Early Latin and late Latin

Michèle FRUYT

University of Paris-Sorbonne (Paris 4)



9. Cyclic renewal

Although the gaps in the classical texts mentioned above are largely due to the lack of attestation, we may also be faced with other linguistic situations. Within the continuous phenomenon of diachronic development, there may occur cyclical renewals which create a degree of discontinuity.

Some early (1st century AD) instances of the causative construction facere + infinitive may be analysed as a cyclic renewal of a lexical group of causative verbs denotating natural states and, in particular, temperature such as cale-faciō + accusative ‘to warm up (something)’. The linguistic community analysed these verbs as including two morphemes. The first one cale-, tepe-, frīge- was synchronically associated with the state verbs calēre ‘be warm’ (calidus ‘warm’, calor ‘warmth’), tepēre ‘be lukewarm’ (tepidus, tepor), frīgēre ‘be cold’ (frīgidus, frīgus). These …e-faciō verbs functioned as the causative counterparts of the state –ē- verbs. The second morpheme °-faciō (passive °-fīō, past participle °-factus) functioned as a causative, grammaticalised morpheme (Fruyt 2011-a, 2011-b).

The first lexical morpheme may be separated from facio by other words (feruēfaciō: Cato Agr. 157,9: ferue bene facito; ārefaciō: Lucr. 6, 962: terram sol facit are; perferuēfio: Var. R. 1,9,2: perferue ita fit ut; cōnsuēfaciō: Var. R. 2,9,13: consue quoque faciunt; excandēfacio: Var. R. 3,4,1: excande me fecerunt cupiditate). Therefore synchronically, the speech community analysed the whole sequence as the freezing of two words within a lexicalized syntagm where the two words had a certain degree of autonomy. The first morphemes, cale, ferue, are, etc., were associated with the Latin synchronic radicals1) and lexical bound morphemes cal- ‘warmth’ (occurring in cal-or, cal-idus, cal-ē-re), feru- ‘boil’ (in feru-ē-re, feru-or, feru-idus), ar- ‘drought, dryness’ (in ar-ē-re, ar-idus), etc. But they did not function as radicals, since they were treated as free forms that could be detached and separated from the second element, the facio verb. For the same reason, neither did they function as stems. The first lexical morphemes were probably associated with process nouns or infinitives2) and considered as non-inflected forms of the related verbs, part of an agglutinative process3) of two words within the same lexical item. The nearest morpho-syntactic situation is a verbal periphrasis or a verbal complexeme4) containing a lexical morpheme (cale, ferue, are) and a grammatical causative morpheme (facio, fio). The calefaciō type verbs played a role in the development of the causative construction faciō + infinitive which is the ancestor of the Romance causative periphrasis (Fr. César fait construire un pont; It. fare; cf. Sp.; etc.). The first occurrence of this cyclic renewal is already documented in the 1st century AD in the technical works of Columella (Fruyt 2011-a: 784-785) in the referential domain of temperature:

  • Col. 12,38,5:
    Mel … inferuere facito.
    ‘warm up the honey until it boils’, ‘bring the honey to the boil’.

In the temperature domain, this construction was used later in prescriptive texts, e.g. in Apicius’s cookery recipes:

  • Apic. 3, 78 (ed. J. André: Apicius, L’Art culinaire, Paris, Belles Lettres, 1987, p. 23):
    Feruere facies et infers.
    ‘bring to the boil and serve’.

This feruere facies stands as a renewal of ferue-facies. We may also mention deferuere facies as a renewal of deferue-facies in:

  • Chiron Mulomedicina Chironis 158 (p. 49, l. 10 Teubner):
    haec omnia mixta in uno rursum deferuere facies.
    ‘having mixted all these (ingredients) together, you will bring them to the boil again.’

The calefaciō verbs themselves have shown continuous productivity from archaic to classical Latin and they are still used in late Latin texts.

This cyclic renewal in the vocabulary of temperature has probably played a role in the general development of the facio + infinitive causative construction in late Latin with verbs of various lexical meanings, e.g. with an abstract semantic value:

  • Lact. Inst. 1,22,14:
    illos aetas facit putare quod non est et hos stultitia
    ‘Age makes the first ones believe what does not exist; but for the second ones, this is due to foolishness.’
    Fr. ‘Les premiers, c’est l’âge qui leur fait croire ce qui n’existe pas; mais les seconds, c’est la sottise.’

Retour au §8 ou Retour au plan ou Aller au §10

1) We use the term radical (Fr. radical latin synchronique) here in order to refer to the synchronic Latin semantic minimal unit with a lexical meaning, in opposition to root, used for an Indo-European root (Fr. racine indo-européenne) as a diachronic morphological element.
2) The final –e at the end of the first morpheme is short, but in a few occurrences it is long. This would prove that the whole morpheme itself could have been associated with the present active infinitive of the related state -ē- verb (calēre, feruēre, etc.).
3) In opposition with a compound verb (opitulārī, possidēre, tergiuersārī, auscultāre), a morphological structure inherited by Latin, but which was not productive in Latin (except for the causative °-ficāre verbs : aedificāre, amplificāre).
4) For the word complexeme (a translation of Fr. lexie complexe in B. Pottier’s terminology), see Fruyt 2011, p. 663, note 11. The calefacio verbs would then have a morphological and semantic status similar to Fr. faire peur.