Spolia Column Shafts in Fortifications


Turning from the antique exemplars to their impact we find a peculiarly Eastern manifestation of the desire to beautify as well as strengthen fortresses: the widespread reuse of column shafts in fortification walls. Some commentators believe that the technique, as well as especially favoured by the Muslims, was actually invented by them[1]; although it is most likely that the techniques were learned from the Byzantines - just as it seems the Muslims learned the use of lead in joints from the same source[2].


Very long lists could be constructed of the use of columns as structural members and as decoration, and their enthusiastic use - conspicuous display in the extreme - constituted a heavy drain on finite supplies. We should bear in mind that they are used in churches and mosques as well as in fortresses, so what may have begun as structural assuredly becomes decorative. Thus a classic example of column shafts used in a Byzantine wall is by the "Queen's Garden" at the very top of Pergamum, where stood the Temple of Faustina.[3] Even given the convenient nearby temple, the manner in which the material has been laid is regular and aesthetically pleasing. Similarly, granite columns are used as ties in the Byzantine wall by the small church on At Meydani, Istanbul; a succession of marble columns is used as floor supports both in the Red Tower at Alanya, and in the outermost square tower of the peninsular fortifications (both sets protruding as bossed decoration); or in the east tower of the south gate of the lower wall at Ankara; or in the east tower of the Gate of Persecutions at Seljuk; columns forming a floor in the mediaeval fortress-cum-house (and still occupied as a house) as Syedra, east of Antalya; columns are used both as decoration and as tie-bars in the Byzantine wall near to the theatre at Side, in the Byzantine citadel at Bursa (and other spolia in the later Turkish walls[4]), and under exactly the same conditions at Theveste, in North Africa; columns are infill both under the arches of the theatre of Side (that is, on the line of the new late walls), in the towers south-east of the theatre, decoratively laid in courses about 3.5m from the ground (conceivably against ramming) and also, once again in regular courses, in the tower by the Baths (now museum).


When the Arabic writer Watwat (died March 1318 AD) writes that Gafsa est une ville batie sur des colonnes de marbre[5], he is presumably referring to justv such a use of column shafts which he saw in the city walls, or in any of the outlying Byzantine forts which surrounded the oasis. In what may be an adaptation of the use of wooden piles, they were generally employed not so much to hold walls together as, it is suggested, to protect foundations and footings against sapping. It is equally possible that they were used as levelling courses; and they also appear in patterns which are clearly decorative – a feature of fortresses such as Caesarea[6] and Aleppo[7], with the Crusaders certainly using the technique, although they were certainly not the first to do so[8]. Creswell[9] notes the use of such column shafts in the Bab an-Nasr in the Cairo fortifications (AD 1087), and accepts that they are intended to prevent sapping, by reference to Maqrizi's comments on Caesarea. But this is not very convincing. At Caesarea, the column shafts are down at today's ground level, which is well over a metre above that of Roman times; nevertheless, their practical use is easily seen. But surely the shafts in the Bab an-Nasr are too high up to serve in any way against sapping? Are they not just decorative, the more so because they are sparse, stand proud so they cast attractive shadows, and are not even all on exactly the same plane? Columns or marble baulks seem to have had limited success, however, against saps seems doubtful, if we accept the account of the 1097 assault on Nicaea, when a breach made with ignited logs was very quickly repaired[10] They do not seem to have been much protection against violent earthquakes[11].


Constantinople may well have provided the initial model for the use of column shafts, as it surely did for the beautification of the Constantinople-facing walls of Nicaea, in the adorning of such city walls with marble. Certainly, the practice is at least as old as repairs (date unclear) to the Sea Walls[12] and Land Walls of Constantinople[13], and possibly in repairs to the Golden Gate.[14] At Mamure Kalesi, rebuilt by Keykubad in the 13th century, there is no sign of column shafts to landward; but the seaward south-east corner tower does have a line of shafts in the foundations (but are these Byzantine, Armenian, or Seljuk?). By analogy, columns may have been used as underwater defences to stop ships approaching shore, as spolia were frequently used for the building of harbours.[15] In neither case is there any reason to believe the usage numinous since, for example at Seljuk and Nicaea, we find shafts and long marble baulks used as header ties, whilst at Nyssa square baulks were set into the wall, but perhaps not column shafts (none are currently visible).


Although it is frequently the case, then, that columns with structural importance are also decorative, in some structures, the column shafts must only be decorative: the Red Tower at Alanya, for example (built by Alaeddin Keykubad), has its shafts widely spaced, conceivably as internal floor supports, but certainly not against ramming. This is consistent with other spolia displayed rather noncholantly in its walls: inside, architectural fragments are set in the internal walls, including a Roman entablature block, and a tomb door. Why such concern? Because Keykubad had competition in wall-building from the fine Hellenistic walls visible as the lower line of fortification up the hill from the Red Tower – which Keykubad himself made good by bringing up to their original height.


Could the use of decorative column shafts be part of a more widespread use of roundel patterns? Can we infer from a reference in the 12th-century traveller Ibn Jubair, describing the Gate of Abraham at Mecca as decorated with des entrelacements dans le stuc qui ressemblent à des troncs de colonne s’enlacant cercle sur cercle[16] that column decoration was common? Yes, at Konya, as we shall see. And there are semi-parallels in the West, for example in the use of “wheel and tree” tile motifs in Tower C at Terracina, studied by Christie & Rushworth[17]; and a fashion for such ornaments is acknowledged, following Cozza, in 5thC Italy; but they are unwilling to accept his suggestion that such motifs (wheel=sun, lunette=rising sun) are Christian symbols of faith, because they are also found on Honorius’ (AD 401-403) retructuring of the Aurelian walls of Rome, and also exist on pre-Christian circuit walls. At Thessaloniki, for example, the western walls have a lot of brick decorations, like blind relieving arches, but close together and clearly decorative. Of the late walls at AcroCorinth, it has been suggested[18] that where in several places the shaft of a Byzantine colonette is built into the wall as a centre to a rosette of thin stones this could be either original decoration, or later plugs for damage. Foss has suggested that cloisonné work, and other careful use of spolia in Byzantine fortifications, may be a sign that the structures were Imperial ones[19]. Thus for a tower in the acropolis at Smyrna, with towers attributed to the Lascarid reconstruction, he notes coursed spoils and some cloisonné, together with a tree-shaped brick design between two sockets which he believes once held spolia. As he points out, Smyrna, Tripolis and Magnesia all have decorative patterns of brick and stone, which is rare; hence the suggestion that since these three castles were imperial foundations, they might be expected to show a more careful and elaborate style of construction than forts erected for the defence of places of only local importance. The identical argument could be extended to Amastris in Paphlagonia, which incorporates Hellenistic masonry, the careful use of decorative spolia, and parallels with the Golden Gate at Constantinople: no grander Byzantine gateway survives in Anatolia, write Crow and Hill, concluding that it was an Imperial naval fortress.[20]


Again, whether the use of columns is decorative rather than structural (or decorative and structural) can sometimes be difficult to determine. To take first a non-military example: at Xanthos, column shafts are used vertically as infill in a Byzantine basilica, in the north transept (or narthex), but presumably plastered over (as ditto the vertical shafts set in the exterior of the apse). In the same basilica, an indication that the builders’ aesthetics differ from our own is the mismatch between the splendid spolia entablature blocks used as door-jambs and entablature from narthex to church, and the careless placement of marble blocks for the paving of the nave. At Aphrodisias, the Byzantine wall at the theatre also uses column shafts to form triangular weep-holes for water drainage, as well as shafts as tie-bars. In this last case, however, the layout is regular, there are column drums set about 5.5m from the ground, and the decorative effect deliberate, since bucrania-and-garland friezes are set in the same wall. At Apollonia (Mysia: near Bursa: on the lake of Uluabat Golu, with the village of Golyazi on top of it), bucrania friezes with a monumental Greek inscription, and column shafts, decorate the walls. At Tlos, the Byzantine wall contains not only bases, capitals, and even sarcophagi, but sections of column shaft (some with capitals attached!), all laid with extreme neatness and regularity. The columns were presumably rolled down the hill to their present location, or conceivably hoisted the 4m from the stadium area below. Work of equal quality is also to be seen in the late Roman wall at Aegina[21].


If columns are generally used horizontally either as tie-bars or as bossed decoration, more exuberant decoration is possible, as in the Byzantine (? or later) fortress on the peninsula of Iasos. Both genres are used but, in one tower, a vertical column shaft is flanked by two lower and adjacent column-ends in an obvious and unmistakable gigantic phallic symbol.




[1] S. LLOYD & D. STORM RICE, Alanya, (Occ. Papers Brit. Inst. Archaeol. Ankara, 4) London 1958 mention (pp. 14-15) the lines of classical column-drums in (the early 13thC) Kizil Kule and footnote;

[2] K.A.C. CRESWELL, Early Muslim architecture, 2 vols, Oxford 1932 & 1940. I, p. 31 for use at Medina; I, p.36 for Kufa; I, pp. 58-9 for the Dome of the Rock; and I, pp. 79-80 for Byzantine usage and methods;

[3] A. CONZE, Pergamum: Stadt und Landschaft (Altertuemer von Pergamon Band 1 Text 2), Berlin 1913, p. 306 & pl. 62.2;

[4] A. GABRIEL, Une Capitale Turque: Brousse/Bursa, Paris 1958, pp. 23ff & plate X.5; and cf. p.28 note 4 for spolia in Turkish gates and walls (now destroyed) such as Saltanat Kapisi, plate X.4;

[5] FAGNAN, Extraits, cit., p. 52; cf. P. Bodereau, La Capsa antique, la Gafsa moderne, Paris 1907, pp. 195ff for the military history of the city and its five forts;

[6] L. MARINO, Chastel abatuz est demi refez: nota sulla fabbrica dei castelli d’epoca crociata in Terra Santa, in Castellum, 27/28 (1987), pp. 17-33. Cf. p. 25, note 20: Maqrisi recounts that Baybars could not take Caesarea because Louis IX had fortified it so well, usando per le murature colonne di granito, da rendere praticamente inutile ogni lavoro di mina;

[7] A. DJEMAL PASCHA, Alte Denkmaeler aus Syrien, Palaestina und Westarabien, Berlin 1918, plate 39 & p. 87: the shafts are balanced size for size, and colour for colour;

[8] cf. A. RABAN, The Harbours of Caesarea Maritima. Results of the Caesarea Ancient Harbour Excavation Project 1980-85, 2 vols, Oxford, 1989, figs III.119, pp. 124-5, 129-30. The discussion of the Crusader Wall and Square Tower (pp. 181ff) makes it clear they believe the Crusaders did indeed use this technique. Also L. I. LEVINE & E. NETZER, Excavations at Caesarea Maritima 1975, 1976, 1979, Final report, Jerusalem 1986, pp. 178ff for The archaeological finds and their relationship to the history of the city;

[9] K.A.C.CRESWELL, Fortification in Islam before AD 1250, in Proceedings of the British Academy, XXXVIII (1952), pp. 89-125; cf. pp. 113-114;

[10] BREHIER, Histoire anonyme, cit., p. 39;

[11] M. H. CHEHAB, Tyr à l'époque des Croisades, II: Histoire sociale, économique et religieux, 2 vols, Paris 1979, II.2 for damage to the walls of Tyre in 1157 (as well as the citadel at Aleppo, and Acre and Tripoli) and again in 1170;

[12] E. MAMBOURY & Th. WIEGAND, Die Kaiserpaelaste von Konstantinopel, Berlin & Leipzig 1934. Column shafts in Plate XII, west Sea Palace; Plate XIV, Bucoleon Palace; Plate XV, later spolia wall at the Western Sea Palace, with entablatures as well. cf. Plate XX, for the use of column shafts to preserve foundations against the sea (with up to three courses of spolia shafts in places);

[13] VAN MILLINGEN, Byzantine Constantinople, cit., p. 78ff for the Gate of Rhegium, or the Porta Rhousiou (of the Red Faction). Van Millingen only gives one illustration, but notes that there are seven shafts employed to form the lintel, posts and sill of the gateway are covered with red wash, as though to mark the entrance with the colour of the Red Faction.

[14] SHAT Génie Article 14: Constantinople; Carton 1764-1911: Anonymous Environs de Constantinople, en Europe, recu le 15 Thermidor An II, fol 13: a l’angle près des sept tours toute cette partie de l’enceinte a été  autrefois réparee avec des colonnes et des pierres d’entablement d’anciens édifices grecs et romains que les Turcs n’ont pas respectés, on les voit sur les parements de la muraille;

[15] FAILLER, Georges Pachymères, Relations Historiques, cit., V.10, where the Emperor builds a port at Blanga, on the Propontis, d'énormes pierres, presumably from nearby Euluthera;

[16] M. GAUDEFROY-DEMOMBYNES, editor, Voyages d’Ibn Jobair, V, Paris 1951, p. 125;

[17] N. CHRISTIE & A. RUSHWORTH, Urban fortification and defensive strategy in fifth and sixth century Italy: the case of Terracina, in Journal of Roman Archaeology, I (1988), pp. 73-88, and fig. 7;

[18] R. CARPENTER & A. BON, The defenses of AcroCorinth and the lower town (Corinth: results of excavations conducted by the American School of Classical Studies in Athens, III part II), Cambridge MA 1936, p. 196: for making decoration of necessity, cf. the repairs to the "third line wall" at Acrocorinth, fig. 135;

[19] FOSS, Cities, fortresses, p. 317 and fig 30;

[20] J. CROW & S. HILL, The Byzantine fortifications of Amastris in Paphlagonia, in Anatolian Studies, XLV (1995), pp. 251-65;

[21] W. W. WURSTER, Alt-Aegina (H. WALTER, editor) Die spaetroemischer Akropolismauer, Mainz 1975, plates 1-3 & beilage 2;