Parsing "oblita carmina"

Question

Vergil wrote (Eclogues IX.51–4), quoted by Draconis in this answer:

Omnia fert aetas, animum quoque. Saepe ego longos
cantando puerum memini me condere soles.
Nunc oblita mihi tot carmina: vox quoque Moerin
iam fugit ipsa.

This passage is spoken by Moeris (in the first person and the third person!). The first two lines mean:

Age takes all, even memory. I recall myself as a boy, often putting away the long Suns [i.e. concluding the long summer days] by singing.

Now, how do we make sense of oblita carmina? I'll post my current thinking in this question; I'm looking for confirmation or correction.

1. The carmina forgot?

Obliviscor is a deponent verb, so oblita is active. Literally, the line would seem to mean:

Now so many songs have, to me, forgotten.

with an omitted sunt. Did the songs forget the summer days? This seems to push figurative speech awfully far.

2. Oblĭta?

Hey, maybe it's oblĭta, past participle of oblinere, to daub over, as with mud. This is not a deponent verb, so we have:

Now so many songs have, to me, been blotted out (as if) by mud.

That at least works grammatically, but does it scan? No. Both ob-lĭ-tă and ŏ-blĭ-tă would yield three short syllables in a row:

Nunc ob/līta mi/hī tot/carmina: / vox quoque / Moerin

3. Deponent with passive meaning?

Could Vergil have invoked poetic license, asking the listener to accept a passive meaning for oblīta?

Now so many songs are forgotten to me.

Does Vergil really take such liberties with grammar, though?

4. The aetas forgot?

Oblīta could agree with aetas, so the sentence has an omitted est:

Now the age has forgotten so many songs for me.

Isn't the subtext of this entire poem that Octavian's confiscation of land for the army has despoiled the land, tradition, and art—that "the din of arms has silenced poetry"? Or at least Vergil's poetry, if his farm was among those confiscated?

Aetas also means a period in history. So Omnia fert aetas, animum quoque has a double meaning: "The present age carries all, even the soul"—the present age referring to the rule of Octavian. Isn't this the kind of subtle political subversiveness that Vergil is famous for?

Just before this passage, there is a quotation from one of Moeris's forgotten songs, celebrating the rise of Caesar's star, which would make "the fields rejoice with crops". The message seems to be: in the present age, of Octavian's confiscation of farms, those songs of Caesar's beneficent leadership are obsolete and forgotten.

Answer 1

Lewis and short cite this very passage as an example of oblisci being used passively. So there's your answer, "forgotten songs": http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0059%3Aentry%3Dobliviscor

Answer 2

The practice of using deponent participles in a passive sense occurs with other verbs besides obliscor, as noted by Ethan Allen Andrews and Solomon Stoddard in A Grammar of the Latin Language: For the Use of Schools and Colleges:

The perfect participles of some deponent verbs have both an active and a passive sense; as, adeptus libertatem, having obtained liberty, or adepta libertate, liberty having been obtained.

So abominatus, comitatus, commentatus, complexus, confessus, contestatus, detestatus, dignatus, dimensus, emensus, ementitus, emeritus, expertus, exsecratus, interpretatus, largitus, machinatus, meditatus, mercatus, metatus, oblitus, orsus, pactus, partitus, perfunctus, pollicitus, testatus, veneratus, ultus.

As far as why this occurs, there doesn't seem to be a definitive reason. If I'm understanding the comment by Harm Pinkster correctly, it appears that it is a matter of certain authors simply choosing to use these words in a passive sense (i.e. something like the "poetic license" that you mentioned):

The use of a number of verb forms of deponent verbs with a passive meaning is widespread and found in all periods of Latin. The number of verbs with which this is possible increases steadily, with individual variation in creativity between authors. (The Oxford Latin Syntax)

Answer 3

Accepting the kind invitation by Ben (Sententiam tuam de passiv[a] resultativ[a] audiendi studiosus sum), here is my opinion on how to analyze oblita carmina (sunt is omitted) in this example. As I told to him and to Draconis (see here), oblita is better interpreted here as an adjectival/resultative participle rather than as a verbal/eventive passive participle. Two additional elements are relevant to get the interpretation of this sequence as an adjectival/resultative passive: the so-called "dative of agent" mihi and the adverb nunc.

Before dealing with the example from Virgil, some sketchy background on the verbal vs. adjectival passive can be useful. As noted here, a common test to distinguish both types of passives is the expression of the agent. Verbal passives typically express the agent via a by-phrase (e.g. a me 'by me'), whereas adjectival passives can sometimes do so via a so-called "dative of agent":

Verbal/eventive passive: (a) A me consilium captum est. ‘The decision {has been taken/was} taken by me’.

Adjectival/resultative passive: (b) Mihi consilium captum iam diu est. ‘I have my decision taken’ Cf. a more natural translation: 'My resolution has long been taken'.

Mihi consilium captum iam diu est. (Cic. Fam. 5.19.2) (NB: Mihi carmina oblita iam diu sunt would be an example of the same type of construction).

Cf. the following relevant remark made by the French latinist Lavency (1985/1997: 160): (captum) est in Mihi consilium captum est is the present form corresponding to the past one (captum) fuit in Mihi consilium captum fuit (note the use of fuit because of the adjectival form of captum), whereas captum est in A me consilium captum est is the past form corresponding to the present A me consilium capitur. Cf. also the very relevant parallelism between these two resultative constructions: Mihi consilium captum est and Habeo consilium captum (NB: a similar but admittedly trickier parallelism can also be found when dealing with gerundives: see here).

As for the different syntactic distibution of ablatives of agent and so-called/miscalled "datives of agent", please see here and here.

The verbal vs. adjectival passive distinction is not always obvious. The presence of nunc in the following example can help the reader to interpret it as an adjectival/resultative passive but the most famous example of this construction (Gallia est omnis divisa in partes tres) is interpreted as a resultative passive only contextually.

Magnamque Graeciam, quae nunc quidem deleta est, tum florebat, institutis et praeceptis suis erudierunt. (Cic. Amic. 13)

Turning back to our example at issue, my opinion is that the verbal/eventive interpretation of the construction ('Now all songs have been forgotten by me') is not so appropriate as the adjectival/resultative one (lit. 'Now I have all songs forgotten').

Draconis and I have also discussed the coappearance of "datives of agent" with finite verbal forms (cf. the typical example neque cernitur ulli from Verg. Aen. 1,440; quite probably, this is the example that will appear in your favorite Latin grammar after having said that these datives typically coappear with participles. E.g., see the link above to Lavency's VSVS). It should be noted that the coappearance of this kind of datives with finite verbal forms is not typically found in Cicero's/Caesar's classical prose but it can be found in poetry and imperial prose (is it a graecism?). This pattern has been said to be not productive by various scholars, compared to the more widespread pattern involving a non-finite/participial form (e.g. see the example above from Cicero). Personally, I find Suárez-Martínez's (2001: 605-606) explanation of finite verbal forms like that of cernitur ulli as quite illuminating (please see Section 5 of his paper): cf. the parallelism between Virgil's famous example neque cernitur ulli with the "subjacent"/expected one neque cretus est ulli. In any case, my conclusion is the following one: a clear(er) example of verbal passive with obliviscor would be one having a by-phrase: carmina a me oblita (sunt). I don't think that this example was natural in Classical Latin. At least, from what I argued for above it seems more appropriate to conclude that carmina mihi oblita (sunt) would sound more natural.

Note that my conclusion is different from OP's above ("Obliviscor is a deponent verb, so oblita is active"). I would not say that oblita is an active participle: it is the participial form of an adjectival/resultative passive, i.e., similar to the one found in Cicero's example above: mihi consilium captum iam diu est.

This said, let me conclude by saying that the distinction between "verbal" and "adjectival" passives is not always easy to be drawn (e.g., for a relevant discussion & critique of this traditional distinction, see this article by Embick (2004)).