SciELO - Scientific Electronic Library Online

vol.117 índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados



Links relacionados

  • Em processo de indexaçãoCitado por Google
  • Em processo de indexaçãoSimilares em Google



versão On-line ISSN 2305-445X
versão impressa ISSN 0254-1807

Scriptura vol.117  Stellenbosch  2018 



The 'upper regions' and the route of Paul's third journey from Apamea to Ephesus



Mark Wilson

Stellenbosch University




Luke's phrase ἀνωτερικὰ μέρη  in Acts 19:1 has been an interpretative conundrum for scholars of Acts for centuries. How Paul came to Ephesus at the start of the third journey is the geographical issue. The article begins with a brief lexical discussion. It then examines each of the proposed routes and their variants. Recent archaeological and hodological research in western Asia Minor as well as new cartographic productions provide fresh insights into these routes. How contemporary Bible atlases portray the route of the journey is then discussed. The methodological tools of Least Cost Path Analysis, Network Analysis, and 3D modelling are next employed to evaluate these routes. Based on these data, the Meander valley route from Apamea to Ephesus is preferred. The article concludes with several insights about Paul's travel and ministry programme in Acts derived from the study.

Key words: Paul's Third Journey; Ephesus; Apamea; Roman Roads; Meander River




Paul stopped briefly in Ephesus on the return portion of his second journey. Ephesus was probably the intended destination for this journey, but at its beginning the Holy Spirit prevented him from preaching in Asia (Acts 16:6).2 Now over two years later the apostle arrived in the capital of Asia to an already functioning church. There he left Priscilla and Aquila and visited the synagogue where he was invited to stay and teach (Acts 18:19-21a). However, Paul was intent on reaching Jerusalem for Pentecost, so he set sail for Judea. Luke's geographical concision in these verses is jarring: in two verses Paul travelled from Ephesus to Caesarea to Jerusalem to Antioch on the Orontes, a distance of approximately 1160 miles/1875 kilometres involving weeks of travel (Acts 18:21b-22). After spending some time in Antioch, he began the third journey. Tannehill suggests that this journey was a new mission in only a limited sense: "In the first two journeys the emphasis was on the founding of new churches. In 18:23 Paul begins a journey to strengthen established churches."3 However, Paul strengthened the Galatian churches at the start of the second journey as well (Acts 16:5), so this is the third time that the disciples in south Galatia are revisited for strengthening (cf. Acts 14:22). Since his destination was Ephesus,4 he took the Via Sebaste westward from Pisidian Antioch. West of Apollonia Paul took the right fork to continue on the Southern or Common Highway from the province of Galatia into the province of Asia.


Crossroad at Apamea

After entering Asia, the apostle arrived at Apamea (modern Dinar), which was situated at a major road junction.5 Apamea, formerly Persian Celenae, was a great emporium and conventus of the province as well as "hub of communication and exchange" between the Aegean region and the upper Anatolian plateau.6 As described by Dio Chrysostum (Cel. Phyrg. 15-17), the assizes held by the governor attracted area residents to Apamea for juridical, commercial, and entertainment reasons.

Despite its strategic location, Breytenbach and Zimmerman doubt whether Paul ever visited Apamea: "It is highly probable that local Jewish resistance to the Pauline mission on the Phrygian escarpment initially made it impossible to extend the Pauline sphere of influence from Lystra and Iconium westwards along the Via Sebaste beyond Antioch to Apollonia or along the main road running north of the Sultan and Karakus. Daglari down to Apamea and eventually to Eumenea."7 This 'obvious explanation' is faulty for several reasons. Paul experienced great opposition from Jewish communities in many cities where he preached but was able to return to them later for the resumption of his ministry. These include Lystra and Pisidian Antioch (Acts 14:21), Thessalonica (Acts 20:2), and Corinth (Acts 20:2). Therefore, Breytenbach and Zimmerman have Paul looping northward through Laodicea Combusta and Philomelium on his second and third journeys to avoid this ' resistance' around Apamea. Although no Jewish community is known from Philomelium (though it likely had one), a Jewish inscription has been identified at Laodicea Combusta, located just northwest of Iconium where Paul had been expelled from the synagogue.8 If local Jewish resistance was influencing Paul's choice of a route, surely travel through these cities much nearer to Iconium would been blocked as well. So Jewish resistance cannot be the reason Paul is prevented from travelling to Apamea.

Breytenbach also doubts that Paul visited Apamea because no evidence of Christianity has been found there or in its surrounding cities until the late second century CE. He attributes this to Paul's inability to extend his sphere of influence westward between Pisidian Antioch and the Lycus valley. Nevertheless, literary and epigraphical evidence, e.g., Montanism and Abercius, dating from 150-200 CE suggests that Christianity was already entrenched in the Phrygian cities around Apamea.9 Pliny the Younger (Ep. 10.96.9) attests to the presence of Christians in the cities, villages, and countryside of Bithynia and Pontus around 112 CE. Yet the earliest Christian inscriptions from the region date to the late third century CE, suggesting that the absence of Christian realia somewhere is not the same as the absence of Christian presence.10 It is therefore viable to postulate that Christianity was established in Apamea by the late first century CE. At Apamea Paul was literally at a crossroad for choosing his route to Ephesus.


Lexical Issues

Paul's passage to Ephesus is succinctly described in Acts 19:1: . The relevant phrase is with being a hapax legomenon in the New Testament. As Barrett notes, "The precise meaning of this phrase is uncertain. The adjective is rare, and is not used elsewhere as a geographical term."11 BDAG (s.v. ) defines the word as "the upper (i.e. inland) country, the interior."12 Its editors suggest a semantic parallel with the use of in the LXX readings of Judith 1:8, (Upper Galilee; NRSV, NETS), and of Judith 2:21, (upper Cilicia; NRSV, NETS).13 The cognate is used in a geographical sense only in 1 Kings 10:22 LXX: (Upper Beth Horon).14 LXX usage primarily signifies a higher elevation. The phrase is variously translated into English as 'interior' (NIV), 'inland country' (ESV), 'upper regions' (NKJV), and 'interior regions' (NRSV, NLt).15

The textual variant is not discussed by Metzger.16 However, Barrett comments: "The nature of the hinterland makes a suitable verb ... but does not make ... unsuitable. The compound verb may have been introduced in order to match "17Fifteen of the sixteen usages of in the New Testament are found in Luke-Acts.18 In nine cases the verb has the geographical sense of going from a higher to a lower place (e.g., Luke 4:31; 9:37), specifically going down from Jerusalem (e.g., Acts 8:5; 9:32; 11:27; 12:19; 15:1; 30; 21:10). Regarding the text of Acts 19:1, Apamea stood at 2887 feet or 880 meters above sea level (hereafter fasl/masl); Ephesus was situated just above sea level. So whichever route Paul used, he descended in elevation toward the coast. 19


The Inner Route through Lydia

Conybeare and Howson suggested that "it is safer to imagine him following some road further to the north," postulating that since Paul was never personally at Colossae (Col. 2:1), he would not have used the 'well-travelled' Southern Highway.20 They then projected an extreme northerly route that came from Tavium in eastern North Galatia before turning westward into upper Phrygia. It then passed near Thyatira and entered the Hermus valley at Sardis. They interpreted the phrase  as meaning the mountains of Asia and Phrygia that contained the headwaters of the Meander and Hermus Rivers.21 However, the headwaters of the Meander are much farther south near Apamea, while the upper Hermus flows past Cadi (modern Gediz) where Paul travelled along Mysia on his second journey (Acts 16:7). If Thyatira were a destination, he would have travelled northwest along the Simav River rather than cross the mountains west of Cadi before dropping south to Thyatira.22 From there, according to Conybeare and Howson, Paul continued southward past Sardis and followed some track across the Tmolus Mountains (Boz Daglar) before arriving in Ephesus. Their map (Fig. 1) of the journey fails to follow the known Roman road system and instead depicts a straight line from the interior to the Aegean coast. Because of the imprecise nature of this projected route, its length cannot be estimated. These able commentators have ignored Anatolian geography with this ill-conceived hypothesis.



A variant is Huttner's suggestion that "a northern route through Philomelium and Sardis would also have offered an appropriate connection with the capital of the province of Asia."23 However, this route from Iconium did not use the Via Sebaste and bypassed Pisidian Antioch, which Paul visited to strengthen the disciples (see previous discussion). It was also an unnecessary detour northward. Huttner claims that Ramsay suggested this northern route, but Ramsay described the northern route as running through Cappadocia and North Galatia,24 not through the Asian cities of Philomelium and Sardis. In fact, Ramsay repudiates this route and favours a southern route that revisits the churches started on Paul's first journey, including Pisidian Antioch.25


The Upper Route along the Pontic and Propontic Coasts

French has proposed an unusual interpretation of Acts 19:1, suggesting that it may "reflect a correction to an earlier omission during his second journey, namely his avoidance of coastal Bithynia (and the Mysian shore of the Propontis and the Aegean sea) but there can be no certainty." In a footnote French explains that since the phrase 'under Crete' in Acts 27:7 indicates the southern coast of Crete, "'upper parts,' therefore, may mean 'northern,' i.e. the Pontic and Propontid coasts."26 Such an itinerary along the modern Sea of Marmara would have taken Paul hundreds of miles/kilometres out of his way, particularly if by the Pontic coast, the Black Sea is meant. Also, there is no further evidence in Acts for such an itinerary nor is there any evidence in Paul's letters of church planting in these areas. About such a speculative proposal one thing is certain: Acts scholars and Bible atlases unanimously agree that this was not the route that Paul took.


The Upper Route through the Cayster Valley

Ramsay became a prominent advocate that the area of the upper Meander and upper Cayster River valleys is what Luke is describing as the 'higher districts.'27 He based his opinion on a geographic point that Phrygia had two sections: High and Low. Since Laodicea was the southwestern extent of Low Phrygia, this ruled out Paul's passage through there. Without disparaging Ramsay's pioneering scholarship, his successor Haspels has delineated the region called the Highlands of Phrygia as actually further north between the ancient cities of Dorylaeum (modern Eskis.ehir), Cotiaeum (modern Kutahya), Nacolea (modern Seyitgazi), and Prymnessus (near modern Afyonkarahisar). She writes that "it was not, however, this remote rock country that claimed most of his [Ramsay's] attention during his 'twelve years' wanderings,' but the lower plains of Phrygia, with the later, prosperous cities."28 So when reading Ramsay's description today that High Phrygia would be the ' higher districts,' he is actually discussing southwestern Phrygia.

The Barrington Atlas calls this highland area, Phrygia Epikteteis, and shows it considerably north of the area supposedly travelled by Paul.29 Mitchell identifies the upper Meander valley and Phrygia Paroreius as sites of the Seleucid settlements of Apamea, Pelta, Blaundos, Hierapolis, and Laodicea.30 So Ramsay's point is not arguable today; both itineraries were partially in lower Phrygia. Schnabel's statement that Paul could not have taken the southern route because it "did not run through Phrygia" is not correct,31 for Laodicea and Colossae were situated at the southwestern edge of Phrygia.32

Schnabel has presented the most articulate discussion regarding the upper Cayster route and provides a brief introduction to the cities "in which he [Paul] could have preached the gospel to people whom he encountered during overnight stays and rest stops."33 He suggests a route departing northwestward out of Apamea toward Eumenea (modern Isikli) 34 and Sebaste (modern Selcikler). At this road junction Schnabel projects Paul next travelling northeast through Acmonia (modern Ahat) and then westward to Temenouthyrai (modern Usak). Unless Paul specifically wanted to visit the Jewish community in Acmonia,35 this would have been a needless detour of over 62 miles/100 kilometres en route to Blaundos (near modern Sulumenli). 36 Least Cost Path Analysis (see below) suggests that a track existed between Sebaste and Blaundos, pointed to by the red arrows on the digital map of Asia Minor (Fig. 2).



This track, also suggested on the Barrington Atlas but not on Calder and Bean or Pauly,37passed north of the deep canyon of the Sindros River (modern Banaz Qayi).38 It provided access from the east to Pepouza and Tymion and intersected the north-south road (Fig. 3) that connected these cities with Temenouthryai.39



Blaundos was situated on an isolated high plateau. Filges writes that "the city was situated several kilometres away from the nearest major trade route. Thus it can be assumed that it was not a regular station for travelling traders."40 Its inaccessibility made it an improbable destination for a travelling apostle. For this leg ending in Blaundos, described by Schnabel as a 'high road,'41 it is unlikely that Paul would have taken it to Ephesus. Thonemann notes that the cities just mentioned are "on the main route towards Sardis from southern Phrygia."42 So the road's orientation was towards the valley of the Hermus River (modern Gediz) and not the valley of the Cayster River (modern Küçük Menderes). This would suggest an itinerary more in line with Monroy's suggestion discussed in the next section.

Schnabel next projects the road proceeding westward across the watershed of the Meander valley and into the valley of the Cogamus (), a branch of the Hermus, south of Philadelphia (modern Ala§ehir).43 Schnabel then suggests that a track crossed into the upper watershed of the Cayster and passed several ancient towns like Tarigya until reaching Koloe (modern Kiraz). The name of the modern Turkish village, Uluderbent ('high mountain pass'), suggests that it was a transit point on a long-used passage, today highway D310. The track continued on the southern side of the Tmolus range until Hypaipa (modern Gunluce), where there was an important road junction.44

The Barrington Atlas and the digital map of ' Asia Minor' show no route from the upper Cogamus valley to Hypaipa.45 Calder and Bean show a direct route from Koloe past Dioshieron (Birgi) to Hypaipa. However, the terrain dictated that this track ascends 820 feet/250 meters from Koloe before descending a tortuous path to Dioshieron. It is no wonder that both the main ancient and modern routes went southwest around a spur of the Tmolus past Neikaia (modern Türkönü) into the valley of the Cayster before turning northwest to Hypaipa (Fig. 4).46 Near modern Odemis. a branch road runs northeast to Birgi, which surely must have been the same situation in antiquity. French has an arrow pointing southwest from Philadelphia (the junction was actually south of the city) and labels this route D20 with a question mark. However, he does not place Hypaipa at the junction of D2 running from Sardis to Ephesus.47



Roads running from Hypaipa linked the cities in the upper Cayster valley with Smyrna as well as Sardis and Ephesus.48 A stadion-stone dating to the Hellenistic period documents a road between the latter two cities. Now in the Ephesus Museum, the marker measures 90 stadia from Ephesus and 410 stadia from Sardis. French suggests that the distance of 500 stadia may calculate the distance between the Artemis temples in each city and that the route constitutes a 'sacred' or 'King's' road between Ephesus and Sardis.49 Yet the fact remains this route still involved a significant climb over Mount Tmolus (Bozdag) reaching an elevation of over 3200 fasl/975 masl. A route looping through Sardis and crossing over the Tmolus was indeed possible, but it would add considerable time and distance to the journey, thus delaying Paul's arrival in Ephesus.

According to Altinoluk, "The road from Hypaipa to Ephesus follows the plain of Cayster, thus being easy to travel."50 The route followed the northern bank to Laris(s)a. After Larisa, Schnabel suggests that Metropolis was the next stop before Ephesus.51However, this track, depicted on the Barrington Atlas, brought the traveller north of Metropolis, so was the route to Smyrna, not to Ephesus. At Larisa the road turned south to Tyrrha/Thyaira (modern Tire) crossing the Cayster just north of that city. It continued on the left bank past the Seleucid tomb at Belevi before arriving in the territory of Ephesus. The route passed the temple of Artemis and entered Ephesus through the Artemis gate near the stadium.

Other reasons for Paul choosing this route have been suggested. Although conceding that the route from Apamea through Laodicea was 'more usual,' Skeel nonetheless suggests that Paul "might prefer the plain of Metropolis through Eumenia and down the Cayster valley."52 A difference in temperature has also been postulated. Keener writes: "Travel through the highlands would have avoided the scorching heat of a land journey around Ephesus in the summer."53 His source is Pliny the Younger (Ep. 10.17.1) who, however, did not make an inland journey. He arrived by sea at Ephesus (cf. Ep. 10.15) and described his travel by carriage along the Aegean coast in the heat of August. The higher elevation along the upper route does produce a variance of four to five degrees centigrade in the summer when Laodicea (Denizli) is compared to Tememouthryae (Us.ak). However, Paul was probably travelling in the spring when temperatures were more moderate. Night temperatures would be more the concern in April-May: those in Denizli range from 48-55° F/9-13°C while those in Us.ak range from 41-48° F/5-9°C.54 Summarizing the overall climatic situation, the upper Cayster route would be somewhat more bearable in the summer while the Meander route would be more comfortable in the spring. But the temperature difference is so minimal that it would hardly warrant a change in itinerary.

Fairchild has suggested that Paul might have taken the longer northern route because he wished "to avoid the same hindrance that prevented him from entering Asia on his second mission."55 However, the nature of that hindrance is not articulated. Acts 16:16 never states that Paul could not enter Asia, but rather that he was forbidden to preach there. As argued elsewhere, the prohibition by the Spirit likely occurred at the road junction east of Apamea after the apostolic party had already entered the province of Asia. Since the road junction for the upper route began at Apamea, Paul would now have been past that earlier place of 'hindrance.'56

A present fact demonstrating that the cities along the upper route through the Cayster valley were secondary is that no archaeological excavations are occurring in them. However, on the Meander route there are ongoing archaeological excavations at every ancient city except Colossae and Antioch ad Meandrum. Uggeri emphasizes the significance of this for the early church: "Le vie di penetrazione del Cristianesimo in Asia sono dunque le stesse del commercio, e di quello in mano ai giudei in particolare... Tali centri risultano dislocarsi, a partire da Efeso, lungo l'ampia e ricca vallata del Meandro, ossia Magnesia, Tralle e le tre citta contigue di Laodicea, Colossae e Hierapolis."57


The Upper Route through Smyrna

Monroy has proposed a variant by suggesting that Paul went to Ephesus either through the Cayster valley or by way of Philadelphia, Sardis, and Smyrna "perhaps for the purpose of visiting these important cities."58 The latter route had its terminus on the Aegean coast at Smyrna and was rebuilt by Manius Aquillius to provide access to Anatolia's interior.59Monroy cites the fourth-century Life of Polycarp as possible evidence for such a visit: it states that Paul stayed with Strataeas during his visit to Smyrna (Vit. Poly. 2). Grant cites this, contrarily, as an example of the ahistorical nature of the document, since Polycarp in his Letter to the Philippians (11.3) states that the church in Smyrna was not founded before the church in Philippi.60 This route is even more circumambulatory than the Cayster option, adding another 100 miles/160 kilometres. The later tradition that Paul established congregations at this time in several of the Seven Churches proves specious. Their establishment must wait until later during his ministry in Ephesus (Acts 19:10).

Even if Smyrna were the goal, taking the road from Apamea via Blaundos was unlikely. When Xerxes marched westward to Sardis in 480 BCE, his route took him from Apamea (Celenae) past Colossae before traversing the Meander river and then crossing into the Cogamus and Hermus valleys (Herodotus Hist. 7.26, 30-32). In 399 BCE Cyrus the Younger marched with his ten thousand in reverse, departing Sardis and then crossing into the Meander valley and visiting Colossae before heading northeast toward Apamea (Xenophon Anab. 1.2.5-6). In the early second century CE Ignatius, the bishop of Antioch, probably travelled this route via Laodicea and Sardis as a prisoner before stopping at Smyrna.61

A variant of this itinerary has been suggested by Uggeri. For Paul's second journey he posits a route coming from Apamea down the Lycus valley. At Laodicea he turned northwest to Hierapolis, Philadelphia, and Sardis before continuing northward to Troas. Regarding this stage route of the third journey, Uggeri writes: "Anche ora le tappe dovettero essere le stesse del secondo viaggio... Pero, giunto in Lidia, non aveva più motivo di deviare verso la Troade e la Macedonia e percio da Sardi prosegui diritto verso ovest per Efeso, seguendo il percorso dell'antica strada persiana."62 The Persian road that Uggeri speaks about was the Royal Road whose terminus was Sardis. If the route continuing west toward Smyrna was taken, that road forked west of Troketta with its left branch turning south to Metropolis through the Karabel Pass on the modern Kemalpaça-Torbali road. After Metropolis Paul would pass Mount Gallesion before arriving in Ephesus.63

At this stage of the third journey Paul's visitation of cities for sightseeing or ministry seems unlikely. So the roundabout itinerary through Sardis and/or Smyrna via Metropolis posited by Monroy and Uggeri is scarcely feasible. Despite the various arguments marshalled for the route along the upper Meander and Cayster valleys, Bérard's trenchant observation written almost a century ago remains valid: "Cet itinéraire, il faut l'avouer, serait pour le moins étonnant. La vallée du Caystre n'est pas une voie d'accès vers l'intérieur elle conduit à un cul-de-sac, et n'a jamais été, en conséquence, empruntée par les grandes routes antiques ou modernes. Pour rejoindre le haut bassin de Caystre depuis la région d'Apamée, il faut traverser un pays montagneux extrêmement accidenté.... Enfin, dans l'antiquité même, nous n'avons aucune reference tant soit peu précise sur un chemin direct traversant cette région."64


The Southern Highway to Ephesus

The shortest and easiest route from Apamea ran westward along the Southern Highway. Even during the Late Bronze Age (1200 BCE), this was the main route between Apamea and Ephesus and the only route depicted in Vaessen's map of the period.65 Strabo (14.2.29) states that "there is a kind of common road (  ) constantly used by all who travel from Ephesus towards the east." Travelled by his source Artemidorus, the route passed through Laodicea and Apamea. Kulzer emphasizes the importance of this highway "which led from Ephesus (Selcuk) through the valley of the Maeander and the inner parts of Phrygia up to the Euphrates, a connection already used in archaic times."66 In his classic volume on Asia Minor, Magie discusses the important cities founded by the Seleucids that were either in or near the basin of the Meander and its tributary, the Lycus. Only Eumenea to the northwest of Apamea was unconnected to this route.67 These cities with a Hellenistic foundation began at Apamea and ran westward to Tralles.

Approximately 40 miles/64 kilometres southwest of Apamea between the Sinaus Lacus (modern Acigol) and Colossae, the route joined the road that Manius Aquillius built in 129 BCE that connected Ephesus and Pergamum to Pamphylia via Laodicea. Presumably in this area Paul would begin to see bilingual milestones naming the governor as well as the distance to Ephesus, the road's caput viae.68 The road entered the Lycus valley above Colossae and then entered Laodicea through the Syrian Gate. It exited through the Ephesus Gate and crossed the Asopus river over a bridge whose piers are still standing.69 West of Laodicea the Lycus valley merged into the larger Meander valley, and the road proceeded past Antioch ad Meandrum, Nysa, and Tralles. At Magnesia ad Meandrum the road climbed over the small pass at Thorax before descending through the triple-arched Pollio Aqueduct and entering the upper city of Ephesus through the northeastern Magnesian Gate.70

After Asia was incorporated into the Roman Empire, its governors regularly travelled to the province's conventus cities to hold court sessions. Laodicea and Apamea were among these cities, so an imperial retinue regularly travelled from Ephesus eastward to conduct such assizes.71 In 51 BCE when Cicero arrived in Asia to serve as governor of Cilicia, he travelled from Ephesus through Laodicea to Apamea (Att. 5.13-5.16).72 This again demonstrates that this was the natural route between the cities. Regarding the Southern Highway as Paul's route to Ephesus, Mitchell writes that "there is no reason to look beyond the natural geographical interpretation of this journey"; from Apamea Paul continued "down the Maeander valley to the west coast."73


The Third Journey and Bible Atlases

The view of Ramsay and later scholars-that the upper Cayster route best accords with Luke's phrase -has been articulated for over a century. Yet no Bible atlases have adopted it. In a review of ten major Bible atlases, only three might suggest this as the route Paul used to travel to Ephesus.74 The Kregel Atlas shows a broad line vaguely tracing a path coming from the northeast above the Meander valley.75Carta's New Century Atlas shows Paul's route north of the Meander River, thus suggesting an inner road (Fig. 5).76 Since the lines do not follow any known roads, the maps reflect only the imagination of the cartographers.



The Oxford Atlas depicts a puzzling combination: the track stays north of the road from Apamea to Colossae, but west of Laodicea it dips into the Meander valley and proceeds westward to Ephesus. 77 Since the route cuts across mountains and not along known Roman roads, the projected route again looks like a cartographer's construct. Schnabel's map is not helpful either since it depicts a route directly from Apamea to Acmonia without passing through Eumenea. It also fails to mark either Temenouthryae or Blaundos, and shows no route through the upper Cayster valley into Ephesus.78 Although the verdict of the atlases surveyed, such as the Zondervan Atlas (Fig. 6), decisively favours the Meander route, their general lack of cartographic accuracy depicting Paul's routes in Asia Minor remains problematic.79



Least Cost Path Analysis

While texts and maps contribute to answering the question about Paul's route, new methodologies such as network analysis can provide more objective data. Historians of ancient trade and economy have recently been analyzing transportation routes and how they developed for largely economic reasons. However, their research has implications for travel in a broader sense. Herzog identifies several aims for archaeology offered by Optimal Path or Least Cost Path Analysis (LCPA):80 1) reconstructs ancient routes and road networks,81 2) identifies the main factors governing construction of known road or segments,82 and 3) identifies locations that occupy a central position in a territory. LCPA assumes that people optimize the costs of frequently-travelled routes, which leads to the real-world equivalent of a least cost path. Such analysis is most frequently generated using GIS software. Once optimal routes are generated, their reliability must then be compared to archaeological evidence such as ancient road sections, bridges, and milestones. An example of LCPA at work is the projected road between Sebaste and Blaundos, discussed earlier. Similarly, the suggested track in Figure 4 between the junction south of Philadelphia and the upper Cayster valley near Koloe avoids the much longer and equally difficult climb over the Tmolus at Sardis.

Itineraries may change due to weather conditions as well as social and religious factors. Since Paul's travel pattern suggests frequent visitation of Jewish communities, would one route be more conducive for making such religious connections? As discussed above, Acmonia had a Jewish community in the first century CE; however, the city was located off the projected inland route. Unless Paul had a specific reason to detour there, this would not be a factor for choosing the Upper Cayster route. Because the inland route passed through a rural area with smaller cities, the likelihood of encountering synagogues was more remote. In contrast, Jewish communities were well-established in the cities along the southern route such as in Laodicea, Tralles, and Magnesia ad Sipylum.

A second type of network analysis is route analysis, defined by ArcMap as "finding the quickest, shortest, or even the most scenic route, depending on the impedance you choose to solve."83 LCPA favors cost as the impedance. If time or distance is the impedance, then the best route is the quickest and shortest. Related to Paul's journey, these are primary concerns. Route analysis can be done using various route planners available on the web or as phone apps. Strava Route Builder ( is designed especially for runners or cyclists. When Dinar (Apamea) and Selcuk (Ephesus) are entered into Strava, only one route is displayed: the southern route down the Meander valley with a distance of 287 kilometres (178 mi.). Google Maps ( is another popular program for route building. When a route from Dinar to Selcuk is requested in Google Maps, the Meander valley option is generated as the shortest at 295 kilometres (183 mi.). To display the inland route via the Cayster valley, a route from Dinar to Selcuk had to be constructed manually, using modern village names that lie along the approximate ancient route (Fig. 7). This option measured 386 kilometres (240 mi.), an overestimation generated because of the difficulty of plotting this route. (For this the Barrington Atlas measured 194 miles/312 kilometres using a map wheel.)







A modern traveller would clearly choose the first option which follows the lower Meander valley. Nevertheless, there are limitations for using these forms of network analysis. Herzog warns that "GIS-based reconstructing prehistoric routes without validating the results on the basis of archaeological evidence or historic sources is mere guesswork."84 However, the autopsy of these routes by the author along with the examination of relevant archaeological realia and texts have hopefully overcome such limitations.85


3D Modelling of the Cayster and Meander Routes

Two-dimensional examples of route building, however, fail to account for elevation differences between the routes. Whereas distance is not changed, speed and ease of travel are clearly affected. Various software is now available to produce 3D route models. The two routes under discussion are now shown using 3D modelling in Geocontext, and the data drawn from these maps are then presented.86

The following chart first provides the data drawn from the 3D route modelling in Geocontext; it then provides the distances calculated in 2D from the Barrington Atlas.



The Cayster route is preferred by some commentators because of its presumed shorter distance. Keener writes that "this would be a more direct route through the hills."88 But as the chart shows, this route at 320.9 kilometres is 18 percent longer than the Meander route at 272.2 kilometres. The route from Sebaste to the Cogamus valley crossed a tableland more rugged than that from Apamea to Colossae. And the Cayster has a narrower valley and meanders more than its namesake. Finally, the climb from the Cogamus valley (610 fasl/185 masl) to cross the pass at Tarigya (2871 fasl/875 masl) is much steeper than that from Magnesia (105 fasl/32 masl) over the Thorax pass (912 fasl/278 masl).

An average daily distance for walking, both today and in antiquity, is approximately thirty kilometres (29.6 km; 18.4 mi). Calculating travel times on the 2D maps of the Barrington Atlas estimates 10.4 days for the Cayster route and 9.3 days via the Meander route. Factoring in ascent from 3D modelling (Cayster 3824 m. vs. Meander 1769 m.) adds time to both itineraries; however, the Cayster route gains more adding 1.4 days. The total time for the Cayster route is now 11.9 days while the Meander route takes 9.7 days. When elevation is factored, the Cayster route takes 2.2 days longer to walk than the Meander route.89 Topographical realities clearly disadvantage any perceived pluses of the Cayster route, and 3D modelling prefers the Meander route.


Concluding Remarks

Delving into an obscure point of biblical geography may appear an arcane exercise, yet it yields some important insights about Paul's travel and ministry programme in Acts. On the second journey the divine impulse to "preach the word" in Asia, particularly its capital Ephesus, remained unresolved. Yet an open door had presented itself in the Ephesian synagogue that Paul must have interpreted as divine guidance to return.90 As in Antioch on the Orontes, Paul was not the founding apostle of the church in Ephesus. Nevertheless, the city's significance as a governmental, commercial, religious, and transportation hub must have beckoned him. As a church strategist he now moves his base of operations westward from Antioch, 91 the third largest city in the empire, to Ephesus, the fourth largest city. Both cities shared numerous demographic advantages, including a large Jewish population. The resumption of Paul's third journey in 19:1, following an episodic window into events at Ephesus during his absence (Acts 18:24-28), indicates that the capital of Asia is his primary focus now. Tannehill emphasizes this point: "Ephesus is not just another stop in a series. It is Paul's last major place of new mission work; indeed, it is the sole centre of mission noted in the last stage of Paul's work as a free man."92 The concise description of Paul's pan-Anatolian journey suggests a compulsion to return as quickly as possible (cf. Acts 20:22).93After his arrival Paul immediately instructs some disciples of John before accepting the invitation, extended months before, to teach in the synagogue (Acts 19:1b-8). To postulate a longer route between Apamea and Ephesus runs counter to the sense of urgency in Luke's narrative at this point.

Spencer importantly points out that the travel texts in Acts "continue to portray the Pauline mission as a team effort."94 Luke seemingly suggests that Paul was travelling alone from Antioch to Ephesus; however, the possibility that he was accompanied by companions should not be discounted. So far in Acts the only journey wherein Paul travelled alone was when the Jerusalem believers sent him back to Tarsus (Acts 9:30). Luke at times fails to list Paul's fellow travellers. For the famine relief journey from Antioch to Jerusalem only Barnabas is named, while Paul includes Titus in his own description of the trip (Acts 11:30; Gal 2:1).95 At the end of Paul's time in Ephesus two of Paul's travelling companions ( ; Acts 19:29), Gaius and Aristarchus, were dragged into the theatre during the riot. Later Gaius is said to be from Derbe while Aristarchus was from Thessalonica (Acts 20:4; cf. 27:2).96 Titus is never mentioned in Acts yet named ten times in 2 Corinthians. He was sent to Corinth from Ephesus as Paul's emissary along with an unnamed brother (2 Cor. 12:18), and after leaving Ephesus the apostle looked for Titus at Troas (2 Cor. 2:1213). After finally meeting him in Macedonia, Titus and his unnamed travelling companion () were sent ahead to Corinth (2 Cor. 8:19). It is possible that Titus and Gaius travelled with Paul to Ephesus from Syria and Lycaonia respectively. If several travel companions were accompanying Paul, the longer and more difficult Cayster route would be less expedient than travelling directly to his destination.

Paul mentions in Colossians 2:1 that he had never personally met the believers in the Lycus valley:  .97 As suggested earlier, there is nothing conclusive in Paul's statement to preclude him from having traversed these cities previously. Dunn construes this verse so, suggesting that Paul "may have passed through Colossae earlier," but that some local believers presently in the church had not met him then.98 Harris argues contrarily that  introduces the general class of persons to which the Colossians and Laodicean belong: those who have not met Paul personally. He continues, "Paul is emphasizing that his struggle was for Christians personally unknown to him yet spiritually dear to him."99 As Thompson notes about Colossae and Laodicea, "the Christian congregations in both cities owe their existence to the efforts of coworkers in Paul's mission."100 Only during Paul's ministry at Ephesus did Epaphras become a believer and return to the Lycus valley, establishing a congregation at Hierapolis also (Col. 4:13).101Since no believers existed when he passed through initially, no one could know him personally. Because Paul had passed through the Lycus valley on this journey, he had a firsthand knowledge of the physical landscape that his audience inhabited. Although he had not seen their face, he had seen their place: the snow-capped Mount Cadmus that loomed over Colossae like a crown and the white travertine cliffs that gleamed north of Laodicea and provided a ledge upon which Hierapolis rested. These images would come to mind as he later penned his letter to the three house churches there. This existential acquaintance with the Lycus valley seems to underscore the anticipation in his closing request to Philemon: "Prepare a guest room for me, because I hope to be restored to you i n answer to your prayers" (Philm. 1:22).

Can a conclusion be reached regarding Paul's route to Ephesus through the "upper regions?" Walker is noncommittal, stating, "We cannot tell whether he travelled down the Lycus/Meander valley - passing places such as Colossae and Laodicea - or came along the north side of Mount Messogis."102 The majority of Bible atlases, as we have seen, prefer the southern Meander route from Apamea to Ephesus. The longstanding suggestion that Paul detoured off the Southern Highway to travel along a minor route has been demonstrated to be geographically and topographically problematic. The cumulative evidence from ancient texts, archaeological realia, network analysis, and 3D modelling overwhelmingly prefers the direct Meander route for Paul's travel to Ephesus from Apamea. Unless a significant reason existed to deviate, none of which is known, the Southern Highway would be the natural route for Paul's third journey.



1 I wish to thank the anonymous reviewers whose comments helped to improve the article significantly. Any errors that remain are my own.
2 For the probability that Ephesus was his intended destination, see Glen L Thompson and Mark Wilson, "The Route of Paul's Second Journey in Asia Minor: In the Steps of Robert Jewett and Beyond," Tyndale Bulletin 67.2, 2016:225.
3 Robert C Tannehill, The Narrative Unity of Luke-Acts, vol. 2. Minneapolis: Fortress, 1990:231.
4 The most direct route from Antioch on the Orontes to Ephesus was approximately 1000 miles/1600 kilometres. William J Larkin Jr., Acts. Downers Grove, IL: InterVarsity, 1994:269, overestimates this distance by 50 percent, suggesting it was 1500 miles/2414 kilometres.
5 It was probably near Apamea that the Holy Spirit's prohibition forced the apostolic party to turn northward to Bithynia; see Thompson and Wilson, "The Route of Paul's Second Journey in Asia Minor," 227-28.
6 Peter Thonemami, TheMaeander Valley. Cambridge: Cambridge University Press, 2011:133; cf. 99-109. For a local guidebook see Mehmet Özalp, From the Province of Marsyas, Gelenia, Apameia Cibotis, Geykler, Dinar. Dinar: Renksel, 2010.
7 Cilliers Breytenbach and Christiane Zimmerman, Early Christianity in Lycaonia and Adjacent Areas: From Paul to Amphilochius of Iconium. Leiden: Brill, 2018:70.
8 Walter Ameling, Inscriptiones Judaicae Orientis: Kleinasien, Band II. Tübingen, Mohr Siebeck, 2004:488-90 nr. 227.
9 Cilliers Breytenbach, "What Happened to the Galatian Christians? Paul's Legacy in Southern Galatia," Acta Theologica 19, 2014:10-11. For Montanism and Abercius, see William Tabbernee, "Asia Minor and Cyprus," in Early Christianity in Context, ed. William Tabbernee. Grand Rapids: Eerdmans, 2014:268-73; for Pliny and Bithynia-Pontus see idem 302-306. If hypotheses 3-4, 5-4, 5-6, and 5-8 advanced by Rodney Stark, Cities of God. New York: HarperOne, 200:81-82, 132-36, are correct, and they are likely so, Apamea would be a certain candidate for a Christian community in the late first century CE.
10 As Longenecker states similarly about those who deny the existence of robust evidence for a Christian presence in Pompeii before 79 CE, "the absence of evidence is not evidence of absence"; Bruce W. Longenecker, "The Empress, the Goddess, and the Earthquake," in Early Christianity in Pompeian Light: People, Texts, Situations, ed. Bruce. Longenecker. Minneapolis: Fortress, 2016: 62.
11 CK Barrett, Acts 15-28. London: T&T Clark, 1998:892. He further suggests that the statement of Herodotus (Hist. 1.177) about Harpagus pillaging lower Asia (^ev vuv xorw Tfj; Acrfy];) while Cyrus did so to the upper country (Ta 5s avw auTfj;) probably functioned similarly to Luke's statement. Lower Asia for Herodotus was the coastal areas of Ionia, Caria, and Lycia (Hist. 164-176); however, the upper country is ill defined. Previously Cyrus was leaving Lydia for Persia (Hist. 157.1) and is next said to be fighting the Assyrians (Hist. 178.1). The use of nsipo; seemingly refers to continental Asia inland from the sea. Herodotus thus provides little help for understanding Luke's description here.
12 It is worth noting that the Hittites also used the term 'Upper Land' (malum elitum) to designate a region in central Anatolia. As Gurney notes, the term is "based on altitude as seen from their capital, Hattusa"; OR Gurney, "The Upper Land, malum elitum," in In Hittite Studies in Honor of Harry A Hoffner, Jr.: On the Occasion of His 65th Birthday, ed. Gary Beckman, Richard Beal, and Gregory McMahon. Winona Lake: Eisenbrauns, 2003:120.
13 Other parallels in the Deuterocanonicals are sv Tfj avw raAiAaia (Tob. 1:2), avw si; Ta ^spy Aiyunrou (Tob. 8:3), and si; tou; avw T6nou; (2 Macc. 9:23).
14 It is alternatively rendered in the LXX as Bai6wpwv t]]v avw (Josh. 16:5; 1 Chron. 7:24; 2 Chron. 8:5) or t]]v avw Bai6wpwv (Josh. 21:22).
15 For comparison the Afrikaans versions read 'boonste landstreke' (1953 Vertaling), 'die binneland' (1983 Vertaling), and 'deur ander plekke in die provinsie Asie' (Die Bybel vir Almal).
16 Bruce M Metzger, A Textual Commentary on the Greek New Testament, 3rd ed. New York: United Bible Societies, 1975.
17 Barrett, Acts 15-28, 893. In the ellipses Barrett gives the manuscript support for each reading which is fully presented in Nestle-Aland 28. 'Hinterland' in discussions of ancient Greek cities refers not to an amorphous interior region but rather to its relationship to a city, as in n6Ai; and xwpa; see Mogens Herman Hansen, "The Polis as an Urban Centre: The Literary and Epigraphical Evidence," in The Polis as an Urban Centre and as a Political Community, ed. Mogens Herman Hansen. Copenhagen: Royal Academy of Danish Arts and Letters, 1997:17-20.
18 Louw and Nida (s.v. 15.107) place xaTsp^o^ai in the same semantic domain as KaxaPaivco, whose translation in Acts 16:8 involves a similar interpretative issue; see Thompson and Wilson, "Route of Paul's Second Journey," 237-38, 238 n. 104.
19 The reverse movement of ascending or going up is seen in Xenophon's title Anabasis with Sardis as the point of departure and inner Anatolia as the region to be traversed.
20 WJ Conybeare and JS Howson, The Life and Epistles of St. Paul, 2 vols. London: Longmans & Roberts, 1856:2:5.
21 Conybeare and Howson, Life and Epistles, 2:6, also n. 2.
22 Thompson and Wilson, "Route of Paul's Second Journey," 235-37.
23 Ulrich Huttner, Early Christianity in the Lycus Valley. Leiden: Brill, 2013:83.
24 Howard Clark Kee, To Every Nation under Heaven: The Acts of the Apostles. Harrisburg, PA: Trinity Press International, 1997:226, seems to hold this view, suggesting that the upper regions "were on the lofty plateau of central Asia Minor"; cf. Gerd Liidemann, Early Christianity according to the Traditions in Acts: A Commentary (Minneapolis: Fortress, 1989:210, also holds that the interior of Asia Minor is what is meant.
25 Huttner, Early Christianity, 83 n. 17; WM Ramsay, The Church in the Roman Empire. London: Hodder & Stoughton, 1893:91-94. Ramsay (93) makes a significant observation about this initial stage: "Why should the narrator, who in other cases describes St. Paul's route with accuracy, leave it entirely doubtful whether he took the northern or the southern route? The reason is that the northern route never occurred to him as a possibility. The route from Syria by the Cilician Gates to the Aegean coast was a familiar and much frequented one; and unless another route was expressly mentioned, everyone would understand that Paul passed through Lycaonia, and not through North Galatia."
26 David H French, "Acts and the Roman Roads of Asia Minor," in The Book of Acts in its Graeco-Roman Setting, ed. David WJ Gill and Conrad Gempf, vol. 2 of The Book of Acts in Its First Century Setting, ed. Bruce W Winter. Grand Rapids: Eerdmans, 1994:55. James B Pritchard and Nick Page, eds. HarperCollins Atlas of Bible History. New York: HarperCollins, 2008:167, depict this Propontid itinerary for Paul's second journey. However, this route is in Bithynia, which Paul was forbidden to enter by the Spirit of Jesus (Acts 16:7).
27 Ramsay first articulated this in The Church in the Roman Empire, 94, and later more briefly in St. Paul Traveller and Roman Citizen. London: Hodder & Stoughton, 1895:265. This opinion is followed by Colin J Hemer, The Book of Acts in the Setting of Hellenistic History. Winona Lake: Eisenbrauns, 1990:120, 187; FF Bruce, Acts of the Apostles, 3rd ed. Grand Rapids: Eerdmans, 1990:405; Rainer Reisner, Paul's Early Period: Chronology, Mission Strategy, Theology. Grand Rapids: Eerdmans, 1998:286; John McRay, Paul: His Life and Teaching. Grand Rapids: Baker, 2007:177-78, David Peterson, The Acts of the Apostles. Grand Rapids: Eerdmans, 2009:528 n. 5, and Eckhard J Schnabel, Acts. Grand Rapids: Zondervan, 2012:783.
28 CM Emilie Haspels, The Highlands ofPhrygia: Sites and Monuments, 2 vols. Princeton: Princeton University Press, 1971:1:11. She states in the Preface (vvii) that she is completing Ramsay's two-volume Cities and Bishoprics of Phrygia by writing the third volume on the Highlands of Phrygia.
29 Richard Talbert, Barrington Atlas of the Greek and Roman World. Princeton, Princeton University Press, 2000:61-62.
30 Stephen Mitchell, Anatolia, 2 vols. Oxford: Oxford University Press, 1993:1:20; for a map of western Phrygia and Lydia, see 1:190.
31 Eckhard Schnabel, Early Christian Mission: Paul & the Early Church, 2 vols. Downers Grove: IVP Academic, 2004:2:1200.
32 Celal Sims.ek, Laodikeia (Laodikeia adLycum), rev. ed. (Istanbul: Ege, 2013), 35. As head of the excavations at Laodicea, Sim§ek writes, "Antik cografyada Lykos Vadisi, Frigya Bölgesi'nin en bati ucunda bulunur" ("In ancient geography, the Lycus valley is found at the most western margin of the Phrygian region.")
33 Schnabel, Early Christian Mission, 2:1200. The cities along this route are described on pp. 1200-1203. Witherington, following FF Bruce, more equivocally states that "Paul may have taken a higher road farther north, leaving the main road at Apamea"; Ben Witherington III, The Acts of the Apostles. Grand Rapids: Eerdmans, 1998:70.
34 Eumenea later became an important Christian centre; see Tabbernee, "Asia Minor and Cyprus," 270-71.
35 See Paul Trebilco, Jewish Communities in Asia Minor. Cambridge: Cambridge University Press, 1991:58-84, 214-21.
36 The primary road from Sebaste ran northwest to Temenouthyrai before dropping southwest to Blaundos. Following this route was again a detour of approximately 47 miles/75 kilometres. Distances have been calculated on the Barrington Atlas using a Scalex Map Wheel and conform with the measurements given on Orbis for distances in this area ( They also approximate the distances given by Schnabel in his chart (2:1198).
37 Talbert, Barrington Atlas, 62; WH Calder and GE Bean, "Classical Map of Asia Minor". Ankara: BIAA, 1958; A-M Wittke, E Olshausen, and R Szydlak, eds., New Pauly Historical Atlas of the Ancient World. Leiden: Brill, 2010:177
38 "Asia Minor in the Second Century CE," Ancient World Mapping Centre (; accessed 7/20/2017).
39 William Tabbernee and Peter Lampe, Pepouza and Tymion. Berlin: de Gruyter, 2008:105-6; 165-67. Two maps show this road and the Roman bridge across the Sindros: 104 fig. 5.11 and 105 fig. 5.12. For a map of the larger region see 2-3 fig. 1.1.
40 Axel Filges, "Buildings and Citizens: Observations from Late Antique and Byzantine Blaundos in Phrygia," in Archaeology and the Cities of Asia Minor in Late Antiquity, ed. O Dally and C Ratte. Ann Arbor: Kelsey Museum of Archaeology, 2011:136. Filges also writes: "There must have been a junction at this major road with a one-way street leading to and from Blaundos" (personal correspondence 1/9/2016); see also Falko von Saldern, "Das Territorium von Blaundos," in Blaundos. Berichte zur Erforschung einer Kleinstadt im Lydisch-Phrygischen, ed. Axel Filges. Tübingen: Wasmuth, 2006:14-16.
41 Schnabel, Early Christian Mission, 2:1200. Elevations are definitely higher along this road. Eumenea sits at 2822 fasl/860 masl while Blaundos is at 2231 fasl/680 masl; conversely, Colossae sits at 1188 fasl/362 masl while Laodicea is at 951 fasl/290 masl.
42 Thonemann, Maeander Valley, 173.
43 Schnabel's geography gets fuzzy here: "At this point, near mod. Gullu, one needs to assume a road running toward the southwest across the eastern Tmolus Mountinas (mod. Bozdaglari), perhaps along the line from Gullu to A§agice§me, Bozalan, Sarigol, Uluderbent and Akpinar. The BAGRW maps neither document nor suggest such a route, which, however, does not prove that such a route did not exist" (Early Christian Mission, 2:1200). First, the Tmolus Mountains do not extend east of the Cogamus valley (see Fig. 2). Of the places mentioned, only Uluderbent and Akpinar are situated in the Tmolus. Second, the Barrington Atlas does show a route on map 62 from Blaundos connecting just south of Philadelphia. However, it is farther north than that projected by Schnabel.
44 The connection between Hypaipa and Sardis is epitomized by the fragmentary inscription found in the Sardis synagogue mentioning a Jew named Samuel, also called Julianos, who was a citizen and councilor of Hypaipa who erected the panel in fulfillment of a vow; see Jack Kroll, "The Greek Inscriptions of the Sardis Synagogue," Harvard Theological Review 94.1, 2001:33 no. 34; 89 fig. 36.
45 Talbert, Barrington Atlas, 56.
46 Veli Sevin, Necla Arslan Sevin, and Sevda Cetin, Neikaia Unutulmus birAntikKent. Odemis,: Odemis. Museum, 2013:30 fig. 10.
47 David H French, Roads and Milestones of Asia Minor, vol. 3 Milestones, Fasc. 3.5 Asia. Ankara: BIAA, 2014:25, Conspectus map 5.1.1 Asia West.
48 For recent investigations of these roads, see Hatice Kalkan, "New Observations on the Road of Ephesus-Hypaepa-Sardeis and Persian Royal Road," International Journal of Human Sciences, 11.1, 2014:456.
49 David H French, "Pre- and Early- Roman Roads of Asia Minor: A Hellenistic Stadion-stone of Ephesus," Arkeoloji Dergisi 5. 1997: 91-95, map 190. One stadion is calculated at 609.006 feet/185.625 meters.
50 S Altinoluk, Hypaipa, A Lydian City during the Roman Imperial Period. Istanbul: Ege, 2013:20. A map on page 251 depicts the city territories of the Cayster valley.
51 Schnabel, Early Christian Mission, 2:1203.
52 Caroline AJ Skeel, Travel in the First Century after Christ with Special Reference to Asia Minor. Cambridge: Cambridge University Press, 1901:137.
53 Craig S Keener, Acts, 4 vols. Grand Rapids: Baker, 2013:3:2815 n. 5153. Regarding the heat, Keener adds that "Paul was probably by now inured to it, in any case."
54 The temperature information for Denizli and U§ak was derived from the weather section of the Wikipedia articles on these cities.
55 Mark Fairchild, "Laodicea's 'Lukewarm' Legacy: Conflicts of Prosperity in an Ancient Christian City," Biblical Archaeology Review 43.2, 2017:33.
56 Thompson and Wilson, "Route of Paul's Second Journey," 227-28.
57 "The ways of penetration of Christianity in Asia are therefore the same as those of commerce, and of those in the hands of Jews in particular... These centres are positioned, starting from Ephesus, along the wide and rich valley of the Meander, namely Magnesia, Tralles, and the three adjoining cities of Laodicea, Colossae, and Hierapolis": Uggeri, "Sulle strade di San Paolo in Anatolia," 159. Uggeri seemingly ignored his own words when he formulated the route of the third journey; see below.
58 Mauricio Saavedra Monroy, The Church of Smyrna: History and Theology of a Primitive Christian Community. Frankfurt am Main: Peter Lang, 2015:162; cf. 243. Again this interpretation is driven by the reading of Colossians 2:1 that "the men of the valley of Laodicea and Colossae did not see Paul personally."
59 David Magie, Roman Rule in Asia Minor, 2 vols. Princeton: Princeton University Press, 1950:1:157.
60 Robert M Grant, "Polycarp of Smyrna." Anglican Theological Review 28, 1946:137.         [ Links ]
61 William R Schoedel, Ignatius of Antioch. Philadelphia: Fortress, 1985:11; Michael W Holmes, ed. The Apostolic Fathers, 3rd ed. Grand Rapids: Baker, 1999:128, map 611.
62 "Even now the stages had to be the same as the second voyage... However, when he arrived in Lydia, he no longer had a reason to deviate to Troas and Macedonia, and therefore from Sardis he went straight west to Ephesus, following the path of the ancient Persian road": Giovanni Uggeri, "Sulle strade di San Paolo in Anatolia: Il secondo e il terzo viaggio," in Seminario di studi Paolo di Tarso: il messaggio, Vimmagine, i viaggi: studi in memoria di Luigi Padovese, ed. Stella Uggeri Patitucci and Luigi Padovese.(Palermo: Officina di studi medievali, 2011:134; cf. 137.
63 Paradoxically the map of the third journey in Uggeri, "Sulle strade di San Paolo," 163 fig. 34, fails to show this route but instead shows it descending through the Meander valley to Ephesus.
64 "This itinerary, it must be admitted, would be surprising to say the least. The Cayster valley is not an inland gateway but leads to a cul-de-sac, and has never been, as a result, travelled by ancient or modern highways. To reach the high basin of Cayster from the region of Apamea, you have to cross an extremely rugged mountainous country... Finally, even in antiquity, we have no precise reference to a direct path through this region": Jean Bérard, "Recherches sur les Itinéraires du Saint Paul en Asie Mineure." Revue Archéologique 6.5, 1935:85.
65 Rik Vaessen, "Cosmopolitanism, Communality and the Appropriation of Mycenaean Pottery," Anatolian Studies 66, 2016:49 fig. 3.
66 Andreas Kiilzer, "The Network of Communication Routes in Anatolia in Late Antiquity and Medieval Times," 7 (paper presented June 15, 2015 at Atalim University, Ankara, Turkey; _in_Anatolia_in_Late_Antiq uity_and_Medieval_Times; accessed 7/27/2017).
67 Magie, Roman Rule in Asia Minor, 1:125-29. Magie carefully delineates the stages of this route, called by Ramsay the "Eastern Highway," in a detailed discussion in 2.789-91 n. 18.
68 This route is shown as E1 on French, Roads and Milestones, Asia, 26 map 5.1.2. Two milestones marking this route have been discovered; see David H French, Roads and Milestones of Asia Minor, vol. 3 Milestones, Fasc. 3.1 Republican. Ankara: BIAA, 2012:36-38.
69 Simsek, Laodikeia, 83-85, 89-91, 95-97.
70 Thonemann, The Maeander Valley, 103-104 n. 11, 279.
71 Christian Marek, In the Land of a Thousand Gods: A History ofAsia Minor in the Ancient World. Princeton University Press, Princeton, 2016:261. However, Marek includes Sardis among the cities along this main road.
72 Cohen uses Cicero to make this same point: "Under the Republic and Early Empire, the Common Road and its western terminus, Ephesos, remained a vital link for communication with Rome"; Getzel M Cohen, The Hellenistic Settlements in Europe, the Islands, and Asia Minor. Berkeley: University of California Press, 1995:58.
73 Mitchell, Anatolia, 2:4.
74 The other seven are: Yohanan Aharoni and Michael Avi-Yonah, The Macmillan Bible Atlas, 3rd ed. New York: Macmillan, 1993, 184 map 252; Paul Lawrence, The IVP Atlas of Bible History. Downers Grove: IVP Academic, 2006:161; Pritchard and Page, HarperCollins Atlas ofBible History, 167; Barry J Beitzel, The New Moody Atlas of the Bible. Chicago: Moody, 2009:261 map 112; John D Currid and David P Barrett, Crossway ESV Bible Atlas. Wheaton: Crossway, 2010:247; Carl Rasmussen, Zondervan Atlas of the Bible. Grand Rapids: Zondervan, 2010:231; John A Beck, Discovery House Bible Atlas. Grand Rapids: Discovery House, 2015:303 map 11.7.
75 Tim Dowley, The Kregel Bible Atlas. Grand Rapids: Kregel, 2002:85.
76 Anson F Rainey and R Steven Notley, Carta's New Century Handbook and Atlas of the Bible. Jerusalem: Carta, 2007:246.
77 Adrian Curtis, Oxford Bible Atlas, 4th ed. Oxford: Oxford University Press, 2009:169.
78 Schnabel, Early Christian Mission, 2:1604, fig. 18. Also confusing is that his two maps of Asia Minor are mislabelled: fig. 18, Asia Minor I is really "Cities and Roads," while fig. 31, Asia Minor II is actually "Provinces and Territories" (2:1617).
79 This lack of cartographic veracity is discussed in Thompson and Wilson, "Route of Paul's Second Journey in Asia Minor," 219, 221.
80 Irmela Herzog, "The Potential and Limits of Optimal Path Analysis," in: Computational Approaches to Archaeological Spaces, ed. Andrew Bevan and Mark Lake. London: Left Coast Press, 2012:1.
81 An example is Philip Verhagen and Karen Jeneson, "A Roman Puzzle: Trying to Find the Via Belgica Using GIS," in Thinking Beyond the Tool: Archaeological Computing and the Interpretive Process, ed. Angeliki Chrysanthi, Patricia Murrieta Flores, and Constantinos Papadopoulos. Oxford: BAR, 2012:123-30.
82 Barbora Weissova and Peter Pavúk, "On Persistency of the Main Communication Routes from Prehistory until Today," Turkish Academy of Sciences Journal of Archaeology 19, 2016:1-21.
83 (accessed March 3, 2018).
84 Herzog, "Potential and Limits of Optimal Path Analysis," 20.
85 LCPA was used to calculate two optional routes on Paul's second journey also; see Thompson and Wilson, "Route of Paul's Second Journey," 234 n. 84. I wish to thank Levent Oral and Cenk Eronat of Tutku Tours, Izmir, Turkey, for the opportunity to visit Dinar and other sites along the upper route in August 2016.
86 Such 3D modelling was presented by Tonnes Bekker-Nielsen in the presentation "Ancient Roads in the Third Dimension" at a conference called Roads and Routes in Anatolia: Pathways of Communication from Prehistory to Seljuk Times sponsored by the British Institute in Ankara in March 2014. I thank Prof Bekker-Nielsen for his patience in teaching me about this methodology.
87 The formula used in the chart is: (d + [L| * 7.92] = T); "Naismith's Rule," Wikipedia, (accessed 2 July 2017
88 Keener, Acts, 3:2815, apparently echoing the opinion of Ramsay (Church in the Roman Empire, 94), states that Paul "preferred the shorter hill road practicable for foot passengers but not for wheeled traffic."
89 For a more detailed discussion of the methodology behind these calculations, see this author's "Paul's Journeys in 3D: The Apostle as Professional Traveller," Journal of Early Christian History 8.1, 2018 na.
90 Mark Wilson, The Spirit Said Go: Lessons in Guidance from Paul's Journeys. Eugene: Wipf & Stock, 2017:70-72.
91 Tannehill, Narrative Unity of Luke-Acts, 230, makes this significant observation: "Paul is not a loner, founding a separate, Pauline church, but a major figure in the one mission that began in Jerusalem and was effectively continued from Antioch."
92 Tannehill, Narrative Unity of Luke-Acts, 231.
93 Wilson, The Spirit Said Go, 91-93.
94 F Scott Spencer, Journeying through Acts: A Literary-Cultural Reading. Peabody, MA: Hendrickson, 2004:192.
95 This of course assumes that the two journeys are identical, which I do. For more discussion on this issue see Ben Witherington III, Grace in Galatia: A Commentary on Paul's Letter to the Galatians. Grand Rapids: Eerdmans, 1998:13-20.
96 The plural MaKeôóvaç suggests that both men were Macedonians; however, the Alexandrian text here is singular (MaKeôóva) suggesting that only Aristarchus was Macedonian. For more on the possibility of dittography here, see FF Bruce, The Acts of the Apostles: Greek Text with Introduction and Commentary, 3rd ed. Grand Rapids: Eerdmans, 1990:418.
97 "For I want you to know how much I am contending for you and those in Laodicea: everyone who has not seen me in person" (author's translation).
98 James Dunn, The Epistles to the Colossians and to Philemon. Grand Rapids: Eerdmans, 1996:129.
99 Murray Harris, Colossians & Philemon. Grand Rapids: Eerdmans, 1991:79
100 Marianne Meye Thompson, Colossians & Philemon. Grand Rapids: Eerdmans, 2005:47.
101 John Polhill, Acts (Nashville: B&H, 1992), 398, suggests that the "most natural route" passed through the Lycus valley and then concludes from Col. 1:7 that the churches there were established by Epaphras, probably during Paul's Ephesian ministry.
102 Peter Walker, In the Steps of Paul. Grand Rapids: Zondervan, 2008:135-136. Messogis is the mountain range separating the Meander and Cayster valleys. The map on page 84 shows a major road north of Messogis coming from the east; however, it deviates significantly from the routes presented in this article.

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons