Rometta


Rometta was certainly a powerful Byzantine castle.  The fortress is thought to have stood on the site of an earlier Greek work.  It fell to the Arabs under Ibn Ammar after a year long siege in 965, nearly 90 years after the fall of Syracuse in 878 and an unsuccessful attack in 884.  In 1038 the Byzantines returned under George Maniakes and, after taking Messina, they moved onto Rometta, which soon fell to their assault after some heavy fighting.  All was lost in 1040 when Maniakes was disgraced in a political coup and the Arabs retook all his gains.  The castle is next mentioned when Count Roger Hauteville (d.1101), after an unsuccessful recce to the district in 1060, occupied Milazzo and Rometta with an army of 160 knights and 700 infantry in February 1061 on his first invasion of Sicily.  After Roger's defeat at Messina, both fortresses were abandoned.  When in May 1061 Roger returned with his brother, Robert Guisard (d.1085), they advanced on Rometta with their ally, Ibn al-Hawas, and the castle was tamely surrendered to them by Ibn at-Timnah's castellan who swore loyalty to Robert on the Koran.  Robert then handed the keys to the city of his brother, Count Roger.  From here they marched on Enna.

In 1081 Rometta was recorded as a part of the diocese of Troina.  The Arab geographer Edrisi, in his 1154 work, The Book of Roger, records the town as a fortress (qal'a).  In April 1168 the Messina rebels ‘occupied Rometta, a strong fortress, after easily overcoming the castellan's loyalty with promises'.  That same year the rebellion swept westwards into Palermo bringing Chancellor Stephen Perche's regime to an end.

During the Swabian period Rometta castle, listed among the castra exempta of Frederick II (d.1250), was state-owned.  The castle remained in the hands of the Crown, although it was garrisoned by Peter Ruffo of Messina (d.1256+) after the death of Frederick II in 1250.  After his defeat at Piazza Armerina in November 1254 it was surrendered by him to the Messinans.  On 3 May 1272 its Angevin garrison was supposed to consist of merely a single knight.  In 1294 King James II (d.1321) granted the castle to Bernard Ferro on condition that he repaired it.  Shortly before 1337 Frederick III (d.1337) made the castle his home.  The castle remained in use and in 1543 the military engineer Antonio Ferramolino proposed strengthening the defences by making the city walls strong enough for artillery and demolishing the houses both on the defences and within 40' of them.  This does not appear to have been done although the castle remained an important military stronghold in the sixteenth century.  During the Spanish reconquest of the island in 1718-1719, the castle was used as a base and as late as 1757 Abbot Vito Amico called Rometta ‘a very expansive fortress'.   It may still have housed artillery this late.

Description
Rometta lies in the mountains some 3 miles south of the coast and occupies a flat crag dominating the surrounding lower lands.  The entire hilltop would seem to have been the Byzantine castle, which would have made it more a defensible city than a castle as thought of today.  That said, the castle which is the heart of the defences at the top of the hill would also appear to be Byzantine in origin and fits nicely into the battleship style Byzantine fortress as described under Aci castle.

The city defences on the hilltop somewhat resemble a 4 legged star fish.  The main gates are to the northeast and southeast, while steep scarps and crags defend the rest of the hill which was apparently also walled.  To the southeast is the Porta Milazzo or town gate.  This is still the main and difficult entrance to the town.  Near this is the square Byzantine church of Maria e Gesu o Badavecchia with its octagonal central tower and Romanesque windows.  The gate itself has been much rebuilt and enlarged to allow motor access.  The walls on either side of it have been much built into, but still show the strength of the site.

At the northeast end of the town site is a ridge on the end of which is the Porta Messina.  This allows access to a ramp than runs down the crag to the southwest before doglegging to the northeast.  The wall is still battlemented and the gate is offset in a polygonal projection.  The outer arch to the north is pointed, while the inner arch is Romanesque.  The surrounding walls are rubble built and contain much Roman brick, some of it laid in levelling layers.  The curtain running back to the southwest follows the cliff top and is still battlemented.  It also contains a battery of small ground floor loops which appear to have had a wooden walkway on top, making up the battlement's wallwalk.

Roughly centrally in the defended plateau is the elongated ‘battleship' site of the castle proper.  This is now misleadingly called the castle of Frederick II, but there can be little doubt that the castle long predates Frederick's reign (1197-1250).  The site is some 600' long and no more than 60' wide.  As such it appears a typical Byzantine plan, designed to keep manpower usage to a minimum.  The castle seems to have consisted of 3 parts.  To the northwest was a tower block, while to the southeast were 2 (or more) conjoined towers now known as the palace.  Between them lay the long body of the castle, the centre of which is now occupied by a waterworks.  There is also a central entrance to the southwest and possibly also opposite this to the northeast.  The curtain wall at this point, where it survives best to the southwest, is almost 6' wide and standing some 10' tall.  The masonry here lacks Roman tiles in its rubble makeup.  It is therefore possibly Norman and certainly later than the 2 complexes at either end of the site.

The palace, sometimes referred to as the keep, seems to be the main residential block of the fortress.  Its larger, or northeast tower, is about 50' by 40' and divided into 2 equal halves by a crosswall.  The smaller southwest tower is about 40' by 30' and both have walls about 5' thick.  With both structures the building technique is the same, rough slabs of limestone encapsulated by well laid layers of Roman tile.  The corners are further strengthened by well laid ashlar around the quoins.  It would seem possible that the castle walls linking this tower to the north were added later, at least at their higher levels, for the curtain partially overlies a first floor Romanesque window to the northwest.  Entrance to the tower was also gained on this side via a central first floor doorway of typical architrave Byzantine style.  Similar style doorways still survive at Aci, Adrano, Belvedere of Fiumendinisi, Caccamo, Calatabiano and Montalbano Elicona in Sicily.  Such doorways can be seen as far north as Llangeais castle along the Loire in France and as far west as Carcassonne castle near Spain.

On the northeast side of the northwest tower at Rometta, the plastering is similar to that found in the tower of the Messina gate, leaving the lines of Roman tiles protruding through the plaster.  The southwest tower has a fine 40 degree sloping plinth at its base and is set both on bedrock and on a wide plateau at the end of the site, rather than being right on the edge of the drop down to town level.  It also only has an entrance to the other tower to the north at first floor level and no other apertures.  Possibly there was a further storey above.

The southwest wall of the northwest tower is mostly down from the junction with the SE tower, but foundations remain.  Internally the northwest tower was once rib vaulted, the remains of the ribs still being partially traceable on the north and south walls.  The basement has mostly been filled in with rubble, but the northeast corner is open nearly to its full depth.  The remaining top section of the north wall may suggest that the tower had a flat wooden roof at this level.  Certainly there is no trace of mural stairs or a further floor.

At the northwest extremity of the castle site are the remains of another tower, about 25' by 20', which is built in exactly the same masonry fashion as those to the southeast.  The tower has a vaulted basement which used to be entered via an inserted doorway to the southwest which has since been partially blocked.  Probably originally access to the basement was only from above.  The only original access to the tower would appear to have been from the southeast where later steps lead up to the first floor entrance.  This was a ‘Byzantine' architrave door with relieving arch.  The chamber entered had Romanesque windows to the north and northeast, while a blocked doorway led to a destroyed room to the southwest.  Traces of the walls for this can be seen at the southern external corners of the tower.  Close examination of the southeast wall shows quite clearly that another chamber bounded the tower at this point and gave access to the door, rather than any stairway.  Indeed the base of the ribbed arch for the vault still survives with the two succeeding courses of fine ashlar as well as the ghost of their compatriots cut into the wall above.

The idea that these plainly pre-Norman features are Swabian is clearly bizarre.  Worse is the fact that one of the ribs which held up the southeast tower's vault has collapsed in the last hundred years or so, although luckily it was photographed before its collapse.




Why not join me at other Sicilian castles?  Information on this and other tours can be found at Scholarly Sojourns.
 

Copyright©2019 Paul Martin Remfry