Jeremy Green

The main objective of the programme was to establish a database or Geographical Information System (GIS) of shipwreck sites in Galle and to investigate these sites where appropriate. Maritime archaeological fieldwork proceeded along these lines and nine sites were recorded by the Galle Harbour Project in 1992.
Initially, the most valuable sources of information on the location of sites were local fishermen. For many sites, only the general proximity was known, so underwater searches were required to locate them precisely. Once a site was located, its position was accurately recorded. A pre-disturbance survey was then conducted to determine the historical significance of each site and the potential for further archaeological investigation. As part of the Galle Harbour

 project, a study of a yatra dhoni model housed in the Maritime Museum in Galle was carried out by Tom Vosmer (1993).The sites discovered during the 1992 season included four 19th century or later iron shipwrecks, two wooden shipwrecks of c. 19th century, the VOC shipwreck Hercules (1661), a pair of abandoned cannon, and a jettison site.

All material collected during the survey was registered on a computer-based artefact registration system. During the first season, Sri Lankan archaeologists Dr Moira Tampoe of the University of Peradeniya and Dr Martha Prickett-Fernando of the Institute of Fundamental Studies assisted the project with the classification of the artefacts recovered. Conservation work was undertaken by staff from the Western Australian Maritime Museum and the Central Cultural Fund.

A group of Sri Lankan trainees underwent training in maritime archaeology and conservation, together with diving instruction.

Following the 1992 season, archival research was initiated by Robert Parthesius in the Netherlands to help to identify the wreck which was thought to be the VOC ship Hercules , wrecked in 1661.

Figure 4. Archaeologist recording cannon on Hercules site.

The second season, in 1993, continued the work of the first season, with additional survey of existing sites and investigation of new sites.

Three new wreck sites were located. Also, at the invitation of the Archaeology Department of Sri Lanka, a brief investigation was undertaken of the wreck site at the Great Basses—a site discovered and investigated by Arthur Clarke and Mike Wilson in 1963. The two new sites were iron shipwrecks, and the third was the wreck of a very well preserved VOC shipwreck, thought to be the Avondster (1659).

A group of Sri Lankan archaeology students were again given archaeological and conservation training. Details of these two seasons’ work and the training programme have been published by Green et al. (1993). During this season Dr Moira Tampoe of the University of Peradeniya assisted with the archaeological identification of artefacts. Conservation work was continued along the same lines as 1992, together with student training.

Figure 5. Recording anchors on the Great Basses site.
For a number of years, the Sri Lanka Government had been considering a proposal to develop a container terminal in Galle Harbour. As a result of these development proposals, an urgent request was made in mid-1995 to begin planning a complete survey and rescue archaeological operation in the Galle Harbour in the areas that were to be affected by the development, either by infill or dredging.

A three-stage programme of fieldwork was proposed, with the first two stages planned for completion in 1996:

Stage I Remote sensing (side-scan sonar) survey of the harbour to identify and locate wreck sites on the sea bed 
Stage II  Investigation and survey of sites discovered in Stage I and further archaeological investigation of known  sites  (E, J and L) to determine their extent and preservation. Archival research to determine more about the sites located.

Stage III  Proposed for 1997. Rescue archaeology programme appropriate to the outcome of Stages I and II and dependent on the development of conservation facilities in Galle.

Sonar survey
The first part of the 1996 field operations was conducted over a two-week period from the Sri Lankan Naval Base. An EG&G side-scan sonar was deployed on a Sri Lankan Navy small work-boat. The course of the vessel was determined using a differential GPS (Global Positioning System) which gave the position of the vessel to an accuracy of about ±4 m. The sonar record covered an area of the sea bed 100 m on either side of the track of the vessel so that the total area of sea bed covered was 62 400 000 square metres. The analysis of the sonar trace then identified targets that were sites of potential archaeological interest and these were numbered as target numbers. Using a CAD (Computer Aided Drawing) package, a large-scale map of the area was printed up at a scale of 1: 5000 and the tracks of the vessel were generated using the data from the data logger and then pasted onto the CAD plan. The calculated positions of the targets were then plotted on the map. A total of about 160 targets were noted for investigation.

Investigation of targets
The second phase of the 1996 fieldwork (Stage II of the programme) followed almost immediately upon the first, with work being carried out over a four-week period. An additional team, comprising Sri Lankan, Australian and Netherlands volunteers and staff, was brought in to assist with the diving operations.

Targets were prioritized by probable significance, generally based on size and structure. On each day, a survey team was sent to locate the identified targets using the GPS, and to mark each target site with a buoy. A diving team then investigated each target site in turn and produced a report on it. If the site was considered to be of archaeological interest, it was designated by a site identification code. The results of the diving showed that many of the targets were rock outcrops, but a significant number were sites which warranted further investigation. At the end of Stage II, twenty-one such sites were identified: three wooden wreck sites, thought to be European (17th or 18th century); a cannon site known to be the VOC ship Hercules (1661); a group of two cannon (possibly a mooring); two stone anchors; a site where ceramics are found in large numbers but it is uncertain if this is a wreck site or a jettison site; and a number of iron anchors and iron wrecks, which formed the remainder of the recorded sites.
Figure 7. The 1996 track plot and anomaly record for Galle Harbour.
Figure 8. The 1997 track plot of the magnetometer and side-scan survey.

Figure 9. Cleaning the large stone anchor from the anchor site.

February 1997
In February a small two-week survey was undertaken using a magnetometer in conjunction with a side-scan sonar to investigate the sandy areas of the harbour. While the side-scan sonar can locate objects on or protruding above the sea bed, it cannot determine what is buried. The magnetometer was used to search two main areas, the eastern part of the harbour where the harbour development was to take place and the area south of the Point of Galle where the Dolfijn was known to be lost. In addition the side-scan sonar was used to cover search areas which had been missed in 1996.

October 1997
In October a large team was again assembled to continue the investigation of Galle sites. The work concentrated on a site in the east of the harbour and the stone anchor site, together with investigation of the two 1997 magnetometer targets. Students took part in the project and received advance practical and theoretical training. An ArcView GIS (Geographical Information System) was developed to incorporate all archaeological and survey information which had been recorded over the years.

Limited excavation of the eastern site was undertaken to determine whether the site has archaeological potential, and to attempt to identify and date the vessel. Searches were made around the site to attempt to locate hull structure noted by a Sri Lankan Sub Aqua Club team in 1996. A small trench was made across the ballast area on the site in an attempt to determine whether there was any underlying hull structure; no structure was noted.

The stone anchors (one large and one broken), found in 1996 were relocated and in the process two new anchors were found: one was particularly interesting as it was a four-hole anchor, similar to anchors found in the Mediterranean. The anchors were re-surveyed, then raised and placed outside the Maritime Museum in Galle. Detailed recordings and stone samples were taken for an attempt to identify the source of the stone and the place of manufacture.

Robert Parthesius
Galle was the second most important harbour of the Dutch East India Company (VOC) in Asia. After the capture of Galle from the Portuguese in 1640, the Dutch established a firm base for their cinnamon trade in Ceylon. Up to 1796, when the English took over the city, the fortified town and its harbour was an important junction of Dutch trade and shipping. Galle was strategically located for traffic between the Arabian peninsula and East Asia, but it also played an important role in the local and regional trade.

The VOC wrecks are an important example of the interaction between history and archaeology. The presence of four (possibly five) identified and well-documented wrecks within the context of one harbour offers the potential for a broad interdisciplinary study of the ships, the harbour, the city, and the organization of the VOC. In an even broader perspective, this case study can be related to questions about the Asian shipping network and its organization.

This interdisciplinary approach would continue a tradition of integrated historical archaeological research on VOC ships. Since the 1970s the Western Australian Maritime Museum has been conducting research on four East Indiamen wrecked on the Western Australian coast. Central to this work was the linkage between the historical records about the ship, crew and cargo and the archaeological information. Sri Lanka has a unique collection of VOC wrecks representing several types of ships in use by the VOC during its presence in Sri Lanka. These sites can be linked directly with the administration of the company the records of which have survived in the archives in Colombo and in the Netherlands. The combination of the shipwreck archaeology and archival research—linking the harbour and the shipping—provides an important opportunity for a general study about trade and shipping, and a detailed study of the specific ships wrecked in the harbour.

Roughly, such a study would be structured in four parts: the use of the harbour; local shipping and trade in the Gulf of Bengal; inter-Asia shipping and trade; and direct trade and shipping between Ceylon and Europe. Detailed questions may be asked about the following aspects:

 Ship types in the inter-Asian trade and the trade between Asia and Europe (both European and local vessels);
 Technical developments during the 17th and 18th centuries;
 Organisation of the shipping in the harbour: repairs, equipment, crew, etc.

The Galle situation, in which archaeological material can be linked directly to the Dutch records in the National Archives of Sri Lanka, the VOC archives in the Netherlands and other Dutch archives in India, will enable us to conduct a multi-disciplinary research programme.

The Bay of Galle forms a natural harbour, protected from the NE monsoon but exposed to the SW monsoon. The entrance to the bay was dangerous because of the many submerged reefs and rocks. Of the six VOC ships known to be sunk in or near the harbour: three were wrecked during the vessel’s arrival at or departure from the bay; two were sunk inside the bay (one by an explosion at anchor and the second after coming adrift from her anchor); and one was wrecked outside the bay while waiting for the pilot to bring her in. The wrecking of these ships is well documented in an extensive archive on the organization of the VOC in Europe and Asia in general and, in particular, the activities of the VOC in Ceylon. These archives survived in the Netherlands and in some Asian countries like Sri Lanka. The National Archives of Colombo contain the so-called Dutch Records. The VOC archive covering the period from 1640 to 1795 includes the administration in Galle.

Figure11. Map of Galle Harbour showing the wreck of the Hercules
From these documents, an important part of the maritime history of Sri Lanka can be reconstructed, making it possible to answer detailed questions about the organization of the VOC’s shipping.

In the 17th century, because of the lack of a suitable natural harbour at Colombo, Galle acted as an important port for transhipment. The Bay of Galle during the NE monsoon offered a safe and protected anchorage for large ships. An intensive network of local trade and shipping existed from earlier times between the coasts of Malabar, Coromandel, Bengal and Ceylon. The supply of local products by inter-regional or inter-coastal trade took place in smaller vessels like the Ceylon-built dhoni, European sloops or old worn-out European yachts, like the Avondster.

The Avondster was wrecked in 1659. On 23 June, the ship was anchored near the Black Fort. During the night, in fine weather, the vessel slipped her anchor and hit the coast NE of the anchorage. Soon after grounding, the old ship broke in two and was submerged in the soft sand. The circumstances of the wrecking were unclear. An eye-witness account, found in the Dutch records of Colombo, tells how a sailor on the deck of the ship discovered the vessel drifting and tried to wake the skipper. However, the skipper was slow in making his appearance, and by the time he ordered the warp anchor to be thrown out, it was already too late. After the disaster, the skipper and the first mate were arrested, convicted and ordered to pay the damage.

The Avondster had been loading cargo for a trip to the coast of India. After the ship was lost, there were no other VOC vessels available to transport the rest of the cargo, which was still on shore, to India. The VOC officials decided that the burghers (free citizens) should be allowed to buy the cargo, but only on condition that it was sold for a fixed price.

Two other ships wrecked in Galle Harbour, the Hercules (1661) and the Dolfijn (1663), are representative of the inter-Asia trade. The Hercules was wrecked as she departed from Galle with a cargo for Batavia. The Dolfijn was lost within sight of the harbour, coming from Surat. Both vessels belonged to the class of yachts that were specially designed for the Asian trade.

In its first fifty years of operation, the VOC developed several types of ship suitable for the different trade areas in Asia. In the Resolutions of the Directors of the VOC in the Netherlands, one can trace the development of the organization of shipping during the 17th century. Originally the Dutch sent large ships to the Indies to collect cargo from all around Asia before returning to Europe. This had many disadvantages, and was soon replaced by a system whereby the VOC operated from a permanent headquarters and central hub in Asia.Figure 12. The Hercules bell found near Gibbet Island in 1992.

In 1619 the Dutch captured the city Jacatra on Java, renaming it Batavia, and it was from here that the VOC’s Asian shipping was organised. Large ships were sent from Europe to Batavia, to collect cargo brought in by smaller ships from all parts of Asia. For many years, the yachts which coursed these routes proved to be also suitable for the trade to India and Ceylon.

The Hercules was such a yacht built for the inter-Asia trade. She departed for Asia in 1655 and stayed there until 1661 when she was wrecked on the rocks near Closenburg on the north side of Galle Bay. In the early morning of 22 May 1661, a small fleet of four ships was ready to sail to Batavia. The fine weather influenced two officials who decided to take the ships out of harbour before any adverse change. There are several reefs in the area which lie hidden just below the surface, making it impossible for ships to enter or leave harbour without a pilot at the helm. After the flute Elburg and the yacht Tholen were guided out of the Bay, the pilot returned and made preparations to take the yachts Hercules and Angelier out of the bay together. An eyewitness aboard the Angelier gave an account of what went wrong after the ships weighed anchor:

"When the crew of the Angelier had weighed anchor and were busy pulling up the sails, quite suddenly a strong cross-wind struck the ship. We managed to fasten the sails again and to throw the anchor. On the Hercules however, who was half a pistol shot from us, things went wrong. I saw that the anchor rope was broken. This seemed strange to me, since this rope wasn’t bad and no other ship in the bay at that moment had the same problem. Still they tried to throw the second anchor, but in this case the end of the rope wasn’t secured to the mast so they lost the second anchor too. Without anchors the ship was now a playing ball of the elements. The bow of the ship turned in the direction of the land and was breaking to pieces on the cliffs a few moments later" (ARA VOC: 1238–9 folio 1137–40).
After the disaster, the pilot was interrogated. He explained that the cross-wind struck them when they were in the process of weighing the anchor; in the ensuing panic, the rope of the anchor stuck between the ship and the rudder, with a catastrophic result. The complete cargo was lost: 1700 packets of fine cinnamon and a cargo of Canarase rice destined for Batavia.

Apart from the type of ships used for the inter-Asia trade, there are also questions about the equipment of the ships. The VOC had strict regulations concerning the equipment of ships sailing between Europe and Asia. From the regulations within Asia, less information is available so far. An interesting issue is the arming of the ships. On the wreck site of the Hercules, thirty-one iron guns were found. This was many more than normal for a ship of that size. However, a list of ships present in Ceylon in 1657, found in the Dutch records in Colombo, also shows a picture of heavily armed ships (SLNA 1/9 Council Minutes 1657–63, 29 Nov 1657). It is clear that the VOC followed a flexible policy on ship armaments, and the ships were more heavily armed if they sailed to areas where enemies were active (Stapel, 1927: 507).

The maintenance of the fleet and the crew was one of the biggest concerns of the directors in the Netherlands and Asia. In the first years of the 17th century, most of the ships made a round trip to Asia without proper maintenance in Asia. With the establishment of the VOC headquarters in Batavia, a permanent repair facility also came into being. With every ship sailing to Batavia the VOC sent large quantities of spare parts and maintenance products, such as masts, sails, nails, and tar. To use the ships as efficiently as possible the directors in Batavia introduced a ‘quality registration system’. Every year the ships were inspected. Ships in a good condition were allowed to make the trip to Europe, moderate ships were not fit any more for such a long voyage but could serve in the China– Japan and India–Arabia routes. Inferior ships were only used for short journeys to the Spice Islands.
Despite the better maintenance and monitoring system, ships were often wrecked or abandoned because of the bad condition of the ship or crew. The history of the yacht Dolfijn is an example of this.

This vessel sailed from Sualijs (Surat) for Batavia on 29 April 1663, after loading packets of yam and letters. Shortly afterwards, the crew found that the ship was leaking badly; even with two pumps they were not able to keep her dry. So they returned to port and examined the ship. After sending the skipper, the high boatswain and the constable’s mate below they found the leak in the powder magazine on the port side. The powder was unloaded and the carpenter was able to repair the leak. On 30 April the Dolfijn sailed for the second time. When they were passing het Hoogelant van St. Jan the ship was strained by heavy sea conditions. Again the crew had to pump day and night to keep her dry. On 3 May the skipper discussed the difficult situation in the scheepsraad (the council of officers). They arrived at a VOC post along the coast of India and asked the merchant Zandtvliet onshore for assistance. They asked for 20 to 25 locals to pump the vessel in case of emergency. This request was unsuccessful because the locals asked too high a price: one pagood (local currency) per month, free water, firewood, and rice. They even asked for a galley and for six months’ pay in advance. So the vessel left without assistance but with 32 packets of amphiaen for Coetegin [Cochin]. The Dolfijn arrived there on 10 May lek maar behouden (leaking but safe). Once again a request for assistance was turned down. The ship was told to sail on to Galle. It was on this part of the voyage that things went really wrong. On 14 May the ship was at 6° 10' North in bad weather when the leak became worse. The skipper decided to anchor in 13 fathoms of water in order not to miss the Bay of Galle. To keep the ship dry, the crew had to install a fourth and an fifth pump. They were so exhausted after constant pumping that they were not able to lift the anchor. After cutting the anchor rope, the Dolfijn sailed along the coast to the Bay of Galle. In the entrance of the bay they anchored and fired several guns as distress signals. The situation became untenable, since even five pumps were not enough to keep the ship afloat and buckets were also needed. The only sensible thing to do was to sail the ship into the bay as quickly as possible in order to save the crew, money and cargo. Again the problem was to lift the anchor because the crew were either fully occupied with pumping or were completely exhausted. Another problem arose when the pilot came on board. He explained that it was impossible to enter the bay because the ship was lying directly in front of a shallow reef and the wind was not favourable. Aware of the seriousness of the situation, the VOC sent assistance from the shore, but by the evening the only thing they could do was abandon ship since the galleries at the side of the cabin were already striking the water.

An eye-witness account of the disaster was found in the dairy of Adriaan van der Meijden, who was a high VOC official in Galle at that time.

"Shortly after the afternoon a ship came sailing in the direction of the bay. Because it was firing its guns constantly we assumed it was in distress. Originally we thought that is was the Archilles sailing from Persia. We gave the pilot Bastiaen the order to go to the ship so far possible because the wind was strong and showery. After several attempts the skipper and former pilot of this Bay, Daniel Harthouwer, succeeded on getting aboard. Eventually the costly ship the Dolfijn was pitiful wrecked in the dark evening [...] The Thonij (pilot vessel) brought the under merchant Meijndert Janssen ashore. He went back on board after handing over the letters from Suratte" [...] (SLNA, Dutch Records: 1/ 2712)
According to the letters the Dolfijn carried a rich cargo. To protect the cargo in expectation of salvage the guards on the klip bij d’ vlaggespil en aen’t nieuwe puntje […] received the order to stop ships nearing the place of the wreck.

Galle was not only an important emporium, but a key staging post for European arrivals and departures. Ships normally arrived from the Netherlands in October or November, before going on eventually to Batavia, and left Ceylon for the Netherlands either in November or in February. For this regular shipping between Europe and Asia, the so-called retourschepen (returning ships) were used. These ships were larger than the ships in service in Asia. They could carry large numbers of people and large quantities of equipment on their outward journey to Asia, and large cargoes of Asian products on their way back. Retourschepen arriving in Galle unloaded their products for Ceylon, underwent repairs, and loaded commodities for their onward journey.

In the Bay of Galle two larger Retourschepen have been wrecked: the Barbesteijn (1735) while sailing into the harbour, and the Geinwens (1776) while sailing out of the harbour. The Barbesteijn sailed on 18 October 1735 from Colombo to Galle. The normal procedure was for ships to anchor outside the Bay and wait for a pilot. Because the weather was too bad to enter the harbour, the Barbesteijn waited for a number of days. In the morning of 22 October the ship broke one of her anchors and drifted in the direction of the shore:

"When the rope of the small bower broke we were driven to land. We fired distress signals, in order to get assistance from the wall. [...] There was a strong force S with a high running sea from the SW, at the second glass both anchors broke and we fell with the bow to the NE in the direction of the shore, against all [instructions], we went into the bay with God’s blessing. We were able to keep her above the blaasbalg (the SE corner of the bay), we sailed through a rough sea, the current dragged us into the bay. We used the foresail and spritsail and prepared the two kedges. We dropped the anchors inside the klip of negen (submerged rocks in the middle of the bay), [...] we still couldn’t hold the ship we dropped another kedge, in the end we tried to stop the ship by throwing guns on ropes overboard, we also saw a boat with anchors but it couldn’t reach us, after a little while we hit the bottom. This was about eight hours, till nine hours we were not leaking, due to the constant bumping we first made water and later also sand. At the evening there was water up to the first deck; that night we made water up to two feet between the decks. In the meantime they tried everything from the shore to help us, it was useless the vessels and the thonij capsized in the breakers."
The Barbesteijn may have been found during the magnetometer survey in February 1997 on the beach of Oenewatte.

The Geinwens was preparing for the homeward journey; she loaded cinnamon and cloth in Ceylon. On 23 October the ship struck a submerged reef during her departure to Coechin to load pepper. Within some hours the ship was floating again but heavily leaking. The master attendant of the Harbour decided to beach the ship to investigate the damage. It turned out that the keel was broken and hence the ship irreparable. After stripping the ship as much as possible she was ballasted with stones and sunk in a shallow part of the Harbour. The idea was to use the ship as an emergency stop for ships drifting towards the shore [SLNA 1/949].

The Galle Harbour Project offers the chance to study VOC shipping in its Asian context. The ships discovered represent different aspects of the Dutch trade with Asia; being well-documented and from an important period they shed light on the function and activities of the Harbour.  [NEXT PAGE]

 WWW Virtual Library Sri Lanka