Capernaum

The place a visitor means by Capernaum is the Franciscan enclosure: the gate, the white synagogue, the excavated street and the memorial church. It is a sanctuary of the Custody of the Holy Land, listed by the Custody's own Christian Information Centre as 'Capernaum: House of Peter and the Eucharistic Promise', with a 10 shekel entrance fee and a friary on site. It is not a national park, no Israeli parks pass covers it, and it shuts for two hours in the middle of the day.

East of it, on the shore, stand the red domes and white walls of the Greek Orthodox Church of the Holy Apostles, built in 1931 on the Patriarchate's share of the site. The Greek Orthodox land was excavated separately between 1978 and 1987 by Vassilios Tzaferis, whose team uncovered a seawall some 200 metres long with two piers running out into the lake, and a hoard of 282 gold dinars from the Arab and Byzantine transition. The monastery's own site says visits are arranged in advance by telephone with the superior rather than by turning up.

Then there is Kfar Nahum (Capernaum) National Park, which is a real Israel Nature and Parks Authority site and a different place. The authority describes its main area as about 80 dunams, or 20 acres, northeast of the antiquities, prepared to complement a visit to the ruins rather than to replace it, with a floating dock that rises and falls with the lake so boats can come in from Tiberias and Ein Gev. Its published hours run to 16:00 and its telephone number is not the friary's.

• Franciscan enclosure: the synagogue, the village street and the memorial church, 10 ILS, closed 12:00 to 14:00
• Greek Orthodox Church of the Holy Apostles: built 1931, east of the antiquities, visits by prior telephone arrangement
• Kfar Nahum (Capernaum) National Park: about 80 dunams northeast of the ruins, run by the parks authority, with the boat dock

The synagogue is the reason the site was found at all: Edward Robinson noted its ruins in 1838 without connecting them to the Gospels, and Charles William Wilson, who made a small excavation inside it in 1866, was the first to identify Tell Hum as Capernaum. The prayer hall measures 23 by 17.28 metres, is floored in white limestone slabs and was divided by sixteen columns on three sides. The stone is not local. Blocks weighing up to four tonnes were brought from quarries several kilometres away, in a village that built everything else out of the black basalt lying under its feet.

The date is genuinely contested, and the contest is worth understanding before you look at the carving. Britannica still gives the second or third century CE, with an older synagogue possibly buried beneath. The Franciscan excavations overturned that from below. The Custody's own account says more than 20,000 coins were found under the floor, along with pottery, and puts completion in the last quarter of the fifth century. The Franciscan guidebook quoted by Vatican News reports more than 30,000 coins on the pavement, dated to 383 to 395, and reads the same evidence as a late fourth-century building. Jodi Magness has argued the construction may be later still, in the sixth century or after. Four positions, one floor. The difference decides whether this is a building of the Mishnaic age or one raised while the octagonal church a minute away already stood.

What nobody disputes is that this is not the synagogue of Mark 1. Under the central nave the excavators found a wide basalt cobble pavement of the first century, too large for a private house, and around the outside of the standing building run formidable walls of squared basalt on a different alignment, clearest at the southwest corner. Corbo thought those walls belonged to the same first-century building as the pavement. Loffreda thought they were an intermediate stage between it and the fifth-century synagogue. The Custody prints the disagreement rather than resolving it.

The carving repays a slow circuit. Two inscribed columns stand at the front, one in Greek recording that 'Herod (the son) of Monimos and his son Justus together with their children erected this column', the other placed by the Department of Antiquities in memory of Fr Gaudenzio Orfali. A separate Aramaic inscription names Alphaeus son of Zebedee son of John. Among the cornice medallions is a block carved with a small temple on a cart, an ancient image of the Ark being carried to Jerusalem. Almost every figurative motif has been systematically chiselled away, which the Custody reads as the mark of an iconoclastic movement that the community here took part in.

Thirty metres south of the synagogue, under the foundations of a Byzantine octagonal church, Corbo's team reopened the ground in 1968 and worked downwards through four phases. A house of the late Hellenistic period. Then, from the end of the first century, one of its rooms plastered floor to ceiling and given over to gatherings rather than to living, its cooking pots replaced by storage jars and lamps. Then, in the fourth century, that room enlarged into a domus ecclesiae and walled off from the rest of the village. Then, in the second half of the fifth century, the whole block levelled and an octagonal church built with its inner octagon directly over the room, floored with a mosaic peacock.

The case for calling it Peter's house is cumulative rather than documentary. Fr Emmanuele Testa analysed 454 fragments of painted plaster from the fourth-century phase; the parks authority's summary counts 173 fragments carrying inscriptions, mostly Greek, with others in Syriac, Aramaic and Latin. They include the name and monogram of Jesus, liturgical phrases such as Amen and Kyrie Eleison, and, the excavators said, references to Peter. Around 380 the pilgrim Egeria was shown a house of the prince of the apostles that had been made into a church, its walls still standing. The Biblical Archaeology Society, reviewing the same evidence, notes there is no definitive proof, calls it layer upon layer of circumstantial evidence, and adds that many scholars now dispute the readings that find Peter's name in the graffiti.

The modern building over all this is a deliberate piece of engineering rather than a shelter. Designed by the Italian architect Ildo Avetta, worked out with the engineer Cesare Pocci and the Technion in Haifa, and built by Solel Boneh under the supervision of the Nazareth engineer Anis Sruji, its main volume was conceived as a ship's hull hovering above the apostle's house. It was consecrated by Cardinal Lourdusamy on 29 June 1990, a date cut into the facade as BEATO PETRO APOSTOLO A.D. MCMXC DICATUM. Visitors see the remains twice: from a path at ancient street level that passes underneath, and from above, through an opening in the floor of the memorial. Corbo died in 1991 and is buried at Capernaum beside the venerated room.

The facade that photographs so well is a reconstruction, and the Custody says so plainly: the synagogue has been partly rebuilt by Franciscan architects using the original blocks that lay scattered over the site. The work began with Orfali, who excavated in 1921 and 1926, published a monograph on the synagogue and started its restoration; the memorial column raised by the Department of Antiquities stands to him because he died in 1926 and the work stopped with him.

The reassembly is honest about itself if you know where to look. Segments of the great arch and tympanum that once crowned the front were laid out on the ground in the area behind the synagogue rather than lifted back into place. The sculpted lintels from the doorways of the prayer hall and the courtyard were replaced in situ. Loose architectural fragments were set out around the site so that visitors could see them, a decision Corbo took alongside the digging, and Fr Eugenio Alliata later catalogued the entire stone inventory.

The same distinction applies at the octagon. The mosaics were lifted in 1968 to allow deeper excavation and to preserve them, so what lies under the memorial church is a stripped stratigraphy rather than a Byzantine floor. Nothing at Capernaum stands above head height as it was found. The village walls run knee to waist high, the synagogue is a facade and a colonnade, and the newest thing on the site is a concrete hull from 1990.

Settlement here began in the second century BC. The village lived off the lake and off the traffic on a branch of the Via Maris that ran to Damascus, which is why the Gospels put both a customs post and a Roman detachment in it. A milestone bearing the name of the emperor Hadrian, who reigned from 117 to 138, was found in 1975 about 100 metres northwest of the synagogue, confirming the imperial road. About 110 metres of the main street, made of small stones and gravel and sloping down to the shore so that it drained into the lake, has been cleared. Houses of undressed basalt set in mud open off it in blocks, several families to a compound, courtyards with bread ovens, and basalt millstones and olive and grape presses left where they were found.

The excavated coins and pottery describe a village that traded with Upper Galilee, the Golan, Syria, Asia Minor, Cyprus, Africa and the Phoenician coast, and, as Loffreda found striking, barely at all with central and southern Palestine. Between the fourth and seventh centuries a mostly Christian community and a smaller Jewish one lived side by side here, which was unusual in a region where communities tended to occupy separate villages: crosses scratched on pottery turn up across the Franciscan sector, and the synagogue was in use at the same time.

The ending is dated differently by different accounts, and the disagreement is not resolved. The Custody's history has the village losing importance from the seventh century, under the Arab period, and permanently abandoned in the thirteenth, with the last fishermen gone no later than the fourteenth. Wikipedia's summary of the archaeological literature gives abandonment in the eleventh century, before the First Crusade. Vatican News, following Loffreda, says the village was abandoned about a thousand years ago, though some Arab families of the Semekiyeh tribe remained on the land until the 1948 war. What all of them agree on is that the site was never built over again, which is the accident of history that leaves a first-century street plan legible at all.