In “word x is case y”, what dictates the verb’s number?

Question

In the languages I am familiar with where verbs are pluralised, if you have more than one subject, the verb has to agree in number to however many subjects there are. However, in Lingua latīna per sē illūstrāta : Pars 1 : Exercitia latīna I, this assumption apparently is not correct.

The exercise

In this exercise, you are supposed to fill in the case and number for the words in question. In questions 7–11 and 14, there are more than one word being asked for, but still the verb chosen by Ørberg is est.

Just to mention it: I have checked the answer against the answer key, and they are all correct. Also, had I known I would post this online, I would have written nicer earlier today.

Questions

My first question is simply why is it est and not sunt? In English I would have expected are, e.g. ‘Pāstōrem and ovem are accusative singular’; in German, I would have expected ‘Pāstōrem und ovem sind akkusativus singularis.’

My second question therefore is: Is the construction chosen by Ørberg actually good Latin? I would love to learn whether any of the ancient grammarians are using similar phrasings, or for that matter: not using similar phrasings. How do the ancient grammarians themselves talk about words like this when describing them grammatically? Do they use plural verbs when talking about more than one word, or do they do as Ørberg and think of them as a single case and use a singular verb?

What I would have expected

There are two possibilities here, as I see it:

Option 1: Thinking of it as a singular case with x parts

‘The case with pāstōrem and ovem is accusative singular.’
Cāsus cum [verbīs] pāstōrem et ovem accūsātīvus singulāris est. Please correct my Latin as need be.

Option 2: Thinking of it as x words

‘The words pāstōrem and ovem are accusative singular.’
Verba pāstōrem et ovem accūsātīvus singulāris sunt.

Answer 1

I didn't find too many examples, but based on what I've seen, I'd expect "pāstōrem et ovem accūsātīvī singulārēs sunt" (or "accūsātīvī sunt singulārēs") to be a possible wording. A plural verb is possible but may not be the only option (see next section).

It seems that when the name of a case or number (or both) are used in a context like "'ratio' est nominativus singularis", it acts as a predicative noun. So it doesn't agree in gender with the subject, but it could be pluralized if you have a plural subject and copula. Servius seems to like sandwiching the copula between the two parts of the predicate in sentences like these.

E.g. Servius's commentary on Virgil contains the sentence

"aonie aganippe" nominativi sunt singulares.

(In Vergilii Bucolicon Librum 10.12.1, PHI)

Maybe it is short for "casûs nominativi sunt singulares", since we find the wording "nam 'rubra crista' longae sunt ultimae, quia ablativi sunt casus" at In Vergilii Aeneidos Libros 9.49.3. The commentary of Pomponius Porphyrio on Horace uses the wording "Non nominativi casus sunt "sumptuosa" et "hostia".

Grammatical number of verbs

Either a singular or plural verb is possible when the subject is two singular nouns joined by "et". The plural because the two nouns together are plural; the singular because the verb may agree with the nearest noun, which is singular.

In AG, as well as in Latin and in many unrelated languages, whenever the agreement is triggered by two or more coordinated phrases, two different constructions are allowed: either the agreement can be controlled by the coordinated phrase as a whole, through a synthetic resolution of the conjoined phrase, or it can be triggered by just one of the coordinated words. Thus, e.g., two coordinated singular nouns can license either plural or singular agreement over the syntactic elements that they control.

("Subject-Verb Agreement with Coordinated Subjects in Ancient Greek: A Treebank-Based Study", Francesco Mambrini and Marco Passarotti, Journal of Greek Linguistics published online 01 Jan 2016)

Plural might be more common, but singular is not incorrect. I don't actually know the intricacies of when you'll generally see singular vs. plural here. I think this topic is covered by Harm Pinkster's Oxford Latin Syntax, based on the answer to this related question: The use of et...et and the following grammar

Mambrini and Passarotti cite the following paper which probably has more specific information on Latin:

• Johnson, Cynthia A. 2013. Multiple antecedent agreement: A comparative study of Greek and Latin. In: Stephanie W. Jamison, H. Craig Melchert and Brent Vine (eds.), Proceedings of the 24th Annual UCLA Indo-European Conference, 67–86. Bremen: Hempe.

Johnson's paper seems to be available here.

There is some more discussion in Johnson's dissertation from 2014 which I haven't finished reading yet: "Deconstructing and Reconstructing Semantic Agreement: A Case Study of Multiple Antecedent Agreement in Indo-European", Cynthia A. Johnson:

For example, Menge and Thierfelder (1953: 6) describe the rules as follows:² when the controllers are animate, the target is masculine plural via Resolution; if the controllers are inanimate, then Nearest Antecedent Agreement is usually (“gewöhnlich”) the preferred strategy, with Resolution to the neuter plural more rare (“seltener”). When the controllers differ in animacy, the target is plural and the gender can be masculine, feminine, or neuter. This is the same story as Kühner and Stegmann (1962), according to Hock (2007).

Some grammarians ignore the question of which strategy is more frequent: Hoffman and Szantyr (1965: 444–5) address only multiple antecedent agreement with inanimate controllers, where either Resolution to the neuter plural or Nearest Antecedent Agreement can occur, with no discussion of which strategy is more likely. Roby (1896: 26), on the other hand, discusses only Nearest Antecedent Agreement as a strategy for any multiple antecedent context. Finally, Lindsay (1936: 4–5) finds that “we have often a Sing. Verb with two subjects” and “occasionally. . . a Plural Verb with ‘A cum B”’ (emphasis mine).³

(page 61)