We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 23 December 2013
The interesting thing about the word for ‘palace’ in Homer is that there is no such word. All the words that mean ‘house’ (δόμος, δῶ, δῶμα, μέγαρον, οἶκος, οἰκία) may be applied to a royal palace, but all of them (and their plurals) may equally well be used of the house of an ordinary citizen, μέγαρον is often translated ‘palace’, or some other word with connotations of kingly majesty. But it too, when it is not more narrowly localised to the living-room, means just a house in general.
Just as there is no separate word to designate a palace in Homer (τὰ βασίλεια occurs first in Herodotus), so a qualifying adjective is never used to indicate that a particular house is ‘royal’ (the first occurrence of such a periphrasis is βασίλειοι οἶκοι Aeschylus, Ag. 157). In fact, to Homer there is no sharp distinction between a palace and a house, except that a king is likely to have a bigger and better home, more sumptuously furnished.
The Iliad and the Odyssey are much concerned with kings. So it is odd that no word serves to designate the king's palace and no other dwelling, especially when the poet so delights in describing the splendour of royal homes (for example, the description of Alcinous' palace and its grounds in Od. vii 81 ff.). The major archaeological sites of the Mycenaean period, in which the main stories of the poems are set, all contain a palace which is easily recognised, even in its ruinous state. Can one imagine a new arrival at a great Mycenaean site needing to be shown which was the palace, as Odysseus was shown at Phaeacia? It seems highly unlikely that the Mycenaeans should have had no special word for a palace, though there is so far no firm evidence for (or against) the existence of such a word in Linear B. If a ‘palace’ word existed, its special significance, if not the word itself, must have been forgotten by Homer's time, when a great many houses might be found within the city wall, as at Smyrna. There may, however, be traces of the ‘palace’ meaning still detectable.
In Table 1, which has been compiled from the concordances of Prendergast and Dunbar, I have tabulated the frequency with which each of the regular words for ‘house’ is used to denote a dwelling occupied by gods, lesser divinities, mortals or animals. Various interesting points arise from this table.
It is apparent that there is a far higher proportion of gods' houses mentioned in the Iliad than in the Odyssey. The figures for δόμος and δῶμα show this particularly clearly. But the difference is probably due entirely to the subject-matter: far more of the action is taken up with scenes among the gods in the Iliad than in the Odyssey.
1 This paper is based on parts of a Ph.D. thesis written at University College, London, with the aid of a scholarship awarded by the New Zealand Uni versity Grants Committee. Special thanks for helpful advice and suggestions are due to Professor T. B. L. Webster, and to Professor H. A. Murray, Dr A. Thornton and Dr G. Morgan. I am grateful to Mr F. B. Knox for the statistical calculations in n. 9.
2 Nausicaa tells Odysseus to ask, when he reaches the city, which is the palace, but she adds (Od. vi 300 ff.):
This sounds like a conflation of two worlds, the traditionally remembered Mycenaean palaces and the little towns of the Geometric period. Perhaps each is introduced here for a purpose: Mycenaean splendour mentioned to prepare for the description of the glories of the palace (vii 82 ff.), and contemporary insignificance implied to prepare for Athena's appearance as guide (vii 18 ff.). Or perhaps, more simply, the poet inadvertently began to speak as though of contemporary houses, and then caught himself up with an attempt to contradict the anachronism.
3 qa-si-re-wi-ja is not now thought to mean a palace, or indeed any sort of house at all (cf. Morpurgo, , Mycenaeae Graecitatis Lexicon 272 Google Scholar s.v.; Palmer, , Phoenix xiv [1960] 182 Google Scholar). Nor is there any sign that wa-na-ka-te-ro can mean a palace (cf. Palmer, , Mycenaean Greek Texts 461 Google Scholar s.v.).
4 Cook, J. M., BSA liii (1958–1959) 1 ff.Google Scholar
5 The purely adverbial forms (e.g. δόμονδε) have been omitted, as some of these probably came to be regarded as words in their own right and so would weight the figures wrongly. Singulars and plurals are given separately, because it cannot be assumed that they will behave identically. The figures for the Iliad and the Odyssey are separated for the same reason.
6 Under ‘gods’ I have included only the Olympian pantheon, with Poseidon and Hades; the latter contra Page (History and the Homeric Iliad 326, n. 8), who believes that there is no personal Hades in Homer. Even if this is true for the time of Homer himself, it is impossible to see how the expression ‘the house of Hades’ could have arisen except as noun plus possessive genitive.
7 ‘Divinities’ means all the more or less supernatural persons who are not included under ‘gods’. These are Calypso, , Circe, , Dawn, (Od. xii 4)Google Scholar, Oceanus, Heracles, the Muses, Polyphemos, and Erechtheus, (Od. vii 81 Google Scholar, cf. Il. ii 546 ff., which indicates, though it does not quite prove, that Erechtheus is a divinity in Homer). It is, however, possible that neither Heracles nor Polyphemos is regarded as more than mortal (cf., for Heracles, , Il. xviii 117 ffGoogle Scholar).
8 I have omitted the use of δόμος once of a temple (Il. vi 89). If the conclusions drawn in this paper are correct, δόμος is in fact the word we should expect to find for a temple, not οἶκος. But in later times either word might be used (e.g. Ar. Ran. 1273, Nub. 600).
9
is used in the Iliad for human or gods' houses 13 times in all; for δόμος, δῶ, δῶμα, δόμοι and δώματα the human and divine total is 94 (μέγαρον and μέγαρα are omitted from this calculation as it is often impossible to be sure whether the entire house or just the living-room is meant; references to houses of minor divinities are also omitted, because of the uncertainties mentioned in n. 7 above). Thus the total number of times these words are used for human or divine houses is 107. Fifty-eight occurrences refer to human houses and 49 to gods' houses. Thus the number of times 
would be expected to be used of human houses is 
and of gods' houses 
In fact 
are used all 13 times for human houses and not at all for gods' houses. The χ2 test (cf.
Moroney, M. J., Facts from Figures [1956] 246–70Google Scholar) gives the probability of this distribution of 
occurring by chance as approximately one in a thousand.
A similar calculation for the Odyssey gives the probability (that 
being used only for human houses is due to chance) as much less than one in a hundred.
If both the Iliad and the Odyssey are taken together, the probability is very much less than one in a thousand. Such a result is normally considered highly significant (cf. Moroney, op. cit. 218).
It may be felt that as the poet's choice of vocabulary was largely at the level of formulae rather than single words, counting occurrences of words is of little value. However, the fact remains that when a god's house was mentioned the poet never chose a phrase containing οἶκος. It has recently been shown (J. B. Hainsworth, The Flexibility of the Homeric Formula) that if he really wanted to use a certain formula there were many ways he could adapt it to fit metrical requirements.
10 Cf. Morpurgo, , Mycenaeae Graecitatis Lexicon 365.Google Scholar
11 Mycenaean Greek Texts 203 ff.
12 The dual meaning of *wic- is fundamental in the IE languages (cf. Skt viç-, Lat. vicus), and there is no certainty which meaning is older.
13 Cf. for example Boisacq, , Dictionnaire étymologique de la langue grecque 195 f.Google Scholar However, Benveniste, (BSL li [1955] 14 ff.)Google Scholar believes δόμος and δέμω are from different roots, and that δόμος in early Greek has as its basic meaning a social or territorial grouping, like οἶκος. But there is no example of δόμος with such a meaning in Homer. All Homeric δόμοι are dwellings in the most concrete sense, and when a household or family is spoken of, the word is οἶκος.
14 Professor Webster has suggested (by letter) that πίων (Od. ix 35) may in fact mean ‘fertile’, its basic meaning, and οἶκος here refer rather to an ‘estate’ (cf. p. 119 above). Compare however πίονος ἐξ ἀδύτοιο, Il. v 512.
15 I have not counted possessive adjectives in these figures.
16 The possible objection to confining the ‘palace’ meaning to δόμος, that the word for the palace of the King of Persia was regularly οἶκος, proves invalid, as of the ten times it is so used (once in Thucydides, nine times in Herodotus) eight refer unmistakeably to the King's household, and the remaining two may do so (Hdt. iv 97.6 and v 31.4).
17 Oddly, this expression does not occur in the Catalogue, where in view of the subject-matter (‘the such-and-such people, who dwelt in…’) one might expect it to be frequent. In fact, however, ναίω and ναιετάω are themselves comparatively rare in the Catalogue (six times with this meaning, out of a total of 33 times in the Iliad, compared with 17 out of 24 for νέμω, and 18 out of 18 for εἶχον in the sense of ‘they lived in’). The scarcity of οἰκία in the Catalogue (once, with τίθημι) is probably to be explained by this stylistic difference.
