Early Latin and late Latin

Michèle FRUYT

University of Paris-Sorbonne (Paris 4)



12. A turning-point for morphology and syntax

While, as we have shown, a lexical turning-point is to be seen in the 2nd century AD with Apuleius and Gellius, some Latinists have also drawn attention to the role of Cyprian in the middle of the 2nd century AD, since this author’s works mark a turning-point in Latin morphology and syntax, with new evolved features.

12.1. The fu- forms

In Cyprian, the evolved morphological type amatus fuerim, amatus fueris, amatus fuerit in the passive perfect subjunctive with a fu- form for the auxiliary (instead of amatus sim, amatus sis, amatus sit) is systematically used in hypothetical subordinate clauses introduced by ‘if’ (Deléani 2007):

  • Cypr. Epist/. 4, 3, 1, line 46:
    Et si incorrupta inuenta fuerit uirgo ea parte qua mulier potest esse, potuerit tamen et ex alia corporis parte peccasse quae corrumpi potest et tamen inspici non potest.
    Fr. ‘Et même si une vierge a été trouvée sans corruption en la partie de son corps qui lui permet d’être femme, rien n’empêche qu’elle ait pu pécher en une autre partie, qui peut être corrompue sans qu’on puisse le voir à l’examen.’ (translation S. Deléani, 2007, 105).
    ‘Even if a virgin has been found uncorrupted in the body part by which she can be a woman, …’

This structure with the past passive participle and a fu- form is documented in earlier texts, but Cyprian plays an important role in the development of this construction. Firstly, the previous occurrences of what looks like this construction may sometimes be a fu- form of the verb sum with an adjectivised past participle. Secondly, in Cyprian, the phenomenon is not sporadic (as it is in the previous authors) but it is systematic and regular in a given syntactic environment. Therefore Cyprian gives us valuable information about the precise syntactic and semantic conditions in which the construction developed.

12.2. The dico quod construction

In the morphosyntactic area, Cyprian’s works display a change in the construction of the verbs meaning ‘say’, ‘know, think’ and, more generally, denotating any process involving a psychic activity. Although the classical constructions (with an infinitive clause and an indirect question clause) are still usual in his works, a new construction also appears in a small number of occurrences, with the subordinating conjunctions quod, quia (sometimes quoniam in Biblical quotations) governing a finite verb in the indicative or subjunctive:

  • Cypr. Eleem. 2, 6-7:
    Hic quoque ostenditur et probatur quia sicut lauacro aquae salutaris gehennae ignis extinguitur, ita eleemosynis atque operationibus iustis delictorum flamma sopitur.

    Fr. ‘Ce texte lui aussi montre et prouve que, tout comme par le moyen du bain dans l’eau du salut le feu de la géhenne est éteint, de même par les aumônes et par de justes œuvres de bienfaisance la flamme des fautes commises est étouffée.’
    (translation M. Poirier, 1999)

Even if this construction is only rarely attested in Cyprian, its use had become usual by that time according to M. Poirier (1999: 76, note 1) and J.-C. Fredouille (1992: 517). Moreover, also by the 2nd century AD, this construction shows a good frequency in the Lemmata situated at the end of the Praefatio in Gellius’s work Noctes Atticae (edition R. Marache, 1967, Paris, CUF, 7-8). This construction is also regularly used in the biblical quotations1) from the Vetus Latina in Cyprian:

  • Cypr. Zel. 11, 7:
    et scitis quia omnis homicida non habet in se uitam manentem.’
    Fr. ‘ ‘et vous savez qu’aucun meurtrier ne garde la vie en lui’.’
    (translation M. Poirier, 2008)
    ‘‘and you know that no murderer keeps life within himself’. ’
  • Cypr. Zel. 15, 8-9: Filius Dei … in euangelio suo dicit: ‘audistis quia dictum est: …’
    Fr. ‘le Fils de Dieu …dit dans son évangile: ‘Vous avez entendu qu’il a été dit: …’’
    ‘the Son of God says in his gospel: ‘You have heard that it has been said: …’’

This evolved feature is not only a calque of the Greek construction in the Vetus Latina, but an internal structural change in Latin, whose origin is already visible in Plautus (Cuzzolin 1994; Hoppe 1985: 146). These data illustrate yet again one of the situations where early Latin gives a clue for a linguistic phenomenon that is developed afterwards in late Latin without being documented in classical Latin.

12.3. The indirect reflexive

Cyprian’s work is also a turning-point in the use of the indirect reflexive pronoun and of the adjective suus referring to the speaker. This logophoric system (Fruyt 1987, 2007, 2014) that explained the use of indirect reflexive in early and classical Latin in the indirect style was typologically rare. It is no longer to be seen in Cyprian. By then Latin had already adopted a more common typological system where the speaker is not referred to by a specific pronoun, but by the usual anaphoric or personal pronouns, as in most of the modern Indo-European languages: Engl. he says he will come; Fr. il dit qu’il viendra.

Moreover, in Cyprian heavy syntactic constraints are already established for the use of se as the indirect reflexive pronoun. The pronoun se functioning as an indirect reflexive was maintained only in a specific morphosyntactic environment: a) with the accusative form se, b) functioning as the subject of the infinitive clause governed by the main verb meaning ‘say’, ‘know’, etc.; c) and this pronoun se was situated next to the governing verb (Fruyt 2014):

  • Cypr. Laps. 2, 30-31:
    Explorandae fidei praefiniebantur dies: sed qui saeculo renuntiasse se meminit nullum saeculi diem nouit.
    Fr. ‘On décidait une date à laquelle leur foi serait mise à l’épreuve, mais celui qui se souvient qu’il a renoncé à ce siècle ne sait rien des dates de ce monde.’ (translation M. Poirier, 2012)
    ‘Dates were settled when their faith would be tested: but the man who remembers that he renounced this world does not know anything about the dates of this world.’

These important limitations on the use of the indirect pronoun were linked to a weakening of the notion of ‘infinitive clause’ in the linguistic awareness of the speech community. The se pronoun, situated near the main verb, behaved not as an indirect, but as a direct reflexive pronoun. The sentence no longer contained two clauses - a main and a subordinate clause - with an indirect reflexive pronoun, but one single clause with a direct reflexive pronoun (Fruyt 2014).

Some of the evolved features attested in Cyprian also appear in other contemporary authors and they may even have occurred earlier in the usual, spoken language. As concerns the indirect reflexive pronoun, some passages in Gellius already show the newly emergent evolved system:

  • Gell. 1, 20:
    Nam cum amicus eum rogaret ut pro re causaque eius falsum deiuraret, his ad eum uerbis usus est: …
    Fr. ‘En effet comme un ami lui (Périclès) demandait de faire un faux serment pour son propre (= de l’ami) intérêt et sa défense, il (Périclès) lui répondit en ces termes:
    … ‘.
    ‘While a friend of his (= Pericles) was asking him to commit perjury in his own (=the friend’s) interest and defence, he (= Pericles) answered: …’.

Moreover, some of these evolved features can be traced back to the texts of the 1st century AD (cf. § 15).

12.4. A compromise solution

According to Poirier (2012, 2014), Cyprian was conscious that, by his time, the language of the ordinary people had already started to evolve and to diverge from the language of more literate people. Since he wanted to be understood by everyone (a constant concern of the Christian Fathers) and although he himself was a literate admirer of the classical language, he chose to use only those subordinate conjunctions that everybody in his time would understand. Therefore, he specialised dum and donec into just two of their earlier uses and he avoided quoad.

In Cyprian, dum is used only with the present indicative expressing simultaneity ‘while, as long as’, while donec is used only with the present and imperfect subjunctives with a terminative value ‘until’. He selected only what was acceptable both in the previous periods and in his time. This is an example of a compromise language adopted by a literate author who was well practised in classical Latin. His contemporary readers were probably not aware of this change, since it was only a limitation of the possibilities of classical Latin.

Retour au §11 ou Retour au plan ou Aller au §13

1) The occurrences of the dico quod construction in the Biblical translation of the Vetus Latina do not have the same status as its occurrences in a Latin text, since they are calques of the Greek construction in the New Testament. Therefore, in order to judge the development of the construction in Latin, we cannot build an analysis based on the Vetus Latina.