Introduction
Acre is situated on the northern Mediterranean coast of Israel, about 13 km to the north of Haifa. Its history since the twentieth century bce has been well-documented. Until the third century bce it stood on a mound called Tel al-Fukhar.Footnote 1 From the Hellenistic period the city developed on the plain to the west of this mound. Acre experienced two important phases during its long existence. The first was during the Hellenistic era, when it occupied the mound, the plain, and the rocky peninsula to the south of the western sector of the plain. The second was during the Crusader period, when it occupied an even larger area. Between these two periods, the city declined, although it was still repeatedly referred to in historical sources.Footnote 2
The object of this article is to trace Acre's urban evolution during the twelfth century. It will be argued below that, during these years, Acre gradually expanded towards the south, namely towards the peninsula. This view, however, is opposed to opinions expressed in earlier studies on the topography of Acre that have focused on the development of the city during the thirteenth century.Footnote 3 According to such studies, the city's perimeter had remained unchanged throughout the previous century: its layout, they have suggested, did not alter between 1104, when the Franks took Acre, to 1187, when they lost it to Saladin. Such conclusions need to be proven, or at least supported, by evidence. However if they do not employ any scrutiny of twelfth-century Acre, they are based on mere assumptions.Footnote 4 Moreover, such views seem to imply that the layout of early-Muslim Acre is fairly well-known. Yet, no study has as yet been published on this topic. Therefore our knowledge remains somewhat vague or uncertain and, hence, such theories amount to a circular argument, along the following lines: since the city did not expand during the twelfth century, it is clear that early-Muslim Acre was as large as its successor. This state of affairs calls for a re-analysis of the entire urban history of this important medieval city.
The idea that twelfth-century Acre was identical with its Muslim predecessor is mainly based on the premise that, during both phases, the city was not walled on the west and the south. This understanding relies on the account of Willibrand of Oldenburg (1211) who described the city as a square fortified by walls in the north and east, whereas the sea protected it from the south and west.Footnote 5 This opinion is rationalised by arguing that sea walls are not an obstacle to be removed during the development of a city. Therefore, if there were no sea walls at the beginning of the thirteenth century, they did not exist before then.
As logical as this argument may seem to be, it should be rejected. As will be demonstrated below, during the eleventh and twelfth centuries the city did indeed possess western and southern walls. These walls apparently stood not on the seashore itself, but at a short distance from it. It will be suggested that they ceased to function during the twelfth century as a result of the city's expansion in these respective directions. Indeed, it is precisely the focus on the thirteenth century that has led scholars to neglect developments during the twelfth century. Documents from the earlier period have been analysed either out of their urban context, or even ignored altogether, despite the fact that some of them (as will be discussed later) clearly refute the view that for almost a century the city's perimeter did not develop.
During the Crusader period, Acre became the main port of the Frankish kingdom, while its Muslim predecessor was one of many ports in the Levant littoral, and certainly not the most important among them.Footnote 6 The fact that the city was rarely mentioned in early-Muslim sources implies that for centuries it was of rather marginal significance. This might be the reason why there is no evidence as to the nature of its commercial activity throughout the eleventh century. During the twelfth century, however, as the city developed into the principal harbour of the Latin kingdom and of the Italian cities in the east, the volume of trade should have grown significantly.Footnote 7 This should have also caused a boom in the city's economic life. Its population should have accordingly increased and the built-up areas expanded. Such an assumption seems to be so obvious that suggesting the opposite - namely that despite the economic growth Acre was stagnant - calls for substantial evidence, which does not exist.
I would like to suggest that, during the pre-Crusader period, the area of the peninsula was hardly occupied. The early-Muslim and Roman-Byzantine city lay to the north of the peninsula and to the west of the Tel. Remains from these periods constitute a small fraction of the total findings in the city area. In many archaeological excavations conducted in the peninsula over the last few decades, hardly any such remains were found.Footnote 8 While they do appear, this is in much smaller quantity than remains from the Crusader and Hellenistic periods elsewhere in the city.Footnote 9 Of course, the fact that archaeological findings were not uncovered in a certain district does not mean that it was not occupied during a specific period of time. However, it obliges those who claim that it was occupied during the period in question to sustain their argument using alternative evidence. In the case of Acre, this is not the state of affairs. There is a substantial body of data that clearly implies that the current reconstruction of Acre's urban history is based on mere speculation and, therefore, it has to be re-assessed.

Acre until the Crusader Conquest
Acre should have been an important city from the Hellenistic period until the Arab conquest during the first half of the seventh century CE, for obviously a marginal city would not have been referred to so extensively as Acre was during these periods. By the same logic, the city declined during the early-Muslim period, when it was mentioned only occasionally. Yet, the abundance of sources about Acre seems to be misleading. It is very difficult to reconcile the view that the city constituted a considerable size with the paucity of findings from both the Roman and the Byzantine periods.
There are, for instance, very few sources on the layout of the early-Muslim city. Al-Muqaddasī (d. ad 985) described in detail the renovation of Acre's port during the reign of Ibn Tulun at the end of the ninth century. When outlining the construction he mentioned a chain which blocked the entrance to the port. The only detail that he noted about the city itself was that its central mosque was encircled by an olive grove.Footnote 10 The fact that a geographer who provided extensive descriptions of the cities of al-Sham dedicated so few words to Acre should not be overlooked. It clearly highlights the city's marginality in the late tenth century.
According to al-Bakri (d. ad 1094), the entrance to the port was under an ancient arch (or bridge) that enabled ships to enter with deployed sails.Footnote 11 But while he referred to an arch, he made no mention of a chain. This construction was also mentioned by al-Muqaddasī, who also referred to a chain and did not point out that ships could enter the port with deployed sails.Footnote 12 The existence of a bridge or arch in the entrance of the port has not been hitherto evidenced by archaeological findings. Yet, the most well-known feature at the port entrance was the chain. The only author who ignored this was al-Bakri; something that should be attributed to the fact that he never left al-Andalus.
The only source which describes eleventh-century Acre in some detail is the book by the Persian traveller Nāsir-i Khusraw (ad 1047).Footnote 13 According to his description, the city was surrounded by very solid walls 2,000 ells in length and 500 in width (an ell = ca. 64 cm), of which the western and southern walls were washed by the sea. Its central mosque stood in the city's centre, and the port was in the south.Footnote 14 The simple understanding of this text suggests that, like other cities in the region, Acre possessed sea walls during the eleventh century:Footnote 15 after all, its western and southern walls are clearly noted. Therefore, arguing the opposite, based only on an implicit deduction from a later text, would seem untenable.
Additional data concerning the city on the eve of the Frankish conquest can be obtained from Ibn al-‘Arabī's description as well as from Crusader-period documents. Ibn al-ʿArabī visited Acre only a few years prior to the Frankish conquest.Footnote 16 In a short entry in his book, he referred to a defence facility in Acre called the Makhras.Footnote 17 This complex was apparently located near the city's southern shore, as the author mentioned that he saw the sea while praying there, which might suggest that the sea was on the south side of the building.Footnote 18 This compound might have been associated with a mosque which later served as a landmark in the Venetian quarter.Footnote 19 But it should not be confused with another complex later known in Latin documentation as Rabattum, probably a Frankish version of the Arabic word rabat which means, inter alia, a defence facility or a pilgrims’ lodge.Footnote 20 While Ibn al ‘Arabī apparently described the Makhras as being near the southern seashore of Acre, the later document refers to a complex in Montmusard, the northern suburb of thirteenth-century Acre. If the name ‘Montmusard’ is indeed a Frankish version of the Arab word mazar (which usually means a saint's tomb), it might be reasonable to assume that the rabat and the mazar were related or even identical.Footnote 21 In other words, the rabat/mazar was the kernel around which the northern suburb of thirteenth-century Acre developed. In any event, if the rabat and the makhras were two different complexes, I would like to suggest that both sets of buildings were already located in Acre in the eleventh century.Footnote 22 The rabat probably stood in an area outside the northern walls of Acre while the makhras was near its southern seashore. The poor topographic data included in Ibn al ‘Arabī's narrative does not enable us to determine whether it stood inside or outside the city. Acre's central mosque, however, did stand in the middle of the city. It must have been an impressive structure because it has been repeatedly described by many geographers and travellers.Footnote 23 These references indicate that during the second half of the eleventh century there were in Acre at least three mosques, two of which were within complexes that combined defensive roles with Muslim pious activities.
It is rather frustrating that the many documents of the Cairo Geniza that mention Acre do not contribute to our knowledge about its topography.Footnote 24 Therefore, as far as the present study is concerned, these references are irrelevant.
According to Nāsir-i Khusraw, the city was of a rectangular shape and its port was adjacent to its southern flank. This implies that, unlike Frankish Acre in which built-up areas encompassed the port from the north and west, during the Muslim period the port adjoined only the southern side of the city. This view is corroborated by a recent underwater survey that shows that early-Muslim remains were not found near the southern breakwater of the port, that during the Byzantine period this region was hardly active, and that the southern breakwater had probably been destroyed and could not protect vessels that anchored in the bay.Footnote 25 Hence, there is no reason to suggest that the port of an important city existed on this site.
Acre during the First Quarter of the Twelfth Century
According to a document found in the Genoese municipal archive, the commune received a third of the city, a third of its port's income, a territory extending about a league outside the city, and an annual payment of three hundred besants for its aid during the conquest of Acre (1104).Footnote 26 Unfortunately, this deed does not delineate the precise limits of the Genoese quarter. Therefore, it fails to provide evidence of the exact nature of the royal grant. It seems, however, that no matter what the size of Acre in 1104 was, this grant constituted much less than a third of the city. As late as 1249 the commune issued an inventory of its properties in Acre.Footnote 27 As this document was published, the limits of the Genoese quarter can be ascertained. The commune's quarter was encircled by the assets of the Venetians, the Pisans, the Hospitallers and the Templars, as well as of many other protagonists in the Frankish kingdom. If the quarter's area had decreased, certainly some evidence for this process should have survived, but it does not. The Genoa archives, as well as those of most of its neighbours in Acre, have survived but they do not include evidence that might suggest that the area of the Genoese quarter diminished during this period. Hence, presumably, at this stage, the area of the Genoese commune was not smaller than in 1104. We may then assume that Genoa received then much less than a third of Acre. This conclusion supports the view of Mayer and Favreau that the document of 1104, which states that Genoa received a third of the city, represents a forgery.Footnote 28
In 1110 and 1123, the Venetians received an area in Acre in return for their assistance during the conquest of Sidon and Tyre.Footnote 29 It also seems reasonable that their share was significantly less than a third of the city. According to the previous analysis of the Muslim city's layout, these quarters were in the south-western district of the city, without access to the port's quay. Thus, it is obvious that the Italian cities' quarters were located in a peripheral area. Presumably, Genoa, and later on Venice, received their sectors in what was a newly-developed part of Acre.
A survey of the Genoese quarter and the inventory list of the Genoese communal property in Acre suggest that the commune's main buildings were located near a central square close to the junction of the major streets of the quarter.Footnote 30 The results are straightforward: the Genoese quarter was built during the Crusader period.Footnote 31 Early-Muslim remains were not found in these areas.Footnote 32 Evidently, therefore, the Genoese quarter was built on unoccupied land.
Walking through the impressive sewage system of Crusader Acre clearly indicates that most, if not all, of the Frankish buildings were built on the bedrock itself. Earlier structures are not to be seen there. All this data points to the peninsula having been built up during the Crusader period – and not before it.
According to this analysis, the peninsula's area was divided between its initial holders: the northern parts belonged to Genoa and Venice, while the south and the west belonged to royalty. Other holders, such as the Hospitallers, the Templars, and the Pisans received substantial grants of territory later, during the twelfth century. This last conclusion is supported by later documentation including donations made by the kings throughout the century in the south of the peninsula.
Acre 1125–1187
On 30 April 1135 the Hospitallers agreed to remove a house that they owned near the upper (northern) portal of the Holy Cross cathedral.Footnote 33 An extensive land exchange then took place in 1149 between Queen Melisend, her son Baldwin the III, and the Hospitaller order. The order received houses built in front of the St John the Baptist church,Footnote 34 evidently the Hospitaller church, and neighbouring land parcels within the city.Footnote 35 In return, it relinquished bathhouses that it possessed in the St Leonard neighbourhood. As a result, the royal house acquired control of territories in the city centre and near its eastern wall, whereas the order was able to control a sizeable compound in the city's north-western region.
St Leonard church appears in a list known as ‘The Pilgrimages and Pardons of Acre’.Footnote 36 This document was composed in ca. 1260 and is organised in a roughly clockwise order starting from the gate of St Nicholas in the eastern wall of the city.Footnote 37 The churches of St Leonard and St Roman, which appear in consecutive order in the list, were granted to the abbey of Mount Sion before 1179.Footnote 38 The church of St Roman is depicted in the north-eastern corner of Acre in the map of Marino Sanudo.Footnote 39 Hence, it seems reasonable to suggest that the quarter of St Leonard was called after the church; that it already existed in 1149, and that the north-eastern corner of thirteenth-century Acre was already inside the walls during the first half of the twelfth century. This view suggests that modifications in Acre's eastern wall during the twelfth century were of a local nature. Kedar's opinion that the course of the eastern wall had been changed during the twelfth century relies on a single case. According to Nāsir-i Khusraw, the spring of Ain al-Baqar was outside the city's eastern gate, while later evidence presumably places it within the walls.Footnote 40 This piece of evidence is not sufficient to support an argument that the whole course of the wall was altered. Ain al-Baqar is about 300 meters to the west of Tel al-Fukhar, the Frankish Toron outside Acre. The fact that it was described by Nāsir-i Khusraw in relation to the city's eastern gate implies that in 1047 they were very near to one another. Therefore, if the spring was included within the walls of the city during the twelfth century, this was the result of a local modification of the wall's course. Such an adjustment took place in 1181, not too far from the main gate of the city.Footnote 41 Thus, it is possible that in 1184 Ibn Jubayr described the spring as being inside the city, even if it was outside the walls in the eleventh century, without necessarily implying a general change in the wall's course between the visits of the two Muslim travellers.Footnote 42
Jacoby, following Kesten, has asserted that the city's main axis ran between a gate in the northern wall to the east of the Hospitaller complex and the port in the south.Footnote 43 The existence of such a street would be plausible if the present old city plan was a continuation of the Frankish city, which in turn followed its predecessor's perimeter. If such was the case in both instances, the present central mosque would be on the same site as the Frankish cathedral and the early Muslim central mosque. Yet, in a recent study, Jacoby himself states that a document of 1198 refers to St Nicholas's Gate in the city's eastern wall as the city's main gate.Footnote 44 This clearly indicates that the city's main artery ran in an east-west direction rather than in a north-south one.
The document of 1149 also refers to the ‘Syrians’ as neighbours of the new Hospitaller area and to nearby land parcels. This means that, after almost half a century of Frankish occupation, the area within the city's walls was not exhausted.Footnote 45 The city did not include only ‘Frankish’ quarters, but also an indigenous population which lived in discernable areas.Footnote 46 These certainly were not quarters of the same nature as those of the Franks. They did not enjoy the same rights as the Frankish quarters, and it is quite clear that their residents were occupants, rather than owners, of the property in which they lived. The existence of such districts in Acre implies both Frankish tolerance of their existence and the wish of non-Frankish people to exercise communal life. Obviously, there were also indigenous persons who lived in a scattered pattern elsewhere in the city.
Three different documents issued by King Amalric indicate that as late as the seventh decade of the twelfth-century Acre was undergoing a process of intensive development. In 1165, the King confirmed a transaction between the Hospitallers and Arnulf of Curbinhi. The document describes the deal as follows:
domos illas quas vendidit eis Arnulfus de Curbinhi, que sunt inter murum urbis Acohn [sic] et mare.Footnote 47
[These houses which Arnulf of Curbinhi sold to them are between the wall of the city of Acre and the sea.]
This document has not been hitherto referred to. It clearly suggests that the view that twelfth-century Acre was walled only on its eastern and northern flanks is unrealistic. If the houses were near the seashore, they could neither be between it and the eastern nor the northern city wall. They should have been near to either Acre's western or southern walls. Of these two, the more likely is the western wall thanks to its proximity to the Hospitaller compound. The policy of military orders was to strive to acquire neighbouring assets.Footnote 48 If this 1165 document refers to such a case, the wall of the Hospitaller compound may also have been the city's western wall. It was not on the seashore itself, but a short distance from it. This analysis strongly indicates that in 1165 Acre quite likely had a western wall.
A year later the king granted and confirmed vast properties throughout the Frankish kingdom to the abbey of Templum Domini. In this case, the document concerned refers, among other properties, to assets within and without Acre.
in Accon quoque domum que fuit Guiberti de Portu, duas domunculas juxta domum Cholet sitas et domos quas Adam Burdula de Templo Domini tenuerat; extra muros etiam Accon, fundum in quo ecclesia Sancti Andree sita est et terra adiacentem cum domibus in ea edificatis et edificandis.Footnote 49
[in Acre also a house which belonged to Guibert of Portu, two small houses near the house of Cholet and a house which Adam Burdula held from the Temple of the Lord. Besides, outside the walls of Acre a marketplace in which is the church of St Andrew and an adjacent area where there are houses under construction and where further will be built.]
This text is even more precise. It suggests that in 1166 St Andrew's Church, which stood at the peninsula's southern end, was outside the walls of Acre.Footnote 50 Chalandon, who published this text, punctuated it with a semicolon before the words ‘extra muros’. In other words, it was his understanding that St Andrew's Church was outside the walls of Acre. This opinion was also shared by Röhricht who published an abridged version of the same document four years later.Footnote 51 It is not impossible to punctuate these sentences differently. Yet, the way that the text was published is the most logical. Both Chalandon, who read the original in the library of Monte Casino, and Röhricht were prominent text experts. The fact that they both understood this text in the way that they did is, therefore, significant. This reading plainly implies that in 1166 there was a wall in the promontory that excluded its southern part. This document too, has hitherto been ignored.
The Pisan quarter was developed after 1168 when King Amalric granted the commune a small area in which to build a church and a public building.Footnote 52 In 1182, the commune received a square near the port.Footnote 53 Two documents from the period between the defeat at Hattin and the re-conquest of Acre record generous grants made to the Pisans by Conrad of Montferrat and Guy of Lusignan.Footnote 54 According to Jacoby, however, there is good reason to believe that these documents were forged. He asserted that the new suburb referred to in these documents should be understood as Montmusard. As a result of these documents the Pisans received most of the city. Therefore, from his point of view, it seems quite clear that these documents could not be any thing other than a blatant forgery.Footnote 55
But while the two deeds are almost identical, there are a few differences between them. The first document records a grant made by Conrad of Montferrat in October 1187, a few months after the Frankish defeat near the Horns of Hattin. A superficial reading of this document would suggest that he granted the commune about three-quarters of the city. At this stage, many important nobles who perished during the battle or were taken prisoner were absent. Presumably, after the defeat people were in such a state of despair that they did not pay heed to details and had little hope of retaking the city. Moreover, it should be borne in mind that Conrad was an outsider, who did not know the city well; therefore the Pisans could easily deceive him. In November 1189, after the beginning of the siege of Acre and the release of war prisoners, Guy of Lusignan, who knew Acre quite well, issued the second document, which is almost identical to the first. In this case, there is no room for arguments that Guy did not know the city, or that he did not expect to take it. Therefore, whether he granted three quarters of the city or not, he did it with full awareness to the possible implications.
Jacoby's theory relied exclusively on the understanding that a new suburb referred to in the document was Montmusard. Yet, he ignored the fact that these documents did not mention Montmusard but an anonymous new suburb. Elsewhere he wrote that Montmusard did not exist as a suburb prior to the Frankish re-conquest of Acre (1191). Therefore, the argument that these documents should be deemed as forgeries is unsustainable. If the suburb did not exist before 1191, how could it be referred to prior to this date? Evidently, if someone forged a document in order to claim rights on a certain site, this should have existed at the time that the forgery took place. Otherwise, what was the point of falsifying the document? Footnote 56
According to the working hypothesis of this essay, the burgus novus in the documents of Conrad of Montferrat and Guy of Lusignan was not Montmusard. Rather, it was located in the south of Acre's peninsula, in the vicinity of the ‘old’ Pisan quarter, established after 1168. This was one of the vacant areas in the peninsula after almost a century of urban expansion. Such an interpretation places the two documents in a clearly urban context. They do not portray the majority of Acre's territory in an odd way, but instead describe a certain expansion of the Pisan quarter in the southern part of the peninsula. This view is corroborated by the fact that all the ecclesiastic institutions mentioned in these documents were in the peninsula. The last sentence in both documents is of a restrictive nature: it fully acknowledged the rights of the Pisans to their quarter and to other properties in Acre, though not to three-quarters of the city. In this state of affairs there is no room for the argument that discredits the documents.Footnote 57 They were just intended to describe a certain expansion of the Pisan quarter, but not three-quarters of the city.
These two documents also refer to properties both outside and inside the Gate of St Andrew. These references clearly support the view that this was a gate in the city's wall and not, the gate of the church itself. Both documents also describe an entire quarter that lay between the gate of St Andrew and the Cape of Wrath. Therefore it is plausible that the gate in the wall was accessed by a street that led directly from the church of St Andrew.
There are good reasons to suggest that this gate was to be found in the wall referred to in the 1166 document that described St Andrew's church as being outside it. To be more precise, the gate probably was at the southern end of the Genoese quarter's south-north artery, identified by Kool as the Carrubium Street.Footnote 58 Accordingly, the Baths' Gate was to the east of this point, near the existing early-modern small bathhouse, in present-day Alexander Moqdon Street. It seems that this wall had been built during the early-Muslim period, when sea walls were a typical feature in all the cities of the Levant littoral.Footnote 59 The Genoese quarter, the first to be established by an Italian commune in Acre, was far from the city's centre, and did not have access to the quay of the port. There is no reason to believe that later grants included more advantageous conditions.
The notion that the documents issued by Conrad of Montferrat and Guy de Lusignan were of a limited scope is also expressed elsewhere. The bishop of Acre invested the Pisan church of St Peter with parochial rights and included a description of the limits of the parochial jurisdiction of the church of St Peter.Footnote 60 This document deals with an area which appears to be even larger than the area described in the two above charters. Yet, it also did not exceed the peninsula's area. In this case the situation is even clearer. Evidently, the bishop of Acre did not give up the rights of his church which extended to almost the whole city. He just invested the Pisan church with parochial rights in a rather small area. Also this document acknowledges the rights of other actors in the vicinity of the peninsula.
This document thus clearly supports the view that the two first ‘Pisan’ documents were not forged. It included vast rights in the peninsula just as the other two did. Its inclusion in the debate would not seem to support Jacoby's view that the two first documents were forged.
To sum up: all three documents referred to the Pisans' areas and rights in the peninsula itself. Montmusard was not the only new suburb established during almost two centuries of Frankish presence in Acre, and there is no way to explain its appearance in such early documents.
One fact is clear: as time passed by, the kings of Jerusalem relinquished most of their estates in the south and the west of the peninsula. This suggests that they considered these areas to be less important, otherwise they would not have transferred them to others.Footnote 61 As a result, the peninsula came to be occupied by important actors in Acre's life. They received their land grants in this specific area due to royal preferences, but also because of the availability of empty land there. Thus, the Italian communes were able to establish spacious quarters, and the Templars and Hospitallers could likewise build their compounds there. This land ownership pattern was unique for this part of the city. Elsewhere in Acre the land mostly belonged to the monarch, to the cathedral, and to small owners.
The Port of Acre
The port of Acre was the most important in the Latin Kingdom of Jerusalem and as such it was a key element in the city's urban layout.Footnote 62 It undoubtedly influenced the city's street network and demanded precious space for the location of port-related facilities. According to many scholars, the quay of the port was at the southern breakwater.Footnote 63 This popular view is based on the assumption that Acre's peninsula was a significant part of the city's urban tissue prior to the twelfth century.
According to this view, the Frankish port functioned in the same location as did the early-Muslim port, which succeeded the Byzantine port.Footnote 64 This understanding is based on a possible interpretation of a Byzantine chronicle and the assumption that the Muslim port was re-established in the ninth century in the same place. This is a very flimsy theory, however, for recent studies clearly suggest that the ninth-century port reconstruction project was carried out in the area of the eastern pier.Footnote 65 Galili et al. have suggested that it was built as a barrier against hostile ships and as an approach road to the Tower of the Flies. In their opinion, it was pointless to build a breakwater because the waves from the east did not endanger anchored vessels.Footnote 66 Underwater archaeological surveys near to the southern breakwater do not support the view that this was the site of an important port.Footnote 67 Byzantine and early-Muslim remains have hardly been found in this area.
The view that the twelfth-century port was on the peninsula's eastern shore implies that many related facilities were also located there. Yet, there is no available evidence that such facilities existed in this area. Briefly, the theory that the port of Acre was on the peninsula's seashore between the Byzantine period and the twelfth century does not rely on sound facts.
The view that the port was in the South of the peninsula also contradicts the document that described the church of St Andrew as being outside the city walls. It also opposes the description produced by Nāsir-i Khusraw who claimed that the port's sidewalls were extensions of the city walls, which were evidently elsewhere.Footnote 68
All the above arguments indicate that during the early-Muslim period the port of Acre was not on the south-eastern shore of the peninsula. The logical possibility is, therefore, that it was at the bay's northern shore, probably by the eastern pier. If it was transferred to the south during the Crusader period, it would be necessary to explain when and why this major change took place.
Gertwagen has suggested that the eastern pier decomposed and ceased to function during the Crusader period. She based her view on a possible interpretation of a single thirteenth-century source.Footnote 69 This could explain a possible transfer of the quay from the eastern pier to the southern breakwater. However, given the fact that the description used by Gertwagen dates from the thirteenth century, it is not applicable to earlier periods. The possibility that the quay was transferred from the eastern pier to the southern breakwater during the thirteenth century needs to be discussed within a clear historical and urban context. It might be associated with the transfer of the royal court from the gate of St Nicholas about the same time.Footnote 70 Nevertheless, its verification or refutation is beyond the scope of this study, and therefore will not be discussed here.
The opinion that the port was in the south-eastern sector of the peninsula also implies that caravans had to cross the entire city. To do this, the city's street network had to enable two-way traffic and it would have been necessary to earmark precious land to house transport animals, warehouses, workshops, a customs house, and so on. There is no evidence for the existence of either a sophisticated street network or port facilities in the peninsula during the twelfth century. The loading and unloading of merchandise and people embarking and alighting in this area makes no sense. For example, Ibn Jubayr estimated that the ship on board which he travelled from Acre back to Europe accommodated about 2,000 pilgrims.Footnote 71 This figure seems to be exaggerated, but certainly there were in this ship, and in many others, a large number of pilgrims. What was the impact of such a multitude on domestic traffic in Acre? During the pilgrimage season, many vessels anchored simultaneously in the port.Footnote 72 Did those passengers and their baggage pass through the city's alleys? How were the local authorities able to regulate the traffic during this particular period?Footnote 73 This is, indeed, no more than an auxiliary argument. If the Franks had found an active port near the southern breakwater, it might have been expected that they would have continued to use it, despite its disadvantages. Yet, as far as the available data indicates, the situation was just the opposite. Therefore, the idea that the Franks transferred the port to a site where accessing it was rather difficult requires much more evidential support.
To sum up this section, we can say that there is enough evidence for the quay being at the eastern pier during the early-Muslim period. It is, therefore, plausible that this was also its location during the twelfth century. The idea that it was elsewhere implies that the city underwent dramatic changes during the twelfth century. It seems impossible to reconcile this view with the opinion that the city did not develop significantly during the same period.
The suggestion that the quay was located near the southern breakwater, in the peninsula's south-eastern region, is based on the idea that all the port's facilities were in a compound, or quarter, known as the cathena.Footnote 74 This name is derived from the chain which ran between two towers in the port's main access.Footnote 75 According to Jacoby, the chain stretched between the southern breakwater and the south-eastern shore of the peninsula, near the present-day Khan al-‘Umdan.Footnote 76 It, thus, created a small basin which, in Jacoby's opinion, was adjacent to all the important port facilities, as well as to the Court of the Chain. He maintains that these were located on the site of the present Khan al-‘Umdan.Footnote 77 A key element in Jacoby's thesis is that the chain itself was near a quarter bearing the same name. Hence, the chain blocked the entrance to a very small area in the most inaccessible corner of the port. This idea raises major questions: what was the point of protecting a small area, which was in any case hardly accessible, while leaving the major part of the port undefended? Moreover, the chain was installed at the port's entrance during the early-Muslim period. This probably was not in the location that Jacoby suggested, for it was a dormant area throughout that period.Footnote 78 Since the chain was installed by Ibn Tulun, it would be reasonable to assume that it blocked the main access to the port itself, rather than to a neglected corner in Acre's bay.
According to Jacoby, an inner bay that has been depicted on the map of Marino Sanudo, near, or even on, the site of the Khan al-‘Umdan is “a clumsy depiction of the western harbour”.Footnote 79 He believes that the Ottoman khan stands on the site of the Court of the Chain.Footnote 80 It is important to emphasise that the medieval foundations atop of which the pre-modern edifice apparently was built are either a mere fairy tale or a sort of popular belief. This area is one of the only sites in Acre that somehow have not been excavated. Hence, since the Ottoman khan was built almost two centuries ago no human eyes saw its ‘medieval foundations’. Gertwagen claimed that drillings in the area of the khan indicate that it was built on solid rock, which suggests that it was not underwater during the Crusader period.Footnote 81 If there was any support for this claim, it would suggest that the khan could have been built on earlier foundations. Yet, according to recent studies no drillings were carried out inside nor outside the khan's area; therefore, they could neither refute nor support the view that this area was on solid soil during the Crusader period.Footnote 82
Jacoby's suggestion that this is a clumsy description of the shore line deserves special attention. Since he wrote his article before the alleged drillings took place, he was in no position to rely on them or disagree with their results. His analysis is based exclusively on the way that he understood the map's manuscripts. For the most part, he considered the map as a rather reliable source. Yet, he emphasised that whenever the map does not concur with archaeological or documented evidence, the latter should be preferred.Footnote 83 The case of the alleged ‘inner basin’ constitutes an exception, for the map's information has been discarded on the basis of assumptions.
It seems that this interpretation was very instrumental in Jacoby's reconstruction of the port area during the Crusader period. The claim that this was the venue of the Court of the Chain is a key element in his hypothesis about Acre's urban layout. If it was underwater, as the maps seem to suggest, he should have had to consider that the court and the port were elsewhere.
N. Makhouly and C.N. Johns, who published an excellent guide to Acre, wrote that an inner basin was still discernable in this region as late as the eighteenth century.Footnote 84 Later studies have proved uncertain as to whether this was the site of an inner basin or not. Boas left it as a pending issue, quoting later references according to which an inner basin on this very site was still discernable when Napoleon besieged Acre (1799).Footnote 85 However as long as the claim that Khan al Umdan had been built on medieval foundations, or at least that during the Crusader period this area was on solid soil, remains unwarranted, it should be considered as an inner basin in Acre's port. This case illustrates the process by which a groundless belief becomes a consensual ‘scholarly’ view. It seems fair enough to state that the fourteenth-century cartographer did not clumsily describe the coastline of Acre. As a result, the ruga cathene, which is a pivotal axis of Jacoby's theory, has to be understood in the narrow sense of the term: the street rather than the quarter of the chain.Footnote 86
To sum up, convincing evidence exists that before and during the twelfth century Acre's port was located on the bay's northern shore. All other theories are just mere speculations by modern scholars.
The Royal Court
The royal court (fonde) was another key element in Acre's topography. Jacoby has suggested that according to a deed issued in 1188 by Conrad of Montferrat, the fonde was near the main land access to the city, namely the gate of St Nicholas.Footnote 87 His view is supported by an explicit clause in Document No. 128 of the Teutonic order archive. This document is a summary of all the transactions in which the order was involved prior to 1243. It includes, inter alia, the following article: ‘item rex Guido dedit nobis veterem curiam prope portam sancti Nicholay’Footnote 88 [and also King Guy gave us the old court near the gate of St Nicholas], which clearly indicates that the royal court was near the gate of St Nicholas. Yet, it also implies that a new court was built elsewhere before 1192. The original document, to which Deed 128 refers, did not mention that the court was an old one.Footnote 89 This conclusion suggests that the royal court was near the gate of St Nicholas throughout most of the twelfth century. Likewise, it implies that the court was transferred to an as yet unknown site. This approach could be consistent with the idea that the quay of the port was also transferred about that same period. The changed location of the quay demanded substantial modifications in the urban plan. The establishment of a new royal court was a natural development in such a case.
Discussion
Twelfth-century Acre is described by modern scholars as a huge city.Footnote 90 Certainly, most of its population did not live on the peninsula. There were the quarters of the Italian cities as well as the Templar headquarters. The Hospitaller compound was just nearby. These key actors in the city's life did not establish populous communities during the twelfth century. Assuming that the port's facilities were also in the peninsula does not leave room for more than a tiny population.Footnote 91 The occupants of the peninsula and the Hospitallers, whose compound was near to it, were strong and wealthy protagonists, but they did not act in a vacuum. The city and its port belonged to the royal house. The kings should have been instrumental in city life, not only as sellers and donors, but also as buyers, barterers, and confiscators. The existing documents mostly referred to them as they deal with transactions in which involved organisations whose archives survived. The city's centre, where major owners others than royalty hardly had a foothold, is hardly referred to in the documentation. This is the result of the partial survival of contemporary archives. Ordinary citizens hardly appear in records, due to the disappearance of secular archives and since they were secondary actors in urban affairs.
Therefore, scholars have to bear in mind that the documented history of medieval Acre is biased. While there is a significant body of data concerning the peninsula, the majority of the city, which lay between the eastern wall and the Hospitaller complex, is much less well-documented. Yet, most of Acre's residents lived in the royal quarters, outside the peninsula.Footnote 92
Archaeological excavations confirm the view that the peninsula was not the scene of a vibrant urban centre prior to the twelfth century. When their findings are combined with the documentation discussed above, the weakness of the argument that Acre was stagnant for about a century clearly stands out.
A central element in the theory that Acre did not develop during the twelfth century is the location of its port. If it was not—as proposed—near the southern breakwater, those who believe that Acre stagnated during the twelfth century have to return to the drawing board. That this point has been so crucial for the theory is reflected in the fact that Jacoby did not hesitate to describe a clear feature in Marino Sanudo's map as a clumsy depiction. I suggest instead that the port's quay, both during the early-Muslim period and the twelfth century, was located near the eastern pier. It seems plausible that this structure was the main scene of the port's activities during the early-Muslim period and the twelfth century. The combinations of these archaeological findings with the urban considerations indicate that the port's main quay was on the eastern pier or close to it, and not in the southern breakwater.
The location near the eastern pier was easily accessed from the funda Regis, which might solve most of the transportation problems. This also provides an explanation for the rather strange land granting policy adopted by the kings in the peninsula. Such considerations would not have been alien to the rulers of Acre during the twelfth century.
Assuming that the quay was near the eastern pier provides satisfactory explanations for other issues regarding the port. Analysis of the navigation possibilities reveals that ships, which entered the port from the south, were unloaded near the eastern pier and continued in an anti-clockwise direction. This method could be applicable for small and perhaps also medium-size ships that arrived from overseas, as well as for boats carrying passengers and goods from and to larger ships that were anchored outside the port. Ships and boats could moor near the eastern pier on either side, a possibility that would make loading and unloading of ships more efficient.
Summary
Twelfth-century Acre was considerably larger than its Muslim predecessor and its port was much busier than the early-Muslim city's harbour. The majority of the city's population lived in the area of the pre-Frankish city. It was, likewise, the venue of a royal residence, a cathedral, and some religious orders. The peninsula, until then a rather large and mostly vacant area, had been settled by the quarters of the Italian cities and the military orders. The monarchs of Jerusalem handed over vast properties in the peninsula to the Italians, the military orders, and ecclesiastical institutions during the twelfth century, and in all probability concentrated their efforts elsewhere. The fact that many thirteenth-century transactions occurred in an area that this article considers as the old district, emphasised royal preferences. The kings began to grant lands in the old district only after the peninsula was entirely occupied.
Consequently, in the twelfth century the port's quay was close to the eastern pier, where it had been located during the early-Muslim period. It was easily accessible from the nearby main land gate and the funda Regis. Moreover, it was more accessible for ships than was the southern breakwater and resolved some essential traffic issues which might have existed if the quay had been located near the southern breakwater.
