A fire for cremation

YAv. nasupāka- (cremation) literally means “(in which a) corpse (is being) cooked.” The corpse represented the highest degree of impurity in ancient Zoroastrianism, and the Wīdēwdād and later texts based on it abound in prescriptions regarding how to avoid any direct or indirect contact with dead matter. As fire was one of the purest elements and the corpse was a key source of contamination, cremation was considered a sacrilege. As such, it makes perfect sense that a list in decreasing order of fires exposed to impurity begins with the most polluted one, that of cremation.

A fire on which (textiles/fur/hides) have been soaked to be dyed

YAv. uruzdipāka-, which translates as “on which (textiles/fur/hides) have been soaked to be dyed,” has no clear etymology. Spiegel translated this as “das Unreinigkeiten kocht” (which cooks impurities), deriving the Avestan (Av.) uruzdi° from the verbal root rudh- “to flow.”Footnote ¹⁰ Darmesteter translated uruzdipāka- as “où l’on a brûlé les liquides impurs” (in which one has boiled impure liquids) based on a comparison with the Pahlavi explanation ātaxš ī nasā-pāg (…) ātaxš ī hixr-pāg, “a fire for cremation (…), a fire on which bodily refuse has been cooked,” in Dēnkard 8.44.46, which resumes V 8.81–82.Footnote ¹¹ Bartholomae followed Darmesteter in his translation, producing “Flüssigkeiten kochend” (cooking liquids), which the former mistakenly linked with the verbal root ³raod-, “to hinder,” instead of ⁴raod-, “to flow.”Footnote ¹² Herzfeld criticized these above translations, opting to instead follow the Pahlavi translation rang-pāg, which he rendered as “fire for color-boiling.”Footnote ¹³ According to Herzfeld, Av. uruzdi° is etymologically related to Av. raoiδita- “red.” Although Herzfeld never mentioned this, Aspandiarji had already proposed this interpretation, as noted by Darmesteter.Footnote ¹⁴ In my opinion, the Parsi scholars were on the right track.

None of the three Proto-Iranian (PIr.) verbal roots—*(H)rau̯d- (to tear, to break), *Hrau̯d- (to grow), and *(H)rau̯d(H)- (to moan, to bewail)—reconstructed by Cheung provide a satisfactory explanation of YAv. uruzdi°.Footnote ¹⁵ A fourth option, however, may solve the problem.

YAv. raoiδita-, “red,” cognate of Vedic (Ved.) róhita-, “idem,” goes back to the Proto-Indo-European (PIE) *h₁reu̯d^h-, “to be red.”Footnote ¹⁶ From its zero grade plus a suffix *-ti°, the first element of compound YAv. uruzdi° can be formed (PIE *h₁rud^h-ti° > Proto-Indo-Iranian (PIIr.) *Hrudd^hi° > PIr. *Hrud^zdi° > YAv. uruzdi°). YAv. uruzdipāka-, literally “cooking red,” most likely designates the process of soaking textiles, fur, or hides to be dyed in red color. Since red, in any of its shades, was one of the most frequently extracted colors from different types of dyes in antiquity, “cooking red” or rather “cooking (textiles/fur/hides) in red (color)” might easily have developed into the general meaning of soaking textiles, fur, or hides in this or any other color to be dyed. If so, the Pahlavi translation rang-pāg, “cooking dye,” for YAv. uruzdipāka- perfectly matches the original. Not only does the etymology of the Young Avestan compound and its Pahlavi translation point to this interpretation, but the ancient technique behind the process points to it as well. As is well known, dyes did not easily adhere to textiles, fur, or hides without the help of mordants, which often consisted of urine in ancient times.Footnote ¹⁷ For a Zoroastrian, as such, a fire on which textiles, fur, or hides were dyed was unquestionably a fire exposed to a high degree of impurity in the form of urine, which is a bodily refuse and therefore considered filthy. Moreover, the degree of impurity could increase if the materials to be dyed were fur or hides, which could still have remains of flesh even when dried. This double source of impurity may also explain why YAv. uruzdipāka was interpreted as hixr-pāg, “cooking bodily refuse,” in Dēnkard 8.44.46. Moreover, from a compositional point of view, it makes perfect sense that a fire exposed to urine (and eventually also remains of flesh) was listed immediately after a fire polluted by a corpse in a cremation, the highest degree of contamination. If my interpretation is right, V 8.82 provides relevant information about the material culture of the Young Avestan period, insofar as it may imply the use of urine as mordant in a specific technique for dying textiles, fur, or hides.

A fire into which cream has boiled over/curdled

YAv. ⁺sairiiehiiaṯ, written sairehiiaṯ in most manuscripts, is a hápax legómenon of unclear meaning and no satisfactory etymology. The Pahlavi translators did not understand it or consider it a technical term, so they left it untranslated. As a result, most scholars dealing with this passage have resorted to traditional New Persian (NP) translations, which interpreted this term as “dung.” For instance, the Pahlavi Wīdēwdād manuscript 4700 renders YAv. sairiiehiiaṯ and its Pahlavi (Phl.) translation as NP az sargin, “from dung,” in folio (f.) 184r, l. 12, and transcribes Phl. <slhyyc> as NP sarāgic “?” in f. 184v, l. 12. Anquetil-Duperron followed these New Persian translations and rendered YAv. sairiiehiiaṯ as “dans lequel on a brûlé des excrémens” (in which one has boiled excrements); Spiegel and Bartholomae translated it as “Unrath” (debris) and “Vorrichtung zum Dörren von Mist” (setup for drying dung) respectively, the latter translation being followed by Wolff’s “Mistdarre” (kiln for drying dung); Darmesteter rendered it as “wherein cowdung has been burnt” and “où l’on a brûlé du fumier” (idem); Scheftelowitz gave the translation “Misthaufen” (dungheap); Anklesaria and Kapadia read its Pahlavi translation as sarîyê-icha and rendered it as “dung” as well, while Moazami, following Darmesteter and Scheftelowitz, read it as sarāsp and translated as “dung.”Footnote ¹⁸

Only de Vaan deviated from this traditional interpretation and translated YAv. ⁺sairiiehiia- as “(pile of) reeds” based on its possible Vedic cognate śárya-, “arrow.”Footnote ¹⁹ Moreover, de Vaan transliterated the Pahlavi translation as <slʾs-c>, transcribed it as sarāh-az, and considered it a mechanical transcription of YAv. ⁺sairiiehiiaṯ. I agree with de Vaan’s interpretation of the Pahlavi translation of YAv. ⁺sairiiehiia- as an attempt to transcribe it in Pahlavi script.Footnote ²⁰ However, I disagree with his interpretation of YAv. ⁺sairiiehiia- as “(pile of) reeds” and propose a different etymology.

In my opinion, YAv. sairehiiaṯ is a corruption of ⁺sairiiehiiaṯ, the ablative singular masculine or neuter of a compound, the first element of which is YAv. sairi° and the second element is YAv. °iahiia- > °iehiia-. The first element of the compound in –i° is part of a Caland system together with Ved. śáras-, “cream, curdled milk,” which is related to Ved. śar-, “to break, to curdle,” and stems from PIE *ḱerh₂-, “to break.”Footnote ²¹ The second element of the compound has as a cognate in Ved. yas-, “to boil,” and goes back to PIE *i̯es-, “to boil.”Footnote ²² If my analysis of this Young Avestan compound is correct, YAv. sairi-iehiia- would designate a fire “into which cream has boiled over/curdled,” probably meaning that the cream boiled over from its container during heating. Milk and cream are produced by pure animals, such as cows and goats, so they are not impure themselves; if they boil over from heating, however, milk and cream can be burned to a residue.Footnote ²³ In such a case, it makes perfect sense for this to be placed on the list after a fire on which textiles have been soaked to be dyed, as the technique implied boiling liquids that could eventually pollute the fire in both instances.

A fire from the potter’s kiln, a fire from a glass kiln

YAv. xumba- “vessel,” attested in both V 8.84 and 85, is probably related to Ved. kumbhá-, “idem.”Footnote ²⁴ YAv. xumba- designates a specific type of kiln used for both pottery (YAv. zəmaini.pacikāṯ, literally “baking in clay,” in V 8.84) and glass blowing (YAv. yāmō.pacikāṯ, literally “baking glass,” in V 8.85).Footnote ²⁵ Despite Harmatta’s important remarks on the possible technique behind this kiln, I do not think the Young Avestan syntagms formed with YAv. xumba- provide enough information to precisely match it with a concrete type of kiln from among the archaeological remains of first millennium BCE southern Central Asia.Footnote ²⁶

A fire from an oven for roasting grain

Scholars have interpreted YAv. aoniiaṯ. haca. parō.bərəjiiāṯ., “from an oven for roasting grain,” in different ways. Anquetil-Duperron translated the syntagm altogether as “ceux qui travaillent en étain” (those who work with tin) and Spiegel as “Erz(e)” (ore).Footnote ²⁷ As regards YAv. aoniia-, the latter was unsure about connecting it with either Ved. aváni- “riverbed” or Hebrew (Hebr.) oni, oniyāh, “vessel.” Bartholomae rendered YAv. aoniia- as “Feuerungs- oder Heizverrichtung” (firing equipment or heater), without specifying which one, and did not venture any etymology; Scheftelowitz opted to link it with Ved. aváni-, “riverbed,” but rendered the Young Avestan word as “Platz” (place).Footnote ²⁸ Mayrhofer doubted this connection and de Vaan denied it, translating YAv. aoniia- as “oven.”Footnote ²⁹ On the other hand, Herzfeld derived YAv. aoniia- from YAv. unā-, “cavity, hole,” attested in V 17.2 and Nērangestān (N) 82.2, and considered it a possible loanword from Semitic, citing Old Babylonian anā, unūtu and Hebr. oni, oniyāh, “vessel.”Footnote ³⁰ Although the link with YAv. unā-, “cavity, hole,” is likely, I would not rule out the possibility of a Semitic origin for YAv. aoniia-, as already advanced by Spiegel.Footnote ³¹

As far as YAv. parō.bərəjiia- is concerned, Bartholomae left it untranslated.Footnote ³² Scheftelowitz understood °bərəjiia- as “grain.”Footnote ³³ Bailey interpreted the compound as “which roasts the food” and linked the Young Avestan first element of the compound parō° with the second element of the Khotanese (Khot.) compound aś-para-, “horse-fodder, lucerne,” and the second element of the Young Avestan compound °bərəjiiāṯ with Ved. bhrajj-, “to roast.”Footnote ³⁴

In my opinion, Scheftelowitz and Bailey were on the right track, but perhaps we can go a bit further. The cognates Khot. aś-para-, “horse-fodder, lucerne,” and Sanskrit (Skr.) bharuja-, “roasted barley,” point not to the general meaning of “which roasts the food” but to a more concrete meaning, “roasting grain.” On the one hand, V 8.86c equates the meritorious action with bringing “as many firebrands as separate plants,” making it clear that YAv. parō.bərəjiiāṯ refers to plants. On the other hand, the Pahlavi translation bē pēš [arzan] brēzān, “before the [millet] roasters,” of YAv. parō.bərəjiiāṯ also indicates that the translators understood this not as food in general but as plants, and, more concretely, as millet. While the Pahlavi translators correctly identified the Young Avestan second element of compound °bərəjiiāṯ, “roasting,” and rendered it appropriately with Pahlavi brēzān, “roasters,” they mistook the Young Avestan first element of compound parō° as the Avestan adverb parō, “before,” subsequently rendering it in Pahlavi as bē pēš, “before.” However, a Pahlavi commentator noticed the mistake and added the gloss arzan, “millet.” As we see, both the Khotanese and Sanskrit cognates as well as the Pahlavi commentary indicate that the Young Avestan compound may refer to roasting grain. If so, YAv. aoniiaṯ. haca. parō.bərəjiiāṯ. would designate a specific type of vessel or oven for roasting grain. It is pure speculation to think that such a vessel or oven may have been metallic, but then V 8.86 would perfectly serve as a link with the next passages, which deal with forges for smelting different metals.

A fire from a forge for smelting gold, silver, bronze/iron, and iron/steel

These passages list fires used in forges for smelting four different metals in decreasing order of economic value: YAv. zaraniia-, “gold”; ərəzata-, “silver”; aiiah-, “bronze/iron”; and haosafnaēna-, “iron/steel.”Footnote ³⁵ The meaning of the two former Avestan terms is considered established, but that of the two latter remains debated.

YAv. aiiah- and haosafnaēna- have been interpreted by Anquetil-Duperron as “steel” and “copper” respectively; by Spiegel as “iron” and “stone”; by Darmesteter and Bartholomae as “iron” and “steel”; by Scheftelowitz as “iron” and “copper”; and by Herzfeld as “bronze” and Spanish (sic!) “steel.”Footnote ³⁶ However, Herzfeld noted that YAv. aiiah- can be interpreted as either “bronze” or “iron.”Footnote ³⁷

Etymology does not help much in clearly identifying the metals intended by the Young Avestan terms aiiah- and haosafnaēna-. The former has a Vedic cognate in áyas-, but the term experienced a semantic shift. Footnote ³⁸ While the Ṛgvedic passages in which the term is attested do not help us establish a clear meaning, the metal was qualified as black and likely designated iron in the Black Yajurveda, as Chakrabarti noted.Footnote ³⁹ The Pahlavi translators rendered YAv. aiiō.saēpa- as Phl. āhanēn dazīdārān, “ironsmiths,” and added the explanation āhangarān ud arzīzkarān, “ironsmiths and tinsmiths.” However, this Pahlavi translation and commentary alone are not a proof enough to simply transfer the meaning “iron” to YAv. aiiah-.

As regards YAv. haosafnaēna-, it was rendered into Pahlavi as pōlāwad dazīdārān, “steelsmiths,” and glossed as āhangarān <ud> cēlāngarān, “ironsmiths and swordsmiths,” but again we cannot uncritically transfer the Pahlavi meaning to the Young Avestan original. Spiegel proposed an etymology for YAv. haosafnaēna-, relating it to YAv. safa-, “hoof (of horses).”Footnote ⁴⁰ Ognibene proposed a metathesis from Old Iranian (OIr.) *spana-, “iron,” to *safna-, “idem,” in YAv. haosafnaēna- and rendered it as “good iron.”Footnote ⁴¹ More recently, Peyrot, Dragoni, and Bernard, quoting a personal communication by Lubotsky, considered this Young Avestan word to be an inner-Iranian borrowing from a Skythian dialect, eventually going back to a form *ham-tsu̯ana- < *ham-ću̯ana-, “iron.”Footnote ⁴² Taking into account that the development of PIr. *ham° as Av. hao° is an ad hoc explanation with no parallel in Avestan, to my knowledge, and that PIr. *ću̯a- would yield Av. spa-, I consider that reconstruction unlikely. In my opinion, the first element of the compound in YAv. haosafnaēna- is derived from the full grade of hu-, “good” (Av. hao°), of which there are several parallels in Avestan, and its second element is an adjectival derivation in –na- from YAv. saf(a)-, “hoof (of horses),” plus the suffix -aēna- with the meaning “characterized by/made of.” If this is correct, YAv. haosafnaēna- would designate a kind of metal, perhaps iron, to make horseshoes.

A fire from a baking oven

As regards YAv. tanūr(a)-, “baking oven,” Spiegel was the first to propose that the term may derive from Semitic and be related to Hebr. tannūr.Footnote ⁴³ Geiger, followed by Darmesteter, reproduced a communication by Halévy at the Société de linguistique of Paris, according to whom YAv. tanūra- was a loanword from Aramaic and related to the Hebrew and Arabic tannūr.Footnote ⁴⁴ Nevertheless, Geiger argued that both the Young Avestan and Semitic word may go back to a common word inherited from an older “türanische Kultur” (Turanian culture), without ruling out that a word from Mesopotamia might have entered the Avestan language prior to the Median period. Scheftelowitz and Herzfeld also proposed that the Avestan term was a loanword from Semitic.Footnote ⁴⁵

The Pahlavi translators rendered YAv. tanūr(a)- into Phl. tanūr, “oven,” and added the gloss nān-pāg, “baking bread,” to make it clear that it was a specific type of oven.

A fire from (an oven for firing) bricks

YAv. dišta-, “(an oven for firing) bricks,” has been differently interpreted by scholars. Anquetil-Duperron understood the term as “deserts”; Spiegel translated the word as “hearth,” derived from the verbal root daēz-, “to build, to heap (up)”; Darmesteter, followed by Scheftelowitz, Bartholomae, and de Vaan, interpreted it as “cauldron,” probably guided by the Pahlavi translation dēg, “idem,” of this word; and Herzfeld understood it as “funeral pyre” and, later, as “charcoal pile.”Footnote ⁴⁶

However, while YAv. dišta- derives from the verbal root daēz-, “to build, to heap (up),” it does not designate a funeral pyre, which was already mentioned at the beginning of the list of fires in V 8.81, but rather the material heaped up (YAv. dišta-) to build, i.e., bricks. Therefore, the professional activity referred to in this passage may be the firing of bricks for construction. The same meaning “brick(s)” applies to YAv. dištā- in V 8.74 a well: āaṯ. mraoṯ. ahurō. mazdå. auua. aētəm. nasupākəm. janaēta. auua. hē. janaiiən. apa. aētąm. dištąm. baraiiən. (And Ahura Mazdā said: “One must extinguish this [fire] in which a corpse is being cooked. They must extinguish it. They must remove these bricks”).

A fire from an oven for preparing buttermilk

The meaning of YAv. taxriiāṯ, written as taxairiiāṯ in most manuscripts, has been debated as well. Anquetil-Duperron translated YAv. aoniiaṯ. haca. taxriiāṯ as “ceux qui gardent les champs” (those who guard the fields); Spiegel translated the term as “flüssigen Erdarten” (fluid types of soil) and linked it with the verbal root tax-, “to flow”; Bartholomae left this word untranslated without proposing any etymology; Scheftelowitz read it as taxairiia-, etymologically connected it with the Greek τέφρα “ash,” and translated it as “sandiger platz” (sandy place).Footnote ⁴⁷ However, the Young Avestan word cannot be related to the Greek one, as the latter goes back to the PIE root *d^heg^u̯h-, “to burn,” which would yield Ved. dah- and Av. daž-, “to burn.”Footnote ⁴⁸ Herzfeld understood this word as a bitter infusion or tisane, and emended it as *saxariia-, which he interpreted as a Young Avestan mention of sugar.Footnote ⁴⁹ The Pahlavi translators rendered the Young Avestan word into Phl. payādagīgān, “foot-soldiers.”

In a later article, Scheftelowitz related the reading taxairiia-, which he interpreted in this instance as an adjective derived from the noun *taxara-, to Gothic þaho, “clay,” thus translating it as “Lehmboden” (clay soil).Footnote ⁵⁰ In my opinion, the connection with PIE *tenk-, “to thicken,” is correct, but it would imply that -ai- in the reading of tax(ai)riia- in several Wīdēwdād Sāde manuscripts is the result of an anaptyctic -a- and an epenthetic -i- from YAv. taxriia-.Footnote ⁵¹ If the latter is the original form and is indeed an adjective derived from PIr. *taxra- (not *taxara-), it would find its exact Vedic cognate in takra-, “buttermilk mixed with water.”Footnote ⁵² If so, it may be translated as “related to (the preparation of) buttermilk,” a suitable epithet for an oven, and would perfectly connect to the next type of fire, which is linked to a path frequented by cowherds.

A fire from a path frequented by herders

YAv. staorō.paiia-, “guarding cattle, cowherd,” is a transparent compound with the first element of YAv. staora-, “cattle,” and the second element an adjective from the Proto-Indo-European verbal root *peh₂-, “to guard, to protect,” > PIIr./PIr. *paH-, “idem.”Footnote ⁵³

A fire from a charcoal burner

The meaning of YAv. skairiia- was also problematic. The Pahlavi translators simply copied the Young Avestan word and explained it with the gloss šubān, “shepherd.” Anquetil-Duperron understood this term as referring to “night watchmen”; Spiegel, followed by Darmesteter, interpreted the word as “campground,” without proposing any etymology for it; Scheftelowitz referred to it as “dried dung,” but also ascribing to Geldner, without further reference, a connection of the Young Avestan word with NP sikār, “charcoal”; Bartholomae left the word untranslated and offered no etymology; and Herzfeld, who translated the term as “hunter,” linked it with YAv. skāraiiaṯ.raϑa-, “hunting in a chariot,” in Yašt 13.108 and NP šikār, “hunt.”Footnote ⁵⁴

Hasandoust retook the connection made by Geldner apud Scheftelowitz of YAv. skairiia- with NP sikār, “charcoal,” and added to it YAv. garəmō.skarana-, a hápax legómenon of unknown meaning that designates a tool to remove the fire from an oven, according to its Pahlavi commentary in V 14.7b.Footnote ⁵⁵ I agree with Geldner and Hasandoust, and I consider that YAv. skairiia- might designate charcoal or a charcoal burner. In my opinion, it falls almost at the end of the list because producing a charcoal burner would imply a very low degree of potential impurity, for instance, if the logs were not dry enough due to being exposed to open air conditions, or eventually none.

A fire from the nearest (place)

The Pahlavi commentators explained YAv. nazdišta-, “nearest (place),” with the gloss šabestān, “private apartment,” which means that the last fire in the list refers to the one kindled at home, i.e., a domestic hearth.