A preliminary version of this paper was presented at the 42nd Annual Conference of the Association of Jewish Studies in Boston (December 2010). I would like to thank Zvi Ben-Dor Benite, Moti Benmelech, Martin Jacobs, and Micha Perry for stimulating discussion on this topic and the anonymous AJS Review readers for helpful comments.

1. Neubauer, Adolf, “Where Are the Ten Tribes? I. Bible, Talmud, and Midrashic Literature,” Jewish Quarterly Review 1 (1888): 20Google Scholar.

2. Though the river's impassability is a standard feature of the Sambatyon legend, as this study will make clear it was by no means an initial component of the legend. While a number of Amoraic sources (Y. Sanhedrin 10:5; Tanḥuma, Addition to par. Shelah, 6 [ed. Buber]; and Midrash Eikha Rabba, par. 2:4, to Lamentations 2:5 [ed. Buber, 112]) refer to those located past the river as prisoners, perhaps thereby implying that it cannot be crossed, this point is first explicitly made in later (how much is hard to say) sources such as the tradition preserved in the eleventh-century Midrash bereshit rabbati, ed. Albeck, Hanoch (Jerusalem: Mekize Nirdamim, 1940; reprinted 1967), 124, ll. 15–16Google Scholar, where God stretches forth the Sambatyon stream before the sons of Moses, closing them up behind it, “in order that no one will be able to pass over to them” (“כדי שלא יוכל אדם לעבור אליהם”). This source in all likelihood preserves substantially older material. This specific aspect of the legend is also prominent in the Eldad the Danite literature. In one version we read that because of the cloud that settles upon the river on Shabbat “it is impossible for anyone to approach it until after the Sabbath” and that “we are not able to come to [those past the river], and they are not able to come out from there.” Haberman, A. M, Kitve R. Avraham Epstein, vol. 1 (Jerusalem: Mossad ha-Rav Kook, 1949), Story 1, 39Google Scholar. In a second version we learn only that owing to the fire that erupts along the river on Shabbat, one cannot come (lit. touch לנגע) within a mile of it (Ibid., Story 2, 55). The exact relationship between the Eldad the Danite literature and Midrash bereshit rabbati remains vexed. Hananel Mack has recenly argued that the latter source copied in this regard from the former literature while Micha Perry has argued the account in Midrash bereshit rabbati “represents the kind of sources Eldad (or someone in his name) used to create his description” of the lost tribes and their Sambatyon environment. See Mack, Hananel, Mi-sodo shel Mosheh ha-Darshan (Jerusalem: Mossad Bialik, 2010), 190–194, esp. 191Google Scholar and Perry, Micha, Mesirat ve-shinui: masoret yed‘a be-kerev yehude ma‘arav ’eropah bi-yeme ha-benayim (Bnei Brak: Ha-Kibbutz ha-Meuḥad, 2011), 71–72Google Scholar. These sources can all be found together in Hebrew in Kitve R. Avraham Epstein 1, 1–211. A helpful collection of the various sources in English translation is presented in Reeves, John C., Trajectories in Near Eastern Apocalyptic: A Postrabbinic Jewish Apocalypse Reader (Atlanta: Society of Biblical Literature, 2005), 208–224Google Scholar. Explicit reference to the halakhic prohibition of crossing the still river on the Sabbath appears to be an even much later development, appearing for the first time, to the best of my knowledge, only in the fifteenth century. I hope to deal with the development of these aspects of the myth in detail in a subsequent study.

3. Even the name of the river varies in the sources: Sabbatikon, Sabbatyon, Sabbatinus, Sabbatino, Sanbatyon, and Sambatyon are all attested. For convenience, when referring in general here to the river I use the most common and best-known variant. The replacement of –bb with –mb (as in Sabbatyon and Sambatyon) is a documented linguistic phenomenon in Indo-European and Semitic languages. See Tcherikover, Victor A., Fuchs, Alexander, and Stern, Menaḥem, eds. Corpus Papyrorum Judaicarum, vol. 3 (Cambridge, MA: Harvard University Press, 1964), sec. XIII, 44Google Scholar.

4. For example, Midrash Aggada Shemot, par. Vayakhel, to Exodus 35:1 (ed. S. Buber, p. 187), where the river draws along stones (“מושך אבנים”), but is not clearly stated to lack water. Tanḥuma, Ki-Tissa, par. 33, to Exodus 34:27 (ed. Zondel, 126) speaks of rocks and sand (“מושך אבנים ו/בחול”), but again does not explicitly deny that it contains water. For that, one apparently needs to wait until the emergence of the Eldad the Danite tradition, where we read of a river which “six days of the week rolls around stone and sand without a single drop of water, and on Shabbat rests” (”והנהר גולל כל ששה ימי השבוע אבנים וחול בלי שום טיפת מים, ובשבת נח“) (Kitve R. Avraham Epstein 1, 29; see also Eisenstein, Judah David; Adler, Elkan Nathan, Jewish Travellers in the Middle Ages (New York: Dover, 1987), 14Google Scholar). David Kaufmann surmised that the Sabbatical river began its course as a river of sand, later understood to flow on weekdays, thanks to the dual meaning of the Hebrew חול (ḥol), which can refer to either secular, profane time (as opposed to holy) or sand. But since it is clear that the earliest sources on the Sambatyon presume a waterway, Kaufmann's reconstruction cannot be correct. Kaufmann, David, “Le Sambatyon,” Revue des Etudes Juives 22 (1891): 285–7CrossRefGoogle Scholar.

5. For example, Tanḥuma, Ki-Tissa, par. 33 “and on Shabbat rests” (“ובשבת נוח”); Midrash Aggada Shemot, par. Vayakhel. Later, the tradition of the resting Sambatyon comes to be supplemented by that of the Sabbatical fish, which rests on the shore all Sabbath long, and by that of the mountain, out of which silver can be mined every day except for on Shabbat. See R. Eleazar of Worms, Perushe siddur ha-tefilah la-rokeaḥ, “Seder Kiddush Leil Shabbat,” 487 n. 119 and Sode raza as cited in Sefer yalkut Reubeni (Amsterdam: Immanuel ben ha-Yashish Yosef Attias, 1700), 11bGoogle Scholar.

6. See footnote 2 above.

7. For example, according to Bereshit Rabba the ten tribes wandered into exile on the other side of the river Sambatyon, whereas Judah and Benjamin are scattered throughout all known lands. See Bereshit Rabba, Vayeiẓei, par. 73:6, to Genesis 30:24 (ed. Theodor-Albeck, 2:850).

8. See the version of the Eldad the Danite tradition discussed by Benite, Zvi Ben-Dor, Ten Lost Tribes: A World History (Oxford: Oxford University Press, 2009), 87CrossRefGoogle Scholar.

9. For the link between the “Sons of Moses” and the Sambatyon, see the various versions of the Eldad the Danite tale; see also Benite, Ten Lost Tribes, 92. The Rechabites, first introduced in Jeremiah 35:1–19, were a clan committed to abstinence from alcohol, agriculture, and the construction of permanent dwellings. A number of Medieval Jewish sources place them past the Sambatyon. See Reeves, 20–45, for brief discussion and bibliography as well as Nikolsky, Ronit, “The ‘History of the Rechabites’ and the Jeremiah literature,” Journal for the Study of the Pseudepigrapha 13, 2 (2002): 185–207CrossRefGoogle Scholar and Nikolsky, Ronit, “The Rechabites in Ma‘aseh Alexandros and in the Medieval Ben Sira,” Zutot 4 (2004): 35–41CrossRefGoogle Scholar.

10. Benite, Ten Lost Tribes, 77, 120.

11. See, for example, Y. Sanhedrin 10:5.

12. Benite, Ten Lost Tribes, 77, 120. Other important recent contributions include Gonen, Rivka, To the Ends of the Earth: The Quest for the Ten Lost Tribes of Israel (Northvale, NJ: Jason Aronson, 2002)Google Scholar; Halkin, Hillel, Across the Sabbath River: In Search of a Lost Tribe of Israel (Boston: Houghton Mifflin, 2002)Google Scholar; Parfitt, Tudor, The Lost Tribes of Israel: The History of a Myth (London: Weidenfeld & Nicolson, 2002)Google Scholar; Cooper, Alanna Esther, “Conceptualizing Diaspora: Tales of Jewish Travelers in Search of the Lost Tribes,” AJS Review 30 (2006): 95–117CrossRefGoogle Scholar; Barmash, Pamela, “At the Nexus of History and Memory: The Ten Lost Tribes,” AJS Review 29 (2005): 207–236CrossRefGoogle Scholar; Veltri, Giuseppe, “‘The East’ in the Story of the Lost Tribes: Creation of Geographical and Political Utopias,” in Creation and Re-Creation in Jewish Thought: Festschrift in Honor of Joseph Dan, eds. Elior, Rachel and Schäfer, Peter (Tübingen: Mohr Siebeck, 2005), 249–269Google Scholar, reprinted in Veltri, Giuseppe, Renaissance Philosophy in Jewish Garb (Leiden: Brill, 2009), 144–168CrossRefGoogle Scholar (all citations here are taken from the former); Busi, Giulio, “Sabbatyon” in his Simboli del pensiero ebraico: lessico ragionato in settanta voci (Turin: Einaudi, 1999), 306–12Google Scholar; Avni, Zvi, “Ha-Sambatyon: Gilgul Masorot,” Mada‘e ha-yahadut 37 (1997): 147–159Google Scholar; Toaff, Ariel, “Il Sambation, fiume vivente,” in Mostri giudei: L'immaginario ebraico dal Medioevo alla prima età moderna (Bologna: il Mulino, 1996)Google Scholar; Loewenthal, Elena, “La storia del fiume Sambation: alcune note sulla tradizione ebraica antica e medievale,” in Biblische und Judaistische Studien: Festschrift für Paolo Sacchi, ed. Sacchi, Paolo and Vivian, Angelo (Frankfurt am Main: Peter Lang, 1990), 651–663Google Scholar. Aaron Rothkoff's EJ entry “Sambatyon” offers a helpful survey of the major references to the river. Finally, less scholarly in nature but useful as a collection of sources is Aviḥail, Eliyahu, Shivte Yisrael: ha-‘ovdim veha-nidaḥim (Jerusalem: Agudat Amishav, 1987)Google Scholar.

13. Benite does devote some attention to these sources (Benite, Ten Lost Tribes, 79–80), correctly noting that neither associates the river with the lost tribes or places it at the ends of the earth. The research on both the Eldad the Danite and Prester John literature is vast. For studies which treat these characters in tandem see Wasserstein, David J., “Eldad ha-Dani and Prester John” in Prester John, The Mongols and the Ten Lost Tribes, eds. Beckingham, Charles Fraser and Hamilton, Bernard (Aldershot: Variorum, 1996): 213–236Google Scholar and the more recent article of Perry, Micha, “The Imaginary War Between Prester John and Eldad the Danite and Its Real Implications,” Viator 41 (2010): 1–23CrossRefGoogle Scholar. For various Sambatyon-related travel accounts, including the diary of David Reubeni, see Adler, Jewish Travellers. Traditions of the Sambatyon are of course hardly limited to Jewish sources. For Samaritan, Christian, and Muslim reports, see the bibliography referred to by Ginzberg, Louis, Legends of the Jews (Hildesheim: Georg Olms Verlag, 2000), vol. 4, 40–78Google Scholar. Specifically on a possible connection between the Samaritans and the Sambatyon, see Grünbaum, M., “Nachträge zu den ‘Bemerkungen über die Samaritaner,’” Zeitschrift der Deutschen Morgenländischen Gesellschaft 23 (1869): 627–630Google Scholar.

14. Benite, Ten Lost Tribes, 80–1 also draws briefly on associations between Saturn and the Sabbath in speculating that elements of the mythology surrounding this god may have influenced the development of the Sambatyon legend.

15. My linkage of the Sambatyon myth to Sabbath discourse is briefly and partially anticipated by Giuseppe Veltri (“‘The East,’” 261), though I disagree with his reading of Pliny as well as with his apparent assumption that the Ten Lost Tribes and Sambatyon legends are inherently conjoined.

16. Maior, Pliny, Historia Naturalis, ed. Mayhoff, Karl Friedrich Theodor (Leipzig: Teubner, 1906), 31.18.24Google Scholar (“in iudaea rivus sabbatis omnibus siccatur”).

17. Josephus, The Jewish War Books V–VII, trans. Thackeray, H. St. J., Loeb Classical Library 210 (Cambridge, MA: Harvard University Press, 1997), 7.96–99; 336–7Google Scholar (also as BJ = Bellum Judaicum). The translation has been modified slightly.

18. On Arcea, see ibid., n. c. Earlier in his Jewish War (7.18; 312–3), Josephus refers to Raphanea as the initial base of the Twelfth Roman legion. For his part, Pliny refers in the Natural History (5.16.74) to a city named Raphana, a member of the Decapolis league, located to the east of Roman Judea. This appears, however, to be a different place.

19. For a somewhat similar instance of Josephus responding to a pagan historian, see Shaḥar, Yuval, “Josephus's Hidden Dialogue with Strabo,” in Josephus Geographicus: The Classical Context of Geography in Josephus, ed. Shaḥar, Yuval (Tübingen: Mohr Siebeck, 2004), 238–256, esp. 25–34Google Scholar. In highlighting a particular, strange-behaving river, both Pliny and Josephus are in a sense responding to the recently deceased Roman philosopher Seneca (died 65) who in his Natural Questions (3.16.1) specifically rejected the value of referring to individual streams: “But why is it that certain springs are full for six hours and for six others dry? It is superfluous to refer to individual rivers, which in certain months are wide and in others narrow, and to seek out an explanation for individual cases, since I can adduce the very same explanation for all of them.” (“Sed quare quidam fontes senis horis pleni senisque sicci sunt? Superuacuum est nominare singula flumina, quae certis mensibus magna certis angusta sunt, et occasionem singulis quaerere, cum possim eandem causam omnibus reddere.”). Seneca, Naturales Quaestiones 1, trans. Corcoran, Thomas H., Loeb Classical Library 450 (Cambridge, MA: Harvard University Press, 1971), 236–7Google Scholar.

20. Pliny, Natural History 1, trans. Rackham, H., Loeb Classical Library 330 (Cambridge, MA: Harvard University Press, 1991)Google Scholar, vii; Carey, Sorcha, Pliny's Catalogue of Culture (Oxford: Oxford University Press, 2003), 15CrossRefGoogle Scholar. On Pliny's relationship to the imperial family, see the articles by Gely, S., Magno, P., and Zucchelli, B. in Atti del Congresso internazionale di studi flaviani (Rieti: Centro di Studi Varroniani Editore, 1983), 313–23, 331–5, 427–432Google Scholar.

21. Murphy, Trevor, Pliny the Elder's Natural History: The Empire in the Encyclopedia (Oxford: Oxford University Press, 2004), 114CrossRefGoogle Scholar. H. Rackham dates the work on the basis of Pliny's description of Titus in his preface as six times consul (“sexies consul”). Rackham, viii.

22. Shaye J. D. Cohen captures the relationship most succinctly: “If any historian was a Flavian lackey, it was Josephus.” Cohen, Shaye J. D., Josephus in Galilee and Rome: His Vita and Development as a Historian (Leiden: Brill, 1979), 86CrossRefGoogle Scholar.

23. See Cohen, Josephus, 85–7 for an indication of why book 7 has to have been written under Domitian. See below for further discussion as to why this probably took place during the early years of his rule.

24. See Veltri, “‘The East,’” 261, where it is argued that rabbinic tradition on the Sambatyon river and Ten Lost Tribes “convinced” Pliny. There is no evidence that Pliny “locates the ten tribes in Judea” (he never mentions them at all!), nor that the Sabbatical river proved for him the Sabbath's status as a natural law. In the same vein, Louis Ginzberg writes that Pliny “agrees with the Rabbis.” Ginzberg, Legends, vol. 4, 407. Such statements are not helpful, for they take as a given a course of transmission that can only be determined through careful investigation.

25. The earliest rabbinic sources which refer to the Sambatyon appear to be Midrash Tanḥuma (which in its core has been placed as early as 400), Midrash Bereshit Rabba (first half of the fifth century), and the Jerusalem Talmud (likewise generally dated to the first half of the fifth century). On the dating of these works see Stemberger, Günter, Einleitung in Talmud und Midrasch, Eighth Edition (Munich: C.H. Beck, 1992), 173, 275–276, 301Google Scholar. The date of the portion (23,8) of Pesikta Rabbati relevant for the present investigation is uncertain. Rivka Ulmer has dated this work in general to fifth/sixth century Palestine, noting that it contains much older material, but also indicates that chapters 21–4, often referred to as Midrash ‘aseret ha-dibrot (Midrash on the Ten Commandments), “did not initially belong to Pesiqta Rabbati,” rather representing a later insertion. Ulmer, Rivka, ed., Pesiqta rabbati: a synoptic edition of Pesiqta rabbati based upon all extant manuscripts and the editio princeps, vol. 1 (Atlanta: Scholars Press, 1997), xiii, xviiGoogle Scholar. In a personal communication (April 29, 2011), Ulmer indicated that she believes the reference to the Sambatyon in this section was a quotation from an early, pre-fifth century text from the Land of Israel. If so, this might make the Pesikta Rabbati passage the oldest rabbinic source on the river.

26. See Natural History 5.70–74 for Pliny's description of Judea.

27. Gruen, Erich S., Diaspora: Jews amidst Greeks and Romans (Cambridge, MA: Harvard University Press, 2002), 48CrossRefGoogle Scholar. The rest of this paragraph is also highly indebted to Gruen. For more on pagan attitudes regarding the Sabbath, see Robert Goldenberg, “The Jewish Sabbath in the Roman World,” in Aufstieg und Niedergang der römischen Welt (ANRW) 2.19.1 (1979), 414–447 and Schäfer, Peter, Judeophobia: Attitudes Towards the Jews in the Ancient World (Cambridge, MA: Harvard University Press, 1997), 82–92Google Scholar.

28. Tacitus, Historiae 5.4.3 (“Septimo die otium placuisse ferunt, quia is finem laborum tulerit; dein, blandiente inertia, septimum quoque annum ignaviae datum”).

29. Gruen, Diaspora, 48. The emperor Augustus, and the writers Strabo, Pompeius Trogus, Petronius, and Martial are among those who were party to this view. For more on this, see the brief discussion in Stern, vol. 1, 341, and in Schäfer, Judeophobia, 89–90 and 245–6, n. 63. On the possibility of actual Sabbath fasting among the Jews of ancient Rome, see Goldenberg, “The Jewish Sabbath,” esp. 439–441 and Ben-Ezra, Daniel Stökl, “Whose Fast Is It” in The Ways that Never Parted, ed. Becker, Adam H. and Reed, Annette Yoshiko (Tübingen: Mohr Siebeck, 2003), 270–1Google Scholar, where additional bibliography is also presented. While Goldenberg stops short of arguing that the Jews of Rome did in fact fast on Shabbat, Margaret Williams has more recently made precisely this claim. See Williams, Margaret, “Being a Jew in Rome: Sabbath Fasting as an Expression of Romano-Jewish Identity,” in Negotiating Diaspora: Jewish Strategies in the Roman Empire, ed. Barclay, John M. G. (London: T&T Clark International, 2004), 8–18Google Scholar. While Williams offers strong arguments in favor of this position, I do not think she proves the matter conclusively. A further possibility which she does not seem to consider is that this was the practice of a small group of Jews in the city (and perhaps beyond it as well) which Roman observers polemically applied to the entirety of the Jewish world. But whether real or imagined, this practice does not bear on the larger fact of negative Roman attitudes vis-à-vis the Sabbath.

30. The sources for the claim of the Jewish refusal to fight on the Sabbath are as follows: Agatharchides, as quoted in Josephus, Contra Apionem 1.209–10; Josephus, Antiquitates Judaicae 12.5–6; Cassius Dio, Roman History, 37.16.1–4; Strabo, Geographica 16.2.40, where the impression is created that the Temple fell into Roman hands on the Sabbath, described “as a day of fasting, when the Judaeans were abstaining from all work” (Stern, vol. 1, 302; see also his discussion on 307); Frontinus, Strategemata, 2.1.17; Plutarch, De superstitione 8.169c.

31. Gruen, Diaspora, 49.

32. Murphy, Pliny, 138.

33. Murphy, Pliny, 142.

34. Pliny, Natural History, 31.18.24 (“dirum est non profluere eos aspicere volentibus, sicut proxime larcio licinio legato pro praetore post septem dies accidit.”).

35. Pline l'Ancien, Histoire Naturelle, Livre XXXI, ed. Guy Serbat (Paris: Société d’édition “Les Belles Lettres,” 1972), 121.

36. Murphy notes elsewhere (Pliny, 142) that Pliny “represents rivers as means of travel, as connections between places.” Here we can perhaps say that the Sabbatical river forms a kind of narrative bridge linking two related, yet somewhat different, topics.

37. Pliny, Natural History, 5.73, Murphy, Pliny, 110. Pliny (5.15.71) specifically writes of how the Jordan loses its fresh waters in the Dead Sea: “aquasque laudatas perdit pestilentibus mixtas.”

38. Murphy, Pliny, 117. For Pliny, Murphy suggests, “The Essenes are to all intents already dead, having refused all interaction with the world” (ibid., 118).

39. Josephus, BJ 2.119–62.

40. Murphy, Pliny, 127. That the Essenes “stand in” for the Jews as a whole is made particularly clear by the element of gratuitous and specious chronology Pliny inserts here: “Thus through thousands of ages (incredible to relate) a race in which no one is born lives on for ever: so prolific for their advantage is other men's weariness of life.” (“ita per saeculorum milia [incredibile dictu] gens aeterna est in qua nemo nascitur: tam fecundia illis aliorum vitae paenitentia est.”). Pliny, Natural History 2, ed. Rackham, H., Loeb Classical Library 352 (Cambridge, MA: Harvard University Press, 1989] 5.15.73; 276–7Google Scholar.

41. Josephus, BJ 7.145; 349–50. See ibid. 7.132–162; 347–355, for an account of the triumphal procession celebrating the conquest of Judea.

42. Josephus, BJ 7.134; 347.

43. For multiple images of this carving, see Pfanner, Michael, Der Titusbogen (Mainz: Verlag Philipp von Zabern, 1983)Google Scholar, Tables 79–80, 85–87, esp. Tables 79.1, 86.4, and 86.6. These last three are figures 1–3, respectively.

For a description of the image, alongside a clear indication as to why it needs to be understood as depicting the god of the Jordan River, see Pfanner, 84, 90.

44. Pfanner, Der Titusbogen, 90. In BJ 7.136; 347 Josephus refers specifically to the presence in the triumphal procession of “images of their [i.e. the Roman] gods”; might this in part be a substitution by Josephus for geographical gods whom he would not have recognized?

45. Murphy, Pliny, 159. This passage in general showcases the prominence of rivers in Roman triumphal display and writing.

46. Murphy, Pliny, 155–6.

47. For the story of these two fighters, and the disaster for Babylonian Jewry which resulted from their actions, see Josephus, Jewish Antiquities Books 18–20, trans. Thackeray, H. St. J., Loeb Classical Library 433 (Cambridge, MA: Harvard University Press, 1997), 18.314–73; 180–209Google Scholar (also as AJ = Antiquitates Judaicae).

48. AJ 18.322.

49. AJ 18.323.

50. Later in this same narrative, at AJ 18.354–6, the brother Anileus is shown to have fought on the Sabbath, and at 18.374, Josephus refers to Jews as “men that despise dangers, and [who are] very ready to fight upon any occasion.”

51. Trevor Murphy has noted that in the Natural History rivers are often described as “famous” (Murphy, Pliny, 145), a practice characteristic of other Roman authors. Josephus, it can be said, seems here to be trying his best to achieve similar renown for the Sabbatical river. On rivers in the Natural History in general, see ibid., 138–148.

52. It is of course striking to encounter Josephus forging geography, in light of his avowed commitment in the introduction to the Jewish War to historical and geographical truth, especially at 1.22 and 30. This may well be linked to the particular conditions in which Josephus wrote the seventh book of this work (site of the Sabbatical river passage) which, as we have seen, took place somewhat later. On Josephus and truthful reporting, see also Shaḥar, Josephus Geographicus, 190–3. Veltri briefly suggests that the Sambatyon's behavior roots the Sabbath in nature. See Veltri, “‘The East,’” 261. It is also true that some have identified Josephus's Sabbatikon with the Nahr-el-Arus of northern Lebanon. In his Physical Geography of the Holy Land (Boston: Crocker and Brewster, 1865), Edward Robinson notes that the stream in his day typically runs every third day, though with some seasonal variation, this not stopping the local population from describing it as the Nahr Sebty (Seventh-Day River) which, as Muslims, they associate with Friday. See Robinson, Physical Geography, 359–60; Neubauer, “Where Are the Ten Tribes?” 20. I have not been able to confirm the accuracy of this discussion. It is of course not impossible that Josephus was aware of such an intermittent stream and made use of it in his report. But even if we assume that this is the case, Josephus would still have needed to craft a description of the river which linked it to the Sabbath (and not to every third day), thus mobilizing it to do ideological work for him. Identification, to the extent that it is at all possible (and one is inclined to agree with Neubauer that it is, in this case, a waste of time), marks merely the beginning of, not a substitution for, analysis. Any local tradition of the sort reported by Robinson, to the extent it actually existed or exists, is probably more likely the result of later exposure to Josephus's narrative than a relic of a tradition to which Josephus was himself exposed. In addition, an early edition of Baedeker's Palästina und Syrien refers to the intermittent stream Fuwâr ed-Dêr, adding that “dies ist der Sabbatfluss der Alten, an welchem Titus vorbeizog,” followed by reference to Josephus's Jewish War. See Baedeker, K., ed., Palästina und Syrien: Handbuch für Reisende (Leipzig: Karl Baedekker, 1880), 439Google Scholar. Once again, the identification is problematic for all the above reasons.

53. His official title was: “Legatus Augusti pro praetore provinciae Iudaeae.” Paulys Realencyclopädie der classischen Altertumswissenschaft, 2nd Series, 12th half-volume (Timon-Tribus), col. 1376.

54. See Steinsaltz, Adin, Talmud Bavli: Massekhet Sanhedrin (Jerusalem: The Israel Institute for Talmudic Publications, 1997), 281Google Scholar (though he nevertheless punctuates the first name as Turnus); also, Paulys Realencyclopädie, ibid.: “In hebräischen Texten ist der Name meist in der Form טוּרְנוֹס רוּפוֹס geschrieben . . . Es ist sehr wahrscheinlich, daß man an Tyrranus anspielen wollte.”

55. “Tyranus Rufus” translates as “red tyrant,” nicely reflecting the rabbinic associations between Rome and Edom and its ruddy progenitor Esau. On this association, see the classic study of Cohen, Gerson D., “Esau as Symbol in Early Medieval Thought” in Jewish Medieval and Renaissance Studies, ed. Altmann, Alexander (Cambridge, MA: Harvard University Press, 1967), 19–48Google Scholar. More recent scholarship suggests that the Jewish equation between Edom and Rome in fact stems from far earlier than Cohen had supposed. For a summary of the evidence to this effect, see Feldman, Louis, “Remember Amalek!” (Cincinnati: Hebrew Union College Press, 2004), 62–67Google Scholar. On the literary character of the Akiva/Rufus dialogues, see Schäfer, Peter, “Rabbi Aqiva and Bar Kokhba,” in Approaches to Ancient Judaism vol. 2, ed. Green, William Scott (Chico, CA: Scholars Press, 1980), 120–1Google Scholar, and, more extensively, in “Rabbi Aqiva und Bar Kokhba,” in Schäfer, Peter, Studien zur Geschichte und Theologie des rabbinischen Judentums (Leiden: Brill, 1978), 65–121, esp. 95–100CrossRefGoogle Scholar. Referring to the Akiva/Tyranus Rufus debates, Schäfer notes that “these are without doubt literary topoi with different speakers, whose historical background is to be found in the Gentile/Jewish controversy” (Schäfer, Approaches, 120). I agree as to the literary character of these sources, but wish to emphasize that the rabbinic tradition rather early on fixed and limited polemical discussion about the Sabbath involving the Sambatyon to Akiva and Tyranus Rufus, a “fact” important for the rabbinic development of this myth.

56. Loewenthal, “La storia del fiume Sambation,” 657: “In queste testimonianze il Sambation diventa un riferimento fondamentale, quasi ovvio e comunque a quanto pare piuttosto comune: Rabbi Aqiva sembra dare per scontato che persino un governatore romano sappia come ‘funziona’ il Sambation” (Italics mine). Benite, Ten Lost Tribes, 80: “It is significant that the river is mentioned in the context of a fictional dialogue with a Roman governor, which suggests that, at this period, the Sambatyon was not uniquely Jewish but belonged in a cultural realm shared by Jews and Romans.”

57. This dialogue is reported in the following texts: Tanḥuma, Ki-Tissa, par. 33; Bereshit Rabba, Bereshit, par. 11:5, to Genesis 2:3 (ed. Theodor-Albeck, 1:92–3); Pesikta Rabbati 23,8; B. Sanhedrin 65b; Sefer she'iltot Rav Aḥai Gaon, She'iltot Bereshit, She'ilta Aleph, 1:4,12 (Vilna: Garber, 1908) Midrash ‘aseret ha-dibrot 23, in Oẓar Midrashim, ed. Eisenstein, Judah David (New York: J. D. Eisenstein, 1915), 455Google Scholar; Yalkut Shimoni, Bereshit, Bereshit, pis. 17, to Genesis 2:3 (ed. Hyman-Lerrer-Shiloni, 1:53); Yalkut Shimoni, Vayikra, Kedoshim, pis. 617:34–38, to Leviticus 19:31 (ed. Hyman-Lerrer-Shiloni, 2:615); Yalkut Shimoni, Devarim, Shoftim, pis. 918:25–29, to Deuteronomy 18:11 (ed. Heyman/Shiloni, 1:376); Midrash Aggada Shemot, par. Vayakhel; Sefer ve-hizhir: le-seder shemot, par. Vayakhel, fol. 122r–v (ed. Israel M. Freimann, 24–34); and is also referred to, among others, by David Kimḥi in his commentary on Genesis 2:3. The Italian sage R. Tzidkiya ben Avraham ha-Rofe (ca. 1210–ca. 1275) does not refer to the dialogue directly in his Shibbule ha-leket, but does mention manna, ba‘ale zikhurim (necromancers), and the Sambatyon as proofs of the Sabbath, as in many versions of it. See Shibbule ha-leket, Inyan Shabbat, sign 126. On the likely tenth-century and Palestinian Midrash ve-hizhir, which extends from Exodus 8:16 through Numbers 5:11, see Stemberger, Einleitung, 307.

58. The two exceptions known to me are the Midrash 'aseret ha-dibrot and She'iltot, where the Sambatyon is placed between the deployment of the case of manna and the raising of dead spirits, perhaps suggesting somewhat lesser importance. Along the same lines, only in the She'iltot does Tyranus Rufus explicitly reject the rabbinic sage's argument, stating that “with regard to this river I don't believe you, since I do not know where it is located” (אף בנהר הזה איני מאמינך שאיני יודע את…” “מקומו), a direct challenge which Akiva does not parry. (Sefer she'iltot, She'iltot Bereshit ). It is as if this source, one of the latest which features this dialogue (it dates to the first half of the eighth century), has itself come to doubt the value of the Sambatyon argument and certainly to have forgotten (if my claim is correct) its Roman origin.

59. Linguistic analysis also points in this direction, for the name Sambatyon in this period was hardly limited to the mythical river. In a chapter entitled “The Sambathions,” the editors of the Corpus Papyrorum Judaicarum note that in Greco-Roman antiquity “persons, buildings, deities, and institutions in various countries were called by names derived from the Hebrew word ‘Sabbath’.” Corpus Papyrorum Judaicarum, 45; the chapter as a whole runs from 43–87. From the early Roman period, the standard forms for such Sabbatical names—Sambathyon and Sambatyon (or close variants thereof) became a remarkably common name for men and women, Jew and more often non-Jew (in Egypt and elsewhere in the diaspora) alike. Remarkably, the Corpus makes no reference to the Sambatyon river as an application of this form. But based on what it reveals, it should come as no surprise that this is the name that emerged for the Sabbatical river. It seems clear as well that the Palestinian rabbis who spoke and wrote about it (for the preponderance and certainly the earliest sources on the river are Palestinian) deliberately selected a name that would have currency in the non-Jewish world. For we have little trace of Sabbatical names in Palestine proper from antiquity (ibid., 44). Interestingly, the personal name Sambatyon declines in third, and disappears in fourth-century Egypt, precisely the period in which traditions about the Sambatyon river first begin to be recorded (ibid., 55). For more on the use of this and very similar-sounding names, see now Ilan, Tal, Lexicon of Jewish Names in Late Antiquity (Tübingen: Mohr Siebeck, 2002–2008), Part I, 213; Part III, 148–154, 186–8Google Scholar.

60. Tanḥuma, Ki-Tissa, par. 33; Bereshit Rabba, Bereshit, par. 11:5; Pesikta Rabbati, 23,8; Midrash 'aseret ha-dibrot, 23; Yalkut Shimoni, Bereshit, Bereshit, pis. 17; Sefer ve-hizhir: le-seder shemot, par. Vayakhel, fol. 122r–v.

61. In the pithy account of B. Sanhedrin 65b, the grave of Tyranus Rufus's father is offered as proof, since it does not bellow with smoke on the Sabbath (“קברו של אביו יוכיח, שאין מעלה עשן בשבת”). This tradition is also mentioned in the second and third above-cited passages from Yalkut Shimoni, as well as that from Midrash 'aseret ha-dibrot 23 (in Eisenstein, Oẓar midrashim) and the She'iltot (albeit in more expansive form).

62. One is here reminded of Seneca's statement (preserved, or at least reported, by Augustine) that “the customs of this detestable race (i.e. the Jews) have become so prevalent that they have been adopted in almost all the world. The vanquished have imposed their laws on the conquerors.” See Augustine, City of God, trans. Bettenson, Henry (London: Penguin Books, 2003) 6.10; 252Google Scholar. Pace Seneca, the rabbis could not impose the Sabbath on the Romans, except perhaps in the afterlife. One wonders if their doing so reflects exposure to Roman anxiety of the sort to which Seneca gives voice.

63. The rabbinic effort to subvert Tyranus Rufus in particular is furthermore reflected in the tradition (preserved in B. Avodah Zara 20a and B. Nedarim 50b) that his wife eventually converted and married, of all sages, Rabbi Akiva himself.

64. Bereshit Rabba, Bereshit, par. 11:5 (“מה יום מימים”).

65. On Josephus's positive regard for the Sabbath, see Weiss, Herold, A Day of Gladness: The Sabbath Among Jews and Christians in Antiquity (Columbia: University of South Carolina Press, 2003), 65–85Google Scholar. Josephus speaks in particular of the Sabbath as a day of gladness in BJ 5.228–233 and AJ 3.151–178 (these passages are briefly discussed by Weiss, 69).

66. Though I have no doubt that Josephus primarily has his Greco-Roman audience in mind here, Williams's argument as to the practice of fasting on the Sabbath, if correct, raises the possibility that Josephus may to some degree be responding to his fellow Roman Jews. Noting that the early Roman Jewish community was largely composed of prisoners of war, Williams argues that its alledged Sabbath fasting reflected communal mourning over the fall of Jerusalem, said to have occurred on this day (Williams, “Being a Jew,” 16). Josephus's description of the Sabbatical river might thus be intended to encourage a less solemn and more celebratory Sabbath observance.

67. Josephus, BJ 7.96–99; 336–7.

68. Josephus emphasizes the significance of the number seven for Jews a bit further on in Book 7, in the course of his depiction of the Menorah. This ritual object had seven branches, he notes, “indicating the honour paid to that number among the Jews” in BJ 7.149; 350–1. See below, n. 91, for further discussion of the significance of the number seven in this context.

69. “. . . Moses introduced new religious practices, quite opposed to those of all other religions. The Jews regard as profane all that we hold sacred; on the other hand, they permit all that we abhor.” Tacitus, The Histories, 5.4.1, cited in Stern, vol. 2, 25.

70. 2 Kings 24:9, 2 Chronicles 36:9.

71. AJ 10.100, 10.136–7; cf. this latter citation with 2 Kings 25:4 and Jeremiah 39:4. On these two cases, see Feldman, Louis H., “The Concept of Exile in Josephus,” in Exile: Old Testament, Jewish, and Christian Conceptions, ed. Scott, James M. (Leiden: Brill, 1997), 156–8Google Scholar.

72. This would be especially true if, as I argued above, the legend is in fact of Roman origin.

73. Weiss notes that “For [Josephus] it was of the essence that on the Sabbath no work, not ‘even the most innocent of acts’ (BJ 2.456) be done”; indeed, he can hardly “refer to the Sabbath without mentioning [this] requirement. Weiss, A Day, 84–5. For another case of violating the Sabbath in order to defend it, see Shaye Cohen, J. D., “Hellenism in Unexpected Places,” in Hellenism in the Land of Israel, eds. Collins, John J. and Sterling, Gregory E. (Notre Dame, IN: University of Notre Dame Press, 2001), 229Google Scholar.

74. See Sifre Devarim, ‘Ekev, pis. 42, to Deuteronomy 11:14 (ed. Finkelstein, 88–9); Sifra, Beḥukotai, par. 1:1, to Leviticus 26:4 (ed. Weiss, p. 110b); and Vayikra Rabba, Beḥukotai, par. 35:10, to Leviticus 26:4 (ed. Margoliot,) all of which regard God's promise of “rain in its season” (Leviticus 26:4) as having been fulfilled in the days of Queen Shelamzion, when rains fell every Friday night (Sifre Devarim 42 reads as follows: “מלילי שבת ללילי שבת כדרך שירדו בימי שלמצו המלכה”; the Sifra has: “…בלילות שבתות מעשה בימי שמעון בן שטח בימי שלומציון גשמים יורדים מלילי שבת ללילי שבת…”; Vayikra Rabba is very similar to the Sifra: בלילי שבתות מעשה בימי שמעון בן שטח ובימי שלמצה המלכה…” “…שהיו גשמים יורדים בלילי שבתות). A similar tradition, reported in B. Ta‘anit 22b, refers to rainfall on Tuesday and Friday nights. These sources are discussed and compared in Ilan, Tal, Silencing the Queen (Tübingen: Mohr Siebeck, 2006), esp. 36–7, 47, 53, 69, 124–5Google Scholar. Ilan translates Sabbath night (ליל שבת) as Saturday night, and hence writes of “rains [which] fell from Saturday night to Saturday night” (37). More likely, either Friday night is intended—and with it the notion that rains fell regularly on Friday nights—or Friday night to Saturday night—in which case, rains fell regularly throughout the duration of the Sabbath. This is implied by the adjacent passage in B. Ta‘anit 22b, on how the rain at night did not interfere with the construction of Herod's temple; rain on the Sabbath eve or Sabbath would similarly not interfere with labor. See also B. Pesaḥim 112b, where these two nights are again grouped together and Adin Steinsaltz's explanation as to why (Steinsaltz, Adin, ed., Babylonian Talmud: Tractate Pesahim [Jerusalem: The Israel Institute for Talmudic Publications, 1997], 483Google Scholar). In drawing this connection I am not suggesting that Josephus was personally acquainted with these sources, which were set down in writing after his time (though Ilan does argue that this rain tradition is ancient; Ilan, Silencing, 124 et passim). In any case, Josephus, as is well known, despised Queen Shelamzion (Ilan, Silencing, 70). My point is simply that what Josephus's Sabbatical river and the rabbinic traditions of the rains have in common is a link between the Sabbath and nourishing waters. In addition, Pirke Avot (M. Avot) 5:6 reports the well which accompanied the Israelites in the desert (later associated with Miriam) as one of ten items created on the first Sabbath eve.

75. Josephus, BJ 7.96; 336–7.

76. Josephus, BJ 7.109; 338–9. On this passage, see Chapman, Honora Howell, “Spectacle in Josephus's Jewish War,” in Flavius Josephus and Flavian Rome, eds. Edmondson, J. C., Mason, Steve, and Rives, J. B. (Oxford: Oxford University Press, 2005), 308–9Google Scholar.

77. Weiss, A Day, 84. Weiss is the only other scholar known to me to have argued that Josephus's disagreement with Pliny and the rabbis regarding the behavior of the Sabbatical river most likely stemmed from agenda and not from an alternate source. He is also unique, so far as I am aware, in (briefly) investigating the context in which Josephus introduces it—Titus's departure from Judea and passage through Syria. But according to Weiss, for reasons which are neither clearly nor persuasively argued, the presence of the Sabbatical river at this point in the narrative reflects nature's accompaniment of the Syrian cities in celebrating Titus's triumph: “While Syrian cities display expensive shows in his honor, nature joins with a display of life-giving waters of an eschatological order” (Weiss, A Day, 84). For reasons that shall become apparent, I find this reading untenable.

78. Aryeh Kasher noted that Josephus's “entire description . . . follows the pattern of an ancient tragedy—the persecutions of Antiochus IV Epiphanies.” See Kasher, Aryeh, The Jews in Hellenistic Egypt (Tübingen: J. C. B. Mohr, 1985), 299Google Scholar. More recently, Joseph Sievers has compellingly emphasized the substantially fictional and, indeed, typological character of Josephus's account in Sievers, Joseph, “What's in a Name? Antiochus in Josephus' ‘Bellum Judaicum,’” Journal of Jewish Studies 56 (2005): 40–46CrossRefGoogle Scholar. Sievers's argument offers powerful support for our treatment of Josephus's presentation of the Sabbatical river, an aspect which Sievers does not address in his discussion.

79. BJ 7.43–5; 318–321.

80. BJ 7.47–53; 320–323.

81. Weiss notes Josephus's discussion of the prohibition of the Sabbath at Antioch but does not connect it in any way with the later discussion of the Sabbatical river. See Weiss, A Day, 80–81.

82. BJ 7.62; 324–5.

83. Weiss, A Day, 84. If Margaret Williams is correct in arguing that the Roman Jews of antiquity fasted on the Sabbath out of mourning for the fall of Jerusalem on this day (Williams, “Being a Jew,” 16–7), then Josephus's narrative could also be seen to encourage a more positive conception of Shabbat as a day which preserves Judaism rather than one which leads to its destruction.

84. The fact that Titus was no longer alive presumably made it easier for Josephus to create the story of his encounter with the Sabbatikon.

85. Michael Pfanner argues that the arch must have been erected after Titus's divination in late 81 or early 82 and shows here how Domitian seems to have made use of it to advance his own stature, probably during the early portion of his reign. He is certainly correct to insist that the arch could only have been completed after this point (Pfanner, Der Titusbogen, 91–2), though it seems perfectly possible that the arch was already under construction when Titus died in late 81.

86. Table 56.45 in Pfanner, Der Titusbogen. Pfanner himself offers further indications as to the striking degree of detail lavished upon the miniature group of Vespasian, Titus, and especially Domitian (see ibid., 91). It is so especially strange to find Domitian more decked out than the divine Titus to whom the arch was supposedly erected such that Pfanner speculates that Domitian himself may have played a role in this aspect of the arch's artistic program. In this regard, he refers the reader to a passage in Suetonius's “Life of Domitian” in which the emperor is alleged to have “boasted in the Senate that he himself had conferred power on his father and brother, while they had merely returned it to him.” Suetonius, vol. 2, trans. J. C. Rolfe, Loeb Classical Library 38 (Cambridge: Harvard University Press, 1992), 8.13.1; 366–7 (translation modified). (“Principatum vero adeptus neque in senatu iactare dubitavit et patri se et fratri imperium dedisse, illos sibi reddidisse . . . ”). It is worth noting in addition that Suetonius explicitly links this declaration to the beginning of Domitian's rule (“When he became emperor . . . ”), possibly constituting further evidence that the Arch of Titus was a project of his very early reign, a supposition which Pfanner puts forward as highly reasonable though improvable (ibid., 92: “sehr wahrscheinlich . . . freilich unbeweisbar”).

87. BJ 7.152

88. The third possibility, that Domitian had actually had such a prominent role in the original procession and that this was recalled independently by both Josephus and the arch seems to be the position of Martin Goodman, “The Fiscus Judaicus and Gentile Attitudes to Judaism in Flavian Rome,” in Jonathan Edmondson, Steve Mason, and James Rives, eds., Flavius Josephus and Flavian Rome (New York: Oxford University Press, 2005), 171. While not impossible, this too appears highly unlikely, given that Domitian played no role whatsoever in the fighting in Judea. On this, see also Cohen, Josephus, 87. Further indication that Josephus is here simply “copying out” what he saw displayed on the arch comes from the observation that only in this one specific passage is Domitian at all mentioned; every other Flavian reference in Josephus's account of the triumph is to Vespasian and Titus alone. Cf. BJ 7.119, 121, 124, and 128.

89. Shaye Cohen points out Josephus's flattery of Domitian throughout Book 7, but does not draw specific attention to the similarity of his description here to the images on the arch. Cohen, Josephus, 87. On page 89, Cohen appeals for further study of the relationship of BJ Book 7 to Books 1-6 and notes that “in particular we should like to know when BJ 7 was written, and why Josephus wrote it.” It is hoped that the observations here offer a modest contribution in this direction.

90. For a discussion which similarly links Josephus's discussion of Antioch to a diaspora ideology, see Shaḥar, Josephus Geographicus, 262. Shaḥar notes in particular that reference to the oikoumene precedes the commencement of the historian's treatment of the Antioch disturbances. Josephus opens by observing that “The Jewish race, densely interspersed among the native populations of the entire oikoumene, is particularly numerous in Syria, where intermingling is due to the proximity of the two countries. But it was at Antioch that they specially congregated . . . ” (The Jewish War, 319; translation modified). Though located near its border, Antioch is nevertheless the locus (pun intended) classicus of the diaspora for Josephus.

91. In this regard, it is striking that Josephus later describes Moses as “the best of all the Hebrews,” noting that “Abraham was his ancestor of the seventh generation.” These comments immediately follow his salvific account of the origin of Moses' name: “Mo” from the Egyptian term for water, “Uses” for “such as are saved out of it.” See Josephus, AJ 2.9.6. The notion of Moses as seven generations after Abraham was widespread in antiquity. See Feldman, Josephus's Interpretation of the Bible, 378, for further examples. On sabbatical notions of Jewish redemption in general, see for example B. Sanhedrin 97a, where there is reference to “the seven-year cycle at the end of which the son of David will come.” According to 1 Chronicles 2:15 (though not 1 Samuel 16:1–23 and 17:12–14), David himself was the seventh son of Jesse, a numerical scheme directly followed by Josephus (6.158–165). The Church father Augustine refers to “the end of the seven days of time” in his homily on Cain and Abel. See Hallo, William W., Ruderman, David B., and Stanislawksi, Michael, eds., Heritage: Civilization and the Jews (New York: Praeger, 1984), 79Google Scholar. It is also worth noting that in the book of Ruth, Boaz who married Ruth is seven generations removed from Peretz, the offspring of Judah and Tamar (and Ruth herself is described at 4:15 as equal to seven sons). For another indication in Josephus of an implicit expectation of Jewish revival, this time based on the book of Daniel, see Josephus, AJ 10.209–210. To be sure, the Antiquities were written several decades after The Jewish War, though it has been noted that “the outlook and concerns of BJ 7 seem closer to those of AJ and Vita” (Cohen, Josephus, 89). Might Tacitus thus be parodying Jewish sabbatical messianism when he writes (The Histories 5.3.2) of Moses “[discovering] abundant streams of water. This relieved [the Hebrews], and they then marched six days continuously, and on the seventh day seized a country…”? Finally, the redemptive character of Josephus's Sabbatikon recalls the desert wadis of Psalm 126, whose periodic flow is a metaphor for Jewish redemption: “Turn our captivity, O LORD, as the streams in the Negev” (Psalm 126:4). To be sure, there is nothing sabbatical or sevenfold in this psalm, though it is interesting to note that in Jewish liturgy it prefaces “Grace after Meals” on the Sabbath and Holidays. I am, however, not certain how far back this association can be traced.

92. Josephus, BJ 7.44; 319–321.

93. Also known as Marcus Julius or Herod Agrippa II.

94. See his long speech to the war-crazed Judeans in BJ 2.345–401.

95. BJ 3.57; 7.43.

96. See above, n. 38.

97. Murphy, Pliny, 142. Comparing Josephus's earlier (BJ 3.51–58) and later (here) descriptions of Agrippa's kingdom and noting his failure to mention Arcea in the former, some scholars have surmised that this city was a late addition to his kingdom. On this possibility, see Cohen, Josephus, 87 n. 9. Whether or not this was the case (and the evidence in fact seems rather weak), it contributes little to our appreciation of Josephus's reference to the Sabbatikon. Literary analysis of the two passages can much better illuminate their significance. In the first, Josephus notes that Agrippa's “country begins at Mount Libanus, and the fountains of Jordan, and reaches breadthways to the lake of Tiberias; and in length is extended from a village called Arpha, as far as Julias.” This is structurally quite similar to his later account, which replaces the Jordan and lake of Tiberias with the Sabbatikon, and Arpha and Julius with Arcea and Raphanea. Thus, if in Book 3 Josephus describes Agrippa's territory from the perspective of Judea, emphasizing the former as the source of the latter's greatest river, the Jordan—appropriately concluding that he has, “with all possible brevity, described the country of Judea, and those that lie round about it”—in Book 7 his interest is in Agrippa's kingdom as lying on the way to Syria, and the Sabbatikon as symbolizing the move from Judea to the diaspora. The second passage's inversion of the first offers further indication as to the geographically artificial, i.e. imagined, status of the Sambatyon. This is also nicely illustrated by the fact that it runs between what appear to be two cities and not, as would be expected for a river (e.g. the Jordan), from spring to sea. This further redounds to the Sabbatikon's political significance of heralding the shift of Jewish fortunes from one civic center to another, i.e. Jerusalem to Antioch. One might in addition wonder if the very names of these cities are intended as symbolic: “Arcea” evokes the Hebrew ארכי ('arkhi), “rule, authority” (deriving ultimately from the Greek ἀρχή, whereas “Raphanea” rings of the Hebrew רפיון (rifyon): “weakness, powerlessness.” It thus appears possible to read the “source” and “mouth,” so to speak, of Josephus's Sabbatikon as indicative of its narrative significance.

98. BJ 7:111.

99. BJ 7:112.

100. BJ 7:113.

101. Shaḥar, Josephus Geographicus, notes that the diaspora takes on a positive moral aspect when Josephus mourns the ruin of Jerusalem. But it is specifically Antioch, as symbolic of the diaspora as a whole, which appears for Josephus to represent the critical counter-narrative to Jerusalem. Why, of all diaspora locales, does Antioch assume this role? Might Josephus find in its very name (so evocative of Antiochus IV) a striking contrast to the Hasmonean story of Jewish independence in Judea? Antioch's special significance for Josephus might explain why alone in the case of this city does he give ample voice to the diaspora Greek-Jewish tension and violence that he elsewhere attempts to downplay. It has been noted, for example, that Josephus does his best “not to spoil the moderate image which he portrayed concerning the tolerant attitude of the Hellenistic cities beyond the borders of Judaea.” Kasher, Aryeh, Jews and Hellenistic Cities in Eretz-Israel (Tübingen: Mohr, 1990), 286Google Scholar; see also 285. Certainly, his de-emphasizing the Greek/Jewish clashes in other diaspora locales during the Great-Revolt while highlighting those in Antioch only serves to emphasize the Antioch/Jerusalem contrast all the more.

102. The phrase is Veltri's, “‘The East,’” 261.

103. The attempt to root and replace the lost Temple menorah in a cosmic likeness may well represent an important parallel to the case of the Sambatyon. In describing the sixth-century Palestinian synagogue poet Yannai's partially-preserved piyyut for Shabbat beha‘alotkha (Numbers 8:14), Steven Fine writes that this “composition argues strenuously that although the menorah itself is lost, God's heavenly menorah—symbol of Divine love for Israel and of Israel herself—is built into the natural realm and is brightly ablaze” (italics mine). See Fine, , Art and Judaism in the Greco-Roman World (Cambridge: Cambridge University Press, 2005), 161Google Scholar. Both the Sambatyon and the menorah, then, offer reassurance—natural or miraculous—for Jewish existence in the face of political tragedy.