¹ The most comprehensive documentation of these beliefs and practices remains Drower, E. S., The Mandæans of Iraq and Iran (Oxford, 1937)Google Scholar, which has yet to be surpassed.

² The standard grammar of their written language remains Nöldeke, T., Mandäische Grammatik (Halle, 1875)Google Scholar. No dictionary of the written language existed until E. S. Drower and R. Macuch published A Mandaic Dictionary (Oxford, 1964), which remains the only such reference. The spoken language remained largely undocumented until Macuch published his Handbook of Classical and Modern Mandaic (Berlin, 1965). The same author produced two additional text collections, Neumandäische Chrestomathie mit grammatischer Skizze, kommentierter Übersetzung und Glossar (Wiesbaden, 1989) and (posthumously with the editorial intervention of G. Dankwarth) Neumandäische Texte im Dialekt von Ahwāz (Wiesbaden, 1993). The first and only full-length grammatical treatment of any dialect of the spoken language remains C. G. Häberl, The Neo-Mandaic Dialect of Khorramshahr (Wiesbaden, 2009). Most recently, H. Mutzafi has addressed the lexicon in his Comparative Lexical Studies in Neo-Mandaic (Leiden and Boston, 2014).

³ All of the chief Mandæan scriptures, and quite a few of their other manuscripts, have been translated into one or more languages of scholarship, although only a few have benefited from critical editions. The principal translations are those of M. Lidzbarski, Das Johannesbuch der Mandäer, vols. 1 and 2 (Giessen, 1915 and 1922) and Ginzā: der Schatz oder das große Buch der Mandäer (Göttingen, 1925). Lidzbarski also produced an edition and translation of an abbreviated portion of the liturgy in his Mandäische Liturgien: mitgeteilt, übersetzt und erklärt (Berlin, 1920), and E. S. Drower published an edition of a complete manuscript of the same with her translation as The Canonical Prayerbook of the Mandæans: Translated with Notes (Leiden, 1959).

⁴ All examples from Mandaic and other languages have been transcribed phonemically. Transcriptions of the literary language do not indicate post-vocalic fricative allophones of the b/g/d/k/p/t stops. At an indeterminate point in its history, the sound system of Mandaic was restructured, resulting in the emergence of a new series of fricative phonemes. These new phonemes are indicated in transcriptions of the spoken language.

⁵ Nöldeke (Mandäische Grammatik, pp. xix–xxiv) divides the history of the written language into two phases, Old and Young Mandaic. In the 1960s, Macuch introduced an alternative periodisation, consisting of ‘Classical’, ‘postclassical’, and ‘Neo-Mandaic’, implying the existence of a standardised register of the language, as elaborated in dictionaries and grammars along the lines of other classical languages, and measuring all manifestations of this language according to their adherence to this register. There is no evidence that Mandaic was ever elaborated in this manner. While Macuch's new periodisation still remains popular with scholars, in light of its deficiencies I have elected instead to distinguish between ‘literary’ (lm) and ‘spoken’ (sm) registers of the language. The boundary between these two registers is not firm; literary forms may appear in speech, and vernacular forms abound in the literature.

⁶ Mutzafi (Comparative Lexical Studies, p. 194) proposes that it emerged from a compound of be(ṯ)- ‘place of’ and lm *kəbāšā ‘subduing’. Another possibility is an adverbial phrase, *be-kbāšā (d-šāmeš) ‘at the subduction (of the sun)’, along the same lines as the frequently attested bə-nehgā ‘at dawn’, and subsequently lexicalised as a noun.

⁷ Macuch, Handbook, p. 511 furnishes this word, but in his Neumandäische Texte he reports that his learned informant Salem Choheili consistently uses the word šuši. As this word employs the lm plural morpheme -i rather than any of the sm plural morphemes, it is evidently a classicism.

⁸ While this word is evidently cognate with Syriac šāʿtā and Arabic sāʿah, the expected form would be *šātā, not šitā, as no regular Mandaic sound rule could produce the latter form. Instead, šitā looks very much like a Babylonian reflex of the same word, from Proto-Semitic *sāʿt- > *šēʿt- > *šit-. It was therefore possibly borrowed from that source, together with šuššu, even if its etymon is unattested within the corpus of Akkadian.

⁹ Macuch, Handbook, p. 241:1 and n. 230. Once again, the fact that the entire voluminous corpus of Mandaic literature lacks so basic a word as “tomorrow” proves the limits of the linguistic perspective offered by textual corpora.

¹⁰ Thursday is also known as yum Hibel Ziwɔ “the day of (the saviour spirit) Splendid Hibel”.

¹¹ Drower, Mandæans, p. 204. I use the term ‘epagomenal’ here with its usual reference to days that do not belong to any month but are nonetheless annually added to the calendar, in order to distinguish them from other less frequent or regular forms of intercalation, such as leap days.

¹² Taqizadeh, S. H., ‘An Ancient Persian Practice Preserved by a Non-Iranian People’, Bulletin of the School of Oriental Studies, University of London, 9, no. 3 (1938), pp. 603–619CrossRefGoogle Scholar.

¹³ Stern, S., Calendars in Antiquity: Empires, States, and Societies (Oxford, 2012), p. 183CrossRefGoogle Scholar.

¹⁴ Blois, F. de, ‘The Persian Calendar’, Journal of Persian Studies, 34 (1996), pp. 39–54Google Scholar [41]. The translation is his.

¹⁵ F. de Blois, ‘The reform of the Zoroastrian calendar in the year 375 of Yazdgird’, in Ātaš-e dorun, the Fire Within: Jamshid Soroush Soroushian Memorial Volume ii, (ed.) C.G. Cereti and F. Vajifdar (Bloomington, il, 2003), pp. 139–145.

¹⁶ All dates prior to 15 October 1582 are provided according to the Julian calendar, unless otherwise noted. According to the Gregorian calendar, the vernal equinox and transit of the sun to Aries both occurred on 21 March 1006 ce.

¹⁷ Drower, Mandæans, p. 87.

¹⁸ K. Rudolph dates the reform to the reign of Yazdgird iii in his ‘Mandaeans III: Interaction with Iranian Religions’, in Encyclopædia Iranica (online edition, 2008), url: http://www.iranicaonline.org/articles/mandaeans-iii (last accessed 8 August 2018). Panaino (‘A “Neglected” Source’, p. 97, fn. 2 and 103, fn. 50) objects to this hypothesis on the basis of the testimony of the Islamic-era sources and the Syriac Martyrium of Gregory, which de Blois (‘Persian Calendar’, p. 47), cites as both a witness to and a terminus ad quem for this reform.

¹⁹ de Blois, ‘The reform of the Zoroastrian calendar’.

²⁰ Stern, Calendars in Antiquity, p. 181.

²¹ A. Panaino (‘A “Neglected” Source for the History of the “Reformed” Zoroastrian Calendar’, in Non licet stare caelestibus: Studies on Astronomy and its History offered to Salvo De Meis, (ed.) A. Panaino (Milano-Udine, 2014), pp. 97–112) cautions against drawing conclusions about the ‘pre-history’ of this calendar (p. 100), which I read to mean prior to the sixth century ce reform.

²² In the Julian calendar, ‘leap years’ occur once every four years, without exception. In the Gregorian calendar, centuries are not leap years, save for divisible by 400; thus, the year 1900 ce was not a leap year, but 2000 ce was.

²³ Petermann, J. H. (ed.), Thesaurus sive Liber Magnus Vulgo ‘Liber Adami’ Appellatus Opus Mandaeorum Summi Ponderis. Tomus I: Pars 1 s. Dextra. (Leipzig, 1867), p. 395Google Scholar, ln. 7.

²⁴ I have reckoned all dates prior to October of 1582 according to the Julian calendar. Kristina Plazonic has developed a web browser-based application to correlate the colophon dates with the Gregorian calendar, which is available at https://goo.gl/uLhx32 (last accessed 29 August 2018).

²⁵ Lidzbarski, Ginzā, pp. 412–414. Lidzbarski translates arbā habšabbā as ‘Wednesday’, which is a plausible reading but obviously imprecise for dating purposes (by comparison, the entry for ap 800 refers to ‘the month of Siwān—Ṣelmi of the Meseneans—on the first day of the month, in the second hour and a half hour’). I am preparing a translation of the entirety of this tractate for publication in the series Translated Texts for Historians (Liverpool).

²⁶ In this regard, Lidzbarski and Taqizadeh is followed by nearly all other scholars who have discussed this passage, e.g. Shapira, D. D.Y., ‘On Kings and on the Last Days in Seventh Century Iraq: A Mandaean Text and Its Parallels’, ARAM Periodical 6 (2010), pp. 133–170Google Scholar [141], and Bladel, K. van, From Sasanian Mandaeans to Ṣābians of the Marshes (Leiden and Boston, 2017)CrossRefGoogle Scholar.

²⁷ Taqizadeh, ‘Ancient Persian Practice’, p. 614.

²⁸ NASA technical publication, ‘Five Millennium (-1999 to +3000) Canon of Solar Eclipses Database’, url: https://eclipse.gsfc.nasa.gov/SEpubs/5MCSE.html, last accessed 18/8/2018. Eclipse Predictions by Fred Espenak, NASA's GSFC.

²⁹ Rothstein, G., Die Dynastie der Lahmiden in al-Hira. Ein Versuch zur arabisch-persischen Geschichte zur Zeit der Sasaniden (Berlin, 1898), pp. 71–114Google Scholar

³⁰ NASA, ‘Five Millennium (-1999 to +3000) Canon’. Within this time frame, there were 10 total, annual, or partial eclipses that were potentially visible from portions of Mesopotamia and western Iran. Of these, only two occurred during Šabāṭ / Dawlā, which ranged from late July to early September during this period: those of Saturday, 9 September 405 ce and Sunday, 20 August 472 ce. Only the latter fell on the fourth Sunday of that month.

³¹ de Blois argues that “the notion that the kings of Persia were able to enforce a correction of the calendar regularly every 120 years is one that hardly seems reconcilable with the chequered political history of the country” (ibid., p. 40), and Panaino similarly considers such an intercalation “highly improbable” (‘A “Neglected” Source’, p. 103), but nonetheless notes that the evidence suggests “a certain synchronization between the occurrence of the spring equinox and the beginning of the New Year was still working in the middle of the 6^th century ad” (p. 102).

³² Stern, Calendars in Antiquity, p. 181.

³³ Petermann, Thesaurus sive Liber Magnus, p. 379, lns. 3–6. This figure was evidently derived by dividing 480,000 into 7 equal parts, which would result in 68,571 years with a remainder of 154 days, 6 hours, 51 minutes, 25 seconds, and 714 milliseconds.

³⁴ Ibid., p. 27, lns. 15–21 and in the parallel on p. 50, ln. 24 to p. 51, ln. 6. This same order, offset by four, reproduces the so-called ‘Chaldæan order’, which establishes the governorship or ‘faces’ of the Decans and the ‘planetary hours’, each of which governs the day in which it falls first. The ‘Chaldæan order’ is determined by the speed with which the planets appear to move through the heavens, from the slowest (Saturn) to the fastest (the Moon); the motivation for the Mandæan offset is not immediately apparent.

³⁵ Ibid., p. 379, ln. 18 to p. 380, ln. 8.

³⁶ Ibid., p. 379, lns. 6–18. Note that the zodiacal eras follow the same order as the Mesenæan months, but without the offset, beginning with Aries rather than Aquarius.

³⁷ Similarly, this same period is described as ‘the end of the age of Mars’ in the Book of John, e.g. Lidzbarski, Johannesbuch, p. 50.

³⁸ Wilde, O., Reviews (London, 1908), p. 128Google Scholar.

³⁹ Taqizadeh, ‘Ancient Persian Practice’, p. 604.

⁴⁰ It is my honour to be given the opportunity to make this small contribution to a volume honouring one whose erudition and contributions to knowledge in this area are unparalleled. Earlier drafts of this article were reviewed by Yuhan Sohrab-Dinshaw Vevaina and Antonio Clemente Domenico Panaino, and were much improved by their helpful criticism and feedback. Any errors of omission, commission, deduction, induction, transliteration, transcription, and/or translation that remain are naturally my sole responsibility.