(There is also ad + gerundive, which also seems to lack any nominal that could be a head: Venit ad vincendum.)

For example, the ambiguity of “ab urbe condita” (‘since the foundation of the city’ vs. ‘from the city founded’) is not merely semantic in Ancient Greek: “he polis hidrumene” (AUC/dominant participle construction) vs. “he hidrumene polis” (attributive reading: the city which has been founded). So the distinction is not merely semantic in Ancient Greek, right?

Hmm interesting.

My Greek is not good enough; you may very well be right.

I suppose placing the participle between article and substantive makes it look more attributive.

Knowing Greek, though, I would expect writes to be quite flexible.

But it may very show a distinction.

Yes, as I told you, it is not obvious which the head is in an AA (some people like you and many others say it is the nominal subject. Other authors like Haug & Nikitina say that the head is the predicate).

I have the article open in a tab...

I'll read it cras hehe.

Ok! Have a good reading! ;-)

@Mitomino Ha, I'll let you know when I have finished it.

I saw a lapidatum example that seemed akin to the kind of gerundives without substantives I mentioned above.

Yes, this kind of evidence is used by the authors to claim that the head in an AA is the predicate.

I would ask what their definition of a head is, but it would probably be more efficient if I read (a section of) the article...

Yes, if you don't mind, reading it first would probably be more efficient. And please feel free to raise any other doubts you may have on this. I find the syntax and semantics of AUC, AA and gerundive constructions fascinating! As I say at the end of my answer to Ben's question, all of them show a common basic syntactic pattern.

@Mitomino Would you call a non-dominant + substantive a "gerundive construction"?

I.e. tibi pecuniam servandam do, where it is assumed that the meaning is non-dominant.

These gerundive constructions are precisely the ones that I find more difficult to analyze. In many cases I find them ambiguous between a dominant and a non-dominant reading. The fact that Joonas does not see a real difference between the readings makes me think that perhaps I'm wrong in seeing this ambiguity (perhaps it's a matter of my Romance eyes...).

Cf. the dominance reading with the verb "dare" plus gerundive in this example: Populus Romanus consuli potius Crasso quam privato Africano bellum gerendum dedit (Cic. Phil. 11.18)

@Mitomino My non-Romance eyes also see a difference. The fact that the gerundive can be omitted in non-dominant constructions (I'll follow you and call them all gerundive constructions) but not in dominant ones does point to some kind of semantic distinction, to me.

@Mitomino OK to my Indo-European eyes, this one is smack in the middle between dominant and non-dominant. I agree that such a sentence can have the exact same type of ambiguity or 'middle-ness' in other Indo-European languages, including English and Dutch.

For example, cf. the French translation of "bellum gerendum dare" given by Lavency in his excellent textbook (VSVS): ‘Le peuple romain confia à Crassus la conduite de la guerre’.

When you say, in English, "they gave Crassus a war to fight", this can mean roughly the same thing as "they have him the opportunity/mandate to fight a war", but also "they handed him this war, in order that he might fight it".

@Mitomino Yes. I remember this translation from one of your posts, and the construction is often mentioned in textbooks.

Yes, it's in these cases where it's difficult to draw the line (if there is one to draw in Latin, as Joonas would probably say).

In any case, it is not clear to me that all is a matter of semantics, i.e. of interpretation. Same in AUC constructions. Although there are no visible morphosyntactic differences in Latin, I think that the dominant reading of the participle in "ab urbe condita" (since the foundation of the city) and the attributive reading (from the founded city) is not just a matter of interpretation/semantics.

@Mitomino Hmm why not?

Semantics are often intricately linked with morphosyntaxis.

They're not independent.

In a way, the boundary is not hard.

Yes, it's a complex issue. Some authors have claimed that in the dominant reading of the participle in "ab urbe condita", there is a "small clause" (cf. en.wikipedia.org/wiki/Small_clause ) involved in its syntax: [SC urbe condita]. This sort of constituent is not postulated when dealing with the syntax of the attributive reading. "Small clauses" have been argued to be appropriate for resultative constructions like They ran [SC the pavement thin].

In fact, some authors have translated this SC analysis (which is also typically used for so-called "unselected object constructions": "The snakes cough [SC the egg shells out])" to Latin. So an attested example like "Serpentes putamina extussiunt" has been analyzed as "Serpentes tussiunt [putamina ex-]".

with subsequent affixation of the prefix "ex-" (the predicate of the Small Clause Result) to the verb "tussiunt". Cf. the ill-formedness of "*The snakes cough the egg shells" & "*Serpentes putamina tussiunt".

There is also debate in the literature as to whether Ablative Absolutes are nominal phrases (NP) or sentential constituents. In the latter case the use of Small Clauses for AA has been claimed by Oniga (2014), i.a. global.oup.com/academic/product/…;

@Mitomino Yet another term.

Any reason why it is called a clause, rather than just a praedicate / praedicative?

From you examples, it looks like another way of analysing separable verbs c.q. verbs with 'special' adverbs?

Yet more complex analyses.

In the wikipedia link above you'll see some evidence that has been put forward in the literature for postulating the term "small clause" in syntax.

Perhaps I'll read it later hehe.

I know some Anglo-Sexon linguists consider any verb + verbal arguments a clause, which I would call a praedicate.

In the syntactic literature there are two competing analyses: the so-called "Small Clause" based approach (cf. Teun Hoekstra's works, i.a.) and the so-called "Complex Predicate" approach (cf. Ad Neeleman's work). Given an example like "The snakes cough the egg shells out", "the egg shells" is an argument of "out" in the SC analysis, whereas it is an argument of "cough out" in the complex predicate analysis.

If we apply it to the AUC construction, it seems that the Small Clause analysis is more appropriate: ab [SC urbe condita]. Crucially, for the SC analysis, "urbe" is not an argument of the preposition "ab" but only of the predicate "condita".

@Mitomino Makes sense.

Both Dutch?

@Mitomino Perhaps those two analyses are not mutually exclusive?

You'll see that Nikitina and Haug claim that in "ab urbe condita", there is a non-finite clause containing "urbe" and "condita", which turns out to be nominalized, i.e. it has the same distribution as an NP. So in a sense what they propose is a sort of a nominalization of a Small Clause. That accounts for the sentential & nominal properties of AUC constructions.

I would say, in one way, urbe is the object of ab, following normal patterns; in another, it is urbe condita together which is the object of ab.

Yes, Hoekstra and Neeleman are Dutch. The former, who died in 1998, was a very important linguist in the Netherlands.

On Dutch Radio 4, the classical station, there is this game where you hear the first few seconds of a piece, and callers are to guess which piece it is. Mijnheer Neeleman always called in when I listened, and he was often correct.

His first name was never mentioned...

@Mitomino Do they propose the same analysis for ablativus absolutus me consule ("when I was consul"), and a me consule ("since my consulate")?

"Me consule" only involves a Small Clause ([SC me consule]), whereas "a me consule" would involve an additional nominalization process of this SC.

Hmm.

> it has the same distribution as an NP

I'm not sure what distribution means except occurrence, so I'll make sure to pay attention when finishing the article.

Here is their analysis of an AUC:

Well, Cerberus, I have to leave. It's been a pleasure to chat with you. Have a good reading of Nikitina & Haug's (2016) paper!

It seems to be a bit of a mixture between syntactic and semantic structure?

@Mitomino Indeed!

Until next time.

@Mitomino Those are indeed good examples! My wondering whether the degree of dominance can make a semantic difference was due to not having any good examples in mind and my dislike of unnecessary classification.

But of course some classification is very useful. I just don't find it fruitful to insist that things always fall clearly in neat classes, especially considering the organic origin of language. Fortunately such insistence doesn't really appear on our site.

@Mitomino Well predicted! :)

Perhaps one way to see it is than the full space of semantics is too much for most if not all languages and peoples to handle. They will work in different projections of that space, conflating some things but not others to keep it to a manageable size. Different languages and peoples and people project in different directions and so find different distinctions meaningful.

I like to understand some matters of Latin grammar through an innate Roman indifference to some things we might consider separate. This mindset might not help everyone but it does help me.

@Cerberus That is an interesting question. The plural of nullus is certainly allowed: Nullae enim lites neque controversiae sunt, quae cogant homines sicut in foro non bonos oratores, item in theatro actores malos perpeti. (Cic. De Oratore 1,26) But in your case I would say nullum errorem sounds better. (Although in English, German, and probably Dutch we would rather use the plural? But here too both is allowed.)

@Joonas: In my question to Cerberus, above, beginning: "Caeser could easily have used..."; followed by a valuable contribution from Mitomino, to Q: latin.stackexchange.com/q/15055/1982, how can this material be linked back to the original Q., on the mainsite, where it needs to be? Linking something to CHAT brings the visitor here, where he may have to search for something that was said 6-months/ a year ago. This is why I don't like moving discussions to CHAT.

@JoonasIlmavirta I absolutely agree.

@SebastianKoppehel Agreed!

@tony Yes, there is a disadvantage.

It is the philosophy of Stack Exchange to discourage discussions in comments, though.

@tony You could copy the link to the starting point of the discussion in chat, then add that link in a comment on the original question.

@tony As @Cerberus says, comments are second class content on this site and are deleted with a low threshold. If you see anything important in a comment, salvage it by editing it into a relevant post or writing a new one.

Chat is indeed more permanent and you can link to a specific message or even select a conversation to keep. The suggestion by @Adam is a good one.

The rule of thumb to comments is that they are temporary and only meant for suggesting improvements to the post.