Mediterranean Bronze-Age Style Galley Under Construction

This item just appeared on the WoodenBoat Facebook page, quoted here in full:

"Adem Ali Yılmaz tells us that this is being built at a high school in Turkey and that the dimensions are length 19 meters, beam 4 meters."

I don't know the provenance of the design: it appears similar to ancient Greek ships, but similar designs were probably used by other nations throughout the Mediterranean. The Persian ships at the battle of Salamis, for example, were roughly similar to the Greek ships against which they fought, and Turkey lay within the Persian empire at that time.


See my comments about the bow of Tim Severin's Argo, then look at the details of the bow construction here. The keel appears to be laminated. The structure of the stem, however, looks pretty authentic. On the other hand, I believe that Bronze-Age Mediterranean boatbuilders used no forms in their shell-first hull construction. If that's the case, then the use of forms and ribbands here is not a historically accurate method.


That issue aside, this is an ambitious and impressive project for a high school and I wish them luck and success. I sure would like to know and see more about this project.

How Accurate Was Tim Severin's ARGO?

The Argo in Istanbul. All images in this post are from The Jason Voyage by Tim Severin. Click any image to enlarge.

In The Jason Voyage: The Quest for the Golden Fleece , Tim Severin describes his recreation of the legendary Greek's quest. As in his other books detailing his reenactments of other legendary voyages, Severin's main objective is tell a rousing tale. While his reenactments all seek to demonstrate that the original voyage described in the legend was feasible -- and thereby lend credence to the legend -- Severin doesn't overdo the argument. He retains his credibility by never pretending to be doing serious science, the way that Thor Heyerdahl did. Instead, Severin combined an  adventurer's urge with a fascination for voyaging legends and a skill at writing exciting first-person narratives into a tremendous career which I can only envy.


To the extent that a date can be ascribed to any truth that lies behind the legend of quest for the Golden Fleece, 1300 BC is the best guess. But no Greek ships of that era have ever been recovered, and the only evidence of their form comes from sketchy, stylized illustrations found on pottery of the era. We do have firm archaeological evidence for Greek ships from several hundred years later, however, and that evidence drove the design of the new vessel, which was drawn by British naval architect Colin Mudie, who had designed the curragh for Severin's previous project, The Brendan Voyage (also discussed in a prior post).


While Jason's Argo had 50 oars, according to the epic poem of Apollonius, Severin's Argo would have but 20, making the vessel considerably smaller and more economical to construct and campaign. Severin felt that if a 20-oar vessel could complete the voyage of reenactment, that would more than demonstrate the feasibility of the original legendary voyage. 

As discussed in a previous post, the form of Greek warships of even Homer's date is debatable, so the form of a ship of Jason's era is even less sure. With this uncertainly permitting flexibility on Mudie's part, he designed the vessel with an eye toward safety, with fairly full lines, high gunwales and even higher ends. The new Argo was an aphract -- a vessel with all the oars on one level -- 54' LOA, 9' in beam, and 3' of draught. As shown above, the hull is lovely in profile, and only moderately slender in plan view. 


All this is a kind of preface to asking: just how close did Mudie come to an accurate reproduction?

Argo struggles against the current in the Bosphorus -- the most difficult rowing of the voyage.

Ships with 20 oars -- the smallest mentioned by Homer several hundred years following Jason's quest -- were used "for ordinary transport and dispatch work," according to Lionel Casson in Ships and Seamanship in the Ancient World. Dispatch-carrying, of course, demands speed. But the new Argo did not prove notably swift. Although driven without too much trouble in calm airs (ten men pulling could move the boat at 3 to 4 knots and make 20 to 30 miles a day), Severin found it nearly impossible to make progress against the slightest headwind, blaming the windage of the high bow. Could Greek military units  have abided such delays in communication, or were they willing to sacrifice safety -- and accept the more frequent loss of a dispatch -- by giving their 20-oared dispatch-runners a lighter, narrower, lower, and generally sleeker hullform? (In contrast, 50-oared ships -- later known as penteconters [i.e., "fifty-ers"] were the "common troop transport" of Homer's day. I suspect that their greater horsepower would have failed to make up for their greater burthen in terms of sustained speed over long distances.) 

Severin holds the construction model of Argo. Note how the extended bow appears to be added on to a bow of more conventional form.

Looking at the profile drawing, the ram-shaped bow appears to be added-on to a conventional bow. From the upward run of the planks near the keel just abaft of the hawsehole, it looks as if the bow has a conventional run below the waterline, and that the extended bow was created by overlaying "cheeks" of partial strakes between the extended stem and the hull's main strakes where they begin to curve inward at the bow. I suspect that a real, conventional, aft-curving stem lies hidden beneath the triangular cheeks, and that the forward-curving stem is a false stem that ties into an extended keel. This appears to be confirmed by the construction model held by Severin in the photo above. (Severin's shipwright followed this model and did not work from plans.) One wonders if the area inside the cheeks, between the two stems, was filled with buoyant material or left hollow as a watertight buoyancy chamber. Either way would appear to be a breeding ground for rot.


I have never seen any depiction or description of ancient Greek ships that suggests this approach, and I don't know of any reason to believe it correct. The more logical approach by far is for the hull's strakes to run directly to the visible stem, with no hidden stem and no cheeks filling out the hollow waterlines. I can only guess that Mudie and Severin thought they'd have better luck finding a Greek shipwright who could build a conventional bow and then add a false front to it, than one willing to experiment with an unfamiliar reverse-curving stem and the attendant unknowns about how to plank up to it.

Argo under sail. A single helmsman manned the two tillers, moving them in opposite directions to steer.

With a shallow keel and no other lateral plane, Argo performed poorly under sail, unable to maintain even a broad reach until it was discovered that by moving virtually all stores and crew onto the after deck, pointing ability improved by 15 degrees. And this was without the heavy bronze fitting that would have tipped the ram on a real warship. Argo was also difficult to turn, the extended bow tending to keep the boat heading in its current direction regardless of the position of the dual side-rudders. (By the way, both rudders broke twice during the voyage, although this may have been a failure of execution, not design.) It's interesting that Viking ships managed with a single side-rudder although, without the extended bow, they were probably easier to steer in the first place.

Planks were joined edge-to-edge with free tenons ("mortise tongues") in matching mortises in the facing plank edges. All was held in place with wood plugs through the planks and the tenons.

As in  Homer's  vessels, the hull was built shell-first, with frames added only after the planking was complete. Plank-to-plank fastening was done with the mortise-and-tenon method, as illustrated above and discussed in a previous post. Although it is an effective fastening method, the high intensity of labor that it requires is well recognized. Severin disagrees, however, claiming that it was quite efficient. But one gets the sense that he says this as a tribute to his shipwright's skill, for he also acknowledges that, where the shipwright handled every other aspect of construction single-handedly, using only hand tools, the free tenons were cut by a helper, who also did the mortising with the help of a machine. 


The "Jason voyage" was successful in its objective of traveling from Greece, across the Aegean Sea, through the Dardanelles, the Sea of Marmara, and the Bosphorus, and along the entire southern coast of the Black Sea, to Poti, Georgia -- a distance of some 1,500 miles. (And Severin later used it  to recreate the tale of the Odyssey in The Ulysses Voyage: Sea Search for the Odyssey.) But partly because the questionable accuracy of the boat in which the voyage was accomplished, the validity of any conclusions that may be drawn from the voyage are questionable as well. 


This is not to imply that the new Argo was poorly designed or in any way deficient. We simply don't know enough about the ships of Jason's era or their performance, and Severin's experiment provides a useful benchmark against which future research can be compared. And all technical issues aside, the book is still a fine adventure, stirringly and sensitively told.

An Abandoned Planking Method

It may seem out of place to examine the shipbuilding methods of the ancient Greeks and Romans in a blog whose subtitle is "small craft outside the Western tradition," given the fact that so much of Western culture is directly descended from those early civilizations. But the boatbuilding methods they employed ultimately went nowhere -- they died out and ended up contributing little to the boatbuilding traditions that we currently recognize as Western.

"Traditional" Western boatbuilding is characterized by one of two planking methods: carvel and clinker (or lapstrake). Carvel is the smooth-skinned method of planking, with the planks set edge-to-edge and attached to a pre-assembled framework. In clinker boatbuilding, each plank overlaps the one below it. Clinker construction proceeded shell-first. That is, after the keel, stem, and sternpost were assembled, the planks were fastened to one another, after which the ribs and other internal stiffening structures were installed. (Some modern artisan builders of lapstrake boats set up a complete temorary framework before planking.)

The ancient Greek and later Roman method differed from both of these. Planking was laid edge to edge, for a smooth skin like a carvel hull's. But construction proceeded shell-first, as in clinker construction. Most interesting was the method used to fasten the planks to one another.

A section of plank recovered from an ancient Greek shipwreck, showing tenons on both edges. Note how they overlap in places. From Lionel Casson: Ships and Seamanship in the Ancient World
CLICK ANY IMAGE TO ENLARGE

Left: schematic showing the method of mortise-and-tenon joinery, with pegs to fix the planks to the tenons.
Right: a plank recovered from an ancient shipwreck, split down the middle, shows how a large percentage of the plank was taken up by the mortises and tenons.
Both images from Lionel Casson: Ships and Seafaring in Ancient Times

As can be seen in the images above, both edges of the planks were deeply mortised, and held in alignment by free tenons. The planks were then drilled through the tenons at both of their ends, and tight-fitting dowels were driven through the holes to hold the tenons in place.

In this amazingly labor-intensive fastening method, the mortises took up a large percentage -- often more than 50% -- of the length of each plank's edge. Sometimes, the mortises were staggered alternately toward the inner and outer faces of the plank, so that they actually overlapped, thus totalling more than 100% of the length of a plank's edge.

Having been used for thousands of years in trade, war, and explortation, it could hardly be said that this method of shipbuilding was a failure. But it was ultimately superseded by methods far less time-consuming to employ. The clinker method was developed in Scandinavia. The source of the carvel method is subject to dispute, but it appears to have originated in the Mediterranean independent of the Greek and Roman tradition. Both ultimately overlapped in northern Europe and made their way to the Americas.

Proposed construction cross-section of a Greek two-level fighting ship. This ship would have used the mortise-and-tenon planking method described here. From Lionel Casson: Ships and Seamanship in the Ancient World   

Homer's Ships - Two Views

Since few of us read much academic literature, and even fewer read the literature of more than one field of interest, our views of the distant past are necessarily colored by the interpretations of experts who do. Where experts agree, and where their interpretations have been promulgated in popular literature or other media, our views may be about as accurate as those of the experts', albeit considerably less nuanced and detailed. But sometimes, even the experts' best guesses can be wrong: for example, tyrannosaurus rex used to be depicted with an upright posture, based on the consensus view of paleontologists. But with more research, that view changed, and the consensus is now for a more birdlike posture, with the torso held horizontal. The new consensus has been sufficiently popularized so that anyone under the age of 30 or so probably pictures the "new" version of the beast while, for some of us over 50, it's hard to shake the image of the upright t.rex we knew and loved in our youth.

Left: tyrannosaurus rex, Beta version
Right: the new, improved t.rex 2.0

Where the experts disagree with one another, however, we non-experts are left to choose among competing visions based on incomplete evidence. Obviously, the experts' evidence is incomplete too, or they would not be in disagreement. But at least they're disagreeing on the interpretation of the best available evidence, while we non-experts can only guess at who's right based on a fraction of the available evidence and, more likely, on which expert's argument is more persuasive to our untutored intellects.

That's why it's hard (for me, at least) to choose between two different interpretations of Greek warships of the 8th century BC, by Lionel Casson (in Ships and Seamanship in the Ancient World) and Bjorn Landstrom (in The Ship: An Illustrated History). This was the period when the Iliad and the Odyssey were first written down. When Homer's epics were composed for oral recitation is a matter of dispute, however. Some believe it may have been only 50 years or so before they were committed to writing; others think it was much closer to the time of the events they describe: some 400 years earlier. In any case, we'll be looking here at the graphic evidence for the vessels from the 8th century BC. Whether that graphic evidence was an accurate depiction of the vessels of Homer's time (whenever that was), and whether Homer was accurate in his descriptions of the vessels of Odysseus's time, are both different questions.

Casson describes the aphract as a vessel with a single bank of rowers, based on painted images on pottery like the one below, labelled "Casson #1." Note the ram bow, the small platforms in the bow and stern, and especially the structure between the ram and the bow platform. This appears to be a shield for a warrior on the platform, and perhaps it is also meant to deflect spray from the rowers. The raised stern would seem to perform the same functions. Although it is obscured, there appears to be a horn-like decoration mounted atop the bow "shield." Note also the white band that runs horizontally beneath the rowers -- we'll address this detail below.

Casson #1: An aphract, a galley with a single bank of rowers, mid 8th-century BC. (Click any image to enlarge.)

Following is Landstrom's lovely illustration of a vessel very much like the aphract. (I "cleaned up" this illustration using a free-download design tool, removing extraneous and overlapping images from the picture for clarity, and knitting together two separately scanned halves into a single image. In doing so, I have made Landstrom's image somewhat less crisp and lovely, but I have not altered his interpretation. I apologize for the amateurish quality of the alterations.) 

Greek galley, from Landstrom. This dugout-based boat seems to logically interpret every feature of the aphract pictured on the pottery fragment in Casson #1 above.

Note that Landstrom's vessel, while lovely and undoubtedly fast, could not possibly be a seagoing ship. It is clearly a boat suited only for the most protected waters. But the oddest thng is that Landstrom's illustration was based on Greek originals that, though similar to the aphract pictured in Casson #1 in many respects, were apparently more substantial vessels.

See the illustration below labeled Casson #4. Casson interprets this illustration as a two-banked warship -- i.e., one with rowers on two levels. Landstrom, in contrast, insists that it is a single vessel with two lines of rowers -- i.e., two rowers per thwart, sitting side by side. A third interpretation is possible: that Casson #4 represents two vessels side by side, seen overlapping in perspective. I make this interpretation based on what clearly (to me) appears to be two bow rams. If my interpretation is correct, then I do not think that the arrangement of rowers in the two boats can be definitively interpreted. While the artist may be showing an early use of perspective (although Landstrom says that this technique was not known at this time), he is almost certainly not attempting to depict such details as rowing arrangements with literal accuracy.

Casson #2: Warship ashore; two views on the same piece of pottery.

Casson #3: Warship cruising with upper level manned

Casson #4: "two banked warship"

Casson believes that Casson #2, above represents a decked vessel, and that rowers would have sat below the deck. (Oars are not in evidence because the vessel is beached.) But while the presence of a deck seems clear, the black hull is virtually unchanged from Casson #1. Is it possible that, rather than a deck/platform, the grid of fine lines above the black hull represents rowing outriggers as shown in Landstrom's illustration?

Casson #3 represents, in Casson's view, a two-banked warship, with rowers shown on the top level only. The broad white area between the main hull and the level on which the rowers are shown would appear to represent a lower bank where more rowers might be seated -- perhaps during a sea battle, leaving the upper level free for combat, and protecting the rowers from missiles. But if we interpret the broad white band as a lower level, then what do we make of the narrow white band in Casson #1?

In Casson #5 below are two more images showing, in Casson's interpretation, the bows of two-banked galleys. Graphically, these are still very similar in most respects to Casson #1, (showing the same ram, "shield," horn ornament, and overall hull shape) but Casson interprets them as quite different vessels; biremes with semi-enclosed rowing cabins vs. open single-banked vessels.

Casson #5: Note: i) the horns mounted on the bow shield, ii) the round, "spoked" decoration on the bow shield, and iii) the apparent presence of an upper deck, i.e., two banks of oars.. 

Here, then, is Casson's interpretation of the vessels shown in Casson #3, #4, and #5:

Casson #6: reconstruction of an 8th Century BC bireme: a true warship with two banks of rowers.

Casson's and Landstrom's interpretations couldn't be more different. Landstrom's is a dugout canoe -- a simple boat not even extended vertically with the addition of washstrakes. Casson's is a full-fledged, sea-going ship, built plank-on-frame, with an upper deck consisting of a fighting platform that runs the length of the vessel amidships, and thwarts to each side of it, upon which a second bank of rowers could sit in good weather or anytime when not engaged in battle. There may be a keelson, but it is very slight, and two wales have been added for longitudinal strength. Note how the round, spoked decoration -- merely painted on in Landstrom's version -- is a large hole pierced through the otherwise solid timber bow, with actual spokes added for decoration. The horns -- actual antlers on Landstrom's dugout -- are probably fabrications on Casson's vessel meant to represent (immensely oversized) animal horns. I doubt they would be real elephant tusks, for their presence in the illustrations is too consistent while elephant tusks would, I presume, have been extremely rare in 8th century BC Greece.

The plank-built ship of Casson's interpretation is obviously more capable than Landstrom's dugout canoe of having crossed the Aegean Sea to make war on Troy. But there is a grace to Landstrom's illustration of a fast dugout that is so appealing that one hesitates to reject it entirely.