Part IV

STRATA AND
LANGUAGE CHANGE

PAIDEIA, A LINGUISTIC SUBCODE

Henry and Renée Kahane

1.NATURALNESS AND PAIDEIA

Vittore Pisani, taking Italian as the exemplar of a literate language providing excellent material for the abstraction of diachronic principles (1977:132-3), challenges the conventional hypothesis that the evolution of that language from Latin to the modern standard represents a transformation from one monolithic stage into another. Each stage of the linguistic development reflects a complex society; thus he perceives the Latin foundation, socalled Vulgar Latin (to him, a phrase, comparable to, say, the Greek Kaine), as the blending, in principle, of two interlocking forces: the language of the literary tradition of the classi colte as against the colloquial conglomerate of archaic or innovative features with its multilingual, non-Latin ingredients absorbed in the course of history from the many cultures that impinged upon Rome, and later Italy. The intertwining of levels had its impact on language change: in many ways the rules of a systemic development were neutralized. Pisani contrasts, e.g. two variants of the same morpheme: Ital. più, with its regular change of Lat. initial PL, as against its bookish congener, plur(ale), with preservation of PL.

Pisani’s linguistic model calls to mind the analogue of a neighboring discipline. More than sixty years ago, the influential German Literaturwissenschaftler Friedrich Gundolf created the model (Schmitz 1965:64-5) of Urerleb-nis and Bildungserlebnis, primordial experience and cultural experience, to isolate the two forces behind Goethe’s creativity (Gundolf 1920:26-7). Uverlebnis is the impact to which man is exposed by virtue of his inner nature [die Erschütterungen, denen der Mensch kraft seiner inneren Struktur ausgesetzt ist]; Bildungserlebnis is the absorption of the past, of tradition, of education, of the milieu [die geistig geschichtlichen Einflüsse und Begegnisse, schon geformte Anschauungen aus Kunst, Wissenschaft, Religion] (49). The forces, in short, are polar: the one is innate, the other derived [ursprünglich/abgeleitet] (26), yet they are intertwined [verwoben] (27).

It seems tempting to transfer this binary model from artistic creation to language change in literate societies. A construct based on these two roots of our existence as speakers, naturalness (for Urerlebnis) and paideia (as we would like, in the humanistic tradition, to translate Bildungserlebnis), would sharpen our insight into the linguistic process. In their opposition both evolve as essentially diachronic concepts: naturalness represents the rules which are accepted through tradition, paideia the exceptions, often the not-yet rules, which may invade the system. The following tentative remarks center on that much neglected subcode, paideia. Particularly apt to demonstrate its force is a specific constellation recurrent in the history of language: the impact of an external influence. Greenberg (1978:3), searching for universals, underlines its often lasting effect: “there is much empirical evidence supporting the view that external influences play an important role in initiating linguistic changes.”

2.THE GREEK BEHIND THE LATIN

Pisani, in his summary outline of Vulgar Latin, hints at a superb exemplum of linguistic paideia, the impact of Greek on Latin (1977:133). Greek exerted its influence on the Latin of the upper classes, partly through multi-faceted transfer, partly through the prescriptive dogmatism of the grammarians. The presence of “the Greek behind Latin” (the term is borrowed from Paul Friedlander [1943-44:270]) has, of course, been commented on for some time, and the significance of the process has been stressed in persuasive statements which open a wide field of investigation. Thus, the well-documented history of Latin by Leumann-Hofmann-Szantyr (1972:88*) characterizes the final period of the ancient world as “bilingual, displaying a confluence and blending of languages.” On the literary-colloquial level, i.e. that of the literarische Umgangssprache, Dietrich (1973a :20n76) believes that the stylistic mark of the Vulgar Latin Bible translations during the first three post-Christian centuries rested, precisely, on a high degree of Hellenization. For Bonfante (1960:174, 182), Italian is thoroughly “a synthesis of Latin and Greek,” and he sees in the innovations involved not so much a natural development as the reflection of a Greek influence; he thus substantiates Bartoli’s metaphor (1925:44) of the blending of spirito greco and materia latina. Similarly, Coseriu (1971:135) sees in the Greek impact “the central problem of so-called Vulgar Latin, i.e. the foundation of the Romance languages”; therefore he formulates a rule of thumb (141): “if a feature appears in Greek and Romance but not in classical Latin, there are solid reasons to interpret that feature as a Hellenism of late colloquial Latin.” Reichenkron, the historian of Vulgar Latin, considered the role of Greek during the entire development of Latin into Christianity and Romance of such significance and volume that he planned (but death intervened) to devote a besonderes Werk to the topic (1965:23).

In concrete exemplification, the following, without our specifying here the strata of borrowing or the ways of transmission, are a few of the Hellenistic models, on the lexemic, morphological, or syntactical levels, that were supposedly transferred to Vulgar Latin and Romance. The various hypotheses were proferred by Pasquali (1927), Bonfante (1960), Coseriu (1971), or Pisani (1974, 1977).

Greek influence, then, is to be suspected behind the Vulgar Latin article, definite and indefinite; behind the large periphrastic verbal subsystem in the perfect, the future, the “progressive” form; in the strengthening of word-order rules to balance the reduction of inflection; in the comparative and the relative superlative formed with plus; in the increase of diminutive suffixes to replace the simplex; in the great number of borrowed lexemes and of caiques.

We have used here terms such as supposedly/hypothesis/ to be suspected, to indicate the still experimental stage of forays into the field of Graeco-Latin blending. The dilemmas are recurrent: do the similarities result from a common heritage, from the Greek impact on Latin, from the revitalization, with restructuring, of existing patterns, or do they represent a mere parallelism?

Recently, in a review of Dietrich (1973a), we singled out a few of the striking similarities between Greek and Latin as an epistemological exemplar of the pros and cons (Kahane 1978:647-8). (a) Neither Imisch nor Bonfante can decide whether or not to ascribe to Greek influence the Latin replacement of case endings by prepositions, (b) The loss of the infinitive in the Balkanic languages and in Southern Italy is interpreted by Sandfeld-Jensen and RohIfs monogenetically as a caique from Greek, but by Togeby polygenetically as an independent feature. (c) The shift of Latin adjectives to nouns (say, discentes ‘disciples’) is seen by one as the imitation of a Greek model, by another as influence from Latin poetry. (d) The frequent irregularity of intertonic vowels in Tuscan has been explained traditionally as an indigenous Italian development, resulting from phonological distribution, dialect conditions, or language levels; Malkiel, however, derives the vocalic fluidity from the effect of colloquial Hellenisms which then drew terminal word segments into its domain. (e) Above all, the emerging article has provoked varying explanations. Trager sees in its expansion a reflection of the democratizing spirit of expanding Christianity; Löfstedt believes it was an internal linguistic process, caused by the weakening of the pronouns and the collapse of the inflectional system. On the other hand, the Greek contribution to the rise of the article is stressed by Wartburg, who discerns in its presence or absence a new way of differentiating concreteness or ab-stractness? but both Rohlfs and Lausberg find the Greek role to be no more than the reinforcement of a preexisting Latin device of identification.

As even these sketchy remarks have indicated, the Greek linguistic influence in Vulgar Latin was, in a certain sense, a diastratic one: it came from and enveloped the upper and the lower levels of society and survived on the basic levels of Romance, the inherited and the “learned.” Here, however, within the frame of this essay, Greek influence will be viewed as a contribution to Western paideia, and in the following we shall try to describe briefly five typical channels of transmission which mediated Greek elements to the speech of the educated elite. In other words, we are interested here in the sociolinguistic act of the transfer.

3.LITERATURE

Observers in antiquity were already aware of poetry as a major channel of Grecisms. Janssen (1941:115) adduces two authoritative testimonia. Quintilian, the rhetorician and critic (1st c.p. Chr.), characterizes the esthetic attraction of Greek for the Latin writer of poetry (Inst, or., XII, 10.33 tr. Butler): “The Greek language is so much more agreeable in sound than the Latin that our poets, whenever they wish their verse to be especially harmonious, adorn it with Greek words.” Horace expresses the same view essentially when he rejects linguistic Hellenomania as inappropriate for the factual jargon of the bar; in doing this he uses the analogue of the sweet Greek and the dry Roman wines (Sat. I, 10.20 tr. Fairclough): “‘But that was a great feat, you say, his [Lucilius’] mixing of Greek and Latin words ... a style, where both tongues make a happly blend, has more charm, as when the Falerniari wine is mixed with Chian.’ In your verse-making only (I put it to yourself), or does the rule also hold good when you have to plead the long, hard case of the defendant Petillius?”

Kroll (1924), Janssen (1941), and Leumann (1947) (all three now available in Italian translation, updated bibli-ographically, in Lunelli 1974) have gathered a considerable corpus of Hellenisms in Latin poetry. Kroll (6) stresses the function of poetry as the early channel of transmission; Leumann (157), its special role in the transfer of syntactical Grecisms. Janssen (108), viewing the problem from a systemic standpoint, observes, quite in agreement with what we quoted from Pisani, that “the syntactic deviations which occur in Latin poetry are often those resulting from Greek influence.” In the following we shall try to sketch, within this particular frame, the history of one morpho-syntactical feature, the adverbial formans -mente, from Greek through Latin into Romance.

Shorey (1910) was the first to be struck by the similarity between the Romance adverbial -mente and a formulaic Greek dative phrase frequent in classical drama which consisted of a noun plus adjective with instrumental function: eudóxōi phrení (Aeschylus) prefigured a Latino-Romance gloriosa mente ; aphóbōi phrení (Aristophanes) an intrepida mente. Grk. phrení ‘with the mind’ was easily replaced by quasi-synonyms, all in the dative: thymôi/ nóōi/gnṓmāi/kardíal/psýchāi. The range widened to less spiritual parts of the body such as cherí dat. ‘hand’ and podí dat. ‘foot’. Semantically, the nouns shared the element of a human involvement; which nuance they represented was not relevant. As Shorey points out (88), the noun functioned as the carrier of the adjective, and the latter was the dominant feature of the nexus.

The Greek expression was paralleled in Latin with, of course, a shift from dative to ablative, as in honesta mente (Seneca) / timido pectore (Plautus). Again, the noun functioned through its case ending as the grammatical, the adjective as the semantic, marker of the string. McCartney (1920) lists numerous records, excerpted for the most part from belles-lettres. The similarities of structure, the semantic range of the nominal element, and the literary use suggest the Greek formula as a model of the Latin, and indeed, various investigators have advocated this genetic relationship: Kroll (1923:275), Bartoli (1925: 88), Vendryes (1950:197-8), Battisti (1954): s.v. mente.

The broad spectrum of ablative nouns sharing the concept of mental disposition was reduced to a single item, mente, and by the sixth century the Romance use of adverbial -mente was in existence (Mihäescu ap. Spitzer 1940:189). Two aspects of this final phase have been stressed. In grammatical terms the development is one from an analytic to a synthetic stage: Baldinger (1965: 709) thus interprets the transition from severa mente ‘dans un esprit sévère’ to sévèrement ‘d’une manière sévère’; Malkiel (1978:128) terms the shift one from composition to derivation, most evident in French where the morpheme ment has lost its lexical autonomy. In semantic phrasing, Lausberg (1972:100) sees in the adverbialization of mente the mechanization of a former psychological content, the disposition of the agent: in, say, l’eau coule doucement human involvement is no longer present; Mihăescu (ap. Spitzer 1940:190) and Spitzer (ibid, and 1925:287) stress the educated, intellectual foundation and connotation of the pattern.

4.TRANSLATION

Translation not only transmits the substance of a different culture, but it also carries some impact of the concomitant language; in other words, the source language remains a reality behind the target language. In the long and multifaceted history of translation certain periods stand out in which the glory of the source language was even cultivated; included among these was that of the early translation of Christian literature from Greek into Latin. The word had a weight of its own, competing for preeminence with the meaning of the text: we are referring here to the debate between the two principles of verbum de verbo and sensus de sensu (Marti 1974:64). Literalism transcended, of course, the merely lexemic, and syntactical features were drawn into the process. The popular character of such literature as the New Testament involves an interplay of levels which is of the greatest interest to the history of language: while the translator enjoys a certain measure of education, the public for whom he translates includes everybody; thus, the newly coined or reawakened or restructured forms are assured a wide and continued reception and retention.

Linguistic change, which is stimulated through translation, chiefly Biblical translation, is, in a schematic description (Kahane 1978:647), a three-step process: First, a bilingual who partakes in a Mischkultur (in the present context, the Graeco-Latin) weaves into his translation such patterns of the source language as are matched by similar forms in the target language, yet deviate in certain ways from the norms of the latter: they may occur less commonly, convey divergent semantic shades, represent a different function within the system, or even be just barely understandable. In step two, the mediator perceives the finished product with his eye and transmits it by word of mouth to the listener’s ear. Third, the listener is continuously exposed, for generations, to the phrasing of the teachings and of the stories, and the former deviation may turn into the accepted standard.

Dietrich (1973a and b) has extensively treated one case in point, the aspectual periphrase, to be + present participle. He evidences the structure for Greek, from Homer to the Byzantine period, as a device of the narrative and descriptive style, e.g. Herodotus (VIII, 137.4) ên gàr ton oîkon eséchōn ho hḗlios ‘for [just then] the sun was shining into the house’. Synchronically, Dietrich sees in the periphrastic paradigm, as against the unauxil-iaried verb, a pattern marked for aspect; he calls it, with Coseriu, Winkelschau, referring to the speaker’s ability to perceive, from his standpoint, a segment (rather than the totality) of an act; in his German rendering of the ancient records he frequently uses an adverb of involvement, gerade dabei sein ‘to be about to/be on the verge of/just happen to’ (thus Mk. 2:18 erant ieiu-nantes ‘they were keeping a fast’). We are less sure than Dietrich seems to be (1973a:195) that the English progressive form am (sing) ing, with its stress on circumstantiality, is not closely akin in function to his Winkelschau. Diachronically, the periphrastic aspect was a tradition in Greek but was alien to Latin where the form was known, to be sure, but the function apparently different: Plautus’ sum oboediens ‘I am obedient’ represents the copula relating a predicative to the subject. With the period of increasing Greek influence and through the translation of Biblical literature, i.e. from about the second post-Christian century on, the Greek pattern superimposed itself on the existing Latin form and in the process introduced a new function. Through Vulgar Latin this function became a feature of the Romance verbal system (so Span, estoy oantando) as well as, possibly, of the Germanic (Kahane 1978:647): Mt. 7:29 ên didáskōn ‘was teaching’ → Lat. erat doeens → OE wees lœrende.

Two of Dietrich’s conclusions support the basic hypotheses of this essay: on the one hand, that the transfer of the Greek periphrastic aspect into Vulgar Latin and beyond was realized above all within the course of Bible translations [besonders im Zuge der Bibelübersetzungen] (1973a:227), i.e. on the level of paideia; on the other hand, that the transfer introduced a deviatory syntactical feature into Vulgar Latin: it had not existed within the norms of either the older phases or of the written level of Latin (224).

5.SYMBIOSIS

A bicultural and bilingual symbiosis is a most favorable environment for the transfer of linguistic features. Often, in such an environment, one of the two cultures, with its language, is dominant, and its styles of life and of speech are apt to filter first of all into the upper stratum of the dominated society. In the present context that situation is exemplified by the Byzantine Exarchate of Ravenna, where, from the middle of the sixth to the middle of the eighth century, Vulgar Latin came under the influence of Hellenism (Kahane 1970-76:440-2). The influence was chiefly lexical, with Greek words flowing into the Latin of that time and that particular region. A circumspect essay by Sylviane Lazard (1976) elucidates the patterning of the process. She registers the Grecisms used by a single man, the Abbot Agnello, author of the Liber Pontificalis Ecclesiae Ravennatis (c. 827-840). Agnello was born in 805, about two generations after Byzantium retreated (in 751) from Ravenna, and thus was probably a good representative of the speech of that episode of biculturalism. He came from a family in which knowledge of Greek was a tradition, and he liked to play (an intellectual hobby which is indicative of his educational status) with the etymologies of Greek words. Lazard sees in him a man of culture and a man who was, to a certain degree, bilingual. His Hellenism reflects (298) “the usage of the educated class of Ravenna at his time.”

The following samples of Agnello’s Hellenisms focus on what is typical in this kind of linguistic and thereby cultural transfer. The lexemes selected had to be truly Ravennate: either they can reasonably be diagnosed as adapted during Ravenna’s Byzantine phase; or, if they were remnants of an earlier period of Latinization, they display some specific feature which ties them to Ravenna: a shade of meaning, a phonological peculiarity, or continued presence in the regional dialects. We shall follow Lazard’s analysis (always indicated), with certain expansions and retouches.

arohiergatus ‘foreman (in a crew of workers)’ < archiergátēs, a Greek term of the period of the Exarchate (290).

ardica ‘narthex of a church’ < (n) árthēka, acc. of nárthēx, with loss, in Greek, of the initial in synizesis with the article; the reduced form attested from Byzantine to modern Greek; the Ravenna variant, ardioa with d, surviving regionally into the Middle Ages, represents a later, voiced stage of artica, likewise recorded for Ravenna, with rendering of Grk. θ as t (268-9; MLatWb, s.v. ardica; Kahane 1970-76:367-8).

argýrion ‘silver, silverplate’ < argýrion ‘silver’, a sheer Hellenism in the West, even recorded in Greek characters in the 9th century; in Greece the same meaning is preserved in such marginal areas as Crete and Cappadocia (290; MLatWb, s.v. argyrion; Kriaras, s.v.; Andriotis, #1147).

bisalis ‘brick’ < bḗsalon; a Greek Latinism of the Koine, still in use; the i of the stem, rendering the iotacistic pronunciation of Grk. n, indicates the Greek provenience of Agnello’s term (290-l; MLatWb, s.v. bessalis; Liddell-Scott, Suppl., s.v. bḗsalon; Kriaras, s.v. bḗsalo [n]).

cereostatus ‘candelabrum’ < kērostátēs; the ecclesiastic term, which appeared from the 6th century on, and above all in Rome, must have been popular in Ravenna as indicated by the surviving regionalism zilostar (269; MLatWb, s.v. cerostatum; Kahane 1970-76:359).

chartularius ‘keeper of the archives of the court’, then an honorific title < chartoulários, itself borrowed from Latin; the term is repeatedly attested with reference to the Byzantine administration of Ravenna (270: Lampe, s.v. chartoularios; Niermeyer and MLatWb, s.v. chartularius; Kahane 1970-76:512).

cherumanica ‘glove, sleeve’ < cheirománikon, attested from the 6th to the 10th century (and so listed in the Etymologicum Magnum, 209.40 Gaisford), a compound of the Latinism maníkion (from manioa ‘sleeve’) and Grk. cheír ‘hand’ (291-2; DuCange, s.v. maníkion; Sophocles, s.v. cheirománikon).

diplois, acc. diploidem ‘kind of cloak of the clergy’ < diplois, -idos ‘cloak affording a double wrapping’; a borrowing well-attested in Late Latin, whose regional records, as late as the 16th century, in Ravenna, Emilia, Senigallia indicate a continued regional use (272; Lampe, s.v. diplois).

dooarium ‘frame of roof’ < dokárion ; the Greek term recorded since 7th-8th-c. papyri and still in use (292; Kriaras, s.v. dokário(n); Niermeyer, s.v. docarium, with a 10th-c. reference).

dromo, -onis ‘a fast vessel’ < drómōn ; recorded in the West by the 5th century and in particular since Theodoric, i.e. in some kind of relation to Ravenna (272; Kahane 1970-76:363).

endothis ‘altarcloth’ < endytē ‘garment, altarcloth’; the term was apparently borrowed in Ravenna in its ecclesiastic use and remained alive there, in its corrupt form, as late as the 11th century (272; Lampe, s.v. endytḗ; Koukoules 11:2,27).

exarohus ‘the exarch of Ravenna’ < éxarohos ‘exarch, viceroy’; the term refers to a specifically Ravennate institution; later this title of a worldly dignitary was transferred to a spiritual one, the archbishop of Ravenna (273; Lampe, s.v. éxarchos; Niermeyer, s.v. exarchus).

glossocomun ‘coffin’ glōssókomon-, in Greek, the meaning ‘Sarcophagus’, an extension of the meaning ‘box, casket’, survived into the Middle Ages, particularly in hagiographie texts (292; Koukoules IV, 190; Lampe, s.v. glōssókomon).

graphia ‘inscription’ < graphía ‘writing, treatise’; the Greek term, recorded in Gregorius Nyssenus (4th c.), was borrowed as ‘(piece of) writing, description, document¹ and quoted as a sheer Hellenism by Isidor of Seville, Orig., 6,9,2 (7th c.) (292-3; Lampe, s.v. graphia; CGL III, 131.49/495.55/511.71; ThLL, s.v. graphia; Niermeyer, s.v. graphía).

manuale ‘candlestick’ < manouálion; the latter a Greek Latinism, from manuale ‘thing held in the hand’, which in Greek was applied to an ecclesiastic object and is still in use (293-4; Lampe, s.v. manouálion).

molohus ‘bolt of a door’ < mochlós, with metathesis; the ancient term survives in such modern Greek dialects as Chios, the Cyclades, and the Dodecanese (294; Andriotis, #4128).

Orphanumtrofium, name of a monastery dedicated to St. Peter orphanotropheîon ‘orphanage’, a neologism of the Justinian period; with the passing of time the name of a Ravennate establishment, which in all probability was founded by Greek monks, lost its meaning and its form: by the 11th century it appears as Offeotrofeum (275).

platavum ‘a kind of vessel’ < pláthanon ‘dish or mold for baking’ recorded in anc. Greek; a Latinized fem. form, *plátana, either derived from the neuter plural pláthana or from a Greek fem. plathánē, is the base form of numerous regional names of vessels spreading from Ravenna (294; Henricus Stephanus, s.v. pláthanon; Kahane 1970-76:391).

Hellenisms such as these exemplify the lexical impact which Byzantium’s colonial government exerted on the speech of the educated citizen of Ravenna. In a semantic resume the borrowings, many of which were neologisms, referred to objects and institutions which may have been new or at least fashionable. The ecclesiastic terminology is characteristic: the clerical cloak, altarcloth, candlestick and candelabrum, and the coffin all have Greek names. The building trade is well represented: frame of the roof, brick, bolt on the door, the architectural feature of the narthex, and the foreman of the crew. Amenities of daily life: glove (or sleeve?), silverware, and baking dish. A new type of ship: the fast dromon. And, of course, the administrative contribution of State and Church: exarch, chartulary, inscription, and orphanage. As to phonology, Agnello’s spelling may echo realistically the actual pronunciation of the Byzantine model: thus, the iotacistic value of Grk. ŋ is reflected in bisalis and in ardica; and, again, ardica reflects the loss, through syntactical phonetics, of initial n in its base form (n) árthēka. In diachronic terms, some of the words were sheer Hellenisms such as argyrion and diplois. Others, such as cereostatus and platanum had popular appeal and survived in the regional dialects. Some, interestingly, were rückwanderer: Latin words that had been borrowed by Greek and now returned, sometimes with semantic restructuring, as Hellenisms: bisalis, chartularius, manuale, and cherumanica, from the hybrid cheirománikon half Greek (cheir) half Hellenized Latin (manikon) .

6.SPECIAL LANGUAGES

With the terminology of a particular area of knowledge the translator faced an often difficult and sometimes even creative task. In the Empire, several professions had a tradition anchored in the Hellenic culture, and the jargons of the various linguistic fields were transferred into Latin in three distinct patterns: direct borrowings or caiques or transformations. Transformations concern us here. Theology, that most technical field of all, where terminology assumes a reality and a vitality of its own, exemplifies the process. Studer (1971), who devotes a fine essay to the interplay, in this event, of source language and target language, stresses a characteristic feature (190): “The transfer of Greek concepts always involved an Umdeutung, a reinterpretation, resulting from the fact that Greek concepts and notions were being transplanted into a new environment.” The following case histories, often discussed, are reported grosso modo after him. They represent three types of such transmission.

(a)Grk. dóxa and Lat. gloria (182-4). The translations of dóxa illustrate a case of polyvalence, Latiniza-tion of a Greek term with a multiple semantic load. After such early renderings as claritas/maiestas/honor, the term gloria was in the ascendant, supported by its profane background, ‘glory, fame, honor’, which corresponded closely to the traditional meanings of dóxa in Greek. In the religious context both dóxa, the model, and gloria, the translation, developed in common such uses as ‘divine power’ and ‘praise of God’. But when dóxa was drawn increasingly into the semantic orbit of light and radiance, gloria did not follow all the way. The reasons are not entirely clear: perhaps gloria, with its connotation of power, was less apt to express the notion of light; perhaps the lexemic competitors gloria/maiestas/claritas/ honor were differentiated and used according to the shade of meaning which each member of the set expressed. Claritas , in any case, continued to be used as the term for the notion of light. Interestingly, the polysemy of doxa as reflected in its renderings had already motivated debate even in antiquity; St. Jerome had to defend his use of terms other than gloria (and gloriatio); similarly, Augustine felt impelled to stress the equivalence of gloria/maiestas/claritas. Both justified themselves with an appeal to consuetudo, the linguistic norm of the Church.

(b)Grk. prósōpon and Lat. persona (184-8). In this example, an early equivalence of the two terms was voided through semantic bifurcation; the Latin member shifted, in contrast to the Greek which was static. Greek Biblical theologians introduced the “prosopographic exegesis”: to demonstrate the difference between Father and Son they identified and differentiated the ‘persons’ to whom the various sayings were ascribed. This method was taken over by the earliest Latin theologians, Tertullian and Novatian, who used the expression ex persona, patterned after Grk. ek prosṓpou. Thus, the term persona came into the dogma of the Trinity, and with Tertullian it became a central concept of this dogma. Then, model and translation bifurcated: In the East, Grk. prósōpon remained restricted, as an exegetic term, to the differentiation between the divine hypostases, but in the West, Christian persona under the influence of its use in the Roman tradition, took on the concrete meaning of ‘personality’, body and soul, and thus subsumed Christ’s divine and human attributes.

(c)Grk. mystḗrion and Lat. sacramentum (180-1). Here, the primary meanings of model and translation were incon-gruent; yet their secondary uses apparently coalesced. In Pauline literature, mystḗrion denoted the secrets of God now revealed in Christ, and secrets known only to the initiated, and in a formula such as mystḗrion tês písteōs (I Ti. 3:9) simply ‘faith’. But mystḗrion still retained something of its pagan past, the mysteries, and its immediate Latin counterpart, mysterium, a borrowing of long standing from Greek, might therefore have been unattractive as the label for the new acts of faith. However, in the early Church the term mystḗrion concretized in the direction of the ‘sacramental’ (Bornkamm 1942:8 32, Mohrmann 1954:149, Braun 1962:437), thus opening possibilities for Latinization. Lat. sacramentum, on the other hand, had been an expression of law in the classical language; it referred to an oath of allegiance, then to a solemn obligation, and with this connotation the term might have become applicable to some ritual of faith. Sacramentum turned into the standard rendering of mystèrion. But sacramentum, under the impact of its etymological root, sagr- ‘holy’, took on uses which went beyond those of mystḗrion and evoked an association with holy things, drawing the Latin term into a new linguistic field. Thus holiness became the dominant feature of the sacramentum, as against the notion of the unspeakable and the secret, which was dominant in Grk. mystḗrion.

7.METAPHOR

The metaphor shifts the meaning of a lexeme (or a string of lexemes) from its usual linguistic field to another, and thus represents an act of deviation: the rules of contextual collocation have been violated. Now, Weinrich (1976), in a most interesting and wide-ranging essay, views the metaphor in the Western languages as a characteristic feature of Western civilization, and even defines the latter as Bildgemeinschaft, a community sharing the metaphorical tradition; he thereby parallels the hypothesis of the tradition of Western literary topoi, splendidly proferred by Ernst Robert Curtius. Somewhere, sometime, of course, the deviatory process of a metaphori-zation which by now has become a commonplace in innumerable realizations within the Western linguistic culture, must have had its origin. In view of the paramount role that Greek has played in the intellectual development of this culture, it seems tempting to probe in that direction. Greek metaphor is a broad and often investigated area. Therefore, we shall try to trace the tradition of metaphorization from Greek through Latin into modern times as exemplified by English, within the specific linguistic field of rhetoric and literary criticism, as described in Van Hook (1905). In individual instances it may be hard to distinguish between a monogenetic and a polygenetic origin; but we are simply trying to present the contours of a tradition.

The metaphorizing level (bildspendend to Weinrich) is the surface; the metaphorized (bildempfangend), the underlying structure. The metaphorized field, in the following sample, is always some aspect of literary judgement; the metaphorizing is taken from the every-day experience of man: nature/the human body/social status/arts and crafts. We have selected examples in which the metaphor is still reasonably alive, i.e., we have avoided learned borrowings, either from Greek to Latin, or from Latin and French to English. The references are to Van Hook.

(a) Nature, with reference to water, heat and cold, light and darkness, weight and height, and flora. Gr. katharós lit. ‘clean, clear’, met. ‘clear, lucid’/L. purus/ E. clear (12): Gr. tholoûsthai lit. ‘get turbid’, met. ‘get muddy’/L. lutulentus adj./’E. muddy adj. (12); Gr. reîn lit. and met. ‘flow’/L. fluere/ E. flow, with the adjectival set Gr. eúrous/ L. pulchre fluens/ E. flowing (13); Gr. thermós lit. ‘hot’, met. ‘vehement’/L. oalidus/E. hot (14); Gr. psychrós lit. ‘cold’, met. ‘without ardor, insipid’ /L.frigidus/E. cold (14); Gr. lamprós lit. ‘bright, shining’, met. ‘brilliant’/L. splendidus/E. shining (15); Gr. skoteinós lit. ‘dark’, met. ‘obscure in meaning, hard to understand’/L. obscurus/ E. dark (15); Gr. embrithḗs lit. and met. ‘weighty’/L. gravis /E. weighty (16); Gr. hypsēlós lit. ‘high’, met. ‘lofty’/L. sublimis/E. lofty (16); Gr. anthērós lit. and met. ‘flowery’/L. floridus/ E. flowery (17).

(b) The human body, with reference to physique, looks, taste, sickness. Gr. ischnós lit. ‘thin, lean’, met. ‘unadorned’/ L. gracilis, exilis/E. meagre (19); Gr. hadrós lit. ‘bulky, strong’, met. ‘powerful’/L. arnplus/ E. full, powerful (19); Gr. kállos lit. and met. ‘beauty’/L. pulchritudo/E. beauty (20); Gr. rṓmē lit. ‘bodily strength’, met. ‘force’/L. vis/ E. force (20); Gr. xērós lit. ‘dry’, in reference to bodily conditions ‘withered’, met. ‘arid, austere’/L. aridus, siccus /E. dry (21); Gr. pikrós lit and met. ‘pungent, bitter’/L. amarus/ E. bitter (28); Gr. oideîn lit. ‘swell’, met. ‘be inflated, bombastic’/L. turgere/ E. swollen (35); Gr. chōlós lit. and met. ‘limping, halting’/L. claudus/ E. limping, halting (21) .

(c) Social status. Gr. dēmṓdēs lit. and met. ‘popular’/L. Vulgaris/E. folksy (27); Gr. tapeinós lit. and met. ‘low’/L. humilis/E. low (27); Gr. ptōchós lit. ‘beggarly’, met. ‘poor’/ L. inops/E. poor (28); Gr. ploúsios lit. and met. ‘rich’/L. opulentus/E. rich (28).

(d) Arts and crafts, with reference to harnessmaking, weaving and embroidery, carpentry, metalworking, masonry. Gr. achálinos lit. ‘unbridled’, met. ‘unrestrained’/L. infrenis/E. unbridled (25); Gr. hyphaineín lit. ‘weave’, met. ‘compose’/L. texere/ E. weave (35); Gr. syneírein lit. and met. ‘string together’/ L. connectere/E. string together (37); Gr. leîos lit. and met. ‘smooth’/L. levis/E. smooth, with the verb Gr. leaínein lit. ‘smooth’, met. ‘polish’/L. polire/E. smooth (38); Gr. trachýs lit. and met. ‘rough’/L. asper/ E. rough (38); Gr. sphyrḗlatos lit. ‘wrought with the hammer’, met. ‘put in shape with intellectual effort’/L. malleatus/E. hammered (39); Gr. hýlē lit. ‘timber for building, the stuff of which a thing is made’ met. ‘subject for a writing’/L. materia/E. stuff (41); Gr. kanṓn lit. ‘mason’s rule’, met. ‘standard’/L. regula/E. yardstick (41).

8.CONCLUSIONS

We have outlined the Hellenization of Latin (essentially Vulgar Latin in late antiquity) as characteristic of linguistic paideia, a non-inherited subcode cultivated by the intelligentsia, the upper classes, and the professionals who built it into the indigenous system of their language. The case history of the Greek impact on Vulgar Latin reveals the contours of the four basic facets of linguistic paideia: the media which typically channel such linguistic features; systemic deviations within the target language, which are represented by these features; the frequent naturalization of such high-level features in the standard language; and the element of universality inherent in the entire process.

(a)The channels. There are customary ways in which the educated absorb a prestige language, whether a high level of their own tongue or a fashionable foreign one. We have isolated here five of such channels of external influence: the literary text, which involves the learner on all levels of training in letters (§ 3, above); translation, the impact of which implies in Bible translation, the case under discussion, a chain from the translator to the mediator to the listener (§ 4); symbiosis, the optimal condition for linguistic transmission, welcomed most eagerly, of course, by the upper strata (§ 5); special languages, whose transfer presupposes a professional acquaintance with practices and concepts easily varying from the milieu of the source to that of the target (§ 6); and metaphors, the constant stimulus to intellectual recreation (§ 7).

(b)Deviations. The many features of the high-level subcode which we have labeled paideia are likely to clash, at least at their inception, with the inherited, “natural” rules of a system as generally practiced. The examples cited in the course of the discussion illustrate varied and typical rule-violating innovations: The adverboid ablative phrase based on mente with its synonyms (§ 3) has been interpreted convincingly, because it adds the feature of motivation to the concomitant verbal action, as a Greek-induced innovative category of intellectual and educated foundation and connotation.—The periphrastic pattern to be + pres. part. (§ 4), common in late Greek, coalesced with a Latin expression alike in form yet differing in function; and by drawing the Latin form into the aspectual paradigm of the Greek contributed to a restructuring of the system of Latin verbal categories and established a new stylistic device of great significance.—The influx of lexical borrowings (§5) is easily the most obvious and the most concrete reflection of contacts with a prestige language; and the more intensive the symbiosis the more extensive is the borrowing. In the early stages, in particular, the foreignisms are only slightly disguised by phonological and morphological adaptation; thus they keep their foreignness within the general lexicon and remain fashionable precisely because of their foreignness.—The transformation of professional terms from Greek to Latin in a discipline as specialized as early Christian theology (§ 6) throws light on the innovative facet of the process. Typically, the lexemes of source language and target language are incongruent in part, and both have a life of their own; therefore in the course of blending the ingrained connotations of a term of the target language may break through, thus deflecting the concept of the source language into a new track.—The metaphor (§7) is deviatory by definition: in the incipient stage of the process, lexemes shift from their ordinary collocation to an unexpected one, usually with an inherent loss of con-creteness. If the metaphors are calqued after the model of another language, as was apparently frequent with Greek figures in rhetoric and literary criticism transferred to Latin, the semantic deviation evolves as a typical feature of linguistic paideia.

(c)Naturalization. We have contrasted naturalness and paideia within a diachronic frame of reference: we have termed inherited linguistic features natural as distinct from acquired ones, which filter in with the flux of linguistic paideia. Clearly, the opposition is most valid at the early stage when features stemming from paideia are apt to be deviatory, representing so-called exceptions. They may, of course, disappear altogether, or they may be incorporated as part of the regular system. Thus, the quasi-autonomous and psychologically loaded mente was mechanized into the adverbial suffix -mente (§ 3).—The new periphrastic pattern, to be + pres. part. (§ 4), restructured the aspectual system of Latin by adding a weighty stylistic device, the isolation of a segment of the action; this was inherited by Romance, as in Span, estoy (canta) ndo.—The mass of Hellenisms flooding into Vulgar Latin (§ 5) represented, as often in such cases, a status symbol of the period; many were just briefly fashionable and went out of use; yet (to stay within our Ravennate sample, a late echo of the Justinian era) some of the borrowings caught on, “trickled down” and remained alive in the dialects of the region, e.g. ardica ‘narthex’, cereostatus ‘candelabrum’, platanum ‘vessel’.—As to the transfer of special languages (§ 6), the tie between model and caique may weaken; in such a case the equivalent of the target language develops its own conceptual offshoot, as shown by the constituent of ‘holiness’ in sacramentum, which replaced the ‘mystery’ of its Greek antecedent, mystḗrion.—The tie between the Greek foundation and our own practice as English speakers is in our intuition least noticeable in the linguistic field of the metaphor (§ 7): our English metaphors seem so natural to us that we refuse to see in them anything but our own creativity: a style is, say, dry because we feel a lack of juiciness, and as speakers we are inclined to rebel against an unbelievable past involving the chain from Grk. xērós through Lat. aridus to our so obvious figure of speech.

(d)Universality. The typology of paideia, the Bildungserlebnis, which we abstracted from the case history of the Greek behind Latin, describes a linguistic event many times repeated and more significant and powerful than ever in our times. Paideia evolves as the universal subcode of the educated, and this subcode functions, on the one hand, as a status symbol for the speaker and, on the other hand, as the speech community’s symbol of modernization. This sociolinguistic force of paideia has its linguistic correlate: the features of the subcode, most obviously in the early stage of their use, are often incongruous with the system of the target language.

We have viewed the relation between the two linguistic levels from varying angles: hierarchically, as code and subcode; structurally, as rule and exception; diachronically, as inherited and acquired; and sociolinguistically, as naturalness and paideia. Spitzer (1925:283) couches what we call paideia in terms of attitude: he speculates that “the speaker simply wants to take on a second nature, that of the educated person.”

REFERENCES

Andriotis, Nikolaos. 1974. Lexikon der Archaismen in neugriechischen Dialekten. (Österr. Akad. d. Wissensch., phil.-hist. Kl.: Schriften der Balkankommission, Linguistiche Abteilung, 22). Wien.

Baldinger, Kurt, 1965. Art. mens. W.v. Wartburg, Französisches etymologisches Wörterbuch, VI : 1, 708-9. Basel.

Bartoli, Matteo. 1925. Introduzione alla Neolinguistica. (Biblioteca dell’Archivum Romanicum, ser. II, vol. 12). Genève.

Battisti, Carlo. 1954. Art. mente. Battisti und Alessio, Dizion-avio etimologico italiano, 1950-57, IV:2425. Firenze.

Bonfante, Guiliano. 1960. Les rapports linguistiques entre la Grèce et l’Italie. Hommages à Léon Herrmann. (Collection Latomus, 44:171-82.) Bruxelles-Berchem.

Bornkamm, G. 1942. Art. mystḗrion. G. Kittel, Theologisohes Wörterbuch zum Neuen Testament, IV:810-34. Stuttgart.

Braun, René. 1962. “Deus Christianorum”: Recherches sur le vocabulaire doctrinal de Tertullien. (Publications de la Faculté des Lettres et Sciences Humaines d’Alger, 41.) Paris.

CGL = G. Loewe and G. Goetz, Corpus glossariorum latinorum, Leipzig and Berlin, 1888-1923.

Coseriu, Eugenio. 1971. Das Problem des griechischen Einflusses auf das Vulgärlatein. Sprache und Geschichte: Festschift für Harri Meier, 135-47. München.

Dietrich, Wolf. 1973a. Der periphrastische Verbalaspekt in den romanischen Sprachen. (Beihefte zur Zeitschrift für romanische Philologie, 140.) Tübingen.

——. 1973b. Der periphrastische Verbalaspekt im Griechischen und Lateinischen. Glotta 51:188-228.

Friedlander, Paul. 1943-44. The Greek behind Latin. Classical Journal 39:270-7.

Greenberg, Joseph H. 1978. Introduction. Universals of human language, ed. by Joseph H. Greenberg, 1:1-5. Stanford, CA.

Gundolf, Friedrich. 1920. Goethe. Berlin. [The first edition 1916]

Janssen, Hendrikus Hubertus. 1941. De kenmerken der Romeinsche dichtertaal. Cited after the Italian translation, Le caratter-istiche della lingua poetica romana. Lunelli 1974:67-130.

Kahane, Henry and Renée. 1970-76. Abendland und Byzanz: Sprache. Realleorlkon der Byzantinistik, ed. by P. Wirth, 1:345-640. Amsterdam.

——. 1978. Rev. of Dietrich 1973a. Romance Philology 31:644-8.

Koukoules, Phaidon. 1948-57. Byzantinôn bíos kaì politismós. 6 vols. Athens.

Kriaras, Emmanuel. 1968. Lexikò tês mesaiōnikês hellēnikês dēmṓdous grammateías : 1100-1669. Thessalonike.

Kroll, Wilhelm. 1923. Rev. of McCartney 1920. Glotta 12:275.

——. 1924. Die Dichtersprache. Cited after the Italian translation, La lingua poetica romana. Lunelli 1974:1-65.

Lampe, G.W.H. 1961. A patristic Greek lexicon. Oxford.

Lausberg, Heinrich. 1972. Romanische Sprachwissenschaft, III: Formenlehre. Berlin.

Lazard, Sylviane. 1976. De l’origine des hellénismes d’Agnello. Revue de linguistique romane 40:255-98.

Leumann, Manu. 1947. Die lateinische Dichtersprache. Cited after the Italian translation, La lingua poetica latina. Lunelli 1974:131-78..

——; J. B. Hofmann; and Anton Szantyr. 1972. Allgemeiner Teil der lateinischen Grammatik. Appendix to Lateinische Syntax und Stilistik (Handbuch der Altertumswissenschaft 11:2, 2). München.

Lunelli, Aldo, (ed.) 1974. La lingua latina. (Testi e manuali per 1’insegnamento universitario del latino, 12.) Bologna.

Malkiel, Yakov. 1978. Derivational categories. Universals of human language, ed. by J. H. Greenberg, 111:125-49. Stanford, CA.

Marti, Heinrich. 1974. Übersetzer der Augustin-Zeit. (Studia et testimonia antiqua, 14.) München.

McCartney, Eugene S. 1920. Forerunners of the Romance adverbial suffix. Classical Philology 15:213-29.

MLatWb = Mittellateinisches Wörterbuch bis zum ausgehenden 13. Jahrhundert, 0. Prinz (ed.) München, 1967.

Mohrmann, Christine. 1954. Sacramentum dans les plus anciens textes chrétiens. Harvard Theological Review 47:141-52.

Niermeyer, J. F. 1976. Mediae latinitatis lexicon minus. Leiden.

Pasquali, Giorgio. 1927. Rev. of Hofmann, Lateinische Umgangssprache. Rivista di filologia e di istruzione classica, n.s. 5:244-50.

Pisani, Vittore. 1974. Indogermanisch und Europa. München.

——. 1977. Tradizione e innovazione nella dinamica della innovazione linguistica. Aspetti e momenti del rapporto passato-presente nella Storia e nella Cultura, 127-40. (Istituto Lombardo di Scienze e Lettere.) Milano.

Reichenkron, Günter. 1965. Historische Latein-Altromanische Grammatik, I. Wiesbaden.

Schmitz, Victor A. 1965. Gundolf: Eine Einführung in sein Werk. Düsseldorf and München.

Shorey, Paul. 1910. A Greek analogue of the Romance adverb. Classical Philology 5:83-96.

Sophocles, E. A. 1870. Greek lexicon of the Roman and Byzantine periods. Boston.

Spitzer, Leo. 1925. Warum frz. énormément und warum romanisch -mente? Zeitschrift für romanische Philologie 45:281-8. [Reprinted in Stilstudien 1:70-84. München, 1961.]

——. 1940. Rev. of H. Mihäescu, Despre începuturile sufixului romanic -mente. Le Français Moderne 8:189-91_;#

Studer, Basilius. 1971. Spätantike lateinische Übertragungen griechischer christlicher Texte und Themen. Hermeneutik als Weg heutiger Wissenschaft, ed. by V. Warnach, 179-95. (Salzburger Studien zur Philosophie, 9.) Salzburg and München.

Van Hook, Larue. 1905. The metaphorical terminology of Greek rhetoric and literary criticism. Dissertation, University of Chigago.

Vendryes, Joseph. 1950. Le langage. Paris.

Weinrich, Harald. 1976. Münze und Wort: Untersuchungen an einem Bildfeld. Sprache in Texten, 276-90. Stuttgart.