Appendix. The Semantics of ἄνθος and ἀνθέω

A number of the passages discussed in this book incorporate the noun ἄνθος, which is commonly translated “flower.” [1] Linguists have however expressed doubts over the semantics of the lexeme: some have suggested that it originally carried a more abstract meaning, such as “surface” or “that which bursts forth.” Therefore, in spite of the usual renderings of ἄνθος by translators, we need to address this controversy if we are to be confident that the relevant Homeric images do indeed refer to flowers.

As we shall see, there are in fact good reasons to endorse more conventional translations of ἄνθος in Homeric poetry. Comparative evidence from other Indo-European languages suggests that the archaic Greek term derived from an earlier Proto-Indo-European root with a floral meaning; and the internal evidence of the Homeric poems offers us still stronger reasons to accept such a meaning for the Homeric noun. While the equivalent verbal root may possess a more general meaning, the primary reference of the Homeric noun ἄνθος is to the concrete concept of “flower(s).”

What is more, George Lakoff’s, Mark Johnson’s and Mark Turner’s analyses of metaphor—and more specifically of the manner in which concrete concepts can be used to explain the more abstract—help us to understand how more abstract connotations of ἄνθος, such as “surface” or “that which bursts forth,” might have developed. [2] Given that the lexeme referred to flowers in the Greek natural environment, it was also able to evoke more abstract qualities associated with such flowers—the bright surfaces of flowers in spring or early summer, or the flowers that burst forth at those times. In this way, the ground was prepared for later, purely metaphorical usages of ἄνθος: it was a relatively easy step for classical authors such as Aeschylus and Pindar to use the noun to evoke surfaces or irruptions more generally.

A number of twentieth-century scholars were led by such classical usages of the noun ἄνθος to doubt that floral meanings were primary for the root ἀνθο/ε-, from which both the noun ἄνθος and the verb ἀνθέω derive. Instead they suggested that the root is associated first and foremost with the notions of “surface” or “top.” William Stanford for instance objects to the rendering of Aeschylus’ image ἀνθοῦν πέλαγος Αἰγαῖον νεκροῖς (Agamemnon 659) as “the deep was all aflower … with corpses”: according to him, associations of corpses with beauty are possible only for “a disillusioned modern fin de siècle.” [3] He suggests that this and other attestations of the root ἀνθο/ε- do not, in fact, convey the meaning “flower” but rather “that which rises to the surface.” Drawing on Stanford’s observations, Kerr Borthwick argues that the image τρίβῳ κατέξαινον ἄνθος Ἀργείων at Agamemnon 197–198 (“[winds] were wearing away the anthos of the Argives with their attrition”) evokes the nap (i.e., the top) of a cloth, and that the metaphor at line 659 refers in fact to the foam of the sea. [4]

Some linguists have drawn similar conclusions from the Homeric uses of ἄνθος. Rahim Raman for instance builds on Stanford’s conclusions in a study of Homeric and classical poetry. In a homogenizing reading of the terms ἄωτος, ἄνθος, χνόος, and ἀκμή, he argues that they all underwent a development from a Homeric meaning of “surface-growth” to a later metaphorical sense of “excellence,” and that at no point were ἄνθος or any of these other nouns primarily associated with flowers. In his opinion, ἄνθος only refers to flowers in the sense of “surface of a landscape” or “top of a plant,” and we should not see instances where the term refers to other types of thing as floral metaphors. [5] Likewise J. M. Aitchison, in an essay focused on Homeric poetry, argues that ἄνθος is not primarily associated with flowers. According to him, a meaning of “growth (up from)” best accounts for all the Homeric usages of the root ἀνθο/ε-. [6] Michael Clarke has recently revisited Aitchison’s interpretation. He argues that early Greek uses of ἄνθος convey the concept of “burgeoning, swelling, upward-spreading motion or growth that bursts into a spreading excrescence.” [7]

As we can see from this summary, while some of these scholars associate the root ἀνθο/ε- with the concept of growth, they all cast doubt on the floral associations of the root. However, when we study evidence from historical linguistics and review the testimony from our Homeric texts, we find good reasons to believe not only that the primary and original meaning of the root ἀνθο/ε- was vegetal, but also that the noun from this root, ἄνθος, has the specific meaning of “flower” in Homeric poetry.

Firstly, the possible cognates of ἄνθος in other Indo-European languages do not point in the direction of abstractions such as “what lies on the surface,” but towards more concrete meanings. They include Frisian åndul, “marsh grass”; Armenian and, “field”; Sanskrit ándhas-, “a herb; the soma plant; grassy ground”; and, most strikingly, Albanian ëndë, “flower.” [8] The semantics of some of these proposed cognates are uncertain. [9] Nevertheless, they are all associated in one way or other with vegetation. What is more, the chief Albanian grammars, which are not always cited by linguists working on this question, confirm the existence of a cognate of Greek ἄνθος with a floral meaning. [10]

On the basis of such evidence, linguists have suggested Proto-Indo-European roots with vegetal meanings. Robert Beekes tentatively offers an original verbal root *h ₂ endh- with the meaning “sprout.” [11] James Mallory and Douglas Adams reconstruct a nominal stem *h ₂ endhes- with the meaning “± flower”: that is, “flower” offers the best approximation of the original meaning of this Proto-Indo-European root, but a more general, vegetal meaning is also possible. [12] What is more, as Adams has pointed out to me per litteras, even if we cannot be certain that Proto-Indo-European *h ₂ endhes- meant “flower,” the presence of cognates with that meaning in geographically adjacent regions, Greek and Albanian, indicates that at the very least these two branches of the Indo-European family shared an early noun meaning “flower.” [13]

This evidence, then, suggests that the floral meaning of the Greek noun ἄνθος is not a later development but rather reflects the early semantics of the term. And as we shall see, the internal evidence of Homeric poetry likewise points us towards more concrete referents for the root ἀνθο/ε-, as opposed to abstractions such as “burstings forth.” Specifically, we observe a distinction between a verbal root associated with vegetation in general and substantival stems referring to flowers. The verbal root ἀνθε-, usually found in participial form, has either a general, vegetal meaning of “flourishing” or a more specific meaning of “blooming,” while in all but a few instances the adjective ἀνθεμόεις and the noun ἄνθος clearly mean “flowery” and “flower” respectively. We have, then, a distinction between verbal and nominal roots similar to that between the verbal root *h ₂ endh-, tentatively proposed by Beekes, and the nominal root *h ₂ endh es-, reconstructed by Mallory and Adams, which have the meanings “sprout” and “± flower” respectively: it is possible that these Homeric roots reflect the semantics of the original Proto-Indo-European forms.

There are three cases in our versions of the Homeric poems where the noun ἄνθος might carry a more general meaning of “vegetal growth”—though in all of these instances a floral meaning is still very much a possibility. It might seem at first sight that the phrase ἄνθεα ποίης at Odyssey 9.449 and Hymn 30.15 has a more general vegetal meaning—“grassy growths” vel sim. But audiences familiar with early hexameter might not have understood it in this fashion. At Theogony 576 the phrase refers to material used for a garland, which we would expect to be made up of flowers. [14] This suggests that it could carry the meaning “flowers of the grassland.” [15] At Odyssey 7.125–126, a general vegetal sense of ἄνθος is likewise possible: πάροιθε δέ τ’ ὄμφακές εἰσιν / ἄνθος ἀμφιεῖσαι (“before them were unripe grapes / sending a bloom [?] around them”). The juxtaposition of ἄνθος with a reference to ripening grapes is surprising: we would expect the flowers to fall off the vine before any fruit appeared. It may be, then, that these lines attribute some general notion of vegetal flourishing to the grapes. But we should remember that the Phaeacian plantations are no ordinary allotments: as we saw in Chapter 5, their plants grow constantly throughout the year, without the need for human tendance; and the description of unripe grapes in lines 125–126 is immediately followed by a reference to mature grapes. Audiences could readily have imagined that such plants, which are not subject to seasonal variation, would exhibit flowers and fruit at the same time.

In other passages, the Homeric noun ἄνθος clearly evokes flowers. In a number of cases, the noun is coupled with references to particular flowering plants. At Hymn to Demeter 6, for instance, Korē is gathering ἄνθεα, specified as roses, crocuses, and violets; later in the hymn the daughters of Celeus are said to have hair like the saffron flower (κροκηΐῳ ἄνθει, Hymn to Demeter 178); at Odyssey 10.304, Hermes digs up the herb μῶλυ, which has γάλακτι … εἴκελον ἄνθος (“a flower like milk”); and at Odyssey 6.231 and 23.158, Athena adorns Odysseus’ head with locks like the flower of the hyacinth (ὑακινθίνῳ ἄνθει). [16]

Similarly, in the relevant passages of Homeric poetry the adjective ἀνθεμόεις is most readily translated “flowery.” It is often used to describe man-made objects: cauldrons (Iliad 23.885, Odyssey 3.440), a mixing-bowl (Odyssey 24.275), and Aphrodite’s earrings (Hymn 6.9). It is very likely that these passages focus on decorative, floral motifs, rather than conveying the more vague notions of “surfaces” or “burstings forth.” [17]

The noun ἄνθος and the adjective ἀνθεμόεις, then, mean “flower” or “flowery” in the majority of instances in Homeric poetry, and even in those cases where they seem at first sight not to do so, audiences might well have perceived a reference to flowers. The verb ἀνθέω in Homeric poetry can also be associated with flowers, but more often it carries general vegetal meanings. [18] In two of the Homeric instances of the verb, a generally vegetal translation of “flourish” seems appropriate: ὄρος ἄνθεον ὕλῃ, “a mountain flourishing with forest” (Hymn 1.9), and ἀνθοῦσαν ἀλωήν, “flourishing vineyard” (Hymn to Hermes 87). In contrast with the vines of Odyssey 7, the vineyard in the Hymn to Hermes is a regular plantation tended by a farmer: we would not expect that its flowers and fruit would appear at the same time. But owing to its associations with flowers, the noun ἄνθος is able to bring out similar connotations in the verb ἀνθέω. At Hymn to Apollo 139, for instance, the juxtaposition of the aorist form ἤνθησ’ with the noun ἄνθεσιν suggests a specifically floral meaning for the verb: “[Delos] bloomed … with flowers.”

From this evidence it seems best to conclude that in Homeric poetry the root ἀνθο/ε- has either a specifically floral or a generally vegetal meaning. Depending on context the verb ἀνθέω can be translated either “bloom” or “flourish.” And in most if not all instances, the substantival forms ἄνθος and ἀνθεμ- refer specifically to flowers.

It is, however, true that in a restricted range of passages the noun ἄνθος and the adjective ἀνθεμόεις might evoke not merely “flowers” but “flowery surfaces”—the sorts of bright, colorful surfaces that carpet Greek spring landscapes. And these passages point us in the direction of a compromise with the findings of scholars such as Stanford and Borthwick. In the Iliad, for instance, the two lexemes describe flowers by the River Scamander: An onlooker—or an audience member imagining the scene—might perceive not merely the individual flowers (captured by the plural ἄνθεα) but also their collective effect as they cover the meadow of the Scamander (suggested by the adjective ἀνθεμόεντι). [19]

The root ἀνθεμ- evokes “flowery” fields in three other passages, perhaps again with suggestions of “flowery surfaces.” At Odyssey 12.159. we once more find the phrase λειμὼν ἀνθεμόεις (this time in the accusative case) in the description of the island of the Sirens; at Hymn to Hermes 96, we encounter “flowery plains” (πεδί’ ἀνθεμόεντα), which appear to anticipate the asphodel meadows of lines 221 and 344; and the “flowery pastures” (νομοὶ ἀνθεμόεντες) of Hymn to Aphrodite 169 contrast with the “grassy pastures” (νομοὶ ποιήεντες) of line 78. [20] A translation simply of “surface” or “with surfaces” would not do for ἀνθεμόεις in any of these descriptions of fields; but “(with) flowery surface(s)” is certainly possible.

I suggest, therefore, that the meaning simply of “surface” (as opposed to “flowery surface”) attributed by Stanford and Borthwick to the root ἀνθο/ε- in Aeschylus represents not an earlier sense of the ἀνθο/ε- root that was then transferred to flowers, but rather a metaphorical development from an original, floral meaning of the term. Flowers carpeted Greek spring landscapes, and this concept of bright surfaces may well have been present in some early uses of the term, including the Homeric descriptions of fields listed above; classical authors would therefore have been able to use the concept of flowers to evoke surfaces more generally. And if as Clarke suggests, the noun ἄνθος conveys the notion of bursting forth in other passages of classical Greek literature, this would represent a metaphorical development from a second characteristic of Greek flowers—their bursting forth in the spring and early summer. [21] These developments would be in keeping with Lakoff’s, Johnson’s and Turner’s analysis of the formation of metaphor: as they show, metaphors map from the more concrete to the more abstract and thereby aid our understanding of abstract concepts. [22]