Zoroastrianism: History, Beliefs, Practices, Avesta, and the Parsi Diaspora

The Gathas are seventeen hymns, and they are the only texts in the Zoroastrian canon that most scholars will confidently say came from a single named historical figure. The language of the Gathas — Old Avestan, the sister dialect to the language of the Rigveda — is archaic to a degree that philologists have been arguing about for a hundred and fifty years, with positions ranging from a confident 1500 BCE (Boyce, in some of her more sweeping statements) to Kellens’ flat refusal to give a date at all. The religion of Zarathustra in the Gathas is austere, ethical, dualistic in a way that has to be carefully read, and very different from the temple-and-priesthood Zoroastrianism of the Sasanian state. The thing usually called “Zoroastrianism” in a popular book is the Sasanian version, the religion that emerged from centuries of editorial work on the Avesta, the codification of twenty-one nasks that are mostly lost, and the political theology of a court that needed a state church. Both are legitimate objects of study. The mistake is to assume they are the same thing.

The accounts I find most useful are Mary Boyce’s three-volume Zoroastrians: Their Religious Beliefs and Practices (1975) and History of Zoroastrianism (co-edited with Frantz Grenet, 1991–), Gherardo Gnoli’s Zoroaster’s Time and Homeland (1980), and Prods Oktor Skjærvø’s The Spirit of Zoroastrianism (2011), which is the clearest short introduction in English. Shaul Shaked’s Dualism in Transformation (1994) is the indispensable book on the variants of dualism. Almut Hintze’s The Zand i Wahman Yasn (2009) is the best thing on the eschatological material. Michael Stausberg’s Zoroastrianism (2022) is a recent balanced overview, and Yuhan Vevaina and the Encyclopaedia Iranica articles are the first places to look for current scholarship on disputed points.

For the connected topics covered in this site, see the clusters on Zarathustra, the Magi, Mithraism, and Ahura Mazda and the pantheon, and the long-tail pages on whether Zoroastrianism is still practised, on Zarathustra, on Zurvanism, and on the Yashts.

A short, opinionated history

The Zoroastrian tradition is, in the most defensible reading, the religion that developed in the eastern Iranian plateau, was carried westward by Median and Persian dynasties, became the religious language of the Achaemenid state without much of an Achaemenid theology, was transmitted orally in the Parthian period while the Sasanians were still an obscure priestly dynasty in Fars, and was codified as a state church in the third century CE. The Sasanian nasks, the Pahlavi Zand, the Bundahishn, the Denkard — all of this is downstream of the Sasanian moment, and most of the nasks are lost. We have the Gathas, the Yasna, the Visperad, the Vendidad, the Yashts, the Khorda Avesta — about a quarter of what the Sasanians had, and that is one of the more substantial losses in the history of religious literature.

The Achaemenid kings, from Darius I onward, invoke Auramazdā at the top of nearly every inscription. The Achaemenid Auramazdā is, on the inscriptional evidence, not the full Zoroastrian pantheon. The Amesha Spentas are not named. Zarathustra is not named. The Gathas are not cited. Darius I in the Behistun inscription (c. 520 BCE) credits Ahura Mazda with his victory and treats the druj — the lie — as the moral opposite of arta, truth, but this is a king, not a theologian. Shaked’s reading of the Achaemenid religion as a Magi-mediated form of what would later be codified as Zoroastrianism is, to my mind, more convincing than the older “Achaemenid kings were fully Zoroastrian” position. Cyrus the Great, who in the Cyrus Cylinder invokes Marduk and not Ahura Mazda, is the awkward test case. The standard reading is that the cylinder is a Babylonian document in Babylonian idiom, and that the absence of Ahura Mazda tells us about the audience, not the king’s private religion.

The Sasanian state made Zoroastrianism official. The high priest, the mobadan mobad, sat on the imperial council; the great fire temples of Adur Farnbag, Adur Gushnasp, and Adur Burzen-Mihr were endowed with land and personnel; the dadestan (religious law) was applied in the courts. Mary Boyce’s “On the Pious Foundations of the Zoroastrian Priesthood” (BSOAS 1968) Boyce’s article on the institutional economics remains the indispensable study of the Sasanian church. Pourshariati’s Decline and Fall of the Sasanian Empire (2008) complicated the older picture by showing how much the wuzurgān — the great noble houses — ran the Sasanian state in alliance with, and often in opposition to, the church. The point matters for dating: a centralised Sasanian church probably emerges under Khosrow I Anushirvan in the sixth century, not under Ardashir I in the third.

The Arab conquest of 651 CE scattered the community. A group migrated, by a route we can only sketch, to Gujarat, possibly in the eighth or tenth century — the traditional date of 716 CE rests on the Qissa-i Sanjan and is not independently verifiable. The Qissa-i Sanjan itself is a sixteenth-century text, the Bayān-i Sanjān of Bahman Kaikobad, and the legends of the Jadi Rana and the sixteen naujvar (noble families) are literary reconstructions, not contemporary records. By the seventeenth century the Parsis were in Bombay, and from the nineteenth they were everywhere from the courts of British India to the boardrooms of the Tata Group.

The Avesta, in layers

The Avesta as we have it is a layered text. The oldest stratum is the Gathas, seventeen hymns in five groups (Yasna 28–34, 43–46, 47–50, 51, 53), in a dialect of Old Avestan that sits close to the language of the Rigveda. Around the Gathas is the Yasna, the central liturgical text; the Visperad, a set of extensions to the Yasna used in the Gahambars; the Vendidad, the priestly code of ritual purity; the Yashts, the twenty-one hymns to the yazatas; and the Khorda Avesta of shorter prayers for daily use. The standard critical edition is Karl Friedrich Geldner’s three-volume Avesta, the Sacred Books of the Parsis (1886–96), still in print and still useful; Helmut Humbach’s work on the Gathas, Jean Kellens’ French editions, and the more recent English work of Skjærvø are the modern scholarly treatments.

The Sasanian Avesta was much larger. The twenty-one nasks were arranged in a Gathic group, a Hadha-Mānthric group, a Dātastān group, and a Hūxtāk group, in a total of 21 books. The contents are partly recoverable from the Denkard (ninth century, an encyclopaedic compendium) and the Bundahishn (ninth or tenth century, a cosmogony in Pahlavi prose). The actual Avestan text survives for about a quarter of the corpus; the rest is preserved in Pahlavi translation and commentary — the Zand — and through Pahlavi encyclopaedias. D. N. MacKenzie’s A Concise Pahlavi Dictionary (1971) is still the working tool for the Pahlavi material.

Beliefs, in order of how central they are

Ahura Mazda — the “Wise Lord” — is the central figure of the tradition. The Gathic Zarathustra addresses Ahura Mazda directly, repeatedly, in the language of a man arguing with his god in the open air. The six Amesha Spentas — Vohu Manah (Good Mind), Asha Vahishta (Best Truth), Khshathra Vairya (Desirable Dominion), Spenta Armaiti (Holy Devotion), Haurvatat (Wholeness), Ameretat (Immortality) — are the divine attributes personified. Each presides over a creation. Asha Vahishta over fire. Haurvatat over water. Spenta Armaiti over earth. The Bundahishn sets out the developed system. Kellens has argued, against the older Boyce view, that Spenta Armaiti was originally a feminine figure of the earth, and that the apparent gender confusion in some sources is real. I find Kellens more persuasive on the philology, but the issue is genuinely open.

The dualism is the question everyone asks. In the Gathas, the opposition is between Spenta Mainyu (the Holy Spirit, Ahura Mazda’s own spirit) and Angra Mainyu (the Destructive Spirit). The Gathic evidence is ambiguous. Some passages read Angra Mainyu as an internal principle of opposition, a kind of choice against asha; others read him as a near-independent attacker. The Zurvanite position, derived from the figure of Zurvan (Infinite Time), is the most explicit: both spirits are born of Zurvan’s sacrifice, with Ohrmazd born of his hope and Ahriman of his subsequent doubt. The orthodox position, as the Bundahishn hammers home, is that Angra Mainyu is co-eternal with Ahura Mazda, has always existed, and will be defeated at the Frashokereti. The Bundahishn was compiled in part to refute Zurvanism, and the orthodox church was, in the Sasanian period, actively hostile. Shaked’s Dualism in Transformation (1994) is still the right place to start.

The eschatology is elaborate. The Chinvat bridge, narrow for the wicked, broad for the just. The soul weighed against its deeds by Mithra, Sraosha, and Rashnu. The four levels of afterlife — heaven, hell, the hamestagan for those whose good and evil are equal, and the mishdana purgatorial state. And at the end, the Frashokereti, the final renovation: the resurrection of the dead, the defeat of Angra Mainyu, the rekindling of the world. The figure of the Saoshyant, the future saviour born of the seed of Zarathustra preserved in a lake in the Elburz, is post-Gathic but central in the later tradition. The idea of a virgin birth of the Saoshyant is a Zand-era development; the Gathas expect a righteous king or a righteous general of Iranian tradition.

Practice

The fire temple, the atashkadeh, is the visible centre of worship. Three grades of fire are recognised: the dadgah (hearth fire maintained by a layperson), the adaran (community temple fire), and the atash behram (the fire of victory, established by a long ritual of consecration that involves sixteen different fires, a varas ceremony that can take a year). The Indian Parsi community maintains eight atash behram fires. Iranian Zoroastrian temples burn sandalwood only in India, not in Iran, where the use of frankincense, myrrh, and other local aromatics is the older tradition; the regional variation is sometimes taken for inconsistency, but it isn’t.

The dakhma — the Tower of Silence, where the dead are exposed to sun and vultures — is the most photogenic and the most endangered practice. The collapse of vulture populations across South Asia since the 1990s, caused by the veterinary drug diclofenac that is toxic to Gyps vultures when they feed on treated cattle carcasses, has effectively ended sky burial at the major Parsi towers. The Malabar Hill towers in Mumbai, the Malabar Hill dakhma set up in the 1670s on a donated site, now use solar concentrators and chemical dissolution, and the controversy over what to do with the bones is ongoing. The total Parsi population in Mumbai is now around 57,000, down from a peak of about 100,000 in the mid-twentieth century, and the demographic problem is the bigger one. The fertility rate is around 0.8 children per woman.

The daily ritual centres on the sudreh (a white cotton shirt) and the kusti (a woven cord tied around the waist). The kusti ritual — untying, reciting the Ahunavaiti Gatha, retying — is performed several times a day, and is the layperson’s main ritual obligation.

The calendar and the diaspora

The Zoroastrian calendar is solar: twelve months of thirty days, plus five gatha days at the end of the twelfth month, in a 365-day year that drifts about a day every four years against the Julian. The months are named after the Amesha Spentas and the yazatas: Farvardin (the Fravashis), Ordibehesht (Asha Vahishta), Khordad (Haurvatat), Tir (Tishtrya), and so on. The Sasanian calendar is the basis of the modern Iranian Solar Hijri, the official calendar of Iran and Afghanistan — the year 1403 in Iran began at the spring equinox of 21 March 2024. The “Yazdegirdi calendar” name is misleading; the last major reform was under Yazdegerd III in 632 CE, but the calendar as used today is a Solar Hijri reform, not a continuous Sasanian survival.

The diaspora is now the largest part of the Zoroastrian world. Los Angeles, Toronto, London, Sydney, and Dubai have substantial communities, fire temples, and cultural organisations. The Federation of Zoroastrian Associations of North America (FEZANA), the World Zoroastrian Organisation, and the Association of Zoroastrians in Berlin are the principal coordinating bodies. The community is small, ageing, and demographically fragile. The Jiyo Parsi programme in India, launched in 2013, is the most visible attempt to reverse the trend. The orthodox position on intermarriage and conversion is the harder side of the tradition to defend in the diaspora, and the disputes are real and ongoing.