You are here

Biblical Archaeology Review 46:2, Spring 2020

Searching for Bethsaida: The Case for El-Araj

By R. Steven NotleyMordechai Aviam

The archaeological site of El-Araj is the best candidate for the biblical town of Bethsaida. This is how we know.

For 38 years (4 B.C.E.–34 C.E.), Herod Philip, the son of Herod the Great, ruled as governor of the northern Transjordan, a region stretching from Mount Hermon (on the modern Lebanon-Syria border) to the territory east of the Sea of Galilee. In about 30 C.E., he founded a city at the southern boundary of his tetrarchy on the shores of the Sea of Galilee. This new city complemented the administrative capital that he had established in the northern portion of his district at Caesarea Philippi-Paneas three decades earlier, in 3 B.C.E, near where his father had previously constructed a temple to Caesar Augustus (Josephus, Jewish Antiquities 15.363). For his new, southern polis, the governor did not start from scratch. This son of Herod the Great selected the existing Jewish village of Bethsaida on the lakeshore and transformed it into his new city. He named it Julias, after the widow of Caesar Augustus and the queen mother of Caesar Tiberius.

Most of these details have escaped the notice of Christian readers, because the New Testament mentions only the name of the Jewish village and not of the polis that replaced it. We are uncertain of the reason for this omission, because elsewhere the New Testament writers do report cities founded under the Herodian dynasty—for example, Tiberias (John 6:23), Caesarea Philippi (Matthew 16:13; Mark 8:27), and Antipatris (Acts 23:31).

Moreover, the city’s very location has remained a mystery until now. One of the challenges for modern-day archaeological investigations is to identify the ancient sites being excavated. Finding an inscription that bears the name of the place is rare. More often, identification involves consideration of different types of evidence that can resemble clues in a mystery novel.

The rediscovery of the lost cities of the Bible has been a slow process that began in earnest in the middle of the 19th century, when advances in transportation made these distant lands more accessible to European and American travelers. Edward Robinson, a biblical scholar from Union Theological Seminary in New York City, was among the first to attempt to locate these forgotten places. He traveled the region by horseback in the 1830s and 1850s, accompanied by Eli Smith, an expert in Semitic languages. Together, they discovered that the Hebrew place names from long ago were often preserved in their modern Arabic equivalents.

Still today a number of locations known from ancient sources elude our identification. Bethsaida-Julias is one of them. Robinson theorized that the New Testament village was situated at et-Tell, the site on a small hill at the edge of the Bethsaida valley, 1.5 miles from the current lakeshore, excavated since 1987 by Rami Arav of the University of Nebraska.a

The et-Tell excavators claim that theirs is the site of ancient Bethsaida-Julias. Yet even in Robinson’s day, not everyone embraced this proposal. Gottlieb Schumacher, an American civil engineer and architect who had resettled in Haifa, argued that et-Tell is too far from the lakeshore to be identified as a fishing village. The site of el-Araj, with its close proximity to the lake and its numerous surface finds, seemed much more likely to Schumacher.

The ancient historical witnesses to Bethsaida-Julias provide ample details regarding the general location of its settlement, a description of its nature, and the early history of its development. Bethsaida is first mentioned in the New Testament (Matthew 11:21; Mark 8:22), as are the sites of Capernaum, Chorazin, and Nazareth. These towns were likely founded as part of an increased settlement in the north during the Hasmonean Dynasty (142–76 B.C.E.).

Both the New Testament and Josephus situate Bethsaida on the lakeshore. Jesus is reported to have traveled there by boat (Mark 6:45). Similarly, Jewish reinforcements sailed to Bethsaida-Julias from Taricheae (Magdala) during the early days of the First Jewish Revolt Against Rome (Life 406). Josephus describes the village of Bethsaida “next to the lake” (Antiquities 18.28), and the Gospel of John states that it was the home of fishermen (i.e., Peter, Andrew, and Philip) who were numbered among Jesus’s followers (John 1:44). Finally, early rabbinic sources speak frequently about the fishing industry at Bethsaida (e.g., Jerusalem Talmud tractate Sheqalim 6, 50a).

One of the more detailed geographical descriptions of the delta surrounding Bethsaida-Julias is Josephus’s story (Life 398–406) of a battle near Julias between the Jewish forces that he led and the army of King Agrippa II. The topographical details Josephus provides about the location of Bethsaida-Julias with regard to its proximity to the opposing troops, the Jordan River, and the Sea of Galilee, are important clues in the search for Bethsaida-Julias.

We learn that King Agrippa II stationed troops under the command of Sulla, the captain of his bodyguards, east of the Jordan River about half a mile (500 stadia) from Julias on the inland road leading eastward to Seleucia and Gamala (Life 398). The aim was to cut off these cities in the lower Golan from the centers of the revolt in Galilee. Josephus responded by bringing 5,000 Jewish soldiers to the western side of the river about 500 feet (100 stadia) from Julias. The historian’s measurements indicate that the opposing forces were not too far distant from each other and were primarily separated by the Jordan River.

Josephus attempted to lure Agrippa’s mercenary forces into an ambush by leading a small unit across the river. When the Jewish forces feigned retreat, Sulla and his men pursued them. In the ensuing fight, Josephus fell with his horse on the muddy terrain and was injured. His men evacuated him to Capernaum and then to Taricheae. The next morning, Jewish reinforcements reportedly sailed from Taricheae to Julias.

This last remark indicates that Bethsaida-Julias was not the site of fighting and that the city must have been located on the lakeshore for the reinforcements to arrive by boat.1 Josephus is our sole source of information about Philip’s transformation of the Jewish fishing village of Bethsaida into a small Roman polis. The historian’s only mention of the toponym Bethsaida, the name of which he reports was changed to Julias, appears in Jewish Antiquities:

And to the village of Bethsaida [located] next to the lake of Gennesar, he granted the dignity of a city by [introducing] a multitude of inhabitants and other fortifications, and he called it Julias after the name of the daughter of the emperor.

—(Antiquities 18:28)

Otherwise, Josephus always refers to Bethsaida by its Hellenized name, Julias. In spite of the attribution, many scholars today assume that the historian has confused Julia, the daughter of Augustus, with Livia, the emperor’s wife, whose name was changed to Julia Augusta in 14 C.E., when she was adopted into the imperial gens (clan). Julia Augusta died in 29 C.E., and it may be that the founding of Bethsaida-Julias was intended by Philip to commemorate the death of the mother of Caesar Tiberius. Philip may have followed the example of his brother Herod Antipas, who renamed Betharamphtha, a city in the Transjordan to “Julias after the name of the emperor’s wife” (Antiquities 18.27).

Josephus also informs us that Philip died at Julias and was carried from there in a funerary procession to his mausoleum (Antiquities 18:108). Josephus does not inform us of the location of Philip’s tomb, and its whereabouts remain unknown.

Bethsaida-Julias continued to exist during the Roman period, when it is attested by two classical authors: Pliny the Elder (77 C.E.; Natural History 5.15) and Claudius Ptolemy (150 C.E.; Geography 5.15.3). To these should be added a few references in early rabbinic literature, where the Semitic name Ṣaidan (Hebrew: ציידן) is used.2

Bethsaida was not a city familiar to the rabbis after the third century C.E. Around 305 C.E., Bishop Eusebius of Caesarea reports the following about Bethsaida in his catalog of most of the places mentioned in the Hebrew Bible and the Gospels: “The city of Andrew and Peter and Philip. It is located in the Galilee next to the lake of Gennesar” (Onomasticon 58.11). Eusebius draws his information from the Gospel of John (John 1:44; John 12:21) and Josephus’s description (Antiquities 18:28). Since he merely repeats details about Bethsaida found in well-known first-century sources and supplies no additional physical description, it seems that by the end of the third century C.E., the hometown of the apostles had been abandoned.

As we shall see, the silence regarding Bethsaida is no accident. In this instance, the absence of evidence may indeed be the evidence of absence. The literary silence corresponds uncannily with a lack in material remains at el-Araj for almost two centuries at the beginning of the Byzantine period.

The next mention of Bethsaida is by Theodosius, a Byzantine pilgrim to the Holy Land in 530 C.E. He succinctly describes his journey from Tiberias to Paneas (modern Banias):

From Tiberias to Magdala, where Saint Mary was born, is two miles. From Magdala to the Seven Fountains (Tabgha), where the Lord Christ baptized the apostles, and where also he fed the people with five loaves and two fish, is two miles. From the Seven Fountains to Capernaum is two miles. From Capernaum it is six miles to Bethsaida, where were born the apostles Peter, Andrew, Philip, and the sons of Zebedee. From Bethsaida to Paneas is 50 miles. There the Jordan emerges from two sources, the Jor and the Dan.3

The importance of this testimony is that a Christian pilgrim was able to speak about a site identified as Bethsaida. Apparently, its location was no longer unknown. It does not necessarily mean that Bethsaida of the Byzantine period corresponds with the earlier Roman city. Examples of Byzantine misidentification are not uncommon. Nevertheless, Christians in the Byzantine period—after about two centuries—began once again to identify a location called Bethsaida.

Theodosius does not include in his itinerary any mention of churches or shrines, only references to New Testament events. In addition, the direction and stops on his itinerary seem not to be haphazard or unique to Theodosius. We read the same journey two centuries later in the visit by Willibald, the bishop of Eichstätt in Bavaria:

From [Tiberias] they went around the sea, and by the village of Magdala to the village of Capernaum, where our Lord raised the prince’s daughter. Here was a house and a great wall, which the people of the place told them was the residence of Zebadee with his sons John and James. And [from Capernaum] they went to Bethsaida, from which came Peter and Andrew. There is now a church, where previously was their house. They remained there one night, and the next morning went to Chorazin, where our Lord healed the demoniacs, and sent the devil into a heard of swine. Here was a church of the Christians. Having performed their devotions there, they went to the place where the two fountains, Jor and Dan, issue from the earth, and flowing down from the mountain are collected into one and form the Jordan.4

Willibald’s reference to Chorazin has led many scholars to assume that he was confused about which places he visited. Since it makes little sense to cross the Jordan River to reach Bethsaida, cross it again to visit Chorazin, and cross it yet a third time to journey north, scholars have assumed that Willibald confused Bethsaida with Capernaum. Accordingly, the Byzantine church in Willibald’s description is sometimes “corrected” and identified with the Byzantine church at Capernaum.

Yet the bishop’s itinerary is not mistaken; he has correctly ordered the places visited, but has mistakenly applied the name Chorazin to the site of Kursi (Latin: Chorsia) with its Byzantine church and monastery on the eastern shore of the lake. That Willibald meant Chorsia (Kursi) and not Chorazin is evident from his description that it was there, where “our Lord healed the demoniacs, and sent the devil into a herd of swine.” The Byzantine Christian tradition of Jesus’s exorcism recorded in the Gospels (Luke 8:26-39) is identified at Kursi and never Chorazin (i.e., Khirbet Karazeh).5 Rightly understood, Willibald provides another Byzantine pilgrimage itinerary around the lake, not unlike the one of Theodosius, but his also includes details about a church built over the house of Peter and Andrew at Bethsaida.

Moving finally to the archaeological evidence, the historical and geographical picture of Roman Bethsaida-Julias that is gleaned from the literary witnesses does not fit the results of more than 30 years of excavations at et-Tell.6 This inspired us to investigate el-Araj, a stronger contender for Bethsaida.

In 2014, a shovel survey (digging and sifting the soil from several 5-by-5-meter squares to a depth of 1 ft) was organized at el-Araj to create a profile of the site and its settlement through datable pottery. We found evidence for settlement in the Roman, Byzantine, Islamic, and Crusader periods—precisely the periods Bethsaida-Julias appears in the literary sources.

In 2016, we launched the El-Araj Excavation Project. With a limited budget, the scale of the efforts was initially small.b Many excavations involve 100 or more volunteers for six weeks. Our team was fewer than 20, working for two weeks.

We initially focused on the area adjacent to the ruins of the “Bek’s House,” a lakeside villa built in the 19th century by Abdul Rahim Bek, a wealthy landlord who owned the entire Beteiha Valley and part of the Golan. A visitor to the home in 1929 reported seeing a colorful mosaic unearthed near the house. Our aim was to uncover this mosaic and determine whether it belonged to the Byzantine period church or an earlier, Roman period structure.

In the upper stratum of our excavation, we found remains of sugar production from the Crusader period. The residents had mostly reused still-standing Byzantine walls. In the 2017 season, we continued to excavate what we now identify as a Byzantine monastery that accompanied a church. This is a common combination in Galilee, similar to what is found at Byzantine Kursi.

Although we had not yet identified the walls of the church, its existence was unmistakably signaled by the discovery of many gold-gilded glass tesserae, which are found only in wall mosaics of ornate churches.

We decided to dig two probes to see if there were Roman remains under the Byzantine floors. Beneath the Byzantine pavement, dated by numerous coins to the fifth century C.E. and probably pointing to the establishment of the monastery, we encountered about 3 feet of different layers of silt from the Jordan River. There were no archaeological objects found in this stratum. Instead, it seems that the lack of material remains coincides with the two centuries of silence in the literary sources, during which the site was abandoned.

Below the alluvial soil, we immediately encountered a compacted layer with Roman pottery and coins that lacked any Byzantine objects. We even found portions of a mosaic floor at 692 feet below sea level—a full 10 feet below what has been generally assumed to be the level of the lake in the first century C.E. Our excavation results have not only challenged previous assumptions about settlement on the lakeshore in the New Testament period but have also given a reason to rethink the level of the lake—the lowest freshwater lake on earth—in Roman antiquity.

The mosaic floor was composed of white and black tesserae in a meander design similar to the floor of the synagogue recently excavated at Magdala, as well as the Early Roman bathhouse at Magdala.c We assume that our mosaic belonged to a Roman-styled bath, which is indicated by accompanying ceramic bricks and vents (tubuli), colored plaster, marble, color stone tiles, and roof tiles. In another probe, a coin of Nero from 63 C.E. dates the stratum to the early Roman period.

The Roman bath is the first evidence of urbanization found in the region. It is not a common feature in a Jewish village. However, we have already seen from the report of Josephus that Herod Philip transformed the village of Bethsaida into a Roman polis and renamed it Julias. The bathhouse aptly reflects what we would expect from Herod Philip’s urbanization of Bethsaida-Julias.

The new finds encouraged the team to enlarge the excavation. In 2018, we excavated an outlying area about 160 feet northeast of the main excavation area. In the new squares, we found no Crusader and little Byzantine settlement. Instead, we quickly encountered Roman walls, pottery, and coins. In addition, this area produced several “discus” lamps, Kfar Hananya-type pottery, and plaster from red Pompeiian frescoed walls.

In 2019, we further expanded our main excavation area with the aim of finally locating the walls of the Byzantine church. We found two more decorated stone fragments from the church, a marble fragment from the chancel screen that is decorated with a wreath, and a limestone fragment of a decorative table with a cross and a floral decoration.

And, after three seasons of excavation, we have finally identified the Byzantine church! It has a southern aisle with a colored mosaic floor, decorated with a geometric design and stretching to a length of 50 feet. In the final days of the season, we excavated the edge of a colorful mosaic frame from the floor of the central nave. The three-stranded braid design in the mosaic resembles those found in other Byzantine churches. It seems that the church will measure no less than 100 by 66 feet.

Some have questioned the significance of what has been called the Church of the Apostles. By itself, the Byzantine church should not be considered evidence for the location of first-century Bethsaida. However, coupled with the extensive and growing archaeological evidence from the earlier Roman period at el-Araj, the church does take on increased importance. We believe that the Christian community had not forgotten the location for the hometown of the apostles Peter, Andrew, and Philip when they reestablished a Christian presence at the site of el-Araj in the fifth century C.E.

As for the church itself, many doubted it existed at all. Several scholars told us that we had misread the pilgrimage reports. They thought Willibald, the Bavarian bishop, was confused when he reported that he visited a church at Bethsaida in 725 C.E., built over the house of Peter and Andrew. Instead, they contended he meant Capernaum and the octagonal Byzantine church there. Those assumptions will now need to be rethought.

The paved courtyards and rooms around the church at el-Araj indicate there was a monastery on the site. We found many coins on the different floors of the suggested monastery, most of which date to the late sixth and seventh centuries C.E. up to the eighth century. In two squares, we probed below the Byzantine pavement and found walls and remains from the Roman period with pottery and coins dating from the first to third centuries C.E. Among these discoveries is also a limestone mold for casting decorated lead fishing weights. This and the many lead weights unearthed in each season indicate the presence of a significant fishing industry at el-Araj.

In the 2019 season, we also opened a new excavation area, about 330 feet north of the main excavation to see the extent of the settlement. Already on the surface, we found Roman period pottery and coins. About 1 foot below the surface were the tops of walls. Some were preserved to 6.5 feet high! We identified two floor levels. The upper one, made of flat stones, dates to the third century C.E. Below it was a plaster (crushed lime) floor which yielded first- to second-century C.E. pottery. Among the coins found in this area, 16 date to the first century C.E. (including three of which might date to the first century B.C.E.), 12 date to the second century C.E., three to the third century, and one to the fourth century.7

We also found fragments of stone (chalk) vessels and sherds of oil lamps, both commonly used among Jews in the first century. Below this floor was a fill of soil that yielded first-century B.C.E. pottery and, surprisingly, also Early Bronze and perhaps Iron Age pottery.

These finds add to the mounting evidence of Jewish life at el-Araj. It probably was the site of a Jewish village that was transformed into a city in the Roman period, precisely as is reported in the New Testament and early Jewish sources. It certainly was not 13 feet under the lake in the Roman period, as the et-Tell excavators have previously contended.

The preliminary results from an electromagnetic imaging survey show that there are domestic buildings and even a possible public building waiting to be excavated in the 3 acres surveyed. If el-Araj was settled in the New Testament period, it lay on the lakeshore between the Sea of Galilee and et-Tell. Hence, it is the more likely location for a first-century fishing village, just as all the literary sources describe Bethsaida-Julias to be.

Next year, we intend to excavate the church entirely with the hope of finding its inscription, a routine feature in similar churches. We also plan to expand our excavation to other areas on the surrounding plain to look for the material evidence for el-Araj in the Roman period. Each season has strengthened the argument that el-Araj is the location for Bethsaida-Julias, and nothing has been found to challenge this identification.8

Although no one has declared the search for Bethsaida-Julias to be concluded, the mounting evidence has made el-Araj the leading candidate for Bethsaida-Julias, the lost city of the apostles.