In English, it is fairly common to write/say such sentences as the following:

What is the possessive case of she?

Should I use who or whom after man?

What is the past participle of run?

These kinds of questions quickly get thorny when I attempt to translate them into Latin. Here are some off-the-top-of-my-head translation options:

  1. Precede with verbum/vocabulum: Quid est casus genitivus verbi ea?
  2. Treat as indeclinable: Debeo qui aut quem post vir uti?
  3. Rework the sentence: Correre quale habet participium praeteritum?

Can anyone offer general guidance for speaking about inflected words outside their original context?


There are several ways of doing this, as you've seen. Obviously, the basic goal is to 'mark' the word in question in such a way that it can easily be understood to stand, in spite of its inflected form, outside the grammar of the sentence, without having to rely on (or rely exclusively on) a 'cheat' such as quotation marks. The use of something like verbum or vocabulum, or Joonas's vox, would do this.

But I want to expand on the option that Joel Derfner has mentioned: prefacing the word – or words – in question with the appropriate case form of the neuter singular demonstrative illud. Although Joel tentatively alluded to this in the context of contemporary spoken Latin, it's also quite common in Classical Latin. (That said, it doesn't work particularly well for your examples, because it assumes some previous or well-known instance of the words that you want to talk about.) By adding illud, you 'mark' what follows and effectively treat it as a neuter singular noun, even if nothing about it is neuter or singular or even a noun. Another neuter singular pronoun (e.g., id or hoc) would also work, but illud is what I'm most familiar with. Although modern texts will generaly print quotation marks around the word or words that are being marked in this way (unless they're Greek), it's the illud, of course, that does the actual work in the Latin text.

One example occurs in Cicero, Epistulae ad Atticum 6.1.11:

in Pompeio te spem omnem oti ponere non miror; ita res est, removendumque censeo illud 'dissimulantem'.

...I recommend that 'dissimulantem' must be removed.

Without the illud (and the helpful quotation marks that the editor has included), the clause could be read to mean only 'I recommend that the dissembling person must be removed.' So, although some grammatical sense could be made of this example, it wouldn't be a logical sense in context.

Next, there's Quintilian, Institutio oratoria 9.3.64, where the illud marks the otherwise-problematic finite verb edico. And since the illud effectively turns the finite verb form into a neuter nominative singular noun, it can itself be the subject of another finite verb form, convenit. In this example, without the illud, the sentence would make no grammatical sense at all as is.

quamvis enim pars [bello] posterior participio insistat, utrique convenit illud 'edico'.

...'edico' goes with both.

Earlier, in Inst. orat. 6.3.62, illud marks the finite verb form petis, and the neuter nominative singular noun that is thus formed is then not only used as the subject of another finite verb form, est, but also modified by a neuter singular adjective, ambiguum. Once again, the sentence would make no sense without the illud.

iungitur amphiboliae similitudo, ut a L. Galba, qui pilam neglegenter petenti 'sic' inquit 'petis tamquam Caesaris candidatus'. nam illud 'petis' ambiguum est, securitas similis.

...For 'petis' is ambiguous....

In Aulus Gellius, Noctes Atticae 18.9.10, illud is used to treat a Greek neuter plural noun as a Latin neuter singular noun:

sed etiam ipsum illud ἔπη, quod significat verba aut versus, non aliunde esse dictum tradunt quam ἀπὸ τοῦ ἕπεσθαι καὶ τοῦ εἰπεῖν. eadem ergo ratione antiqui nostri narrationes sermonesque 'insectiones' appellitauerunt.

But even ἔπη itself....

And illud can even be used in this way, not just with individual words, but with phrases or clauses, as in Cicero, Epistulae ad familiares 9.22.1:

quid est enim illud
'quae mulier una'
quid, inquam, est
'usurpat duplex cubile'?

For what is 'quae mulier una'? What, I say, is 'usurpat duplex cubile'?

In all the preceding examples, the illud – and therefore the word or words after it that are effectively treated as a neuter singular noun for grammatical purposes – is in either the nominative or accusative. This makes sense, since it's in these forms that the demonstrative is most recognizable as neuter, as brianpck has noted in a comment. Still, the same technique can also be used when the grammar of the sentence requires some other case.

In fact, the example that I'm most familiar with – the first example that really made me sit up and take notice of this technique – is Pliny the Younger, Epistulae 3.16.13, where the illud phrase is in the ablative case (ablative of comparison):

videnturne haec tibi maiora illo 'Paete, non dolet', ad quod per haec perventum est?

Do these words seem greater than 'Paete, non dolet'...?

In this example, the words after the form of illud are actually a famous quotation. So, although I've been omitting any explicit translation of the form of illud for all previous examples, treating it as just a marker, I could follow the practice of many translators here and in other similar examples, and translate it as 'that famous saying,' 'that well-known utterance,' or the like.

In Varro, De lingua latina 8.10, we find another ablative (of instrument/means), where the form of illud is strengthened by a form of idem:

sic quod dicimus in loquendo 'consul fuit Tullius et Antonius', eodem illo 'et' omnis binos consules colligare possumus, vel dicam amplius, omnia nomina, atque adeo etiam omnia verba, cum fulmentum ex una syllaba illud 'et' maneat unum.

...we can join together every set of two consuls by the same 'et'....

Then in Aulus Gellius, Noctes Atticae 1.21.4, we find a genitive:

sed enim cum Favorino Hygini commentarium legissem atque ei statim displicita esset insolentia et insuavitas illius 'sensu torquebit amaro', risit et: 'Iovem lapidem,' inquit 'quod sanctissimum iusiurandum habitum est, paratus ego iurare sum Vergilium hoc numquam scripsisse, sed Hyginum ego verum dicere arbitror....'

...the unusualness and unpleasantness of 'sensu torquebit amaro'...

And finally, in C. Caesius Bassus, De metris, we find a dative (with the adjective par):

hoc ipsum imminuendo paulatim tria haec metra quae rettuli manifestiora faciam: 'tu ne quaesieris quem mihi quem tibi': hoc par est illi 'Maecenas atavis edite regibus'.

...this is the same (metrically) as [that famous verse] 'Maecenas atavis edite regibus.'

  • Excellent examples! Do you happen to know if "illud" is only used in the nominative and accusative, as in your examples? It seems like the other cases--which aren't recognizably neuter--would be less useful for disambiguation. – brianpck Feb 15 at 13:38
  • @brianpck, It didn't register with me that all my examples were either nominative or accusative, so thanks for pointing that out. Since other cases are also found, I'll add some more examples this evening. – cnread Feb 15 at 20:39
  • I've changed this to the accepted answer: it's really thorough, has great examples, and isn't restricted to medieval Latin. – brianpck Feb 16 at 3:55

In Medieval Latin, the word "ly" could be paired with the relevant term, which could then be treated as an indeclinable term.

See for example the passages from St. Thomas Aquinas written in the 1250's:

Quia ly se potest esse ablativi casus, et tunc simpliciter vera est: et est sensus: genuit alterum se, idest alterum a se. Vel potest esse accusativi casus; et tunc vel facit simplicem relationem . . . (Super Sent., lib. 1 d. 4 q. 2 a. 2 expos.)

Si enim cum dicitur, Deus genuit Deum, qui est Deus pater, ly pater construitur appositive ad ly Deus, locutio falsa est: quia tunc ly Deus restringetur ad standum pro persona patris . . . (Super Sent., lib. 1 d. 4 q. 1 a. 3 ad 4)

In the first example "ly se" allows "se" to function as the subject of the sentence even though it is in the accusative or ablative. "Ly" seems to act a lot like mention marks in English. For instance, I would be inclined to translate the first example: "Since "se" can be in the ablative case . . . "

  • Fascinating! I wonder whether this is related to the use of ille that I discussed in my answer. – Joel Derfner Apr 15 '16 at 5:07
  • 1
    This is really cool--any idea how common this is? I've read a lot of Aquinas (though no, not the Commentary on the Sentences) and never came across this. – brianpck Apr 15 '16 at 13:17
  • @brianpck It seems to have been generally adopted by the scholastics; see this book. – Nathaniel is protesting Apr 15 '16 at 17:24
  • I got my laugh for the day: "Tantum cuiuslibet hominis praeter Sortem quilibet non asinum c et alterum c ipsiusmet hominis nigrum contingenter incipit esse" – brianpck Apr 15 '16 at 17:38
  • @brianpck Using this search engine reveals that the Commentary on the Sentences contains the most instances, followed by the Lectura on the Gospel of John. To see the breakdown of instances across works, enter "ly" into the search box and click "works." Note that some works by authors besides Aquinas are also included in the search. – SAG Apr 19 '16 at 4:28

A few times in spoken-Latin circles I've come across this kind of locution:

In sententiá "Ódió virum quí canem meum siccávit," opportébatne vóce "quem" pró illó "quí" útí?

However, my experience is of limited enough scope that you probably shouldn't take this answer as definitive.

  • So the general principle would be to add either vox or a kind of pseudo-article ille before the inflected word? – brianpck Apr 14 '16 at 13:56
  • I'm not sure it's an either/or thing (it might be), but yeah. (Out of curiosity: why a pseudo-article rather than just an article? We have the same thing—"In 'I hate the man who assassinated my dog,' should I replace the 'who' with 'whom'?"—not the most usual way of phrasing it, perhaps, but certainly a possibility.) – Joel Derfner Apr 14 '16 at 14:17

The most important aspect here is clarity, and your first option is best for that.

I recommend against treating declined words out of context as indeclinable (casus generalis), unless you mark them with quotes of some kind. With pronouns and other common words the reader might not parse them the way you intend. Reading qui and quem in your second example sentence as declined pronouns makes the sentence a mess.

Quotes around the inflected word (or any other phrase for that matter) make the indeclinable status clear. To clarify the syntactical role of the indeclinable, I would add a suitable noun in front of it. That is, write Debes voce "qui" uti or In locutione "…" accusativo utimur. If the case of the auxiliary word would be nominative or accusative or it would come with a preposition, leaving it out does not confuse too much. In other situation the risk of confusion is significant.

I have been taught to use vox for "word" and verbum only for "verb", but I do not always do so. Unless you need to discuss pronunciation, it might be better to reserve verbum for verbs and use vox or vocabulum for words in general.

Suggested translations of your examples:

  1. What is the possessive case of she?
    Qualis est casus genetivus vocis "ea"?
  2. Should I use who or whom after man?
    Debeone adhibere vocem "qui" aut "quem" post vocem "vir"?
    Utro uti debeo post vocem "vir": "qui" aut "quem"?
  3. What is the past participle of run?
    Quale est participium praeteritum verbi "currere"?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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