39 M. G. Pearson, ‘Snowstorms in Scotland, 1782–1786’, Weather, 28 (1973), pp. 195201; John Kington, The weather of the 1780s over Europe (Cambridge, 1988); Hubert Lamb and Knud Frydendahl, Storms of the North Sea, British Isles and northwest Europe (Cambridge, 2005). See also weather data from Skansen in Torshavn, which supports these general tendencies. FNA, Skansenprotokollir, 17825 and 17859.

40 For a longer discussion on why and how the contraband entrepôt was run, see Anna Knutsson, ‘The rise, fall, and profitability of illegal trade at a Faroese entrepôt, 1767–1787’ (forthcoming).

41 For food, see, for example, National Records of Scotland, Edinburgh (NRS), E382/62; for travellers, see, for example, FNA, Skansenprotokollir, 17825 and 17869.

42 Ferguson, Shipwrecks of Orkney, appendix C.

43 Ibid., p. 25.

44 George Eunson, The ancient and present state of Orkney, particularly the capital borough of Kirkwall (Newcastle upon Tyne, 1788), pp. 48–9.

45 DNA, Rentekammeret (Treasury), 303, Journaler for Færøerne og Grønland (Journals for the Faroe Islands and Greenland), 1771–1805, no. 63, 7 Apr. 1772; J. C. Svabo, Indberetninger fra en reise i Færøe, 1781 og 1782 (Reports from a journey to the Faroe Islands, 1781–2) (Copenhagen, 1959), p. 296.

46 Svabo, Indberetninger, p. 299.

47 ‘North Ronaldsay’, Northern Lighthouse Board, https://www.nlb.org.uk/lighthouses/north-ronaldsay/.

48 Ferguson, Shipwrecks of Orkney, pp. 25–8.

49 Eunson, Ancient and present state of Orkney, p. 48.

50 ‘Sumburgh Head’, Northern Lighthouse Board, https://www.nlb.org.uk/lighthouses/sumburgh-head/.

51 See, for example, Caledonian Mercury, 15 Feb. 1775; Hibernian Journal; or, Chronicle of Liberty, 5 Jan. 1776; Lloyd’s List, 9 May 1777, supplement 858; Caledonian Mercury, 28 Feb. 1778; Lloyd’s List, 17 Nov. 1780, no. 1216; Caledonian Mercury, 25 Nov. 1780.

52 For the weather in 1783, see Lamb and Frydendahl, Historic storms, pp. 87–8; Kington, Weather of the 1780s, esp. chs. 4 and 5. For more on the Laki eruption and its consequences, see Geoffrey Hellman, ‘The Laki volcanic eruption of 1783–1784: a reappraisal and reinterpretation of the consequences of the event in Europe. Villain or fall guy?’ (PhD thesis, Rennes 2, 2021).

53 The logbook of the entrepôt does not fully correspond to the logbook held at Skansen; while the discrepancy is not very significant – usually a date might differ by only one to four days – it is not possible to know on exactly what day a vessel actually departed from the islands.

54 DNA, Færøske Toldregnskaber, vol. 4, Dec. 1783.

55 NRS, CS238/R/3/49.

56 Orkney Archive, Kirkwall (OA), D8/1; OA, D14/1/11.

57 See ships leaving the Faroes and records of the daily weather in FNA, Skansenprotokollir, 1782–5, vols. 31–2.

58 Markus A. Denzel, The Hamburg marine insurance, 1736–1859, trans. F. Streng (Leiden, 2022), ch. 5. Unfortunately, although not surprisingly, there are no insurance estimates for the Faroe Islands, but see pp. 132 and 156 for data for the seasonal fluctuation in relation to other ports in the northern part of the North Sea, in Norway and in the British Isles.

59 Assignation Rosenmyre [sic] Flor & Co. to Geo. Andrew, 13 Jan. 1786, NRS, GD237/21/60; James Hay to William Gray, 5 Oct. 1778, Shetland Archives, Lerwick (SA), SC12/53/5, fo. 36v.

60 NRS, CS238/R/4/22, attached cargo slip.

61 NRS, CS238/R/3/49, CS238/R/3/51, CS238/R/4/22, CS238/R/4/44, and GD237/21/60 (the last contains a reference to a case at the Court of Sessions that I have not yet been able to find in the session papers).

62 NRS, CS238/R/3/49.

63 ‘Unto the right honourable Lord Stonefield, the representatives of Donald Ross at Culaig’, NRS, CS238/R/3/49, p. 8.

64 James R. Coull, ‘The haaf fishery of the Shetland Islands: an inevitable method of organising fishing in the eighteenth and nineteenth centuries?’, Northern Scotland, 24 (2004), pp. 53–73, at pp. 535.

65 NRS, GD46/6/101.

66 NRS, CE86/2/3, 31 Dec. 1777 (Sibylla and Hopewell); 1 Apr. 1778 (Sibylla); 23 May 1778 (Bella); 3 Jan. 1784 (Charming Keaty).

67 NRS, CE86/2/3, 23 May 1778 and 3 Jan. 1784. Norman Morrison’s visits to the entrepôt occurred on 17 May 1776, 6 Sept. 1776, 12 Sept. 1777, 6 Apr. 1778, 16 May 1778, 9 Nov. 1778, 12 Mar. 1779, 13 Feb. 1781, 30 May 1782. DNA, Færøske Toldregnskaber, vols. 1–4.

68 Lochbroom, County of Ross and Cromarty, in John Sinclair, The statistical account of Scotland, vol. 10 (Edinburgh, 1794), p. 465.

69 NRS, CE86/2/3, 30 Dec. 1780.

70 Coull, ‘Haaf fishery’, p. 56; James R. Coull, ‘The development of herring fishing in the Outer Hebrides’, International Journal of Maritime History, 15 (2003), pp. 21–42, at p. 27.

71 NRS, E382/62, Charter Party between Rosenmeyer & Flores Co. and William Lovie of the Friendship of Elgin, 1 May 1782; OA, Margaret accounts, D3/188.

72 The National Archives, Kew, UK (TNA), T1/644, Royal George cutter, 28 Dec. 1787.

73 TNA, T1/644, Glasgow, 8 Jan. 1788.

74 Ralph Davis, The rise of English shipping in the seventeenth and eighteenth centuries (Liverpool, 2012), p. 128.

75 Captain Andrew Irving to Alexander Bain (merchant at Caithness), 17 Sept. 1781, Stornoway, OA, D3/370. There is a similar case from 1783 involving one Captain Laurence Calder, but he was proposing a contraband journey to Hamburg: see Laurence Calder to William Hay, 20 Aug. 1783, SA, D40/9/31.

76 Robert Sandison to William Watt Jr, 28 Oct. 1776, OA, D3/85 (quoted in Francis Wilkins, The smuggling story of the northern shores (Wyre Forest, 1995), pp. 389).

77 Sandison to Watt, 7 Nov. 1776, OA, D3/85 (quoted in Wilkins, Smuggling story, pp. 38–9).

78 Beverly Lemire, ‘Monopoly claims and moral economy’, in Veronika Hyden-Hanscho and Werner Stangl, eds., Formative modernities in the early modern Atlantic and beyond (London, 2023), pp. 145–70.

79 Mackenzie, Orcades. Mackenzie was in part building on an earlier mapping effort by Greenville Collins, Great Britain’s coasting pilot (London, 1693).

80 No one by the name Paterson appears in the entrepôt’s records between 1775 and 1785.

81 Caledonian Mercury, 25 Nov. 1780.

82 Mackenzie, Orcades, p. 13 and map iv.

83 Caledonian Mercury, 2 Dec. 1782.

84 Mackenzie, Orcades, p. 8, map ii.

85 Caledonian Mercury, 2 Dec. 1782.

86 TNA, T1/664, Customs Letters – Collectors to Board, Prince Royal cutter, 26 Dec. 1787.

87 From Alexander Logie’s letter book, quoted in W. R. Scott, ‘The trade of Orkney at the end of the eighteenth century’, Scottish Historical Review, 10 (1913), pp. 360–8, at pp. 3645. This letter book has unfortunately been lost since 1913 but I am still working on tracking it down.

88 See, for example, expenses paid to locals for unloading, piloting, and onward transport: letter of 18 July 1779, OA, D3/104; OA, D3/209, expenses, Nov. 1780.

89 For example, after the wreck of William Gray’s sloop Elizabeth in 1781, Gray returned another five times to the entrepôt in five different vessels: 29 Mar. 1782, 11 July 1782, 17 Aug. 1782, 16 Sept. 1782, and 7 Nov. 1783. After losing his sloop Bella in Dec. 1783, Allen Morrison returned in July 1784 aboard the Charming Keaty. DNA, Færøske Toldregnskaber, vols. 34. For Eunson, see OA, D3/199.

90 Assignation Rosenmyre [sic] Flor & Co. to Geo. Andrew, 13 Jan. 1786, NRS, GD237/21/60.

91 TNA, T1/511.

92 NRS, CE1/16/1–3, 29 Sept. 1779, fo. 205.

93 TNA, T1/664, fos. 21, 23, 26, 27.

94 NRS, CE86/2/3, 13 Feb. 1783. The Prince of Wales cutter was also ordered on a similar ‘cruize’: ibid., 15 May 1783.

95 NRS, E382/84/7. For a similar case concerning John Brown and the Friendship, see Caledonian Mercury, 2 Dec. 1782.

96 David Rich to William Watt Jr, 9 June 1775, OA, D3/209.

97 We find that many of the accounts concerning overland smuggling also date from the winter months. See, for example, OA, D3/209, expenses, Nov. 1780.

Figure 0

Figure 1. Smuggling vessels leaving the entrepôt by month of departure, 1775–85.Source: DNA, Toldregnskaber, Reviderede Regnskaber, Færøske Toldregnskaber, vols. 14 (1775–85).

Figure 1

Figure 2. Percentage of vessels each year that left the entrepôt during the winter months (October to March), 1775–85.Source: DNA, Toldregnskaber, Reviderede Regnskaber, Færøske Toldregnskaber, vols. 1–4 (1775–85).

Figure 2

Figure 3. Originating ports of 217 Scottish ships that arrived at the entrepôt, 1775–85.Source: DNA, Toldregnskaber, Reviderede Regnskaber, Færøske Toldregnskaber, vols. 1–4 (1775–85).