The Theseus Complex: A Retrospective on the Design, Build, and 1990 Sea Trials of the Olympias

Kat Moore

University of Arizona

Movement and migration is an essential part of the human experience both today and in the past. We commute to work with bikes, buses, and cars. To explore new places, we can hop on an airplane and be in a completely different county in a matter of hours. We could even hop on a boat and make pit stops at various destinations. This type of casual exploration that we enjoy today would be impossible without technologies developed in the ancient past. Seafaring today relies on a host of modern developments, engines and navigation systems that rely on electricity. Importantly, the skills of craftspeople of the past and their knowledge built to bring us current ocean-traversing abilities. 

One of the most notable seafaring vessels is that of the Greek trireme, or trieres, the ship style that endured from the late Bronze Age to the 18th century (Coates 94). Though the trireme is probably the most well-known ship from antiquity, no trireme has been recovered.1 These ships were most likely around the end of the 7th century BCE, their size, number of crew members, speed and cost presumably increased from their maritime predecessors like the two-banked warship (Coates 94; Casson 81). Eventually, the trireme would overtake its predecessors as the premier warship by the late 6th century BCE (Casson 81). Archaeological evidence for the origins of these ships is relatively scant but seafaring is prevalent throughout the works of ancient authors. One such author, Thucydides, attributed the Corinthians as the first Greek builders of the trireme (Casson 81).2 The trireme was the preeminent instrument of naval power in the Mediterranean during the 5th and 4th centuries BCE. Various sources detail elements of Athenian construction and the city’s creation of their fleet and eventual maritime dominance under Themistocles (Mcarthur 484). Access to material resources significantly contributed to the continued power of the Athenian navy, holding power over timber imports being one such example (Mcarthur 487). Archaeological evidence for shipbuilding as an Athenian industry primarily comes from the site of Piraeus, a port located 7 kilometers from Athens proper. It is from Piraeus that the size of these warships can be determined, an essential piece of knowledge if one were to attempt to assemble one of these storied vessels.

1. In 2018 the Black Sea Maritime Archaeology Project discovered a wreck that is a contender for an extant trireme. The ship materials are too fragile to recover, but radiocarbon dating dates the ship to the 4th century BCE. 

2. The Phoenicians are also contenders for the origination of the ship style, as attributed in chapters of the Bible. 

To undertake the reconstruction of the past is an enormous task, to do so accurately is another beast in and of itself. For archaeologists, reconstruction primarily occurs through the interpretation of materials left behind, an analysis of abandonment and assumption. To engage in experimental archaeology is yet another way to evaluate the past. The inclusion of “experiment” may seem odd as an archaeologist would need to produce a testable hypothesis to fit the general criteria of an experiment. But importantly studies of the past lend themselves to imitative experiments, tests that seek to capture the processes of people who lived. Robert Ascher, who wrote on this burgeoning field in the early 1960s, highlights that this form of research grapples with the questions of how “matter is shaped, or matter is shaped and used, in a manner simulative of the past” (793). The simulative ability of experimental archaeology encourages a researcher to imagine a human-driven process rather than what remains. Or, in the case of the experimental trireme, The Olympias, what ancient authors tell us what once existed.

The Olympias is an experimental trireme developed by scholar John Morrison, naval architect John Coates, and writer Frank Walsh. In 1982, the three worked together to establish the Trireme Trust and three years later the building of the ship commenced with the aid of the Hellenic Navy (“The Trireme Trust”). The ship took two years to build by craftsmen in Piraeus, from 1985-1987. The Olympias underwent multiple sea trials in 1987, 1990, 1992, and finally in 1994 for the project documentation. The ship was transported to Britain to celebrate 2,500 years of democracy and used to transport the Olympic Flame. Trips were organized with amateur rowers and passengers from 2016-2018 (“The Olympias Trireme Sails Again”). After 2018, however, the ship was deemed as not being seaworthy due to a lack of upkeep and is exhibited in a dry dock at the Naval Tradition Park. 

The project sought to evaluate four goals for the experimental trireme. First, Coates and Morrison aimed to finally resolve arguments about the design of the vessel. Many scholars debated the ordering of the oar system and believed that the three-level system to be unseaworthy. Instead, many proposed different orientations of rowers as seen with the three rowers on one level in the extant 16th century trireme alla sensile (Shaw 2). Second, and the primary research focus, the experimenters wanted to test the performance and limitations of the vessel at sea. If the vessel was seaworthy, then their third focus could be tested: the simulation of naval operations in antiquity.3 Last, in their documentation of the process and publication of the data, the researchers aimed to highlight the skills needed to create and operate this iconic Ancient Greece war ship.

Figure One: Image of the “The Lenormant Relief” in the Acropolis Museum. Accessed 23 April 2024. https://www.theacropolismuseum.gr/en/relief-depicting-trireme-lenormant-relief.

With ancient evidence considered it was also necessary for Coates to balance the outlined historical and modern requirements. To meet the agreed historical requirements, 170 oars between 9 and 9½ cubits long needed to be assembled (Shaw 21). The body of the ship would have a maximum beam of 5.5 meters with a maximum length of 36 meters long, the length of the ram not included (Shaw 21). The hull of the ship was to be built before the innerworkings, typical of ships in antiquity but different from our modern shipbuilding techniques today (Shaw 21). In addition, the hull would need to be designed to accommodate for the hypozomata, one of the features that secured the lightweight ship against the water. Though exact rigging and sail information was notably absent from the ancient sources, the sails would still need to only utilize the oars to be used on passage only in insufficient or contrary winds (Shaw 21). Additional considerations more in line with comfort and storage were also an essential part of balancing historical and modern aspects of the vessel.5 The experience and comfort of the oarsmen, in particular, was paid attention to. Oarsmen seats feature greased cushions, protection in battle by screens of leather or matting, and required a specific height and size for the thalamin airports which were closed by leather ascomata (Shaw 21). Storage considerations were predominantly focused on the sails as the masts and sails needed to be stored within the vessel in anticipation of conflict (Shaw 21). The modern considerations were less detailed, more like guidelines. Meeting historical requirements was the most crucial however Coates and Morrison included estimations and desires of their rowers. The Olympias was designed to enable 170 men to pull and maneuver with oars effectively while also being constructed as short as possible to fit all the rowers (Shaw 21). General strength requirements were also included in the design as the trireme needed to also be strong enough for beaching and slipping (Shaw 21). Furthermore, the mass and wetted surface area needed to be as small as possible for optimal performance under oar (Shaw 21). Across both requirement types, Coates and Morrison put forth that the time to build the Olympias would have ideally modeled the actual capacity of Athenian shipbuilders of the past, with an estimation of 100 full vessels able to be constructed in a year (Shaw 21). However, as the project moved into the building phase, other considerations were undertaken that possibly accounted for the two-year construction of the Olympias.

3. One could argue that Coates and Morrison effectively ran two different experiments: the imagining and building of the Olympias and the sea trials of the warship.

4. Aechylus, Herodotus, Thucydides, and Xenophon are some of the more notable literary figures that Morrison and Coates drew inspiration from. In addition, segments from The Odyssey played a corroborative role for general functioning of ships.

5. Not discussed in this paragraph are the Herculean efforts of Owain Roberts, who designed the sailing rig for the Olympias without the scant sourcing that Coates had access to. As his own process could be and is an entire publication on its own, it is important to note that Roberts built the rig mostly off analogy with reference to the survival of merchant ships. 

Before the vessel could begin full construction, it was necessary for the team working on the trireme to provide proof-of-concept models. These models included a partial, full-size mock-up of the oar-system (Figure Two). As mentioned previously, debates over the oar system for the ancient trireme were at the heart of the controversy; hence, testing the feasibility of the structure was essential to even move forward with the experiment. Additional models included a small-scale model of the whole ship and ram, a hull model tested in a ship tank in Athens, and a one tenth scale piece to secure funding. In various publications, Coates and Morrison emphasize that material tests and structure tests were critical in finalizing aspects of the trireme design, most critically with stress tests of the plank tenons to evaluate weight capacity of the ship they aimed to build (Shaw 27). Cost estimates for the build of the trireme are hard to pin down, however Coates and Morrison pointed to the scaled piece of the trireme as the most expensive aspect of the project, which had a starting cost of 700,000 

pounds (Shaw 27). Through these models and material tests, the trireme team was able to secure funding and confidence from their investors to begin the build in 1985.

The material considerations and necessities for a modern recreation of an ancient ship were essential to the building of The Olympias. Athenian dominance of maritime trade assured that different timber was sourced from around the Mediterranean to maintain their fleet (Mcarthur 484). However, the materials of the distant past proved much harder to obtain than anticipated; Douglas fir was utilized in the hull planning instead of more regionally proximate pines and the laminated oak was replaced by the cheaper and stable iroko (Coates and Morrison 88). The timbers for the framing system for the trireme’s hull proved to be a challenge as the specific cut of wood for structural integrity was substituted with iroko as well to maintain the anatomical importance (Coates and Morrison 88). To resist strain under repeated use, the fastenings—prone to bending—were made of Quercus virginiana, the hardest known oak (Coates and Morrison 88). In fact, only the oars were crafted from the more authentic silver fir (Coates and Morrison 89). Once completed, the Olympias was 36.8 meters (m) long 5.62 with a beam of 5.5 m wide, a draught of 1.25 m, a displacement of 47 tons while full and 25 tons when empty (Grimm 252-253).6 

In addition to changes in the material sourcing for the wood of the trireme, changes from ancient materials to modern materials were undertaken. Most notably, linen ropes were deemed too dangerous to be employed on The Olympias as they intended to operate the ship. In tests with linen ropes, Coates discovered that linen, as a material type, relaxes under a consistent amount of tension. As such, daily tightening and even switching out the ropes would have been critical for the operation of these ancient vessels; however, for practicality and the anticipated upkeep of The Olympias the hypozoma ropes were substituted for low-stretch polymer (Coates and Morrison 89). Safety was a major concern of the build and continued to be important in the sea trials, as the introduction of people into the actual space of the trireme brought about additional safety issues.

6. See Figure Three for Coates’ plan of the Olympias. It is worthwhile to note that Coates and Morrison retrospectively note that the trireme should have been closer to 40 meters in length. For their design purposes, the length of the trireme is fifteen times longer than deep so by increasing the length the hull and oar system would have been larger, thereby providing the rowers with more room for operation.

After its completion in 1987, the trireme set sail, undertaking the first sea-trial that very same year. These trials are what make this experimental archaeology project incredibly ambitious in design and in practice; one could even argue that the trials themselves are a separate experiment entirely. Building the ship was in and of itself a feat however the primary goal of the experiment was to “rediscover [The Olympias] performance and to learn what was involved in her operation at sea” (Coates and Morrison 87). Although the Olympias set sail in 1987, the 1990 sea trials are the most expansive in record when the project was still headed by Coates and Morrison.

Figure Four: Chart of the 1990 Olympias Long Proving Couse Voyage. Image from The Trireme Project: Operational Experience 1987-90 Lessons Learned, editor Timothy Shaw, Oxbrow Monographs, 1993, page 41.

apart from a single thalamin, Charles Hirschler (Shaw 39). Minor repairs to the Olympias occurred before the long proving course trial and a sound system was outfitted to aid communication between the pulling masters and the crew (Shaw 39).

On day six, the Olympias’ voyage to Tolo commenced, provided with an escort, the HN Kriti, by the Hellenic Navy. The initial journey to Tolo was completed as planned in three days, with the return to Poros changed from the planned three days to one, mostly due to the uncomfortable and unprepared accommodations on the HN Kriti. The first three legs of the journey were from Poros to Metochi (14 nautical miles), Metochi to Ververouda (19 nautical miles), and Ververouda to Tolo (17 nautical miles) (Shaw 40-41). While the changes in the expedition plan allowed for simulating different conditions for data collection, the choice to not complete the same course back to Poros was a missed opportunity in terms of replication. In fact, replication of conditions was one of the main struggles of this experiment and experiments of this type. Weather conditions, especially changes in wind direction and speed, impact seafaring with extreme prejudice. The voyage back to Poros was especially emblematic of this experience, as seen in Figure Four, where the sailing conditions varied from calm to light wind and more consistent wind speeds (Shaw 41). In fact, the Olympias needed assistance from the Kriti, eventually being towed by the larger, modern vessel until released to finish the voyage into Poros’ channel (Shaw 41-42).

By the culmination of the expedition to Tolos, essential data on the operation and maximum speed of the simulation of an ancient vessel were able to be recorded. With three full levels of rowers without sails at a maximum of 30 strokes per minute (spm), the Olympias reached 5.4 knots (Shaw 40). With two levels of rowers under sails at 39 spm, the trireme reached 7 knots, closer to the historical estimations (Shaw 40). It is more accurate to note that this result was achieved with two levels rowing for 4 and a half hours with alternating support as rowers rested in the thalamin seats for 20 minutes (Shaw 40-41). Though maximum speed was a critical aspect tested for in this experiment, this long trial also illuminated additional aspects of the operation of this ancient warship. 

Comfort and safety were essential considerations that emerged as problems with this longer trial. Discomfort certainly arose due to conditions outside of the operation of the vessel, notably regarding the accommodation of the participants in the trial. The first night spent on the HN Kriti was uncomfortable and revealed an ill-preparedness best encapsulated by Shaw’s comment that “unfortunately the caterers in Kriti did not expect the crew of the Olympias to be as hungry as they were, nor were the toilet arrangements able to cope with so many people” (40). Later stops noted that some participants stayed on the Kriti, some slept in hotels, and the most enthusiastic were able to obtain permission to sleep on beaches, just as soldiers of antiquity might have (Shaw 40). One could argue that these adverse living arrangements might have added to the authenticity of the trials, as adversity is an aspect of warfare. One could also call it poor planning. 

Discomfort also emerged on the Olympias, especially for those unfamiliar with the seated position of rowing. However, the North American contingent brought seat cushions, enough for the 170 rowers, made of slow springback foam that were designed by NASA (Shaw 39). These comfort items were essential, as Shaw noted that “the pads were a godsend as they practically abolished the problem of blisters on the rowers’ behinds” (39). While briefly touched on in Shaw’s commentary, the health and comfort of these rowers reveal important information about the impact of operating machines of war on the human body.8 In consideration of the human body, the safety of the participants was an important element assessed in these trials, seemingly headed by the requirements of the Hellenic Navy. For the long trial, all participants were required to wear lifejackets; in addition, disaster drills prepared members for what to do in case of a hull breach (Shaw 40, 43). In designing these drills, it was found that the thalamins were most at risk if the trireme were to be swamped and as such a future reconstruction would need an escape hatch for these participants.

7. This estimation may be off as the records of distance in the experiment encountered some instrument difficulty. However, this calculation is the sum of the legs of the journey doubled to account for the journey back to Poros.

8. If exit interviews with the participants were recorded, I was unable to find them. But having a record of the perspective of the operators of these machines is an element sorely missed in this experiment.

The Olympias is a feat, but what do the experiments tell us? Well, critically, the three-level design is seaworthy and operational, and 170 rowers can operate the vessel with guidance from a pulling master. Boris Rankov’s future trials replicate and surpass the top speeds achieved in the earlier trials. With these technical aspects achieved, what might be missing from the experiment? 

Notably absent from the Olympias project are the records of the assembly of the ship and who exactly the shipbuilder in Piraeus was. As highlighted earlier, to test the abilities of a vessel, the ship needs to be built. As all major contributors to the project are mentioned and named, it is odd that there is a glaring absence of such an essential figure in official documentation of the experiment. Coates and Morrison make mention of advisements they undertook from the shipbuilder they worked with but unlike the sea trials and the design phase, the dearth of records of the two-year building process renders the building of the vessel a bit moot, especially for those interesting in the production processes of shipbuilding in the past. This lack seems a bit shortsighted but is understandable within the confines of the research context. The primary goals of the experiment focused on testing function over the processes of production. Following Ascher’s model, it is evidence that the simulative element was focused on the experience of operating the trireme not on the assembly of the ship. Hence, the lack of information on the shipbuilder of Piraeus who spent two years on this vessel. 

Attention and attitudes drive documentation and publication. As this build was one of the first large-scale forays into experimental archaeology, the focus driving the project differs from processes valued and explored in archaeology today. Where 40 years ago, the function and ability of objects of the past was prioritized, now the steps of making are focused upon. But it remains that without the documentation of the build we’ve once again lost the opportunity to corroborate and evaluate the steps of creating an ancient trireme. Where researchers might turn their attention now is to the repair of the vessel as the retirement of the Olympias in 2018 necessitated maintenance for its use as an educational tool. It is through the continued upkeep of this vessel that researchers can tackle an oft-forgotten element of production: the care and preservation of human tools and knowledge through multiple generations.

Ascher, Robert. “Experimental Archaeology.” American Anthropological Association, vol. 63, 1961, pp. 793-816.

Casson, Lionel. Ships and Seamanship in the Ancient World. Princeton, Princeton University Press, 1971.

Coates, John, and John Morrison. “Authenticity in the Replica Athenian Trieres.” Antiquity, vol. 61, no. 231, 1987, p. 87. ProQuest, https://ezproxy.library.arizona.edu/login?url=https://www.proquest.com/scholarly-journals/authenticity-replica-athenian-trieres/docview/1293639266/se-2.

Coates, John, and John Morrison. The Athenian Trireme. Cambridge, Cambridge University Press, 1986. 

Coates, John. “Reconstructing the Ancient Greek Trireme Warship.” Endeavour, vol. 11, 1987, pp. 94-99.

Grimm, Oliver. Großbootshaus – Zentrum und Herrschaft: Zentralplatzforschung in der nordeuropäischen Archäologie (1.-15. Jahrhundert), Berlin, New York: De Gruyter, 2006. https://doi-org.ezproxy4.library.arizona.edu/10.1515/9783110202847. 

McArthur, Mills. “Athenian Shipbuilders.” Hesperia: The Journal of the American School of Classical Studies at Athens, vol. 90, no. 3, 2021, pp. 479-532. Project MUSE, https://doi.org/10.2972/hesperia.90.3.0479.

Trireme Olympias the Final Report: Sea Trials 1992-4: Conference Papers 1998. Edited by Boris Rankov, Oxford, Oxbow Books, 2012. 

The Trireme Project: Operational Experience 1987-90 Lessons Learnt. Edited by Timothy Shaw, Oxford, Oxbrow Books, 1993.