To determine when the phenomenon of coin deposits in ancient synagogues started, it is also important to know the construction dates of the (different phases of the) buildings. However, unlike churches of the Byzantine period, ancient synagogues rarely have dated dedicatory inscriptions. Thus, these buildings must be dated using other means. In the past, synagogue buildings often were dated based on their art-historical and architectural styles. “Galilean” type synagogues, especially the one at Capernaum, were originally dated to the 2nd and 3rd centuries, based on stylistic parallels with temples in Syria and Asia Minor. In 1971 Gideon Foerster wrote:
“The late second and third century A.D. dating is founded on architectural and stylistic parallels in contemporary Roman art and architecture in Syria and Asia Minor. The synagogue in Capernaum is in harmony with the classical architectural concept that stresses the outer appearance of a building. In contrast, Byzantine architecture concentrates on the interior (e.g. the lavish mosaic pavement of the synagogue of Hamath-Tiberias, which dates to the first half of the fourth century A.D.). Though in some remote areas in Syria and Asia Minor classical concepts of the exteriors of buildings seem to continue even in the Byzantine period, the architectural details are non-classical. At Capernaum, however, the capitals, friezes, cornices and other architectural details belong to types well-established in the late second and third centuries A.D.”
Scholars claimed that the architecture and decoration of the Capernaum synagogue are so different from those at Hammath Tiberias and Beth Alpha (dated to the 4th to 6th centuries) that they could not possibly have been constructed at the same time. They also believed that historical considerations supported these dates. Foerster writes:
“In the second century, the Jewish authorities together with a large number of Jews left Judea and settled in Galilee after two wars against the Romans. The prosperous condition of Jewish communities, as a result of their political, economic, and, not least, spiritual strength was the proper background for unusual building activity, of which the Capernaum synagogue can serve as one example.”
Scholars like Jodi Magness, however, have rebutted these theories over the last decades, pointing out that the building style of Galilean-type synagogues was in use for hundreds of years and can be found in houses and churches in Syria through the fifth and even sixth centuries CE. Furthermore, differences in architecture between synagogue buildings are influenced by factors such as regional styles or local topography and building materials, and not necessarily by the period of construction (see “The ancient synagogue and its components”). Magness argues that the archaeological evidence should be interpreted in its own right; the dating of a building should not be dictated by historical trends deduced from written sources. Only after the chronology of each ancient synagogue has been determined on the basis of the archaeological evidence can we accurately reconstruct the historical setting of ancient Jews. Magness has been at the forefront of this, often heated, dating debate, insisting that ancient synagogues should be dated scientifically on the basis of ceramics and coins found in and under the buildings. She insists that only a close examination of the archaeological material can provide us with the correct construction, renovation, and destruction dates for these buildings. Each synagogue building should be dated independently of typological or historical considerations, on the basis of well-excavated and thoroughly published archaeological evidence. However, although I agree with this methodology and also apply it in this study, not all archaeologists are convinced, and synagogues continue to be dated based on stylistic considerations, or historical events. In this study, I have tried to assess the dating debates for each building, choosing the date that I believe is the most accurate based on the published stratigraphy and the site’s archaeological materials. However, sometimes a date was determined by the original excavators many decades ago and was never re-evaluated. In other cases, the archaeologists admit that their date is problematic, but they have no other means of determining the exact time of construction. In both these cases, for the sake of the data needed in the database, I have chosen to use the dates provided by the excavators. Readers should note that as new evidence comes to light in the future or pottery types are re-dated, these dates might need to be adjusted.
 Exceptions are the Beth Alpha synagogue (according to a dedicatory inscription in the mosaic, built during the reign of Roman Emperor Justinus, which is either Justin I (518–527 CE) or Justin II (565-578 CE)), the Nabratein synagogue (a dedicatory inscription found on the lintel of the main door reads: “Built four hundred and ninety four years after the destruction of the Temple under the leadership of Hanina ben Lizar and Luliana bar Yuden,” or 564 CE), and the synagogue of Gaza-Maiumas (a dedicatory inscription in the south aisle reads: “Mena-ḥem and Yeshua, sons of the late Isai/Issi, wood merchants, as a sign of thanks for the most holy place, this mosaic (we) have donated in the month of Loos [July-August], 569 [of the era of Gaza = 508/9]”). See Naveh 1987, Nos. 13 and 43; Werlin 2012, pp. 332–334.
 This scholarship is sometimes called the “first school,” and started with Kohl and Watzinger, who dated Galilean synagogues on the basis of their architectural style and historical context (Kohl and Watzinger 1916; Aviam, 2001, p. 166). This school still exists, especially among Israeli archaeologists, who, for example, often link the construction of synagogue buildings in the Galilee to the thriving age of the rabbis in the 2nd and 3rd century CE (see for example Ma’oz 1996, p. 422; Dar 1999, p. 31).
 Many of the Syrian buildings cited as comparanda have inscriptions that date them primarily to the second and third century CE (Leibner 2018, p. 10). For an overview of the history of the tripartite synagogue typology into the Galilean type, broadhouse type, and Byzantine type, see “The ancient synagogue and its components”.
 Foerster 1971, pp. 207–208.
 Avi-Yonah 1981, p. 61: Avi-Yonah objects to a 4th century date for the synagogue of Capernaum since that would mean it was contemporaneous with the nearby synagogue at Hammath Tiberias, a building that is very different in style.
 Foerster 1971, p. 208.
 Magness 2001b. She refuted the architectural dating method by using the pottery and coins found in the synagogues at Capernaum, Gush Halav, and Meroth as case-studies, showing that these buildings are much younger than previously thought. Magness belongs to the so-called “second school” of thought, who state that in the field of archaeology, the terminus post quem provided by coins and other artifacts cannot be ignored (Magness 1993, 2005b, 2007a, 2012a; Aviam, 2001, p. 166).
 Magness 2001a, p. 35.
 Magness 1993, 2001b, and 2012a in which she re-dates the ancient synagogues of Meiron, Gush Halav, Chorazin, Capernaum, and Khirbet Shema’ based on the published stratigraphy of the excavations, as well as the pottery and coins. In recent years, Magness has started her own excavations in the Galilean synagogue at Huqoq and the results of this excavation will undoubtedly contribute to the discussion. The fact that the synagogue would have been dated to the 2nd-3rd century CE based on stylistic considerations but can actually be dated to the 5th century based on archaeological evidence shows that Galilean-type synagogues cannot automatically be dated to the second-third centuries based on stylistic considerations alone (Magness 2016, 2018, 2019b).
 Magness 2001b, p. 90, see also Chen 1986b.
 Foerster 1989; Tsafrir 1995. See also Amir 2012, who rebutted Magness’ assessment of the Galilean architectural style being used in Syria in the 5th-6th century.
 For example, many synagogue destructions in Israel have been linked to earthquakes that occurred in the region in the Late Roman and Byzantine periods. However, it is often difficult, if not impossible, to prove if an earthquake was the reason for a building’s destruction. An earthquake could have happened decades after a building was already deserted for other reasons, causing construction stones to crack, giving archeologists the false impression that this was the reason for the abandonment. For a critical analysis of using seismic activity as the cause of archaeological destruction layers, see Karcz and Kafri 1978.
 For example, Daniel Schindler recently re-evaluated the Galilean pottery from the Late Roman and Byzantine period in his PhD. dissertation. In his work, he re-dated many of the local pottery types, meaning that the associated archaeological levels from this region should be redated (Schindler 2017).