Ships and Navies

[Sharur]

"The Age of the Galley: Mediterranean Oared Vessels Since Pre-Classical Times" (Conway Maritime Press, 1995) deals with everything concerning galleys and galley-like ships and their precursors in the Med; archaeology, theory and reconstruction. It's fully referenced and an amazing textbook on its subject, with many technical discussions. Robert Gardiner is "merely" the editor for a team of 15 other authors in this case. The book is volume two in Conway's monumental twelve-volume "History of the Ship" series.

The relevant notes about it from Amazon's webpage for the 2004 edition, http://www.amazon.co.uk/The-Age-Galley-Mediterranean-Pre-classical/dp/0851779557 today are:

"The Age of the Galley charts the development from the earliest paleolithic craft to the classical tireme and its Roman and Byzantine successors. As a warship, the galley survived the coming of the three masted sailing ship, and later chapters are devoted to its mediaeval and Renaissance successors. An unprecedented line-up of over one hundred leading maritime historians and specialists from around the world has been assembled to ensure that the work is informative, authoritative and fully international in it outlook. Essays are also included on related but more general aspects and themes of the period under review such as material resources, battle tactics, shipbuilding, gunnery, exploration or technology."

There's also a useful customer review on the same page.

Note that although Amazon have John Morrison listed as its editor/author, he was actually the Consultant Editor (zoom-in on the cover illustration to find out!). Also, the hundred-plus experts from the blurb refers to the number of authors for the entire History of the Ship series, not just this text...

[Erpingham]

Does Gardiner's book feature the creation of models, or is it limited to engineering theory on paper?

Depends what you call models.  Can't see any physical models on display but there is evidence of computer modelling, especially the bits on oar mechanics.

If you can get hold of it, do.  You'll find a good selection of topic coverage, use of textual and artistic evidence, engineering studies and some good stuff on logistics.  It goes up to the Renaissance, so you get medieval and Byzantine stuff too.

The wikipedia page on Galleys has a good bibliography too (though the article still needs some work)

[Patrick Waterson]

Merci beaucoup: physical models was indeed what I meant.  My local library will be getting a visit shortly. 

Meanwhile, if I may trouble an owner of the book (or reader with a good memory), what are the volume's contributors' conclusions about the relative efficiency of single-man and multi-man oar systems?  My own simplistic view of the subject would suggest that a multi-man arrangement is not efficient because only the inmost man can pull the full distance, the others adding power for only part of the stroke.

Patrick

[Erpingham]

Meanwhile, if I may trouble an owner of the book (or reader with a good memory), what are the volume's contributors' conclusions about the relative efficiency of single-man and multi-man oar systems?

Yes, though the mechanics of multiple men-per-oar systems is more thoroughly explored for medieval and renaissance galleys.  The one for classical vessels is more of a set of initial notes.  Both confirm your thought that the second and third men on an oar don't put in as much to the stroke (the difference varies depending on the rowing system).  Shaw, who writes the classical chapter, believes that travelling any distance, the rowing team would have rotated stations.  Shaw gives some consideration to pull/push systems (pushers are less efficient again but may have been needed on some of the larger polyreme oars) and seated v. standing (he thinks classical galleys could use three seated men).  The reasoning (which is around stroke lengths, gearing etc.) I leave someone who understands rowing or naval architecture to explain more fully.

As I think I understand Shaw, the trireme gets the most efficiency from its rowers, vessels with double manned oars next and the power per man drops away from there (but with the caveat that a multiman oar team could rotate, giving greater endurance - although your primary set up almost certainly had a experienced lead rower in the No.1 position, as per renaissance galleys).

[Jim Webster]

I think we have to remember that naval engineering is always a balance (more armour, bigger guns, bigger engines, faster ship, more armour, bigger guns, bigger engines.............)

Do you want a ship that does 15 knots for half an hour or a ship that will cover 200 miles in a day?

efficiency is a tricky thing to get a grasp of in this subject, you can be wonderfully efficient and no use as a warship 

Jim

[Patrick Waterson]

Do you want a ship that does 15 knots for half an hour or a ship that will cover 200 miles in a day?

The ideal Hellenistic warship would do both: the first under oars (for combat) and the second under sail (for distance travel).  If a classical fleet wanted to get from point A to point B the rowers put up their feet and the deck crew put up the sails*.  Once you were at point B and about to fight, the deck crew put the masts and sails on shore and the rowers put their blades in the water.

*The hemiola and trihemiola were exceptions, being designed specifically to use oar and sail simultaneously, giving them a slight speed edge in most situations.  They were much prized for reconnaissance, and also much used by pirates.

Thanks, Anthony, for outlining Shaw's conclusions.  Very helpful. 

One reason why I am less than happy with hypotheses about multi-oar arrangements and oarsmen relieving one another in different positions (hmmm ...) is that oarsmen, as I understand it, were really only active in a fight, and trying to shift positions in the middle of a fight is not a recipe for good control of the ship.  Naturally, if the wind failed while undertaking strategic travel, the oarsmen would have to put their backs into it, but they would only need 'one-third power' available for cruising, so a trireme could have one bank rowing while the others rested with oars shipped, then change banks after a certain time interval and thus keep going indefinitely.  Multi-man oar systems cannot do this except by shipping every second oar, which constrains the stroke of those ahead and behind, or by rowing with the forward or aft oars only, which makes the ship take the water at an angle and increases drag.

While instinct is not the most conclusive of arguments, my instinct is that Hellenistic designers went for maximum efficiency, not second best.  One man per oar is most efficient, and design effort would, I feel, have focussed on getting as many oar banks into/onto a ship as could be managed in order to avoid becoming tangled up in the law of diminishing returns that seems to be the bane of multi-man oar systems.  It seems to me that the real bar to accepting the one-man-one-oar principle is not architectural but conceptual: until one can successfully visualise how the oar banks were organised, it is hard to envisage how multiple banks could have been arranged.  (In this connection one recalls that the late John Coates, the man behind Olympias' rowing arrangements, was confidently told by many academics that his understanding of the trireme as a three-banker could not work in practice - or so it says on the Trireme Trust website.  It is a pity he never managed to apply his skill and knowledge to reconstructing a quinquereme.)

It may be worth noting that in Polybius I.20, the Romans, who were quite familiar with triremes, were unable to envisage how to construct a quinquereme until an example fell into their hands.  This ship was copied and served as the pattern for every Roman quinquereme until late in the First Punic War when a significantly faster quinquereme fell into Roman hands, and copying that gave them a warship which won the war at the battle of Aegusa (Aegates Islands).

Polybius I.21 tells us that the Romans set up a special rowing school for quinquereme crews:

"They made the men sit on rower's benches on dry land, in the same order as they would sit on the benches in actual vessels: in the midst of them they stationed the Celeustes, and trained them to get back and draw in their hands all together in time, and then to swing forward and throw them out again, and to begin and cease these movements at the word of the Celeustes."

This implies (at least to me) that each man was responsible for his own oar, and the idea was to avoid getting them tangled.  What also seems to be implied is that rowing a quinquereme is a more precise and demanding process than rowing a trireme.

All of which suggests to me that there was more to a quinquereme than just a trireme with longer oars and deeper benches on the top row. 

Patrick

[Erpingham]

One reason why I am less than happy with hypotheses about multi-oar arrangements and oarsmen relieving one another in different positions (hmmm ...) is that oarsmen, as I understand it, were really only active in a fight, and trying to shift positions in the middle of a fight is not a recipe for good control of the ship. 

Naturally, if the wind failed while undertaking strategic travel, the oarsmen would have to put their backs into it, but they would only need 'one-third power' available for cruising, so a trireme could have one bank rowing while the others rested with oars shipped, then change banks after a certain time interval and thus keep going indefinitely.  Multi-man oar systems cannot do this except by shipping every second oar, which constrains the stroke of those ahead and behind, or by rowing with the forward or aft oars only, which makes the ship take the water at an angle and increases drag.

While instinct is not the most conclusive of arguments, my instinct is that Hellenistic designers went for maximum efficiency, not second best.  One man per oar is most efficient, and design effort would, I feel, have focussed on getting as many oar banks into/onto a ship as could be managed in order to avoid becoming tangled up in the law of diminishing returns that seems to be the bane of multi-man oar systems. 

Patrick

Firstly - rotation of oarsmen.  This is hypothetical - there doesn't seem to have been any evidence quoted.  However, he is describing arrangements for long range cruising under oars.  This was perhaps more common than you allow - galleys had a fairly basic sailing rig - in winds to light or strong or in the wrong direction, or the current was a problem (going up the Dardanelles for example) they used oars. 

Secondly, oarsmen may have rotated a bank at a time as you suggest.  They may have used quarter rowing like a renaissance galley - the oars were divided into two pairs of quarters, fore and aft.  The pairs took it in turns to row.  I don't understand your point about multi-rower oars not being able to row by the level.   The oar system on a three level quinquereme would be the same as a standard trireme - the oars would be longer, the hull wider and taller but it's rowing system would be the same.

I think you have missed Jim's point on efficiency.  You are assuming only one measure of efficiency - the amount of the oarsman's strength used in the stroke.  The taller a ship gets, the longer the oars have to be if the oarsman is to maintain a sitting position.  The longer the oar the heavier and harder to wield it is.  Longer oars and taller ships add weight.  Height shifts the centre of balance, which makes the vessel more prone to roll and rolling plays havoc with a smooth efficient rythm, as oars coming clear of the water mess things up.  Also, what was the ship for?  Agile, with a tight turning circle and fabulous burst speed or a steady platform to put marines and engines on?  So, I am tempted to agree they would go for the most efficient design, only I'm with Jim that they are juggling a lot more factors and making appropriate design compromises along the way.

[Patrick Waterson]

While it is true that the taller a ship gets, the longer the oars have to be if the oarsman is to maintain a sitting position, this is only part of the story.

Consider this picture:

http://www.mlahanas.de/Greeks/war/images/TriremePlan2.jpg

Note how the oars of the trireme are all the same length.  This miracle of geometry is achieved by placing the benches ever nearer the hull of the ship, and providing an outrigger for the top row of oars.  I shall extend my cervical-spine-supported anatomy to the extent of suggesting that this principle would have guided Hellenistic shipbuilders, so that what we should look for in polyremes is cunning outrigger arrangements that leave the oars all the same length.

With that in mind, spot what is wrong with this arrangement:

http://upload.wikimedia.org/wikipedia/commons/7/79/Penteres.png

Correct: instead of being staggered outward, the benches are directly superimposed, and the hull is too wide, too deep and too tall.

Now imagine the hull reduced to about half the cross-section in the illustration (in both dimensions) and the benches staggered as per the trireme.  What immediately emerges is that the lowest row of oars reach out the greatest distance: the oars of the upper banks meet the water inboard of those of the lowest bank.  The higher the row, the closer the oar-blade to the hull.

This brings its own law of diminishing returns: when an oar is held at a significant angle, it becomes harder to operate effectively.  Hence, beyond a certain number of banks one has to go outwards and upwards.  Outwards is preferable.  When one goes upwards the rule about keeping all oars the same length goes out of the window, as correctly observed.

From http://www.mlahanas.de/Greeks/GiantShips.htm:
The Leontophoros of Lysimachos was a "eight" (probably around 100 metre long ship) with 100 men rowing each file and therefore 800 men from each side, and with 1200 fighters on the deck and 2 commanders for the fighters.

This suggests to me a wide deck (bearing in mind previous discussions about how many Persians can dance on the deck of a trireme), as one might expect if covering an extensive outrigger arrangement.

Interestingly enough, the 'forty' of Ptolemy Philopator had the same 3:4 troop-to-rower ratio as Lysimachus' Leontophoros.  From the same site:

Athenaeus gives us a detailed description of a very large warship, built by Ptolemy Philopator (c. 244– - 205 BC). It was 130m (420 feet) long, 18m (57 feet) wide, and 22m (72 feet) high to the top of her gunwale. From the top of its sternpost to the water line was 24m (79.5 feet). In comparison the length of the Titanic was 243 m and the largest Oil Tanker around 485 m. It had four steering oars 14m (45 feet) long. It had 40 tiers of oars. The oars on the uppermost tier were 18m (57 feet) long. The oars were counter-balanced with lead to make them easier to handle though its size alone would impress and the broad deck would make a excellent weapons platform. It had a double bow and a double stern and carried seven rams, of which one was the leader and the others were of gradually reducing size. It had 12 under-girders 275m (900 feet) long. The ship was manned by 400 sailors to handle the rigging and the sails, 4000 rowers and 2850 men in arms for a total of 7250 men.

Note the 57-foot oars for the topmost oar bank, as long as the ship was wide.  This ship pushed the absolute limits of polyreme construction, and was more of a vanity piece, being moved, as Plutarch tells us, 'only with difficulty and danger'.  The sixteen, or perhaps the twenty, seems to have been the largest battle-capable warship.

The question is: how far is it practicable to go outwards adding rows of benches and outrigger supports before one has to go up?

Patrick

[Erpingham]

Patrick, I am always impressed by your powers of lateral thinking   I don't think I've ever seen the argument that you extended the rowers outwards before. 

Not being a naval architect, I can't answer your question.  What I can do is is raise the issue of evidence.  We do have a number of sculpted depictions of hellenistic and Roman warships which show the form of the outrigger.  It does not seem to fit the idea of of a tall overhang.  Another problem is none of the many illustrations we have of warships seems to feature anything with more than three tiers of oars.  Admittedly, we have no illustrations of ships with three oar levels saying "This is a five, six, eight etc." (AFAIK we have only one picture of a ship rated higher than three where it is written on it what it is - a Roman four - and it doesn't show the oar arrangement).  It is therefore possible that no-one ever chose to illustrate big ships. 

Overall, however, I think the current interpretation of multiple rowers to the oar best seems to fit the evidence available.

[Patrick Waterson]

Thank you, Anthony. 

If for a moment we look at the matter onomasiologically*, would not the classical Greeks and Romans have referred to a trireme as a trireme because it had three banks of oars (as opposed to two or four)?  Given that starting-point, would they not have distinguished a three-bank, five-bench vessel as something like a 'pentatrieres' rather than as a 'penteres', and referred to triremes with multi-man oar banks throughout as something like 'hypertrieres'?

*I am sure we all remember this means starting with a concept and looking at how it is expressed in words.

Painting myself into my own little corner , I shall hold with one man, one oar until either we find an original source reference to multi-oar polyremes or find incidental evidence that unambiguously points that way.  So far, I have encountered none (although in fairness it has not yet been a long journey).

It is quite possible that big ships were illustrated, but bearing in mind how few representations of the trireme have survived (this being the most common form of warship for most of the classical period) it is perhaps unsurprising that we lack depictions of the larger vessels.  We might still encounter useful evidence if a villa from the Augustan period belonging to, say, Agrippa turns up with floor mosaics of Actium, for example.  Coinage is frustratingly ambiguous, not least because die engravers undoubtedly had their own views about the amount of work involved.

Changing the subject slightly, I shall sneak in a word about ramming tactics (yes, ramming was an art in itself).  Mediaeval dromons tended to mount the ram high, above the waterline, prizes apparently being more desirable than kills.  In the classical era, the ram went below the waterline and made a hole in the part of the hull that lets in water.  Rhodians used to dig the bow in deeper just before a ram, backing water with their forward oars in order to make the bow dip just prior to impact.  While effective, this could be taken too far.

At the Battle of Chios in 201 BC (Polybius XVI.3) King Philip V of Macedon was using a dekeres as his flagship, and happened to ram a Rhodian trihemiola (sail-and-oar trireme), dealing it a fatal blow.  So far so good, but when his ship tried to back away, it found itself stuck - the dead trihemiola simply would not shift.  The problem seems to have been that the ram of the dekeres had smashed through the trihemiola's keel, so the two ends of the ship pressed in upon the keel and would not let go.  Philip's embarrassed flagship was sunk by two delighted quinqueremes, which happily rammed this helpless target of a lifetime without getting stuck themselves.  Fortunately for Philip he had left the fight to his admiral, Democrates, and so survived to lose to the Romans on land at Cynoscephalae four years later.

Patrick

[Erpingham]

Painting myself into my own little corner , I shall hold with one man, one oar until either we find an original source reference to multi-oar polyremes or find incidental evidence that unambiguously points that way.  So far, I have encountered none (although in fairness it has not yet been a long journey).

I do recommend you read a bit of Morrison and Coates work - there are enough classical references to keep you very happy.   However (spoiler alert) they do come down on the idea that the rating is based on the number of files of oarsmen.  This allows them to explain the hemiolia (one & a half) and triehemiolia (two & a half) as having one or two full files running fore-aft, with a half file in the centre of the ship.  You get a slightly more cumbersome (IMO) explanation if you use the number of men per half-room argument.  Anyway, I'll let you research some more and we can switch to ramming tactics

[Andreas Johansson]

If for a moment we look at the matter onomasiologically*, would not the classical Greeks and Romans have referred to a trireme as a trireme because it had three banks of oars (as opposed to two or four)?  Given that starting-point, would they not have distinguished a three-bank, five-bench vessel as something like a 'pentatrieres' rather than as a 'penteres', and referred to triremes with multi-man oar banks throughout as something like 'hypertrieres'?

My experience is, for reasons of callow youth, restricted to latter-day armaments industries, but I wouldn't feel confident in assuming the naming scheme necessarily made perfect sense. I would not be greatly surprised if it turned out that the "three" in trireme and "five" in quinquereme counted somewhat different things.

[Jim Webster]

I'd agree with this, we have to remember that names evolve and are obvious to those who are there at the time.
Also we have names which don't fit into the scheme because they have a different source, Lembus and Liburna for example

Jim

[Andreas Johansson]

I note incidentally that the (somewhat confusing) WP page on Hellenistic warships cites Morrison 2004¹ for the claim that "It is known from references from both the Second Punic War and the battle of Mylae that the quadrireme had two levels of oarsmen". Anyone with the thing at hand to tell us what the ancient sources are?

1. Morrison, John S. (2004). "Hellenistic Oared Warships, 399–31 BC", p7. In Gardiner, Robert. Age of the Galley: Mediterranean Oared Vessels since pre-Classical Times. Conway Maritime Press. pp. 66–77. ISBN 978-0-85177-955-3

[Patrick Waterson]

It will be interesting to see how the claim in Morrison (in Gardiner) turns out, as I am unaware of any such reference in the Battle of Mylae (264 BC).  (Polybius I.23)

As for Gaius Duilius, he no sooner heard of the disaster which had befallen the commander of the navy than handing over his legions to the military Tribunes he transferred himself to the fleet. There he learnt that the enemy was plundering the territory of Mylae, and at once sailed to attack him with the whole fleet. No sooner did the Carthaginians sight him than with joy and alacrity they put to sea with a hundred and thirty sail, feeling supreme contempt for the Roman ignorance of seamanship. Accordingly they all sailed with their prows directed straight at their enemy: they did not think the engagement worth even the trouble of ranging their ships in any order, but advanced as though to seize a booty exposed for their acceptance. Their commander was that same Hannibal who had withdrawn his forces from Agrigentum by a secret night movement, and he was on board a galley with seven banks of oars which had once belonged to King Pyrrhus. When they neared the enemy, and saw the "crows" raised aloft on the prows of the several ships, the Carthaginians were for a time in a state of perplexity; for they were quite strangers to such contrivances as these engines. Feeling, however, a complete contempt for their opponents, those on board the ships that were in the van of the squadron charged without flinching. But as soon as they came to close quarters their ships were invariably tightly grappled by these machines; the enemy boarded by means of the "crows," and engaged them on their decks; and in the end some of the Carthaginians were cut down, while others surrendered in bewildered terror at the battle in which they found themselves engaged, which eventually became exactly like a land fight. The result was that they lost the first thirty ships engaged, crews and all. Among them was captured the commander's ship also, though Hannibal himself by an unexpected piece of luck and an act of great daring effected his escape in the ship's boat. The rest of the Carthaginian squadron were sailing up with the view of charging; but as they were coming near they saw what had happened to the ships which were sailing in the front, and accordingly sheered off and avoided the blows of the engines. Yet trusting to their speed, they managed by a manœuvre to sail round and charge the enemy, some on their broadside and others on their stern, expecting by that method to avoid danger. But the engines swung round to meet them in every direction, and dropped down upon them so infallibly, that no ships could come to close quarters without being grappled. Eventually the Carthaginians turned and fled, bewildered at the novelty of the occurrence, and with a loss of fifty ships.

Can anyone see any mention of a quadrireme?  The only ship I see specified is the hepteres [seven-banker] the Carthaginians had acquired from Pyrrhus.

Patrick