On the etymology of προοίμιον (prooímion)

2021.11.01, re-written 2021.11.07 | By Gregory Nagy

John Roddam Spencer Stanhope (1829-1908), Penelope at her loom (1864). Image via Wikimedia Commons.

§0. This essay connects—and updates—arguments I have presented in previous writings that are listed under my name in the Bibliography below and that are hereafter abbreviated as PH (1990), PP (1996) PR (2002, 3rd ed. 2021.10.01), HC (2009|2008), HPC (2010|2009), HH (2011), and MoM (2015). In these writings, I argue—accretively—that the ancient Greek compound noun προοίμιον (prooímion), which is conventionally translated as ‘prelude’, has a basic meaning that can best be explained in terms of not one but two simplex nouns, οἶμος (oîmos) and οἴμη (oímē). The meanings of all three forms, I argue, can be traced back to metaphors centering on the activity of weaving a web on an upright warp-weighted loom, much as Penelope weaves her web at her own loom— only to re-weave after each of her nightly un-weavings. Also, a related word that will interweave into my argumentation is ὕμνος (húmnos), often translated as ‘hymn’ or, more generally, ‘song’, which can likewise be traced back to a metaphorization of weaving at a loom. Relevant to these metaphorized meanings, in terms of my argument, are two other accretive writings of mine that I list in the Bibliography, each one of which I cite here by repeating all the initial words in their all-too revealing titles: “To trace a thread of thought” (2015.06.03) and “About a perfect start for a world-wide web of song” (2020.08.07). In the paragraphs that follow, I will epitomize also from these two writings, not only from the earlier writings mentioned at the start. In the epitomes here, I will keep at a bare minimum my references to the relevant secondary sources that I have amassed in the course of my earlier writings as listed in the Bibliography.

§1 (via PH 12§33). I start with traditional Greek myths where the very meaning of prooímion is mythologized. An archetypal form of the prooímion is performed by the god Apollo in the Homeric Hymn to Hermes 502: Apollo first struck up the lyre and then ὑπὸ καλὸν ἄεισεν ‘he sang beautifully, in accompaniment [hupo-]’. In this context prooímion can be translated roughly as ‘prelude’, but I prefer to use here the more neutral Latin borrowing, ‘prooemium’. The prooímion is a framework for differentiated virtuoso singing by an individual performer known in the Classical era of Athens as the kitharōidós ‘lyre [kithárā] singer’. The noun pro-oímion means, literally, ‘the first part of the song [oímē]’—just as the noun pró-nāos means ‘the first part of the temple [nāós]’. Just as the prónāos is not a temple in front of a temple but rather an organic part of the temple itself, so also the prooímion, if we consider its etymology, is not a song that is sung before another song but rather an organic part of the song itself. I see here an all-important fact that proves to be essential for understanding the basic meaning of prooímion. Let the record show that this fact, once highlighted by Hermann Koller (1956:191), is generally ignored in most published accounts about the meaning of prooímion.

§2 (via PP 63). If pro-oímion means, literally, ‘the first part of the song [oímē]’, I need to say more about oímē (οἴμη), which I have just translated as ‘song’. I argue that oímē is metaphorically the thread or, better, threading of the song, if it is true that the noun oímē stems from a verb-root *sei– referring to fabric-work. Here I am following the argumentation of Marcello Durante (1976:176–177) and thus disagreeing—at least in part—with what we read in the etymological dictionary of Pierre Chantraine (DELG under the entry οἴμη). As Durante argues (p. 177), the metaphor that is built into the Greek word pro-oímion is parallel to what we see in the Latin word ex-ordium, likewise referring to a prelude, which can likewise be interpreted etymologically as the first part of weaving a web (Nagy, PP 63n20; PR 72, 81; HC 2§92). And the ex– of exordium, indicating the origin of the threading, is comparable to the “genitive of origin” that contextualizes oímē in Odyssey 8.74 as the starting-point for the threading of the songs sung by the blind singer Demodokos (Nagy 2020.08.07 §0, picking up from Nagy 2015.05.27 §2).

§3 (via PP 63 continued). Making a related point, Durante (1976:176–177) argues persuasively that oímē (οἴμη) is cognate with another noun referring to fabric work, oîmos (οἷμος). If we consult in this regard the relevant discussion by Chantraine (again, under the entry οἴμη), we find that he shows some hesitation about the relationship between oímē and oîmos, but at least he makes it clear that the two words must be reconstructed as *hoímē and *hoîmos, as is made evident by the Attic by-form of prooímion, which is phroímion. The aspiration of p as ph in the by-form shows that the root of –oim– must be *-soim-, not *-oim-, and that this root therefore cannot be derived from *-oim– as attested in the verb eîmi ‘go’ (such a derivation is argued by Maslov 2012; on the meaning ‘go’ as a secondary semantic development from the idea of threading, I refer to my analysis in HC 2§92). I conclude, then, that oîmos and oímē can be interpreted etymologically as ‘song-thread’ and ‘song-threading’; in fact, the cognate of Greek oîmos in Old Icelandic is seimr, meaning ‘thread’ (again, HC 2§92, recontextualized in Nagy 2020.08.07 §3).

§4 (via PP 63–65). The metaphor implicit in the etymology of prooímion, where making songs is equated with a process of threading songs, is explicit in Pindar’s reference to Homeric reciters at the beginning of Nemean 2 (lines 1–3), where rhaptá ‘threaded together’ is applied to épē in the sense of poetic ‘utterances’: ῞Οθεν περ καὶ Ὁμηρίδαι |  ῥαπτῶν ἐπέων τὰ πόλλ᾿ ἀοιδοί | ἄρχονται, Διὸς ἐκ προοιμίου ‘[starting] from the point where [hóthen] the descendants of Homer [Homērídai], singers [aoidoí], most of the time begin [árkhesthai] their threaded-together [rhaptá] words [épē], from the prooímion of Zeus …’. The same metaphor is implicit in the etymology of the actual word for a professional Homeric performer, rhapsōidós, ‘he who threads together [rháptein] the song(s) [aoidḗ]’ (Schmitt 1967:300–301; Durante 1976:177–179; Nagy PH 1§21). This metaphor of threading together the song(s) is comparable to a related metaphor in archaic Greek traditions, that of weaving the song(s), which is in fact so old as to be of Indo-European linguistic provenience (Schmitt 1967:298–300). An example is this phrase of Pindar (F 179): ὑφαίνω δ᾿ Ἀμυθαονίδαισιν ποικίλον ἄνδημα ‘I weave [huphaínō] a varied [poikílos] headband [that is, of song] for the Amythaonidai’. As we see from such passages, song is being visualized as a web, a fabric, a textile—Latin textilis, from texere ‘weave’. Or, to use only for the moment an English word that no longer retains its metaphorical heritage, the song is even a text, Latin textus, again from texere (Schmitt 1967:14–15).

§5 (via PH 12§33). The prooímion or prooemium took the form of a prayer sung to a given god who presided over the occasion of a given seasonally recurring festival where the song was performed in competition with other songs. A clear reflex of this form can be found in the actual structure of the Homeric Hymns (Koller 1956:174–182, 195–206; elaborations in Nagy 2011). In fact, Thucydides (3.104.3–4) uses the word prooímion in referring to the version of the Homeric Hymn to Apollo that he knew (Koller 1956:173–174). That the dramatized context of these Hymns is one of seasonally recurring festivals where contests in song are held is clear from the use of hṓrā ‘seasonal time’ in Hymn 26.12–13 and of agṓn ‘contest’ in Hymn 6.19. That these Hymns are morphologically preludes, with the inherited function of introducing the main part of the performance, is illustrated by references indicating a shift to the performance proper, such as μεταβήσομαι ἄλλον ἐς ὕμνον ‘I will shift to the rest of the song [húmnos]’ at Homeric Hymns 5 (verse 293), 9 (verse 9), and 18 (verse 11). That the meaning here is ‘the rest of the song’, not ‘another song’, was effectively argued by Koller (1956.174–182), and the argumentation has been continued in my own work (especially in Nagy 2011). Relevant is the working definition of prooímion by Quintilian (Institutio oratoria 4.1.2): quod οἴμη cantus est, et citharoedi [kitharōidoí] pauca illa, quae antequam legitimum certamen inchoent, emerendi favoris gratia canunt, prooemium cognominaverunt… ‘that oimē is song and that the kitharōidoí refer to those few words that they sing before their competition [certamen] proper, for the sake of winning favor, as “proemium” [prooímion]…᾽. Here I note a most telling parallel, analyzed by Koller (1956:193), who shows that the certamen ‘competition’ of the kitharōidoí, that is, their agṓn, corresponds to agṓn ‘competition’ among rhetoricians, as metaphorized at the beginning of a celebrated speech by Demosthenes, On the “Crown”: πρῶτον μέν, […] τοῖς θεοῖς εὔχομαι πᾶσι καὶ πάσαις, […] τοσαύτην [= εὔνοιαν] ὑπάρξαι μοι παρ’ ὑμῶν εἰς τουτονὶ τὸν ἀγῶνα ‘First of all, I pray to all the gods and goddesses that as much good will [as I have accorded to the community] will also be accorded to me from you, for this present competition [agṓn]’.

§6 (via PH 12§34). Quintilian’s reference to ‘those few words’ of commencement as sung by the kitharōidoí ‘lyre [kithárā] singers’ is belied by the proportions of some of the larger Homeric Hymns, which had evolved into magnificent extravaganzas that rival epic in narrative power, as in the case of the Hymn to Apollo. It is in fact legitimate to ask whether the Homeric Hymns, especially the larger ones, were functional preludes. For now, however, it is enough to stress that they were formally just that, preludes. As we have seen, Thucydides refers to the Hymn to Apollo as a prooímion (3.104.4–5). Even the Hesiodic Theogony, with its even more imposing proportions, is morphologically a prooímion. So also is the representation of the first theogony ever sung, an archetypal performance of lyre singing by the god Hermes, as described and paraphrased in Hymn to Hermes 425–433 (commentary in PH 12§29). The crucial concept here is anabolḗ ‘prelude’, closely parallel to the concept of prooímion. Hermes sings his theogony ἀμβολάδην ‘in the manner of a prelude [anabolḗ]’ (426), just as the song started by Apollo’s lyre is elsewhere described in terms of ἀγησιχόρων (…) προοιμίων ἀμβολάς ‘the preludes [anabolḗ plural] of chorus-leading [hāgēsíkhora] prooemia [prooímion plural]’ in Pindar Pythian 1.4.

§7 (via PH 12§35). Still, the medium of the Homeric Hymns, which is poetry recited in dactylic hexameter, is several stages removed from the medium of kitharōidíā, which involves not recitation but singing to the accompaniment of the kithárā. We have to step back and ask what form the prooímion ‘prelude’ would have had in the context of kitharōidíā ‘lyre singing’ and, further, what kind of performance can be expected to have followed the prooímion in this same context of kitharōidíā.

§8 (via PH 12§36). Part of the answer is to be found in the lyric traditions ascribed to Terpander, reputedly the founder of the first katástasis ‘establishment’ of traditions in songmaking at Sparta (“Plutarch” On Music 1134b). This figure is credited with “inventing” the melodic patterns of kitharōidíā, which were called nómoi in this context (On Music 1132d). And there is more to it: as I show elsewhere (PH 3§9), a figure like Arion, the archetypal kitharōidós ‘lyre singer’ for the polis of Corinth (Herodotus 1.23), is specifically represented as performing a nómos when he sings a monodic performance to the accompaniment of his lyre (Herodotus 1.24.5). The traditional association of kitharōidíā with the concept of nómos, which we may interpret most generally as a lyric composition that followed a set mode or melodic pattern (PH 3§11), must be compared with a traditional saying, variations of which are strikingly attested in the idiom of Plato, that a prooímion ‘prelude’ presupposes a nómos (Plato Timaeus 29d; Republic 7.531d, 532d; commentary by Koller 1956:183). This association is made explicit in Plato Laws 4.722d and following: the prooímion is a ‘prelude’ to the nómos in the specific context of kitharōidíā (Koller pp. 183, 188).

§9 (via PH 12§37). Another part of the answer to the question about the actual form of the prooímion can be found in the diction of Pindar. So far we have noted the association of prooímion and nómos in the general context of kitharōidíā, but we have yet to see a reference to this association in a specifically choral context of kitharōidíā. The references in Plato clearly presuppose a monodic rather than choral context. Turning to the choral context, however, where we see singing and dancing, not just solo singing, let us consider the wording in two passages taken from Pindar, Nemean 5.21–26 and Pythian 1.1–4. In Nemean 5, we see the representation of a khorós ‘chorus’ of Muses (23) where this ensemble is specifically singing (ἄειδ’ 22), and in their midst is the god Apollo himself, taking control as he strikes up a lyre that is heptáglōssos ‘having seven languages’ (24), leading the choral performance of ‘all sorts of nómoi’ (παντοίων νόμων Nemean 5.25). As I observe elsewhere (PH 3§13), the seven-string lyre, supposedly the “invention” of Terpander, could fit a wide variety of set melodic patterns, called nómoi, within a new interrelated system reflecting Panhellenic synthesis. Here too, in the passage from Pindar, these melodic patterns are explicitly called nómoi. But in this case the nómoi that are represented as being performed are not monodic, which was the case when Arion sang his nómos (again Herodotus 1.24.5), but clearly choral: it is the khorós ‘chorus’ of Muses who are actually singing the nómoi (again Nemean 5.21–26). Moreover, this ensemble of Muses is represented here as actually singing the words of the prooímion. Although the word prooímion is not used in this passage, the phraseology of the Muses’ paraphrased words (Nemean 5.25–26) is perfectly in accordance with the proper syntax and rhetorical strategy of attested prooemia. The chorus of Muses is represented as performing not just the subsequent nómoi but also the prooímion that is expected to introduce a nómos, with Apollo’s overall control being represented simply by his act of striking up the lyre.

§10 (via PH 12§38). We may now supplement the testimony of Pindar Nemean 5.21–26 with that of Pythian 1.1–4. This passage, another tour de force in descriptive compression, pictures the lyre of Apollo, as the player strikes up a tune (4), straightaway being heard by a stylized ensemble of singers/dancers—let us call it a chorus—who start dancing (2) as soon as the lyre gives off its sḗmata ‘signals’ to be heeded by aoidoí ‘singers’ (3), and what results are anabolaí ‘preludes’ (4) of prooímia ‘prooemia’ (4) that are described as hāgēsíkhora ‘chorus-leading’ (4). In short, the diction of Pindar gives indications that even in terms of choral performance the prooímion ‘prooemium’ precedes what is being consistently called the nómos.

§11 (via PH 12§39). This Pindaric picture, however, of a prooímion as if performed by the chorus is idealized. Another example of such idealization is Pindar Nemean 3.1–12, where the chorus members are described, five lines into the composition, as if waiting for the voice of the Muse, which is to be their cue to start their performance; then, at lines 10–11 of the composition, the Muse is invoked to start (ἄρχε 10) the húmnos ‘song’ (11), while the “I” of the lyre player, the persona of the composer, distributes the song to the chorus members and to the lyre (11–12). The word for “lyre” here is twelve lines into the composition, and yet the context itself presupposes that it had started the whole performance, just as the chorus has been presumably performing the entire composition ever since the first line. Still another example is Pindar Nemean 2.1–3, where the beginning of the composition describes the prooemium of a performance without being a prooemium itself, in that no divinity has been directly invoked to start the performance. The prooemium being represented in Nemean 2.1–3 is specifically the prelude of a Homeric performance: the Homērídai ‘Sons of Homer’ (2.1), who are described as the aoidoí ‘singers’ (2.2.) of ‘threaded-together words’ (ῥαπτῶν ἐπέων 2.2), are said to ‘begin’ (ἄρχονται) their performance by ‘starting from a prooemium of Zeus’ (Διὸς ἐκ προοιμίου 2.3). The first word of Nemean 2, ὅθεν ‘starting at the point where…’, is transitional, to be expected after the given divinity has already been invoked. Then, at the very end of the composition, the chorus as polîtai ‘polis-dwellers’ (24) are called upon to ‘lead’, as conveyed by the verb exarkhō (25), in celebration. In a functional prooemium this verb could be expected at the beginning of the performance.

§12 (via PH 12§40). The stylized prooemia in Pindar, then, are idealizations. It is as if the traditions of differentiated monodic composition and performance had never happened. Yet the context of monody had already developed the form of the prooímion far beyond its native choral context of striking up the lyre for the chorus. There is evidence (collected and analyzed in PH 12§§41–44) that what we call the prooímion had already undergone, by Pindar’s time, a vast stretch of evolution in traditions of composition and performance in monodic song and even in poetry. This evolution serves as backdrop for the use, in Thucydides (3.104.4–5), of the word prooímion in referring to the Homeric Hymn to Apollo.

§13. In a previous essay, “The Earliest Phases in the Reception of the Homeric Hymns” (Nagy 2011, abbreviated here as HH), I explain at length the point I just made about the word prooímion as used by Thucydides. In the course of my explanation (HH 329–330), I offer an overall analysis of the relationship between this word prooímion and another relevant word, húmnos, as used in the Homeric Hymns and, once, in the Homeric Odyssey (earlier analysis in HPC I§§188–223, MoM 4§§69–183). I will elaborate further in an essay that follows this one in Classical Continuum, but I need to preview, already here, a point that pertains directly to the present essay.

§14. I argue that the word húmnos, like the words oîmos / oímē / prooímion, is also derived from wording that refers to the making of fabric, or, to say ot more simply, to the activity of threading. As I once noted elsewhere (HC 2§93), the word oîmos is interchangeable with the word húmnos at verse 451 of the Homeric Hymn to Hermes, where we see the attestation of both οἶμος ἀοιδῆς and ὕμνος ἀοιδῆς in the manuscript tradition. This interchangeability, which I take to be formulaic, needs to be compared to the cognate expression ἀοιδῆς ὕμνον at verse 429 of Odyssey 8. I interpret the combinations of húmnos and oîmos with aoidḗ ‘song’ in these expressions as references to the ‘weaving’ of song or, more basically, to the ‘threading’ of song. Relevant to this interpretation is the context of oímē at verse 74 of Odyssey 8: I see a metaphorical reference there to the initial part of performing a song, that is to the ‘initial threading’ of a song.

15. I sum up, for now. So far, I have argued that the meaning of oîmos or oímē as ‘song’ results from a metaphorical extension: the idea of making song is being expressed metaphorically through the idea of making fabric, that is, threading. As for contexts where oîmos and oímē seem to mean ‘way, pathway’, I argue that such a meaning is a result of further metaphorical extension: here the general idea of moving ahead from one point to another is being expressed metaphorically by applying the specific idea of threading one’s way from one point to another. Whereas the metaphorical extension here goes from the specific—threading— to the general—going, moving ahead— we can also find examples of a reverse metaphorical extension, from the general to the specific, as in the case of the Homeric word metabaínein ‘moving ahead’, where the general idea of moving ahead is applied to the specific idea of shifting from one beginning to another beginning in the the process of weaving a song, as it were. In the essay that follows this one in Classical Continuum, I plan to elaborate on such shifts in the starting of a song—or, better, in the re-starting of a song, as symbolized by the daily re-weavings of the web that gets un-woven night after lonely night by Penelope.


Chantraine, P. 2009. Dictionnaire étymologique de la langue grecque: histoire des mots, ed. J. Taillardat, O. Masson, and J.-L. Perpillou. With a supplement “Chroniques d’étymologie grecque,” ed. A. Blanc, Ch. de Lamberterie, and Jean-Louis Perpillou, 1–10. Paris. Abbreviated as DELG.

Durante, M. 1976. Sulla preistoria della tradizione poetica greca II. Risultanze della comparazione indoeuropea. Incunabula Graeca 64. Rome.

Koller, H. 1956. “Das kitharodische Prooimion: Eine formgeschichtliche Untersuchung.” Philologus 100:159–206.

Maslov, B. 2012. “The Real Life of the Genre of prooimion.” Classical Philology 107:191–205.

Nagy, G. 1990. Pindar’s Homer: The Lyric Possession of an Epic Past. Baltimore. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Pindars_Homer.1990.

Nagy, G. 1996. Poetry as Performance: Homer and Beyond. Cambridge. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Poetry_as_Performance.1996.

Nagy, G. 2002. Plato’s Rhapsody and Homer’s Music: The Poetics of the Panathenaic Festival in Classical Athens. Cambridge, MA, and Athens. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Platos_Rhapsody_and_Homers_Music.2002.

Nagy, G. 2009|2008. Printed | Online version. Homer the Classic. Hellenic Studies 36. Cambridge, MA, and Washington, DC. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Homer_the_Classic.2008.

Nagy, G. 2010|2009. Printed | Online version. Homer the Preclassic. Berkeley and Los Angeles. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Homer_the_Preclassic.2009.

Nagy, G. 2011. “The Earliest Phases in the Reception of the Homeric Hymns.” The Homeric Hymns: Interpretative Essays, ed. A. Faulkner, 280–333. Oxford. http://nrs.harvard.edu/urn-3:hlnc.essay:Nagy.The_Earliest_Phases_in_the_Reception_of_the_Homeric_Hymns.2011.

Nagy, G. 2015. Masterpieces of Metonymy: From Ancient Greek Times to Now. Hellenic Studies Series 72. Washington, DC. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Masterpieces_of_Metonymy.2015.

Nagy, G. 2015.05.27. “An experiment in the making of a Homer commentary: Taking a shortcut in analyzing the first song of Demodokos in Odyssey 8.” Classical Inquirieshttps://classical-inquiries.chs.harvard.edu/an-experiment-in-the-making-of-a-homer-commentary/.

Nagy, G. 2015.06.03. “To trace a thread of thought starting from a Homeric song that seems to have no ending.” Classical Inquiries. https://classical-inquiries.chs.harvard.edu/to-trace-a-thread-of-thought-starting-from-a-homeric-song-that-seems-to-have-no-ending/.

Nagy, G. 2020. Second edition, online version, of Nagy 2002. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Platos_Rhapsody_and_Homers_Music.2020.

Nagy, G. 2020.08.07. “About a perfect start for a world-wide web of song.” Classical Inquirieshttps://classical-inquiries.chs.harvard.edu/about-a-perfect-start-for-a-world-wide-web-of-song/.

Nagy, G. 2021.10.10. Third edition of Nagy 2002. Classical Continuum. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Platos_Rhapsody_and_Homers_Music.2021.

Pagliaro, A. 1953. Saggi di critica semantica. Messina and Florence.

Pierini, R., and T. Palaima, eds. 2020.08.07. “MASt @ CHS – Friday, June 26, 2020: Summaries of Presentations and Discussion by Elizabeth Barber, Thomas Palaima, Rachele Pierini, and Brent Vine.” Classical Inquiries. https://classical-inquiries.chs.harvard.edu/mast-chs-friday-june-26-2020-summaries-of-presentations-and-discussion/.

Palaima, T. “The Etymology of Mycenaean pe-re-ke-u and a-ke-te-ri-ja and Related Forms,” §§3–32 (bibliography §32).

Barber, E. “The Weaving Process in Ancient and Today’s Textile Industry and Some Notes on Mycenaean and Alphabetic Greek Textile Lexicon,” §§33–50 (bibliography §50).

Pierini, R. “Linguistic idiosyncrasies and chronology on Linear B tablets from the RCT at Knossos,” §§51–68 (bibliography §68).

Vine, B. “Commentary,” §§69-72 et passim (bibliography §72).

Schmitt, R. 1967. Dichtung und Dichtersprache in indogermanischer Zeit. Wiesbaden.

One response to “On the etymology of προοίμιον (prooímion)”

Leave a Reply