The study of Latin syntax in a systematic way was particularly a feature of the late 19th century, especially in Germany. For example, in the 3rd edition of Gildersleeve's Latin Grammar (1895), the reviser, Gonzalez Lodge, mentions 38 scholars whose works have been used in its revision; of these 31 wrote in German, five in English and two in French. (The English scholars include Roby and Lindsay).
In the twentieth century, the German tradition was continued with the publication of two very comprehensive grammars: the Ausführliche Grammatik der lateinischen Sprache by Raphael Kühner and Karl Stegmann (1912, first edition 1879), and the Lateinische Grammatik by Manu Leumann, J.B. Hofmann, and Anton Szantyr (revised edition Munich 1977, first edition 1926). Among works published in English may be mentioned E.C. Woodcock's A New Latin Syntax (1959). More recently, taking advantage of computerised texts, three major works have been published on Latin word order, one by the American scholars Andrew Devine and Laurence Stephens (2006), and two (adopting a different approach) by the Czech scholar Olga Spevak (2010 and 2014).
Latin word order
The word order of Latin, which was not mentioned at all in Kennedy's Revised Latin Primer (1871), and received seven pages in Gildersleeve and Lodge (1895), has recently been the subject of much attention and some controversy.
Latin word order is relatively free. The verb may be found at the beginning, in the middle, or at the end of a sentence (although most often at the end); an adjective may precede or follow its noun (vir bonus or bonus vir both mean 'a good man'); and a genitive may precede or follow its noun (hostium castra 'the enemy's camp' or castra hostium = 'the camp of the enemy') (the latter is more common). There are stylistic differences between Latin authors; for example, Devine and Stephens note that while Caesar always writes castra pōnit 'he sets up camp', Livy more often writes pōnit castra. There are however certain constraints; for example, in prose a monosyllabic preposition such as in 'in' generally precedes its noun (e.g. in Italiā 'in Italy'). Moreover, even though adjectives can both precede and follow the noun, there is a tendency for different kinds of adjectives to take different positions, for example adjectives of size usually come before the noun (magnā vōce 'in a loud voice', rarely vōce magnā), while "modifiers that are more important than their noun or that specify it" (e.g. Via Appia 'the Appian Way') usually follow it.
Theories of word order
To explain these facts there are two main schools of thought. One, represented by Devine and Stephens (2006), argues from the point of view of generative grammar, and maintains that Latin prose has a basic underlying "neutral" word order, from which authors deviate for reasons of emphasis, topicalisation, rhythm, and so on. According to Devine and Stephens (p. 79), the basic order in "broad scope focus sentences" is as follows:
- Subject – Direct Object – Indirect Object / Oblique Argument – Adjunct – Goal or Source Argument – Non-Referential Direct Object – Verb
In the above scheme, two kinds of direct object are distinguished. Non-referential direct objects are those which form a ready-made ("precompiled") phrase with the verb, such as impetum facere 'to make an attack', lēgātōs mittere 'to send ambassadors', signum dare 'to give a signal', etc. These tend to go immediately before the verb, while ordinary direct objects tend to go after the subject.
"Adjunct" refers to adverbial phrases such as gladiō 'with a sword', adverbs of time, place or manner, comitative phrases such as cum suīs servīs 'with his slaves', and so on.
"Goal and Source Arguments" are phrases such as in castra 'into the camp' or ex urbe 'out of the city'. These tend to go closely with the verb.
The other approach, represented by Olga Spevak (2010), examines Latin word order from the point of view of functional grammar. This approach rejects the idea that there is a basic underlying word order in Latin, but seeks to explain word order in terms of a series of hierarchies – personal (1st person before 2nd, human before animals or things, etc.), semantic role (agent before patient, etc.), familiarity (given before new, definite before indefinite, etc.) – and considerations of topic and focus.
Research into Latin word order has been greatly accelerated by the recent development of computerised texts of Latin authors. For example, entering the phrase castra hostium into the Perseus PhiloLogic search engine of the University of Chicago instantly generates 18 examples from Livy and 4 from Caesar, whereas hostium castra generates 5 examples from Caesar and 3 from Livy.
Examples of Latin word order
The order of words found in Latin authors is often very different from that which is met with in books for beginners. One way of emphasising a word is to reverse the usual order. For example, in the famous opening sentence from Caesar's Gallic War, the usual order of numeral and noun trīs partīs 'three parts' is reversed to emphasise the number 'three':
- 'Gaul, considered as a whole, is divided into three (parts)'
In the following example, Cicero, instead of writing the usual order dē locō superiōre impetum faciunt ('they make an attack from higher ground'), draws attention to the word impetum 'attack' by moving the verb faciunt 'they make' to earlier in the sentence:
- lit. 'immediately several men, (armed) with weapons, on my client make from higher ground an attack'
Similarly, Livy, instead of writing urbem posterō diē ingressi sunt 'they entered the city the next day', writes:
- lit. 'having entered the following day the city'
Another technique for adding emphasis is separation, as in the following example from Nepos, where the phrase magnam pecūniam 'a large sum of money' is split up by other words:
- 'for he was carrying with him a large sum of money'
The technical term for this kind of separation is "hyperbaton" (Greek for 'stepping over'); it is described by Devine and Stephens as "perhaps the most distinctively alien feature of Latin word order".
In the following example from Cicero, there is not only a long hyperbaton (cruentum ... pugiōnem 'blood-stained dagger') but also the verb (tollēns 'raising') is brought forward of the subject (Brūtus):
- 'immediately Brutus, raising high the blood-stained dagger, called out "Cicero" by name'
Hyperbaton is frequent in poetry. The following examples come from the opening lines of Virgil's Aeneid:
- 'who first from the shores of Troy ...'
- 'came to Italy and the Lavinian coast'
- 'on account of cruel Juno's (saevae Iūnōnis) unforgetting anger (memorem īram)'
Considerations of rhythm and elegance also play a part in Latin word order. For example, Pliny the Younger begins a letter as follows:
- 'it is a great crop of poets this year has brought'
In this sentence, the object (magnum prōventum poētārum 'a great crop of poets') has been brought forward to make it the topic of the letter. The other striking feature is the order annus hic for the more usual hic annus 'this year'. Two reasons which might be suggested are Pliny's fondness for ending a sentence with the rhythm − u − − u − and also no doubt because of the elegant assonance of the vowels a-u-i a-u-i in the last three words.
Gender and number
Latin has three genders (masculine, feminine, and neuter) and two numbers (singular and plural).
- Masculine : hic est fīlius meus: 'this is my son'
- Feminine : haec est fīlia mea : 'this is my daughter'
- Neuter : hoc est corpus meum: 'this is my body'
The same three genders are also found in the plural:
- Masculine : hī sunt fīliī meī : 'these are my sons'
- Feminine : hae sunt fīliae meae : 'these are my daughters'
- Neuter : haec mea sunt: 'these things are mine'
As can be seen from the examples above, adjectives and pronouns have to agree with nouns in both gender and number. This is true whether they are used attributively ('the long road') or predicatively ('the road is long'). Participles such as captus/capta 'captured' and the numbers ūnus 'one', duo 'two', trēs 'three' also show agreement.
In Latin, words referring to males are always masculine, words referring to females are usually feminine. (An exception is scortum (neuter) 'a whore'.) Words referring to things can be any of the three genders, for example mōns 'mountain' (masculine), arbor 'tree' (feminine), nōmen 'name' (neuter). However, there are certain rules; for example, nouns with the suffixes -a (unless referring to men), -tiō, -tās are feminine; the names of trees, islands, and countries, such as pīnus 'pine', Cyprus 'Cyprus', and Aegyptus 'Egypt' are also usually feminine, and so on. Some nouns such as parēns 'parent' can vary between masculine and feminine and are called of "common" gender.
- 'father and mother are dead (masc.)'
- 'the wall (masc.) and gate (fem.) had been struck (neut.) by lightning' (lit. 'touched from the sky')
However, sometimes the adjective may agree with the nearest noun.
Some nouns are used in the singular only, others (called plūrālia tantum) only in the plural. Examples of plural nouns with singular meaning are castra 'a military camp', litterae 'a letter', vestīmenta 'a set of clothes', quadrīgae 'a four-horse chariot'. For counting these a special set of numbers is used: ūnī, bīnī, trīnī, quadrīnī, quīnī, instead of the usual ūnus, duo, trēs, quattuor, quīnque:
- 'my daughter Tullia came to me ... and delivered three letters'
- 'Octavius surrounded the town with five camps'
Nouns, pronouns, and adjectives in Latin change their endings according to their function in the sentence. The different endings are called different 'cases'. (Case endings of a similar kind are also found in other languages, such as Greek, German, Russian, Finnish, Sanskrit, and Turkish.)
The five cases most commonly used in Latin and their main meanings are as follows (the cases are presented here in the order familiar in Britain and countries influenced by Britain ever since the publication of Kennedy's Latin Primer in the 19th century, as opposed to the traditional order – Nom, Gen, Dat, Acc, Voc, Abl – still used in America and Germany):
- Nominative : is rēx : 'that king' (Subject, or Complement (e.g. 'he is that king'))
- Accusative : eum rēgem : 'that king' (Object, or Destination)
- Genitive : eius rēgis : 'of that king'
- Dative : eī rēgī : 'to that king', 'for that king'
- Ablative : eō rēge : 'with that king' (also 'by, from, in')
(A small line, called a macron, over a vowel indicates that it is pronounced long.)
Two other, less frequently used cases are the Vocative (ō rēx 'o king!), which in many nouns is the same as the Nominative, and the Locative, which is used mostly with the names of cities (e.g. Rōmae 'in Rome') and a very limited number of ordinary nouns (e.g. domī 'at home').
Examples of case use
For the most part the use of cases is quite straightforward. The following examples from Caesar show the cases in use in a basic sense:
- 'Curio (Nom.) sends ahead Marcius (Acc.) to Utica (Acc.) with the ships (Abl.)'
Here Cūriō as subject of the verb is nominative, Mārcium as direct object is accusative; Uticam is also accusative as it is the destination; and nāvibus 'with the ships' has the ablative ending, which means 'with'.
- 'Pompey (Nom) from Luceria (Abl) sets out to Canusium (Acc)'
Here Pompeius is subject (Nom.), Lūceriā shows another meaning of the ablative ending, namely 'from', and Canusium is accusative of destination.
- 'Caesar (Nom) gave a signal (Acc.) to the soldiers (Dat)'
Here mīlitibus, although it shares the same ending as the ablative nāvibus in the previous example, is clearly dative, meaning 'to or for the soldiers'. (Usually when a noun has an ambiguous ending such as -ibus it will be interpreted as dative if it is a person, ablative if it is a thing.)
However, the description of the use of cases is not always simple. The classification of the uses of the dative alone takes up nearly twelve pages in Woodcock's A New Latin Syntax and ten pages in Gildersleeve and Lodge. For example, when asking someone's name, a Roman would say:
- quid est tibī nōmen? (Plautus)
- 'what's your name?' (lit. 'what is for you name?)
This is an example of the Dative of Possession, as in:
- 'he had two daughters' (lit. 'to him there were two daughters')
Another surprising idiomatic use is the "Dative of the Person Affected":
- 'I haven't stolen anything from you' (lit. 'for you')
- 'don't fight with (lit. 'for') two people at once'
- 'for no one was my arrival a trouble or expense' (lit. 'for a trouble')
- 'he did not persuade him (lit. 'for him')'
- 'he ordered him' ('gave an order to him')
Frequently, to make the meaning more precise, a noun in the accusative or ablative is preceded by a preposition such as in' in, into', ad 'to', cum 'with', or ex 'out of'. This is especially so if the noun refers to a person. For example: ad rēgem (Acc) 'to the king' (with a verb of motion), ā rēge (Abl) 'by the king'; in urbem (Acc) 'into the city', in urbe (Abl) 'in the city', cum eō (Abl.) 'with him'.
Four prepositions can be followed by more than one case, depending on their meaning. These are in 'in' (Abl), 'into' (Acc.); sub 'under' (Abl.), 'to the foot of' (Acc.); super 'over, above' (Acc.), 'concerning' (Abl.); and subter 'under' (usually with Acc.)
However, when the meaning of an accusative or an ablative is clear (for example Canusium (Acc) 'to Canusium', nāvibus (Abl) 'with the ships', posterō diē (Abl) 'on the following day'), the case ending alone is sufficient to give the meaning. Unlike in Greek, prepositions are not used in Latin with the dative or genitive.
Prepositions almost always precede their noun or pronoun, except that cum 'with' follows a personal pronoun, e.g. mēcum 'with me' and sometimes a relative pronoun (quīcum, quōcum and cum quō are all possible for 'with whom'). There are occasional exceptions, especially with two-syllable prepositions after pronouns, e.g. haec inter (Virgil) 'in the midst of these'.
Sometimes when the noun has an adjective it is placed before the preposition for emphasis, e.g. magnā cum cūrā 'with great care' (Cicero), but this is not an invariable rule. Occasionally also the opposite order (noun-preposition-adjective) may be used in poetry and later prose, e.g. silvā lupus in Sabīnā (Horace) 'a wolf in the Sabine forest', metū in magnō (Livy) 'in great fear',
Latin has six main tenses in the indicative mood, which are illustrated below using the verb facere 'to make' or 'to do':
- Present : faciō : 'I do', 'I am doing'
- Future : faciam (2nd person faciēs): 'I will do', 'I will be doing'
- Imperfect : faciēbam : 'I was doing', 'I used to do', 'I began to do'
- Perfect : fēcī : 'I did', 'I have done'
- Future Perfect : fēcerō : 'I will have done'
- Pluperfect : fēceram : 'I had done'
The verb sum 'I am', which is irregular, has the tenses sum, erō, eram, fuī, fuerō, fueram. Some verbs (conjugations 1 and 2) instead of the Future -am, -ēs, -et etc. have a different Future ending in -bō, -bis, -bit, e.g. amābō 'I will love'.
Idiomatic uses of tenses
(a) The Present tense is often used in narrative in a past sense, especially for events that are sudden or unexpected. This is known as the "historic Present":
- 'immediately several men make an attack on my client from higher ground'
(b) The Present is also usually used after dum 'while', even when referring to the past:
- 'while she was fleeing, her cloak (vēlāmina) slipped from her back (tergō) and she left it behind'
- 'this nation was brave as long as Lycurgus's laws were in force'
- 'I've been rolling this rock (hoc saxum) long enough (satis diū) now!'
- 'which they had been desiring for a long time now'
- 'he has been in my debt for many years' (lit. 'there are many years when he is in my debt')
- 'for a long time there was silence'
- 'he lived in our house for many years'
(e) The Perfect fuit 'it was once', 'it used to be' is also used for a situation which is no longer in existence:
- 'there was once a statue of Attus to the left of the senate house'
- 'I have left a horse and a mule for you at Brundisium'
(g) Certain verbs, such as ōdī 'I have come to hate' (= 'I hate'), meminī 'I have remembered' (= 'I remember'), and nōvī 'I have come to know' (= 'I know') are used in the Perfect tense but have the meaning of a Present tense. The Pluperfect of these verbs has the meaning of an Imperfect:
- 'I remember that I was present'
- 'he did not know Catiline, since the latter was at that time governor of Africa'
- 'you will die, if you utter a sound!' (lit. 'if you will have uttered')
- 'everyone began shouting at once'
- 'as a matter of fact I was sitting there as I was writing this to you' (i.e. 'I am sitting there as I write this to you')
Passive and deponent verbs
In addition to the active voice tenses listed above, Latin has a set of passive tenses as follows:
- Present : capior : 'I am captured', 'I am being captured' (by someone or something)
- Future : capiar (2nd singular capiēre or capiēris) : 'I will be captured'
- Imperfect : capiēbar : 'I was being captured', 'I used to be captured'
- Perfect : captus sum : 'I was captured', 'I have been captured'
- Future Perfect : captus erō : 'I will have been captured'
- Pluperfect : captus eram : 'I had been captured'
The infinitive of a passive verb ends in -ī or -rī: capī 'to be captured', audīrī 'to be heard', etc.
Corresponding to the active periphrastic tenses is a form made the gerundive, e.g. capiendus erat 'he was to be captured', i.e. 'he had to be captured'. (See below on the gerundive.)
The Perfect participle (captus/capta etc.) which is used to make the Perfect, Future Perfect, and Pluperfect tenses changes according to the gender and number of the subject:
- 'the citadel has been captured by the enemy!'
Most of the verbs ending in -or are true passives in meaning (i.e. they represent actions which are done by someone or by something). However, there are a few which are ambivalent and can be either active or passive in meaning, such as vertor 'I turn' (intransitive) or 'I am turned', volvor 'I revolve' (intransitive) or 'I am rolled':
'meanwhile the sky turns and night falls upon the Ocean'
In addition, there are a few verbs such as proficīscor 'I set out', sequor 'I follow', cōnor 'I try' which despite their passive endings only have an active meaning. These verbs (which have no active counterpart ending in -ō) are called deponent verbs:
- 'he himself set out for Italy'
- 'he ordered this man to follow him'
In Latin, unlike English, only the direct object (not the indirect object) of an active verb can be made the subject of a passive verb. It is not correct to say in Latin 'the soldiers were being given their pay' but only 'pay was being given to the soldiers':
- 'pay was being given to the soldiers'
Another unusual feature of Latin, compared with English, is that intransitive verbs such as eō 'I go', veniō 'I come', and pugnō 'I fight' can be made passive, but only in a 3rd person singular impersonal form:
- 'they go into an ancient forest' (lit. 'going is done')
- 'on the seventh day they reached Carthage'
The subjunctive mood
As well as the indicative mood illustrated above, which is used for stating and asking facts, and an imperative mood, used for direct commands, Latin has a subjunctive mood, used to express nuances of meaning such as 'would', 'could', 'should', 'may' etc. (The word mood in a grammatical sense comes from the Latin modus, and has no connection with the other meaning of 'mood', in the sense of 'emotional state', which comes from a Germanic root.)
The subjunctive is often translated with 'should', 'could', 'would', 'may' and so on, but in certain contexts, for example indirect questions or after the conjunction cum 'when' or 'since', it is translated as if it were an ordinary indicative verb.
Often in English the subjunctive can be translated by an infinitive; for example the literal imperāvit ut īret 'he ordered that he should go' becomes in more idiomatic English 'he ordered him to go'.
There are four tenses of the subjunctive, which are as follows:
- Present : faciam (2nd person faciās) : 'I may do', 'I would do', 'I should do' (also simply 'I do')
- Imperfect : facerem : 'I would be doing', 'I should do' (in a past context) (also simply 'I was doing')
- Perfect : fēcerim : 'I may have done' (also: 'I did')
- Pluperfect : fēcissem : 'I would or should have done' (also 'I had done')
The verb sum 'I am' has the following four tenses in the subjunctive mood: sim, essem, fuerim, fuisset.
The verb possum 'I am able' similarly has possim, possem, potuerim, potuissem.
Volo 'I want is also irregular, with tenses velim, vellem, voluerim, voluissem.
The Imperfect and Pluperfect subjunctive tend to be used either where the context is past, or (in the case of the Imperfect) when the speaker is imagining an unreal situation which might potentially be true now (e.g. 'if I were rich').
Uses of the subjunctive
The subjunctive has numerous uses, of which only a few can be given here. One (known as the 'potential' subjunctive) states what potentially may, might, would, or could happen. The negative of this kind is nōn:
- 'this may perhaps seem harsh'
- 'if only he were here already!'
- 'let's live, my Lesbia, and let's love'
- '(he said that) they would easily find the place where he was'
It also applies to all indirect questions:
- quārē id faciam, fortasse requīris (Catullus)
- 'perhaps you ask why I do this'
When used for indirect speech or an indirect question, the subjunctive is translated as if were the corresponding tense of the indicative.
Subjunctive after conjunctions
The subjunctive mood is very frequently used in subordinate clauses following conjunctions.
- 'when they are silent, (it is as if) they are shouting'
- 'while I was sitting sadly at home, Venerius came running up'
- 'in view of the fact that these things are so' / 'since this is so'
- 'he did nothing to help me, though (or: at a time when) he could have done'
When followed by the indicative, the conjunction ut can mean 'as' (e.g. ut fit 'as generally happens') or 'as soon as' or 'when' (ut vēnī 'as soon as I came'). But with the subjunctive ut has the meaning 'that' or 'so that'.
- '(Hannibal) came to Crete so that there he could consider (in order to consider) where he should go to next'
- 'he ordered him to go round (lit. 'that he should go round') all the doors of the building'
- 'and he had built it in such a way that in all parts of the building it had exits'
After sī 'if', the subjunctive expresses an imagined or unreal situation:
- 'which, if I had been killed, could not have happened'
- 'if they were to come to life and talk to you, what answer would you be making?'
After nē 'that not', the subjunctive can express a negative purpose:
- 'so that he would not be able to escape from here, the ephors blocked up the doors'
It can also introduce a negative indirect command:
- 'they requested him not to move his camp any nearer to them'
- 'fearing that he might be handed over to the enemy'
- 'he who obeys modestly, seems to be the sort of person who one day is worthy to rule'
- 'they sent ambassadors to Rome to ask for help'
Another meaning is 'in view of the fact that' (causal), as in the following example, said jokingly of a consul who was elected on the last day of the year:
- '(Caninius) was of amazing vigilance, in view of the fact that he didn't see any sleep in the whole of his consulate!'
- 'Verginius waited until he had a chance to consult his colleague'
- ōderint dum metuant (Accius)
- 'let them hate, provided that they fear'
The conjunctions priusquam and antequam both mean 'before (something happened)'. If the event actually happened, the verb is usually in the indicative mood; but when the meaning is 'before there was a chance for it to happen', the verb is subjunctive:
- 'he fortified the hill quickly, before it could be noticed by the enemies'
The conjunction quīn (literally, 'how should it not be?') is always used after a negative verb or the equivalent, typically 'there is no doubt that', 'who does not know that...?', and so on. The words following quīn are always positive and usually state what was actually the case:
- 'I have no doubt that all your friends will have written to you'
- 'who does not know that there are three kinds of Greeks?'
Another usage is after a negative verb such as 'I can't help doing' or 'he did not refrain from doing':
- 'I can't do otherwise than to thank you'
- 'Antiochus did not refrain from publishing a book against his own teacher'
Equally it can be used in sentences of the kind 'A didn't happen without B also happening':
- 'up to now I have not let a day go past without dropping you a line'
- 'he wasn't even able to save himself without saving the Republic and you at the same time'
In sentences like the following, there is potential for confusion, since the quīn clause, though positive in Latin, is translated in English with a negative:
- 'there was no one in Lilybaeum who did not see it'
- 'it was quite impossible that Cleomenes would not be spared'
In the following context, the words after quīn express not what actually happened but what very nearly happened:
- 'nor were they far from being expelled from the camp'
The imperative mood
The imperative mood is used for giving direct orders. The active form can be made plural by adding -te:
- 'give me a thousand kisses, then a hundred!'
- 'give me your right hands and your oath!'
Deponent verbs have an imperative ending in -re or -minī (plural):
- 'the gates are open: depart!'
- sequiminī mē hūc intrō ambae (Terence)
- 'follow me this way inside, both of you'
An imperative is usually made negative by adding nōlī(te) (literally, 'be unwilling!'). However, in poetry it can sometimes be made negative with the particle nē:
- nōlīte mīrārī (Seneca the Elder)
- 'don't be surprised'
- 'do not terrify me, who am already scared, obscene birds!'
- 'if anything happens, write to me'
It is also possible to express a request or order by using the Present or Perfect subjunctive, the Future indicative, or expressions such as cūrā ut 'take care to...', fac ut 'see to it that...' or cavē nē 'be careful that you don't...'
Latin has three infinitives in the active voice, and three passive. Since faciō is irregular in the passive ('to be done' is fierī, taken from the verb fīō 'I become'), they are here shown using the verb capiō 'I capture':
- Present : capere : 'to capture, to be capturing'
- Perfect : cēpisse : 'to have captured'
- Future : captūrus esse : 'to be going to capture'
- Present : capī : 'to be captured'
- Perfect : captus esse : 'to have been captured'
- Future : captum īrī : 'to be going to be captured'
The infinitives of sum 'I am' are esse, fuisse, and futūrus esse (often shortened to fore). Possum 'I am able' has posse and potuisse, volō 'I want' has velle and voluisse. Neither of these verbs has a Future infinitive, and the Present infinitive is used instead.
Uses of the infinitive
- 'to live is to think'
- 'we consider to be in error, to be ignorant, to be deceived as something shameful'
It can also be used, as in English, dependent on an adjective, or with verbs such as possum 'I am able' or volō 'I want':
- 'it is a sweet and glorious thing to die for one's country'
- 'I can't bear it'
It can also be used, as in English, with verbs such as iubeō 'I order', vetō 'I forbid', patior 'I allow', volō 'I want' and so on, where the main verb takes an object in the accusative case:
- 'I want you to know this'
However, other verbs of similar meaning, such as imperō 'I order', persuādeō 'I persuade', and hortor 'I urge', are not used with an infinitive, but with ut and the subjunctive mood:
- 'he is urging me to write to the senate' (lit. 'that I should write')
Infinitive in indirect statements
A very common use of the infinitive in Latin, in which is differs from English, is its use for indirect statements, that is for sentences where a subordinate clause is dependent on a main verb such as 'he says', 'he knows', 'he pretends', 'he believes', 'he thinks' and so on. In Latin, instead of 'they pretend that they want', the idiom is to say 'they pretend themselves to want':
- 'they pretend that they want peace'
Similarly 'I'm glad you've arrived safely' becomes 'I am glad you to have arrived safely':
- 'I am glad you have arrived safely'
In this construction, the subject of the infinitive (sē, tē in the above examples) is in the accusative case.
So common is this construction in Latin, that often the verb 'he said' is simply omitted if it is clear from the context, the accusative and infinitive alone making it clear that the statement is reported:
- 'a terrible thing had happened (she said)'
Other ways of expressing 'that'
Not every subordinate clause which starts with the conjunction 'that' in English is translated with an accusative and infinitive. In some contexts ut with the subjunctive is required, for example after a verb of happening:
- 'it happened by chance that some ambassadors of King Prusias were dining in Rome'
- 'I omit the fact that he chose that house for himself'
This type of clause with quod (which became que in modern French, Portuguese and Spanish and che in Italian) gradually took over from the Accusative and infinitive construction and became the usual way of expressing indirect speech in modern Romance languages which are descended from Latin.
- Present : faciēns (pl. facientēs) : 'doing'
- Perfect : factus : '(having been) done'
- Future : factūrus : 'going to do'.
Thus, there is no passive present or future participle, and no active past participle. In deponent verbs, however, the Perfect participle is active in meaning, e.g. "profectus", 'having set out', "conatus" 'having tried'.
The verb sum 'I am' has no Present or Perfect participle, but only the Future participle futūrus 'going to be'.
Uses of participles
- 'he was doused with boiling water'
- occīsōs sepelīvit (Eutropius)
- 'he buried the dead (those who had been killed)'
More frequently, however, a participle is more like a verb, and can often replace the first of two verbs in a sentence:
- Caesar Cascae bracchium arreptum graphiō trāiēcit (Suetonius)
- 'Caesar grabbed Casca's arm and stabbed it with his writing instrument'
Literally, 'Caesar with his writing instrument (graphiō) stabbed the arm (bracchium), which had been grabbed, for Casca' (Cascae here is probably Dative of the Person Affected.)
Mostly commonly, participles are translated into English using '-ing' (or 'having'), 'when', or 'who':
- 'and when he tried to leap forward he was slowed down (tardātus) by another wound'
- 'Lepta came running'
- 'drawing his sword, he came to Lucretia, when she was sleeping / who was sleeping'
In this last example the phrase strīctō gladiō (lit. 'with drawn sword') illustrates a common idiom of putting a noun and participle in the Ablative case to represent the circumstances of the main event. This idiom is referred to as an "Ablative absolute": Another example is:
- 'when the signal was given (lit. 'with signal given'), they made an attack on the enemy'
- 'although it can't see itself, the eye discerns other things'
- 'they raised a sign from the wall that the town had been captured' (lit. 'of the town having been captured')
Normally a Present participle represents an action which is simultaneous with the main event ('he came running'), and a Perfect participle represents one which has already happened ('after drawing his sword'). In the following example, however, the Perfect participle represents the result following the main action:
- 'she tore her hair, making it loose'
Participles are much commoner in Latin than in English. Sometimes multiple participles can be used in a single sentence:
- noctū lūmine appositō experrēcta nūtrīx animadvertit puerum dormientem circumplicātum serpentis amplexū. quō aspectū exterrita clāmōrem sustulit. (Cicero)
- 'in the night, in the light of a lamp placed nearby, the nurse, who had woken up, noticed that the boy, while he was sleeping, had been wrapped around with the coils of a snake; terrified by this sight, she raised a cry'
The gerundive is a verbal adjective ending in -ndus (-nda etc. if feminine). It is usually passive in meaning (although a few deponent verbs can form an active gerund, such as secundus 'following' from sequor 'I follow') The usual meaning of the gerundive is to state that something needs to be done or must be done:
- nunc est bibendum (Horace)
- 'now drinking must be done' (i.e. 'now we must drink')
- Catō inexpiābilī odiō dēlendam esse Carthāginem ... prōnūntiābat (Florus)
Because it is passive in meaning, the gerundive is usually formed from transitive verbs. However, intransitive verbs such as eō 'I go' and persuādeō 'I persuade', which be used passively in an impersonal construction, can also have an impersonal gerundive, ending in -um:
- 'I have to go to Arpinum'
- persuādendum iūdicī est (Quintilian)
- 'the judge has to be persuaded'
- 'they sent the military tribune Lucius Septimius to kill Pompey'
- 'Datames handed this man over in chains to Mithridates for him to be led to the King'
The gerund is a verbal noun ending in -ndum (accusative), -ndī (genitive), or -ndō (dative or ablative). Although identical in form to a neuter gerundive, and overlapping the gerundive in some of its uses, it is possible that it has a different origin.
Gerunds are usually formed from intransitive verbs, and are mainly used in sentences such as the following where the meaning is 'by doing something', 'of doing something', or 'for the purpose of doing something'. A gerund is never used as the subject or direct object of a verb (the infinitive is used instead).
- 'by coming here, you have saved the army'
- 'alkaline water is good for drinking'
- 'weather suitable (idōneam) for sailing'
- 'for the sake of sacrificing, I climbed up to Delphi'
- 'by saying incongruous things laughs (rīsūs) are raised'
They can also be formed from deponent verbs such as ingredior 'I enter':
- 'for others fear of the enemy gave them the boldness (audāciam) to enter (lit. of entering) the river'
- 'dry wood (lignum) is a suitable material for striking fire'
The supine is a rarely used part of the verb ending in -tum or (in some verbs) -sum. Although it is identical with the accusative case of verbal nouns such as adventus 'arrival', mōtus 'movement', etc., it differs from them in that it is a verb as well as a noun, and can sometimes take a direct object.
The supine is normally used to express purpose, when combined with a verb of going such as eō 'I go' or mittō 'I send':
- 'Maecenas goes to play a game, Virgil and I to sleep'
- '(the girls) come to watch, but they also come so that they can be looked at themselves'
In the following example it takes a direct object:
- 'they send ambassadors to Caesar in order to ask for help'
There is another form of the supine, an Ablative in -ū, found with certain verbs only. But this cannot take an object. It is used in phrases such as mīrābile dictū 'amazing to say', facile factū 'easy to do':
- 'it is easier in the saying than in reality'
- Devine, Andrew M. & Laurence D. Stephens (2006), Latin Word Order. Structured Meaning and Information. Oxford: Oxford University Press. Pp. xii, 639. ISBN 0-19-518168-9. Google books sample. See also reviews by M. Esperanza Torrego and Anne Mahoney.
- Gildersleeve, B.L. & Gonzalez Lodge (1895). Gildersleeve's Latin Grammar. 3rd Edition. (Macmillan)
- Greenough, J.B. et al. (1903). Allen and Greenough's New Latin Grammar for Schools and Colleges. Boston and London.
- Kennedy, Benjamin Hall (1871). The Revised Latin Primer. Edited and further revised by Sir James Mountford, Longman 1930; reprinted 1962.
- Kühner, Raphael; & Karl Stegmann (1912) . Ausführliche Grammatik der lateinischen Sprache
- Leumann, Manu; J.B. Hofmann, & Anton Szantyr (1977) . Lateinische Grammatik. Munich.
- Nisbet, R.G.M. (1999). "The Word-Order of Horace's Odes. Proceedings of the British Academy, 93, 135-154.
- Rose, H.J. (1924). Review of J. Marouzeaux (1922), "L'Ordre des Mots dans la Phrase latine: I. Les Groupes nominaux". The Classical Review, vol. 38, issue 1-2.
- Spevak, Olga (2010). Constituent Order in Classical Latin Prose. Studies in Language Companion Series (SLCS) 117. Amsterdam/Philadelphia: John Benjamins Publishing Company, 2010. Pp. xv, 318. ISBN 9789027205841. Reviewed by J.G.F. Powell in the Bryn Mawr Classical Review
- Spevak, Olga (2014). The Noun Phrase in Classical Latin Prose. Amsterdam studies in classical philology, 21. Leiden; Boston: Brill, 2014. Pp. xiii, 377. ISBN 9789004264427. Review by Patrick McFadden.
- Walker, Arthur T. (1918) "Some Facts of Latin Word Order". The Classical Journal, Vol. 13, No. 9, pp. 644–657.
- University of Chicago Perseus under PhiloLogic searchable corpus. Perseus under PhiloLogic home page
- Online version of Allen & Greenough's Latin Grammar
- Online version of Gildersleeve & Lodge's Latin Grammar
- Devine & Stephens (2006).
- Spevak (2010); Spevak (2014).
- Three female research students whose results are reported in Walker (1918) are a minor exception.
- Gildersleeve & Lodge (1895), p. 431.
- Gildersleeve & Lodge (1895), p. 430.
- Devine & Stephens (2006), p. 126.
- Gildersleeve & Lodge (1895), p. 432.
- Walker (1918), p. 648.
- H.J. Rose (1924)
- Spevak (2014), pp. 212ff.
- Spevak (2010), pp. 2ff.
- Perseus under PhiloLogic home page
- Gildersleeve & Lodge (1895), p. 429.
- Walker (1918), p. 648.
- Caesar, B.G. 1.1.1
- Cicero, Mil. 29.
- Livy, 5.41.4
- Nepos, Hannibal, 9.2.
- Divine & Stephens (2006), p. 524.
- Cicero, Phil. 2.28.
- Nisbet (1999).
- Virg. Aen. 1.1
- Virg. Aen. 1.2-3
- Nisbet (1999), p. 137.
- Virg. Aen. 1.4.
- Gildersleeve & Lodge (1895), p. 429; Walker (1918), p. 652.
- Pliny, Ep. 1.13.1
- "a clausula of cretic plus cretic, a favorite with Pliny": Selatie E. Stout, Scribe and Critic at Work in Pliny's Letters (1954), p. 150. (The symbol – stands for a long syllable, and u for a short one.)
- Matt. 17.5
- Matt. 26.26
- Virgil, Ec. 9.1
- Kennedy (1930) , pp. 14-15.
- The appendix of Kennedy's Revised Latin Primer, pp. 221-225, has a series of rhymes to assist in learning the rules for gender.
- Gildersleeve & Lodge (1895), p. 183.
- Terence, Eun. 518
- Livy, 32.29.1
- Cicero, Att. 11.17
- Caesar, B.C. 3.9
- Blake, Barry (1994). Case. Cambridge Textbooks in Linguistics: Cambridge University Press.
- Christopher Stray (1996), Grinders and Grammars: A Victorian Controversy (The Textbook Colloquium).
- Caesar, Bell. Civ. 2.24.1.
- Caesar, Bell. Civ. 1.24.1
- Caesar, Bell. Gall. 7.27.
- Woodcock (1959), pp. 38-50.
- Gildersleeve & Lodge (1895), pp. 218-230.
- Plautus, Pseud. 634 and 639; Pre. 697; Trin. 886.
- Plautus, Poen. 84
- Plautus, Aul. 635; cf. Gildersleeve and Lodge (1895), p. 219.
- Kühner & Stegmann (1912), p. 319.
- Catullus 62.64
- Woodcock (1959), pp. 48-9.
- Cicero, Ver. 2.1.16; cf. Woodcock (1959), p. 48.
- Woodcock (1959), pp. 41-2.
- Seneca, Ep. 70.10
- Nepos, Hann. 12
- Greenough et al. (1903), pp. 131-136.
- Gildersleeve & Lodge (1895), p. 267.
- Gildersleeve & Lodge (1895), p. 267.
- Virgil, Aen. 8.671
- Cicero, Inv. 1.50
- Walker (1918), p. 651-2
- Gildersleeve & Lodge (1895), p. 267, note 1
- Horace, Car. 1.22.9
- Livy, 9.37.11
- Gildersleeve & Lodge (1895), p. 88.
- See Gildersleeve & Lodge (1895), pp. 154-167.
- Cicero, Mil. 29.
- Ovid, Met 4.55
- Gildersleeve & Lodge (1895), p. 365.
- Cicero, Tusc. 1.101
- Gildersleeve & Lodge (1895), pp. 157, 159.
- Terence, Eu. 1084
- Livy, 24.29
- Cicero, Fam. 15.14.1; cf. Gildersleeve & Lodge (1895), p. 372.
- Gildersleeve & Lodge (1895), p. 158.
- Livy 7.10.1
- Cicero, Tusc. 5.113
- Livy, 1.36.5
- Gildersleeve & Lodge (1895), p. 159.
- Cicero, Fam, 16.9.3.
- Cicero, Fin. 2.55
- Cicero, Cael. 10
- Gildersleeve & Lodge (1895), pp. 373; 380-381
- Livy 1.58.3.
- Woodcock (1959), p. 15.
- Cicero, Verr. 2, 2, 188
- Gildersleeve & Lodge (1895), pp. 166-7.
- Cicero, Att. 1.10 (6)
- Livy, 9.24
- Virgil, Aen. 2.250
- Gildersleeve & Lodge (1895), pp. 110-114.
- Caesar, B.G. 3.35.2
- Nepos, Alc. 10.5
- Gildersleeve & Lodge (1895), p. 152.
- Livy, 1.12
- Woodcock (1959), p. 43.
- Virgil, Aen. 6.179
- Livy, 26.42
- Woodcock (1959), p. 83.
- Woodcock (1959), p. 89.
- Cicero, Ver. 2.5.7.
- Woodcock (1959), p. 87.
- Cicero, Fam. 11.25.2
- Woodcock (1959), p. 85.
- Catullus, 5.1.
- Woodcock (1959), pp. 223ff.
- Nepos, Hann. 12.3
- Catullus 85
- Gildersleeve & Lodge (1895), pp. 370-373.
- Cicero, Cat. 1.21
- Woodcock (1959), pp. 187ff.
- Cicero, Ver. 2.4.32.
- Gildersleeve & Lodge (1895), pp. 374-5.
- Cicero, Dom. 142 et passim
- Gildersleeve & Lodge (1895), p. 375.
- Cicero, Att. 9.13.8
- Woodcock (1959), pp. 108ff.
- Nepos, Hann. 9.1.
- Woodcock (1959), pp. 101ff.
- Nepos, Hann. 12.4
- Woodcock (1959), pp. 114ff.
- Nepos, Hann. 12.3
- Gildersleeve & Lodge (1895), p. 392.
- Cicero, Sest. 49
- Cicero, Fin. 4.22.61; cf Gildersleeve & Lodge (1895), p. 384.
- Nepos, Paus. 5.2.
- Caesar, B.G. 4.9.1
- Woodcock (1959), pp. 144-147.
- Nepos, Hann. 9.1
- Gildersleeve & Lodge (1895), p. 353.
- Cicero, Leg. 3.2.5
- Woodcock (1959), p. 109.
- Livy, 5.35.4
- Gildersleeve & Lodge (1895), p. 406.
- Cicero Fam. 7.30.1
- Gildersleeve & Lodge (1895), p. 367.
- Livy 4.21.10
- Gildersleeve & Lodge (1895), p. 368.
- Accius, quoted in Cicero, Off. 1.28.97
- Gildersleeve & Lodge (1895), pp. 369-370
- Caesar, B.C. 1.54.4
- Woodcock (1959), pp. 141-4; Gildersleeve & Lodge (1895), pp. 356-357.
- Cicero, Fam. 5.8.1
- Cicero, Flacc. 64
- Cicero, Fam. 10.24.1 (letter from Plancus)
- Cicero, Ac. 2.4.12
- Cicero, Att. 7.15.1
- Cicero, Mil. 30
- Cicero, Verr. 2.5.140
- Cicero, Verr. 5.104
- Caesar, B.C. 2.35.4
- Catullus 5.
- Livy 1.58.7
- Cicero, Cat. 1.10
- Terence, Hec. 793
- Seneca the Elder, Controv. 7.7.2
- Virgil, Aen. 12.875
- Gildersleeve & Lodge (1895), p. 174.
- Cicero, Att. 10.1.3
- Gildersleeve & Lodge (1895), p. 176.
- e.g. Allen & Greenough (1903), p. 262
- Gildersleeve & Lodge (1895), p. 167; Woodcock (1959), p. 14.
- Gildersleeve & Lodge (1895), p. 334, note 3.
- Gildersleeve & Lodge (1895), p. 331, note 3.
- Gildersleeve & Lodge (1895), p. 334.
- Kennedy, p. 162.
- Cicero Tusc. 5.111
- Cicero, Off. 1.18
- Horace, Carm. 3.2
- Cicero, Att. 7.2.7
- Cicero Att. 1.18.6
- Cicero, Fam. 10.31.4
- Cicero, Att. 14.15.2
- Terence, Ph. 286
- Livy 1.58.5
- Gildersleeve & Lodge (1895), p. 355.
- Nepos, Hann. 12.1
- Gildersleeve & Lodge (1895), p. 329.
- Cicero, Cluent. 66.188
- Woodcock (1959), p. 71.
- Woodcock (1959), p. 77.
- Cicero, Ver. 2.1.67
- Eutropius, 2.11
- Suetonius, Jul. 82.2
- Suetonius, Jul. 82.2
- Cicero, Fam. 3.7.4
- Livy, 1.52.8
- Woodcock (1959), p. 73-4.
- Caesar, Gal. 1.52.3
- Woodcock (1959), p. 73.
- Cicero, T.D. 1.67
- Woodcock (1959), pp. 75-6.
- Livy, 4.34.1
- Virgil, Aen. 12.870
- Cicero, Div. 1.79
- Woodcock (1959), p. 158.
- Horace, Carm. 1.37
- Florus 184.108.40.206
- Woodcock (1959), p. 158.
- Woodcock (1959), p. 159.
- Cicero, 13.9.2
- Quntilian, 7.3.15
- Woodcock (1959), p. 164.
- Caesar, Civ. 3.104.2
- Nepos, Dat. 4.5
- Woodcock (1959), p. 159
- Woodcock (1959), p. 159.
- Livy, 7.35; servāstis is a contraction of servāvistis.
- Pliny N.H. 31.32.59
- Caesar, B.G. 4.23.1
- Livy, 42.42
- Woodcock (1959), pp. 161-2
- Cicero, de Or. 2.289
- Livy, 22.56
- Kennedy, p. 165.
- Seneca, N.Q. 2.22.1
- Horace, Serm. 1.5.48
- Ovid, Ars 1.1.99
- Caesar, B.G." 1.11.2
- Kennedy, p. 167
- Woodcock (1959), p. 112.
- Livy, 40.35.13
- Woodcock (1959), pp. 112-3