Moorish Bridges of Andalucia Joseph Dahmen  ,   Contents  Acknowledgements  Introduction  Overview of Bridges  I. Bridges of Cordoba Puente Romano Puente Nogales Puente over the arroyo Cantarranas  II. Road from Cordoba to Sevilla Puente de la Tejera sobre el Guadiato Puente Quebrado sobre el rio Bembezar Puente sobre el rio Guadalbacar  III. Bridges of Granada Puente del Aljibillo Arco del Darro Puente sobre el Rio Genil Puente de Pinos Puente Puente Arabe  General Characteristics of Moorish Bridges  Conclusion  Appendices  Bibliography Acknowledgements This paper presents research on bridges built in southern Spain during the eighth through the twelfth centuries AD. I travelled to eleven existing bridges in Andalucia in January  to collect information about their geometry and structural characteristics. The work was a collaboration with Amber Frid- Jimenez, who took the majority of the photographs. She also set the type and performed the graphic design required by the paper. The research was advised by Professor John Ochsendorf and funded by a travel grant from the Aga Khan Program for Islamic Architecture at the Massachusetts Institute of Technology. It is intended to provide the groundwork for subsequent scholarly inquiry into the influence of Moorish builders on bridge design within the Islamic world and Western Europe.  Introduction The advanced state of Islamic architecture, mathematics, and astronomy dur- ing the Umayyad and Abbasid periods of rule in the middle east is well-docu- mented. However, the contributions of Islamic builders to bridge design have rarely been studied. As O’Conner states in the conclusion of his exhaustive study Roman Bridges, “…it appears that no accurate review has ever been pub- lished of bridge construction in the post-Roman period… such a review should pay closer regard to regions with civilizations which continued in the Post-Ro- man period, such as Turkey or Moorish Spain.” (O’Connor 188). Moorish domination on the Iberian peninsula lasted from their arrival in 711AD until their dispersal at the hands of the Spanish monarchs Ferdinand and Isabel in 1492. The centers of power shifted during this time between Cordoba, Sevilla, and Granada, which became the three principal cities of Andalucia, whose name is inherited from the Arabic al-Andalus (Collins 163-224). Along with mosques and palaces erected during the almost 800 years of Muslim rule, the Moors also built or restored a network of roads, which required bridges across the rivers lying in their paths. In some cases, such as probably occurred in the bridge at Cordoba, this meant repairing existing Roman bridges that had dete- riorated since the breakdown of Roman administration on the Iberian penin- sula around 200AD (Richardson 231-235). In numerous other instances, such as the three existing Moorish bridges between Cordoba and Sevilla, the Muslim inhabitants of Spain constructed bridges where there had been none previously. The attempt to distinguish the Moorish lineage in the bridges of Andalucia is complicated by the fact that in many cases, the bridges were in service long after the departure of the Moors (of the eleven bridges investigated, five are still carrying traffic). Their continued use means that in addition to the work that might have predated the structure, one must also attempt to distinguish Moor- ish work from alterations occurring over the thousand-odd years since comple- tion of the structure.  During a literature review at the outset of the project, I was fortunate to find a work of Basilio Pavón Maldonado entitled Tratado de arquitectura hispano- musulmana, published in 1990. This source was suggested by a member of Archnet who responded to a query posted about Moorish Bridges. Maldonado has written extensively in Spanish on Moorish bridges, dams, and irrigation structures, and the first of two volumes, entitled Agua, functioned as a guide as we travelled around Andalucia looking for the bridges included in the study. A full treatment of the dimensions and geometry of each bridge is summarized in appendix 1 at the end of the document. The bridges investigated in the study consist of the following: Name Location Puente Aljibillo Granada Arco del Darro Granada Puente sobre el Rio Genil Granada Puente de Pinos Puente Pinos Puente Puente Romano Cordoba Puente Nogales Cordoba Puente Califal sobre el Cordoba arroyo Cantarranas Puente Califal sobre el Cordoba Guadiato Puente Quebrado Hornachuelos Puente Sobre el rio Setefilla Guadalbacar Puente Arabe Ronda These bridges of Andalucia, generally built between 711 and 1400 are of particular interest because of their cultural and geographical location. Andalucia was for many years the locus of a peaceful and fertile collaboration between Muslim and Christian scholars(Collins 200-207). As such, it was a prime interface in the transfer of knowledge between two empires. A bridge is a technological expression of a culture. I hope that this field research will lay the groundwork to begin to appreciate the transfer of advanced knowledge and method from the Muslim to the Christian world in the area of bridge design.  Overview of Bridges The bridges visited fall into three major geographical areas: those in and around Cordoba, those lying on the Moorish road that follows the valley of the Guadalquivir river to connect Cordoba to Sevilla, and the bridges of Granada. The descriptions below follow the approximate chronology of the center of Muslim political power in Spain. They begin with the bridges in and around Cordoba, the original seat of Muslim power on the Iberian Peninsula. After this they treat the bridges between Cordoba and Sevilla, before proceeding to the bridges of Granada, the location of the last Muslim emirate on the Iberian Peninsula.  I. Bridges of Cordoba The Moors crossed the strait of Gibraltar into the present city of Tarifa, in Spain, in 711 AD from what is now Morocco. From Tarifa they travelled over- land approximately 220 kilometers to the Roman city of Corduba, establishing a regional caliphate which functioned for several hundred years as a outlying colony of the Umayyad dynasty administered from Damascus. (Collins 151). When the power shifted from Umayyads in Damascus to the Abbasidds in Baghdad, the Cordoban Abd al-Rahman declared himself caliph in 929, ef- fectively establishing the independence of Cordoba. The Caliphal period lasted from 929 to 1031, during which lavish palaces and public buildings were constructed in Cordoba. (Collins 171). Puente Romano Puente Romano Early accounts by Moorish travellers arriving in Cordoba indicate that they were impressed by the ruins of a seventeen span Roman bridge crossing the Guadalquivir River (Maldonado 96). The original bridge, which lies on the Via Augusta, may have been built in the time of Augustus (O’Connor 104, who cites Gazzola). Maldonado interprets the chronicles of Julius Caesar to sug- gest that a Roman bridge of boats or heavy timber supported by baskets filled  with stones predated the masonry structure, but also theorizes that the bridge may have been built during the time of Octavius (Maldonado 94). During the Moorish domination of the city the bridge connected the Torre de la Calahor- ra, a fortress tower on the south eastern bank, to the mezquita, or mosque, on the northwest bank of the Guadalquivir. Whatever the lineage of the bridge across the Guadalquivir prior to the 8th century, Muslims arriving in Cordoba after 711 found a bridge partially destroyed, probably by the passage of time and floods. Arab chronicles written during the time of the caliph of Damascus Umar ben Adb Aziz state that the bridge was reconstructed by al-Samh, who apparently came to Andalucia to undertake the repair in 719-720, after receiving authorization from the caliph. Notably, al Samh offered to repair the masonry of the bridge using stones from the existing walls surrounding city, which were in turn to be replaced with bricks if suit- Puente Romano able stone was unavailable (Maldonado 95). Presumably the walls were from Roman times, like the bridge itself. Thus the portions of the bridge repaired by the Moorish builders might be expected to blend well with the remnants of the bridge surviving intact from Roman times, making it difficult to accurately assess which parts of the bridge were rebuilt by the Moors. At present the wide range of sizes and shapes of the arches, as well as subtle and obvious differences in design of breakwaters both upstream and downstream, attest to the centuries of repairs and modifications made to the bridge since its initial construction, both before and after the period of Muslim domination. In various locations in the piers and arches the original masonry of the bridge is interrupted by later repairs of what appear to be a soft sandstone, similar to that employed in other Moorish bridges between Cordoba and Sevilla. These repairs sometimes incorporate baked bricks of a dimension common to the bricks used in the Puente Romano mosque of Cordoba. The bridge is large, consisting of 17 spans, which are generally each 10 meters long. There is considerably more variation in the piers, which begin close to 10 meters wide on the side of the mosque and diminish to closer to 7 meters at the tower end of the bridge. This might be accounted for by the greater amount of current on the mosque side of the river; according to Maldonado floods have carried away sec- tions of the mosque side of the bridge as recently as the 16th century (103). A catalog of the spans, their shapes, with speculation on the builders responsible for the various sec- tions, arches, and breakwaters, can be found in Appendix III  at the end of this report. A complete study of this bridge is beyond the scope of this investigation but might begin to acknowledge the contribution made by Muslim builders to a bridge which is commonly considered exclusively Ro- man, in name and historical pedigree. The bridge currently carries pedestrian traffic across the Guadalquivir, but until recently it carried vehicular traffic, which from casual inspection it seems easily capable of doing. Puente Nogales The mosque of Cordoba, built on a site occupied previously by a visigothic church, functioned as the administrative center of Andalucia just after the arrival of the Moors on the Iberian peninsula in 711 until 1031. Eight kilome- ters northwest of the city lies Madinat al-Zahra, a palace complex constructed between 936 and 945 by Abd al-Rahman III for use as a summer palace (Col- lins 171). The Medinah, or village, is today the site of an extensive archaeologi- cal dig. Numerous sites of historical importance exist between Cordoba and Madinat al Zahra, including norias, acueducts and bridges. I will focus on two bridges lying on two separate roads which led from the mosque to the summer palace. Puente Nogales The first bridge, puente Nogales, consists of three spans which cross a stream which was barely a trickle during our winter visit. Although all three spans are intact, the bridge has experienced moderate deterioration over the course of its life. A small number of concrete voussoirs have been installed, presumably by archeologists, to prevent further deterioration of the structure. A fabric mat is installed between the stones of the spans and the soil on top of them, suggesting that the bridge has been the site of an archaeological investi- gation within the past five years. The bridge lies amidst a field currently under cultivation. A great multitude of round stones measuring in the vicinity of twelve centimeters were  Puente Nogales observed scattered in the furrows of the field on either side of the bridge extending out in a straight line in both directions. These were likely to be of the part of the pav- ing of the original Moorish road from the mosque in Cordoba to the Madinat al-Zahra. The bridge consists of a span of 2.9 meters flanked on either side by arches of 2.2 meters, with piers approximately as wide as the two outside spans. The voussoirs of the center span vary between seventy to ninety centimeters high, making them very large for the relatively short distance of the spans. The bridge is 4.5 meters wide and has no visible break- waters up- or downstream. Stones of dimensions up to 120cm long by 50 cm high of unknown depth (but not likely to be less than 25 cm) were observed in the spandrel area between the arches. Although all three spans of Puente Nogales appear today to be semi- circular, there is ample reason to believe that at least the two flanking arches, and perhaps the central span, are horseshoe shaped. A ledge of a depth of Puente Nogales approximately 20 centimeters, consistent with the bottom return of the horseshoe shape was visible at the base of the left hand arch. A small amount of excavation at the base of the right-hand span revealed a similar ledge. While the variability of river flows make it difficult to reliably predict a standard rate of soil accumulation (Jimenez-Camino, 2005), it seems reasonable to suggest that soil was deposited about the bridge, making what were originally horse- shoe arches appear today to be semicircular. The bond of the stonework at 