RESUMESMiguel Canas. - Scribonia Caesaris et le stemma des Scribonii LibonesPlusieurs points font debat dans le stemma des Scribonii Libones?: l'identite des deux premiers epoux de Scribonia Caesaris, celle des enfants auxquels cette derniere donna le jour avant d'epouser Octavien, la nature de son lien de parente avec L. Scribonius Libo (RE?20), consul en 34?av. J.-C., enfin le nombre de fils de ce dernier personnage. Une relecture prudente des sources permet de conclure que Scribonia Caesaris eut pour enfants, de l'un de ses deux premiers maris, lequel etait un P.?Cornelius Lentulus Marcellinus, petit-fils par son pere de P. Cornelius Marcelli f. Lentulus (RE?230) et d'une Cornelia des Scipiones Nasicae, la Cornelia de l'elegie IV, 11 de Properce et au moins l'un des deux P. Cornelii consuls respectivement en 18 et en 16?av. J.-C. En outre, l'aspect du stemma des Libones se precise?: Scribonia Caesaris etait soeur de L. Libo (RE?20) et amita des deux fils de celui-ci.Romain Garnier. - Sur la famille du grec ss's Dans l'etude qui va suivre, on se propose de retracer l'histoire de la famille du gr. ss's faire paitre?. Ce groupe d'aspect archaique ne possede aucune parente clairement identifiable au sein de la famille indo-europeenne, et la comparaison tourne court. Cependant, s'il s'avere infructueux de chercher une racine verbale susceptible de rendre compte des faits attestes, il n'est pas exclu de pouvoir depasser l'aporie etymologique que souleve le gr. ss's en admettant une neo-racine d'emergence grecque?: il y aurait eu essor d'un nivellement lexical, avec la disparition d'un suppletisme herite, d'ou la necessite d'une neo-apophonie productive, conduisant a l'illusion d'une racine verbale authentique. Il s'agit de trouver la forme-pivot heritee, sur laquelle se fonde toute la famille du gr. ss's .Pierre Ragot. - Hittite ?assikzi/?assikkanzi ?se rassasier?, grec homerique ?sa?/ e?a? ?id.? et ?d a satiete: reflexions nouvelles a l'appui d'un rapprochement etymologique discuteLe rapprochement du verbe homerique ?sa?/ e?a? ?se rassasier? avec la famille antolienne representee notamment par le verbe hittite ?assikzi/?assikkanzi ?id.? invite a reconstruire en indo-europeen une racine *h2es(-k)-/*h2s-(e)h2. Cette hypothese contraint le linguiste a reinterpreter la forme *h?(-d)-, a laquelle on fait traditionnellement remonter l'adverbe ?d a satiete?, comme un degre zero renouvele d'apres le degre plein *h?- et a postuler, dans la prehistoire du grec, l'existence d'une racine alternante *?h-?-/*?h-?(-d)- dont le grec ancien aurait conserve un vestige dans l'expression formulaire ?de?a? ?d # (E?203).Ulrike Roth. - ?In uilicationem relegauit: Petronius, Satyrica 69.3L'article soutient l'idee que le chatiment de Trimalchio dans Satyrica, 69, 3, consistait davantage en une degradation de ses fonctions professionnelles que dans un transfert dans un autre lieu, comme la campagne.Nathalie Rousseau. - ?essa's t? d t p?ste: ?peninsules syntaxiques? chez les poetes tragiques grecsOn trouve chez les poetes tragiques des adjectifs hypostatiques (crees a partir de syntagmes prepositionnels) accompagnes d'un complement au genitif qui fonctionne en realite comme le complement du substantif qui constitue le second terme de l'adjectif, ainsi que le montre l'etude detaillee de plusieurs passages dans lesquels ces constructions apparaissent. Ce tour rare en grec, et jusque-la inexplique, releve en fait de ce que Cl. Hagege a decrit pour d'autres langues sous le terme de ?peninsularite syntaxique?, et qui connait plusieurs degres d'acceptabilite selon les langues, et selon les locuteurs d'une meme langue. L'analyse precise de la fonction syntaxique des adjectifs hypostatiques presentant une telle construction, en lien avec l'emploi que l'on appelle traditionnellement ?predicatif?, permet d'expliquer pourquoi ce tour, au-dela de son caractere poetique, est principalement represente par des formes hypostatiques, ces dernieres etant susceptibles, au premier stade du passage du syntagme a l'adjectif, d'etre deja pourvues des caracteristiques formelles des adjectifs, mais non encore de leurs proprietes semantiques.ABSTRACTSMiguel Canas. - Scribonia Caesaris and the stemma of the Scribonii LibonesSeveral points of the stemma of the Scribonii Libones are much debated: the identity of Scribonia Caesaris' first two husbands and of the children to whom she gave birth before her marriage to Octavian, the nature of her kinship with L. Scribonius Libo (RE?20), consul in 34?B.C., finally how many sons the latter had. A careful reading of the related sources makes it possible to conclude that Scribonia Caesaris was mother, by one of her first two husbands, who was a P.?Cornelius Lentulus Marcellinus, grandson by his father of P. Cornelius Marcelli f. Lentulus (RE?230) and a Cornelia of the Scipiones Nasicae, to the Cornelia of Propertius' elegy IV, 11 and at least to one of the two P. Cornelii who became consuls respectively in 18 and in 16?B.C. Moreover, the shape of the stemma of the Libones becomes clearer: Scribonia Caesaris was sister to L. Libo (RE?20) and amita to his two sons.Romain Garnier. - About the word family of Greek ss's The following paper is intended to explain the origin of Greek ss's "to feed", which lacks any reliable cognate within PIE, however archaic it seems to be. Although it is unsuccessfull to seek a PIE verbal root which may account for the attested facts, one may elucidate the desperate etymology of Greek ss's by assuming a secondary root, which would have arisen within Greek itself: an innovating lexical levelling would have ruled out the inherited root-suppletivism, so that it became necessary to produce a new ablauting pattern, giving the idea of a genuine verbal root. The point of this contribution is to find out the very inherited form, which the whole family of Greek ss's is based on.Pierre Ragot. - Hittite ?assikzi/?assikkanzi "to be satiated", Homeric Greek ?sa?/ e?a? "id." and ?d "(to eat) one's fill": new considerations in support of a disputed etymological connectionAs it has been convincingly proven that the Anatolian family, reflected especially by the Hittite verb ?assikzi/?assikkanzi "to be satiated", was connected with the homeric verb ?sa?/ e?a? "id.", it is possible to reconstruct the Indo-European underlying root as *h2es(-k)-/*h2s-(e)h2. On this assumption, the scholar is compelled to regard *h?(-d)-, on which there is a general agreement that the adverb ?d "(to eat) one's fill" goes back, as a new zero grade remodelled from the full grade *h?-. He would equally have to concede the existence in prehistoric Greek of an alternative root *?h-?-/*?h-?(-d)-, the remnant of which Ancient Greek could have persisted in the formula ?de?a? ?d # (E?203).Ulrike Roth. - "In uilicationem relegauit": Petronius, Satyrica 69.3The article argues that the punishment of Trimalchio at Satyrica 69.3 consisted more importantly in a downgrading of his professional capacity than in a transferral to another location such as the countryside.Nathalie Rousseau. - ?essa's t? d t p?ste: "syntactical peninsulas" in Greek TragedyIn Greek Tragedy, adjectives derived from a prepositional phrase ("prepositional governing compounds", or "hypostases") are sometimes associated with a genitival complement, which is in fact linked with the substantive forming the second part of the adjective, as shown in the detailed study of several passages where this construction appears. This rare and as yet unexplained turn of phrase exemplifies what Cl. Hagege has described, in the case of other languages, under the name of "syntactical peninsulas"; its receivability depends both on the languages and on the different speakers of one language. A precise analysis of the syntactical function of these adjectival hypostases, in connection with the usually called "predicative use", explains why this turn of phrase, besides its poetic nature, is mainly represented by hypostases. When they are at the first stage of changing from prepositional phrase to adjective, they may already bear the formal characteristics of adjectives, but not yet their semantic properties.