I'm stuck while reading Ennius' "Cūrantēs magnā cum cūrā", written in dactylic hexameters.

I added to the text some macrons and caesuras that are of my own. I scrupulously respected what little I knew about Ennius' latin, especially long vowels in active.indicative.present.3S (e.g. in seruāt). Caesuras are largely cosmetic here : please do not take them too seriously. Feel free to correct me if necessary !

What bothers me is line 96:

[096] laeua uolāuit auis. | Simul aureus exorītur sōl

[096] […] on the left flew a bird. Meanwhile a golden sun appeared […]

I can't scan this line as a regular dactylic hexameter. Everything happens as if exorītur sōl should be read exoritur sōl. But exorītur comes from exorīrī, hence my amazement.

What I'm missing ? Thanks !

update : There's something interesting about exoritur. Virgil has exoritur (not exorītur), e.g. in Eneid 2.213:

exoritur clāmorque uirum clangorque tubārum.

The verb seems to hesitate between several stems : see oreris in Ovid Metam. 10.166. :

tū totiēns oreris uiridīqu(e) in caespite flōrēs.

[080]    Cūrantēs | magnā cum cūrā | tum cupientēs
[081]    regnī | dant operam | simul auspici(ō) auguriōque.
[083]    [-] Remus auspiciō | sē dēuouet | atque secundam
[084]    sōlus auem seruāt. | At Rōmulus pulcher in altō
[085]    quaerit Auentīnō | seruāt genus altiuolantum.
[086]    Certābant urbem | Rōmam Remoramne uocārent.
[087]    Omnibus cūra uirīs | uter essēt induperātor.
[088]    Exspectant uel utī | cōnsul cum mittere signum
[089]    uolt | omnēs auidī | spectant ad carceris ōrās
[090]    quam mox ēmittat | pictīs ē faucibus currus
[091]    sīc exspectābat | populus | atqu(e) ōra tenēbat
[092]    rēbus | utrī magnī | uictōria | sit data regnī.
[093]    intereā | sōl albu(s) recessit | in īnfera noctis.
[094]    Exin candida sē radiīs | dedit īcta forās lux
[095]    et simul ex altō | longē pulcherruma praepes
[096]    laeua uolāuit auis. | Simul aureus exoritur sōl
[097]    Cēdunt dē caelō | ter quattuor corpora sancta
[098]    auium | praepetibus sēsē | pulchrīsque locīs dant.
[099]    Cōnspicit inde sibī | data Rōmulus esse priōra
[100]    auspiciō regnī | stabilīta | scamna solumque.
  • auium(098) is bisyllabic (confer Ernoux, Recueil de textes latins archaïques, p. 150)
  • same problem with quattuor(097), also bisyllabic (confer Ernoux), some editors read "quattor".
  • "albus" must be read "albu" since "[f]inal /s/ [...] was weakly articulated in Old Latin." ("The Blackwell history of the Latin language, p.98")
  • seruāt(084, 085) : Old Latin (not Latin) had a long here. See e.g. "The Blackwell history of the Latin language, p.98" speaking about the "charateristic Latin shortening of long vowels before [...] /-t/ (3rd person singular)"
  • 2
    Well, it’s not my insight. I have rearranged the listing order, but I took all information from Ernout’s book, pp. 147-152, which cites Niedermann, Une loi rythmique proethnique en latin, Mélanges de Saussure, Paris, Champion, 1908, as the original source.
    – Dario
    Commented Mar 13, 2019 at 21:53

2 Answers 2


My source for this answer is A. Ernout, Morphologie historique du latin, Paris 1974.

1. orĭtur/orītur

Verbs in -iō (originated by a well-known indo-european -ye/o- suffix attached directly to the verbal root) belong either to the third (ĭ) or the fourth (ī) conjugation, according to a set of rules:

  1. Monosyllabic radical, ī: sciō, scīre.

  2. After a long syllable, ī: dormiō, dormīre; rūgiō, rūgīre.

  3. After two short syllables, ī: aperiō, aperīre.

  4. After a sequence of long+short, ĭ: cōnspiciō, conspicere (only compounds.)

  5. After one short syllable followed by a stop, ĭ: capiō, capere; faciō, facere.

  6. After one short syllable followed by l, n, r, v, mostly ī: veniō, venīre, with several deviations and uncertainties (morĭor, mori but archaic moriri; pariō, parere but archaic parire...)

  7. Verbs derived from nouns have a tendency to generalise ī, contrary to the above rules: sĭtiō, sĭtire.

Orĭor belongs to the uncertainties of point 6. In the infinitive, only oriri is attested, but forms like oreris, orĭtur, orĕrētur are the normal ones. Ennius is archaic, and uses morīmur instead of classical morĭmur, (Ann. 392), but also horĭtur (Ann. 432) and, as you noted, orĭtur. So, no problem here.

Developments in Romance languages show that the forms in actual spoken usage must have been more varied than those codified by classical prose.

2. seruāt

In classical times, the a was shortened and we have amăt, servăt, etc., but an ā is to be expected from etymology and we do find it in archaic times, in Ennius as in this quotation, but, e.g., also in Plautus: fundum alienum arāt, incultum familiarem deserit (Asin. 874, troch. hept.).

Hope this helps.

  • 1
    "Hope this helps". It does, indeed !
    – suizokukan
    Commented Mar 13, 2019 at 10:12

Here are some remarks concerning the marked vowel lengths, caesuras, and scansion. Many of them are tangential to your actual question, but I hope they are useful. Most of the caesuras and macrons are correct; I hope my analysis does not come across as too negative for picking mistakes. First remarks:

  • I should stress that my point of view is that of classical Latin, and may not be entirely appropriate for Ennius. I am not as familiar with older Latin.
  • Some authors use macrons to denote not only long vowels but also long syllables. I have assumed that you have used them for vowel length only. (I only ever use them so myself, but it is good to be aware of the alternatives.)
  • The answer to your actual question seems to be that the length of i in the stem of oriri and derivatives can vary, and these verbs seem to be somewhere between conjugations three and four.
  • Caesuras might be just cosmetics for you, but as you have marked them, I find it beneficial for other readers to comment on them.

[080] Cūrantēs | magnā cum cūrā | tum cupientēs

Caesuras are typically — I would even say always, but there may be exceptions — placed within a dactyl, not between them. Therefore the second caesura seems invalid to me. Caesuras are typically (≈ always) confined to the second, third, and fourth feet of a verse. The two possible points for a ceasura here are before and after magna. You can consider it to have both caesuras. I would only put the one after magna so as to split the verse more evenly. (I do not typically put a caesura if there are several options in the same verse, like in 86.)

[081] regnī | dant operam | simul auspici(ō) auguriōque.

The same caesura rules from [080] apply: The two possible positions are right before and after operam.

[084] sōlus auem seruāt. | At Rōmulus pulcher in altō

The vowel before the third person ending -t in servat, audit, and such is short. Therefore the macron in servat should not be there. Scansion does require the syllable to be long (heavy) there, though. This is achieved by a longer pause, and my experience suggests that this is particularly common with punctuation. (Ennius may not have had punctuation like we do, but a pause for the shift from one sentence or thought to another still makes sense.)

[085] quaerit Auentīnō | seruāt genus altiuolantum.

There is a spurious macron in servat.

[089] uolt | omnēs auidī | spectant ad carceris ōrās

There should be no caesura in the first foot.

[091] sīc exspectābat | populus | atqu(e) ōra tenēbat

Here I would only use the second caesura so as to emphasize it. Namely, the last syllable of populus needs to be scanned long, and a sole caesura helps there.

[092] rēbus | utrī magnī | uictōria | sit data regnī.

The only two caesura possibilities are before and after magni.

[093] intereā | sōl albus recessit | in īnfera noctis.

Curiously, the second syllable of albus needs to be scanned short here. Whether this is excessive poetic licence is a matter of taste, but it certainly helps that it is an unstressed syllable for both prose and meter.

[094] Exin candida sē radiīs | dēdit īcta forās lux

Scansion requires that the first syllable of dedit is short. And indeed, the vowel is short. This is not a present form of dedere but a perfect form of dare.

[095] et simul ex altō | longe pulcherruma praepes

There is a macron missing from longe. The meter suggests it, and long -e is also standard morphology.

[096] laeua uolāuit auis. | Simul aureus exorītur sōl

Indeed, the i in exoritur has to be scanned short. I, too, expected it to be long. At first I thought this is mere poetic licence, but then I checked that Ovid would use a short i as well. The corresponding active form has a short i (as is always the case before a word-final t), and the i is short in i-stem verbs of the third conjugation.

My dictionary suggests that exoriri is a mixture of third and fourth conjugations, unlike oriri which is pure fourth. Exploring the morphology of prefixed versions of oriri would make a nice separate question. I can only say that the short i seems to be correct, but I cannot give the whole paradigm with confidence.

[098] auium | praepetibus sēsē | pulchrīsque locīs dant.

The first caesura looks misplaced. Both sides of sese work.

More interestingly, avium seems to require a long (heavy) first syllable. This can be achieved by reading it as avjum, with a consonantal i and only two syllables.

[100] auspiciō regnī | stabilīta | scamna solumque.

The second caesura falls between feet. A caesura only fits on both sides of regni.

  • Thanks for your detailed answers. I improved my text (see dedit*(094), *longē*(095) and my remarks about *auium*(098), *quattuor*(097), *albus*(093)). I beg to differ about *seruāt. See my remarks.
    – suizokukan
    Commented Mar 10, 2019 at 21:39
  • @sumelic : I beg to differ. I and my colleagues use macrons only to mark vowel length : it's the most untrivial information we need to guess syllable length.
    – suizokukan
    Commented Mar 10, 2019 at 23:11
  • 1
    @suizokukan Thanks for the updates! I added some remarks to the start of my answer. I can't tell when the a in servat shortened or if it was ever long in the first place; older Latin is not my strongest suit. I hope this answer is of use anyway.
    – Joonas Ilmavirta
    Commented Mar 11, 2019 at 12:40
  • yes, thank you very much for everything you noticed in my post, it was helpfull !
    – suizokukan
    Commented Mar 11, 2019 at 21:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.