Where is the Castle of the Grand Magne?

The puzzle?

In the year 1249 Prince Guillaume de Villehardouin was the Frankish master of vast swathes of the Morea (present day Peloponnesus or Peloponnese). The heart of his power was in the south of the Morea, except, that is, one intractable area, "…the people told him that the zygos of the Melingi was a great drongos and had strongly fortified passes and large towns and arrogant people who respected no lord." from The Chronicle of the Morea

Obviously this didn't please the Frankish Prince and he took council and this told him that he must build another castle similar to that at Mistra which he had just constructed to contain the Melingi in their mountains.

"Thereupon the prince himself made a tour on horseback, following the directions of the people of the land, and he passed Passavas and journeyed to Maine; there he found an awesome crag on a promontory. Because he found it very pleasing, he built there a castle and named it Maine, as it is still called." from The Chronicle of the Morea

Well yes, and no. In the time that the chronicler* was writing it would have been possible to identify this castle, generally called Grand Magne, but since the early 14th century, when The Chronicle of the Morea was written, we have lost the exact location of the castle. Such a lacuna in our knowledge causes scholarly fascination which leads to speculation and research and a smidge of controversy. Historians continue to debate and speculate on the exact location of Grand Magne with at least five locations in Mani claiming one or other scholarly champion.

* I'm not linguist enough to really go into the ins and outs of the debate as to whom the chronicler actually was - though I have read any number of articles on the subject. There are a number of versions of the Chronicle of the Morea - which basically covers the years 1204-1292. The Versions are ; Greek (in a verse form), French in prose, Italian and Aragonese and various odd fragments embedded (i.e. plagiarised) in other chronicles. The tone of the Chronicle is pro-Frank, it glorifies the invaders. This leads some to think that it must have been written by a Frank. But it was written in the 14th century when the Franks and native Greeks had been interbreeding for decades, therefore it is possible to have been written by a Greek then translated by a Frank. Or the writer and translator could be one and the same person - they were certainly aware of the societal patterns of Frankish Morea. There is even a contemporary medieval term for one of mixed Frankish and Greek blood.

Vardounia Castle on the eastern slopes of the Taygetus - medieval - probably Frankish

With the further establishment of Frankish rule in the southern Peloponnese in the 1240s and '50s on both sides of the Taygetus Mountains Guillaume de Villehardouin and his associates began to build castles. One could almost say that this was in their blood and they would have brought with them the traditions and skills of castle building that they had developed in France, Italy and the Holy Land. One of the major problems facing the Franks was less the overt but seriously weakened threat of the Greek/Byzantine forces but more the constant raids undertaken by the originally Slav tribe of the Melingi (there are many spellings and interpretations of this name - the most appealing is that as they were blonder haired than the existing population the name connates honey, 'Meli') who are asssumed to have lived in the foothills and mountains of the high Taygetus.

If the Melingi's early history is a subject for academic debate there is even less consensus on who they were by the 13th century. They appear to have been in the area which we call the Exo Mani (i.e. North West Mani) since the dark ages. Had they kept their Slav identity throughout the intervening 6 centuries? It is unlikely. And although some portray them as wild outsiders lurking in the mountains surrounding Greek towns and villages one has to remember the words of the Chronicle of the Morea which specifically mentions the "large towns" of the Melingi. The spread of Slavic place names in Mani (the tell tale -itsa ending is just one clue) and common sense point to the area being an amalgam of Greeks and Slavs. The latter were probably, by this period, speaking Greek and of the Orthodox faith, the area being converted in the 9th century. The Melingi had allied themselves to the Greeks in the defense of the Morea against the Franks some 50 years before. They were however very independent - a theme running throughout Mani history, and a thorn in the side of any foreign power who wished to create a hegemony in the area.

The typical way to dominate a contumacious and hill dwelling population in this period was to build castles that surrounded and encircled them. The Norman kings of England used this method with an eventual if success in Wales. The major castles we know that were established or reanimated from earlier institutions by the Franks in the area around the Taygetus and the Melingi were Kalamata, Mistra, Beaufort (Leuktro or Leftro), Passavas and Le Grand Magne. These are referred to in contemporary documents.

Seven other fortified sites will feature in this account: Zarnata, Kelefa, Vardounia, Kastro tis Orias near Kipoula, Tigani and Porto Kaiyo. See the map below for locations.

As Prof. J.M. Wagstaff pointed out in his persuasive paper on this subject, there are a number of conditions which must be fulfilled before a site can claim to be Grand Magne.

It has to be in Mani and to the west of Passavas

It has to be atop a fearsome crag

It has to be on a promontory overlooking the sea

It must make some sense with regard to encircling the Melingi and in relation to the other Frankish fortifications in the area

It should have some signs of 13th century building or archaeological remains

I shall start with a brief description of each of the fortifications around which everything revolves and then try and sketch out the salient points of the historical debate.

N.B. I've not revised this page for a number of years and in the interim I've read, thought more and visited more of the sites. It will therefore change in detail over the summer of 2005, though my conclusions have not altered radically.


Home of the Villehardouin family (Guillaume de Villehardouin was born here) this castle is still extant on a small rise to the north of the old town of Kalamata and now contains a modern outdoor theatre just below the main heights. The opening hours are vague and visitors, in my experience, few and far betweeen. It overlooks, to the NE the modern 'delights' of Kalamata bus station. The castle controlled the roadstead of Kalamata (there is no real port save that contructed by man) and the entrance to the Langada pass over the Taygetus to Sparta. It also would be able to control the passage into the north-western Mani across the foothills to Kardamili. It goes almost without saying that Kalamata was and remains the main city of the lush and fertile Messenian plain. There would have appeared to have been a castle here before the coming of the Franks and it is not a contender for the appellation of Grand Magne.


The castle at Mistra is built at the summit of a vertiginous hill. Below it on the eastern slopes was built the later city of Mistra. On the western and southern sides the ground drops almost vertically into a gorge which divides the hill of Mistra from the bulk of the Taygetus. It not only affords stunning views over the Lakonian plain but also is not far south of the spot where the Langada pass from Kalamata comes onto the Lakonian plain. It was chosen by Guillaume de Villehardouin as a site for a castle for precisely these reasons and that it was part of the Taygetus range - home of the troublesome Melingi tribe - and yet separate from that range due to the chasm to the west.


This is a teaser, as the name is fascinatingly vague, "Lovely Fort" anyone? Generally this is accepted as being known as Leuktro or Leftro. This castle is supposed to have been situated near to the small modern resort of Stoupa. The small landing place at Stoupa was probably less important to its development than its natural defensive position on a small, isolated but steep and flat topped hill 80 metres high. This outcrop dominates the plain of Gisterna (springs or cisterns) and is near enough at the end of the old route over the Taygetus via Milia and the Panagia Giatrissia pass. Built on the site of a presumably ancient acropolis there is little left of the castle and the site is extremely overgrown.


Magnus von Stackelberg sketched the view from just north in 1812 and a print of this appeared in his La Grèce: Vues pittoresques et topographiques. Paris. 1834. The notes which accompany this, identify the fortifications on the summit being one of a local Maniate chieftain, '...sur son sommet couronné de rochers sont des murs ruines et un grosse tour servant de residence au chef independante se la contrée qui, de haut de ce donjon fait flotter avec orgeuil son etendard …les murs de la forteresse en ruines ont d'une epoque plus rapprochée de nous, mais plusieurs futs de colonnes des inscriptions effacées des monnaies et des debris de statues semblent prouver que la etait l'Acropole de Leuctres, renomme du temps d'Homere.'

Beaufort depicted c.1812 by Otto Magnus von Stackelberg. Published in his volume of views of Greece. View from the north looking south over the plain of Gisterna.The same view (from slightly more to the west) 2005.

Nearby is the anchorage of Skala/Stoupa. After building Grand Magne to the south to control the drongos (area/district) of the Melingi - according to the Chronicle of the Morea, themselves pointed out to Villehardouin that a castle at Leftro would make the cordon even more secure - presumably there were divisions between different groups of Melingi and this lot wanted to curry favour. Beaufort was reported in the Chronicle of the Morea to be half way between Grand Magne and Kalamata - if one does the calculations Grand Magne should be somewhere near Itilo.

Vardounia or Bardounia

At the eastern side of the main Taygetus ridge Vardounia is situated just to the north of the small village of Ag. Nikolaos. It is not mentioned in any contemporary sources nor in many later histories of castles in the Peloponnese, save a short mention in Paradissis who ascribes it to the Venetians, a brief note in Leake, Finlay gives it a mention, and it is swiftly dealt with in H.A. Ormerod's description of Vardounia in the Annual of the British School at Athens in 1909. Ormerod at least says , "…the large castle of Bardounia close to Hagios Nikolaos testifies to its importance in the middle ages…" (ABSA 16. 1909-10. p.67) . In fact there is a mention of Vardounia in connection with the Kladas revolt of 1480/81 and Jacomo Barbarigo, a Venetian Provveditor of the Morea just before this wrote to his masters, The Signoria in Venice,

" The brave ser Manoli Kladas and Krokondilos his brother...continue to hold the castle of Vordounia, near Mistra, the key to the Mani. They have lost all their relatives in Your Signoria's service, and have sustained every labour and loss in your name."

The Kladas family were also Capi (Captains) of the Zygos area of Outer Mani - just the other western side of the pass which Vardounia castle controlled to the east. When Venice made peace with the Ottomans in 1478 they handed Vardounia back to the Turks. Much to the Kladas family's ire. They revolted - retook the castle and Zygos and were only chased away into exile by a large Ottoman force in the spring of 1481. A few years later Krokondilos Kladas returned and took the castle back under his control where it remained until his death at the hands of Ottoman troops at Monemvasia in 1490.

There is also mention of Vardounia in Evliya Celebi's accounts of his travels in the Morea in the period 1668-70 which have only recently (1994) been translated into Greek. Evliya gives an extremely accurate description of the castle and its location. He states, "It was taken by Mehmet The Conqueror". In other words this confirms that it was extant and occupied by the Greeks of the Desperate of Morea before the Ottoman conquest of the area in 1460.

Vardounia's main importance would have been in controlling the other end of the route over the Panagia Giatrissa pass from Beaufort. It is not a candidate for Grand Magne but its position has a bearing on the location of that castle.

Vardounia Castle

There is a recent article on Vardounia (Burridge 1995) which eloquently points to the likelihood that the architecture and siting of the castle indicates a probable Frankish occupation - though this has to remain speculation as differences between Frankish and Byzantine building techniques are blurred if not completely non-existent. The site is easy to find - though ask for simply 'To Kastro' - locals don't understand the Vardounia bit. Bear off to the west off the main Githeon/Sparta road and drive up into the red earthed foothills of the Vardounia Mountains. Stop in the very pleasant platea of Ag. Nikolaos and by walking to the edge of the square next to the local council building you will spot the castle on a crag to the north. There is an easy track to the foot of the castle crag but the undergrowth in and around the castle is extremely well established and as there are a number of cisterns in the castle extreme care is advised.


Zarnata is not mentioned as a Frankish castle and many historians believe, mistakenly, that it is primarily a Turkish fort - although the name certainly exists in medieval documents. However there are a number of reasons for believing that it may have been part of the Frankish cordon around the Melingi. Firstly it does have ancient, if not prehistoric foundations and Andrews is certain that that some of its enciente walls are of medieval if not specifically Frankish construction. Secondly it is mentioned by Evliya Celebi who witnessed the Turkish siege, capture and rebuilding of the fort around the year 1670, that it was first captured by Sultan Mehmet II. Therefore it predates the Turkish occupation of the Peloponnese in 1460. Thirdly it is at the end of another old road over the Taygetus which basically makes use of the Rindomo Gorge. Controlling the small but fertile Kambos plain would have been important as well as observing any forays from the Melingi in the mountains and one cannot see any controlling power in outer Mani leaving such a fortress unused.

Zarnata under Venetian control 1685-1715. The fortress was the centre of Venetian power in the Outer Mani during the 'Venetokratia'

P.A. Phourikes even postulated that Zarnata was Grand Magne which goes to show that it has to be added into the equation even if its inland site eventually excludes it from the likely candidates for Grand Magne.


Mentioned in numerous sources Passavas (or Passava - one reads both) controls the low pass between Areopoli and Githeon. Signposted beside the main road Passavas is on top of a (now) thickly wooded hill it is identified by Leake as the site of the ancient 'city' of Las. Whereas it has obviously been re-used many times by intervening ages including by the Turks, there is the stump of a minaret in the grounds, and by the Greeks in the War of Independence it is known to have been a Frankish castle built circa 1222 by Jean de Nully though probably built on the site of the earlier ancient acropolis. The enciente walls are still relatively intact but probably mostly Turkish and any traces of a keep have disappeared. The name is thought to come from the French war-cry"Pas Avant". After the disaster of the Pelagonian Fields, the capture of Villehardouin and his highly expensive ransom Passavas does not appear to have been part of any recorded hostage deal but when the de Nully family tried to re-occupy it they found it seized by the Greeks. It obviously is not Grand Magne due to its inland position and clearly identified name but has a bearing on the probable location of that castle.


Kelefa (the name means 'bald' or 'leprous') is atop a huge bluff overlooking the Bay of Itilo A large rectangular construction very similar to Passavas, it is usually ascribed to the rather inept 17th century fortress building skills of the Turks and was certainly one of their corks on the bottle of irascible Maniates. It's present layout is certainly Turkish and dates from the expedition of the Ottomans into Mani in 1670 when they re-imposed their sway over Mani. Evliya Celebi, an Ottoman official and luckily also a talented travel writer recorded how the Turks decided to dominate the bay of Itilo with a fortress and forestall the trade in Muslim slaves by the inhabitants of Itilo and to prevent the use of the bay by the Venetian fleet.

The interior of Kelefa today

Kelefa was seized by the Venetians in 1685 and lost again to the Turks in 1715 and at some stage after that was deserted. Not only does it control the western end of the vital pass between Areopoli and Githeon but also the bay of Itilo - one of the few half decent harbours on the western shore of the Mani peninsular (even then it is not a recommended berth for modern yachts). On a north south axis it commands any movement between the Exo and the Mesa Mani. If one takes the reference in the Chronicle of the Morea that the castle of Beaufort/Leftro was equidistant from Kalamata and Grand Magne then Kelefa fits the bill.

Porto Kaiyo

Porto Kaiyo is the safest of the deep Mani harbours and lies on the eastern side of a narrow isthmus and although the modern rather straggly village is unimpressive, on the north side of the bay are the remains of the castle. There have been some claims (Traquair for one) that this was Grand Magne mainly due to the understandable propensity for western engravers and mapmakers working on hearsay and their imaginations to place a fortress at the tip of the peninsular and call it after the area. In general the fort is not large and is thought to be of Turkish construction - possibly in the 1570s when the Turks made a big effort to control Mani. The position is good if one wants to dominate the harbour but has little impact on the surrounding countryside and could easily be dominated by land based artillery. Apart from which it is on the east coast of Mani - whereas Grand Magne should be on the west.


Towards the end of the Mani peninsular there is a large western bulge in its shape. This is the Cavo Grosso or Great Cape. Just before this is the relatively sheltered bay of Mezapos. This sheltered nature is assisted by the long mini peninsula of Tigani which sticks out from the north of the Cavo Grosso like a frying pan (from whence it gets its name). The long handle of Tigani is a low feature of broken, sharp stones and salt pans but at the northern tip there is a small rise - insignificant except from a boat, and this area was obviously once a walled town.

The Tigani - or 'frying pan'. The Byzantine citadel was at the end of this barren peninsula

There are signs of houses, cisterns and churches though little above the foundation level remains apart from some of the walls surmounting the cliffs. The impressiveness of the site has lead some observers to claim it as Grande Magne. But the remains are, if anything, a site of late antiquity or early Byzantine. The format of the churches, for example, is basilical - a style seldom after the 700s. The position, whereas relatively easy to defend, cannot dominate the surrounding deep Mani peninsular. If one wished to 'sally forth' from Tigani it would have taken a good half hour just to have got the the base of the promontory. Simpson and Waterhouse followed by Greenhalgh and Eliopolous tend to favour Tigani as the site of Grande Magne and others have followed suit.

Ano Poula

At the western edge of the Cavo Grosso is the Makryna ridge. Most of the eastern edge of the ridge is made up of lowish cliffs - on the western side it plummets hundreds of metres into the Messenian Gulf. Towards the northern end of the ridge overlooking the village of Kipoula is a complex of walls. Those on top of the eeastern cliffs have firing slits in them - though the walls are low and the firing slits even lower. My own feeling is that one is looking at an eighteenth century refuge for the locals and their flocks and that the slits are for muskets, not crossbows. However having walked over it with Panayiotis Katsafados I know he disagrees (in the nicest possible way). Panayiotis has written a scholarly book on the castles of Mani (Ta Kastra tis Mainis. 1992) and did a prodigious amount of field work and his feeling is that Ano Poula is Grand Magne. However the site is difficult to interpret from the many remains of past buildings dotted along the ridge, many of which are ancient (with a fair sprinkling of 11th to 13th century churches) and it could well be the site of the acropolis of ancient Hippola, mentioned in Pausanius. As pointed out to the immediate west are the vast cliffs of the Cavo Grosso - no landing place for miles and there are no walls on the seaward side. Ano Poula dominates the Cavo Grosso but its influence on the Sangias let alone the part of the Taygetus where the Melingi lived would have been minimal. Also one has to ask if it fits the description. If Guillaume de Villehardouin was riding round the Mani looking for a site for a castle it is unlikely he would describe this site as an awesome crag - unless that is, he had taken his horse aboard ship.

Kastro tis Orias

This is on the southern end of the Makryna Ridge, the highest point of the Cavo Grosso, above the village of Dri. Both Bon and Rogan plump for this site as the most likely site of Grand Magne. However the site is devoid of any major walls and quite bare. Panayiotis Katsafados has scoured the site and can see no justification for Bon's conclusion.

After all that let us concentrate on what Guillaume de Villehardouin was trying to do. He had a group of contumacious locals, living in the area which approximates to the Exo Mani, who are giving him grief. He decides to build a chain of fortresses around the area, both confining and dominating the area. There are keys passes and routes across the Taygetus The main passes then and today are between Kalamata and Sparta and Githeon and Areopoli. These would have been controlled by the castles at Kalamata and Mistra in the north and at Passavas and …well what in the south? The obvious answer is Kelefa. To make even more sense of the encirclement the Melingi themselves suggested a fort at Gisterna or Leftro. This controls the small but fertile plain but is also at the end of another pass. Nowadays not used greatly and having no proper road. But Ringlia just to the south east of Leftro/Stoupa is the start of a kalderimi path which crosses the Taygetus above Milia by the monastery of the Panagia Giatrissia and drops down via a valley to Githeon and the Laconian plain. This route is defended by the castle at Vardounia which Peter Burridge has pointed out has distinctive Frankish or at least medieval features.

Map showing the castles in relation to the supposed area dominated by the Slavic derived tribe of the Melingi - Passes over the Taygetus are marked in purple

So if we were Guillaume de Villehardouin we'd put a fort somewhere that supports the encirclement and additionally controls the southern pass over the mountains.

"…he passed Passavas and journeyed to Maine; there he found an awesome crag on a promontory. Because he found it very pleasing, he built there a castle and named it Maine, as it is still called." He passed Passavas - and then went to the far south of Mani before deciding to build a castle? I think not. Kelefa not only controls a vital east west pass, held at the other end by Passavas but also completes the encirclement of the Melingi.

But…Kelefa has a number of problems when matched against Prof. Wagstaff's criteria.

It is certainly west of Passavas and in Mani. Indeed the position can be further identified by the Chronicle of Morea's assertion that the castle of Beaufort was half way between Grande Magne and Kalamata. This fits somewhere in the Itilo, Limeni area, not a far distant location as the Cavo Grosso or Porto Kaiyo.

Kelefa today

The awesome crag bit depends on where you stand and look at it. From Itilo it looks on the same level until one realises that the chasm of the Milolangada gorge separates it from you - as it would the Melingi from the castle of Grande Magne. From the east it is rather unimpressive but stand on the seashore at Neo Itilo and look up and there is your awesome crag. If, as is likely, any Frankish castle had a keep on top of it then it would be truly impressive and the view from the castle dominates the surrounding land and sea. Look at it from Limeni and the large mass could, at a pinch, be a promontory and Coronnelli's etching of Kelefa (taking into account that this is a transcription of a sketch) certainly makes it look like a promontory. Interestingly Evliya Celebi's 1670 account of the village of Kelefa says, "Between Kelefa and Oitylo tumbles the stream of Karya with its cold water in a precipitous ravine. We continued downwards for an hour. A precipitous crag is found there".

It is, however, in strict geographical and topographical terms, not on a promontory. In fact only Tigani really fulfills this criteria, Kastro tis Orias at a stretch. Equally Kelefa fails on the evidence on the ground. There are no 13th century remains, or rather there has never been any specifically identified. But this is a difficult task. All the experts have tried to identify the difference between Byzantine, Frankish and Venetian brickwork and none of them, Traquair, Bon or Andrews has come to any convincing conclusion or taxonomy. Also, as Bon points out, there is no mention of Grande Magne = Kelefa in contemporary records. Or is there?

We get into a confusion here as we have evidence that there was an earlier castle of Maina. It is mentioned by the Byzantine Emperor Constantine Porphyrogenitus in the 10th century and by Benedict of Peterborough in the 12th century, who reported "super Gulfum de Witum illum est castellum bonum et forte quod dictur Maine", or "Above the Gulf of Vitylo is situated a good and strong castle called Maine". Witum as Vitylo isn't as clear cut as it may seem as it is possible that the Gulf of Witum/Vitylo may refer to the whole of the Messenian Gulf. The castle of Maine is generally accepted to be a separate from Grande Magne and some think that the ruins at Tigani was Maine. But this doesn't necessarily follow.

To confuse things further the Chronicle of the Morea was written in both French and Greek. There has been a considerable amount of academic controversy over which version was the first, Greek seeming to be in the lead. In the Greek version Grande Magne is known variously as 'Megali' or 'Palaia' Maina, in other words either 'great' or 'old'. This could point to the two Maina castles being one and the same and, pure speculation, if Villehardouin saw a useful site, which once was called Maine but had fallen into disrepair, rebuilt it more resplendent and bigger than before wouldn't he have the right to call it Grand?

These descriptions are both from before the foundation of Grand Magne by Villehardouin. This doesn't necessarily have to distract us. Many castles have been re-animated by later military forces. Passavas is a good example and so to is Zarnata and if one set of people assume a location is good for defence then it generally follows that, saving a complete revolution in weaponry, succeeding generations and ages will pick the self same spot. It doesn't state in the Chronicle of the Morea that there were no remains or existing castle at Grande Magne. Then again it doesn't state that there were. An inconclusive argument.

Kelefa has no identified 13th century remains. It has already been commented that experts cannot agree on a strict taxonomy of building styles and the site is overgrown. That stated, it is accepted that the present incredibly crude stone walls are Turkish, probably from the 1670s. No-one has ever excavated at Kelefa and, knowing the fixed propensity of Archaeological schools in Greece to use their permits for ancient sites to the practical exclusion of medieval or later period remains, it is unlikely to be dug over in the near future. (To be fair to these organisations the British School at Athens, for example, receives just three permits per year).

Coronelli's depiction of Kelefa. c. 1690. In this the 'awesome' crag of Kelefa and its dominating position overlooking the bay are clearly shown

The Turkish fortifications, little altered by the Venetians, are those of a frontier fort, not one designed or intended to withstand a contemporary siege train but rather to keep the locals in order. It has very much the same shape as Passavas which is it is accepted is, in its present state, the remains of a Turkish edifice, the Frankish features having disappeared over the centuries and any keep (a feature which was anyway declining in use in the 13th century in western Europe) would have been destroyed as far too obvious a target for artillery. Indeed one commentator (Kougeas, quoted in Kreisis) has pointed to a dismantling of of many of the Frankish castles, including Beaufort, Magne, Koutiphari and Passavas by the Byzantine ruler, Manuel II Palaeologus (reigned 1391-1425), to prevent the local population from seizing them for their own ends.

There is mention of the documentation concerning the Kladas revolt of 1480 which refers to Korkondilos Kladas forcibly retaking his Mani properties. Mehmet II wrote that Kladas "...has gone out and aroused the men of Zigo and have made him capo. They have killed my timar-holders who were in the area and they have killed the subasi of Megalo Mani"...A possible reference to the Castle of Grand Magne but in the Italian spelling. In the intervening centuries the identification of the bluff of Kelefa may well have been disassociated from the appellation Grand Magne.

There are records in Venetian documents and portolans (old navigatory maps) to places called the Castle of Maina or simply Maina (notably lacking any Grand or Megalo conotations) after the medieval period but none of these identify the location or precisely equate it with Kelefa. There are a number of maps from that period but all are partly whimsical, were usually concocted long after and far away from the events they portray (often sieges of Ottoman castles by Venetians) and are bafflingly inaccurate.

A 16th century map of an attack by Venetian galleys and forces on the 'Fortezza di Maina'. This took place in June 1570 when galleys under the command of Quirini attacked and took the castle at Porto Kayio which the Turks had just contructed. The Venerians blew up the walls but within a few years the Ottomans had re-established the fort.

However the coastline was observed in the 17th century by various travellers. Those early scholars of ancient Greece George Wheler and Jacob Spon sailed round the coast of Mani in 1675. This was just after the massive Turkish assault on Mani in 1670 and the two were informed that the Turks had built two forts on the coast - from which one could assume they built them from new. But an earlier visitor to the area was the Turkish chronicler and traveller Evliya Celebi. Celebi (1611-1684?) was a sort of civil servant cum envoy and indefatigable traveller. He travelled most of the Ottoman lands between 1630 and his death around the year 1682 and his experiences are recorded in his massive Seyahatname (Traveller's Tales).

His account is that after the Turkish onslaught on Mani in the summer of 1670 the Grand Vizier recommended that the Bay of Kelefa be fortified and that the fort "already there be built anew". Celebi is not always a reliable source but in this case we know that he was not just an observer. As a trusted official Evliya Celebi was sent, in 1671, by the local commander, Ali Pasha, to Albania to recruit workmen to re-build the castles of Zarnata and Kelefa. Celebi is not one hundred percent reliable but his embroideries on the truth, when they are recognised by experts, are usually whoppers - he claimed, for example, to have met the Hapsburg Emperor Leopold I and visited many north-western lands under his patronage - or rather bits of arcane and unlikely folklore which as a storyteller par excellence he couldn't resist retelling. This casual mention and his known involvement with the rebuilding programme of Zarnata, which certainly was re-built rather than built new, and Kelefa in Mani points to the likelihood of his account being correct. If our translations are correct (and here it must be admitted there is yet no definitively accurate manuscript or folio of The Seyahatname to work from), then there was a castle at Kelefa before the Turks re-animated the site.

From all the evidence at our disposal then, and an informed rationale of common sense it would appear that the prime candidate for the site of Grand Magne is Kelefa. This cannot be an absolute assertion as historical records have yet to come up with either conclusive archaeological or documentary proof.


This page was based primarily on the three papers, by Kriesis, Wagstaff and Burridge listed below which many may find difficult to locate. In May 1998 having just read the aforementioned papers I visited many of the sites and came to broadly the same conclusion as them. I am also indebted to Malcolm Wagstaff for letting me have sight of his translation of Evliya Celebi's account of Mani.

Peter Burridge

The castle of Vardounia and defence in the Southern Taygetos

in The Archaeology of Medieval Greece. Peter Lock and G.D.R. Saunders (eds.). Oxbow Monograph 59. Oxford. 1996. (pp. 19 - 28). ISBN 1900188031

A. Kriesis

On the castles of Zarnata and Kelefa

Byzantinische Zeitschrift. 56. (1963) pp. 308 - 316.

J.M. Wagstaff

Further Observations on the Location of Grand Magne

Dumbarton Oaks Papers. 45. (1991) pp. 141 - 148.

P.S Katsafados

Ta Kastra tis Mainis

Athens. 1992 ISBN 960 220 265 4


For references to other sources, Bon's Le Morée Franque, Andrews' Castles of the Morea, Rogan's Mani, Greenhalgh and Eliopolous' Deep into Mani, Paradissis and Leake and many more see the main bibliography page. The quotes from The Chronicle of the Morea come from Laurier's translation.