The fortress of Pythion / Крепостта Питион

The fortress of Pythion is located at the village that bears the same name, 15km north of Didymoteichon. The village was identified with 'Pythion' or 'Empythion' of the late Byzantine sources already in the 1920s. In the 50s and 60s it was identified with Egri Kuleli Burgaz or II Bey Kalesi of the Ottoman itineraries and geographers, toponyms which were later identified with "Empythion".



The archaeoligist Ch. Bakirtrzis was the first to deal with it as the ταμιειov ("treasury, storage and private quarters") of John Kantakouzenos, while the architect M.Korres was the first to deal with it as a prominent example of military architecture in the 70s and 80s. There have been several publications, both scientific and popularized concerning the monument.

The fortress is divided in an outer and inner enceinte. The former used to have a tower at every free-standing corner. The latter occupies the eastern edge of the hill and had three towers, one of which functioned as the last resort for defense (donjon) and is the oldest part of the fortress. The smaller tower and the arched gateway (portal) were constructed later, simultaneously with the inner wall. The outer enceinte followed.


The larger (northern) tower is almost square in plan (14,75/80m X 14,65/70m) and preserves three floors. It is approximately 17m high from its crepidoma. The thickness of the walls ranges from 2,35 to 2,60 m.

The only entrance to the tower lies on the eastern side slightly off the ground. On the left side, as you enter it, there is a notch for the bar. On the other side there is a similar notch for the bar to plug. In the center of the room there is a square pier (side : 1,60/1,70m), from which spring four arches (free arches, henceforth). They end in the middle of the four walls, thus forming two arches on each wall (side arches, henceforth). Thus, every floor is covered with four domical vaults resting on four arches and four pendetives.

Over the free arches of the second floor and under the third floor, as well as over the arches of the third floor lengthy vaulted spaces are formed. These spaces were used as underfloor storage area and therefore the masonry overlaying the free arches was minimized.

A narrow staircase formed within the east wall and covered with a vault link the floors. On the south side of the ground floor there are two narrow windows, built in at a later stage. On the second floor there are two openings on the western and southern sides and three on the eastern side. Only the latter are wide and arched. The rest are narrow, slightly widening towards the interior, with a straight lintel, a discharging arch on the inside and a sill leaning slightly towards the inside. The narrow windows, rising 1,50m over the floor served lighting and ventilation purposes. They did not allow for supervision or arching and could not therefore have been arching slits. On the third floor there are two narrow windows on the West Side and a wide arched one on the East Side. From the narrow windows of the West Side only the southern one is on such height to allow for vigilance and arching. Three of the windows of the eastern wall are opened on the outer side of the staircase. Their main purpose is to provide light to it. Secondly, they light the chambers of the floor and allow for vigilance of the inner wall. There is only one window over the gate, which controls the approach to it, while a rectangular opening at its sill functions as a 'murder-hole'. There is a fireplace and a small, built-in cupboard on the southern wall of the third floor.

On the south side of the third floor there is an exit leading to the walk of the intermediary wall.


Around the summit of the donjon there is continuous machicolation still preserved to considerable height, which used to have a walk with battlements and 'murder holes'. There is sound evidence to confirm that further floors were originally designed. Furthermore, their construction must have started, but it is not known (and is no of vital importance) to which height they were built and if they were ever completed. There are also remains of smaller corbels at the bottom of big openings, indicating defensive balconies. They were closed on all three sides and had openings on the floor. They allowed for safe shooting against anyone approaching the tower and foremost against anyone approaching the crepidoma. Two such defensive balconies existed on the eastern side of the third floor and one on the second and the third floor of the northern side. It is obvious that both the machicolations and the defensive balconies were the most important defensive elements of the tower.

The visitor ought to imagine the tower originally detached. It should then be considered as a part of a fortification system. The East Side, whose defense attracted the utmost care, is yet most delicate in its structure. The wide gate, the staircase, the windows and the defensive balconies, although designed to increase the defense of the gate and the side, reduce its mass and weaken the wall through piercing it at several points. Its delicacy is evident by its state of preservation: the major damage points of the tower are concentrated on this side. It is also especially susceptible to attack, as there is plenty of space in front of it for the deployment of offensive forces, something impossible on the southern and northern side.

Returning to the interior, we observe some minor interventions that took place during the construction of the tower. On the ground floor the central pier was enlarged and the thickness of the pilasters in the middle of the walls as well as the free arches were reinforced with other arches- this time with flat arch-stones - under their intrados. The southern chambers were walled and plastered internally with hydraulic mortar in order to be used as a cistern. One clay-pipe lead through each of the narrow windows of the southern side of the ground floor. The windows were later walled. Despite the lack of conclusive evidence, we presume that the cistern was supplied with rainwater from the roof, which was carried through the joint clay pipe covered in mortar on the south wall of the tower. The western pipe was to be embedded in the transverse wall, but was slightly shifted in order to stand free. Only its end entered the tower at an angle and was embedded in the transverse wall. This remark aims at showing the succession of works in the castle. While the transverse wall was being built, the conversion of the two chambers of the ground floor into a cistern had already been completed or was still in progress. No overflowing has been located. Openings on the domes covering the cistern allowed for the drawing of water with buckets.

The transverse wall blocked the western, narrow window of the South Side of the second floor. Today, a small-scale destruction in the area of the gate allows the light to enter, but prohibits observation from within.

Judging from the facilities of each floor, we realize that the third floor should most likely have been the residential quarters of the castle owner. To form a complete picture of the living in the castle, we should mark that no sanitary or cooking facilities have been found in the existing edifice or its surrounding.


The small tower almost square in plan (side length c. 7.40 / 7.30) is preserved in 20m of height, measured on its southern side from its crepidoma and shows substantial diminution. The tower consists of four floors, all single chambered. On every floor four arches and four pendetives support a domical vault. The ground floor has no windows and was probably used as a dungeon. A hole on the floor of the second storey permitted the descent to the ground floor via a wooden ladder. There is no communication between the rest of the floors as there is separate access to each of them. There is a wide arched window on the south side of the third floor and on the West Side of the fourth, both looking onto the outside of the castle. A small and narrow rectangular opening on the North Side of the third floor allows for the observation of the area right in front of the portal. Finally, an opening on the South Side of the third floor remains unaccounted for, as it is not clear whether it was a window or the door to a balcony, the latter being more likely. We tend to believe that at the end of the forth floor there was machicolation on the side looking outside the castle and a fifth floor on top of it.

The small tower served exclusively military purposes. The space between the two towers is filled with a 2.65m wide wall, in which a portal leading to the inner enceinte is opened. The transverse wall was built at such an angle to the larger tower, so that the walkway would coincide with the exit of the third floor of the larger tower.

The portal was three-storey-high. A continuous barrel vault covered the ground floor (length 4.60m, width 3.30m.). Two arches on pilasters lower than the vault (the keys of the arches were at the springing of the vault) divided the portal into two rooms (1.86m and 1.48m. long respectively). 'Murder holes' might have existed over the rooms. Behind the exterior arch there was a door, which was locked with a bar installed while the castle was being built. The draw bar hole on the left as you enter is 3,64m deep. On the other side there is a shallow socket for the bar to plug. A second two-leafed door existed behind the interior arch.

There is a second floor over the vault (internal dimensions 3.58X2.95m). Four arches and four pendetives support a domical vault. The third floor is a closed defensive gorbelled balcony for the safeguarding of the gatehouse.

The wall of the inner enceinte is 2.20-2.40m thick. The south wall preserves 28.5m and its maximum surviving height is 9m. The northern curtain is preserved in 10.15m and maximum height 9.30m. On the internal side blind arches support the wall walk. Along the south curtain there were wooden-roofed auxiliary buildings.

As for the circulation of the complex, there is only one staircase on the inner side of the intermediary wall, providing access to the second floor of the portal. The first step of the staircase was much higher off the ground and it was to be reached only from the court level with a portable ladder. The second floor of the small tower was only accessible through the wall walk of the inner enceinte. A corridor in the transverse wall connected the second floor of the portal with the third of the tower. There is a window on the eastern side of corridor. The fourth floor was accessible only through the wall walk of the intermediary wall. In fact the latter provided communication between the third floor of the larger tower and the fourth floor of the smaller one. The presumed fifth floor was accessible only through a staircase built outside the eastern side of the fourth floor and once again through the wall walk over the portal. Finally, access to the defensive balcony of the portal was possible from the intermediary wall with a staircase.



It is obvious that each and every unit and section of the unit of this unique fortress was self-sufficient in terms of its defense. The fortress was accessible only from a point on the West. The steep slopes of the hill prohibited any offensive from another point. To intrude the inner enceinte the attackers had to violate the only gateway with its two gates. To conquer the larger tower they had to violate its gate under the heavy charge of the defenders. The latter could shoot easily and safely mainly from the larger tower and secondly from the smaller one and the intermediary wall. To control the other floors of the larger tower the attackers had to climb the narrow staircase, where there was room only for an armed soldier. To conquer the smaller tower the invaders had to fight their way on each floor breaking through narrow, steep and easily defendable accesses.

The fortress is built from bulky, big, medium-sized and small stones, bricks of three sizes and brick fragments. The connecting mortar composition includes chalk, big-grained sand and a small amount of pebbles. The faces of the walls were dressed with coarsely cut stone in horizontal bands. The dimensions of the stones gradually decrease towards the upper parts. The smaller tower is built with medium size stones throughout and it gives a fairly uniform impression, notwithstanding the use of some smaller or larger stones. In the construction of the larger tower stones of every kind have been used. The bulky, big and mid-sized stones of the lower part of the tower, and especially the crepidoma, the ground and the first floor seem to be the majority. The nucleus of the walls is filled with medium and small-sized stones and mortar. The excellent cohesion of the faces and the nucleus leads us to believe that they were built simultaneously. As the work had to proceed at the same pace on both faces, we are quite sure that provided there was ample labour and building material, more than two-three bands could be built a day. The walls have wooden grids at various levels.

The corbels of the defense balconies were built with stones and bricks, while those of the machicolation around the larger tower are built with bulky cantilevered stones. The floors are indicated in both towers with bands of three brick courses, which do not traverse the wall. The three bands of the larger tower exhibit the 'recessed brick' technique. The same technique is employed at the face of the arches of the openings of the larger tower. The brick bands of the smaller tower do not exhibit the same technique. The vaulted parts are built with bricks. Arches, vaults, pendetives and domes were all exclusively built with bricks. Centering was employed in the construction of arches and barrel vaults, whereas the pendetives and the domes were built without it. The 'brick filled mortar joined' technique was used in the vaulted parts of the larger tower. It is also to be found in the vaulted parts of the portal and only scarcely and randomly in the smaller tower.

With the exception of the arches of the second and the third floor, which are slightly pointed, the rest of the arches in the fortress are circular.

All evidence leads to the conclusion that the fortress of Pythion is an exceptional example, very much so the best example of Byzantine military architecture in Greece and among the best of the 'white period' in the Balkans (with exception of the two main centres of the empire). Compared briefly to other fortresses in the Balkans, it becomes rather obvious that very few can be considered superior: a) Yedi kule and Rumeli Hisar in Constantinople, both projects of Mohamed II were built for gunpowder wars, b) the Manasija and the Genoese Galata towers were taller then any other. If we take into account that the Manasija tower with its characteristics (ground plan surface of almost 214 sqm and 35m high, single-chambered interior, wooden floors, entrance 10.65m higher from the court level, its 104 machicolations and the isolation from the wall walk) represents 'the peak of Serbian medieval military architecture', then any comparison is easily understood.

We will not come across this 'four-part' arrangement of the larger tower almost anywhere else.

Neither will we find the brick-built roofing. The rest of the edifices mentioned were not aristocratic establishments. They were more and foremost military buildings. The prerequisites were therefore totally different. The Pythion castle was the only example that was both a military project and a symbol, which meant to provide shelter and security to a leader of a power in decline, who still though was fully aware of his leading part in History.

* * *
The features of the fortress that dominated and overshadowed anything else have always been the size and the climax of its volumes. The dimensions of the towers and the walls, especially the height - much bigger in the past - and their impenetrable mass add to it a sense power and magnificence. Anyone in the plain gazing the castle at the tip of hill would be in awe. That sense is masterly expressed in Nicephoros Gregoras' description of the castle: "Kantakouzenos undertook expensive works in the castle of Pythion .... and managed to make it look as if suspended in the air". The climax of the big shapes provided the castle with some harmony, the larger and the smaller towers and the connecting wall divided the fortress in two equal parts. The walls formed a ring, which connected the towers. The latter, much higher than the walls seemed to guard the large central tower. This, finally, mightier and far taller than the rest dominated the complex in the middle of its constituents.

When were the parts built? The masonry and the 'brick filled mortar joined' technique point to a dating in the Paleologan period. Dendrochronological examination of samples from the larger tower has shown that the wood used in the framework came from trees felled in 1331. The succession of the individual parts is distinct and has been pointed out often in the literature as well as in this paper. It is clear that the differentiation in specific building techniques (e.g. the employment or absence of the 'concealed brick' technique) do indicate different dates for the individual parts. The builder and owner of the castle let us know that it is his own work. We shall follow him in his adventures and strange paths of his life in order to define, as closely as possible the date of the construction of the fortress. The peak of his activity and involvement in the political and military developments of the decaying empire lies between c. 1320 and 1354, and that must be the timespan of the erection of the fortress. When the first civil war broke out in 1321, he was about 25 years old, not only wealthy but an acknowledged dignitary of the Empire in Thrace. When Andronicus III reigned (1328-1341) in Constantinople, the Megas Domestikos was active either by his side or at the edges of the Empire. In his lengthy memoirs, though, the Historiai, when he narrates the beginnings of the second civil war (c.1341), he informs us that the castle a) αυτώ, περιόντος ΕΤΙ βασιλέως, ωκοδόμητο αυταίς κρηπίσιν and β) πολλοίς ΠΡΟΤΕΡΟΝ χρόνοις ΕΞ aυτών κρηπίδων εδείματο. These quotation indicate that the castle was erected when Andronicus III was alive and reigned either with his grandfather (1321 splitting of the Empire, 1328 he became the only ruler) or own his own (1328 - 1341) and very likely many years before 1341. Nikephoros Gregoras provides similar, yet slightly different information: a) ΕΚ πολλού Kαντακουζηνός ... ο βασιλεύς πολλαίς χρημάτων δαπάναις, ωχύρωσε and b) ΤΟ ΤΟΥ Πυθίου φρούριον... γεγονός ερείπιον. The first quatation agrees with the information from Kantakouzenos. The second, though, without contradicting Kantakouzenos on the erection of the castle from scratch, adds that there was a ruined castle on the hill. These sources combined with the data of particular construction methods as well as with the dendrochronological evidence lead us to the conclusion that the castle is a work of the 1330s. What remains unaccounted for -at least to us- is Gregoras information of an existing castle, already in ruins at the time of Kantakouzenos. We cannot rule out the possibility that some kind of ruins did exist at that spot and the engineer of Kantakouzenos cleared the site before proceeding to the erection of the fortress.

The sources do not enlighten us on the destiny of the fortress. It was conquered by the Turks, either in 1359 or 1360, before the fall of the two main centers by the river Evros, namely Didymoteichon and Adrianople. Neighbouring Didymoteichon was in Turkish control in 1361, while Adrianople fell to the Turks around 1369. The stones of the gateway and the surrounding wall have been ruptured by fire, which might very well have been caused by the burning of the door or even a bigger one. If that fire had taken place at the time of its conquest, it is obvious that the castle was defended before it fell to the Turks. The region between Didymoteichon and Adrianople became under the full control of Sultanate in the 1370s. The completion of the Ottoman conquest put an end to the role of the castle in the defense system in the region. Excavation data reveal that whatever its use after the conquest, it was not long-lasting. After its abandonment, it was looted and quarried. What suffered most were its brick-built parts. These developments led to its drastic dilapidation. In the 20th century it was used briefly as a shelter. The cistern was used for dung storage while the rest of the ground floor as a stable and the first floor as a barn.

After the consolidation of the Ottoman control in the region the floors with machicolations and the defense balconies were demolished. Gradual degradation and damage due to weather and man-inflicted reasons followed suit. The dilapidation process contributed equally as much. The residents of the village dislocated a great number of bricks from the domes of the interior. The mortar was being crumbled by the frost and the birds nesting in the hollows. The gravest damage is located at the portal and the NE corner of the larger tower.

A series of studies have been carried out for the conservation and the promotion of the fortress. Conservation works have been under way since 1993. The existing parts of the outer wall were the first to be conserved. The exterior of the small tower and the portal followed. The work on the exterior of the larger tower started in 2000. The reconstruction of the brickwork in the interior is the next to be undertaken. Once the work is completed an exhibition in the larger tower is planned. It will include: l.the findings of the excavations in the inner court. 2. plans, drawings and photographs concerning the fortress and the conservation works. 3. Plans, drawings and photographs concerning the defense of the Evros valley and late military architecture




Материалите за текста са взаимствани от тук :http://alex.eled.duth.gr/archaic/puthio/index.html

Снимки: Ангел Йорданов