Fig 1 - uploaded by Katharina Rieger
Content may be subject to copyright.
Map of Northern Libyan Desert with the investigation area of the Eastern Marmarica-Survey (map A.-K. Rieger)
Contexts in source publication
Context 1
... plain, stony, and uniform, mainly a dry (full) desert. However, to the North, vivid harbour cities like Paraitonion (Marsa Matruh), Leuke Akte (Ras el-Hekma) and Ainesisphyra (Sidi Barrani) are situated on the Mediterranean coast. 300 km to the south, the Oasis of Siwa constitutes a center of ex- change and production in the Northeastern Sahara ( Fig. 1). To the west and east lie the fertile regions of the Nile Valley and the Green Mountain, the Gebel el-Akhdar in the Cyrenaica. Nevertheless, even in the semi-arid Marmarica, it was possible to cultivate crops on favourable spots in its northern parts, based on water harvesting systems 2 . Due to the economically significant regions ...
Context 2
... (Fig. 7b) 14 . They may have been fireplaces, since some of the stone cobbles show signs of firing, but they show neither typologically recognisable forms nor The northwestern building at Abar el-Kanayis and its courtyard wall, covered by drifted sands. At the northern end of the building the higher debris heap marks the chain of rooms (cf. Fig. 10), view to northeast (photograph O. Klammer) MDAIK 68, 2012 chronologically diagnostic characteristics. For one campsite (Fig. 7c) people used the stones from the buildings, indicating a terminus post quem after their abandonment in the 6 th century CE. The scattered surface finds consist mainly of pottery sherds dating to Graeco-Roman ...
Context 3
... southern sides (Fig. 4). On the northern side, a smaller courtyard completes the building. The northern part of the complex was investigated more closely by opening seven trenches ( Fig. 9, trench 2, and 6 to 11). The building material consists of a lacustrine limestone that heavily weathers when exposed to cli- matic impacts, and of mudbricks (Fig. 6b, 10, 11). The double-faced base walls are made of well cut lime- stone ashlars, filled with cobbles and earthen, loamy mortar (Fig. 10b, 12). They reached a height of at least 1.20 m, as the preserved four courses of ashlars in wall 61 and 113 testify (Fig. 12b). The base walls of ashlars were erected directly on the natural rock (58, only ...
Context 4
... more closely by opening seven trenches ( Fig. 9, trench 2, and 6 to 11). The building material consists of a lacustrine limestone that heavily weathers when exposed to cli- matic impacts, and of mudbricks (Fig. 6b, 10, 11). The double-faced base walls are made of well cut lime- stone ashlars, filled with cobbles and earthen, loamy mortar (Fig. 10b, 12). They reached a height of at least 1.20 m, as the preserved four courses of ashlars in wall 61 and 113 testify (Fig. 12b). The base walls of ashlars were erected directly on the natural rock (58, only with a thin levelling layer beneath, 57 in case of wall 113, 42 in case of wall 11, 69 in case of wall 61, and 93 in case of wall 10; ...
Context 5
... limestone that heavily weathers when exposed to cli- matic impacts, and of mudbricks (Fig. 6b, 10, 11). The double-faced base walls are made of well cut lime- stone ashlars, filled with cobbles and earthen, loamy mortar (Fig. 10b, 12). They reached a height of at least 1.20 m, as the preserved four courses of ashlars in wall 61 and 113 testify (Fig. 12b). The base walls of ashlars were erected directly on the natural rock (58, only with a thin levelling layer beneath, 57 in case of wall 113, 42 in case of wall 11, 69 in case of wall 61, and 93 in case of wall 10; Fig. 14a, b, 12b, 17c) without any construction pit due the negligible soil strata on the desert margin, which generally do ...
Context 6
... 10b, 12). They reached a height of at least 1.20 m, as the preserved four courses of ashlars in wall 61 and 113 testify (Fig. 12b). The base walls of ashlars were erected directly on the natural rock (58, only with a thin levelling layer beneath, 57 in case of wall 113, 42 in case of wall 11, 69 in case of wall 61, and 93 in case of wall 10; Fig. 14a, b, 12b, 17c) without any construction pit due the negligible soil strata on the desert margin, which generally do not exceed 30 ...
Context 7
... that, mudbrick is used for the mural con- struction ( walls 1, 10, 11, 61, 83, 95, 113; Fig. 11). A re- markable amount of straw and rachis of barley (Hor deum vulgare) was found to serve as tempering for the mudbricks (e. g. of wall 96, Fig. 13) 15 . Loamy soils and water were available on site in the depression; but the temper -barley straw -had to be brought to Abar el-Kanayis, a place on the desert fringe, where the sebhka soil and water conditions are not at all suitable for crop ...
Context 8
... the staircase on the northeastern wall of the rooms, the usual narrow width of 1.0 m is doubled to 1.9-2.0 m (wall 89, Fig. 9, 12c, 16a) -likely to carry a corridor in the upper part. The three different base wall dimen- sions correspond to four, five or eight lines of mud- bricks in the rising construction, since the mudbricks measure 24 × 24 cm (Fig. 11b), which is slightly smaller than mudbricks found in settlements in Eastern Mar- marica on the Northern Tableland 16 . In some instances, courses of mudbricks are still preserved (up to four courses on wall 95, and four bricks side by side, on wall 10, Fig. 11a). The stone fundament is raised high enough to fulfill its purpose as ...
Context 9
... of mud- bricks in the rising construction, since the mudbricks measure 24 × 24 cm (Fig. 11b), which is slightly smaller than mudbricks found in settlements in Eastern Mar- marica on the Northern Tableland 16 . In some instances, courses of mudbricks are still preserved (up to four courses on wall 95, and four bricks side by side, on wall 10, Fig. 11a). The stone fundament is raised high enough to fulfill its purpose as protection of the loamy building material from the humidity of the un- derground and occasional splash waters. In a later phase (see below, pp. 151-153) roughly cut blocks and rubbstones were used for secondary walls (12 and 64; Fig. 10a, 14a) that are rather negli- ...
Context 10
... four bricks side by side, on wall 10, Fig. 11a). The stone fundament is raised high enough to fulfill its purpose as protection of the loamy building material from the humidity of the un- derground and occasional splash waters. In a later phase (see below, pp. 151-153) roughly cut blocks and rubbstones were used for secondary walls (12 and 64; Fig. 10a, 14a) that are rather negli- gently stacked. The stone-built parts of the later walls reach a maximum of 1.0 m in height; no traces of the rising parts are preserved, but presumably they were also made from ...
Context 11
... walls form three rooms on the northern side of the courtyard in the primary layout of the entire complex -each of them covers 15 to 20 m 2 -and is likely to be mirrored on the southern side. The south- eastern room opens to the outside through a large gateway (Fig. 9, 15a). This room may have been con- nected to the large courtyard by a presumed door- way in wall 61. A doorway to the building was found at its eastern part (Fig. 9). No inner doorways connect the rooms. Two doorways lead from the small court- yard (see below, pp. 152-153) in the northern part of the complex to the northwestern rooms and ...
Context 12
... room may have been con- nected to the large courtyard by a presumed door- way in wall 61. A doorway to the building was found at its eastern part (Fig. 9). No inner doorways connect the rooms. Two doorways lead from the small court- yard (see below, pp. 152-153) in the northern part of the complex to the northwestern rooms and were later blocked (Fig. ...
Context 13
... long northern wall backing the rooms is ad- joined by stairs leading to a second storey. Six steps are preserved, with complete slabs forming the steps, each reaching a height between 20 and 25 cm (Fig. 16a, b). A presumed seventh step led to the top of the wide stone base along the northwestern parts of the build- ing (wall 89), that served as landing and/or corridor. The stairs were blocked by wall 107 in a later phase - we assume, that the second storey was not in use any more at this time (Fig. 16a). The stone foundations of this part of ...
Context 14
... each reaching a height between 20 and 25 cm (Fig. 16a, b). A presumed seventh step led to the top of the wide stone base along the northwestern parts of the build- ing (wall 89), that served as landing and/or corridor. The stairs were blocked by wall 107 in a later phase - we assume, that the second storey was not in use any more at this time (Fig. 16a). The stone foundations of this part of the northern wall have a width of 1.9- 2.0 m and thus have the capacity to carry a second storey made accessible by the staircase (Fig. 12c, 16a). The debris of collapsed construction material reaches the highest extension in this part of the build- ing, which corroborates the assumption of a ...
Context 15
... that served as landing and/or corridor. The stairs were blocked by wall 107 in a later phase - we assume, that the second storey was not in use any more at this time (Fig. 16a). The stone foundations of this part of the northern wall have a width of 1.9- 2.0 m and thus have the capacity to carry a second storey made accessible by the staircase (Fig. 12c, 16a). The debris of collapsed construction material reaches the highest extension in this part of the build- ing, which corroborates the assumption of a second storey above the northwestern rooms of the building (Fig. 6b) 17 . Whether there had been a second staircase to the northwestern rooms of the upper floor leading up from the large ...
Context 16
... 6b) 17 . Whether there had been a second staircase to the northwestern rooms of the upper floor leading up from the large courtyard, was not recognisable af- ter a cleaning of wall 61. A second courtyard is adjacent to the northern part of the building, surrounded by wall 13, which was made to the same high quality standard as walls 10 and 61 ( Fig. 9, 12c). It is based directly on the natural bedrock. The courtyard itself is only filled with blown sands (Fig. 17a, b), corresponding to the stratigraphy in the large courtyard (Fig. 17c); the only artificial fea- ture is the levelling layer, that represents the living floor ( Fig. 17a), similar to the rooms defined by walls 10, 11, 61, 95 ...
Context 17
... the large courtyard, was not recognisable af- ter a cleaning of wall 61. A second courtyard is adjacent to the northern part of the building, surrounded by wall 13, which was made to the same high quality standard as walls 10 and 61 ( Fig. 9, 12c). It is based directly on the natural bedrock. The courtyard itself is only filled with blown sands (Fig. 17a, b), corresponding to the stratigraphy in the large courtyard (Fig. 17c); the only artificial fea- ture is the levelling layer, that represents the living floor ( Fig. 17a), similar to the rooms defined by walls 10, 11, 61, 95 and 113). A floor in the literal sense of the word was not recognisable in the first phase of the building. The ...
Context 18
... A second courtyard is adjacent to the northern part of the building, surrounded by wall 13, which was made to the same high quality standard as walls 10 and 61 ( Fig. 9, 12c). It is based directly on the natural bedrock. The courtyard itself is only filled with blown sands (Fig. 17a, b), corresponding to the stratigraphy in the large courtyard (Fig. 17c); the only artificial fea- ture is the levelling layer, that represents the living floor ( Fig. 17a), similar to the rooms defined by walls 10, 11, 61, 95 and 113). A floor in the literal sense of the word was not recognisable in the first phase of the building. The trenches in the courtyards yielded no finds. According to the masonry's ...
Context 19
... was made to the same high quality standard as walls 10 and 61 ( Fig. 9, 12c). It is based directly on the natural bedrock. The courtyard itself is only filled with blown sands (Fig. 17a, b), corresponding to the stratigraphy in the large courtyard (Fig. 17c); the only artificial fea- ture is the levelling layer, that represents the living floor ( Fig. 17a), similar to the rooms defined by walls 10, 11, 61, 95 and 113). A floor in the literal sense of the word was not recognisable in the first phase of the building. The trenches in the courtyards yielded no finds. According to the masonry's characteristics and the general layout we assume that this smaller court- yard was built ...
Context 20
... the northern corner of the smaller courtyard lies a single room separated by a stone wall and enclosing an area of 1.6 × 1.4 m with an entrance (0.6 m wide) in the southeastern wall ( Fig. 9, 12c, d). A mudbrick ped- estal or bench is added to its northwestern wall 13 (0.5 m deep). Since the stone wall is erected with a butt joint to the courtyard's wall 13, the room was an addition, but its chronological relation to the court- yard appears to be contemporaneous, since the level- ling layer is the same as for wall 13. Based on the ...
Context 21
... the northwestern building, a substantial layer of destruction debris (parts of collapsed walls, burnt mudbricks, ash and charcoal accumulations), recognisable all over the northern part of the build- ing, divides the life span of the building in two stages (Fig. 14, 18a-c): After its erection and first phase of use, the later reuse of the building that is recognisa- ble in the construction of several walls on a higher level can be considered a second phase 19 ...
Context 22
... a first step of stable constructions on the site, the outer walls 10 and 61 running from southeast to northwest as well as 83 and 113 running southwest to northeast were erected, since they are based directly on the bedrock (Fig. 12b, 14, 17). The shorter walls that separate the single rooms (11 and 95) were brought up in a second step of this construction phase, since more substantial levelling layers of compacted soil Eastern Marmarica-Survey were spread underneath the ashlars (36 in Fig. 14b). Especially wall 95 is interwoven with wall 10, i. e. built in one sequence ( ...
Context 23
... southwest to northeast were erected, since they are based directly on the bedrock (Fig. 12b, 14, 17). The shorter walls that separate the single rooms (11 and 95) were brought up in a second step of this construction phase, since more substantial levelling layers of compacted soil Eastern Marmarica-Survey were spread underneath the ashlars (36 in Fig. 14b). Especially wall 95 is interwoven with wall 10, i. e. built in one sequence ( Fig. 9). The same applies to the cor- ner room in the smaller ...
Context 24
... of the 3 rd century CE. The intensive use of the site from Roman Imperial times onwards is emphasised by several finds of Roman local Egyptian Amphorae (AE 3.1) mixed into later layers (e. g. K9, K11, K32, K40 from the debris in the corner room). Within the same level containing material from the destruction a foot of an imported Greek Amphora (Fig. 21, K12) was found, made in the 1 st or 2 nd century CE, while Ptolemaic material does not appear in any stratigraph- ical layer. This composition of finds in the archaeolog- ical contexts of the building suggests that its con- struction took place already in later Roman Imperial times, however not even before the second half of the 3 rd ...
Context 25
... of the building and the site took place in the 6 th cen- tury CE, ending with a sweeping fire that struck the entire complex and destroyed large parts of it 20 . A thick layer of debris (up to 60 cm deep), formed by burnt mudbrick, ash, and collapsed walls testifies to this moment of catastrophe at the cistern site on the Marmarica-Plateau (Fig. 14, 18). The finds in the de- struction layers (= end of Phase 1) were blackened and some sherds were secondarily overfired, which points to the high temperatures that occurred. An al- most complete Late Roman Amphora (K16, Fig. 21) from Cilicia, dating to the second half of the 6 th cen- tury CE is the youngest piece in the destruction lay- ...
Context 26
... ash, and collapsed walls testifies to this moment of catastrophe at the cistern site on the Marmarica-Plateau (Fig. 14, 18). The finds in the de- struction layers (= end of Phase 1) were blackened and some sherds were secondarily overfired, which points to the high temperatures that occurred. An al- most complete Late Roman Amphora (K16, Fig. 21) from Cilicia, dating to the second half of the 6 th cen- tury CE is the youngest piece in the destruction lay- ers, brought to the site before the moment of de- ...
Context 27
... el-Kanayis − northwestern building Trench 6 Section 6/1 (NW) Section 6/1 (NW) quent accumulation of levels by blown sand over time, with some bricks, pottery, and stones ( Fig. 17). Above that, the drifted sands covered the courtyard and rooms after the abandonment of the building; the same is true for the smaller courtyard on the northwestern side, where only layers of continuous use appeared. The reasons for the heavy fire and collapse of the building in Byzantine times remain unknown. It may have happened due ...
Context 28
... destruction material was levelled, forming a layer of 40 to 50 cm in depth, for a new utilisation of the building, where the rooms' layout was reshaped (Phase 2). The rough construction technique of the new walls differs remarkably from the refined ones of Phase 1 (see above, p. 145 and Fig. 10, 12b). Again the pottery finds within the contexts belonging to this phase are low in number: A highly fragmented, but almost complete jug (K21, Fig. 22) probably dating to Late Roman times and a hand-made cooking pot were found next to residual finds, mainly locally pro- duced Roman amphorae (AE 3.1). In some of the rooms walls were set ...
Context 29
... 22) probably dating to Late Roman times and a hand-made cooking pot were found next to residual finds, mainly locally pro- duced Roman amphorae (AE 3.1). In some of the rooms walls were set against the former ones of Phase 1 (e. g. wall 64), maybe as a measure of stabili- sation, others were inserted to create new, smaller rooms (f. e. wall 12, Fig. 12). Doorways were blocked, so that the passage between the rooms and into the smaller courtyard in the northeast was not possible any longer (Fig. 9, 10a, 15b). Access was provided on a higher level by small doorways facing southeast. The second storey presumably was out of use in this phase -at least the stairways (context 125) did not ...
Context 30
... 3.1). In some of the rooms walls were set against the former ones of Phase 1 (e. g. wall 64), maybe as a measure of stabili- sation, others were inserted to create new, smaller rooms (f. e. wall 12, Fig. 12). Doorways were blocked, so that the passage between the rooms and into the smaller courtyard in the northeast was not possible any longer (Fig. 9, 10a, 15b). Access was provided on a higher level by small doorways facing southeast. The second storey presumably was out of use in this phase -at least the stairways (context 125) did not function anymore, since they were blocked by the later wall 107 ( Fig. 9, 16a). The four burials found in the excavated area were placed in the ground during ...
Context 31
... rooms and into the smaller courtyard in the northeast was not possible any longer (Fig. 9, 10a, 15b). Access was provided on a higher level by small doorways facing southeast. The second storey presumably was out of use in this phase -at least the stairways (context 125) did not function anymore, since they were blocked by the later wall 107 ( Fig. 9, 16a). The four burials found in the excavated area were placed in the ground during Phase 2 of the building, which began directly after the destruction since their pits do not cut any layers of drifted sands 22 . Even though the building seems to have lost its character as a roadside inn, the cisterns and the route must have been in use in ...
Context 32
... period of reuse was very limited, and the final abandonment of the building happened at the end of the 6 th century CE (end of Phase 2) since the youngest pottery pieces from that phase can be dated to this time (K15, Fig. 21). A piece of palm tree wood (Phoenix dactylifera), coming from the first level of drifted sands above the living floor of the building is dated to the 7 th century CE by C14-dating and confirms that the build- ing was subsequently covered by sands from this time onwards. It is difficult to gain a more detailed insight into the ...
Context 33
... (39 %; MNI: 32) and Coarse Ware (33 %; MNI: 27) are the main groups on the site, followed by hand-made pottery (15 %; MNI: 12) -only Northern Libyan Desert Ware -, Fine Ware (6 %; MNI: 6) and wheel-made Cooking Ware (7 %; MNI: 8) (Fig. 19a) 26 ...
Context 34
... wares prevail over imported ones, but a few Amphora imports, as well as Fine Ware and Cooking Ware imports, can still be observed, especially in Late Roman times (Fig. ...
Context 35
... Amphorae (Fig. 21, both have the same beige-brown colour with white and black inclusions, reddish ones, quartz and some mica and are of South Italian origin, probably from Calabria or Sicily 43 . The Keay LII Amphorae were made in the 5 th and the beginning of 6 th century CE and be- came immediately abundant in the Eastern Mediter- ranean 44 . South ...
Context 36
... spectrum of forms is very limited (Fig. 21, 22, K18-K30). Closed vessels are represented mainly by Type: Two-handled Jug/Filter Jug? Parallels: M. Egloff, Kellia, pl. 67.2 (beginning of 5 th century CE); Tanis: L. Bavay/J.-L. Bovot/O. Lav- igne, La céramique tardive, fig. 24. C o n t e x t : E n d o f P h a s e 1 . K26 -ID 1140, Bef. 33, DM: 10 cm. Fabric: F2. Surface abraded. ...
Context 37
... quantity of Egyptian wheel-made Cooking Ware (Fig. 22, K31, K32) is very rare. Only two examples of two-handled globular wide-mouth cooking pots with flat, inside slightly undercut rim K31 made of the local 'calcareous' clay (see above, p. 157) are represented in the material next to one single lid made of alluvial clay. All are of uncertain dating occurring in contexts with a terminus ante/ad quem ...
Context 38
... piece of a Roman lamp (Fig. 23, K41) of un- known, but non-Egyptian origin was found. The base carries incised letters [ARU(…)P], the "A" is made with a cross-stroke. The fabric is beige, silty and very fine. Only five coins were found at Abar el-Kanayis, four of them in the layers of drifted sands, where mixed ma- terials from Late Roman to recent times appear. Ac- ...
Context 39
... dates cer- tainly represent part of the vegetal diet of the travel- lers. A peach kernel (Fig. 24b) (Prunus persica) points to the existence of fruit tree gardens either in Siwa or in the coastal area. Whether it remained in Abar el-Kanayis as part of a batch or the provision of a trav- eller is unclear. The remains of barley (Hordeum vul gare, Fig. 13), cultivated on the Northern Tableland to an extent that allowed it to be sold 82 , however, were used as mudbrick temper (see above, p. 144) on the site. Whether one considers this context of the cereal MDAIK 68, 2012 to be also proof of its consumption in Abar el-Kanayis or even its export to Siwa, is debatable. A piece of palm tree ...
Similar publications
Citations
... The population in Marmarica has lived there since the Late Bronze Age up to present, partly sedentarily, partly nomadically. However, according to the findings, a peak in agricultural production in Graeco-Roman and Byzantine times allowed a surplus economy [64][65][66]. Hence, one marker for marginality, the economic potential of a region, is affected by human impact. ...
... People living here from the Bronze Age to Graeco-Roman times followed combined strategies of livestock breeding and agriculture, and were therefore prepared for both drought years and good years. However, landscape archaeological studies, pursuing a spatially large scale and diachronic approach, broadened perspectives to socio-spatial organization, economic surplus production, as well as to intra-and inter-regional connections that helped with understanding how the region functioned in antiquity in a way that was far from 'marginal' (Figure 10) [61,63,65,66]. ...
... However, fallows or the crops on the fields in drought years could be used as grazing areas, as was the case with the steppe zone south of the tableland. The yields and returns from agricultural production (barley, grapes, figs, and only little wheat) and livestock breeding amounted to a surplus, whereas marginal areas are normally considered to only allow the inhabitants a subsistence economy ( Figure 15) [62][63][64][65]96]. This surplus production was reconstructed mainly from the existence of numerous pottery production sites along the coast and on the tableland dated to between the second century BCE to the fifth century CE, and from the peak in the number of settlements ( Figure 16) [64,65] (p. ...
Arid regions in the Old World Dry Belt are assumed to be marginal regions, not only in ecological terms, but also economically and socially. Such views in geography, archaeology, and sociology are—despite the real limits of living in arid landscapes—partly influenced by derivates of Central Place Theory as developed for European medieval city-based economies. For other historical time periods and regions, this narrative inhibited socio-economic research with data-based and non-biased approaches. This paper aims, in two arid Graeco-Roman landscapes, to show how far approaches from landscape archaeology and social network analysis combined with the “small world phenomenon” can help to overcome a dichotomic view on core places and their areas, and understand settlement patterns and economic practices in a nuanced way. With Hauran in Southern Syria and Marmarica in NW-Egypt, I revise the concept of marginality, and look for qualitatively and spatially defined relationships between settlements, for both resource management and social organization. This ‘un-central’ perspective on arid landscapes provides insights on how arid regions functioned economically and socially due to a particular spatial concept and connection with their (scarce) resources, mainly water.
The Marmarica, an arid region in NW-Egypt between the Jebel el Akhdar to the west and the Nile Valley to the east, offers rich evidence for understanding the interlinkages of scarce natural resources, above all water and soil and their human utilization in antiquity. Analysing the natural hydrological regime depending on rainfall, soils and topography in the region and the man-made interventions lies at the heart of this landscape archaeological study. Integrating evidence from various disciplines (hydrology, geomorphology, soil science, archaeobiological methods, ceramic studies, evaluation of literary sources based on papyri) and from various periods (late 2nd millennium BCE to 7th century CE) allowed for the reconstruction of the ancient water management and the related life-strategies.
An assessment of climatic conditions and morphological features of the Eastern Marmarica is provided, where wadis, alluvial fans, but also lateral wadi slopes and to a certain extent even parts of the vast tableland plains represent favourable geomorphological units for water harvesting and hence, agricultural production. Yet, a characteristic that sets the Marmarican systems apart from those in other arid regions are cultivated terrace systems (run-in areas) that are located on lower lateral valley slopes and even on the tableland plains.
The results provide insights into the long-term responses of the inhabitants in antiquity to catchment hydrology by water harvesting and the effects of their interventions, the adapted livelihoods, regional specialization of rural production and intra- and interregional exchange of goods. However, new questions arose regarding i) the implications of the ancient water management strategy for the social organisation of the local people, ii) the destinations of locally produced pottery, iii) the habitational and land-use patterns before the Graeco-Roman period, iv) the role and range of climatic shifts, and v) reasons for the decline of the sophisticated runoff management.
Mobility, from a historical perspective, comprises a broad variety of movements of people. This
paper focuses on the mobility of multisited groups, based on the preconditions that an arid environment
imposes on mobility. The Eastern Marmarica-Plateau in NW-Egypt in Graeco-Roman times (5th century
BCE to 4th century CE) serves as case study of the various ways in which people in such landscapes
were mobile, and what we can infer from the aspect of mobility about their social practices. In order to
elucidate these issues, the discussion centres on archaeological-historical methodology and theoretical
implications of mobility and the wayfaring of inhabitants of arid lands. Moreover, the question is pursued
of how routes emerged and how habitual knowledge was cemented in order to establish the trails used
by generations to follow. The paper shows the interconnection and similarities between the mobility of
multisited communities and the seemingly so different mobility of crop-growing groups, according to
the exploitation, availability and exchange of resources and goods. As a general concern, the alleged
dichotomy of sedentary and nomadic mobility is challenged and replaced by a more open concept of
space and place. Mobility, the complexity of interactions and hence the routes themselves are not so
much shaped by fixed socio-cultural ascriptions of life-strategies
