Medieval Shiloh—Continuity and Renewal

Abstract

The present paper deals with the development of cult in Shiloh during the Middle Ages. After the Byzantine period, when Shiloh was an important Christian cult place, it disappeared from the written sources and started to be identified with Nebi Samwil. In the 12th century Shiloh reappeared in the travelogues of Muslims, and shortly thereafter, in ones by Jews. Although most of the traditions had to do with the Tabernacle, some traditions started to identify Shiloh with the tomb of Eli and his family. The present study looks at the relationship between the practice of ziyara (“visit” in Arabic), which was characterized by the veneration of tombs, and the cult in Shiloh. The paper also surveys archeological finds in Shiloh that attest to a medieval cult and compares them with the written sources. In addition, it presents testimonies by Christians about Jewish cultic practices, along with testimonies about the cult place shared by Muslims and Jews in Shiloh. Examination of the medieval cult in Shiloh provides a broader perspective on an uninstitutionalized regional cult.

1. Introduction

Maintaining the continuous sanctity of a site over historical periods, and even between different faiths, is a well-known phenomenon: It is a well-known phenomenon that places of pilgrimage maintain their sacred status even after shifts in the owners’ faith (Limor 1998, p. 9). However, at times this process of “sanctity preservation” does not go smoothly and some of the original traditions disappear, at least from written sources. This void continues in some cases for a short period and in others for centuries, until the sites reappear, apparently attesting to a local oral tradition passed between the generations. This is what happened in the case of Shiloh [Khirbet Seilun, خربة سيلون]: It disappeared from the written sources in the Byzantine period and reappeared only toward the late 12th century AD.1

The relevant literature on the Shiloh excavations is fairly extensive. We will mention only the major works that assisted us in the current article. The Danish delegation that carried out digs in Shiloh in the 1920s and 1930s published several articles during the excavations (Kjaer 1927, 1930, 1931). A summary of the excavation report by the Danish delegation was published in two volumes: The first volume contains the pre-Hellenistic findings (Buhl and Holm-Nielsen 1969), while the second describes the finds from the Hellenistic period and on (Andersen 1985). In the early 1980s another excavation was carried out by Israel Finkelstein (Finkelstein et al. 1993).

From those excavations and other literature research we know that until the Bar Kochba Revolt, Shiloh was the location of a Jewish village with hints of a sacred past that included mentions of the Tabernacle that had been stationed there but with no testimonies of actual religious rites. After the Bar Kochba Revolt, the site was transformed from a Jewish town into a polytheistic town, and in the Byzantine period Shiloh became a place with Christian ritual significance. This is evident from its mention in Christian travel literature in the 4th century AD, and particularly when exploring the impressive archeological church findings (Schwartz and Shemesh 2019). At least three churches that operated concurrently were established in Shiloh and even continued to operate during the early Muslim period until the earthquake in 749 AD.

Since then, the site of Shiloh was not mentioned in the sources during the centuries that separated the Byzantine period from the end of the first Crusader period. Moreover, the identification of Shiloh “moved” to Nebi Samwil, and Shiloh’s original site was forgotten by the travelers, who had found a more convenient and closer alternative (Andersen 1985, p. 18; for the reasons for this transformation, see Schwartz and Shemesh 2019, p. 14). However, it should be mentioned that we find continuation in the medieval village of Shiloh through the archeological findings, such as Abbasid residential buildings and olive press (Hizmi and Haber 2014, pp. 107–8, 110–11; Livyatan Ben Arie forthcoming).

2. Paper Goals

Most of the studies dedicated to the site at Shiloh deal with the ancient period, particularly the biblical period, but the current article shall seek to emphasize the site’s later periods. The purpose of this article is to explore the ritual prosperity, albeit of a fairly short duration, of the site at Shiloh particularly under Islamic rule, and for this we shall seek to compare Jewish, Muslim, and Christian traditions and explore the archeological finds at the site. The discussion will center on the narrative contained in the traditions and on the specific structures described in them, and less attention shall be devoted to the history of Shiloh as a medieval village.

3. The Historical Sources

3.1. Muslim Sources

The Islamic faith saw itself as the legitimate heir of the biblical past, which included the books of the New Testament and the Midrashic literature, and as such acted to Islamicize the physical environment of Eretz Israel. Ancient traditions were embraced by Islam to prove its supremacy, and it saw its victories as a continuation of the biblical tradition (Frenkel 2008, pp. 148–49).

This appears to have been the case in Shiloh as well. The first medieval evidence we have regarding the village of Seilun is from 1173, when it was mentioned by traveler Ali al-Harawi, who described the way from Shechem to Jerusalem, mentioning Seilun among the villages: “Saylun is a village that contains the Mosque of the Ark of the Covenant (al-Sakīna), and also the Rock of the Table (Ḥajar al-Maida)…” (al-Harawi 2004, p. 66). Al-Harawi was the first traveler to record in writing the oral traditions of A-Sham (Syria) in his book Kitab al-Isharat ila Ma’rifat al-Ziyarat. Although the author’s family originated from Herat in Afghanistan, Ali himself was born in Mosul, Iraq. Nevertheless, Josef Meri showed that al-Harawi’s special interest in the traditions of the A-Sham region constituted an act of local patriotism versus the Crusader conquest (al-Harawi 2004, p. XXVIII). This first mention of Shiloh was not accidental, and see the discussion below on the reasons for its appearance. At around 1225, the well-known geographer Yaqut al-Hamawi quoted al-Harawi almost verbatim (Yaqut 1957, pp. 299–300). Subsequently Iranian physician, astronomer, geographer, and the author of the Marasid in 1300 AD Zakaria al-Cazwini, in circa 1250 AD, (al-Cazwini 1849, pp. 136–38) also brought the same report (Le Strange 1890, p. 527; on this treatise see Tollmacheva 2003, p. 1302), but the two latter may have copied from each other and all drawn from al-Harawi. These testimonies indicated that in the vicinity of biblical Shiloh there was a village named “Seilun,” and the residents of the village had traditions related to the Tabernacle, the place of the Divine presence, and some table on which we shall elaborate further below.

3.2. Jewish Sources

At around these same years Shiloh was “rediscovered,” not only in the works of Muslim travelers but also in those of Jewish travelers from the West. R. Samuel ben Shimshon (Rashbash), who accompanied R. Jonathan of Lunel, one of the greatest sages of Provence, in his immigration to Eretz Israel, reported in 1212 AD that “From there (= between Beit El and the Ai) we went to Shiloh and we saw there the place where the tabernacle was…” (Samuel ben Shimshon 1929, pp. 7–8). Samuel indeed did not explicitly mention the Arabic name Seilun, but it seems logical that this was the same place, north of Beit El. Samuel only noted that this was the location of the Tabernacle, but he did not refer to any specific structure.

In the mid-13th century AD several of the tombs mentioned in the early Byzantine period, such as the tomb of Eli the Priest, reappeared (Schwartz and Shemesh 2019, p. 7). It seems that the renewed mention of Eli’s tomb was related to the phenomenon of the ziyara (زِيارة, lit. “visit”) that flourished in this era in Eretz Israel. The custom of the ziyara included visiting holy places in general and the tombs of the pious in particular, and this was a common medieval custom both in Judaism and in Islam (see Goldziher 1971; Reiner 1988, pp. 317–20; Bilu 1998; Taylor 1999; Talmon-Heller 2002; Meri 2002; Lichtenstein 2007; Petersen 2018). Orientalist and researcher of Islam Ignaz Goldziher, author of the most fundamental and comprehensive article on the subject, claimed that the origin of the custom was polytheistic and that it preserved ancient traditions (Goldziher 1971, pp. 255–60). In contrast, Joseph Sadan explained that following the Crusader conquest access to some of the holy places was blocked, leading to a flourishing of new holy places in areas outside the Crusader kingdom (Sadan 1979). Similarly, Emmanuel Sivan (1991) showed that the Crusader conquest led to a rise in the sanctity of Eretz Israel and of Jerusalem and firmly established the undisputed status of Jerusalem as the holy city. In contrast to these researchers, Daniella Talmon-Heller claimed that the reason for the increase in the veneration of tombs stemmed from internal Islamic motives and primarily the growing influence of the Sufi stream that perceived the Sheikh as a wali (وَلِيّ, lit. “holy man”) while still alive, from whence it was a short step to transforming him into a saint once he had died (Talmon-Heller 2002, p. 275).

The works that accompanied the ziyara differ from regular descriptions of voyages and mainly include the “log” genre (on this see Howard 1980, pp. 18–20). According to Elchanan Reiner (2003), the “log” is a “dry” pedantic record that follows a defined geographical route, such as: “In Arbel Nitai Ha’Arbeli” [בארבל ניתאי הארבלי] (Yaakov Hashaliach 1969, p. 53). If this was the state of affairs, it is possible that the author of the treatise was not himself familiar with the physical location of the tombs at Shiloh, rather only assumed that the tomb of Eli was at Shiloh and perhaps also the tombs of his family. This may also be the reason that these records contain no description of the structure or geographical location of the tomb. This is the difference between travelogues and other genres called “guide” or “narration” (Howard 1980, p. 18), which include more information as well as personal impressions.

Hence, Shiloh was mentioned in such “logs” beginning from the 13th century AD. It is indeed necessary to emphasize that not all the testimonies mention the name “Seilun,” but the preservation of the tradition concerning Shiloh and comparisons to Muslim and Christian works indicate that they all refer to the same place.

In the compilation “Simaney Hakvarot” (“Signs of the Graves”) by Yaakov Hashaliach (Jacob the Messenger), who was affiliated with R. Yechiel of Paris, composed between 1258 and 1270 AD (see Prawer 1988, pp. 230–32), it is written: “In Shiloh Eli the High Priest and his two sons may they rest in peace” (Yaakov Hashaliach 1969, p. 53). According to this record the place was located between Kfar Hares and Jerusalem, and it is highly likely that this was Seilun. Eli’s two sons, Hofni and Pinhas, appear here for the first time with their father. It is interesting that despite the negative image of Eli’s sons in the Scriptures the travelers still chose to note the location of their graves as those of holy people, similar to the other graves in the records of Yaakov Hashaliach, which designate holy people.

Eli’s tomb also appears in a lamentation called “Kivrey Avot” (“Tombs of the Fathers”), whose author is unknown and which was dated to the 12th (Prawer 1988, p. 176) or 14th (Reiner 1988, p. 244) century AD. This record also contains the well-known phenomenon of multiple tombs around a central tomb, which we will discuss below: “To the two sons of Eli the Elder in Shiloh and he has a place and there is where the mother of Ichabod is buried” (Marmorstein 1925, p. 36).2 This is in fact the most extensive list of tombs that can be identified in Shiloh and it includes the mother of Ichabod, the daughter-in-law of Eli the Priest who died in childbirth at Shiloh (I Samuel 4:18–21). This list of tombs is also included in the last Jewish mention of Shiloh during the period studied, in the “Sefer Yuhasin” (“Book of Genealogy”) from 1442 AD: “In Shiloh Eli and Hofni and Pinhas and the mother of Ichabod” (Zacuto 1857, p. 228). In any case, the renewed mention of those buried at Shiloh, some or all of them, indicates that during the Ayyubid period Shiloh went through a certain transformation and resumed its maiden status as in the early Byzantine period, and the testimonies that refer to it mention the place of the Tabernacle, the place of the altar, and the tombs of the pious associated with it, i.e., Eli the Priest, his sons, and daughter-in-law.

A record from the Cairo Genizah (Ilan 1997, pp. 85–87),3 from the 14th or 15th century AD (Reiner 1988, p. 257, note 123), contains a list of holy places and their description. Regrettably, the part on Shiloh was severely damaged and rendered nearly indecipherable. Despite the terrible state of the text it is possible to conclude that Shiloh was identified as one of the holy places in these centuries.

The first to have clearly made a connection between the Muslim and Jewish traditions seems to have been Ashtori Ha-Parḥi, who traveled the land between the years 1314 and 1322 AD and wrote about it and its associated religious precepts (on Ashtori Ha-Parḥi see at length in Schwartz and Ashtori 2019). With regard to Shiloh he wrote:

Shiloh is three hours south of Shechem in a straight line and slightly to the east and it is on the way to Jerusalem, and on the road from Shechem to Jerusalem Shiloh is to the left and it is at the end of the first third of the way and called Seilun. And know that just as it is noted at the end of the Book of Judges4it remains today, where at first you will find to the right Lebonah, called Lubin. Continuing another one thousand cubits, you will see on your left a spring and a path, take it to the east and slightly south about one hour and you will arrive at Shiloh. And there is still today there a dome naming it Qubbat al-Sakīna, and near it a place called Maydat Bani-Israel which means luḥot (boards)”).

(Ashtori 1996, pp. 67–68)

Several things are evident from this description:

• Shiloh was not on the main road, and therefore it was necessary to deviate from it, apparently near the spring under the winding ascending path called Aqbat Luban. Ashtori traveled from Beit Shean to Shechem and from there southwards. It appears to us that it is possible to conclude from the above that the Mamluk khan, Khan Luban, did not exist in the time of Ashtori, as if it had he would have chosen to mention it as a more prominent point of reference than the spring. Mahmoud Hawari thought that it was built around the early 14th century AD, based on other similar khans, but he was not any more specific (Hawari 2001, p. 20). Therefore, it seems that the construction of the khan should be dated later or no earlier than the second quarter of the 14th century AD (Cytryn-Silverman 2010, pp. 127–30). See, for example, the words of Bonficius Stephanus in 1552 AD, who mentioned the place clearly: “Fifteen miles north of al-Birah there is a large hostel in the valley with a spring outside…” (Stephanus 1875, p. 252; Robinson and Smith 1856, p. 89).
• Ashtori did not mention any tombs on-site. This matter of little mention of tombs is characteristic of the works of Ashtori Ha-Parḥi. Ashtori’s main aim was to outline Jewish history through all the holy places (Schwartz and Ashtori 2019, pp. 232–37).
• Seeing that Ashtori had to explain Shiloh’s location to his readers in great detail, it may be assumed that its location was not familiar to his contemporaries, even those familiar with the land, as we do not find such a detailed description in his book regarding any other site. Nevertheless, it is more likely that Ashtori wanted to show the strong connection between the description in the Book of Judges and the land in his own days.
• The most important point pertaining to our matter is the mention of Kubat al Sakina and Maydat Bani-Israel in Arabic, very similar to the terms used by the Arab geographers. As we shall see below, the word kuba (in Latin cupa = “dome,” and in Arabic kuba [قُبَّة] = “dome” or “vault”) seems to denote the Tent of Meeting. The word kuba (قُبَّة) in Arabic usually denotes a tomb, and see Joshua Blau (2006, p. 522): “A tomb, particularly of a holy man, with a domed roof.” Since the Arabic phrase includes the term Divine presence rather than the grave of a certain holy man, this appears to be an exception.

Notably, Ashtori distinguished between a domed structure symbolizing the dome of the Divine presence, and a “place” called Maida Bnei Israel, whose meaning is debatable. Ashtori’s translation of the word maida as “tablets” was utilized by Joseph Braslavsky to show that Ashtori was unfamiliar with the Arabic language (Braslavi 1954, p. 266). In our opinion, however, this is a more complex translation. The Arabic word maida (مَائِدَة) means “table” or “feast.” In Provençal, Ashtori’s original language, a “table” also means “tablet,” as the ancient table was named for the tablet on which food was placed (Levy 1909, p. 358).5 When Ashtori wanted to translate into Hebrew the Arabic word maida, instead of “table” he chose the word “tablets”.6 He may have recognized an allusion to the Tablets of the Covenant that were in the ark at Shiloh, or he may have produced an erroneous translation.

In Islam the concept of Divine presence (سَّكِينَة) is directly related to the Tabernacle of the Testimony, which has the same root in Arabic (سكن) (Fahd 1995, p. 889). It is evident from here that this word has other meanings of tranquility and mercy in the Quran as well, and even a meaning that resembles the Hebrew concept of kavod (honor), i.e., a type of Divine revelation. This idea has already been discussed by Goldziher (1893) and Clermont-Ganneau (1899, p. 300), and Blau (2006, p. 303). The word al Sakina is also mentioned with regard the Ark of the Covenant, as written, for example, in the Quran 2:248: التَّابُوۡتُ فِيۡهِ سَکِيۡنَةٌ. Compare also to the words of Nasir Khusraw: “There is still another gate (to the Haram Area), and it is called Bab as Sakinah [al Sakina]; and in the hall adjacent thereto is a mosque that has many Mihrabs (or prayer-niches). The door of the entrance thereof is barred, so that no one can pass through. They say that the Ark of the Shechinah, which God—be He exalted and glorified! has alluded to in the Kuran, was once placed here, but was borne away by angels” (Le Strange 1890, p. 180).

Moreover, the Tent of Meeting is called kubat al-zaman in Arabic (قبة اللزمن) (Cardahi 1891, pp. 546–47). So this could have the double meaning of a domed holy place and the location of the Tabernacle. Reiner noted that the “table stone” noted above by al-Harawi was mentioned, for instance, by Yaqut, and appears to be used as a phrase designating the altar (Yaqut 1955, p. 284; and see Dozy 1881, p. 627). The stone at the Mount of Olives, which in the early Arab period marked the stone of the ascension of the Divine presence, i.e., one of the stops of the Divine presence after the destruction of the Temple, is also called maida (Gil 1992, p. 628). See also in Reiner (2000, p. 59), who noted there that the nearby place in which the cow was burned on the Mount of Olives is also defined as an altar, as noted by Menachem the Hebronite in 1215: “The altar made by Ezra” (Menachem the Hebronite 1969, p. 37). The record of “yitgadal sofer hanesiut” in the 14th century AD brings a similar description: “The Mount of Olives above a large [shelf?] of hewn stone made of stone … that is the monument of the Divine presence” (Ilan 1997, p. 133). Notably, churches contain a type of table–altar called, among other things, “holy table,” “altar,” or, in Greek, θυσιαστήριον’ (New Testament in Greek, Matthew 23:18; translated by Liddell and Scott 1940, p. 812). In our opinion, however, in this case al-Harawi meant a real table, as he further explained: “And truly the table is in the Church of Zion” (al-Harawi 2004, p. 66). Al-Harawi may not have understood the aim of the local tradition, and instead of referring to the destroyed altar of Shiloh he thought that this was the table from the Sura of the Table (سُورَة الْمَائِدَةِ),7 just as Ashtori appears to have misunderstood the meaning and searched for the word “tablets” as the interpretation of maida.

Several years after Ashtori Ha-Parḥi referred to Shiloh, Isaac Chelo described the site in writing:

From this town you travel to Shiloh, called Seilun. There are the tombs of Eli the High Priest and his two sons Hofni and Pinhas, a very fair monument where Jews and Muslims light candles constantly. One kabbalist old man sits by this gravestone. He is from Ashkenaz and he supports himself by copying holy books that he copies, such as Sefer Habahir by R. Nehunya ben Hakana, Sefer Habitachon by R. Yehuda ben Beteira, Sefer Hayetzira ascribed to R. Akiva, and others.

(Chelo 1919, pp. 102–3)

There is an argument among the researchers concerning the trustworthiness of this description. Shalom (1934) and Ish-Shalom (1935) declared, each for his own reasons, that the compilation is a forgery, but Michael Ehrlich (2008) contended that the main part of the description is authentic, although certain additions must be filtered out. (On Shiloh see Shalom 1934, pp. 35–36, and Ehrlich 2008, p. 71). Thus this treatise must be treated with caution, distinguishing between the realistic material and the legendary material it encompasses. On one hand, the fact that the only the object it mentions as existing on-site is a monument to biblical characters, alluding to the ziyara custom whose main feature was visiting graves, is conspicuous; on the other, he makes no mention of the Tabernacle, the ark, or the altar, and there is no physical description of the monument. Moreover, it is notable that he speaks about a sanctity that was common to both Jews and Muslims, on which we will expand below. However, his description of the old man who sits there copying “holy books” is hard to accept. This description is occasioned by the sanctity of the site, and from a literary perspective this writing can be interpreted in one of two ways: either the old man came to the site to be blessed by its sanctity, or the old man enhanced the sanctity by writing holy books.

3.3. Christian Sources

As stated, in the Middle Ages Shiloh was identified in Christian sources as located at Nebi Samwil (Elitsur 1984, p. 80, note 31). This was attested to by quite a few Christian travelers, such as Johannes of Würzburg in 1170 (John of Würzburg 1890, p. 9), Thietmar during 1217–1218 AD (Thietmar 2012, p. 111), and Burchard of Mount Zion at around 1280 AD (Burchard of Mount Sion 2012, p. 280). In the 14th century, however, Christian travelers too began to notice the different identification of Shiloh by Jews and Muslims. The first Christian source to draw our attention to this matter was John Fedanzola, a Franciscan monk who journeyed Eretz Israel in the 1320s:

Shiloh, about which Blessed Jerome says in his book on the distances of locations … Therefore, those are in error who say and hold Shiloh to be that place near Jerusalem at the 4th milestone, which is called St. Samuel. For that is in the tribe of Benjamin; while that about which Jerome speaks is in the tribe of Ephraim. From there, that one is at a distance from Shechem (which is called Nablus) of 18 miles; this other one about which Jerome spoke, is at a distance from said Shechem of 10 miles. In that one, there is no vestige or sign of the dwelling of the tabernacle; in this other, outside the city a half mile and toward the south, is a certain house with interior columns, very marvelous, which is called the House of God (Latin: Domus Dei), where there is still a certain window the size of the Ark of the Covenant; and the Hebrews say that there stood the said Ark. This house is held in such reverence by the Jews that they do not dare enter into it; but I, with my companions entered and diligently examined it, and I believe firmly that to be the place where the Tabernacle of the Alliance stood for a long in Shiloh and rightly is said of it: “We found it in the fields of the forest” and this is referred to God, found in the Tabernacle, in the Holy of Holies, at Shiloh. In fact, that place is very solitary and woody.

(Fedanzola 2003, p. 25)

Fedanzola’s description, detailed and dramatic as it is, attests that he had himself been to Shiloh and was apparently alluding to the dome of the Divine presence mentioned by al-Harawi and Ashtori, which he called “God’s house” (Domus Dei). (On the relationship between Fedanzola and Ashtori Ha-Parḥi see Schwartz forthcoming). In fact, he accepted the Jewish identification of the site, and on this issue returned to the original Christian identification by Eusebius and Hieronymus (Schwartz and Shemesh 2019, p. 4).

Several years later, in 1335 AD, the Augustine monk Jacopo of Verona gave a close description:

From Michmash we arrived at one village, located on a tall mountain, that is called Shiloh and it is where the Ark of the Covenant was stationed for a lengthy period and the Israelites would hold their assemblies there. To this day the Jews have great veneration for it and it is a one day’s walk from Jerusalem in the direction of Shechem.

(Jacopo da Verona 1895, p. 267–68)

Shiloh’s location, one day north of Jerusalem in the direction of Shechem, suggested that Jacopo was referring to Seilun. He mentioned the Ark of the Covenant that had been stationed there and the Jewish veneration of the site. He may have heard about the Jewish veneration from his Jewish guides, whom he mentioned explicitly and who attested to Jacopo’s appreciation for the “Jewish faith” (Braslavi 1954, p. 130).

In the early 15th century AD also Ghillebert de Lannoy, a Flemish diplomat and traveler, described the Jewish custom of “praying at the site of what used to be the Temple,” and even called the road from Jerusalem northwards the “road of Shiloh,” rue de Sylo (Ghillebert de Lannoy 1840, p. 61). In many French versions of the Bible Shiloh is described as the Temple.8

Franciscan monk Stephanus wrote in 1552 as follows: “And not far off on the right is Shiloh where are altar and ruined church are seen” (Stephanus 1875, p. 252; Robinson and Smith 1856, p. 89). Namely, this was an inactive site, at least for the Christians. It is unclear whether Stephanus was referring to ancient sources, such as Hieronymus, or whether he had been there himself. From the 16th century AD and on, there are sporadic mentions of the place, and most note merely the site’s general direction, for instance, Troilo’s identification of Shiloh at Khan Luban in 1657 AD (Robinson and Smith 1856, p. 89; Andersen 1985, pp. 18–19). No one described any ritual acts performed at the site, and it appears that from then until the time of Edward Robinson we have no testimonies of pilgrims who came to the site (Robinson and Smith 1856, p. 89). An interesting testimony to the Tel’s rediscovery by Europeans is a Swiss coin found at the Tel, dated 1851 (Kjaer 1927, p. 204).

Hence, it may be seen that during the 14th century AD pilgrims still disagreed as to the identification of Shiloh: Most Christians identified Shiloh as located at Nebi Samwil, while the Jews identified it at Seilun near Shechem. A small number of Christians, however, understood the geographical logic of Shiloh’s identification at Seilun and accepted it. With the decline of the village of Shiloh all mentions of the site disappeared as well.

As an aside we shall only say that the Samaritans, too, identified Shiloh as located at Seilun, as evident from the mention of Shiloh by this name in the Samaritan book of Joshua, written in Arabic in 1362 AD (Conder 1876, p. 190).

3.4. The Archeological Finds

With the Muslim conquest in 638 AD and during the beginning of the early Muslim period, Shiloh continued to exist as a Christian town (Hizmi and Haber 2014, p. 111; Livyatan Ben Arie, forthcoming). We know this because some of the churches were active during the Umayyad period as well (Schwartz and Shemesh 2019, p. 12).9 However, the end of their public use, probably after the earthquake of 749 AD, and their transformation into residences showed that they were no longer used as churches in the Abbasid period. Are there archeological indications of the rituals mentioned by Muslim, Christian, and Jewish travelers? From the Muslim periods two uncovered ritual structures remain as well as one covered structure (Figure 1).

3.5. Jamiʿ al-Sitin

The one structure is called at present Jamiʿ al-Sitin (جامع الستين), and it is the southernmost structure at Shiloh (Conder and Kitchener 1882, p. 369; Kjaer 1930, pp. 79–81; Kjaer 1931, pp. 86–87; Yeivin 1973, pp. 168–62; Andersen 1985, pp. 78–81) (Figure 2 and Figure 3). The first researcher to describe the place was Robinson, and he said that it was called the Mosque of Shiloh (Robinson and Smith 1856, pp. 85–86). Then Charles Wilson called the place “Jamia el Arbain” (Wilson 1873, p. 38). Clermont-Ganneau insisted that the name was “Jame es Sittin” and explained that the word sittin resembles the word shekhinah [al-Sakina] (Clermont-Ganneau 1899, p. 305). This structure was identified by many researchers as a synagogue (Kjaer 1930, p. 80; Luria 1947, pp. 171–75; Braslavi 1954, p. 303; Andersen 1985, p. 81; Ilan 1991, p. 251). Braslavsky saw the special form of the synagogue as an attempt to emulate the Tabernacle’s animal-skin sheets, and Kier even recognized it as the tomb of Eli. Hans Kjaer noted that according to the local elders it was named thus for the 60 martyrs killed at the site at some period. The lintel of the structure, which in the 19th century was still situated above the doorway, fell down subsequently, reminding these researchers of the lintels of Galilean synagogues. However, Eitan Klein has shown, quite persuasively, that this was not an original lintel intended for a building, rather the side of a burial trough from a burial cave that was secondarily adapted for use as a lintel (Klein 2011, pp. 172–73). The lintel was stolen from the site by an antiques thief, was found buried in a nearby field, and was subsequently transferred to the Rockefeller Museum (Damti 1970, p. 6; Yeivin 1981, p. 18).

The researchers’ conjecture regarding identification of the structure as a synagogue was based on the assumption that in medieval times Jews resided at the site or at least had a synagogue or ritual place there, as mentioned by the travelers. In fact, from the Roman period on we have no knowledge of a Jewish presence on-site, and during the period under consideration it was solely a Muslim holy place, and only occasionally did Jews visit and develop a similar tradition. This state of affairs was characteristic of most holy places in Mamluk Eretz Israel. Very few places were maintained by Jews in practice, and Jews usually only visited existing Muslim places of worship (Reiner 1988, p. 253). At the same time, it is not possible to prove unequivocally that a small number of Jews, such as merchants, did not live there, but, as stated, we have no evidence of this.

Jamiʿ al-Sitin, like Jamiʿ al-Yatim (below), contains clear elements of spolia taken from the church, such as chancel pillars and Corinthian capitals (Andersen 1985, pp. 77–81). Two Arabic inscriptions were also found on the site, but regrettably they were not deciphered and subsequently disappeared (Yeivin 1972; Yeivin 1973, p. 161).10 In our opinion, al-Harawi and Ashtori’s mention of the “mosque of the Divine presence” and “dome of the Divine presence,” respectively, probably related to this mosque (for a full discussion on this subject see Schwartz et al. forthcoming).

3.6. Jamiʿ al-Yatim

The second structure (Figure 4 and Figure 5) is called Jamiʿ al-Yatim (جامع اليتيم), and it is located immediately at the foot of the Tel, to the south. This name too is unclear. In 1847 John Wilson noted that the place was called “Mazara,” and perhaps this name preserved the custom of the ziyara on-site (Wilson 1847, p. 294). Robinson did not give the name of the destroyed mosque (Robinson and Smith 1856, p. 86). In 1873 Charles Wilson described the place as Conrad Schik, called it Jamiʿ a-Daim, translated as “the eternal” (Wilson 1873, p. 38). Three years later Conder wrote that the place was called “Jamiʿ a-Taim,” interpreted as “mosque of God’s servant” (Conder 1876, p. 192). For more about the mosque and the elements taken from the surrounding churches, such as columns and chancel, see Andersen (1985, pp. 48–49). Below Jamiʿ al-Yatim two churches were discovered, one on top of the other, which we denote “the northern churches” (Schwartz and Shemesh 2019, p. 8). An Ayyubid coin found in the foundation ditch of the mosque may indicate that it was established in that same period (Magen and Aharonovich 2012, p. 205).

In our opinion, this mosque was used by the residents of Seilun and its good state of preservation derives from its late date of usage. The village itself was located on the Tel, so it is the most logical place for the village mosque.

3.7. Open-Air Weli

The third structure is a holy Muslim structure of the open-air Weli type. In present times it is no longer visible, but the Danish excavators contended that it had been a medieval Muslim ritual structure as it contained a mihrab (Figure 6 and Figure 7); (Kjaer 1927, pp. 209–11; Andersen 1985, pp. 76–77).11

An open-air Weli is an unroofed holy Muslim place, usually round (on the features of the open-air Weli see Canaan 1927, pp. 60–63). At present there is only a hollow in the ground that attests to its place near the Byzantine winepress to the west of the Pilgrim’s church. The site of the open-air Weli was excavated by a Danish delegation in the 1920s, and they concluded that the site was medieval. Eight graves were also found in the southeastern corner of the facility (Kjaer 1927, p. 211; Andersen 1985, p. 77). The archeologist Aage Schmidt, who was the living spirit of the Danish excavations from their onset (Buhl and Holm-Nielsen 1969, p. 10), suggested identifying this type of structure with a similar altar in Petra, Jordan, and tried to argue that the altar was ancient, as were the altars in Petra, whereby the Muslims had preserved a Christian or even polytheistic site. (Figure 8; Kjaer 1927, p. 211).

It may be suggested that in their words on the maida al-Harawi and Ashtori Ha-Parḥi were referring to the same open-air Weli, and the difference between al-Harawi’s hajr al-maida (= “table stone”) and maida Bnei Israel attests to a different conception of the same “site”: the table in the Quran, which descended in God’s hands, in Islam, and the altar of the Shiloh tabernacle in Judaism. In the publications of the Danish delegation it is evident that the Arabs preserved the name al-mede in the field surrounding the Weli structure (Kjaer 1927, p. 209).

4. Discussion

Considering the above, the question is why did the ancient tradition of Shiloh emerge in the 12th century AD of all times, as written by al-Harawi on one hand and the Rashbash on the other? First, it must be said that the words were indeed written down only in the 12th century AD, but they reflect an early well-regulated tradition at the site, and the written version only describes the encounter of western pilgrims with the existing custom (Reiner 2003, p. 313). We mentioned above the different opinions with regard to the ziyara, and in our opinion the record of the rituals at Shiloh can be added to this process. As stated, Sadan detected in the emergence of the Muslim rites a counter-response to the Crusader conquest (Sadan 1979, p. 33). Following the Crusader conquest, access to some of the holy places was blocked, which led to the appearance of new sites in areas outside the Crusader kingdom. A similar direction was taken by Sivan (1991) as well.

If we continue this line of thinking, it appears that it is necessary to take into account al-Harawi’s biography, and it is not incidental that he was the first to mention the tradition concerning Shiloh. As stated, Meri proposed that al-Harawi, who also served as an official in the Ayyubid government in Aleppo and Hama, had a specific interest in the traditions of Bilad a-Sham as an act of local patriotism versus the Crusader conquest (Meri 2002, pp. XXI–XXVII). In our opinion, the recording of the Shiloh traditions in writing should be seen in a wider context grounded in the rising emphasis on Eretz Israel and its traditions in the Ayyubid period, which explains the Muslim tradition of Shiloh’s sanctity. Indeed, in al-Harawi’s time (1173 AD) Shiloh was in the territory of the Crusader kingdom, but Ronnie Ellenblum (1998, p. 245) has already shown that the population in Shiloh and its vicinity was Muslim. In fact, northwards of the Aboud-Sinjil line there was no Frankish population, rather only a Muslim population, and therefore the special interest in this area (see also Levy-Rubin 2019).

The routes of the ziyara in Eretz Israel developed at a fairly late stage and they seem to have become fully established only in the Ayyubid period, i.e., after 1187 AD (Reiner 2003, p. 313), concurrent with the Muslim ziyara in Syria, which flourished during about the same years as well (Talmon-Heller 2002, p. 268). Reiner (1988, p. 253) also showed that the foundations of the ziyara were in the “regional cults.” (On the regional saint’s cult in general see the introduction by Werbner 1977, pp. IX–XXXVII. On regional rites in the Galilee see Reiner 1996, 2012). The timing of the annual Jewish ziyara was from 15 Iyar in Meron to 28 Iyar in Rama, and it was identified as located in Nebi Samwil (Reiner 1988, pp. 370–72; Elizur 2007, pp. 177–80). Shiloh is in the vicinity of the road that connects Meron with Rama, and the proximity to this route seems to have brought Jewish pilgrims to it beginning from the time that the annual ziyara was established in the Ayyubid period. This phenomenon of the establishment of traditions in holy places that had been forgotten and then regained prominence is familiar from several other places on this route as well, such as Joseph’s Tomb (Schwartz 2018) and the Tomb of Habakkuk (Lisovski 2007). At the same time, we are also familiar with the opposite phenomenon wherein holy places that were no longer on the route of the ziyara were fated to be forgotten and to disappear from the written sources (Reiner 1988, pp. 233–34).

Reiner (1988, pp. 289–9) tried to link another written source to the Jewish ziyara in Shiloh, one that is called the “Perek Tzomot” (Chapter of Fasts): a list of the death dates of various biblical figures that was usually attached as an appendix to manuscripts of the Scroll of Fasting and was first printed in the book Halakhot Gedolot (Elizur 2007, p. 4). In this compilation significant annual dates were assembled, including a mention of the death of Eli and his sons on 10 Iyar: “On the tenth of Iyar Eli the Priest and his two sons died and the Ark of the Covenant was taken captive” (Elizur 2007, p. 177, note 77). The ancient version of this compilation is from the 9th century, and a slightly different version was dated by Reiner to the 13th century AD (Reiner 1988, p. 289). However, Shulamit Elizur showed that this version too is very ancient (Elizur 2007, p. 123, note 50). Moreover, in some manuscripts of Eretz Israel sources other dates are given for the death of Eli, in the month of Sivan (Elizur 2007, pp. 176–77).12 If so, the development of the “Perek Tzomot” should not be directly linked to the emergence of the ziyara.

Although there is no direct relationship between the issues, the death dates of Samuel, Joshua, and Eli were all around the months of Iyar and Sivan, concurrent with the Jewish annual ziyara. It can be concluded both from records of journeys and from Jewish travelogues that the ziyara passed through Samaria, as various sites in Samaria appear in it. It can be proven that some of those who participated in the annual ziyara visited Shiloh as well, as mentioned by Christian pilgrims in the 14th century AD, which contributed to the rising status of the holy site.

The development of the tomb of Eli, concurrent with identification of the location of the Tabernacle and of the altar, indicates a stage in the formation and shaping of a holy site. Although it would have been possible to remain with the biblical story of the significance of the place of the Tabernacle, the pilgrims who visited the site chose to add related biblical figures, since they could address their prayers directly to these figures (Reiner 1988, pp. 260–62). Although Reiner specifically excluded Shiloh from the holy sites associated with tombs, in our opinion it is not possible to disregard the inclusion of the site in the list of tombs, sometimes even with no mention of the location of the Tabernacle, whereby only the tomb of Eli was mentioned, for instance, in the report by Yaakov Hashaliach and in the “Tombs of the fathers” lamentation mentioned above. Another possible reason is that the place of the Tabernacle did not become “sanctified for all generations” and the Tabernacle itself moved to other places, versus the graves of the pious in which their physical body remained, according to tradition.

This process of “personifying” a place is characteristic of many tombs of holy people and is comprised of two stages. In the first stage a holy place was mentioned as knesset rather than as a tomb (kever), for instance, in the ancient Galilean record in the Parma manuscript (Reiner 1988, p. 233, note 51; Reiner 2003, p. 317), and only in the second stage was there a transition to identifying holy tombs. In addition, the phenomenon of multiple secondary sites at one site, tombs or other elements, is a familiar phenomenon in the evolvement of a sacred site (for instance, in Hebron; (Reiner 1988, p. 309)). In the case of Shiloh the situation is more complex and, to begin with, no tomb was identified on-site, rather the site of the Tabernacle replaced the knesset of the first stage, and at the same time the location of the altar (the maida) was also identified. The tomb of Eli symbolized the second stage. The multiple graves of Hofni, Pinhas, and the mother of Ichabod also fit the scheme presented by Reiner for the development of a central tomb. Once the site becomes established family members are gradually identified around the holy person, such as sons and a wife, as happened in this case as well (Reiner 1988, p. 237). Rashbash and Ashtori Ha-Parḥi did not follow this pattern, apparently under the influence of the Muslim tradition; they returned to the initial identification of the site of the Tabernacle and did not refer to the graves. This approach may have derived from the objection to worship of graves in Judaism (Lichtenstein 2007, pp. 314–32). The other Jewish travelers focused on the biblical figures buried on-site.

As stated, when exploring the historical sources and archeological finds it appears that the worship at Shiloh may have been common to both Jews and Muslims, and, at least in the case of Fedanzola, also to Christians. If such a ritual indeed took place at Shiloh it was not the only case. Examples of shared rituals are familiar from Ein el-Baqar in Acre, Reuben’s tomb in Roma in the Galilee, and the ceremonies at Meron held in the month of Iyar (Reiner 1988, pp. 254–55). Kedar (2001) discussed the different modes of collaboration at ritual sites, and in our opinion the case of Shiloh is an example of the model of coexistence rather than of elimination and competition. No specific shared ceremony is known in the context of Shiloh, rather only that it served as a shared place of prayer.

5. Conclusions

The traditions at Shiloh did not disappear from the site although the place was not mentioned in any sources at least from the 6th century to the late 12th century AD. This is an uncommon phenomenon, and in many biblical places we find chains of traditions through the periods like Jerusalem, Hebron, etc. It appears that local residents continued to preserve oral traditions until these were first written down by al-Harawi in the 12th century AD. If so, nearly two thousand years after the exile of the Israelite Kingdom, people identified Shiloh (apparently precisely) and were aware of traditions concerning the site even in the absence of a reasonable chain of delivery, and this is evidence of an exceptional phenomenon. Other Muslim travelers and most of the Jewish travelers embraced these traditions. The establishment of the ziyara during these centuries and the location of Shiloh on the route from Rama to Meron seem to have expedited this process.

Some of the Christian travelers noticed the Jewish veneration for the site of the Tabernacle and some were even persuaded that the tradition was justified, such as Fedanzola, but these were the minority. Most medieval Christian travelers continued to accept Nebi Samwil as the site of Shiloh. The current location remained a Muslim village, and the Jews took part in common rituals together with local residents. Once the village declined and disappeared in the Ottoman period, the various pilgrims who had visited disappeared as well, and only in the 1830s did Robinson suggest the identification anew.

In our opinion, the medieval ritual sites in Shiloh can teach us about the culture of holy places in the Holy Land. Their unique feature is the fact that each is not a place of burial, rather an ancient ritual site that contains the ancient sanctity within it, and secondary traditions related to the biblical narrative developed there, such as the tradition of the altar. Nonetheless, in the case of this biblical historical site as well we see that most of the Jewish pilgrims focused on the biblical figures related to the place and their graves as part of the personification phenomenon mentioned above. It is to be hoped that additional excavations in medieval ritual sites at Shiloh will reveal further details concerning the essence of this ritual and the beliefs of those who prayed there.