Oh the horror, the horror..

The text of Osprey New Vanguard 225 Roman Republican Warship, is a horror show. I originally read it as a glad fan of ancient ships eager to glean anything new or different,  but..

The previous 7 blogposts I have made addressed the colour plates concocted for this book. As is the Osprey format it is these that sell the books. The text should also be examined. The main incentive for examining the text closely was provided by the examples of plagiarism I found in connection with the illustrations. What is lying in the long grass of the text ?

If one then reads the book as I would students' submissions then I am AMAZED that it came to press in its current form. That is, if it was proof-read or subject to editorial examination. I do not now believe that Osprey have a meaningful editorial staff. The author is writing in a second language and a lot can be accepted on that basis. Shakespeare he is not. However, grammar, punctutation and consistent style can all be marked down  a long way.

I am in the process of making a short summary of the book's text.

A is for Aaaaarrrggghhh!

To deal quickly with plate A from Osprey New Vanguard 225.

This FIVE is one of the best reconstructions in the book. Oh, but it is not Rava's work.
It is a straight copy from Jeff Burns' illustration for John Warry's 'Warefare in the Classical World' with some twiddly bits added.

Jeff Burns' quinquereme 1980

 This bireme -liburnian has an oarbox and too many oars. This ship would be 36metres long, the length of a trieres !

Still More Fish in the Barrel!

Let us now turn to Plate B of Osprey's New Vanguard 225. I will keep it as short as possible.

Plate B

Ship 1. A 'triremis' (sic)

There is a bolt-shooter mounted in front of the corvus. How is it lowered past this ?

The corvus itself is wrongly drawn - Polybius states it is secured on the pole by a slot in the plank floor of the bridge. Here it stands on its end.

There is no evidence for a corvus being mounted on a trieres - which is what this ship purports to be. trieres were so sensitive in their balance that a single man moving across the deck could disturb the stroke of the thalamian oarsmen. The weight of the corvus in raised position would give a lot of wind resistance and potential for overbalancing on this light ship.

The steering oar - pedalium - is not hydrodynamic. it has a thick leading edge and a strong rib down the middle.

The caption text is poor. It uses the Italian words for the rowers - not the latin nor relevant Anglicisation. 'zigiti' instead of zygians or zygitoi.' Talamiti' instead of thalamians or thalamoi. Traniti is replaced later in the caption by' Thranite'. This is a sign of poor or no proof-reading.

The term 'orders' is used for the different levels of oars. Why ? Maybe a mistranslation from the Latin ordo/ordines for ranks. Ranks of rowers ? Some writers do  use 'files' of rowers.

The parexeiresia is confused with the oarbox. The oarbox replaced the parexeiresia. The upper level of oars rested over a rail projected out from the ship's side - this is the parexeiresia. In the third century BC it was replaced by a fixed closed box built along the ship side and the oars were worked through ports in this.

A ventilation course is mentioned in the caption but completely absent in the illustration.

The most worrying error in the caption are  the references to oars.

It is stated that the ship has 170 oars. Why, then, is the ship drawn with 178 oars ? Why ?

Oar lengths are given as
traniti - 12 ft in length
zigiti - 'managed rows of about 10ft.'
talamiti - 'formed the lower row with oars 6ft long'.

Oars on a radius can be same length

Read carefully. I shall say this only once. The key breakthrough in working out how a trieres is constructed and rowed was the realisation that the oarsmen should be arrranged on a radius so that the oars were all the same length ! Read any paper or book about Olympias and you cannot avoid this fact.

Lower man with a 1,8m oar ?

Oh, yes and one could also think for a moment as to how useful an oar 6ft (1,8m) could be to a rower who is positioned so as to require at least 4,66m of oar to have an effect .





To finish, it is stated that the ship - a trieres - carries 120 soldiers. The most a trieres is recorded as carrying is 40 soldiers plus the deck crew. Confused with a Five ?



 Ship 2. A 'Quadriremes'(sic)

 Now, is this from Alba Fucentia - as stands on a previous photo of the graffito or Alba Lucentia ? The author  cannot make his mind up. Thanks, Osprey editorial staff.

This ship is very special. This ship defies all research of the last thirty years to resurrect a nineteenth century idea about ancient  ship construction. This ship is a 'Four', that is a tetreres or quadrireme.

Much ink and almost blood has been spilled to determine that the '4' in the name of this ship does NOT mean it had four banks of oars. On the contrary, a tetreres should have TWO banks of oars, each double-manned. How thi stype of ship can be illustrated this way in 2015 is difficult to understand. Unless the closely collaborating author and illustrator do not have a clue what they are doing and there is no editorial control.

There is also confusion here either in language or understanding, when a parados is referred-to as a 'guard-rail'. The parados is a narrow ledge outside the guard-rail where men can stand.

Again. something appears in the caption, but not the illustration - poor collaboration again. The latticed ventilation course is referenced but not drawn in the picture.

The last sentence is a pointer to the language difficulties of producing a book in a second language. BUT IT IS PRECISELY HERE THAT THE EDITORIAL STAFF SHOULD HELP !

'The upper oars (zygian) would emerge from the hull over the topwale while the oarports would serve the Thalamian(sic) oars.'

The very fact that the ship is drawn with FOUR oar rows destroys the terminology of thranian-zygian-thalamian.  In any case the sentence is confused and cannot be elucidated by looking at the picture. The upper oars would normally be termed thranite.

In fact, the top two rows of oars seem to come out of holes in the top of an oarbox ! ? Do they bend down to the water over the edge of the oarbox ?



 'I know nothing about the 'oars'

Curiouser and Curiouser...

Osprey New Vanguard  225 keeps giving.It gives one so much more faith in the illustrators of  Look and Learn and their ilk.

Salamis a la Look and Learn

 Despite the blurb, Osprey push out fodder for eager boys of all ages just as Look and Learn did.

Plate H is all about a Liburnian.

The Liburnian was a type of lembus. That is, a lighter ship. They were heavier than most lembi because they were cataphract - i.e. they had a deck which covered the oarsmen and hold, and they could be  'boxed-in' - i.e. the oars were worked through ports to further protect the rowers in battle and from the weather. Possibly those used by the Romans were inspired in their design by the vessels of Illyrian pirates. Pirates prized speed and manoeuverability in a compact, light ship that did not require a large crew.

The plate caption tells us that the reconstruction is based upon a liburnian of Agrippa from Isernia.
The relief fragments, from the Santa Maria del Monache National Museum at Isernia, on the eastern side of Italy, are shown, in cropped photographs on page 39.  Found in the River Sordo, near the Ponte Nuovo. Probably from a tomb monument, and to be dated in the years after the battle of Actium (31 BCE), there is nothing to link this relief with Agrippa as far as I can find.The only features the reliefs contribute are the prow volute ornament, the proembelion and the ram decoration.

If we examine the rest of the caption we learn that the ship depicted has 82 oars 'dispersed in two orders'(whatever that means), that its length is 'about 108ft(sic)', and that its crew is 114 oarsmen, 10-15 sailors and 40 marines.

No sailors or marines are depicted in the plate.

We also learn from the caption that  ..
'The ram which the word pointed suggests confirms the epithet  of Propertius as rostrata and  the presence of an armament to be used as occasion might demand for defence or offence.'

Now, call me a curmudgeonly old nit-picker but in 1995 John Morrison, writing in 'Hellenistic Oared Warships 399 -31BC' in Conway's 'The Age of the Galley' wrote thus..

'The ram, which the word 'pointed' suggests and which Propertius' epithet rostrata confirms as a regular characteristic of the liburnian, adds an armament to be used as occasion might demand for defence or offence.'

Morison's construction is a little odd. A ram is essentially an offensive armament. The chances that D'Amato lighted by chance upon the same phraseology is slim. Him being Italian and all that.

Earth to Osprey's editorial staff. Come in Osprey's editorial staff, Come in Osprey's editorial staff....

If we do the unmentionable and count things shown we find that there are, on each side, 20 single-manned oars and 21 double-manned oars on each side.

Firstly, there is NO evidence for differently-manned oars in a dikrotic ship unless we are talking about heavier polyremes from a Five and up. Dikrotic ships - biremes -  are known from the seventh century BC down to medieval times. The only suggested case for double-manned oars combined with single-manned is that of the hemiolia, where the mid-half of each side on a MONOKROTIC ship could have had double-manning. Plate H here has a radical interpretation which is not explained or backed-up. Or it is plain wrong.

Secondly, add 20 times 1, to 21 times 2. This gives 124 in my system of arithmetic. There are places for 124 rowers in this picture. The caption says there should be 114. Plain wrong.

With all those men crammed into the ship let us see if there is space for them.

Space to work the oars

Space in a galley is crucial. There must be enough space to work an oar effectively. We know from antiquity and from the Olympias project that the separation between oars should be in the order of a metre 21 oars on a single level needs 21 metres. The ship shown here is 108ft long (why feet ?). This is 32,4 metres. We do not know if this is waterline length or maximum length. If the distance between the outermost oar ports  is 20 metres this would give a ship, as depicted, of 27 metres hull length. The picture is too short. A Four is generally reconstructed as having 88 oars - 22 on one level. A Four is generally reconstructed as being 37 metres long at the waterline. Rather longer than this ship which has just about the same number of oars in a level. The reconstruction is plainly wrong.

The ship shown in plate H has an oarbox. This is a straight-sided closed extension to the hull in which oarports are pierced. Both levels or oars are shown rowed through the oarbox. No ship of this size could have two levels of oars rowed through the oarbox. There is not enough space available inside the ship to allow oars to penetrate the box side and hit the water without having wildly different oar lengths. The smallest ships with oarboxes so large are Fours or maybe trihemiolia.
The most likely configuration for a liburnian with an oarbox is that one level would penetrate the oarbox and the second issue from its underside. In fact, we have no evidence for a liburnian with an oarbox. They can be boxed-in in that they have sides with protection for the oarsmen but oarboxes ? No. In addition, it is essential that an oarbox is straight, because then the oars can be the same length. this drawing has the oarboxes tapered at the stern which defeats their function.
The liburnians from Trajan's Column speak volumes.

Incorrectly positioned oarsmen but no oarbox. These would have been decked and the lattice is ventilation in the sides. The men are sculpted larger than life-size compared to the ships.

Should I go on ? I will.

The profile shows a crennelated bulwark topping a layer of interlocked shields. Where is this structure in the plan view ? Is it represented by the red line running fore-aft ? If so, why would a protective deck bulwark be located where no one can stand behind it unless they tread on a rower's head ? How do they get the shields ?

The caption states that the liburnae were 'decked and boxed-in'. In the context of galleys, 'decked' usually means cataphract - i.e. the deck covers the ship. Plate H shows a ship with an incomplete deck. It is at odds with the caption.

All these details stirred some dusty entry in my mental archive. I checked New Vanguard 225's bibliography and, as one may expect, there stood John Warry's excellent 'Warfare in the Classical World'. In this book is an illustration of a Liburnian, by Clive Spong or Jeff Burns.

Liburnian from John Warry's Salamander book 1980

Clive Spong or Jeff Burns' Liburnian has 48 oars per side. A liburnian with more oars than a Four !

BUT Warry wrote and his illustrator worked before 1980. Much has been discovered since.

If we take the ratios of the oared-length to the total-ship-lengths on these illustrations we can see they are similar.  Rava 1.44,   Spong-Burns1,36.

If we compare the pictures we see a tapered oar-box, alternate single then double-manned oars, an incomplete deck with an arrow shape and a similarity in prow plan, one side of the oar benches unoccupied., no plan representation of the bulwark shown in profile.

It gives the impression of a colouring-in exercise on the earlier drawing.

If we compare the details given alongside the drawing it is conclusive.

Even though the number of rowers changes, the length is kept the same ?

Why would one do that when many interested readers will have Warry's book on their shelves ? How could one concoct a ship which is essentially a trieres - having three men in each interscalmium and call it a Liburnian.

It is not unknown for artists to 'reference' others' work. It is undersandable for an illustrator who has to churn out a lot of stuff quickly to use cut-and-paste. What is taking the piss out of the buyers and readers of Osprey books is when they boldly state

'expertly pieced together from written sources, archaeology and artefact evidence. With meticulously researched artwork..'

(Osprey New Vanguard 225 REPUBLICAN ROMAN WARSHIPS 509-27BC)

We know from Sr. Rava's comment,previously blogged about here, that the author and illustrator collaborate closely to produce the Osprey illustrations. Neither can avoid credit for Plate H.

Jackson, Embleton, Doughty, Lawrence et al who lavishly illustrated the weekly comics of the past were working, apparently, in a similar vein to those hired by Osprey so many years later. Much more is known but this is not always applied due to pressures of time. At least Look and Learn did not claim to be the last word in historical and archaeological accuracy. Look and Learn was an exciting weekly magazine for young people and the broader-minded adult. Is Osprey doing anything more?

Let's row away from this shipwreck (backwards) ! (Don Lawrence)

Oh , I almost forgot. Here is John Coates' drawing of a Liburnian. This is a light version, they could be a little larger, but most had 50 or so oars. his detailed technical drawing and its supporting evidence have been available since 1994. 'expertly', 'meticulously' .. give me strength...

Is this real ?

In response to a post from 2015 when I made a critical remark in the first review of New Vanguard225 I got this comment.

Im sorry, but if you think that the illustrators at Osprey arent controlled by the author of the book and the editor, youre completely wrong and you dont know what youre talking about. All technical detail are pedantically verified twice, firstly on the pencil sketch and on the color plate, after. Trajanic Testudo included, read by yourself the extract from the brief (italian language): "Gli schieramenti Romani visibili potranno essere: la testuggine di fonte 20 o le truppe disposte in avanzata di fonte 21;" Giuseppe Rava-Illustrator on Osprey : Republican Roman Warship


Translation of the Italian (google)...
'The visible Romani camps (fighters? .me) will be: the tortoise source 20 or the troops stationed in advanced source 21'

The 'source's must be some reference material for the artist to work from.

I do not know if it was really from Sr. Rava.

If it was then it proves that the execution of the artwork was as it was meant to be after the author's requirements. The artist has followed a detailed brief and is only responsible for the style of his delivery which has, presumably, been accepted when he was signed-up for the book. Sr. Rava has illustrated a lot of books and several for Osprey.

One must conclude that it is actually Sr D' Amato who has the interesting interpretations of the ancient material.

And, of course, the Osprey editorial staff.

J 'syrACCUSE!

The saga continues. It is mainly because the Osprey NewVanguard225 has been screaming from my bookshelf for a year or so. It seems to take up more space than an encyclopedia. Partly becaue when I first saw the book I was simply glad the subject had been addressed. This I somewhat regret. I hope to lay the ghost by writing these blogposts.

 Last time plate D got it. The Quinquereme also illustrated in plate D is not error-free but for now I will jump to plate E- 'The Siege of Syracuse 212B.C.'.

What a dog's dinner. Apart from the dress of the sailors at the stern (see previous blogpost).

..warning for the faint-hearted

The most salient problem is the theme of the picture, which is the deployment of a sambuca against the walls of Syracuse.

If you are in doubt as to what a sambuca is look here.



The sambuca of the siege of Syracuse was an arrangement of ships and a ramp for mounting the walls and was so-named because it looked like the musical instrument. Note the soundbox - the ships - and the neck - the ramp - linked by ropes, the strings.

The caption to the plate is largely an excerpt from a translation of Polybius (Historia VIII,4).
By this means the caption author(s?) are hoist by their own petard. The contraption in the plate bears little resemblance to Polybius' description. The ship the thing is mounted-on is also odd.

OK let's go through Polybius' recipe and see how one should make a sambuca.

Nice try but no drinking while blogging!

 SAMBUCA DEL D'AMATO-RAVA
 1) Take ONE quinquereme.
2) A ladder four feet wide (1.2 metres ) with a side-railing.
3) The next bits need two ships. Here we have only one !? Skip over..
4) The affair is raised by men in the stern pulling lines which run through a block at the masthead. In the plate there are precisely four sailors in the stern. One is musing on the massive cable in front of him. The others are making themselves look busy to avoid having to pull on it. The end of the cable is indistinctly terminated in the deck. If you follow the cables to the masthead it is apparent that the sambuca itself is not connected to the stern. The cables illustrated are th emast-stays. The forward mast-stays terminate in thin air or off the ship on land? The sambuca hangs in the block suspended from the foremast and a single cable runs down therefrom to the deck. Unmanned. The size of the sambuca makes it unlikely it could have been the foremast that supported it. Polybius must mean the mainmast.In any casehe clearly says the sailors hauling it up are in the stern of the ships.
5) The platform at the end of the ladder was protected by wicker screens on three sides which were thrown off when the escalading troops rushed up to get onto the wall. The wicker screens are still in place in the plate.

SAMBUCA DEL POLYBIO
 1) On a pair of quinqueremes lashed together,
2)Mount over the junction a ladder 1.2metres wide and very long to project before the ships.The ladder is roofed-over and has side-railings.
3)Arrange functional tackle running from the ladder over the mainmasts to enable sailors in the sterns of the ships to raise the ladder with the aid of sailors in the bows who will use poles.
4)The ladder is equipped with a platform at the end occupied by four men protected by wicker screens on three sides.
5) When the platform is in place above the wall then the screens are thrown down and the main escalading party rushes up the ladder and onto the wall.


How long was the sambuca ?
To get maximum lift the ladder must have had a line fitted as near to its extremity as possible. Maybe 2metres behind the tip or immediately behind the platform. When elevated this point cannot have been raised higher than the top of the mast which was the fulcrum.

The mast top block of a quinquereme would be at about 12 metres over the waterline.
Solving the triangle for a hypothetical slope in action for the ramp of 45 degrees...

The sambuca was in the order of 17 metres long. Polybius says it projects a long way forward of the ships and so it does.

Putting all this together we get an arrangement somewhat like this...

Check this with Polybius

The sambuca in red projects forward as it lies ready to be raised.
The lines to raise it run from behind the landing platform over the mainmasts and to parties of seamen (B) in the sterns ready to haul it up.
In the bows (A) of each ship are parties of seamen with poles to help raise and position the ladder.
Troops wait on deck ready to swarm up the ladder when the landing platform is in place on the walls of Syracuse.

While plate E is down let us kick it some more.
The scale of the ship - a quinquereme - is gross. The timbers on the tower at the ship's stern are approximately 20cm or more thick which maybe they have to be because it is armoured with metal shingles and occupied by Chinamen and a bolt-thrower. The height of the deck above the waterline is well over 3 metres judging by the height of the men on the deck.

 It has a strange gangway built onto the side of the bow. There appear to be men marching up the bow ornament. The pedalion disappears through the oarbox which has two holes in it for some reason. In this situation the pedalion cannot be lifted out of the water by angling it back or outward to any great degree. There is no ventilation course of louvres or apertures for the oarsmen who would soon expire.

Plate E : Too-close-up

One should not look landward because the Syracusans are using one of those fairground crane toys to attack the Romans are are shining searchlights on them to set them on fire. The Romans fight back with geometrically impossible combinations of sambucæ.



On the back of this book - Osprey new Vanguard 225 is written ' With dazzling, meticulously researched artwork, it examines Republican Rome's warships...'
Maybe not.
Messrs Connolly and McBride must be gently rotating in their respective Eternities.

'THE TIME HAS COME' THE WALRUS SAID........

'Don't cry. It's only an Osprey.'

A few things to straighten out about ancient ships. Yes, still banging on about the Osprey New Vanguard 225 'Republican Roman Warship'. On the principle that bad medicine is better in small doses, here is another teaspoonful.

How Big is a Ship ?  OR  Is that an Almost-Trieres?

Palazzo Barberini ship (No relation. Ed.)

Plate D of the book has an illustration of a 'hemiolia-trihemiolia'.
'It has been reconstructed with oars at two levels.'

Said to be based upon a mosaic at the Palazzo Barberini mosaic and the Palazzo Spada stern along with the Lindos stern and the Samothrace prow.

The thing that hit me between the eyes was the statement that the ship is 'about 60ft long.'Sixty feet is about 18 metres! Olympias - a trieres- is 37 metres or so. hmmm.

Next the statement that the ship has' 52 oars a side'. He takes a calculation from Morrison ?(which I have never come across) that there were three tiers of oars 26+26+13. ? What ?  26 plus 26 plus 13 is 65. Last time I checked. I think it still is. And..'A quarter double manned '. What ?

Next one looks at the illustration and counts laboriously that there are shown..
26 oars in the upper row.
27 in the second row.
13 in the lower half row.
Giving 66.

Plate D from New Vanguard 225 - This is 60ft long.

This is a train wreck of a caption. On wonders if it is the author or the illustrator or both who can take credit. It also shows the woeful standard or absence of any technical editing or proof-reading at Osprey.

A hemiolia is a ship with one-and a half men per half-oar-room. The extra half gave extra speed but also allowed 1/3 of the crew to down oars and be ready to fight while the ship could still make way. It probably descended from pirate vessels. It had almost the speed and agility of a dikrotic vessel but required fewer crewmen (about 50 rowers) and was lighter. The hemiolia had all oarsmen on about the same level, the half-file being inboard in the widest part of the hull. It is possible the mid-half was double-manned instead of them having their own oars.

Trihemiolia

The trihemiolia is a ship with two and a half men per half oar-room. The extra half gave extra speed and allowed 1/5 of the crew to down oars and be ready to fight while the ship could still make way. It probably descended from pirate vessels and was favoured by the Rhodians. It had almost the speed and agility of a trieres but required fewer crewmen (120 rowers or so as opposed to 170 for a trieres).
The trihemiolia  had oarsmen on three levels as per a trieres but the upper two would be closer vertically than in a trieres.

And you cannot get 3 rows right ? - don't think about justifying this thing...

There was no 'hemiolia-trihemiolia'. Just as there was no 'trieres-'tetreres'. The point of the name is that a hemiolia is 'a ship with a half(extra) file of oarsmen. Halfway between a monokrotic and dikrotic pentekonter. A trihemiolia is ' a trieres with a half-file of oarsmen'. Halfway between a dikrotic ship and a trieres.

Reconstructing this hybrid ship with oars at two levels is meaningless. If it is a hemiolia it has oarports at two levels - one a half-row- or one, and oarsmen at the same level. If it is a trihemiolia it has oarports at three levels and oarsmen at three levels.

Lastly, the number of oars and the length of the ship.

Morrison and Coates estimate the oars on a hemiolia at 50 and on a trihemiolia at 120.

Rava-D'Amato state there were 130 on their hybrid. And draw it with 132. This gives 132 men in a crew which was trying to save on personnel whereas the trihemiolia had just 120. But wait ! We must add the 'quarter double-manned'. This raises the total of rowers to 162 or 163. A saving of just 7 or 8 men over a trieres, hardly worth it.

And length. Here, size is not everything but it is a lot. My initial bugbear is that it is quoted in feet. In a European publication in the 21st century.... It is really easy ,online even,to find the length of a hemiolia at 21metres and a trihemiolia at about 32metres. 60ft is 18 metres. 'Go figure' as they say in their American accents.

John Coates' trihemiolia

 And the deck is wrong. Oops..