Elamur: **Mova** · [Parse](https://mova.elamur.ai/free/en/parse/) · [Dictionary](https://md.dictionary.elamur.ai/free/en/) · [Translate](https://md.translate.elamur.ai/free/en/) · [Research](https://md.research.elamur.ai/free/en/) · [MCP](https://mcp.elamur.ai/) · [Groupants](https://md.groupants.elamur.ai/free/en/)

[Mova](https://md.mova.elamur.ai/free/en/) · [this page in HTML](https://mova.elamur.ai/free/en/docs/languages/la/08-prepositions-particles/)

[home](https://md.mova.elamur.ai/free/en/) / [docs](https://md.mova.elamur.ai/free/en/docs/) / [languages](https://mova.elamur.ai/free/en/docs/languages/) / [la](https://md.mova.elamur.ai/free/en/docs/languages/la/) / 8. Prepositions, conjunctions and particles

# 8. Prepositions, conjunctions and particles

Latin marks most relations between words by case endings ([chapter 7](https://md.mova.elamur.ai/free/en/docs/languages/la/07-cases/)). Prepositions add precision to the cases, above all to the accusative and the ablative, and grow in number and weight from the classical language to the Romance languages, as the endings wear away. This chapter gives the prepositions and the case each governs, the conjunctions, and the small particles that carry negation, questions, emphasis and the links between sentences. The subordinating conjunctions and the moods they take are in [chapter 10](https://md.mova.elamur.ai/free/en/docs/languages/la/10-subordination/).

## Contents

1. Prepositions with the accusative
2. Prepositions with the ablative
3. *In, sub, super, subter*: accusative or ablative
4. Place and direction without a preposition
5. Postpositions and the position of the preposition
6. Coordinating conjunctions and enclitics
7. Negation
8. Questions
9. Particles of emphasis and discourse
10. Prepositions in late and medieval Latin
11. How Mova handles it

---

## 1. Prepositions with the accusative

Most prepositions take the accusative. Their basic meanings are spatial: motion towards, position beside or around, crossing (Bennett 1918 §141).

| preposition | meanings |
| --- | --- |
| *ad* | to, towards; at, near; for (purpose: *ad pugnandum*) |
| *ante* | before (place, time) |
| *apud* | at the house of, among, in the works of (*apud Platonem*) |
| *circa, circum* | around |
| *contrā* | against, opposite |
| *ergā* | towards (of feelings) |
| *extrā* | outside |
| *inter* | between, among, during |
| *intrā* | within |
| *iuxtā* | next to; according to (late) |
| *ob* | on account of; before (*ob oculos*) |
| *per* | through; by means of (a person); during |
| *post* | behind, after |
| *praeter* | past, beyond, except |
| *prope* | near |
| *propter* | on account of; near |
| *secundum* | along; following; according to |
| *suprā* | above |
| *trāns* | across |
| *ultrā* | beyond |

*Per* with a person names the agent through whom a thing is done, as against *ā/ab* with the ablative for the agent of a passive: *per lēgātōs* "through the envoys".

## 2. Prepositions with the ablative

| preposition | meanings |
| --- | --- |
| *ā, ab, abs* | from, away from; by (the agent of a passive: *ā patre amātur*) |
| *cum* | with (company, manner) |
| *dē* | down from, from; about, concerning |
| *ē, ex* | out of, from; since; in accordance with |
| *prae* | before, in front of; in comparison with; because of (with negatives) |
| *prō* | in front of; on behalf of; instead of; in proportion to |
| *sine* | without |
| *tenus* (after its noun) | as far as |

*Ā* and *ē* stand before consonants, *ab* and *ex* before vowels and *h* (and often before consonants too); *abs* survives in *abs tē*. *Cum* is written after a personal or relative pronoun as one word: *mēcum, tēcum, nōbīscum, quōcum* (Bennett 1918 §142).

## 3. *In, sub, super, subter*: accusative or ablative

Four prepositions take the accusative for motion towards and the ablative for rest (Bennett 1918 §143):

|  | accusative (motion, direction) | ablative (rest, position) |
| --- | --- | --- |
| *in* | into, onto, against, towards: *in urbem īre* | in, on: *in urbe esse* |
| *sub* | under (motion), up to, just before (time): *sub montem* | under, at the foot of: *sub monte* |
| *super* | over, above, beyond: *super caput* | about, concerning: *hāc super rē* (mostly poetry) |
| *subter* | beneath (motion) | beneath (rest, poetry) |

> In principio creavit Deus cælum et terram. (Vulgate, Gen. 1:1) — *in* + ablative, a point in time. Terra autem erat inanis et vacua, et tenebræ erant super faciem abyssi. (Gen. 1:2) — *super* + accusative, position over.

In the classical texts of the CIRCSE treebank (Seneca's tragedies and Tacitus' *Germania*, UD 2.18 training set, counted 2026-10-22), *in* takes the accusative 84 times and the ablative 65 times in 451 prepositions.

## 4. Place and direction without a preposition

Names of towns and small islands, *domus* and *rūs* take no preposition: the accusative marks direction, the ablative separation, the locative position ([chapter 7](https://md.mova.elamur.ai/free/en/docs/languages/la/07-cases/), §6).

> *Rōmam īre* "to go to Rome", *Rōmā venīre* "to come from Rome", *Rōmae esse* "to be in Rome"; *domum, domō, domī*; *rūs, rūre, rūrī*.

## 5. Postpositions and the position of the preposition

A preposition normally stands before its noun, but:

- *tenus* and, in some phrases, *versus* follow it: *Taurō tenus* "as far as the Taurus", *Rōmam versus* "towards Rome";
- *cum* is attached after pronouns (§2);
- in poetry and artistic prose a preposition often stands between an adjective and its noun: *magnā cum laude*, *quā dē causā*, *hīs dē rēbus*;
- a monosyllabic preposition is often placed after a relative or demonstrative: *quem contrā*.

## 6. Coordinating conjunctions and enclitics

|  | conjunctions |
| --- | --- |
| and | *et* (neutral), *-que* (enclitic, binds closely: *senātus populusque*), *atque, ac* (adds weight; *ac* before consonants), *et … et* "both … and" |
| or | *aut* (exclusive alternatives), *vel* (free choice, "or if you like"), *-ve* (enclitic), *sīve … sīve* |
| nor | *neque, nec*; *neque … neque* "neither … nor"; *et nōn* only for a contrast on one word |
| but | *sed, vērum* (correcting), *at* (objecting), *autem* (second word: "but, now"), *tamen* "nevertheless" |
| for, so | *nam, enim* (second word) "for"; *itaque* "and so", *igitur* (usually second word), *ergō* "therefore" |

*-que* is written after the first word of its member, which in a prepositional phrase is the noun, not the preposition: *in forōque* or *inque forō*. A Latin sentence rarely begins without a link to the one before: *autem, enim, igitur, vērō, nam, itaque* or a relative pronoun at its head (the *relative connection*: *quae cum ita sint* "since these things are so"). Chapter [9](https://md.mova.elamur.ai/free/en/docs/languages/la/09-simple-clause/) gives its place in the clause (Bennett 1918 §§341–346).

## 7. Negation

| word | negates |
| --- | --- |
| *nōn* | a statement or a word: *nōn venit* |
| *haud* | a single word, mostly an adjective or adverb: *haud dubiē* "without doubt" |
| *nē* | a wish, a command, a purpose, a fear ([chapter 10](https://md.mova.elamur.ai/free/en/docs/languages/la/10-subordination/)): *nē timueris* "don't be afraid" |
| *nē … quidem* | not even (around the stressed word): *nē tum quidem* |
| *nēmō, nihil, nūllus, numquam, nusquam, neque* | nobody, nothing, no, never, nowhere, and not |

**Two negatives make a positive.** *Nōn nūllī* is "some", *nōn nihil* "something", *nēmō nōn* "everybody", *nōn nēmō* "somebody". Unlike French ([fr chapter 8](https://md.mova.elamur.ai/free/en/docs/languages/fr/27-negation/)) and much of Romance, classical Latin has no negative concord: a second negative cancels the first. A negative followed by *neque … neque* or *nē … quidem* stays negative, an exception.

## 8. Questions

| particle | asks | example |
| --- | --- | --- |
| *-ne* (enclitic, on the questioned word, usually first) | a neutral yes/no question | *Vīdistīne?* "Did you see?" |
| *nōnne* | expects "yes" | *Nōnne vīdistī?* "Surely you saw?" |
| *num* | expects "no" | *Num vīdistī?* "You didn't see, did you?"; in indirect questions: "whether" |
| *utrum … an* | either … or | *Utrum vīdistī an nōn?* |
| *an* | or (in a double question); "or is it that …?" |  |

Latin has no word for "yes" or "no". An answer repeats the verb (*Vīdī.* "Yes." *Nōn vīdī.* "No."), or uses *ita, sānē, etiam, vērō* for yes and *nōn, minimē* for no. The interrogative pronouns and adverbs are in [chapter 4](https://md.mova.elamur.ai/free/en/docs/languages/la/04-pronouns/) (Bennett 1918 §162).

## 9. Particles of emphasis and discourse

The short words that colour a sentence: *quidem* "indeed, at least" (after the stressed word); *vērō* "truly, but"; *equidem* "I for my part"; *sānē* "certainly"; *tamen* "yet"; *quoque* "also" (after its word: *tū quoque*); *etiam* "also, even; yes"; *iam* "now, already"; *modo* "only, just now"; *dēnique* "finally". Many of them are second-position words: they follow the first word of their clause (Bennett 1918 §347).

## 10. Prepositions in late and medieval Latin

From the imperial period on, prepositions take over work that the cases did. The case after a preposition weakens with them, as the case endings merge in speech ([chapter 13](https://md.mova.elamur.ai/free/en/docs/languages/la/13-late-latin/)):

- *dē* + ablative replaces the partitive genitive (*ūnus dē discipulīs*) and gives the French *de*;
- *ad* + accusative replaces the dative (*dīxit ad eum* in the Vulgate beside *dīxit eī*);
- *per* and *prō* widen in meaning; *iuxtā* takes the sense "according to";
- new compound prepositions appear: *dē post, dē intus, ab ante* (source of French *avant*);
- the accusative and the ablative are confused after prepositions in the less learned documents.

On our data (UD 2.18 training sets, counted 2026-10-22), the CIRCSE classical texts never have *ad* with an ablative noun (51 times with an accusative). In the LLCT treebank, eighth- and ninth-century charters from Lombard Tuscany, *ad* governs an ablative-marked noun 474 times against 1 606 accusatives, and *cum* an accusative 166 times against 491 ablatives. The same treebank shows the formulae of the charters: *In Dei nomine regnante domno nostro Carulo rege Francorum et Langobardorum …* "In the name of God, in the reign of our lord Charles, king of the Franks and the Lombards …".

## 11. How Mova handles it

**In the UD trees** (UD Latin LLCT and CIRCSE):

- a preposition is `ADP` attached to its noun as `case`, and the noun takes the relation of the whole phrase (`obl`, `nmod`). The noun's `Case` feature carries the accusative or the ablative, and it is the only place where the two uses of *in* (§3) are told apart;
- the enclitic *-que* is split off as a token of its own, `CCONJ` attached as `cc` (319 times in the CIRCSE training set). In the LLCT charters the medieval spelling *que* for *quae* is far more often the relative pronoun (`PRON`, lemma *qui*: 285 times) than the enclitic (66), so a tokenizer must not split every final *-que*;
- *nōn* is `PART` with `Polarity=Neg`, attached as `advmod:neg` (509 times in the LLCT training set, 2026-10-22).

**In Mova.** Latin goes through Mova's English parser after `en xpos-fill` (LLCT dev LAS 87.35 with its own tags, 87.80 with gold tags, 2026-10-22). The tag set it maps to has no case, so the parser does not see the case of the noun after *in*, and the frames reader would not know direction (*in urbem*) from position (*in urbe*). What a Latin frames reader needs first, in order:

1. the noun's case (from a Latin morphology, which Mova does not have) to read *in, sub, super* as motion or rest;
2. the agent of a passive in *ā/ab* + ablative, and *per* + accusative for the intermediary;
3. the double negation of §7 read as positive (*nōn nūllī* = some), the reverse of the French and Romance rule;
4. the question particles *-ne, nōnne, num* read as the speech act and the expected answer.

The open parsers measured on the same treebank (Stanza LLCT 95.92, UDPipe 2 95.03, both trained on it; [Open parsers: witnesses and tests](https://md.mova.elamur.ai/free/en/docs/repo-docs/parsers/)) show how far a parser with Latin morphology goes.

- [← previous 7. The cases and their uses](https://md.mova.elamur.ai/free/en/docs/languages/la/07-cases/)
- [next → 9. The simple clause: agreement and word order](https://md.mova.elamur.ai/free/en/docs/languages/la/09-simple-clause/)
- [↑ back to Latin (la) — a reference manual for Mova](https://md.mova.elamur.ai/free/en/docs/languages/la/)