Shared burdens: The General Average adjustment of the Jan Maria

The Jan Maria was a Dutch schooner that, in 1883, en route from the Baltic to its home port, was forced to jettison cargo during a fierce storm, which resulted in a General Average (GA) procedure. GA refers to a method that redistributes damages that were deliberately inflicted by the master, in order to save the ship and its cargo, among all those parties that benefited from the action. The report of this procedure of the Jan Maria has been preserved and is exceptional in its completeness. It offers a unique view on a complex procedure and also on the coping mechanisms of small-scale entrepreneurs in a volatile business environment. In this article, I will explain the principle and functioning of GA, and I will argue that, in spite of financial innovations like marine insurance, GA remained an important part of risk management.

Average' (GA) procedure was set in motion. 1 This procedure redistributes the costs among all parties involved in very specific cases of damages incurred during mercantile shipping. The origin of the term 'GA' is not clear and its true origins will probably never be known. Some scholars argue that the term is derived from the French word for port ('havre'), others claim it comes from the Greek word for cargo or from the Arabic áwâr (damage). 2 GA can only be applied if damages are deliberately inflicted by the master and the crew with the express objective of salvaging ship, lives and cargo. The resulting damages are then borne by all parties benefiting from the acts. So, for example, if a master decides to jettison a part of the cargo to lighten the ship to outrun pirates, or he orders the main mast to be cut during a storm, it is not just the owner of the jettisoned cargo or the shipowner who bear these costs, as the damages are borne by all parties involved. GA is a concept based on mutuality, on sharing burdens: sharing the risks of long-distance maritime trade. 3 Although GA is little known outside the world of trade and transport, it is a concept that had been in use for centuries before the Jan Maria's unfortunate journey took place. 4 What's more, it is still in use today. 5 In spite of the emergence of other methods of risk management, GA has not been supplanted, not even by marine insurance. 6 Marine insurances did not exist when GA emerged, but since early modern times have been commonly used to transfer the risks of long-distance trade. Even though marine insurance developed into a profitable and efficient market where many risks and virtually all sums 7. For example, the Amsterdam marine insurance market was known as the market where all risks were insurable, no matter the size of the insured sum. For more on marine insurance, its emergence and early development in the Low Countries, see: J. P. Vergouwen, De geschiedenis der makelaardij in assurantiën hier te lande tot 1813 (The Hague,  . The ERC project AveTransRisk addresses these issues in an comprehensive and internationally comparative study focused on GA. 10. The focus of this article is GA as applied in Amsterdam. 11. Also spelled 'dispatch', 'dispache' and 'despach'. could be insured, GA persisted. 7 Whereas marine insurance has been extensively researched, by legal researchers, historians and social scientists, GA has hardly received academic attention. Most publications regarding GA have a legal character and consist of a description of the relevant regulations, or are manuals or reports of adjudicated GA cases. 8 There has been little research into the importance of GA in the development of trade and transport, on choices and behaviour of shipowning companies, merchants or masters, or of the effect of GA on marine insurance markets. 9 Apart from, in particular, legal publications, little has been published about GA and the procedure of average adjustment. Before being able to analyse the significance of GA, or the way it is related to other means of risk management, we need to lay the groundwork: what exactly is GA and how does an average adjustment work? This is the aim of this contribution: to explain and illustrate GA, as a concept, the legal framework and the GA procedure (that is known as average adjustment). 10 The basis of this paper is the dispach of the Jan Maria, which has been preserved in its entirety. 11 A dispach is the final report and annexes relating to the GA adjustment, in this case the Jan Maria's. What happened to the Jan Maria was not extraordinary in any way, as it must have happened to hundreds, if not thousands, of ships ploughing the Baltic, North Sea, Mediterranean and other bodies of sea. A master and his crew, forced by fierce storms, or threatened by pirates or privateers, would decide to jettison part of the cargo, or intentionally inflict damage to the ship, to prevent the sinking or taking of the ship, to save the lives of crew members and the cargo. Even though this must have happened to countless ships, very few dispaches have been preserved. This is probably due to the fact that, in particular as of the nineteenth century, these dispaches were held 12 by individual shipowning companies rather than formal courts. In fact, the dispach regarding the Jan Maria is the only known copy of a seemingly complete dispach from the nineteenth century in The Netherlands. 12 This dispach of the Jan Maria's GA adjustment is a good example of what Edoardo Grendi has named the 'exceptional normal'. 13 The completeness and quality of this record makes it exceptional, even though it refers to the most mundane of cases: an ordinary ship on a routine voyage with a run-of-themill cargo gets into mortal danger during an autumn storm on the North Sea. 14 The dispach consists of a summarizing report, including the calculation of the damages and relevant values and a great number of appendices which include invoices, valuations and translations of trial transcriptions. 15 The dispach not only shows us how the whole GA procedure took place -who was involved and the costs of the process -but also gives us an idea how a medium-sized shipowner conducted the business of long-distance maritime trade in The Netherlands during the late nineteenth century. In essence, the Jan Maria offers a unique insight into the coping strategies of small-scale entrepreneurs in shipping.
As has recently been argued in a supplement of Past & Present by John-Paul Ghobrial, global history cannot give us all the answers based on generalization. 16 On an aggregate level, providing we have the data, GA dispaches and rulings can reveal much about types, prices and volumes of merchandise, mercantile networks, trading routes, shipping models and crew numbers, nationality of masters and their crew, meteorological data and judicial variation between ports. 17 However, these data are not yet available. For now, we need to populate history with real people, as Tonio Andrade has advocated. 18 And these real people lived in turbulent times: steam was supplanting sail and wood was being replaced by iron and steel. Not only were journey times shortened, shipping companies were faced with increasing investment amounts as well as expensive maintenance. Tramp shipping became increasingly important. 19  risky and one of the methods to deal with these inherent risks was GA. What can we learn from this particular dispach about the relationship between master and shipowner, the duality of the master's role as guardian of ship and cargo, of the many choices shipowners and merchants were faced with in a highly volatile market? And how did a concept like GA, tested and tried, fit into the quickly changing shipping world of the nineteenth century? Considering the ongoing debate among historians on how microhistory and global history may be combined, or how microhistory may possibly enrich global history, it may be tempting to argue that the case of the Jan Maria is an example how the microscale can best describe larger, global issues that were at play in nineteenth-century maritime trade. This article does not aspire to do so. As little has been written about GA, the procedures, the calculations and about those who were involved in the whole process, I will do just that: analyse what GA is in essence and its potential for economic analysis using the unfortunate Jan Maria as a case study. In this paper, I will explain the concept and procedure of GA and address issues like: how and when was it applied? who were involved and how was it regulated? I will argue that GA was an important part of risk management of shipowners and merchants and that, in spite of financial innovations like marine insurance, it remained important for those active in mercantile trade. In the next section, I will explain the principle and functioning of GA before moving on to the unfortunate events that led to the Jan Maria's GA adjustment. I will conclude with some remarks and suggestions for future research.

General Average: Principle, regulations and enforcement
In the second half of the nineteenth century, when the crew of the Jan Maria was battling to save the ship, its cargo and their lives, GA procedures had been a well-known concept among masters, shipowners and merchants for centuries. 20 GA has a long and impressive lineage dating from Antiquity, and its long heritage was still affecting regulations and procedures that were used for the Jan Maria in the nineteenth century. All well-known (late) medieval and early modern regulations, like the Role d'Oleron, the Wisby and Gotland laws, dealt with jettison and other forms of averages. 21 With the passing of time, as more parties became involved with varying interests, the resulting further complexity required increasing regulation. 22 In the early modern period, all across Europe various amendments and additions were made to the regulations, however the basic principle remained the same: all those taking part in a journey by sea were to contribute pro rata (based on the value of their goods or ship) in case of intentionally-inflicted damage in order to prevent more damage. It was key that the damage was inflicted intentionally: the accidental breaking off of a mast was not considered GA. It was also crucial that the objective of the action was to prevent further damage. Additionally, the actions taken needed to be successful -if a ship sank nonetheless, then there was no GA case and all parties would then bear the costs individually. If, for example, a master of a ship with cargo from five different merchants discovered halfway through the voyage that part of the cargo of one merchant had not been securely packaged and was damaged, or if the crew damaged the ship while docking, this was considered Particular Average (PA). However, if that ship got into trouble during a storm, the master would consult the crew and, if present on board, the merchants, before taking measures. If they decided to cut the mast in order to prevent the ship from capsizing or sinking, the costs of the damages to the ship would be carried by all those benefiting from this action. 23 There were strict rules to which the master had to adhere in order to prevent unnecessary damages or fraudulent behaviour, which varied depending on countries. In The Netherlands he was obliged to jettison the cargo that was heaviest in weight and lowest in value first. 24 Some rules were the same all across Europe, and also remarkably resilient across the centuries. Amongst these, the master had to record the fact that he had consulted his crew, the reason for the jettison and the amount of jettisoned goods. 25 He was also obliged to give a statement in the first port of arrival in the presence of local authorities. In this statement he would explain the circumstances -invariably it was the fiercest storm or the fastest and most ferocious pirates that had forced them to cut a mast or to jettison any cargo. Usually a number of crew members or others present needed to confirm the truthfulness of his account. This statement was the first step of the GA procedure, the average adjustment. 26  i.e. all contributing values combined, was 10,000 guilders, that meant for every 100 guilders, one was to contribute five guilders. If there were three parties involved, shipowner A with a total contributing value of 6,500 guilders, merchant B with value of 2,500 guilders and merchant C with an interest valued at 1,000 guilders, this would mean that A would contribute 325 guilders of the total of 500 guilders in GA damages, B would pay 125 guilders and C would contribute 50 guilders. The above is, of course, a very simplified example; if there were several incidents with various actions that led to GA, the sequence of the events was crucial to calculate the correct values. The more parties involved, the more complex the GA adjustment would be. 32. Unfortunately, a large part of the Chamber's records has been lost. The records that relate to the eighteenth century show that in total 9,000 average cases were adjusted by the Commissioners. Whether this constituted all GA cases that related to the port of Amsterdam century, GA adjustments and calculations were handled by 'wise men' in Amsterdam, of whom we unfortunately know very little. 27 In 1598, the Chamber of Insurance and Average (Kamer van Asseurantie en Avarij) was established in Amsterdam and tasked with handling GA cases. 28 The Commissioners of the Chamber would require parties to hand over all relevant information. This included invoices of their merchandise, the bill of lading and, if applicable, the insurance policy. These documents were necessary to confirm the value of the merchandise as stated by the merchants. The Commissioners would first determine the damages: what was the damage to the ship and to the cargo? Masters and shipowners would often increase the amount of damages. To prevent them from including 'normal wear and tear' of the ship and its fittings, the Commissioners would use tables with standardized lengths of ropes per type of ship. 29 The second step was to determine the total value of the journey: the value of the ship, the freight fees and the cargo, which combined made up the so-called contributing interests or values (dragende waarden). Here, parties were inclined to give lower values, as the amount of their final GA contribution was based on the value of their assets. In the nineteenth century, it was customary to use the value of the goods at the port of destination to determine the value. 30 After the total contributing values were calculated, the Commissioners would determine the omslag, the amount that was to be paid per guilder or per 100 guilders. 31 The Amsterdam Chamber was active from 1598 until the end of the eighteenth century, when the institutional structure of the Dutch Republic was upended in the wake of revolution that swept across Europe. 32 or if a number of cases were dealt with informally, for example, via arbitration, is not clear. By the time the Jan Maria sailed the Baltic, the relevant regulations were no longer merely municipal. The Batavian Period and the French Annexation changed the regulatory and institutional structure profoundly in the area that is now The Netherlands. The old institutions, including the Chamber of Insurance and Average, were supplanted, although the introduction of new institutions and national laws was slow at best. The newly formed Kingdom issued its first Commercial Code (Wetboek van Koophandel, WvK) in 1838, which henceforth would regulate and structure the country's economy. However, between the collapse of the structures and institutions of the ancient regime and the establishment of those of the new Kingdom, there was a regulatory vacuum, especially regarding certain parts of economic life. 33 A Court of Commerce (Rechtbank van Koophandel) was instated in 1811, but nevertheless, two lawyers, Mr. Mozes Asser and his son, Mr Tobias Asser, both active as Average Adjusters (dispacheurs), feared GA adjustments would suffer due to the absence of an efficient regulatory structure. They therefore designed a standard form of agreement specifically for GA adjustments, the so-called Amsterdamsche Compromis. 34 This contract regulated the process of adjustment of GA damages in Amsterdam. The Amsterdam Stock Exchange authenticated the Compromis, which meant it had officially been accepted as custom. 35 The original version was issued in 1811 and stated that the final report, the dispach, was to be drawn up by the average adjusters who were to be formally appointed and sworn in before they took up their duties. 36 The Amsterdamsche Compromis was a claim of the shipowning company (reder) on the other parties involved (mainly the owners of the cargo). Although this was not in line with Dutch legal tradition, in which the Commissioners of the Chamber of Insurance and Average were independent arbitrators, the Compromis tallied with the practice and custom in London at the time, where an average adjustment was considered a claim from the shipowner on the owners of the cargo. 37 Several parties lobbied for a specialized court to handle the complex GA cases along the lines of the former Chamber of Insurance and Average, but their pleas were to no avail. 38  '1. All that was given to enemies or pirates for the liberation of ship and cargo; 2. Anything of ship or cargo that was jettisoned for the common good; 3. Cables, masts, sails, and other tools that had to be cut or broken for the same purpose; 4. Anchors, ropes and other items that were let slip overboard (slippen) for the same purpose; 5. Damage to goods still on board that were damaged during jettison; 6. Damage to the ship to facilitate the jettison of goods or to advance the discharging of water and any damages that then resulted due to the water; 7. The healing, keeping and compensation of all those aboard who were wounded or maimed during the defense of the ship; 8. Compensation or ransom for those who were taken prisoner or enslaved while in the service of the ship or cargo; 9. Pay and rations of the crew during enforced stay in port of refuge; 10. Pilotage and other port expenses of port of refuge; 11. Rent of storage during repairs in port of refuge; 12. Expenses to reclaim the ship and cargo after both have been taken; 13. Pay and rations of the crew during the aforementioned taking of the ship and cargo in case they are being released; 14. Expenses of unloading lighter ships in case the ship was forced to seek shelter in a port or river by weather, enemies, or pirates, or any other cause to seek refuge, including the loss or damage to cargo due to the unloading or the lighter ships or the re-loading of the ship; 15. Damage to ship or cargo do to enforced beaching of the ship to prevent the taking of loss; 16. Expenses and pay to reverse the beaching; 17. Loss or damage to cargo that had been loaded in lighters due to an emergency; 18. Pay and rations of the crew when the ship, after having sailed, is being detained or forced to seek shelter; 19. Bottomry premiums of bottomry loans that were taken out to cover GA damages; 20. Insurance premium to insure bottomry premiums or the loss on the enforced sale of cargo in order to cover GA damages in a port of refuge; 21. Expenses of average adjustment; 22. All expenses, including pay and rations of the crew due to a quarantine that could not be foreseen when bills of lading were was taken in Amsterdam by those dealing with GA adjustments in daily life and as a result, the Average Committee (Avarij-Commissie) was instated the same year as the first version of the Amsterdamsche Compromis was issued. The committee was to ratify (homologeren) dispaches that were drafted by the average adjusters or, if applicable, adjudicate disputes regarding the apportionment among the various parties of the GA damages. 39 In 1838, the Commercial Code was issued, and GA was regulated in detail in its Book 2, chapter 11. In one of the first articles, it is stated that if parties have not made other contractual arrangements, then GA (and PA) should be dealt with according to the following rules. These clearly stipulated the difference between GA (Gemeene avarij or Avarij-grosse) and PA and determined that the value of the vessel, the amount paid in freight fees, and the value of the cargo was to contribute to the amount of GA damages. In the following articles the various types of GA were defined, including the better known forms of 'jettison of cargo' and 'cutting of the masts or ropes', to less common incidents such as the wounding or death of crewmen while defending the ship and ransom paid to pirates or enemies. 40 Again, the law emphasized that the actions had to be signed; 23. In general, all damages that were borne of an emergency, deliberately initiated and as direct consequence suffered, after required consultation, and were made to safeguard the common well-being of ship and cargo. All other expenses were considered Particular Average and were to be borne by an individual ship owner or merchant, General Average was to be carried by ship, freight and cargo.' Wetboek van Koophandel (1838) intentional and in the interest of both the ship and its cargo. 41 Other articles focused on whether certain costs were to be accepted as GA -for example, the circumstances under which fees for pilots or lighter ships were to be included as damages for the common good. 42 The final articles of the Commercial Code (articles 722-740) related to the actual procedure, calculation and allocation of GA. According to article 722, the calculation and apportionment of the GA damages was to be undertaken in the harbour of destination, unless parties had agreed otherwise. In practice, it seems that the administrative handling of GA took place in the city where the cargo was originally destined. The reason was that at this port it was easiest to determine the value of the merchandise, which was necessary for the GA calculations. Moreover, this was generally the location where merchants, shipping agents and shipowners finalized the voyage in administrative and financial respects. Dealing with the GA procedure in this location was thus the most sensible. If, however, the GA adjustment was handled abroad, the Dutch authorities would honour foreign laws and customs. Finally, there was a time limit on claiming GA: all claims had to be made no later than two years after the incident. 43 Even after the Commercial Code was issued in 1838, GA adjustments continued to be handled in accordance with the Amsterdamsche Compromis. 44 In fact, the text of the Compromis clearly stated that signatories agreed to have the GA adjusted in accordance with the Amsterdamsche Compromis and that the Commercial Code was rendered 'inactive' unless parties appealed the outcome of the dispach. In spite of this 'discounting' of national law, both the Amsterdamsche Compromis and the Average Committee in 45 Amsterdam seem to have enjoyed a formal status: the Average Committee even held 'court' in the municipal Palace of Justice. 45 Between the implementation of the Commercial Code in 1838 and the incident with the Jan Maria not many changes were made to Dutch GA laws and rules. There were, however, other developments that affected how shipowners, merchants, shipping agents and insurers dealt with GA. As the nineteenth century progressed and trade and transport were unequivocally transformed by technological innovations, political developments and economic integration, it became increasingly difficult to operationally handle the variety of GA regulations -not only between countries, but also between ports of the same country. The various GA regulations, especially in terms of procedure, had created a muddled maze of national and even local regulations for shipowners, merchants, shipping agents and underwriters to deal with. Clearly, this was not an ideal situation in a world where trade and transport were becoming increasingly international and integrated. The Dutch provisions of the 1838 Commercial Code were far more generous than those of other countries and jurisdictions, and with ever growing integration of maritime trade these differences between national legislative codes were increasingly causing problems. 46 In 1860, there was an attempt to unify the varying international regulations. At an International Conference held in Glasgow delegates of a number of seafaring countries debated the proposed regulations regarding GA. Mr Eduard Rahusen, a lawyer and average adjuster, was one of the Dutch representatives. His report about the conference plainly shows the considerable differences between national laws and the difficulties of devising a set of unified rules. 47 In spite of the disparities and problems, the delegates did accept a number of generally accepted rules which would be the starting point for further international legal convergence. In his report about the Glasgow conference, Rahusen made it clear that he did not agree with several of the new rules that were to govern GA proceedings internationally, as they would leave too much room for debate and could lead to fraudulent behaviour. He felt the Dutch regulations were less ambiguous and thus superior. 48  These rules were subsequently applied by the dispacheurs in The Netherlands. The Amsterdamsche Compromis was being used well into the twentieth century as it merely structured the relationship between the various parties involved and the procedure in case of a conflict. The Compromis did not state which costs and damages were accepted as GA or which circumstances and actions were considered GA. 50 In 1879, an important revision of the Amsterdamsche Compromis was put forward by marine underwriters and vehemently opposed by Mr Eduard I. Asser. 51 The underwriters proposed that in the future adjusters would no longer be working on behalf of the shipowners but on behalf of all parties involved. Asser argued that the fact that the adjusters acted on behalf of one of the parties -the shipowning company -had prevented many legal procedures in the preceding six decades. The proposed text was more in line with pre-1810 Dutch legal traditions, where the adjusters were supposed to be independent arbitrators, but clearly did not correspond to the process of average adjusting in London, where an average adjustment was considered a claim from the shipowner(s) on the owners of the cargo. 52 Asser contended that the new text left too many options for litigation and disputes, but in spite of his disapproval, the new text was adopted in 1880. 53 It was this version of the Amsterdamsche Compromis that was used during the GA procedure of the Jan Maria. 54

The last voyage of the Jan Maria
Before the unfortunate night in December in 1883, when the Jan Maria was towed to the port of Geestemünde, the ship had already had an eventful life. The wooden, threemasted schooner was commissioned by Peter Johnston in 1873 and built in the shipyard of Rand & Berger in Manitowoc, Wisconsin, USA. The ship, then named Thistle, was 138.2 ft long, 8.02 ft wide, had a depth of 11.6 ft and was originally designed to sail the Great Lakes. 55 The dimensions of the ship, as most other ships built in the area, were 56. The Canalers or Channelers were not very popular among sailors due to their cramped quarters and bad performance in stormy weather. To counter the ships' notorious instability, they were mounted with centreboards, a feature for which North American shipbuilders were known. limited by the size of the locks of the Welland Canal, hence the name 'Canallers' for this type of ship. 56 The ship was launched in September 1873 and initially did sail the Great Lakes, as intended. However, due to the enlargement of the Welland Canal locks, larger, more cost-efficient ships started sailing the Great Lakes, putting smaller ships, like the Thistle, out of work. 57 The Thistle was towed to the Atlantic, where it served on a North-South coastal route before crossing the Atlantic to Scotland in 1876. In 1877, the ship made a second Atlantic crossing with a cargo of 4,000 barrels of resin. 58 The original plan was for the ship to continue to the West Indies, but during its intercontinental crossing, it was heavily damaged and needed extensive repairs on its arrival in the Netherlands. 59 In spite of the original plans, the ship was sold 'for a good price', as a newspaper reported. 60 Van Vliet & Co, a relatively small shipowning company based in the Dutch port of Nieuwediep, became the ship's new owner and renamed it Jan Maria. 61 The new master, Mr J. D. Kunst, left the port of Maassluis on 7 June 1878, before the formal sale was concluded, en route to Riga. The Jan Maria's passing of Elseneur (current Helsingø, Denmark) was, as was customary, recorded in the press on 14 June. They arrived in Riga on 23 June for the first time -many journeys to the Baltic would follow, mostly direct from the Netherlands but occasionally via England. 62 On average, it would take the Jan Maria a week to reach Elseneur from its home port in Nieuwediep. From there on it was usually another seven to eight days before they reached Riga, the port mostly frequented by Kunst and his crew of eight men. The ship was usually ready to return after 10 days although, in the autumn, when the weather turned, they would sometimes begin their journey home after a mere five days. 63 A 'normal' return trip would take the Jan Maria 40 days altogether and the ship usually managed to make three of these return voyages per year, depending on the start of the season and the weather conditions.
Although the available resources are limited, it seems that the Jan Maria's cargo was always handled by Vinke & Co, a well-known shipping agency (cargadoor) from Amsterdam. Established in 1860 and specialized in the Baltic wood trade, Vinke & Co handled wood transport from Sweden, Norway, Finland, Russia and Archangel. 64 There seems to have been a good working relation between the cargadoor, Vinke and Van Vliet & Co, the company that owned the Jan Maria. Albertus Vinke was known to visit the Exchange every day; on Mondays, after his daily visit, he would go to the so-called Polish Coffeehouse, a meeting point for anyone involved in the Baltic timber trade. Afterwards, every Monday, he would meet and dine with Klaas van Vliet. Undoubtedly, they discussed business in general, the various opportunities and threats of their trade, and perhaps they closed deals -all based on the information gathered throughout the day at the Exchange and the Coffeehouse. 65 On 30 October 1883, the Jan Maria left the port of Riga for its return voyage to the Netherlands. Neither the master nor its crew were able to save the ship's journal and the only available information regarding the journey consists of what was included in the GA dispach. 66 The dispach comprises a report, including a summary of the incident, an overview of the damages and a great number of appendices. 67 Thus we know that the Jan Maria was on its way to Zaandam, where the firm that owned its cargo of beams, was located. The cargo consisted of 1,509 beams, which was equivalent to 156½ tults. 68 The freight rate charged was 32 guilders per tult, so in total a little over 5,000 guilders would be paid by the owner of the freight, G. Kamphuijs & Zoon, for the transport of the beams from Riga to the Netherlands. In addition, the Jan Maria carried 588 firs which were owned by the master, Mr Kunst. 69 Finally, the ship carried a limited amount of ballast (iron). After having left Riga, the ship got caught in a storm and after five days the master decided to seek shelter in Osterrisoër (current Risør, Norway) where a number of repairs were carried out. 70 It was also during this enforced stay that both the firs that were owned by the master and the freight fees were insured. Vinke & Co acted as the agent for Van Vliet & Co and commissioned insurance brokers Brak & Moes of Amsterdam to insure the freight fees (also called 'freight') for a total of 5,000 guilders at 5¼ per cent. Van Vliet also dealt with the insurance for the master's firs for 1,000 guilders. Both policies were dated 5 December 1883. 71 There is no information whether the ship itself was insured. However, considering the completeness of the dispach in which all relevant documents were included, this seems highly unlikely as there is no reference to this in the dispach. Perhaps the sustained damage had rendered the ship uninsurable or too costly to insure. 72 On 6 December, the Jan Maria set sail again, bound for Zaandam. Before leaving the port of Osterrisoër, the master made a statement to account for the repairs in the port of refuge. This statement, also included in the dispach, states that the ship was in good condition. 73 However, on 12 December it got caught in another storm, this time close to the light beacon of the north German island of Borkum. According to the dispach, the storm was accompanied by heavy rainfall that by the afternoon had developed into a fullblown hurricane: a 'sea stretching to the heaven and outraged'. Several parts of the ship were damaged, including the sloop. At 6pm, the deck load and water barrels started to float and it was then that the master and his crew decided to jettison some of the deck load. The pumps were working through the night, the crew continued to jettison the deck load and they managed to keep the ship afloat. 74 The next day, the storm abated and even though the situation had improved, after another consultation with his crew, master Kunst decided to go to the nearest port. However, it never came to that: the storm rose again, the ship started taking in water, the remaining deck load started to float and on top of it all, the first mate suffered a leg injury. 75 Even though the pumps were working nonstop, the ship started to incline and the deck load went overboard. The crew sought shelter in the mizzenmast and spent a night in 'mortal fear, afraid the ship would be torn apart'. In the morning, they managed to get the attention of a passing steamship, the Lahneck from Bremen, which was en route to Spain. The master, the injured first mate and three sailors were taken on board the Lahneck. The remaining crew members of the Jan Maria were saved by another passing ship, the Diana. The battered and heavily tilted Jan Maria and its remaining cargo, was towed to the port of Geestemünde by the Diana. 76 It was then that the administrative handling of the unfortunate incident commenced and the GA was adjusted.

A General Average adjustment in practice
The GA dispach of the Jan Maria is not unique in itself, but in its survival and comprehensiveness. Although in the Municipal Archives of Amsterdam and the archives of the NEHA in Amsterdam there are summaries of verdicts of GA cases dating back to the late sixteenth century, these are concise texts, without any adjoining original documents. 77 The Jan Maria's dispach, however, seems to be complete and is therefore a valuable source of information about the procedures governing GA and the parties that were involved in these proceedings. It also gives anecdotal information regarding trade and transport on one of The Netherlands' main shipping routes.
The dispach consists of a collection of documents of 128 pages in total, starting with a report of the incident, how it was handled and the financial conclusion. The first report gives a summary of the whole case in six neatly written pages. This is followed by a tabular overview of the financial data: the damages, the contributing values and the contribution 78 per party (to which I will return later). Finally, there is a collection of appendices, including a signed copy of the Amsterdamsche Compromis about the GA adjustment, and a statement made by master Kunst in the first port of arrival. As Kunst and part of his crew were brought to Dover, England, by the Lahneck, his statement was made to English authorities, in the presence of the Dutch vice-consul who also verified the truthfulness of the translation of the statement. 78 What follows are statements relating to the salvaging of the cargo from the damaged ship, legalised translations of German court documents, reports regarding the auction of the ship and part of the cargo, the freight bill, a valuation of the cargo and various invoices and receipts. Apart from the Amsterdamsche Compromis, where a number of law clauses were specifically excluded, there are no references to any laws or stipulations. Apparently, the guidelines and regulations that governed the average adjustment were well known among parties and needed no specific references.
Kunst's statement, supported by his crew, turned out to be crucial for the procedure that followed. According to their account, jettisoning the deck load was a deliberate action, taken to salvage the ship and its cargo, and was successful and should thus be treated as GA. They clearly stated that, after having deliberately jettisoned the deck load, the situation stabilized. The second incident, which could also be regarded as a deliberate action (seeking shelter in a port of refuge, after a compulsory consultation of the crew), ultimately led to the ship becoming uncontrollable and heavily tilted. This was, according to their statement, a separate incident from the jettison of the cargo. Had they not been this specific, then the conclusion would have been that their actions had not been successful, as the ship had to be towed (and was later classified as a total loss) and in that case there was no GA case, but all damages would then have to be borne by the individual owners of ship and cargo.
In the port of Geestemünde, the whole procedure to deal with the heavily damaged Jan Maria was set in motion by the Ambstgericht, which appointed harbour master F. von Bülow and surveyor J. D. Ihlder to assess the damage of the ship. J. H. von Hasseln was the Strandvogt of Geestemünde and as such represented the authorities in most dealings regarding the settlement. Von Hasseln commissioned the company of Marcus, Durkee & Gundersen of Bremerhaven to draw up the report regarding the German part of the procedures. Von Bülow and Ihlder concluded that the expected repairs to the ship would amount to Mark 56,837. 79 Consequently, during a hearing before the Ambtsgericht on 30 January 1884, the ship was declared unworthy of repair. The ship was subsequently sold on 5 February for Mk 5,800 (approximately 3,415 guilders) to D. F. Ahlsen & Sohn, a cement manufacturer in the town of Itzehoe. 80  public auction organized by the Strandvogt, the ship's iron ballast was sold for Mk 700. The firs that belonged to master Kunst were also auctioned. Little was left of them as most had been jettisoned during the storm, the remaining firs only generated Mk 41.16. 82 The total costs, as calculated and reported by Marcus, Durkee & Gundersen amounted to Mk 9,900.15. This included the fee for salvage company Bugsir Gesellschaft Union Bremerhaven, of Mk 5,500, which had been determined by the Königlich Preussisches Standambt on 13 February 1884. The fee was, as was custom, approximately two-fifths of the revenue of the sale of the ship and cargo. 83 The total costs incurred also included, for instance, those of Kunst's stay at the Lehrkes Hotel in Geestemünde, several legal fees and the costs of telegrams. The ship, its inventory and the other items generated in total Mk 7,347.46. In the meantime, Mr Bosse, a shipping agent in Bremerhaven, had arranged for the cargo of beams the Jan Maria was carrying to be transported to its original destination. Two ships, the Ceres and the Pietertje, carried 1,131 beams in total, rather less than the original 1,509 beams. The remainder had been lost, most probably when the deck load floated overboard. 84 The costs of this alternative transport to Zaandam were initially paid by the firm of Mr Bosse, who was reimbursed by the owner of the ship, Van Vliet. 85 According to the dispach, the previously mentioned Mr Ihlder also acted as representative of the insurers of the ship's cargo. Apparently, Kamphuijs had the load of beams insured, although we have no information regarding the insured value or the premium paid. 86 Ultimately, after the case of the Jan Maria was concluded in Geestemünde, Van Vliet & Company paid the difference between the revenues and costs of Mk 2,552.69 to Marcus, Durkee & Gundersen. 87 After everything was settled in Germany, all documentation, authenticated by the German authorities, was sent to the average adjusters in Amsterdam -the company of Kool & Rahusen. 88 Although the original destination of the Jan Maria was Zaandam, Amsterdam was the nearest port with average adjusters and it was also where Vinke & Co, the shipping agent that acted as intermediary for the insurances, was based. When the case was transferred to Amsterdam, the GA procedure was started when the Amsterdamsche Compromis, as described above, was signed by the master and the owner of the cargo, G. Kamphuijs & Zoon from Zaandam, on 1 April 1884. 89 In the Compromis, the signatories stated, even though they did not acknowledge that there was in fact a case of GA up front, that they commissioned the average adjusters to make up the final account of the average and, if relevant, to allocate GA damages among the parties. They also agreed that any dispute that was to arise was to be put before the Average Committee of Amsterdam (Avarij-Commissie te Amsterdam). 90 It was custom in Amsterdam to accept any accounts or verdicts that were made up abroad and to use these accounts as a basis for the final average adjusters' report. 91 Thus, the report made by Marcus, Durkee & Gundersen was the starting point for the GA assessment and calculations by Kool & Rahusen, 92 who first had to determine the total damages and costs and then to categorise these as either PA or GA. The dispacheurs determined that total damages and costs of the Jan Maria's unfortunate incident amounted to 9,663 guilders. 93 The firs of the master that had been jettisoned were not included in the damages -most probably because they were loaded on deck. There were two instances when the costs of jettisoned goods were usually not included in the damages: in case they were stowed as deck load or when they had not been included on the bill of lading (vrachtnota). Kunst's firs were recorded on the freight bill but only a small number of firs were ultimately unloaded in Geestemünde, in all probability because most had been jettisoned. 94 Of the total damages, f 3,303 was categorized as PA. A distinction was made, for example, between the PA relating to 'the ship' and the 'shipowning company' (Rederij): the costs of the company's representative travelling to Geestemünde, when it was thought that the crew had been killed, was assigned to the shipowning company rather than to the ship. 95 The same was done with the costs of Mr Kunst's stay in the hotel in Geestemünde: these were allocated to the shipowning company. On the other hand, the costs of registering the sale of the ship were allocated to the ship and not the shipowning company. 96 The costs of both the ship and the shipowning company were ultimately borne by Van Vliet.
The costs of transporting the remaining cargo from the port of Geestemünde to Zaandam by means of the Ceres and the Pietertje amounted to 2,103 guilders in total, which was divided as PA between Van Vliet and Kamphuijs, as Kamphuijs had not yet paid the entire amount of freight. The remaining cost of the total, 6,360 guilders, was judged to be GA and this included, most importantly, the salvage fee (f 3,240). Other expenses that were classified as GA included costs of unloading and guarding the cargo (approximately f 850), of pay and rations (f 395), and of costs of telegraphs, travel and the assistance of Vinke & Co. The costs of the average adjusting were also added: 189 guilders for the adjusters, Kool and Rahusen, and 21 guilders for the Average Committee to ratify the dispach. 97 The GA now needed to be distributed among the parties benefiting from the actions taken by Mr. Kunst and his crew (see Figure 1).  After having determined the damages, the average adjusters determined the contributing interests. As was stipulated in the regulations, and as was customary in Amsterdam's averij grosse-calculations, three main components were to contribute to the mutual costs: the ship, the cargo and the freight. 98 In the case of the Jan Maria, this meant the ship, the master's firs, the cargo of beams owned by Kamphuijs & Zoon, and the shipowning company. The leading principle for the adjusters was to determine 'what has been salvaged?' This was particularly relevant for the beams and for the amount of freight. As part of the cargo of beams was lost, the contributing interest of the beams was based on the number of beams that was salvaged (1,131), not the original number that had been loaded (1,509). The contributing interest of the freight was also based on the salvaged cargo. Master Kunst's firs were included in the total contributing interests, even though the damages did not qualify as 'costs'. These stipulations were undoubtedly meant to dissuade merchants from transporting goods without a proper freight bill (for example, by making a deal with the master); in case their goods were then jettisoned during a GA incident, they would not be compensated for their loss, but they were nonetheless obliged to contribute to the mutual damages. The contributing interest of the master's firs was a mere 20 guilders. 99 The Jan Maria's value was equal to the price the wreck was sold for at the auction minus the PA, which came to 3,748 guilders. With the contributing interest of the freight, it was taken into account that not all beams had ultimately been transported and that a part of the freight had been pre-paid by Kamphuijs. Also, the 'normal' costs of the crew's wages and the PA were deducted to come to the freight's contributing interests.
Thus, the total contributing interests of the various components came to 7,430 guilders. The Averij Grosse had been calculated to be 6,360 guilders. As the latter was 85.6 per cent of the total contributing interests, the contributing interests of the components were multiplied with this percentage. So, the contributing interest of the ship of 3,748 was multiplied by 85.6 per cent to come to 3,208 guilders. This was the amount with which the shipowner contributed to the total costs of the general average (see Figure 1). 100 As Vinke & Co had handled Van Vliet & Co's insurance coverage, they forwarded the final invoice to Brak & Moes, the insurance broker. After having received the payment from Brak & Moes, Vinke & Co wrote Van Vliet & Co on 8 July to let them know that a total of 4,020 guilders had been received from the broker. Of this amount, f 998 was designated to compensate Mr Kunst for the loss of his firs, which had been insured for f 1,000. The remaining 3,038 guilders was compensation for Van Vliet for the loss of its freight revenues. Vinke & Co deducted their own fee of 40 guilders, with which the case of the Jan Maria was formally concluded. 101

Concluding remarks
General Average is an intriguing concept -it has outlasted numerous legal developments and adjustments, the emergence of other methods of risk management and major transformations in trade and transport. In the nineteenth century, when profound changes affected trade and transport, institutions were put to the test. Many institutions were supplanted, but the concept of GA persisted. Yet, in spite of this resilience, it is hardly known outside the world of average adjusters and maritime legal experts. Marine insurance has, although not consistently through time, received far more attention, including from academics. 102 Is it because scholars are more interested in novel concepts and constructs? Perhaps we are more intrigued by the demise of institutions, and the reasons behind their downfall, than a concept that seems to have worked efficiently in the wings, adapting to changes if necessary but persisting nonetheless. This was not always a smooth ride as, especially as of the 1820s, the necessity of GA was questioned because, as was argued, marine insurance made it redundant. However, as was the case with the uninsured Jan Maria, not every ship or all cargo was in fact insured. Advocates of GA also countered that GA was the only method that would guarantee minimal overall damages, which would ultimately benefit shipowners, merchants and underwriters. 103 The master, his skills and the choices he made (after the compulsory consultation of his crew) were, of course, critical. 104 Regulations were focused on guiding and controlling the master: his options were limited by guidelines and stipulations, so that the most likely outcome was the one with minimal total damages. 105 Still, the case of the Jan Maria shows that even within this restricted setting, master Kunst managed to influence the procedure in a way that was beneficial to his employer. By emphasizing, in his statement, that, after the voluntary jettison, the ship was stable but they nonetheless decided to go to a nearby port, the master made sure that the damages of the jettison would be regarded as GA. If the ship had made it to port on its own account, the repairs and the stay in the port of Geestemünde would also be considered GA. The master was undoubtedly aware of the fact that his firs, the cargo and the freight were insured, but that the ship itself was not insured. By making sure that his actions were considered GA he tried to avert financial doom for his employer.
The dispach of the Jan Maria is an 'exceptional normal' -although thousands of these dispaches must have been written, only a few remain. By studying these in the aggregate we may learn more about trade patterns, merchant networks, pricing and the efficiency of markets. By comparing dispaches from various countries and port cities, we will gain insight in possible diverging and converging regulations and laws and if the relevant 'players' (most notably, the masters) were aware of these legal disparities. This single dispach is only a small window into the world of a small shipowner, a shipping 106. The barrels that the Merkur was carrying were used for the biergarten (beer garden) of the salvage company. agent and a wood trader and their small, informal network. It shows us that some chains of international trade were already linked in perhaps unexpected ways: when the locks in a canal in North America were enlarged, rendering a certain type of ship unprofitable, this ship then became the best option for a small shipowner in the Netherlands. The dispach also tells us that, in spite of a world that was already strictly regulated by commercial codes and additional laws and regulations, individual choices within these formal constraints still affected the ultimate outcome. When Van Vliet was faced with the loss of both the Jan Maria and one of its other ships, the Express in 1883, the company may well have decided that the time of sailing to the Baltic was well and truly over. They sold their two remaining ships in 1884 and, as they probably did not have the means to invest in steamships, they became shipping agents -they converted to a business model that required fewer investments and with viable options. In doing so, they undoubtedly benefited from their existing network, their expertise and knowledge of trade routes, ships and trading partners abroad. Entrepreneurs like Van Vliet did not have the means to try to alter conditions or regulations, but rather they adapted their business strategy. They changed with the tides, shifting to other markets, focusing on a niche, adjusting their prices or, as in the case of Van Vliet, exiting the market altogether.
As for the Jan Maria: in actual truth it was not a total loss at all, as the ship was stripped of its masts and sails, renamed Merkur, and continued its life as a lighter on the river Elbe, shuttling barrels of cement. During a storm in 1892, it sank with a cargo of 3,000 barrels. The wreck was towed and sunk off the coast of Hamburg, where it remains today. 106