5

An "ergative verb" is a verb that can either take two nouns (a subject and an object) or only one (a subject), where the object of the two-noun form corresponds to the subject of the one-noun form.

That sounds more complicated than it is: an easy example is "break" in English. If used with only one noun, the subject is the thing that breaks: "the window broke". But if used with two nouns, the object is the thing that breaks: "Marcus broke the window".

Does Latin have any verbs which can interchange the roles of their subject and object like this?

Improve this question
9
  • 2
    This is not what "ergativity" means (at least not in its technical sense). Ergativity is when the agent of a transitive verb is in an oblique case and its logical direct object is unmarked, as in Basque, Georgian, Urdu etc. etc.
    – fdb
    Sep 10, 2018 at 9:51
  • 2
    @fdb: the question seems to be about the phenomenon of “ergative verbs”, not about the terminology, which Draconis did not come up with (see the discussions on the Wikipedia article). Many linguistics terms have multiple meanings
    – Asteroides
    Sep 10, 2018 at 15:39
  • 2
    This is indeed rather unfortunate but that’s how things are in linguistics. Ergative is also used to describe English verbs like break: He broke the vase. The vase broke. This is particularly common in some theories of formal syntax, e.g. generative syntax (GB, MP). This use is not common in linguistic typology though.
    – Alex B.
    Sep 10, 2018 at 23:13
  • 2
    amazon.co.uk/Oxford-Latin-Syntax-Pinkster-2015-10-20/dp/… citing books.google.com.au/… . @alex-b, this warrants an answer. And an example.
    – Nick Nicholas
    Sep 11, 2018 at 3:41
  • 3
    @fdb That seems unnecessarily harsh; the terminology might be less than ideal, but that's nothing new. (See: gerunds vs gerundives, participles vs derived nouns…)
    – Draconis ♦
    Sep 11, 2018 at 16:37

2 Answers 2

Reset to default
5

If "ergativity" is to be (mis)understood à la Burzio (1986), i.e., as "unaccusativity" (https://en.wikipedia.org/wiki/Burzio%27s_generalization ), yes, Latin has been claimed to present so-called "split intransitivity", whereby the typical distinction within intransitive verbs (the one between unergatives vs. unaccusatives/Burzio’s ergatives) can be claimed to apply not only to English (e.g., see the classical work by Levin, Beth; Rappaport-Hovav, Malka (1995). Unaccusativity: At the Syntax-lexical Semantics Interface. Cambridge, MA: MIT Press.) but also to Latin: e.g., https://www.academia.edu/6483921/Dos_tipos_de_intransitividad_en_lat%C3%ADn_sintaxis_y_sem%C3%A1ntica

Within the set of English unaccusative verbs, it is true that many of them correspond to the anticausative labile type exemplified by the anticausative The glass broke (cf. the causative John broke the glass; NB: this phenomenon is often referred to as ‘causative alternation’). Lability, which involves no morphological mark, has been said to be particularly rare in languages with “middle inflection” (Ancient Greek, Latin, Arabic, Kartvelian): see http://aletuchiy.narod.ru/handouts_articles/Transitivity/handout_Lancaster_new.pdf . Probably, many of you will not be sympathetic with the fact that Latin is said to have middle voice. Well, this hypothesis is currently being defended by many linguists: e.g., for a very recent discussion, see Pinzin’s Doctoral dissertation Stuck in the Middle: a morphosyntactic analysis of the deponent verbs from Latin to Romance (pdf downloadable at http://dspace.unive.it/handle/10579/12877 ), where you’ll also find an interesting criticism of Embick’s (2000) idiosyncratic syntactic approach to deponent verbs).

Some examples of labile verbs in Latin follow:

Foris aperit. 'The door is opening'. (Pl. Pers. 300)

Ecquis has aperit foris? ‘Is anyone going to open this door?’ (Pl., Most. 900)

(cf. the non-labile use: ecce autem commodum aperitur foris ‘and look, at this very moment the door is opening’(Pl. Mil. 1198))

Hannibal post Cannensem pugnam [castraque] capta ac direpta confestim ex Apulia in Samnium moverat ‘After the battle of Cannae and the capture and plunder of the camp, Hannibal had moved at once out of Apulia into Samnium’ (Liv.23.1.1)

Hannibal silentio movit castra ‘Hannibal quietly moved the camp’ (Liv. 27.2.10)

Improve this answer
1
  • Googling the example of "Foris aperit" led me to a section in The Oxford Latin Syntax, Volume 1, (2015), by Harm Pinkster, that seems to have some discussion of this kind of verb; Pinkster mentions the example of grandio (§5.32, p. 281). Pinkster calls these "decausative" usages.
    – Asteroides
    Mar 31, 2019 at 2:05
4

I'm not going to use the term "ergative" because I don't understand its exact definition and when it is appropriate (and as the comments mention, it has another definition, so it can be confusing to use this terminology).

Passive in form "verba communia"

Something that seems to match what you describe on the surface level is something called "verba communia". Apparently, for some transitive deponent verbs, an identical form can be can be used intransitively with a passive meaning ("Features, Syntax, and Categories in the Latin Perfect", by David Embick, p. 194).

Embick gives the following example:

(11) Ab amīcīs hortā-rē-tur.
by friends urge-IMPERF.SUBJ-PASS.3SG
‘He was urged by friends.’ (subjunctive)
(Varro in Prisc., GL II 387,2)

Fraenkel 1964 cites Gellius as giving utor, vereor, hortor, and consolor as examples of "verba communia"; and testor and interpretor are also mentioned similarly.

To clarify a point that was brought up in the comments, I'm not claiming that this passive construction was productive or applicable to all transitive deponent verbs: I haven't seen a source that makes that claim. Embick says "some deponents like hortor may also appear in passive syntax".

Wright ("Passive morphology in Late Latin") mentions the alternative of analyzing the verbs used in this passive construction as forms of non-deponent alternatives to deponent verbs (e.g. horto) (p. 71, 77), which would imply that this passive construction should only exist for deponent verbs that had alternative non-deponent active forms. If you consider deponent hortor and non-deponent horto to be separate verbs, then hortārētur in Embick's example could be analyzed as only being a form of the second.

That said, I also feel like English break (tr.) and break (intr.) could be analyzed as separate verbs, so I'm not sure whether this disqualifies "verba communia" as an example of something similar in Latin.

Apparently, a similar ambiguity applies to the perfect participle of many deponent verbs, according to this Reddit post that quotes Gildersleeve 167 n. 2.

Improve this answer
4
  • What is Embick's example supposed to prove? See Pinzin (2017: 32; cf. ref. above): "To deduce from this single example that deponents in general can be passivized is not adequate. Even if Priscianus quoted Varro faithfully, this would only prove that hortor ‘I exhort’, during the early I cent. BCE, could be passivized, not that all deponents could (...) Even taking into consideration this single occurrence, it is a fact that the overwhelming majority of deponent verbs do not present passive occurrences. This fact must be explained and derived from a coherent analysis of these verbs".
    – Mitomino
    Mar 31, 2019 at 1:19
  • By the way, your well-formed example from Embick (2000) recalls an (allegedly, in my view) ill-formed example of Ablative Absolute pointed out by my Italian colleague Renato Oniga: *hortatis militibus: cf. books.google.es/…
    – Mitomino
    Mar 31, 2019 at 1:28
  • @Mitomino: Although Embick's argument about the ability to passivize "hortor" is debatable, I was mainly interested in the example of the same form being used for both meanings. I've edited to add a link to and summary of a separate analysis that I found; I'd be interested in hearing whether you think the interpretation mentioned in Wright's paper is more correct. I also added some other examples of verbs that are said to be "verba communia": I'm confused about why Pinzin talks about deducing something from a "single example", since there seem to be at least some other similar examples.
    – Asteroides
    Mar 31, 2019 at 1:59
  • Yes, Sumelic, I think your qualification above based on Wright is correct: "If you consider deponent hortor and non-deponent horto to be separate verbs, then hortārētur in Embick's example could be analyzed as only being a form of the second". In contrast, I do not "feel" that break (tr.) and break (intr.) can be analyzed as separate verbs.
    – Mitomino
    Mar 31, 2019 at 2:55

Your Answer

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

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