On the absence of impersonal passives of deponent verbs

Question

In a previous post there's a discussion on an intriguing example of a passive construction of a transitive (allegedly) deponent verb: Ab amīcīs hortārētur (Did Latin have any ergative verbs? ). The cautious parentheses are added because of the existence of the non-deponent forms horto, -are: e.g., cf. http://micmap.org/dicfro/search/gaffiot/hortor

As a follow-up question from this discussion, it is worth pointing out that impersonal passives of intransitive deponent verbs like *Mortuum est beate ‘{One/people} died happily’ are ill-formed, compared to ok Pugnatum est acriter. The explanation of this contrast has been argued to be related to the fact that the former is an unaccusative verb, whereas the latter is an unergative one (vid. https://en.wikipedia.org/wiki/Impersonal_passive_voice ; for the well-known distinction between unaccusative vs. unergative verbs, vid. https://en.wikipedia.org/wiki/Unaccusative_verb ).

This distinction notwithstanding, I was wondering why the previous impersonal passive construction with the deponent verb mori (*Mortuum est beate) is ill-formed whereas Moriendum est omnibus is well-formed. Any thoughts on this contrast?

Here are some additional remarks drawn from my comments on Cerberus's interesting answer/contribution below: intransitive agentive verbs (e.g., pugnare) can appear in impersonal passive constructions but intransitive non-agentive (i.e. "lexically passive") verbs (e.g., mori) cannot. In short, it seems that intransitive verbs with an already "lexically passive meaning" cannot appear in a syntactic passive construction (e.g., into an impersonal passive construction). NB: the ill-formed example *Mortuum est beate is an impersonal passive construction but the well-formed example Moriendum est nobis is also an impersonal construction (but, crucially, it is not passive).

I don't think you misanalyzed the construction/the example from Pliny. Notice that my example above of impossible impersonal passive is that of the unaccusative verb mori. Impersonal passivization has been said to be a good diagnostic to divide unergatives (e.g., to fight) from unaccusatives (e.g., to die). See "test of unergative verbs" in https://en.wikipedia.org/wiki/Impersonal_passive_voice — Mitomino, Mar 31 '19 at 03:30
(I deleted my previous comment because I checked L&S and saw that it lists the passive use of "testatum" under the collateral form "testo"). — Asteroides, Mar 31 '19 at 08:16
Your example from Pliny was very interesting. Anyway, besides the typical (but perhaps not systematic; see above) absence of passives of transitive deponent verbs, I'm also quite interested, as noted above, in why Mortuum est beate (on the impersonal reading!) is ill-formed compared to the well-formedness of Moriendum est nobis, which is also an impersonal construction. — Mitomino, Apr 03 '19 at 19:42
@tony I'm sorry but I can't remember Pliny's example put forward by sumelic (now Asteroides). The comment was deleted. — Mitomino, Apr 03 '20 at 15:41
@tony: I'm pretty sure it was "Nam patrem illarum defunctum quoque perseverantissime diligo, ut actione mea librisque testatum est; cui nunc unus ex tribus liberis superest, domumque pluribus adminiculis paulo ante fundatam desolatus fulcit ac sustinet." I think a moderator could check the deleted comment. — Asteroides, Apr 04 '20 at 10:58
Two parts (1/2 ) Just for reference I place here the deponents verbs I could extract from L&S with impersonal passive examples): medeor; tumultuor; cunctor; assentior;
continue below in part 2 — d_e, Jun 04 '21 at 21:42
@d_e Thanks for these examples. Note that some of them seem to have an active correlate as well: medeo(r), tumultuo(r), cuncto(r), etc. E.g., cf. http://micmap.org/dicfro/search/gaffiot/tumultuor It seems that the relevant semantic condition for an impersonal passive construction to be well-formed is agentivity. This could explain why tumultuo(r) (but not morior) enters into this construction. — Mitomino, Jun 04 '21 at 23:50

score 3 · Answer 1 · answered Apr 03 '20 at 03:30

3

I have a thought: the gerundive (along with gerund and present participle) is exceptional in deponent verbs. Whereas all other passive forms of deponents (both past passive participles and passive endings of the present stem) are translated 'actively', this does not apply to the gerundive, which is still translated 'passively' as normal, e.g.:

Ovidius, Metamorphoses XIII.193:

mittor et ad matrem, quae non hortanda, sed astu decipienda fuit ...

"Then I was sent to the mother, who was not to be exhorted, but deceived by craft."

It is not "who was to exhort", but "who was to be exhorted". So it does not surprise me very much that the gerundive should behave differently in other aspects as well. But this is somewhat less than an answer to your question.

answered Apr 03 '20 at 03:30

Cerberus

19,914
3
57
110

I fully agree with you that the gerundive forms in your example from Ovidius must be translated "passively". It seems to me that the impersonal construction with the gerundive of mori (Moriendum est nobis) is not interpreted this way, i.e., "passively". – Mitomino Apr 03 '20 at 15:37
In contrast, the impersonal passive construction with the past participle of mori (*Mortuum est beate) sounds quite odd precisely because we're using a verb with a "lexically passive meaning" (mori) in an impersonal passive construction. Typically, intransitive agentive verbs can appear in impersonal passive constructions but intransitive non-agentive (i.e. "lexically passive") verbs (e.g., mori) cannot. In short, it seems that intransitive verbs with a "lexically passive meaning" cannot appear in a syntactic passive construction (e.g., into an impersonal passive construction). – Mitomino Apr 03 '20 at 15:52
Note that the conclusion from my previous comment is that the ill-formed example *Mortuum est beate is an impersonal passive construction but the well-formed example Moriendum est nobis is also an impersonal construction (but, crucially, it is not passive). – Mitomino Apr 03 '20 at 15:53
@Mitomino: I agree with you that moriendum est nobis is semantically not 'passive', unlike exhortanda est mater. All I wanted to suggest was that perhaps there was something odd about the gerundive with deponent verbs, but it may very well not be relevant, if we concede that the gerundive with inagentive deponent verbs is in fact quite regular, i.e. it does not swap the praedicate frame. – Cerberus Apr 03 '20 at 17:14
@Mitomino I do have a question about Mortuum est beate: how do we know this is not possible? Admittedly, a cursory search did not turn up any examples of unagentive deponent verbs used impersonally, but can we be sure? – Cerberus Apr 03 '20 at 17:16
The question you've raised is typical: since we don't have native competence of Latin, how do we know that "X" is not possible in this language? (Same with similar questions: e.g., how do we know that a resultative construction like We drank the teapot dry is ungrammatical in Latin? For an answer, see this book: https://global.oup.com/academic/product/the-morphosyntax-of-transitions-9780198733287?cc=es&lang=en& ). It seems to me that the methodology we can use to provide a solution to these questions is similar. For example, linguistic typology and comparative grammar usually help a lot. – Mitomino Apr 03 '20 at 17:30
@Mitomino: Yes, of course it is difficult to prove anything about a dead language, especially that something should not be possible. If that book is your source, you might want to add a quotation where it says that at some point (though I'll take your word for it!). – Cerberus Apr 03 '20 at 18:26

score 1 · Accepted Answer · answered May 19 '19 at 20:05

1

Impersonal passive deponents are rare, but not unknown. See Pinkster, Oxford Latin Syntax, 1.5.34.

As forms that occur more than once, he lists morabitur, sortiri, and medeatur. He refers to a fuller list in Flobert, Pierre (1975). Les verbes déponents latins: Des origines à Charlemagne. Paris: Les Belles Lettres.

answered May 19 '19 at 20:05

Kingshorsey

6,505
15
21

Thanks! It is indeed interesting to compare Pinkster's (2015: 284) quote that "Impersonal passive deponent forms are rare" with the one he also makes immediately below: "Gerundive forms of deponent verbs with a passive interpretation are found throughout Latinity". So WHY is there a contrast of productivity here, i.e., between impersonal passives made out of alleged deponent verbs (e.g., cf. medeor AND medeo) and impersonal constructions made out of Gerundive forms of deponent verbs? – Mitomino May 19 '19 at 20:51
Assuming that Pinkster’s admittedly rare examples of passive deponent forms cannot be reduced to false cases (e.g., cf. medeor and medeo), probably there is an agent involved in these rare cases, like in non-deponent unergatives (Pugnatum est) and unlike in deponent unaccusatives (*Mortuum est beate). In that case the contrast could have to do with unergatives (deponents or not) and deponent unaccusatives. – Mitomino May 19 '19 at 21:00

score 0 · Answer 3 · answered Mar 31 '19 at 08:01

0

'Mortuum est beate' can both be translated as he, she or it. So it can mean: 'It has died happily', which doesn't refer to a person, but is semantically equivalent to an animal in English.

If you mean why it can't refer to an inanimate object, then that's something which depends on the meaning of the verb, as inanimate objects obviously can't die. So the answer primarily lies in the meaning of deponent verbs.

It also isn't true that there are no deponent verbs which are impersonal, considering 'conatum est' for example means: 'It has been tried/attemped', which refers to the thing which has been attempted and not the person. Same with the sentence 'pugnatum est', which is a non-deponent verb but is used similarly.

answered Mar 31 '19 at 08:01

Stallmp

153
3

2

Do you know of any quotes that could be used as an example of impersonal "conatum est"? – Asteroides Mar 31 '19 at 08:25
Stallmp, I'm afraid that there is a confusion here. Of course, I'm not saying that Mortuum est is impossible when its (elliptical) subject is a neuter nominative (an animal, a plant or whatever). I'm referring to the IMPERSONAL construction (see the wikipedia link above). Besides this, I'm wondering why impersonal Moriendum est nobis is well-formed, compared to Mortuum est beate, which is ill-formed. – Mitomino Apr 03 '19 at 15:55
Well, have you read the next paragraph? – Stallmp Apr 03 '19 at 16:02
My objection to your final paragraph is essentially the same raised by sumelic above. Regarding conatum est, are you sure it can be considered an example of an impersonal passive construction or ... rather an example of a passive construction with an elided nominative Hoc? In any case notice it is not the same as impersonal pugnatum est It is well-known that unergative verbs like "to fight" do enter into the impersonal passive construction: see https://en.wikipedia.org/wiki/Impersonal_passive_voice. Note that (Hoc (bellum)) pugnatum est is a different (not impersonal!) construction. – Mitomino Apr 03 '19 at 16:21
That is to say, (Hoc bellum) pugnatum est is the same as (Haec pugna) pugnata est. These passive constructions are not impersonal but passive constructions with an elided nominative subject: hoc bellum and haec pugna, respectively. – Mitomino Apr 03 '19 at 16:32
To be fair, I think the answer to your question would be the same as the answer to the question: 'Why is 'this word' masculine?'. There is no real explanation for this, this is just purely induction and you aren't even sure yet whether this is the case or not. Even if it is, like I said, there might not even be an explanation for it. – Stallmp Apr 03 '19 at 16:36
1

Ok, so we disagree. I don't think that the question of why the impersonal passive of a deponent verb like mori (mortuum est beate) is ill-formed (compared to the well-formedness of an impersonal passive of a non-deponent verb like pugnare: pugnatum est) is like the question of why this or that word is masculine. This said, if your potentially interesting example conatum est used as passive ('it has been tried/attempted') is actually found in (Classical?) Latin texts, I'll be happy to take a look at them. Thanks! – Mitomino Apr 03 '19 at 17:01
Wasn't referring to that question, but to 'I was wondering why there are not impersonal passives of deponant verbs'. And I have seen it before, I will be able to quote it once I find it, which I've seen a few years ago at school I'm afraid. – Stallmp Apr 03 '19 at 17:19
Ok. Thanks for your clarification. – Mitomino Apr 03 '19 at 17:22
@Mitomino: Yourself is most insistent that "moriendum est nobis" is not passive. When rendered into English, "we must die", this is correct; but, in Latin, to the Romans, it would always be "for us it-ought-to-be-dead (dying; death)--the passive-- wouldn't it? How are things in Barcelona? – tony Apr 14 '20 at 10:24
@tony Yes, my surmise is that moriendum est nobis is not passive. Otherwise, given what I've said in my post (see also the 3rd link above), I'm afraid that one cannot give an explanation of why the truly passive construction mortuum est beate is ill-formed. Regards from Barcelona. – Mitomino Apr 15 '20 at 22:10

On the absence of impersonal passives of deponent verbs

3 Answers3

Linked