The Mawzaʿ exile

The hardening began under Imam al-Mutawakkil Ismāʿīl (1644–1676). After the affair of Suleimān Jamāl, a Sabbatean agitator whom the imam judged to be usurping authority and had executed, Jewish property was confiscated and Jews were barred from owning land; here too belong the 1667 decrees on headgear and distinctive dress. The expulsion was carried out by his successor al-Mahdī Aḥmad ibn al-Ḥasan (1676–1681).

In 1677–79 synagogues were destroyed. The great Kenisat al-ʿUlamāʾ, "the synagogue of the sages," inside the walls of Sana'a was turned into a mosque and named Masjid al-Jalāʾ—"the mosque of the expulsion." 10 August 1679 = 2 Rajab 1090 AH is the expulsion decree; one of very few dates in Yemenite Jewish history known to the day.

Then the driving on foot of almost the whole of Yemenite Jewry to Mawzaʿ, on the Tihāma coastal plain some twenty kilometres from Mocha: eleven days' walk from Sana'a, downhill, out of a mountain climate at 2,250 metres into the heat of the lowland. What killed was not massacre but climate, distance and destitution.

In 1680 (in some districts 1681) return was permitted, and the reason was economic: Muslim subjects complained of the lack of farm implements—only Jewish smiths made them. Jews were not readmitted to the walled city and were told to build outside it: so arose Qāʿ al-Yahūd.

Numbers and losses. The chronicles say two-thirds died; one account says ten in a hundred returned; another records some eighty dead on a single desert stage of one journey. Jacob Saphir, collecting oral testimony on the spot in 1859, wrote that "two-thirds of them succumbed and perished." Scholarship treats "two-thirds" as a chronicle topos and cannot verify it—but no one disputes that the losses were catastrophic. The figure should be printed as reported. What was lost materially: Torah scrolls, volumes of the Talmud, biblical codices and midrashic manuscripts were burned. That is why the surviving pre-1679 Yemenite manuscript stock is so thin—and why the Cairo Genizah says more about medieval Yemen than Yemen does.

Four corrections. The expulsion was not universal: the communities of the far east of the country—Nihm, al-Jawf, Khawlān—were not expelled, because their local tribal patrons refused to execute the imam's order; dhimmī status in Yemen ran through tribal protection (jiwār), which sometimes overrode the imamate. "A year in exile" is a Sana'a-centric simplification: the dates vary by district—the expulsion is 1679, the returns 1680–81, and some communities never re-formed at all. The cause was not "the Jews' Sabbateanism"—that is the imam's rationalisation; the scholarly account is Zaydī doctrinal hardening, the imam's claim to purify the Arabian Peninsula of non-Muslims on the authority of the hadith that two religions do not coexist in Arabia, reinforced by the political shock of the Sabbatean agitation of 1666–67 and by the consolidation of Qāsimid power after the expulsion of the Ottomans in 1635. And Mawzaʿ was neither a prison nor an island: it was an existing Tihāma town.