Chapter 6. The King and the Hearth: Six Studies of Sacral Vocabulary Relating to the Fireplace

In the Electra, of Sophocles, Clytemnestra dreams that Agamemnon has come back from the dead to the realm of light (417-419; ἐς φῶς 419). The king seizes the skêptron ‘scepter’ (σκῆπτρον 420) that had once been wielded by him, but which is now held by the usurper Aegisthus (420-421), and he places it firmly into the royal hearth, the hestíā (ἐφέστιον | πῆξαι 419-420). From the hearth, there then grows out of the scepter a shoot so vigorous that it covers with its shade all the kingdom of Mycenae (421-423). [1] The focus in the inquiry that follows is this very symbol of the hestíā ‘hearth’ as the generatrix of the authority that is kingship.
The general symbolism of the Greek noun hestíā and of the goddess Hestia, who is the personification of the hearth, has been studied by Jean-Pierre Vernant as a model of Greek society in general and of the family in particular. [2] Vernant draws our attention to the traditional themes of Hestia’s virginity (Homeric Hymn [5] to Aphrodite 21-32) and immobility (e.g. Homeric Hymn [29] to Hestia 3). [3] He explains these themes in terms of the exogamous and patrilocal ideology of Greek society at large. Whereas in real life women are as a rule mobile, being {143|144} shifted from hearth to hearth in the exogamous pattern of Greek marriages, myth presents an opposite image, that of the virginal and immobile goddess Hestia, who conveys the ideal, the myth, of an unbroken male line, an ever-renewed cloning of the father, by way of the paternal hearth. [4] On the level of family, the very legitimacy of generation, of reproduction, is warranted by the paternal hearth of the family. [5]
On the level of archaic Greek society in general, legitimacy is seen as a prolongation of the paternal line. Legitimacy is a global symbol for society, inasmuch as the body politic is embodied in kingship. The symbolism is at work in the vision taken from Clytemnestra’s dream, where the legitimate king is seen as sprouting from the hearth. And we must keep in mind that the Mycenaean Royal Hearth, as pictured in this dream, is destined to become the Public Hearth of the polis. [6] It is in fact a distinguishing feature of the Classical city-state that the Public Hearth is housed in the prutaneîon ‘presidential building’: [7] in Athens, for example, the arkhon’s authority is said to be derived from the Common Hearth (Aristotle Politics 1322b), and he actually resides in the prutaneîon (Aristotle Constitution of the Athenians 3.5). [8]
This symbolism of the Greek hestíā ‘hearth’ as the generatrix of authority is a matter of Indo-European heritage. Turning to the evidence of other Indo-European languages, specifically the hieratic diction of such disparate organizations as the Atiedian Brethren of Umbrian Iguvium and the Brahmans of the Indic Vedas, we shall find some striking convergences with the Greek model. Since we are dealing with societies that spoke cognate languages, I am encouraged to see in such convergences the actual traces of cognate religious attitudes, or even of cognate institutions.
The form that is central to my six studies of sacral vocabulary relating to the fireplace is the Indo-European root *h2es-. As a verb, *h2es- must have meant something like ‘set on fire’—or so we might infer from the comparative evidence of various Indo-European languages that we are {144|145} about to examine. Let us begin with the Hittites. Purely on phonological grounds, we may expect the root *h2es- to survive in the Hittite language as ḫaš-, and there is indeed an attested Hittite noun ḫašša- meaning ‘sacrificial fireplace’. [9] This noun, it is generally agreed, is related in form to Latin āra ‘sacrificial fireplace, altar’. [10]
The problem is, there is also a Hittite verb ḫaš- meaning not ‘set on fire’ but ‘beget’. [11] Despite this semantic anomaly, I propose to relate this Hittite verb has- ‘beget’ to the noun ḫašša- ‘sacrificial fireplace’. As I hope to show in due course, the actual context for a semantic relationship between the concepts of “beget” and “fireplace” may be latent in the heritage of myth and ritual.
There is a related problem, I suggest, in the semantics of the Hittite noun ḫašša-, meaning ‘king’, which has been connected in some studies with the verb has- ‘beget’. [12] In what follows, I shall be arguing that both this noun and ḫašša- ‘sacrificial fireplace’ are derived from the same Hittite verb has- ‘beget’. [13] Already at this point, we may note an analogous semantic relationship, even if we fail to understand as of now the precise application of the notion “beget.” The English noun king and the German cognate König stem from a Germanic formation *kuningaz. This noun is a derivative of *kun- (as in Gothic kuni ‘race, family’), a root with cognates in Latin gēns, genus, gignō, etc. [14] We may note especially the meaning of gigno as ‘beget’ (e.g. Ennius Annals 24S). As I also hope to show in due course, the actual context for a semantic relationship between the concepts of “beget” and “king” may be latent in the heritage of myth and ritual. {145|146}

Greek Hestia, Latin Vesta, Indic Vivasvat

According to Georges Dumézil, the root *u̯es- of Greek hestíā ‘hearth’ (ἑστία) and Latin Vesta, Roman goddess of the hearth, has a cognate in the Indic form vi-vás-vat-. [15] The mythical figure Vivasvat (vi-vás-vat-), as we have already seen, is the first person ever to receive fire on earth, by virtue of being the first sacrificer on earth; he is ipso facto the ancestor of the human race. [16] In Vedic diction, to say sádane vivásvataḥ ‘at the place of the Vivasvat’ (Rig-Veda 1.53.1, etc.) is the same as saying ‘at the sacrifice’. Vivasvat, father of Yama (10.14.5, 10.17.1), is formally and thematically cognate with a figure in the Zoroastrian Avesta, Vīvahvant-, father of Yima. Vīvahvant was the first person ever to prepare Haoma (Yasna 9.3-4). The association of Iranian Vīvahvant with Haoma is crucial because the Indic Vivasvat likewise has special associations with Soma (Rig-Veda 9.26.4, 9.10.5, etc.), and, further, because Soma/Haoma (from Indo-Iranian *sauma) constitutes the Indo-Iranian sacrifice par excellence. [17]
The Indic form vivásvat- is an adjective derived from the verb vas-, with the attested meaning ‘shine’. [18] The Vedic god of sacrificial fire, Agni himself, is called the Vivasvat at the morning sacrifice, as Uṣas the goddess of dawn appears (Rig-Veda 1.44.1, 7.9.3). [19] Uṣas the Dawn is in turn called the feminine equivalent, Vivasvatī:
dídr̥kṣanta uṣáso yā́mann aktór
vivásvatyā máhi citrám ánīkam
Rig-Veda 3.30.13
at the coming of Uṣas from the darkness,
they yearn to see the great shining visage of Vivasvatī
When the fire-god Agni begot the human race, his “eye” was vivásvat-:
imā́ḥ prajā́ ajanayan mánūnām
vivásvatā cákṣasā dyā́m ca apáś ca
Rig-Veda 1.96.2 {146|147}
he [Agni] begot this progeny of men [and]
with his shining [vivásvat-] eye, the sky and the waters
In Vedic diction, the causative stem janáya- is used indifferently to denote either ‘beget’ or ‘create’. For another example of janáya- in the sense of ‘create’, I cite the following verses, again concerning the fire-god Agni:
tvám bhúvanā janáyann abhí krann
ápatyāya jātavedo daśasyán
Rig-Veda 7.5.7
your sound is heard, as you create the world,
O Jātavedas [Agni], helpful for progeny
The macrocosmic principle inherent in Agni, god of sacrificial fire, is anchored in a belief that the rising of the sun is dependent on the kindling of the sacrificial fire. The sacrificers pray as follows:
ā́ te agna idhīmahi
dyumántaṃ devājáram
yád dha syā te pánīyasī
samíd dīdáyati dyávi
Rig-Veda 5.6.4
may we, Agni, kindle
your bright, ageless fire,
so that your wondrous brand
may shine in the sky
In fact, it is Agni whom the sacrifìcers implore to make the sun ascend the sky (Rig-Veda 10.156.4). The Śatapatha-Brāhmaṇa puts it even more bluntly ( without the morning sacrificial fire, there would he no sunrise. The macrocosmic cákṣas- ‘eye’ of Agni in the passage cited above, Rig-Veda 1.96.2, is clearly the sun (cf. also 6.7.6). With the sun, Agni ajanayat ‘created’ or ‘begot’ the world and mankind. To repeat, the epithet of this solar symbol cákṣas- is vivásvat-, derived from the verb vas- ‘shine’.
By now we have seen three important contexts for the adjective vivásvat- in Vedic poetry:
  1. epithet of Agni, god of sacrificial fire
  2. epithet of Agni’s eye, the sun, when he begot, mankind {147|148}
  3. name of the first sacrificer on earth, ancestor of mankind
From these Vedic contexts of vivásvat-, then, it appears that the usage of the Indic verb vas- was appropriate to three parallel themes: the shining of the sun, the kindling of the sacrificial fire, and the begetting of progeny. Furthermore, as we have also seen, vas- implied creation as well as procreation.
To repeat, Dumézil argues that the root of this Indic verb vas- is cognate with the root *u̯es- of Greek hestíā ‘hearth’ (ἑστία) and of Latin Vesta, Roman goddess of the hearth. [20] Going beyond Dumézil’s position, we may consider the possibility that this root *u̯es- could be reconstructed further as *h2u̯es-, despite the absence of any phonological trace of word-initial *h2 before *u̯ in Greek *u̯estiā, whence hestíā ‘hearth’ (ἑστία). [21] This reconstruction is not essential to the main points still to be raised. Still, if it turns out to be valid, then the root *h2ues- of the Greek noun hestíā ‘hearth’ may possibly be interpreted as a variant of the root *h2es- as in the Hittite noun ḫašša- ‘hearth’—and, I would add, in the Hittite verb ḫaš- ‘beget’. Such a root-variation *h2es- vs. *h2u̯es-would be in line with an Indo-European pattern attested in a series of possible examples {148|149} shaped CeC(C)- vs. Cu̯eC(C)-. [22] To repeat: given that Indic vas- ‘shine’ conveys simultaneously the themes of the shining sun, the kindling of sacrificial fire, and the begetting of progeny, the reconstruction *h2u̯es- of this root, entertained here simply as a remote possibility, would make it a formal variant of *h2es-, as in Hittite ḫaš- ‘beget’ and ḫašša- ‘sacrificial fireplace’. [23]
The Indic verb vas- ‘shine’, which I have tentatively reconstructed as *h2u̯es-, has a noun-derivative uṣás- ‘dawn’, which in turn can be reconstructed as *h2us-os-. There is an e-grade variant, h2eus-os-, attested in Latin aurōra ‘dawn’ and in Greek aúōs/ēṓs (Aeolic αὔως/ Ionic ἠώς) ‘dawn’. [24] According to this scheme, there is a possibility that both Latin and Greek have words for the macrocosm of ‘dawn’ built from the root *h2eu̯s- and for the microcosm of ‘sacrificial fireplace’ built from the same root, but with a different configuration: *h2u̯es- as in Greek, hestíā (ἐστία) and Latin Vesta. [25] {149|150}
The semantic connection between the macrocosm of dawn and the microcosm of the sacrificial fireplace is explicit in the Rig-Veda, where the coming of dawn is treated as an event parallel to the simultaneous kindling of the sacrificial fire (1.124.1, 11; 5.75.9; 5.76.1; 5.79.8; 7.41.6; etc.). The link or dūtá- ‘messenger’ between dawn and the sacrificial fireplace is the fire-god Agni:
tvā́m íd asyā́ uṣáso vyùṣṭiṣu
dūtáṃ kr̥ṇvānā́ ayajanta mā́nuṣāḥ
Rig-Veda 10.122.7
at the lighting-up of this dawn [uṣás-],
men [= “descendants of Manu”] [26] have sacrificed, making you [Agni] the messenger [dūtá-]
In the stanza immediately following (Rig-Veda 10.122.8), the Vasiṣṭha-s (‘the Best’) are described as archetypal sacrificers who summoned Agni to the sacrifice. These same priestly Vasiṣṭha-s are also the first to waken Uṣas ‘Dawn’ with their songs of praise (7.80.1). Elsewhere in the Rig-Veda, it is Uṣas who awakens men for the morning sacrifice (e.g. 1.113.8-12), as opposed to the converse theme where the sacrificers awaken Uṣas:
yāvuyád dveṣasaṃ trā
cikitvít sunr̥tāvari
práti stómair abhutsmahi
Rig-Veda 4.52.4
by songs of praise, with awareness,
we awakened you [Uṣas]
who ward off the foe, O Sūnr̥tāvarī! [27] {150|151}

Radical *h2es- and Latin āra

Hittite ḫašša- is comparable in both form and meaning to Italic *āsā/*ăssā ‘sacrificial fireplace, altar’, as in Latin āra, Umbrian asa, Osean aasaí (locative). [28] The length of the radical vowel, guaranteed by Latin āra and Osean aasaí, may be a secondary Italic development. [29] If the original Italic root is *ăs-, we may then reconstruct *ăs(s)-ā- from *h2es(s)-oh2-, vs. *h2os(s)-o-, as in Hittite ḫašša-.
Also apparently related to Latin āra and Hittite ḫašša- is a series of Germanic derivatives nouns with root *as-/*az- (from *h2es-). The following list shows some of the most plausible examples:
  • Old Norse arinn ‘sacrificial fireplace’, from *az-ina- (cf. also the Finnish borrowing arina ‘hearthstone’);
  • German Esse ‘smith’s fireplace’ = ‘forge’, from *as-jōn; likewise Old High German essa, Old Norse esja (cf. also the Finnish borrowing ahjo ‘fireplace’);
  • English ash (es), from *as-kōn; likewise Old English aesce, Old Norse aska, Old High German asca.
For the meaning of English ashes, we may compare the Indic cognate, ā́sa- ‘ashes’. [30] This masculine noun can be reconstructed as *h2os-o-, {151|152} which also fits Hittite ḫašša- ‘sacrificial fireplace’. [31] For a parallel to the semantic contrast of ḫašša- ‘sacrificial fireplace’ vs. ā́sa- ‘ashes’, we may compare Lithuanian/Latvian pēlenas/pȩlns ‘domestic fireplace, hearth’ (singular) vs. pelenaĩ/pȩlni ‘ashes’ (plural).
Another possible reflex of the root *as- (from *h2es-) occurs in Greek ás-bolos/as-bólē, traditionally translated as ‘soot’. [32] A clear example is the following passage, where a woman is being blamed for laziness about her household tasks:
οὔτε πρὸς ἰπνὸν ἀσβόλην άλευμένη
Semonides F 7.61-62 W
Nor would she sit by the oven, [thus] avoiding the asbólē.
The preceding survey of noun-derivatives of root *h2es- points to the basic form of a verb. The intransitive sense of this verb, I suggest, is ‘be on fire’, as we may infer from the following correspondences in the Indic evidence:
  • kā́ma- ‘desire’   from   kam- ‘be desirous’
  • śā́ka- ‘power’    from   śak- ‘be powerful’etc.
  • ā́sa- ‘ashes’      from   *as- ‘be on fire’ (?)
We may consider the following semantic parallel in Lithuanian and Latvian;
  • pelenaĩ ‘ashes’  from   *pel- ‘be on fire’ [33]
  • pȩlni ‘ashes’
There is another possible attestation of root *h2es- as verb in Latin {152|153} ardeō, ardēre ‘be on fire’, which can be reconstructed as *ā̌s-edh- [34] plus stative suffix *-ē-. [35] Similarly, I reconstruct Latin āreō, ārēre ‘be dry’ as *ās plus stative suffix -ē-, without the segment *-edh-. We may compare the reconstructed doublet *ā̌s-edh- (ardeō, ardēre ‘be on fire’) vs. *ās- (āreō, ārēre) with Greek phleg-éth-ō (φλεγ-έθ-ω) vs. phleg-ō (φλέγ-ω), both meaning ‘be on fire’; also the solar names Pha-éth-ōn (Φαέθ-ων) vs. Phá-ōn (Φά-ων), both meaning ‘bright, shining’. [36] Within the semantic framework of cause, that is, fire, and effect, that is, no water, it is easy to imagine a development in the meaning of *ās-ē- from ‘be on fire’ from ‘be dry’. In Tocharian there is a verb as- (presumably from *h2es-) meaning ‘become dry’. [37] For another aspect of the semantic factor “no water” in *ā̌s-, we may consider the usage of the original participle of Latin ardēre, that is, assus, which means ‘roasted, broiled’, that is, ‘cooked without water’ as opposed to ēlixus ‘boiled’. [38]
An alternative reconstruction of ardēre, *ā̌si-dhē-, is disadvantageous because there is no convincing morphological justification for an *-i-. [39] Nor will it do simply to assume that ardēre is derived from āridus ‘dry’, by way of a syncope of -i-. In attested Latin, the formal and functional correlates of stative adjectives in -idus stative verbs in -ēre, not -(i)dēre:
  • calidus ‘hot’       calēre ‘be hot’        (calor ‘heat’)
  • tepidus ‘warm’   tepēre ‘be warm’    (tepor ‘warmth’)
  • āridus ‘dry’        ardēre ‘be on fire’   (ardor ‘burning’)
I fail to see how an adjective āridus meaning ‘dry’ could motivate a derivative ardēre meaning ‘be on fire’, especially when there already exists a stative verb ārēre meaning ‘be dry’: [40]
The semantic distinction between {153|154}
  • ardēre ‘be on fire’     from     *ā̌s-edh-ē-
  • ārēre ‘be dry’            from     *ā̌s-ē-
would be an illustration of Kuryɫowicz’s so-called Fourth Law of Analogy, [41] in that the more evolved form has the basic meaning and the basic form has the more evolved meaning. The basic form in this case, however, that is, *ā̌s-, may still retain the basic meaning of ‘burn’ in the noun-derivative ārea, which means ‘ground, space free of buildings or trees’. The association of this word with trees seems to be the earlier situation, as in the following context:
liber ab arboribus locus est, apta area pugnae
Ovid Fanti 5.707
the place is free of trees, an area suited for battle
Presumably, the ārea was originally a place where trees and bushes had been burned clear for the purpose of farming. We may compare Lithuanian ìš-dagas ‘arable land’, derived from the verb dèg-ti ‘burn’. [42]
Besides retaining the basic notion of ‘burn’ in the Latin nominal derivative area, the root *ā̌s- also retains this notion in the Latin noun-derivative āra ‘sacrificial fireplace, altar’; attested with the same meaning are the Oscan cognate aasaí (locative singular) and the Umbrian cognate asa. The Italic form *ā̌ssā-, which I reconstruct further as *h2es(s)oh2-, [43] is directly comparable with the Hittite form ḫašša- ‘sacrificial fireplace’, from *h2os(s)o-. [44]
Like Hittite ḫašša-, Latin āra is consistently associated with fire, as in this example:
adolescunt ignibus arae
Virgil Georgics 4.379
the altars light up with the fires {154|155}
The Oscan cognate of Latin āra, namely āsā-, is actually combined with an explicit adjectival derivative of pūr- ‘fire’ (cognate of Greek pûr- = πῦρ ‘fire’) in the locative phrase aasaí purasiaí ‘on a fiery āsā- (147 A 16, В 19 Vetter). We may compare, too, the Umbrian sacral formula pir ase antentu ‘let him put fire on the ā̆sā- in the Iguvine Tables (IIa 19-20, III 22-23).

Latin altāria and adolēre

There is a latent trace of the connection between fire and Latin āra in the formation altāria ‘sacrificial fireplace, altar’. This neuter plural noun arose from an adjectival alt-āri-; the first part is traditionally connected with the root of adoleō, while the second is explained as the adjectival suffix -āli-, with dissimilation of the -l-. [45] Instead, I propose that altāria is a Bahuvrīhi compound meaning ‘whose *ā̌s- is nurtured’. I note the incidental explanation in Paulus ex Festo 5 ed. Lindsay: altare, eo quod in ilio ignis excrescit ‘called altāre because fire develops there’; to be contrasted is the folk etymology recorded in Paulus ex Festo 27: altaria ab altitudine dicta sunt ‘called altāria on account of the altitude’.
As justification for my interpretation of the alt- in altāría as the verbal adjective of alō ‘nurture’, I cite the common Latin expression ignem alere ‘nurture fire’, that is, ‘keep the fire going’. [46] The posited Bahuvrīhi compound *al-to- + *ā̌s-i- ‘whose *ā̌s- is nurtured’ has numerous morphological parallels in Indic, of the type hatá-mātr̥- ‘whose mother is killed’; [47] among the examples from the Rig-Veda, I single out the semantically crucial iddhā́gni- (1.83.4, 8.27.7), analyzed etymologically as *idh-ta- + *agni- [48] and meaning ‘whose fire is kindled’. Some morphological parallels in Latin itself are such compounds as uersipellis ‘whose skin is changed [literally ‘turned’]’. In Plautus Amphitruo 123, Jupiter is so described for having assumed human form; elsewhere, uersipellis designates ‘werewolf (Pliny Natural History 8.34; Petronius 62). We may compare too the epithet altilāneus ‘whose wool is nurtured’, a specialized word used in the Acts of the Arval Brethren to describe sacrificial sheep {155|156} (a. 183 I 24 ed. Henzen). In Virgil Aeneid 12.169-170, a sacerdōs ‘priest’ is sacrificing an intosam bidentem ‘unshorn sheep’, and the setting is flagrantibus aris, ‘flaming altars’ (ārae). The Servian commentary adds (ad locam) the following explanation: quam pontífices altilaneam uocant ‘and the priests [pontífices] call it [the unshorn sheep] аltilānеа’. For the reconstructed notion of lānam alere ‘grow wool’ underlying the compound altilāneus, we may compare the attested notions of capillum alere ‘grow hair’ (Pliny Natural History 24.140) and pilos alere ‘grow hair’ (35.47).
As for the attested Latin notion of ignem alere ‘nurture fire’ = ‘keep the fire going’, I have found at least three parallels in the Indic traditions. The first of these is an abstract noun derived from *al- ‘nurture’, which has survived passim in the Rig-Veda with the specialized sense of designating the wood with which the fire is kindled. The word for such wood is aráṇi-. From the etymological point of view, I am proposing that the aráṇi- is the ‘nurturing, nourishment’ of the fire. [49] Since Latin alō can be used in the sense of ‘nurture [the embryo] within the uterus’ (e.g. Varro De re rustica 2.4.13, Gellius 12.1.6, Paulus ex Festo 8, etc.), it may be viewed as a comparable theme that the fire-god Agni of the Rig-Veda is born daily from firesticks called aráṇi-s (3.29.2, 7.1.1, 10.7.9, etc.). [50] Produced from the aráṇi-s, Agni is a newborn infant, hard to catch (5.9.3-4). [51] An epithet of Agni is mātaríśvan- (1.96.4, 3.5.9, 3.26.2), which means ‘swelling inside the mother’ (śvan- from śū- ‘swell’); as the Mātariśvan, Agni “was fashioned in his mother” (ámimīta mātári 3.29.11). [52] In short, I conclude that the aráṇi- is the alma māter, as it were, of fire. {156|157}
This theme brings us to the second of the three Indic parallels to the Latin notion of ignem alere ‘nurture fire’ = ‘keep the fire going’. There is a neuter noun alāta- ‘firebrand, coal’, attested for example in the Mahābhārata, which can be interpreted etymologically as ‘the nurturing one’, derivable from an earlier form *ala-. [53] As the third and final example, I note that the Indic root *al- survives in the post-Vedic word for ‘fire’, an-ala-, which has been interpreted as an original adjective meaning ‘insatiable’; [54] we may compare Greek án-altos (ἄν-αλτος) ‘insatiable’, as in Odyssey xvii 228. There is ample thematic evidence that Indic poetic traditions represent fire as the prime insatiable element. [55] As long as a fire is kept going, it must be fed, and it always needs more: hence an-ala-, ‘the insatiable one’.
I reconstruct the -al- of Greek án-altos ‘insatiable’ as *h2el- (the án- would reflect *n̥- added at a stage when initial *h2- was already lost). This root is also to be found in the causative formation *оl-éi̯-е/о- attested in Latin adoleō and Umbrian uřetu. [56] In the case of Latin adoleō, the sequence ol presents a phonological problem: word-medial ol should survive as ul, as we see from the borrowings crāpula from κραιπάλη and anculus from ἀμφίπολος. In archaic Latin, granted, we do see sporadic traces of ol for ul (popolom, Hercolei, etc.), but the consistency of the form adoleō and the total absence of *aduleō is puzzling. There is a similar crux with subolēs and indolēs. Faced with these phonological problems, one expert finds himself forced to assume morphological interference with phonological change, in that sub-olēs ‘offshoot’ and ind-olēs ‘inherent nature’ must be derivatives of olēscō ‘increase, be nurtured’. [57] In Festus 402 (ed. Lindsay), we read suboles ab olescendo, id est crescendo, ud adolescentes quoque, et adultae, et indoles dicitur ‘subolēs: from olēscō; as also adolēscentēs, adultae, indolēs’, an explanation followed by illustrative citations from Lucretius (4.1232) and Virgil (Eclogues 4.49).
We now turn to the actual meaning of adoleō, as also of Umbrian uřetu. Although this verb is visually translated as ‘burn’, Latin adoleō am be interpreted etymologically as ‘nurture’ in terms of a causative formation. As contextual affirmation of this etymology, let us test this transla{157|158}tion in the following passages, where we should note as well the consistent collocation of adoleō with derivatives of the root * ā̌s-:
cruore captiuo adolere tiras
Tacitus Annals 14.30
nurture the āra-s with the blood of captives
igne puro altaria adolentur
Tacitus Histories 2.3
the altāria are nurtured with pure fire
sanguine conspergunt aras adolentque altaria donis
Lucretius 4.1237
they sprinkle the ārа-s with blood and they nurture the altāria with offerings
castis adolet dum altaria taedis
Virgil Aeneid 7.71
while…nurtures the altāria with pure pitch-pine [58]
I propose that the idea behind these expressions involving adoleō is that the sacrificial fireplace is being “nurtured” by being kept lit with flames and, indirectly, with the material consumed by the flames. Where ad-ol- is actually combined with alt-āria, the collocation of -ol- vs. alt- can be said to reflect an inherited figura etymologica. We may compare the definition in Paulus ex Festo 5 (ed. Lindsay): altaria sunt in quibus igne adoletur ‘altāria are places in which there is adolēre with fire’. For the sense of “nurture,” we may compare the use of adoleō with penātēs, a name for the gods of one’s native sacrificial fireplace:
flammis adolere penates
Virgil Aeneid 1.704
to nurture the penātēs with flames
Servius explains (ad locum) that the verb adolēre is equivalent in usage to {158|159} augēre ‘increase’: adolere est proprie augere. We may compare, too, the formal opposite of abaleō, adaleō, meaning ‘cause to atrophy, check the growth of, abolish’.
In Umbrian, the causative formation *ol-éi̯-e/o- is attested in the sacral formula
pir persklu uřetu
Iguvine Tables III 12; cf. IV 30
with a prayer, let him nurture the fire
The  imperative uřetu corresponds formally to Latin (ad-)olētō; for the change from -l- to -ř-, we may compare Umhrian kařetu ‘let him call’, from *kal- as in Latin calāre ‘call’. Semantically, Umbrian pir…uřetu is comparable with the Latin combination ignem alere. [59]
In the case of Latin adoleō, its formal and functional connection with alō became eroded, so that the contextual association of adoleō with the notion of burning promoted a less restricted and etymologically inaccurate usage. Consequently, adoleō in the simple sense of ‘burn’ became capable of taking direct objects designating material meant to be burned, as in the following:
uerbenasque adole pinguis et mascula tura
Virgil Eclogues 8.65
burn fertile boughs and male frankincense
Besides the formation *ol-éi̯-e/o- of adoleō, which we are translating as ‘nurture’, Latin has also preserved a stative-intransitive type *ol-ē-, plus iterative suffix *-sk-e/o-, in the verb adolēscō ‘become nurtured, grow’; the stative-intransitive *ol-ē- is also attested in adolē-faciō ‘cause to be nurtured’, which occurs specifically in the context of thunderstruck trees in the Acts of the Arval Brethren (arborum adolefactarum, a. 224.16). Even the verb adolēscō is attested in the context of burning:
adolescunt ignibus arae
Virgil Georgics 4.379
the altars are nutured [= light up] with the fires
We may compare also the Swedish verb ala 'be on fire'.
The participle of adolēscō ‘bесоme nurtured, grow’ had evolved in meaning to become adolēscēns ‘adolescent’, and in this function a clearly attested formal variant adulēscēns has been preserved. [60] Thus the function of the word as a noun tolerates the expected phonological development from ol to ul that is suppressed in the function of the word as an adjective, a participle.

Latin focus

Besides the designation of “sacrificial fireplace” by way of āra, a less specialized designation for “fireplace” is focus, which is attested in not only sacral but also domestic contexts:
inde panem facito, folia subdito, in foco caldo sub testu coquito leniter
Cato De re rustica 75
make a loaf, place loaves, and bake slowly on a warm hearth under a crock
In this case Cato is giving a recipe for making the cake called libum (cf. also De re rustica 76-2). Another clear example of focus meaning ‘domestic fireplace, hearth’ is the following:
munda siet. uillam conuersam mundeque habeat. focum purum circumuersum cotidie priusquam cubitum eat habeat
Cato De re rustica 143.2
She [the uilica ‘housekeeper’] must be neat, and keep the farmstead neat and clean. She must clean and tidy the hearth every night before she goes to bed.
As for the sacral rises of the focus, we may consider the testimony of Varro:
sane Varro rerum diuinarum refert inter sacratas aras focos quoque sacrari solere, ut in Capitolia Ioui lunoni Мinervuае, nес minus in plurimis urbibis oppidisque, et id tam publice quam priuatim solere fierinec licere uel priuata uel publica sacra sine foco fieri. quod hic ostendit poeta
Servius Auctus on Virgil Aeneid 3.134 {160|161}
Indeed, Varro (Rerum diuinarum) reports that amidst the ārae that are consecrated, focī too are regularly consecrated, as in the Capitolium to Jupiter, Juno, Minerva; likewise in most cities and towns; and that this is regularly done both publicly and privately;…and that it is not allowed to perform public or private sacrifices without a focus. Which is what the Poet [Virgil] shows here.
Varro’s report on the use of the focus in the Capitolium can be directly linked with the mention of the derivative word foculus in the Acts of the Arval Brethren, year A.D. 87: the setting is in Capitolio (a. 87 I 2), and the promagister of the brethren is presiding (I 2 and following); after the preliminary sacral proceedings (I 2- 7), “on the same day and in the same place” (eodem die ibidem in area I 18), the same promagister does the following:
ture et uino in igne in foculo fecit
Acts of the Arval Brethren a. 87 I 19 ed. Henzen
he made a sacrifice with incense and wine on the fire on the foculus
In the Acts of the Arval Brethren the uses of the āra and the foculus, both located in luco ‘in the grove’, are in complementary distribution when it comes to the sacrifice of pigs and cows: the porcae piāculāres are regularly immolated at the āra and the uacca honorāria, at the foculus. [61] We may compare the following statement:
quae prima hostia ante foculum cecidit
Valerius Maximus 1.6.9
the first sacrificial animal that fell before the foculus
Unlike the āra, the focus/foculus is optionally movable, [62] as the following passages attest:
adde preces positis et sua uerba focis
Ovid Fasti 2.542
add prayers and the appropriate words at the focī that are set down {161|162}

posito tura dedere foco
Ovid Fasti 4.334
a focus was set down and they offered incense [63]

crateras focosque ferunt
Virgil Aeneid 12.285
they take away the craters and foci
praetextatum immolasse ad tibicinem foculo posito
Pliny Natural History 22.11
to make an immolation while wearing the praetexta to the accompaniment of a reed-player, with a foculus set down
bona…conserauit foculo posito in rostris adhibitoque tibicine
Cicero De domo sua 123
he consecrated the possessions…,with a foculus set down at the rostra and with a reed-player summoned for the occasion
tu…capite uelato…foculo posito bona…consecrasti
Cicero De domo sua 124
you consecrated the possessions…with head veiled and with a foculus set down
Liberalia dicta, quod per totum oppidum eo die sedent sacerdotes Liberi anus edera coronatae cum libis et foculo pro emptore sacrificantes
Varro De lingua latina 6.14
Festival of Liber: throughout the town on that day, the priestesses of Liber, old women wearing ivy on their heads, sit with cakes and a foculus, and they sacrifice [the cakes] for any purchaser.
The nature of the focus/foculus is strictly ad hoc. In Cato’s De re rustica, for example, the foculus is catalogued simply as a rustic utensil (11.4, 16.3). Any place or thing on which a fire is started qualifies as a focus, as we see from the following summary: {162|163}
quidquid ignem fouet, focus uocatur, siue ara sit siue quid aliud in quo ignis
Servius on Virgil Aeneid 12.118
Whatever fosters [fouet] a fire is called a focus, whether it be an āra or anything else in which fire is fostered [fouetur]. [64]
Such a wide range of applications is also illustrated by the semantic development of Latin focus into the Romance word for “fire” itself, as in French feu, Italian fuoco, Spanish fuego, and so on.
In light of what we have already seen of this Latin noun focus, with its strikingly expansive semantic range of contextual settings, I bring this section to a close by taking note of a striking gap in the history of the Latin language. That is, there is no known etymology for focus. [65] Without making the results of the preceding observations on focus depend in any way on what now follows, I suggest that there may be an etymological connection between the noun focus, this premier word for ‘fireplace’, and the verb faciō, which not only means ‘do’ or ‘make’ in a secular dimension but also serves as a premier word for ‘sacrifice’ in the dimension of the sacred. This is not to say that focus can be explained as a direct reflex of a primary Indo-European noun-formation. Rather, the point is that focus may perhaps represent a secondary Italic noun-formation, just as the present tense of the verb faciō represents a secondary verb-formation. The -c- of the present-tense faciō is a secondary extension from the -c- of the primary perfect formation, fēcī, direct cognate of Greek thêka (θῆκα) ‘placed’. The inherited meaning of ‘set, put, place’, as explicitly preserved in the Greek cognate, helps explain the meaning of ‘sacrifice’ in faciō (e,g. Virgil Aeneid 8.189), and in the compound sacri-ficō (e.g. Plautus Poenulus 320). More important for now, it also helps explain the traditional collocations of focus with verbs meaning ‘set, put, place’, as we have seen immediately above in the list of contexts illustrating the movable nature of the focus. If we are to explain focus as a {163|164} noun somehow derived from the verb faciō, we have to posit the generalization of the secondary -c- as part of the verb-stem fac- at an early enough stage that it could generate noun-derivatives of the type focus. [66] Moreover, we would have to posit that the -o- of focus is secondary as well, since there seems to be no way to derive a sequence like foc- directly from the Indo-European root *dheh1(-k-)/*dhh1(-k-). Still, there is room for positing an Italic or even Latin stage of derivation, with the secondary -o- of focus being perhaps shaped by analogy. [67]

Umbrian ahti- and aso-

Corresponding to Latin focus, the Umbrian word for “movable fireplace” is ahti-, [68] which is etymologically an abstract noun *ag-ti- meaning ‘carrying’ (from a verb cognate with Latin agō, agere); for the form and the semantics, we may compare Latin uectis ‘bolt, lever’, which is likewise an original abstract noun *u̯egh-ti- ‘carrying’ (from the verb attested as uehō, uehere). The usage of the word ahti- as ‘movable fireplace’ is not necessarily a feature of the Umbrian language in general: rather it is a specialized feature of the repertoire of sacral texts managed by the Atiedian Brethren of the Umbrian city of Iguvium, as recorded in the set of inscriptions known as the Iguvine Tables. The ahti- is central to the religious life of Iguvium, as is evident from the rites {164|165} described in Iguvine Tables III 1 and following. After pir ‘fire’ is kindled on the way leading arven ‘to the field’ (III 11-12), and after this fire is later placed ase ‘on the altar’ which is vuke ‘in the grove’ (III 21-22), then a sacrifice is made iuvepatre ‘to Jupiter’ at the right side of the altar (III 22-23) on behalf of the following:
  • fratusper atiieřies      ‘for the Atiedian Brethren’
  • ahtisper eikvasatis    ‘for the ahti-s eikvasatis
  • tutaper iiuvina           ‘for the people of Iguvium’
  • trefiper iiuvina           ‘for the tribus of Iguvium’ [69]
Such a hierarchy of values is a most dramatic illustration of the importance of the ahti- to the community. This Umbrian collocation of vuke ‘in the grove’/ase ‘on the altar’ with ahtisper ‘for the portable fireplaces’ is comparable to the Latin collocation of in luco ‘in the grove/in ara ‘on the altar’ with in foculo ‘on the portable fireplace’ in the Acts of the Arval Brethren. [70]
In what we have just seen quoted from the Iguvine Tables, the ablative plural ahtis is combined with the postposition –per, which is parallel to the Latin preposition prō ‘for, on behalf of’. This combination of ahti- and –per is semantically parallel to the Latin phrase prō ārīs focīsque, as in the following examples:
sibi pro aris focisque et deum templis ac solo in quo nati essent dimicandum fore
Livy 5.30.1
that they were going to have to fight it out on behalf of the ārae&focī, the sacred precincts of the gods, and the soil on which they were born
pro patria pro liberis pro aris atque focis suis cernere
Sallust Catiline 59.5
to fight it out on behalf of the patria, the household-members, and their ārae&foci {165|166}
We may also compare the highly emotional and affective tone of ārae&focī, as in Against Catiline 4.24; De domo sua 106, 143; In Pisonem 91; Pro Sestio 90; and so on.
The sacral importance of the movable fireplace in Umbrian society is apparent not only from the euphemistically abstract etymology of ahti- and from the importance of the ahti- in the hierarchy just quoted from Iguvine Tables 23-24; it is apparent also from the usage of the word in the Atiedian fire ritual. The Iguvine Tables contain two versions of this ritual, one written in the native Umbrian alphabet (Ib 10-16) and the other, in the Latin alphabet (VIb 48-53). A careful study of the parallel texts reveals several new details about the sacrificial fireplace in Italic ritual. For the sake of convenience, the texts are divided here into sections A’ to H’ (native alphabet) and A to H (Latin alphabet), on the basis of inherent divisions in subject-matter. In what follows, I print the Latin alphabet in italics, and the native alphabet in roman, marked off by brackets.
VIb 48-53 Ib 10-16
A pone poplo afero heries A’ {pune puplum aferum heries}
When he wishes to perform a lustration When you swish to perform a lustration of the people
B avif aseriato etu B’ {avef anzeriatu etu}
He shall go and observe the birds Go and observe the birds
Ca ape angla combifianšiust C’ {pune kuvurtus}
when he has announced the angla When you have returned
Cb perca arsmatiam anouihimu  
He shall put on the perca arsmatia  
Da cringatro hatu D’ {krenkatrum hatu}
He shall hold the cringatro Hold the {krenkatrum}
Db destrame scapla anouihimu  
He shall put it on the right shoulder  
E pir endentu E’{enumek pir ahtimem ententu}
He shall place fire Then place the fire in the {ahti-}
Fa pone esonome ferar pufe pir entelust F’{pune pir entelus ahitmem}
When that in which he has placed the fire is brought to the sacrifice When you have placed the fire in the {ahti-}
Fb ere fertu poe perca arsmatiam habiest  
The one who has the perca arsmatia shall carry it  
Fc erihont aso destre onse fertu  
The same shall carry the aso on his right shoulder  
G ennom stiplatu parfa desua G’ {enumek steplatu parfam tesvam}
Then he shall pronounce a parfa-bird on the right Then pronounce a {parfa}-bird on the right {166|167}
H seso tote iiounine H’ {tete tute ikuvine}
For himself and the people of Iguvium For yourself and for the people of Iguvium
Here ends the fire ritual. Then follows the banishment ritual at Acedonia, starting with ape acesoniamebenust ‘when he has come to Acedonia’ in the Latin alphabet (VIb 52-53), matched by {pune menes akeřuniamem} ‘when you come to Acedonia’ in the native alphabet (Ib 15-16).
There are several details to be noted about the fire ritual. To begin, the expression poe perca arsmatiam habiest ‘the one who has the perca arsmatia’ (Fb = VIb 50) is a tabu periphrasis occurring elsewhere, too (VIb 53, 63; VIIa 46, 51), to designate the arsfertur/{ařfertur}, who is the chief sacrificer in the cult of the Atiedian Brethren (VIa 2, etc.). Henceforth, he will be designated as “Adfertor,” the latinized equivalent of the Umbrian title. [71] The perca of the Adfertor is something that he wears, as is evident from the use of perca with anouihimu ‘put on, wear’ (Cb and Db, = VIb 49) and from its association with ponisiater/{puniçate} (VIb 51/Ib 15), a word apparently related to Latin pūniceus ‘dyed with purple’. [72] In the fire ritual described at Ib 10 and following, two officials called {prinuvatu} are to accompany the Adfertor, and they are to have {perkaf…puniçate} (Ib 15). Likewise, in the fire ritual described at VIb 48 and following, two prinuatur are to accompany the Adfertor, and they are to have percaponisiater (VIb 51); meanwhile, the Adfertor himself has perca arsmatiam (VIb 49, 50). The periphrasis designating the Adfertor as ‘the one who has the perca arsmatia’ is restricted to those parts of the Iguvine Tables that are written in the Latin alphabet. From those parts written in the native alphabet, the identity of the {ařfertur} with this man ‘who has the perca arsmatia’ (poe perca arsmatiam habiest) becomes obvious: in Ib 41-42 (native alphabet), it is the {ařfertur} who chases a sacrificial heifer while the two {prinuvatu} chase two; in VIIa 51-52 (Latin alphabet), the sacrificial heifers are chased by poe perca arsmatiam habiest ‘the one who has the perca arsmatia’ and the prinuatur.
In the parallel texts for the fire ritual presently under consideration, the tabu periphrasis poe perca arsmatiam habiest in the Latin alphabet version (A-Η) cannot be contrasted directly with a counterpart in the native alphabet version (A’-H’), since in one case the Adfertor is instructed in the grammatical third person (A-Η) whereas in the other he is instructed in the second person (А’-H’). There is, however, a direct contrast in the {167|168} devices used by the two texts in referring to the movable fireplace: version A-Η, which shows a reluctance to name the chief sacrificer of the Atiedian Brethren by title, also shows a reluctance to use the brethren’s word ahti- ‘movable fireplace’, the equivalent of Latin focus. As we have seen, section E has pir endendu ‘he shall place fire’ while section E’ has {enumek pir ahtimem ententu} ‘then place fire in the {ahti-}’. We should note, too, the tabu periphrasis in section Fa, pufe pir entelust, in place of the word ahti-: instead of some direct statement, such as “when the ahti- is brought to the sacrifice,” we see in section Fa this periphrasis: pone esonome ferar pufe pir entelust ‘when that in which he has placed the fire is brought to the sacrifice’. In the version with the Latin letters, then, specific words dealing with the cult officers and cult objects of the Atiedian Brethren are treated with special caution; perhaps the same factor of caution explains the regular use of the third person in instructing the Atiedian sacrificers, as opposed to the second person in the version with native letters.
Besides being more circumspect, the instructions in version A-Η are also more precise and detailed than in version A’-H’. Greater detail may imply less familiarity with the prescribed way of doing things; consider section Db, where it is specified that the sacrificer must place the garment cringatro on his right shoulder; in section D’, by contrast, it had sufficed to prescribe that the sacrificer must hold the {krenkatrum}. (This garment cringatro/{krenkatrum} is comparable to Latin cinctus or cingulum.) Presumably, the stark prescription of section D’ was enough of a reminder about what to do next; section Da, by contrast, also prescribes that the sacrificer must hold the cringatro, but further specification has to follow in Db about what to do with it, namely, to put it on. The reason for putting the cringatro specifically on the right shoulder becomes apparent later: the sacrificer who puts on the cringatro/{krenkatrum} is none other than the arsfertur/{ařfertur} ‘Adfertor’ (cf. Cb: the same sacrificer is putting on the perca arsmatiam). The Adfertor then proceeds to place fire in the ahti- ‘movable fireplace’ (cf. E/E’, Fa/F), which at that point “is brought to the sacrifice” (esonome ferar: Fa). The one who brings the ahti- to the sacrifice is the Adfertor himself (cf. Fb), and he carries it on his right shoulder (cf. Fc). It appears, therefore, that the garment called cringatro/{krenkatrum} may have served to shield the Adfertor’s right shoulder from the heat of the ahti- which he was to carry. Presumably, this ahti- was some kind of brazier: we may compare the brazen cribrum used by the Vestal Virgins as a movable fireplace, described as follows:
ignis Vestae si quando ìnterstinctus esset, uirgines uerberibus adficiebantur a pontifice, quibus mos erat tabulam felicis materiae tamdiu terebrare, quousque {168|169} exceptum ignem cribro aeneo uirgo in aedem ferret
Paulus ex Festo 94 ed. Lindsay
Whenever the fire of Vesta was interrupted, the Virgins were beaten by the pontifex; their custom was to bore a tabula of fēlix māteria until a fire could be taken and brought in a brazen cribrum to the sanctuary by a Virgin.
We may compare the usage of tabula ‘board’ here with the following instruction In Iguvine Tables IIb 12: {tafle e pir fertu} ‘саrrу the fire there on a tafla’, where Umbrian tafla is the equivalent, of Latin tabula. We may note too that the wood used to kindle the fire is called māteria. The form of this noun suggests that it is derived from māter ‘mother’. [73] In addition, the māteria is described with the word fēlix, an adjective designating fertility and the power of nurturing. Immediately comparable in theme is the expression ignem alere ‘nurture fire’ [74] and the cognate Indic theme of the fire-god as Mātariśvan, that is, the one who is nurtured inside the mother. [75] We may note too that the enclosure of the Indic Gārhapatya, the domestic fireplace, is actually called yóni- ‘female genitalia’ (Śatapatha-Brāhmaṇa [76]
In the same set of instructions where the text of the Iguvine Tables, quoted above, studiously avoids use of the Atiedian word ahti- to designate ‘movable fireplace’ (E-F), there does occur a synonym, spelled aso:
Fc erihont aso destre onse fertu
Iguvine Tables VIb 50
the same [= the Adfertor] shall carry the aso on his right shoulder
Because of the specification of the right shoulder in section Fc, what is not directly mentioned by name in Fa (pufe pir entelust ‘that in which be has placed the fire’) has to be mentioned, again, and this time it is done not by periphrasis but by use of an equivalent word for ‘movable fireplace’. This Umbrian word aso is apparently not part of the Atiedian sacral vocabulary, and it is probably for this reason that it could be written out in the tabu-conscious ritual instructions of VIb 48-53, whereas ahti- was not mentioned directly but by periphrasis. [77] Just as Umbrian asa {169|170} (Iguvine Tables IIa 38, etc.) can be reconstructed as *ā̌ssā-, so also aso (VIb 50) from *ā̌sso-. Removing the factor of geminated *s, we may reconstruct *aso- as *h2es-o-; in other words, I propose that Umbrian aso is the cognate of Hittite ḫašša- ‘sacrificial fireplace’.

The Meaning of Hittite ḫaš-/ḫašša-/ḫaššu- from the Standpoint of Myth and Ritual

Among the noun-reflexes of the Indo-European root *h2es-, our survey has suggested that the semantic basis is the notion of ‘fireplace’:
  • Hittite ḫašša-                     sacrificial fireplace
  • Indic ā́sa-                          ashes
  • Old Norse arinn, etc.         sacrificial fireplace
  • German Esse, etc.            smith’s fireplace
  • English ashes, etc.            ashes
  • Greek ἀσβόλη ἄσβολος     soot
  • Latin āra, altāria                 sacrificial fireplace, altar
  • Oscan aasa-                      sacrificial fireplace, altar
  • Umbrian asa                      sacrificial fireplace, altar
  • Umbrian aso                      sacrificial fireplace, altar (movable)
The semantically anomalous reflexes of *h2es- remain the Hittite verb ḫaš- ‘beget’ and noun ḫaššu- ‘king’. In light of the myths and rituals that we have surveyed, however, these meanings fit the broader context of the sacrificial fireplace as the generatrix of kingship and the authority of kingship, which has been all along the focus of this inquiry.
In this connection, we may add that the formula which the Hittite ḫaššu- uses in referring to himself is dUTUši ‘my sun’, as in the Autobiography of King Ḫattušiliš III (passim). This usage seems distinctly Hittite, in that there is no corresponding mechanism for designating “ego” + first person singular in Akkadian texts (where the expected form would have been ŠAMŠI ‘my sun’ + third person singular). In the Royal Funerary Ritual of the Hittites, [78] which features the cremation of the ḫaššu- ‘king’ and offerings at the ḫašša- ‘sacrificial fireplace’ (passim), one of the prime recipients of these offerings is the great state god dUTU ‘sun’; after the ḫaššu- has died, he joins this very god dUTU. [79] In fact, after the ḫaššu- has died, he himself becomes a god. [80] This belief also seems {170|171} distinctly Hittite, as we may see from the attenuated Akkadian translation of the following Hittite statement spoken by King Muršiliš II:
Otten 1958.120
[when] my father became a god
We may contrast the parallel Akkadian version:
Otten 1958.120
when my father went to his destiny
We may compare also the following prayer:
n[u-?]ka-ru-ú ma aḫ-ḫa-an an-na-za ŠÀ-za ḫa--ša-[a]n-za e-šu-un n[u-m]u-kán DINGIR-YA a-ap-pa a-pu-u-unZI-an an-da ta-a-I [nu-m]u tu-el ŠA DINGIR-YA ZI-KA am-mu-uk [ ] IGI-an-da [a]t-ta--ma-aš an-na-aš ḫa--ša-an-na-aš x x x [Z]IḪI.A ki-ša-an-ta-ru
Otten 1958.123-124
Already when I was begotten [ḫaš-] from the inside of my mother, then you, my god, put this “animus” [ZI = ištanza-] in for me; and may your divine “animus” become for me the “animī” of my father, mother, and “gēns.”
This theme brings us back to our starting point, that is, the Indo-European pattern of thought that links the rising of the sun at dawn as parallel to the kindling of the sacrificial fire. This parallelism, as we have seen, is explicit in the ritual language of the Vedas and it is implicit in the possible affinity between Indo-European roots in words for ‘dawn’, notably Greek ēṓs and Latin aurōra, and in words for ‘hearth’, notably Greek hestíā and Latin Vesta. In other words, the possibility remains that the macrocosm of dawn and the microcosm of sacrificial fire are designated with variants of the same root, with *heu̯s- for ‘dawn’ and *hu̯es- for ‘fireplace’. {171|172}
If indeed the Hittite ḫaššu- ‘king’, as we have just, seen, considers his identity to be that of the sun, it follows that the begetting of the king is parallel to the kindling of the sun; in that case, the Hittite verb ḫaš- ‘beget’ is thematically connected to the Hittite noun ḫaššu- ‘king’. From the etymological point of view, ḫaš- may then be translated as ‘kindle, light up’. For an example of the reverse in semantic development, we may consider English kindle, which had meant ‘beget’ (Middle English), then ‘set on fire’; another example is Old Norse kveikja ‘beget, kindle’ (the noun kveika means ‘fuel’). Finally, we may compare the Latin noun adulēscēns/adulēscēns ‘young man’, the participial origin of which reveals the built-in metaphor that we have already examined in detail: ‘becoming nurtured as fire becomes nurtured’.
Such an etymological interpretation of Hittite ḫaššu- as the ‘one who is lit up, kindled’ is reinforced by a well-known theme in Italic myth, concerning ritual fire. The protagonist of this myth is the Roman king Servius Tullius, whom Georges Dumézil has singled out as representing the features of the ideal king from the standpoint of patterns in Indo-European mythmaking. [81] If indeed it is valid to claim that Latin āra is related to Hittite ḫašša- ‘sacrificial fireplace’ and that Latin focus is the functional correlate of the āra, then the following myth of Servius and the focus is decisive:
non praeteribo et unum foci exemplum Romanis litteris clarum: Tarquinio Prisco regnante tradunt repente in foco eius comparuisse genitale e cinere masculi sexus eamque, quae insederat ibi, Tanaquilis reginae ancillam Ocresiam captiuam consurrexisse grauidam; ita Seruium Tullium natum, qui regno successit; inde et in regia cubanti ei puero caput arsisse, creditumque Laris familiaris filium; ob id Compitalia ludos Laribus primum instituisse.
Pliny Natural History 36.204
I will not pass over a famous example of the focus in Roman literature. In the reign of Tarquinius Priscus, they say that there suddenly appeared in his focus a genital organ of male sex out of the ashes, and that it impregnated Ocresia, who had sat there. She was an enslaved handmaiden of Queen Tanaquil. Thus was Servius Tullius born, and he succeeded to the kingship. When he was a boy sleeping in the palace, his head caught on fire, and he was believed to be the son of the Lār familiāris. For this reason he was the first to institute the Compitalia Games for the Lārēs.
This version can be supplemented with another: {172|173}
namque pater Tulli Volcanus, Ocresia mater praesignis facie Corniculana fuit.
hanc secum Tanaquil sacris de more peractis iussit in ornatum fundere uina focum.
hic inter cineres obsceni forma uirilis aut fuit aut uisa est, sed fuit illa magis.
iussa foco captiua sedet. conceptus ab illa Seruius a caelo semina gentis habet.
signa dedit genitor tunc cum caput igne corusco contigit, inque comis flammeus arsit apex.
Ovid Fasti 6.625-634
For the father of Tullius was Vulcan, and Ocresia of
Corniculum, distinguished in beauty, was his mother.
When the sacred rites were enacted, according to tradition,
Tanaquil ordered her to pour wine into the ornate focus.
At this point, among the ashes, there was, or seemed to be,
the male form of something indecent. More likely there was one.
Ordered to do so, the slave girl sat at the focus. Conceived
by her, Servius has the seeds of his gēns from the sky.
His father gave a sign, at the time when he touched his head
with flashing fire, and a flame lit up in his hair.
In this remarkable passage the preoccupation of the myth with a ritual context is especially clear. There is also a lengthy account of the same myth in Dionysius of Halicarnassus Roman Antiquities 4.2.1-4. Romulus and Remus themselves were begotten likewise, according to a myth recorded by Plutarch (Romulus 2.4-8). The same goes for Caeculus, founder of Praeneste and ancestor of the distinguished gēns Caecilia (Servius on Virgil Aeneid 7.678). [82]
I close by citing once more a striking detail from the myth about the begetting of Servius, the Italic king par excellence, from the sacrificial fireplace. As we have seen in both versions just quoted, there is an outward sign that warrants the truth of the king’s being generated from the embers of the hearth. To mark the moment that his kingship is revealed, the head of Servius literally lights up. The radiant visage of the king, the ideal human, is a theme that may be linked with the etymology that I have already suggested for Greek ánthrōpos (ἄνθρωπος) ‘human’, that is, {173|174} ‘he who has the looks of embers’. [83] In line with a ubiquitous theme of myth making, that the first human is the first king, [84] this image of the king with a visage glowing from the fire of the hearth is a symbol for the never-ending search of myth to grasp the celestial affinities of humankind. [85] We come back, full circle, to Clytemnestra’s dream in the Electra of Sophocles (417-423): in the vision, to repeat, the king of Mycenae places his scepter into the royal hearth (419-420), and from it grows a shoot so vigorous that it covers with its shade all the kingdom of Mycenae (421-423). The hearth, as we have seen, is the focus for reveries about the father’s generation of a son without the exogamous intermediacy of a female outsider. The female insider, the ultimate generatrix, is in such reveries the hearth itself, and from it emanates the essence of authority, made manifest in the radiant visage of the ideal king.

Appendix. Conflicting Semiotics of Cremation, Inhumation, Exposition: An Iranian Case in Point

In Avestan usage, the root *ā̌s- (from *h2es-) is attested as the component *ah-(ya-) in the compounding of sairiia- plus *ahiia- = sairiie.hiia-. The word sairiia- designates the dried manure used as a proper funerary resting place for the corpse, Vendidad 8.8; the word sairiie.hiia- is attested only once, Vendidad 8.83. It has been traditionally interpreted to mean ‘apparatus for drying manure’, [86] where the semantics of root *ā̆s- correspond to what we find in Latin ārēre ‘be dry’, Tocharian as- ‘dry’, and so on. [87] I propose an alternative explanation of sairiie.hiia-, interpreting its etymology as ‘apparatus for burning manure’. As we shall see, this interpretation helps explain some crucial details of conflicting Iranian ideologies concerning the funerary practices of cremation, inhumation, and exposition. {174|175}
According to Zoroastrian precepts, exposing a corpse to be eaten by dogs and birds is the proper funerary procedure, rather than cremation or inhumation (Vendidad 8 passim). I draw attention here to the contrast with the Indic custom of cremation (e.g. Rig-Veda 10.16, etc.), [88] and with the Greek customs of cremation and inhumation, as discussed earlier. [89] More specifically, I also draw attention to the contrast, established in the Iliad, between the sacredness of cremating a corpse, which I have argued is considered the key to successful afterlife, [90] and the abomination of exposing a corpse to be eaten by dogs and birds, a custom cited at the very beginning of the Iliad (I 4-5) and pervading the rest of the epic as the ultimate image of inhumanity, a symbolic threat to the very afterlife of the deceased. [91]
Zoroastrian ideology, in symmetrical contrast, not only sanctions the exposition of the corpse to dogs and birds; it also singles out the custom of cremation as an abomination, and there are elaborate protective rituals for the true believer to follow in the event that he should come upon ātrǝm nasupākǝm ‘a corpse-cooking fire’ (Vendidad 8.73 and following). Such clear provisions for the eventuality of discovering the practice of cremation suggest that this funerary procedure, though forbidden by the Zoroastrian norm, was widespread m various areas of Iranian society.
In one instance the people of an entire region are singled out for traces of this particular aberration from orthodoxy: in the first book of the Vendidad, a tract against daēuua-s ‘demons’, the Zoroastrian religious community is represented by sixteen regions of Iranian society, and from among these, the thirteenth “best” region, called Čaxra ‘The Chariot-Wheel’, is described as being tainted with the practice of “corpse cooking” (Vendidad 1.16). We may compare these other aberrations from Zoroastrian orthodoxy:
The tenth “best” region, called Haraxvaitī (= Old Persian Harahuvatī in the Behistun Inscription, = Arachosia), is tainted with the practice of corpse burying, that is, inhumation (Vendidad 1.12).
The sixth “best” region, called Harōiuua (= Old Persian Haraiva, = latter-day Herāt), is tainted with the practice of keening or funeral dirges (Vendidad 1.8; sraskǝmča driuuikāča ‘weeping and howling’). {175|176}
In the latter case we may note again a symmetrical contrast with Greek customs, as reflected in the Iliad, which not only describes the practice of keening or funeral dirges as the norm but equates the very essence of this institution with the characterization of its main hero, Achilles, that man of constant sorrow. [92] The Zoroastrian ideology shows the converse, where the very building that is designated for the exposition of the corpse is conceived as the Tower of Silence in Parsee usage. [93]
At the end of the first book of the Zoroastrian Vendidad (1.20), it is pointed out that there are other regions in the Zoroastrian community-at-large besides the sixteen that are formally listed. As for the choice of the sixteen “best” regions and their arrangement in descending order of value, the desired effect is to symbolize the geographical spread of Zoroastrian orthodoxy. [94]
At the top of the list in Vendidad 1 are those regions that were the first to accept Zoroastrian orthodoxy:
  • 1st:       Airiianǝm Vaējah = Ariana
  • 2nd:      Suγδa = Sogdiana
  • 3rd:       Mouru = Margiana
  • 4th:       Bāxδī = Bactriana
  • 5th:       Nisāiia
  • 6th:       Harōiuua = Arīa
It has been argued the best Zoroastrian region of all, “the Aryan Vaējah,” homeland of Zaraθuštra = Zoroaster, is to be identified as Xvārizm = Chorasmia. [95] The Avesta explicitly connects Zaraθuštra with “the Aryan Vaējah” (Yašt 5.17-18, 104), and it was at the river Dāitiiā, closely associated with this region, that Zaraθuštra made sacrifice (Yašt 5.104, 15.2). The precise localization of “the Aryan Vaējah,” which counts as the sacred space of Zoroastrianism itself, seems to have varied in the course of time, following the shifting localizations of power and influence, and it seems clear that Chorasmia, even if it merits the title “the Aryan Vaējah,” was not the only region to be described this way. [96] The point remains, in any case, that the six regions heading the list of Vendidad 1 are apparently to be located in East Iran, visualized as contiguous with {176|177} each other, and that they are the nucleus of Zoroastrian orthodoxy, from where it spread to regions such as Čaxra.
On the steppes of Central Asia in general, of which East Iran forms a part, the poorly wooded terrain makes cremation impractical, and it is no coincidence that the alternative custom of exposition is a characteristic feature of the peoples living in the Central Asiatic steppes, including the Mongols. [97] Since the nucleus of Zoroastrian orthodoxy is to be located in the East Iranian steppes, it follows that the Zoroastrian custom of exposition was an areal feature acquired by the East Iranians from their Central Asiatic neighbors. As Zoroastrian orthodoxy spread, the custom of exposition came into conflict with that of cremation, such as practiced by the people of Čaxra. The specific mention of “corpse cooking” as the plague of Čaxra (Vendidad 1.16) suggests that the inhabitants clung to an older custom that was difficult to uproot. [98] Given the clearly attested Indic custom of cremation (e.g. Rig-Veda 10.16, etc.), [99] the Iranian attestations of non-Zoroastrian “corpse cooking” suggest an Indo-Iranian pedigree for the custom of cremation as opposed to exposition.
There is also direct evidence that the Zoroastrian custom of exposition was generally preceded by that of cremation; the actual Zoroastrian word designating the place built for exposing the corpse is daxma (Vendidad 5.14, 8.2), which from an etymological point of view means “burning” (whence “place for burning, cremation”) from verb dag- ‘burn’ (as attested in Yasna 71.8, etc.). In other words, I am arguing that the original place of the funeral pyre was converted into the place of exposition, without so much as a change in the word used to designate the place itself. [100] There are instances where the word daxma at least implies a {177|178} place of cremation: in Vendidad 7.49-58 (cf. also 3.13), we see variations on the theme of an “illegal” daxma, described as a place frequented by (daēuua-s ‘demons’, fiends who are the primordial enemies of Ahura, head of the pantheon. In line with the opprobrium of “corpse cooking,” it seems that the daēuua-s, are being described as actually devouring the dead who are “cooked” at the daxma (Vendidad 7.55).
The fact that Zoroastrian teaching holds cremation to be an abomination has a bearing on the context of sairiie.hiia- in Vendidad 8.83. In Vendidad 8.81-9, there is a catalogue of merits to be gained by bringing various kinds of fire to the central fire of purification; the more impure the fire, the greater the merit. The reasoning behind this mentality, as reflected to this day by the ritual practices of the Zoroastrian Parsees, has been described as follows: [101]
Since the goal of all these procedures is to obtain a fire that is as pure as possible, the question remains why it is necessary to use for the procedures, among other things, the fire that is the most impure that one can imagine, that is, the fire that has burned a cadaver. Clearly, the reason is that the goal is also to deliver the fire from its impurity, to save it.
In Vendidad 8, the most impure fire of them all is “corpse-cooking” fire:
yō atrǝm nasuta dāitīm gātum auui auua.baraiti
Vendidad 8.81
who brings corpse-cooking fire to the prescribed place…
In this case, the person who brings such impure fire to the central fire of purification merits 10,000 firebrands.
The second in rank among all impure fires is described as follows:
yō atrǝm uruzdipākǝm dāitīm gātum auui auua.baraiti
Vendidad 8.82
who brings fluid-cooking fire to the prescribed place…
In this case, the person who brings such impure fire to the central fire merits 1,000 firebrands. The reference to “fluid” here seems to concern fluids emanating from the body: Dēnkart 8.46 offers the explanatory description hixr pāk ‘excrement cooking’. [102] {178|179}
The third in rank among all impure fires is described as follows:
yō atrǝm sairiie.hiiaṯ hača dāitīm gātum auui auua.baraiti
Vendidad 8.83
who brings fire from the sairiie.hiia- to the prescribed place…
In this case, the one who brings such fire merits 500 firebrands.
From then on, the catalogue lists fires destined for secular uses, such as the fire from a potter’s fireplace (Vendidad 8.84), from a goldsmith’s fireplace (8.87), from a baker’s fireplace (8.91), and so on. Last on the list is the fire that is easiest to bring, namely, “from the nearest place”:
yō atrǝm nazdištat hača dāitīm gātum auui auua.baraiti
Vendidad 8.96
who brings fire from the nearest place to the prescribed place…
In this case, the bringer merits 10 firebrands.
The essential question remains: why does fire from the sairiie.hiia- rank so high in degree of abomination that it should be listed directly after fire for burning the body and after fire for burning fluid discharge from the body? The answer may well be concealed in the use of sairiia- ‘manure’ as a resting place for the corpse:
auua.hē gātūm baraiiǝn ātriiehe vā sairiiehe vā
Vendidad 8.8
they should bring for him [= the corpse] as a place either ashes or manure
The context shows that this practice follows the dictates of Zoroastrian orthodoxy, just like the practice of exposing the corpse in the daxma. Yet the daxma, if my argument holds, was at an earlier stage the place of cremation, not exposition. Similarly, I propose, sairiia- ‘manure’ was at an earlier stage a fuel, or an ingredient in the fuel, for cremation. In the Zoroastrian orthodoxy, use of the term daxma was retained but converted to designate the place of exposition rather than cremation. Similarly, I suggest, any use of manure as fuel for cremating the corpse would have to be converted: the body is to be laid out on manure, but neither the body nor the manure may be burned. We must note that the custom of using manure as an ingredient for cremation lies survived in latter-day {179|180} India. [103] Moreover, manure is the common domestic fuel in latter-day India. If the custom of using manure for fuel is of Indo-Iranian provenience, then Avestan sairiie.hiia- may have at an earlier stage designated simply a place where manure was burned.
To sum up: Zoroastrian orthodoxy prescribes manure as a resting place for the corpse; since corpse burning is forbidden, it follows that manure burning should also be forbidden because of the surviving association of manure with the resting place of the corpse. Because of this association, the use of manure for secular fuel may be forbidden along with its use for cremating the corpse.
In fact, the custom of burning manure may be of Indo-European provenience: we may consider the Latin noun fimus ‘manure’, apparently derived from -fiō as in suffiō ‘fumigate’ (Cato De re rustica 113.1 ) or ‘burn for the purpose of fumigation’ (Pliny Natural History 28.42, etc.); we may compare the root-formation *dhu̯Ǐ- of suffiō with the *dhū- of fūmus ‘smoke’. [104] We may note also the reports about the stercus ‘manure’ that is ritually swept out of the precinct of Vesta, Roman goddess of the domestic fireplace (Varro De lingua latina, 6.32 Festus 344 Lindsay). [105]
The subject of fumigation brings this presentation to a close. And aptly so, since the very concept of fumigation is pertinent to the focus of the entire study, the setting of the sacrificial fireplace. I cite the formation of fūmigo ‘fumigate’, which happens to be parallel to pūrgō ‘purify’, from an earlier pūrigō (as in Plautus Miles 177). [106] Following Rudolf Thurneysen, [107] I interpret pūrgō/pūrigō ‘purify’ as derived from an underlying expression *pūr agere ‘carry fire’, formally parallel to rēmigō ‘row’, derived from an underlying expression rēmum agere by way of the intermediate formation rēmex, rēmigis. Against the conventional rejection of Thurneysen’s positing an underlying *pūr agere ‘carry fire’, [108] I cite the Umbrian collocation pir ahtimem ententu ‘place fire in the ahti-’ in Iguvine Tables Ib 12, where pir (<*pūr) is combined with the abstract noun ahti- (<*ag-ti-) derived from a verb surviving in Latin as agere. [109] The Umbrian ahti-, receptacle of the sacred fire, is the source of purification for the community. {180|181}


[ back ] 1. In Iliad I 233-237, this same skêptron ‘scepter’ is viewed as a thing of nature that has been transformed into a thing of culture; commentary in N 1979a.179-180, 188-189. Here in Sophocles Electra 421-423, the transformation is in the other direction. On the cult of Agamemnon’s skêptron at Khaironeia, where its local name is the dóru ‘wood, shaft’, see áphthiton ‘imperishable’ as applied to the skêptron at Iliad II 46, 186.
[ back ] 2. Vernant 1985, esp. pp. 165-169; cf. also Gernet 1968.387.
[ back ] 3. Vernant pp. 156-157.
[ back ] 4. Vernant pp. 163-165.
[ back ] 5. Concerning the ritual of Amphidromia, where the naming of the newborn child literally revolves around the hestíā, see Vernant pp. 189-195: for the naming to be formalized, the father runs around the hearth carrying his new baby on the fifth day after birth and then sets down the child in the sacred area thus circumscribed (scholia to Plato Theaetetus 160e; Hesychius s.v. Δρομιάμφιον ᾖμαρ; scholia to Aristophanes 758). On the Eleusinian ritual concept of the παῖς ἀφ’ ἑστίας ‘boy from the hearth [hestíā]’ (Harpocration s.v. ἀφ’ ἑστίας Anecdota Graeca 204.19 ed. Bekker), see the discussion of in Vernant pp. 164-168 and the updating of sources in Burkert 1983.280n31.
[ back ] 6. Vernant p. 187.
[ back ] 7. Vernant pp. 181, 186.
[ back ] 8. Vernant p. 186.
[ back ] 9. Cf. Beneviste 1962.14.
[ back ] 10. Benveniste p. 14.
[ back ] 11. For a survey of various etymologies that have been proposed for this Hittite verb, see Tischler 1983.191-194.
[ back ] 12. See the survey in Tischler pp. 207-209, who lists other suggested etymologies as well.
[ back ] 13. The Hittite verb ḫaš- ‘beget’ is spelled with a single s in the third person singular (ḫa-a-ši) and with a double s in the third plural (ḫa--ša-an-zi). Such derivatives as ḫaššatar ‘begetting, gēns’ show double s (ḫa--ša-tar) and so too the proposed derivative ḫaššu- (ḫa--šu-). By contrast, consider the Luvian and Palaic adjective wašu- ‘good’, with single s, which seems to be derived from the verb wašš- ‘be agreeable’, attested in Hittite with double s. Instead of wašu-, however, the Hittite word for ‘good’ is aššu-, with double s. Even if we are not prepared to explain them, it is important to note the existence of such s/ss variations. I find a similar s/ss problem in the contrast of Latin āra and Umbrian asa, both meaning ‘sacrificial fireplace, altar’. Like Latin, Umbrian rhotacizes single intervocalic *-s-, so that we have to reconstruct an inherited Italic contrast of *āsā vs. *ā̆ssā in order to account for the respective Latin and Umbrian forms. Again, I merely note the existence of this s/ss variation, rather than attempt an explanation.
[ back ] 14. Benveniste 1969 2:85.
[ back ] 15. Dumézil 1954.34-35.
[ back ] 16. See p. 104.
[ back ] 17. See pp. 104ff.
[ back ] 18. Dumézil 1954.34.
[ back ] 19. Cf. p. 104, where Rig-Veda 7.9.3 is quoted.
[ back ] 20. Dumézil 1954.34-35. In Ionic and Other Greek dialects as well, there is a variant of Attic hestíā (ἑστία) namely histíā (Ionic ἱστίη): see DELG 379. I draw attention to the raising of *e by *i after the labial *u̯ in histíā, to be reconstructed as *u̯istíā. Such replacement of *e by *i in the vicinity of labials is a feature of the “standard” Mycenaean dialect: see Householder and Nagy 1972.784-785. In other words, there is a possibility that the variant form histíā is a reflex of the “standard Mycenaean” dialect of the second millennium B.C. For a working definition of “standard Mycenaean,” see Risch 1966 (updating in Risch 1979). For an alternative explanation of the *i in histíā, see Vernant 1985.199-200.
[ back ] 21. The absence of *h2 before *u̯ in a hypothetical Greek formation *u̯estiā may conceivably be explained on the basis of a combination of morphological and phonological factors. First, the morphology: the suffix-formation of *u̯estiā, as DELG 379 points out, suggests that this noun is derived from an adjective *u̯esto-, or possibly from *u̯estā-. Formations like *u̯esto- are typical of what we find in the second half of compounds. Which brings us to the second consideration that of phonology. It appears that laryngeals (*h1, *h2, *h3) can disappear without trace in the second half of compounds (see Beekes 1969.242-243 for a list of examples; cf. Mayrhofer 1986.125, 129, 140). If, then, we accept the argument in DELG that the Greek noun *u̯estiā may be a derivative of compound-formations, then we may expect phonological instability in the retention of laryngeal reflexes. Even more important, there is another factor leading to a pattern of phonological instability in Greek *u̯estiā: in many inscriptions featuring dialects that normally retain initial *u̯, the expected digamma (= *u̯) of *u̯estiā is not spelled out (DELG 379). What seems to be at work here is the force of analogy: we may posit a pairing, in Greek usage, of hestíā (ἑστία) with another word designating ‘fireplace’, eskhára, (ἐσχάρα); it is clear that never had an initial *u̯, as DELG 379 makes clear. Besides the semantic convergences between these two words for ‘fireplace’, hestíā and eskhára, we should note, already at this point, an important divergence that is pertinent to the semantics of other words for ‘fireplace’ to be studied later on in this presentation: unlike the hestíā, the eskhára is potentially moveable: see Risch 1981 [1976].537 (also DELG 379-380).
[ back ] 22. Where C = consonant. For the formulation of this Indo-European pattern of root-variation, see Kuryɫowicz 1927 (pace Benveniste 1969 1:22-25). Some suggested examples of CeC(C)- vs. Cu̯eC(C)-: *h1esu- vs. h1u̯esu-, as in Greek ἐυ- ‘good’, Hittite aššu- ‘good’, vs. Indic vásu- ‘good’, Iranian (Avestan) vohu- ‘good’, Luvian and Palaic wašu- ‘good’; *teks- vs. *tu̯eks-, as in Indic takṣ- ‘fashion’ vs. Indic tvakṣ- ‘fashion’, Iranian (Avestan) θuuaxš- ‘fashion’: *h1ers- vs. *h1u̯ers, as in Indic arṣ- ‘flow’, Hittite arš- ‘flow’ vs. Indic varṣ- ‘rain’, Greek ἕρση/ἐέρσν ‘dew’, Hittite warša- ‘dew’.
[ back ] 23. Moreover, the reconstruction *h2u̯es- may possibly fit the Hittite ḫuiš- ‘live’ (if, however, Luvian ḫuit- is a cognate, on which see Tischler 1983.264-266, then this connection is to be rejected). If Hittite ḫuiš- is conceivably cognate with Indic vas- ‘shine’, the meaning ‘live’ rather than ‘shine’ would be in line with the semantics of Hittite ḫaš-, meaning ‘beget’ rather than ‘set on fire’, despite the meaning of ḫašša- ‘sacrificial fireplace’. We may note that the suffix of the Hittite adjective ḫuišwant- ‘alive’ is cognate with that of the Indic adjective vivásvat-. The root *h2u̯es-, this hypothetical reconstruction from Greek hestíā ‘hearth’ (and conceivably from Hittite ḫuiš- ‘live’), is not to be confused with the *h2u̯es- of the Greek aorist áesa (ἄεσα) ‘spend the night’, which is a variant of *h2eu̯s- as in the Greek present iaúō (ἰαύω) ‘sleep’. The distinctness of the roots is reflected in Indic, where vas- ‘spend the night’ (present third singular vásati) is conjugated differently from vas- ‘shine’ (present third singular uccháti).
[ back ] 24. Alternatively, we may conceivably reconstruct Greek aúōs/ēṓs as *h2us-os- instead of *h2eu̯s-os- (cf. Peters 1980.31-32). In any case, the spelling of Aeolic αὔως reflects an underlying *au̯u̯ōs; the gemination of *u̯ reflects the Aeolic device for poetic lengthening of initial syllables with shape *Vu̯- (where V = vowel), on which subject one of the most informative works remains Solmsen 1901. By constrast, the Attic-Ionic device for poetic lengthening of initial syllables with shape *Vu̯- is not gemination of the *u̯ but lengthening of the *V-: *hau̯ōs to *hāu̯ōs to ἠώς in Ionic (Homeric), ἕως in Attic. For a summary of the diachronic motivations for poetic lengthening of the initial syllable, see Kuryɫowicz 1956.264-269; also Householder and Nagy 1972.754. Alternatively, *au̯u̯ōs and *āu̯ōs may perhaps be direct phonological reflexes of *ausōs.
[ back ] 25. Just as the posited root-variant *h2ues- may have survived in Greek as hestíā ‘hearth’ (ἑστία), without a phonological trace of *h2 (see n21), so also the root-variant *h2eu̯s- apparently survives as heúō (εὕω) ‘singe’, again without a trace of *h2. Cognates of Greek heúō are Latin ūrō ‘burn’ and Indic óṣati ‘burn’. In this case the loss of *h2 in the Greek reflex of the posited root *h2eu̯s- may be attributed to the secondary nature of e-grade formations of the type heúō/ūrō/óṣati, apparently built on the zero-grade of the root (on such a pattern of derivation, see Kuryɫowicz 1969.221). Parallel to secondary verb-formations of the type óṣati ‘burn’, Indic preserves secondary adjective- and noun-formations of the type –óṣa-, as in dur-óṣa- ‘hard to kindle’ and óṣa-dhī ‘plant’. I interpret the latter form as a compound consisting of the roots uṣ- (from *h2us-) ‘light’ and dhā- (from *dheh1-) ‘put, place’, meaning something like ‘light-emplacement’; see p. 103. For thematic evidence in support of this etymology, see pp. 102ff. For a survey of the etymological possibilities of óṣa-dhī ‘plant’, including the one suggested here, see again Minard 1956.268. We may compare the semantics of the English idiom “set on fire.”
[ back ] 26. More on Manu at pp. 110ff.
[ back ] 27. The root nr̥t- ‘dance’ in the epithet of Uṣas, sūnr̥tāvari may be compared with the collocation of khoroí ‘dances’ and Ēṓs ‘Dawn’, as at Odyssey xii 4. For more on the goddess Eos ‘Dawn’ and her relation to the dance, see Boedeker 1974.58-63, 87-88.
[ back ] 28. Cf. p. 145n13 above.
[ back ] 29. As we may possibly infer from such contrasts as in Latin ăcuō vs. ācer.
[ back ] 30. For an especially interesting attestation, let us consider the context of ā́sa- ‘ashes’ at Śatapatha-Brāhmaṇa, where ā́sa- is being described as a creative substance. The ā́sa- is what becomes of the aṅgā́ra-s ‘coals’ mentioned in Śatapatha-Brāhmaṇa, where the aṅgā́ra-s are in turn described as the creative substance from which the priests known as the Áṅgiras-es originate (cf. EWA 48). Cf. also Aitareya-Brāhmaṇa 3.34, and Śatapatha-Brāhmaṇa,, We may compare the forms and meanings of Indic aṅgā́ra- ‘coal’ and Áṅgiras-, the name for the fire-priests, with the forms and meanings of Greek ánthrax ‘coal’ and ánthrōpos ‘human’, which I interpret etymologically as ‘he who has the looks of embers’. In a future study I hope to connect this proposed etymology with the context of thūmálōps ‘piece of burning wood, charcoal’ in Aristophanes Acharnians 321, which may be a veiled reference to a local anthropogonic tradition; see especially ánthrakes ‘charcoal’, as associated with the Acharnians, at lines 34, 332. On the anthropogonic theme of First Man as First Sacrificer, see pp. 70, 110-111. (On the possibility of a related theme, that of First Man as First Mántis ‘Seer’, see p. 198n120.) I draw attention to two of the mock-names of members of the chorus of Acharnians: Marīládēs ‘son of embers’ (Acharnians line 609), derived from marī́lē ‘embers of charcoal’ (as at line 350: see p. 198n120), and Prīnídēs ‘son of holm oak’ (line 180; on the connections of anthropogony with the material of wood, see p. 198n120). The latter description of the Acharnians is coupled with stiptoí ‘tough’ (line 180), in the sense of “compressed by treading down,” like the charcoal used by smiths (Theophrastus On fire 37; cf. Sommerstein 1980.176); moreover, the ánthrakes ‘charcoals’ are described as prī́ninoi ‘made of holm oak’ (ἀνθράκων πρινίνων 668). Thus the “tough stuff” that men are made of is also the stuff of charcoals (on the wood used for charcoal by the Acharnians, see also Sommerstein p. 171). A pervasive theme of the Acharnians of Aristophanes is the ridiculing of the Acharnians’ feelings of solidarity and even affection toward the ánthrakes ‘charcoals’ (325-341), which they treat as animate beings, as their dēmótai ‘fellow district-members’ (349; cf. 333).
[ back ] 31. The lengthening of the radical vowel in ā́sa- is secondary: see Kuryɫowicz 1968.282-283 on the phenomenon known as Brugmann’s Law.
[ back ] 32. I accept the etymological interpretation “Aschen-wurf”: Schwyzer 1939 1:440; for the absence of -o- between -ἄσ and βολος-, cf. ἐπεσ-βόλος, κερασ-βόλος.
[ back ] 33. As in Old Church Slavonic polěti ‘be on fire’.
[ back ] 34. For an example of a Latin form in which rhotacism precedes syncope, cf. ornus ‘mountain-ash tree’, from *Ŏrenos from *Ŏsenos (cf. Old Slavonic jasenĭ ‘ash tree’), on which see DELL 469. Also Leumann 1977.96, 99, who contrasts the type ornus with the type pōnō (from *po-sinō). Consider also Faler-nus (from *Falis-isnos; cf. Falis-).
[ back ] 35. The Indo-European stative *-ē- is familiar even from the internal evidence of Latin: calēre ‘be hot’, tepēre ‘be warm’, albēre ‘be white’, etc.
[ back ] 36. Cf. p. 235. For other such doublets, see Schwyzer 1939.703.
[ back ] 37. We may compare the secondary length in the perfect and causative as ās- (vs. present as-) with the long radical vowel of Latin ārēre.
[ back ] 38. Cf. DELL 51-52.
[ back ] 39. This rejected etymology figures among those entertained by Sommer 1914.66-67; note Sommer’s argument that arfet, putatative formal equivalent of ardet, is a textual corruption.
[ back ] 40. Granted, a verb like gaudēre presupposes an original formant *gāu̯id-, latent in the participle gauīsus; note, however, that there is no trace of any *ārid- in the original participle of ardēre, which is, to repeat, assus (again, DELL 51-52).
[ back ] 41. J. Kuryɫowicz’s “Fourth Law”: “Quand à la suite d’une transformation morphologique une forme subit la différentation, la forme nouvelle correspond à sa fonction primaire (de fondation), la forme ancienne est réservée pour la fonction secondaire (fondée).” Quoted from Kuryɫowicz 1966 [1945-1949].169.
[ back ] 42. For further semantic analogues, cf. Reichelt 1914.313-316; Reichelt’s interpretation of āra differs from the one presented here.
[ back ] 43. For the secondary character of the lengthened radical vowel overt in Latin and Oscan, cf. the lengthening in ārēre; also, cf. again such ă/ā variations as in ăcuō vs. ācer, etc.
[ back ] 44. Cf. pp. 144-145.
[ back ] 45. DELL 24.
[ back ] 46. Cf. Cicero De natura deorum 3.37; Livy 21.1.4; Pliny Natural History 2.236; Ovid Metamorphoses 10.173, Remedia Amoris 808; Tacitus Germania 45, Annals 15.38, Histories 3.71; etc.
[ back ] 47. Cf. Whitney 1896.446. For the phonological development from *alto- ā̌si- to altāri-, with deleted *-o-, cf. the types magn-animus, rēm-ex, etc.
[ back ] 48. Whether Indic agní- ‘fire’ originates from *egni- or *n̥gni- is irrelevant to the reconstruction of iddhā́gni-.
[ back ] 49. The derivation of aráṇi from *al- ‘nurture’ is communicated as a possibility by R. Hauschild to M. Mayrhofer; cf. KEWA s.v. (this possibility is more recently rejected in EWA s.v.). There are numerous morphological parallels to the proposed derivation of aráṇi-, in the reconstructed sense of ‘nurturing, nourishment’, from *al- ‘nurture’: I cite for example Rig-Vedic dhumáni- ‘blowing’, from dham- ‘blow’ (for a list of other such examples, cf. Wackernagle and Debrunner 1954.207); cf. also p. 186n37 below. For an interesting analogue to the proposed semantic specialization of an abstract noun like aráṇi- into the concrete designation of the wood with which the fire is “nurtured,” I cite the post-Vedic meaning of taráṇi-: it is no longer abstract ‘crossing’—from root tar- ‘cross’—but rather concrete ‘ship’ or concrete ‘sun’. In other words, the abstract notion of the act of crossing becomes specialized as the concrete notion of the means of crossing. (For a discussion of the process whereby abstract nouns become concrete, see N 1970.63-65, 68-70; cf. also Ernout 1954.179-83 ch. 6, “Passage de l’abstrait au concret.”) On the mythical themes of the sun’s crossing from the realm of light to the realm of darkness and back, see N 1979a.192.210.
[ back ] 50. Rig-Veda 3.29.2 is quoted at p. 103.
[ back ] 51. I note in passing that there are comparable themes concerning the infant Hermes in the Homeric Hymn to Hermes.
[ back ] 52. More on Mātariśvan at pp. 103ff.
[ back ] 53. For morphological parallels, cf. the Rig-Vedic derivative śáryāta- from śárya- ‘reed, arrow’; the derivational expansion of the base from -a- to -āta- is especially marked in names of plants or trees, as in āmrāta- ‘Spondias mangifera’, derived from āmrá- ‘mango’ (cf. Wackernagel and Debrunner 1954.269).
[ back ] 54. Schulze 1966 [1927].215-216. Cf. EWA 70.
[ back ] 55. Schulze pp. 215-216.
[ back ] 56. The *h2 disappears before *o. The etymology of Umbrian uřetu as *olētōd, equivalent of Latin (ad-)olētōd ‘let him burn’, is suggested en passant by Thurneysen 1907.800.
[ back ] 57. Leumann 1977.86.
[ back ] 58. A sacrifice is being described, at which Lavinia’s hair seems to catch on fire.
[ back ] 59. Note, however, the absence of Latin "ignem adolēre."
[ back ] 60. Cf. DELL 23.
[ back ] 61. Acts of the Arval Brethren aa. 90.49-50; Domitian-era C I 2-5; 105 II 7-9; 118 I 59-62; 120.36-37; 155.32-34; M. Aurelius-era E 1-2; 183 II 21-22; 218 a 17-19; 240 (= Dessau 9522) II 4.
[ back ] 62. For a standard accounting from an abidingly useful manual: Wissowa 1912.475.
[ back ] 63. Context: before sailing on, the retinue of Claudia Quinta pauses to sacrifice a heifer.
[ back ] 64. The words fouet and fouetur here imply an etymological connection between verb foueō ‘foster’ and noun focus; but note the lengthened o in fōculum, the derivation of which from foueō seems assured by the collocation of the two words in Plautus Captivi 847. Thus we must distinguish between fōculum, derivative of foueō, and fŏculus, derivative of focus (as in Cato De re rustica 11.4). For attestations showing the distinct vowel lengths, see DELL s.v. foueō. The meaning of fōculum, conveying the notion of “fostering” or “nurturing” fire, is analogous to that of altāria as discussed above. Perhaps the phonological closeness of fōculo- and fŏculo- is influenced by their semantic parallelism, in that both words designate a movable fireplace. It is conceivable that this parallelism masks an earlier form *faculus, derived from *facus, and that *facus was reshaped as focus on the model of an analogized foculus. In which case, focus could perhaps be explained as a secondary derivative of faciō, meaning something like ‘setting’. More on this possibility in the discussion that follows.
[ back ] 65. DELL 243.
[ back ] 66. Cf. Umbrian façia, faç(i)u, fakust, facurent, which are functional equivalents of Latin faciat, facere, fēcerit, fēcerint respectively. Moreover, there may be a trace of e-grade in the Umbrian imperative fetu, also spelled feitu. This form cannot correspond to Oscan factud, the Umbrian cognate of which would be *faitu: we may compare the Umbrian imperative aitu with the corresponding Oscan actud. Thus it is possible to reconstruct Umbrian fe(i)tu as *fekitōd. An argument against the alternative possibility, *fēkitōd, is that the inherited ē of Umbrian is regularly spelled i in the Latin alphabet as opposed to e in the native Umbrian alphabet (e.g. filiu in the Latin alphabet vs. feliuf in the Umbrian). Yet, what corresponds to the native Umbrian spelling fetu/feitu is the Latin spelling fetu/feitu/feetu, never *fitu.
[ back ] 67. See n64. Furthermore, the possible derivation of focus from the root of fac- may have a formal parallel: the noun iocus ‘jesting word(s)’ can be derived from the verb iaciō ‘throw, hurl’, with root *i̯eh1(-k-) as in the perfect iēcī. For the semantics, we may compare Greek epes-bólos, literally ‘thrower of words’, as in Iliad II 275: here the epithet is applied to Thersites as an exponent of blame poetry, which is characterized by words of damaging ridicule (on which subject see N 1979a.253-264, esp. p. 264). The *-k- of zero-grade *i̯h1(-k-) in present-tense iaciō is again a secondary extension from the perfect iēcī, just as in faciō and fēcī. Even in Classical Latin, iaciō is frequently used with direct objects denoting things said (e.g. contumeliam ‘insult’ in Cicero Pro Sulla 23). On the semantics of Umbrian iuka ‘sacred words, formula’: Poultney 1959.199 and Borgeaud 1982.190. Lithuanian juõkas may be a borrowing from German jōk, in turn a borrowing from Latin by way of Studentensprache: LEW 197.
[ back ] 68. For my interpretation of ahti-, I have been guided by the critical discussion of textual evidence in Devoto 1937.267-268, 385-386; also Poultney 1959.165.
[ back ] 69. On the Umbrian word trifu- ‘tribe’ (cognate of Latin tribus), see p. 278.
[ back ] 70. See p. 161. For other parallelisms between the Iguvine Tables and the Acts of the Arval Brethren, see Vine 1986.
[ back ] 71. For Celtic parallels to the Italic concept of “Adfertor”: Borgeaud 1982.31, with bibliography.
[ back ] 72. See Ernout 1961.126.
[ back ] 73. See p. 106.
[ back ] 74. See p. 155.
[ back ] 75. See p. 156; also, p. 106.
[ back ] 76. See pp. 105-106.
[ back ] 77. This interpretation of Umbrian aso is offered as an alternative to the one found in the handbooks, where aso is understood as ‘roast meat’ or the like (cf. Latin assum). The contextual disadvantages of the latter interpretation are apparent from the discussion by Devoto 1937.268-269 and Ernout 1961.111.
[ back ] 78. The texts of the Royal Funerary Ritual have been collected by Otten 1958. For a survey of the specific passages dealing with the afterlife of the king, see Otten pp. 113, 119-120.
[ back ] 79. Otten pp. 113, 119-120.
[ back ] 80. Otten pp. 113, 119-120.
[ back ] 81. Dumézil 1943.
[ back ] 82. For a conscientious collection of testimonia about the Italic myths, see Alföldi 1974.182-185; also Bremmer and Horsfall 1987.49-53. Cf. Brelich 1949.70, 96-100; also Dumézil 1966.69n1, 320-321, who adduces Indic parallels; for example, one particular epic figure is begotten by the fire-god Agni in the Gārhapatya ‘domestic fireplace’ (Mahābhārata 3.213.45ff).
[ back ] 83. See p. 151n30.
[ back ] 84. A classic on this subject is the two-volume work of Christensen 1918 and 1934.
[ back ] 85. Such a theme may prove to be the key to understanding the etymological relationship of Greek anḗr (ἀνήρ) ‘man’ with nṓrops, a Homeric adjective glossed as λαμπός ‘bright’ in Hesychius (e.g. s.v. νῶροψ) and used as a formulaic synonym of aíthops ‘with looks of fire, fiery-looking’ in Homeric diction (e.g. postvocalic νώροπι χαλκῷ vs. postconsonantal αἴθοπι χαλκῷ. On the presence and absence, respectively, of initial a as reflex of laryngeal *h2 in e-grade anḗr (*h2ner-) and o-grade nōr- (*h2nor-), see Beekes 1969.75-76.
[ back ] 86. Bartholomae 1904.1565.
[ back ] 87. At pp. 153ff., I have argued that the semantics of Latin ārēre ‘be dry’ are secondary, and that the root *ā̌s- of this verb can be reconstructed as meaning ‘burn’, as reflected in the derivative noun ārea.
[ back ] 88. Cf. Caland 1896. This is not to say, of course, that cremation is the only type of Indic funerary practice.
[ back ] 89. See pp. 85, 129.
[ back ] 90. See pp. 86ff.
[ back ] 91. Cf. N. 1979a.224-227.
[ back ] 92. Full presentation of the argument in N 1979a.94-117.
[ back ] 93. Cf. Humbach 1961.99.
[ back ] 94. Cf. Nyberg 1938.313-327. There is considerable disagreement about the precise location of many of the places mentioned in Vendidad 1; cf. e.g. Gnoli 1980.23ff.
[ back ] 95. E.g. Nyberg p. 326.
[ back ] 96. Cf. Davidson 1985.93-94, 101, with further bibliography (esp. Duchesne-Guillemin 1979.63; also Gnoli 1980.91ff).
[ back ] 97. Nyberg p. 310.
[ back ] 98. Nyberg. pp. 321-322.
[ back ] 99. Cf. Caland 1896. To repeat, this is not to say that cremation is the only type of Indic funerary practice.
[ back ] 100. The standard etymology of daxma as “burning” (whence “place for burning, cremation”) from verb dag- “burn”, as we find it in e.g. Bartholomae 1904.676, has been challenged by Hoffmann 1975 [1965].338, on the basis of arguments presented by Humbach 1961, who shows that daxma in e.g. Vendidad 7.49 and following designates something like a “mausoleum,” that is, a roofed and sealed building in which the corpse is sheltered from the elements, as distinct from the open-sky format required by the orthodox Zoroastrian place of exposition. And yet, Humbach himself points out (p. 1010) that daxma in e.g. Vendidad 5.14, 8.2 designates an orthodox Zoroastrian place of exposition. In other words, the referent of this word daxma in the language of the Avesta can very from the orthodox place of exposition to the anti-orthodox “mausoleum” (for examples of New Persian daxma in this sense of “mausoleum” in the epic tradition of the Shāhnāma of Ferdowsi, see Humbach p. 100). Given this range of variation, we may continue to posit yet another nonorthodox variant among the referents of daxma, that is, a place where the corpse is cremated rather than exposed to the elements. And I am arguing that this particular variant represents the earliest meaning of daxma.
[ back ] 101. Duchesne-Guillemin 1962.82 (my translation).
[ back ] 102. Bartholomae 1904.1533.
[ back ] 103. Cf. Dubois 1924.485; also Gonda 1960.130.
[ back ] 104. Cf. DELL s.v. suffiō.
[ back ] 105. Cf. Dumézil 1959a.97-98.
[ back ] 106. In this connection, we may compare the obscure gloss exfir in Paulus ex Festo 69 Lindsay, where pūrgāmentum seems to be equated with suffitiō: [ back ] exfir, purgamentum, unde adhuc manet suffitio
[ back ] 107. Thurneysen 1912-1913.276-281; cf. Leumann 1977.550.
[ back ] 108. E.g. DELL s.v. pūrgō.
[ back ] 109. See pp. 164ff.