Speakers and Speeches in the Argonautica: Running the Numbers

When characters in narrative fiction engage in direct speech, we read via a narrator’s quotation the perspectives, thoughts, and interpretations of those characters on events (past, ongoing, and prospective) in the storyworld in which they operate. And when they don’t speak? We might have difficulty reading them. It is a potential problem highlighted early on in the Argonautic narrative when with the Argonauts assembled but before the voyage gets underway, we face Jason on the beach thinking to himself.

ἔνθ᾽ αὖτ᾽ Αἰσονίδης μὲν ἀμήχανος εἰν ἑοῖ αὐτῷ
πορφύρεσκεν ἕκαστα κατηφιόωντι ἐοικώς·

A.R. 1.460-1

There again Aeson’s son, helpless, brooded within himself over each thing, like a man mute with worry.

The Argonaut Idas is less than impressed by what we are seeing (and not hearing) and demands to know what Jason is thinking (and if it matches with his own mind-model of cowardice).

‘Αἰσονίδη, τίνα τήνδε μετὰ φρεσὶ μῆτιν ἑλίσσεις;
αὔδα ἐνὶ μέσσοισι τεὸν νόον. ἦέ σε δαμνᾷ
τάρβος ἐπιπλόμενον, τό τ᾽ ἀνάλκιδας ἄνδρας ἀτύζει;

A.R. 1.463-5

‘Son of Aeson, what plan are you turning over in your mind? Speak your thought amongst us! Does fear attack and overwhelm you, the fear that distresses feeble men?’

Idmon, another Argonaut, intervenes. With the quarrel that breaks out between Idas and himself, the opportunity to hear Jason is lost. On that incident/missed opportunity, Hunter (1993: 19) comments, ‘Appearances give no access to any simple, unmediated “truth”: you cannot tell with any certainty what someone is thinking or what their mood is from their facial expression.’ Unless telepathic we can never tell with certainty, but if we are to mind-model characters satisfactorily, we need material with which to construct a framework. That said, Idas’ outburst which runs for another six lines and initiates a back and forth between himself and Idmon does give us insight into how Idas’ mind works and how he himself mind-models e.g. silence suggesting fear or weakness leading to a need to show strength.

‘Whereas some 45% of the Iliad, 67% of the Odyssey and 47% of the Aeneid are in the direct speech of characters – the high Odyssey figures being largely due to Odysseus’ narrative of his adventures in Books 9-12 – only 29% of the Argonautica falls into this category.’
(Hunter 1993: 138)

Richard Hunter’s Argonautica percentage tallies with mine in the first table below, and for the Homeric epics Rutherford (1992: 58) has similar figures: ‘discounting the special case of the narrative portions of books 9-12… 6,835 lines of direct speech [out of 12,103 in the Odyssey] … The corresponding figures for the Iliad are 7,018 out of 15,690.’ But what might these line-counts and percentages mean for our reading of the Argonautica? What are the implications for the reading experience of a narrative in which characters talk less often? That ‘less often’ depends on a degree of familiarity with Homeric epic to measure degree of difference but what interests me for the here and now is just the 29% itself – in its composition, in breaking that composition down, in looking for patterns and anomalies that suggest routes for future exploration.

29%… My fear at this percentage would be its implications for the reader’s experience of qualia, of the degree to which the reader experiences ‘what it’s like’ to voyage on the Argo to Colchis and back. As pointed out by David Herman in his case study of Hemingway’s ‘Hills Like White Elephants’, in a behaviourist mode of narration ‘presenting only overt, surface behaviors of the characters and omitting narratorial commentary on more or less fugitive internal states (dispositions, thoughts, attitudes, memories, etc.), as the characters’ conversation unfolds in the story a rich context of felt experience emerges’ (Herman 2009: 147). This is not to equate the Argonautic mode of narration with Hemingway’s, only to point out that speech is always potentially fertile ground, and perhaps the more then the better. Still, we’re not going to get our hands dirty here but zoom up and out, turning text into tables and aiming to mark from above some spots for future digging.

Speakers and Speeches in A.R. 1-4 (5,835 lines)


This table is modelled on Nick Lowe’s Speakers and Speeches in the Iliad (Lowe 2000: 117). However, I’ve ordered speakers here by speech frequency rather than alphabetically; I haven’t differentiated between Greeks, non-Greeks and Gods; and I have added the text percentage column (29.25%).

In Lowe’s gaming analogy, characters are players and the spatial element of their narrative universe is the game board. The Iliad has two clearly defined player groups in Greeks and Trojans with the Gods operating at a ‘control level’ (Lowe 2000: 36-60). My first table here is simply all speakers together, looking for stand-out performers. Jason is a runaway leader, with Medea a clear second. Scanning down the list, the reader might be struck by the lack of lines given to ‘names’ like Heracles and Orpheus or wonder why Argus and Phineus, characters encountered on the outbound voyage, have so much to say for themselves. Jason’s totals are very respectable and transposed to Lowe’s Iliadic table would place him fifth on speech frequency and seventh on number of lines spoken (in both cases behind Zeus who is nowhere to be seen on the Argonautic table where instead Hera is the most prominent Olympian).

My table has a top ten featuring five male and five female characters (two Colchians, two Olympians and a Lemnian) and a top five featuring three characters who only enter the narrative after 42% percent of the available text time has elapsed (Argus coming into play at A.R. 2.1093). What about just the Argonauts?


29.25% has dropped to 9.96%. Jason (and the Argonauts)? He has one more speech than all the other Argonauts combined and he’s hogging 57% of the available lines. Whilst we’re not here analysing speech content, perhaps Idas should not have been so quick to go on the offensive (or perhaps that’s just Idas).

In the Catalogue of Heroes (A.R. 1.23-233) there are fifty-four names on the roll-call of those who answer Jason’s invitation. In this table of speakers there are fourteen names following Jason’s, supplemented by three instances where speech is direct but not attributed to a specific Argonaut (2.145-153, 4.1251-1258, 4.1458-1460).

They still do things. Whilst poor Polydeuces is propping up the table with his three lines, those three lines are an acceptance of the barbaric Amycus’ challenge and in the boxing match which ensues Polydeuces does his talking with his fists. He has his moment. Other Argonauts are operating on a restricted time scale e.g. Heracles and Polyphemus are removed from play in Cius in Book 1 following Hylas’ abduction. Polyphemus’ announcement of that disappearance to Heracles and their decision to stay and search for Hylas ends their Argonautic journey. Thirty Argonauts might have nothing to say (directly) but all ‘stories have a necessarily finite cast’ (Lowe 2000: 28) and ‘characters in a story world divide into stars and walk-ons: full players, and comparatively faceless and incidental extras’ (Lowe 2000: 46).

Speech matters because speech helps us both flag a character’s importance and give a character substance. Who are the stars, the support, the extras? Frequency (the number of opportunities a given character speaks) ‘indexes a character’s importance in plot’ and length (how much that character has to say for themselves) ‘is more a measure of their reactive and psychological interest’ (Lowe 2000: 116). Of course, we’d need to look more closely at content to assess the mind-modelling potential of length, taking into account e.g. the opportunity it also allows for exposition. Only a closer reading will allow us to determine not only who speaks and for how long, but also who gets the best lines. E.g.

‘The striking difference even a single brief speaking scene can make is highlighted in the narrative difference between the faceless, passive Chryseis, a mere token in the game, and the otherwise parallel Briseis – whose status as a minor player, the Greek mirror-image of Andromache, is memorably impressed by her solitary speech in [Iliad] XIX (Lowe 2000: 115-5).’

However, (in this post at least) we’ll resist the temptation to zoom in and instead break down the assembled table into its component books. What stands out?

Speakers and Speeches in A.R. 1 (1,362 lines)


Speakers and Speeches in A.R. 2 (1,285 lines)


Speakers and Speeches in A.R. 3 (1,407 lines)


Speakers and Speeches in A.R. 4 (1,781 lines)


In overall frequency, the ratio between Argonauts and Others is roughly equal in Book 1, 2 to 1 in favour in Book 2 and 2 to 1 against in Books 3 and 4 (Colchis and the voyage home). In Book 2, two non-Argonauts feature prominently in both frequency and length: Phineus and Argus. As noted above, we’d need to look at content and consider how much of their character-text is exposition, embedded narrative, prophecy (in Phineus’ case) before being able to evaluate how their speech contributes to mind-modelling them as individuals. Still the figures themselves mark them as worthy of further investigation. Jason’s own numbers spike in Book 3 as does the % Text in comparison with the other three books. Book 3‘s average percentage (39.6%) came closest to the Iliadic average percentage (45%). And somebody else is topping the charts in Book 3 (and Book 4): Medea.

If we compare just the figures from the Argonauts’ arrival in Colchis when Medea is potentially ‘in play’, Jason is no longer the epic’s dominant speaker.

Medea and Jason in A.R. 3-4 (3,188 lines)


Who gets to speak? How often do they speak? How much do they have to say? When do they get to speak? Jason’s speech acts spike in Book 3 because of Medea, because he needs her help, because of the dialogues they share in Books 3 and 4. What we can also view at this remove from content is where character-text occurs.

Speech Distribution in A.R. 1-4 (5,835 lines)


Here we have red blocks of character-text alternating with blue blocks of narrator-text. In Book 1, following the Catalogue of Heroes, we can see the initial cluster of speeches before the voyage gets underway then that lengthy expanse of blue before arrival at Lemnos. Following this, more blue interrupted by a solitary speech (Mopsus 1.1092) before the book ends with a final burst. There are only three more speeches in Book 2 than in Book 1, but with no introductory Catalogue there is a more even distribution, and thanks to Phineus’ appearance in the narrative, a hefty chunk of red prophecy as we’re a quarter of the way in.

As already noted, Book 3 is exceptional in its speech acts and what we see in the distribution is a marked increase in clusters of dialogue exchanges; from the Olympian goddesses deliberating as the book begins, through conversations amongst the Argonauts, the audience with Aeetes, Medea’s conversation with Chalciope and finally onto her meeting with Jason, characters are talking to one another. The large blue block which sees out the book contains summaries of Medea and Jason’s returns from their meeting, before more scenic treatments of Jason’s preparations for the trial and then the trial itself.

Up until that point, however, the reading experience of Book 3 features a markedly high concentration of speech acts (48.7% of available text time), including three of the poem’s five monologues: 3.464-470 (Medea on returning to her room), 636-70 (Medea on waking from her dream), 727-39 (Medea contemplating suicide). Direct Speech Acts are the closest approximation of Story Time to Real Time (characters speak at the same speed we do, see Measuring Arrows in Time), therefore all those clusters of red slats indicate scenic presentations. This has implications for our immersion as readers, standing there as ratified participants watching and listening to situations unfolding in (more or less) Real Time (supplemented by e.g. Book 3’s usage of simultaneous action).

Three monologues. The only other two monologues come at the beginning of Book 4: Medea’s short lament before she leaves her room a final time (4.30-33) and then a short and scornful speech from the watching Moon as Medea flees to the Argo (4.57-65, the Moon does address her directly though she’s a little far away for us to consider it a potential dialogue, leaving the reader the only effective/affected participant). If we are not allowed direct access to a character’s thoughts, the monologue in which characters articulate those thoughts can be invaluable to our mind-modelling efforts and Medea offers us four.

However, looking the table again, any reader expectation of a sustained increase in speech acts appears misguided and that the situation on Colchis was an anomaly rather than a development. The distribution pattern on the return voyage is a return to the pattern of the outbound – speech bursts sporadically interspersed in those great washes of blue. This might lead to a sense of distance, of being kept out, of having to do more guesswork. What caught my eye on the table was the blue expanse beginning at 4.421 interrupted by a solitary voice (much like the post-Lemnos pattern in Book 1). Here that speaker is Peleus proposing a plan to escape their Colchian pursuers (4.495-502). Now, moving just a little closer to identify speakers, the last speaker prior to this was Medea concluding a dialogue exchange with Jason which precedes their murder of her brother Apsyrtus (a non-speaking Colchian). The next speaker after Peleus is Circe when the Argonauts arrive at her island (preceding Odysseus’ footsteps but following the Odyssey’s). At this point I was curious as to when the next Argonaut spoke. The answer? Anonymous at 4.1251-58 when the Argo is stranded on the shoals of Libya.

The voyage home is a strange one. Phineus prophesied the way to Colchis but left their return to the gods (2.420-2). The Argo’s voyage in the final book takes us up the Istros to the Adriatic, through central Europe into the Mediterranean, then down the west coast of Italy to Libya before onto Crete and finally into the Aegean.

‘[Book 4] is characterized by scenes of flight, of despair, and of deception, but also by an eerie other-worldliness … which we have good reason to believe was as experimental when Apollonius composed it as it seems to us now. As the Argonauts confront one such τέρας after another, readers too are forced to stretch their imaginations to encompass the new and the strange: Apollonius makes all of us fellow-travellers with the Argonauts.’
(Hunter 2015: 5)

Potentially the strangeness is being reinforced by our disengagement for large sections of narrative from explicit commentary by the participants? Again, something to investigate.

We could look at other distribution patterns. Gendered Speech Distribution? Within that passage of relative Argonaut/male silence, the speakers from 4.411-4.1251 run Medea-Peleus-Circe-Hera-Hera-Thetis-Thetis-Medea-Medea-Arete-Alcinous. But that possibility only takes me back to Lemnos.

Speakers and Speeches in the Lemnian Episode (A.R. 1.609-909)


Here we find two of Jason’s seven speech acts in Book 1. His average speech length of 8 here is down from a 9.9 average for Book 1 and a 10.4 average for the epic in its entirety. The Argonauts have as many speeches and more to say (just about) than anyone else in Book 1 as a whole. Not on Lemnos.

The Argonautica is an episodic narrative. Each landfall allows at least an opportunity for interaction with the inhabitants, for new experiences (and novel narrations?). Still, the default/anticipated reader-experience would be one more (than less) alongside the characters we have already flagged as our protagonists.

Star Trek is episodic. Suppose after being introduced to Kirk, Spock, McCoy etc. (those emphatically male original players), the Enterprise’s maiden voyage brought us to an inhabited planet. Suppose at that point we jump from the ship to the planet and spend most of the episode with its inhabitants until Kirk finally pitched up to say a few lines and the Enterprise then goes on its way to the next episode. Would that coincide with or deviate from our expectations generated by both the narrative so far and our experience of similar narrative beginnings?

That’s an exaggerated hypothetical analogy of the Lemnian experience (in this Argonautica). It’s not that the Argonauts do nothing on Lemnos, it’s just that (I maintain) the Lemnian women do more, they take up more text, our exposure to them within the confines of the episode is greater. The speech figures above could lend support to this claim, which I’ve made previously (See Adjusting the dynamics of narrative interest). We could also bring in e.g. devices for initiating and preventing perceptual shifts (Stockwell 2002: 53) and note their distribution in the text. Or we could look at the contextual: e.g. the locational speech distribution in a linear reading of Argonautica 1.

pasted image 0

There are five locations in which Direct Speech occurs in Book 1: Iolcus/Pagasae before the launch, Lemnos, Cyzicus, Cius, and finally a first occasion for dialogue aboard the Argo. Distribution in the first setting is reasonably/relatively even. We hear the reactions of the locals to the assembly of heroes, then the exchange between Jason and his mother Alcimede. The next cluster is the leadership ‘debate’ (Jason and Heracles speak) which is followed by Jason’s prayer to Apollo before the argument between Idas and Idmon noted above. Then the characters fall silent, until after the narrator’s preliminary exposition on Lemnos, we find that cluster of speakers and speeches when the Lemnian women hold an assembly. Then the largest speech block of the book occurs when Hypsipyle gives her own exposition to Jason (and reply to the narrator). And after Lemnos? Isolated blips before Telamon and Jason exchange words aboard the ship.

It’s quite striking that once the expedition is underway, the Argonauts at every landfall have remarkably little to say (directly). The places they land, the people they meet, the sights they pass – what are they thinking? what are they feeling? Won’t somebody say something?

Well, yes; the narrator, the narrator who has carried us from Lemnos through the Hellespont and into the Propontis almost without interruption from the cast. Richard Hunter follows up his own percentages by noting that whilst the transition from oral to written epic plays a part, it [the reduced Argonautic percentage] ‘must also be viewed in the context of the insistent authorial voice’ (Hunter 1993: 41). The Argonautic narrator is speaking for them. Fortunately for us, this narrator is prominent, intrusive and involved (not behaviourist).

‘One very clear and important pattern in the Argonautica is the assimilation of the narrator’s experiences to the characters… The narrator often feels emotions analogous to those of the characters and in various ways their situation often reflects his.’
(Morrison 2007: 306)

All then is far from lost; it’s just more complicated. And it’s probably evident from previous posts that I’m not overly interested in statistic collecting, in amassing these figures which take us far from the path of song. Still, once assembled (and in the process of assembly) they did succeed in drawing attention to potential difficulties, in highlighting areas of interest and in prompting speculations.

E.g. How often does Jason initiate communication? How often is he responder? His last three speeches in Book 1 are responses (twice to Lemnian queen Hypsipyle and then when accepting Telamon’s apology). These are his only speeches since his prayer before departure (1.411). The next speech he offers unprompted is to the prophet Phineus (2.411, a fortuitous line correspondence). Phineus certainly has a lot to say but what is he saying? Hunter’s assessment of his prophesying is not promising: ‘the pedantic precision of his instructions’ (Hunter 2015: 7). On the other hand, Andrew Morrison has noted intriguing similarities over the expanse of Phineus’ speeches between Phineus and the narrator e.g. scholarly character; geographical and ethnographical detailing; and use of ‘break-offs’ (Morrison 2007: 306-7). Peleus has the highest frequency of speeches amongst the Argonauts with five. A cursory glance suggests they’re largely hortatory in function; he’s the ‘strong voice in the dressing room.’ Still, going back to that 29.25%, all speeches available deserve more than a cursory glance.

In ‘Speakers and Speeches in the Argonautica: Modes of Speech’, we’ll zoom in again with these problems/questions in mind. We’ll look at direct and indirect discourse and consider the further implications of angled dialogue narration. We’ll bring in Positioning Theory (Herman 2009:55f.). And we’ll tackle some specific examples of speech acts, evaluating content. Yes, length does flag a speech’s potential for ‘reactive and psychological interest’ but for that potential to be realised, we’ll need quality as well as quantity.



Herman, D. (2009) Basic Elements of Narrative, Oxford.
Hunter, R. L. (1989) Apollonius of Rhodes: Argonautica Book III, Cambridge.
Hunter, R. L. (1993) The “Argonautica” of Apollonius: Literary Studies, Cambridge.
Hunter, R. L. (2015) Apollonius of Rhodes: Argonautica Book IV, Cambridge.
Lowe, N. J. (2000) The Classical Plot and the Invention of Western Narrative, Cambridge.
Morrison, A. D. (2007) The Narrator in Archaic Greek and Hellenistic Poetry, Cambridge.
Rutherford, R. B. (1992) Odyssey. Books XIX and XX, Cambridge.
Stockwell, P. (2002) Cognitive Poetics: An Introduction, London.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s