From proper names to common nouns Italian ‑ismo/‑ista and Ancient Greek ‑ismós/‑istḗs formations

[ Abstract: This paper analyses both Italian and Ancient Greek nouns that have proper names as lexical bases, seeking to describe the similarities and differences between the two languages in terms of the morphological processes involved and the lexical meanings of the derived nouns with respect to the proper names. In both languages proper names and common nouns can be combined with the suffixes It. - ismo /- ista , AG - ismós /- istḗs to form common nouns. Lexical borrowing via Latin explains the development of some Italian items in - ismo /- ista , but it cannot account for their extremely high productivity in Italian, which is comparable to that of AG but not at all to that of Latin, where these suffixes are much less productive. The general topics dealt with in this paper focus on the derivation from proper names and how the mean-Dieses


Introduction
This study aims to investigate if there is a specific grammar for proper names, in particular in the field of morphology, or more precisely, in nominal derivation. We will concentrate on a class of derived nouns in a language with an open corpus, i.e. Italian (from now on It.), and in a language with a closed corpus, i.e. Ancient Greek (from now on AG). At stake here are the nouns formed from proper names with the suffixes -ismo/-ista in Italian and respectively with -ismós/-istḗs in AG. In both languages, these suffixes are highly productive (see below Section 3). Furthermore, they combine not only with proper names but also with common nouns, adjectives and other lexical categories. By means of the combination of the suffixes with various types of lexical bases, we will be able to compare the behaviour of proper names, on the one hand, and other lexical categories, on the other, and to determine the differences and the similarities of the nouns derived from proper names with respect to those derived from other lexical bases. Moreover, in spite of their distance in space and time, these languages are comparable in regards to the high productivity of the suffixes concerned here as well as the semantic values of derived nouns.
The aims of this research are twofold. On the one hand, we describe the derivational processes concerning the suffixes investigated here when they combine with proper names in order to find out differences and similarities between the two languages. On the other hand, we are interested in the relationship between proper names and common nouns and we wonder whether derivation with the suffixes AG -ismós/-istḗs and It. -ismo/-ista is different when the lexical base is a proper name and when it is a common noun.
The paper is structured in the following way. In section 1 we will motivate the decision for choosing It. and AG. In Section 2 the crucial problem of the meaning of proper names will be discussed. In Section 3 the high productivity of both AG -ismós/-istḗs and It. -ismo/-ista will be broadly illustrated, with special reference to all possible kinds of lexical bases. In Section 4 the focus will be on the different meanings of AG -ismós/-istḗs formations (Section 4.1) and It. -ismo/-ista (Section 4.2) with regard to proper names. Finally, in Section 5 we will give some concluding remarks.

The meanings of derived nouns
The meanings of the nouns which derive from proper names range from 'siding on the (political, philosophical, religious) part of X' to 'acting like X, imitating X' for nouns in -ismo/-ismós, and from 'follower of the (political, philosophical, religious) part of X' to 'imitator of X' for nouns in -ista/-istḗs. Some examples of nouns in -ismo/-ismós are It. gollismo 'Gaullism, French political stance based on the thought and action of Resistance leader and later President Charles de Gaulle', AG philippismós 'siding with Philip' for the former type, and It. amletismo 'acting in a doubtful, irresolute and somehow mysteriously melancholic way like Hamlet, Hamletism', AG akkismós 'prudery, acting like Akkṓ [proper name of a vain woman]', for the latter one. As far as the nouns in -ista/-istḗs are concerned, examples of the former type are: It. gollista or gaullista 'Gaullist', calvinista 'Calvinist, follower of Calvin's religion', petrarchista 'Petrarchist, a poet writing in a manner characteristic of Petrarch' and AG puthagoristḗs 'follower of Pythagoras', aleksandristḗs 'partisan of Alexander', and examples of the latter one are It. oblomovista 'one who has an attitude such as Oblomov' and AG lakōnistḗs 'one who imitates the Lacedaemonians' (with an ethnonym as a lexical base). More details on the productivity of these types are provided in Section 3. Even though the variation of meanings concerned here may seem inconsistent, it is possible to suggest a classification of the different meanings.
The choice of Italian and AG for our investigation is due to the fact that there exists another highly productive word formation process in both languages which is correlated with those studied here, that is the formation of verbs in -ízein in AG and in -eggiare/-izzare in Italian: e.g. AG philippízein 'to side with Philip' (from the proper name Phílippos 'Philip'), akkízein 'to imitate Akkṓ' (from the proper name of a vanish woman), hellēnízein 'to speak Greek/to Hellenize', It. petrarcheggiare 'to follow the literary trend of Petrarch', catoneggiare 'to imitate Cato', and americaneggiare 'to imitate American people'/ americanizzare 'to Americanize'.
Certainly there is a difference between It. and AG concerning the morphological transparency of the two processes (that forming nouns and that forming verbs). In AG, the relationship between the two processes is formally and semantically transparent, the suffixes -ismós/-istḗs being etymologically composed by the verb suffix -iz-and the noun suffixes -mós and -tḗs. 3 In It., instead, there is only a semantic relationship but not a formal one between the noun suffixes -ismo/-ista and the verb suffixes -eggiare/-izzare as a result of phonetic changes that occurred in the verb suffix from Lat. -izāre/-idiāre to It. -izzare/-eggiare. In spite of this difference, a high productivity of both morphological processes and a consistent similarity in the semantic values of derived items can be observed in both languages. The comparison between the two languages is even more interesting because the high productivity of the two derivational processes in both It. and AG cannot be explained in terms of diachronic continuity. By using etymologically correlated suffixes, both languages have developed similar derivational processes which create new lexemes from proper names. At this point it is interesting to mention the case of German, where there are the -ismus formations, but the verbs which are formed from proper names and are comparable to the It. -eggiare verbs are normally created by conversion (cf. Donalies 2000): e.g. thomasmannen, heideggern, poppern, barzeln, brahmsen, möllemannen, wagnern, fringsen, morsen, röntgen, haidern, goethen, kulicken, töpfern, schwätzern. 4 2. The crucial problem with proper names: the lexical meaning The first problem we are faced with concerns the meaning of the derived nouns in It. -ismo/-ista and AG -ismós/-istḗs which have a proper name as lexical base. According to Bergien (2011Bergien ( , 2013, we assume that proper names have a connotative value, and that the meaning of the proper names is built "online" and depends on the cultural meaning and the shared knowledge of the participants to the communication. 5 The cultural specificity of proper names is also underlined by Thurmair (2002a) who argues that every culture has a specific thesaurus of proper names, i.e. persons, institutions, places, brand names, etc. the knowledge of which is necessary to participate in public relevant discourse. Obviously that knowledge varies from culture to culture and requires an extensive linguistic competence of the connotations related with specific proper names. We can distinguish proper names of universal, occidental, European, or language specific, e.g. Italian or German, relevance. Especially in journalistic texts there is a continuously varying inventory of However, even a German native speaker, in order to understand the formations, needs a comprehensive, specific knowledge of the person and the characteristics, behaviour etc. for whom he/she is famous for, denoted in the morphological base. Only in that way it is possible to decide if the verb belongs to class (a) 'to behave, look like, talk, write or think like X' as in the case of thomasmannen or to class (b) 'to conduct or act following the methods or ideas of X' as in the case of fringsen 'to steal food and fuel in a justified manner'; for the classes cf. Donalies (2000) and Section 2 below. 5 It is well known that the topic of the meaning of proper names has been a longstanding problem for both logicians and linguists. Among the philosophers and logicians, Mill (1843), Russell (1940), Gardiner (1954) and Kripke (1980) must be mentioned at least; among the linguists, interesting reflections on the subject have been proposed by Jespersen (1924), Pulgram (1954), Jakobson (1957), Kuryłowicz (1966), Kleiber (1981); for an overview of the different topics concerning the proper names cf. Gary-Prieur (1994), Vaxelaire (2005) and Anderson (2007).
currently famous people and the ability to reach the correct interpretation presupposes an up-to-date knowledge often with a short "expiration date". It is not enough to know who we are talking about, it is also necessary to identify the specific characteristic or action in question.
As far as the It. nouns in -ismo/-ista are concerned, their meanings are quite variable, but two broad meaning classes may be identified, depending on the connotative values of the proper names which function as the lexical bases of the derived nouns. In the first class, the proper name stands for an attitude, a behaviour that is typical of the proper name's possessor: e.g. It. bovarismo 'bovarism' and bovarist 'a person subject to Bovarism' have the proper name [Emma] Bovary as the lexical base and are easy to understand for those who have read the novel Madame Bovary by Flaubert and know the attitude of the protagonist Emma Bovary. 6 In the second class, the proper name stands for the political party, the philosophical or literary school, the scientific approach which has been initiated by or associated with the proper name's possessor: for instance, It. platonismo 'Platonism' and platonista 'Platonist' are easy to understand, because they designate the philosophical doctrine and its followers, the proper name Plato standing for the philosophical doctrine initiated by Plato himself. So, in the first case the proper name Bovary stands for a specific attitude or behaviour of its possessor (Emma Bovary, the character created by Flaubert), while in the second case, the proper name Plato stands for that which has been created by the possessor of the proper name (Plato, the Greek philosopher), that is, his philosophical doctrine. This difference concerning the "meaning" of the proper names clearly emerges in the following examples, where the proper names are not used as referential labels to denominate the possessor of the proper names, but have connotative values 7 which refer to the attitude/behaviour of Emma Bovary, in the first case, and to the philosophical writings by Plato, in the second one: (1) Connie Benjamin è la Bovary del villaggio, sposata a un ciabattino, con due figlie; una volta sognava di diventare una cantante. 'Connie Benjamin is the Bovary of the village, married to a cobbler, with two daughters; once she dreamed of becoming a singer' (Gore Vidal, Il canarino e la miniera. Saggi letterari 1956-2000, Fazi Editore, Roma, 2003 (2) a. [Marsilio Ficino] nella sua traduzione di tutto Platone (1484) attribuiva al filosofo tutte le lettere tràdite, escluse la prima e la tredicesima.
' [Marsilio Ficino] in his translation of the whole Plato (1484) ascribed all the traditionally conveyed letters to the philosopher, except the first and the thirteenth' (http://www2.unipr.it/~pieri/platone_colli.htm) b. San Paolo è senza dubbio un platonico, ma sicuramente non aveva mai letto Platone. 'Saint Paul is undoubtedly a Platonist, but certainly he never has read Plato' (Ettore Perrella, La psicanalisi dopo la psicanalisi, FrancoAngeli, Milano, 1999, 125) The two values of proper names just illustrated can be recognized in both German verbs derived from proper names by conversion, for instance schrödern 'to behave, to act like Gerhard Schröder' and thomasmannen 'to write novels in the same way that Thomas Mann did' (Wengeler 2010: 86; cf. also Donalies 2000 for a classification of German verbs) and Italian verbs derived from proper names by means of the suffix -eggiare (e.g. catoneggiare 'to behave like Cato' and petrarcheggiare 'to write poems in the same way that Petrarch did').
In the first type, the connotative value of the proper name refers to the quality of the proper name's possessor which is considered as the most relevant one within the speech community. As for Cato in It. catoneggiare, it refers to the moral intransigence and severity of the Roman politician, since Cato is known for this quality. In the second type, the reference of the proper name is to something which has been created by the proper name's possessor: e.g. It.
petrarcheggiare means 'to write poems in the same way that Petrarch did'. The difference between the two classes of derived verbs has been explained by La Fauci (2006) and (2010) in terms of figures of speech: in the first type, the proper name is an antonomasia, while in the second one the proper name is a metonymy. The semantic difference has also a syntactic counterpart, the subject being necessarily a human in the first case, but not in the second one: e.g. i sonetti petrarcheggiano 'the sonnets are written in the style of Petrarch's poems'.

The databases of this research
For this study, we have created two databases, one concerning It. and one concerning AG. The It. corpus is constituted by the lexemes collected from the It. dictionary Zingarelli (2003), for a total amount of ca. 1570 nouns with -ismo suffix (175 with proper names) and ca. 1380 nouns with -ista suffix (90 with proper names). Some formations with -ismo and with -ista have to be excluded from the total numbers, because of the lack of a potential base or the fact that they are further derivations (prefixation or composition). This leaves us with 1282 -ismo and 1241 -ista formations. The distribution of the most frequent lexical categories occurring as a base is the following:

Proper names 14% 7%
Verbs 1,8% 1,3% As far as AG is concerned, we collected data from both the Greek-English Lexicon by Liddell, Scott & Jones (1996) and the Reverse Index by Buck and Petersen (1945). We collected ca. 860 -ismós forms and ca. 530 -istḗs forms. The nouns formed from proper names are fewer than It.: they amount to 45 nouns in -ismós and 30 nouns in -istḗs, that is, approximately 5% in both classes (45/860 nouns in -ismós and 30/530 nouns in -istḗs). Some of them are difficult to understand, because sometimes the contextual (i.e. extra-linguistic) conditions in which these forms arose and were used are unknown. In order to overcome this gap, investigation on AG texts can be useful. The research within the texts has been carried out by using the electronic research tools available on the website of the TLG (Thesaurus Linguae Graecae: http:// stephanus.tlg.uci.edu/) covering all texts of Greek literature from antiquity to the Byzantine era.
• In both languages, there are also derived nouns from compounds and idioms, e.g. It. liberoscambismo 'freetradism', liberoscambista 'follower of freetradism' (from the compound libero scambio 'free trade') and AG tetrapodismós 'a going on all fours', tetrapodistḗs 'one who goes on all fours' (from the compound tetrápous 'quadruped'); It. doppiopesismo 'tendency to apply double standards, bias', doppiopesista 'someone applying double standards' (doppio peso 'double weights' referred to the idiom usare due pesi e due misure 'to operate double standards') and AG skorakismós 'contumely' (es kórakas 'go and be hanged', lit. [go] to the ravens'). 8 This overwhelming variety of possible bases is obviously a neat contradiction to Aronoff's (1976) Unitary Base Hypothesis (UBH), i.e. the idea that derivational suffixes are limited to bases of one lexical category (cf. also Scalise 1986: 138-141). However the suffixes discussed here are not the only contradiction to the UBH: the It. modifying suffixes (diminutive, augmentative and pejorative suffixes) combine as well with various lexical categories, as Necker (2006) decidedly shows. Nevertheless, the variety of lexical bases occurring with the suffixes concerned here appears to be unique, exceeding even the modifying suffixes. Let us refer to Plag (2004) for a critical approach indicating a large number of English derivational suffixes choosing more than one base category i.e. in neat contradiction with the UBH. The word formation patterns, object of our study, are basically not part of evaluative/expressive morphology as suggested by an anonymous reviewer (cf. Scherer 2019 for a distinction of expressive and evaluative morphology). In contrast to the the It. modifying suffixes (diminutive, augmentative and pejorative suffixes) which belong definitely to expressive morphology, the negative or positive evaluative meaning of our forms depends on the context and the intention of the speaker as a general possibility of all speech, but does not concern the meaning of our derivatives per se. As d'Avis (2016: 115) points out in his study on pejoration of generic sentences: (...) obviously not all generic sentences are pejorative. (...) I would say that the two important points that lead to an interpretation of an utterance based on a generic sentence as pejorative are an appropriate context and the clearly recognizable intention of the speaker.
According to Scherer (to appear), the German formations in -ismus with proper names of politicians are not negatively evaluated as such.

Beyond dictionaries: the productivity of -ismo/-ista in Modern Italian
Dictionaries do not show the whole picture as far as the productivity of the analysed suffixes is concerned. We collected 152 deonymic neologisms with -ismo and 57 with -ista on the website of the Italian-language Treccani Encyclopedia (www.treccani.it), e.g. albertosordismo (from the proper name of the Italian actor Alberto Sordi), ignaziolarussismo (from the proper name of the Italian politician and former Minister of Defence Ignazio La Russa), blairismo (from the proper name of the British politician and former Prime Minister Tony Blair). The large number of neologisms in -ismo/-ista found in Italian newspapers is a neat proof of the "on-line" productivity of formations having proper names as a lexical base. See, for instance, draghismo (from the proper name of the President of the ECB Mario Draghi) in (3) and merkelismo (from the proper name of the German politician Angela Merkel) in (4).
(3) Una curiosità interessante che le prime Considerazioni di Draghi suscita[no], riguarda la natura terzista del draghismo, la sua tendenza a muoversi in uno spazio centrale tra i due schieramenti, cui dare anche una soggettività politica. 'An interesting fact emerging from Draghi's initial considerations concerns the third-part nature of Draghism, his tendency to move within a central space between the two parties, also giving it political subjectivity' (Il Foglio 30/05/2006) (4) Il merkelismo è ambizioso, proprio perché non sente il bisogno di giustificarsi in termini ideologici tradizionali. 'Merkelism is ambitious, precisely because it doesn't feel the necessity to justify itself in traditional, ideological terms' (La Stampa 11/09/2013) As far as other types of proper names are concerned, within the Treccani Encyclopedia we find also neologisms with e.g. toponyms (cognismo 'Cognism' from Cogne a town in the Aosta Valley precisely as site of the Cogne Homicide) and ergonyms (cocacolista 'Cocacolist' from the beverage Coca Cola, ikeismo 'Ikeism' from the ready-to-assemble furniture retailer Ikea) as morphological base. As far as AG is concerned, evidence of the "on-line" productivity of -ismós/-istḗs with proper names is given by Byzantine grammarians which attest many nouns formed from proper names and dating back to Classical and Post-Classical Greek, whose meanings had already become unusual and sometimes unclear for contemporary speakers and then needed to be collected and glossed by grammarians, e.g. akkismós 'vain attitude, prudery' from the proper name Akkṓ of a woman who was known for her vain attitude. Thus, we can suppose that the nouns studied here had a high produc-tivity in both languages throughout the centuries. In AG, the suffixes appear to be productive starting from Classical Greek, while in Italian there are only a few new formations in -ismo/-ista before the 17 th century (cf. Necker & Tronci 2014 for more details). Before describing data in more detail, we would spend some words about the notion of productivity. As is well-known, this notion has received much attention in recent studies on morphology and scholars have taken different positions towards quantitative and qualitative approaches to productivity (cf. Bauer 2001 andPlag 2006). In this research we adopt a qualitative approach to productivity: "[w]hen we call a morphological pattern productive, we mean that this pattern can be extended to new cases, can be used to form new words" (Booij 2005: 68; cf. also Gaeta & Ricca 2015). As far as our databases are concerned, data collected from dictionaries provide type frequencies which are a clear indicator of past productivity. Besides, the research in the Treccani Encyclopedia, as for It. database, and in the texts collected in the TLG, as for AG database, provides more information about the potentiality of the investigated patterns (cf. Štekauer 2014).

Ancient Greek 9
Let us start with a description of the AG state of affairs concerning the nouns in -ismós/-istḗs with a proper name as a lexical base. We will concentrate on the first attestations and we will try to define how these nouns are created, which are their meanings and how they develop along the centuries (from Classical Greek to Post-Classical Greek). Then, we will concentrate on It. nouns in -ismo/-ista formed from proper names and we will try to determine the differences and the similarities between the two languages.
In AG, the nouns in -ismós/-istḗs having proper names as lexical bases are rare. The reason for this is to be found in the fact that these formations are occasional: they were created by speakers and used in some limited temporal and spatial contexts, but sometimes they were not registered within ancient dictionaries and therefore there is no evidence for their ephemeral existence. This state of affairs does not only concern AG, but also It., where several shortlived neologisms will have disappeared before they could be registered within dictionaries. For languages with closed corpora, such as AG, this situation is even more evident. For It. as a modern language however, with the diffusion of the internet nowadays, it is easy to find evidence of occasional formations (nothing gets lost on the Web).
One of the first and more interesting attestations of the nouns in -ismós in AG illustrates how these forms are created. The noun philippismós 'siding with Philip' is related to the verb philippízein 'to side with Philip', but has no parallel formation in -istḗs. The noun is attested within a Demosthenes speech, cf. ex. (5), and its creation is a sort of reply to the verb philippízein, which is attributed by Aeschines to his political enemy Demosthenes, cf. ex. (6) In the noun philippismós as well as in the verb philippízein, the proper name designates Philip in his role of leader of a political party. So, when Demosthenes says that the Pythia 'sides with Philip', it means that the Pythia endorses the political party of Philip. Here the noun Philip is used metonymically: the name of the leader stands for his political party. This relationship between proper name and derived noun is frequent in those nouns in -ismós/-istḗs that designate a political party, a philosophical school or doctrine, a religious sect or a literary trend. We may roughly call this semantic relationship "ideology"-type. In some cases, the noun in -istḗs which is related to this kind of nouns in -ismós designates the follower of the political party, the philosophical school or doctrine, the religious sect or the literary trend, as in the following examples: A similar semantic relation can be found in those nouns in -ismós that do not designate the philosophical school or doctrine in an abstract way, but some specific practices of the philosophical school or doctrine. In this case, the noun is in the plural, as in puthagorismoí 'subtle thoughts', lit. 'Pythagorean thoughts', that is the kind of thoughts typical of Pythagoras' school.
As for political parties, the noun Aleksandristḗs is an interesting counterpart of the noun philippismón, both of them showing the "personalization" of the political power: (9) all'hoûtoi mèn hṓsper ánthropon automoloûnta pròs toùs polemíous ep'autophṓrōi tòn theòn eilēphótes, seirás te tôi kolossôi periéballon autoû, kaì kathḗloun pròs tḕn básin, Aleksandristḕn kaloûntes. 'Whereupon, as if the god had been a common deserter caught in the act of going over to the enemy, they encircled his colossal figure with cords and nailed it down to its pedestal, calling him an Alexandrist' (Plutarch, Alexander 24.7) Several nouns in -ismós/-istḗs designating political parties derive from ethnic names, which cannot be properly considered as proper names, but share some properties with them (for an overview on ethnonyms cf. 'At about this time a revolution took place in Thasos, and the partisans of Lacedaemon and the Laconian governor Eteonicus were driven out of the island' (Xenophon, Hellenica 1.1.32) Another kind of relationship between proper names and derived nouns in -ismós/-istḗs concerns the designation of attitudes and behaviours that are commonly associated with the person who bears the proper name. In this case, the proper name has a connotative value, because it evokes a specific attitude of the person bearing the proper name. The association between proper name and its connotation is shared knowledge within the social group in which the derived nouns in -ismós/-istḗs are created and used. In this case too, ethnonyms are frequently found as lexical bases. See for instance the noun lakōnistḗs 'imitator of Lacedaemonian manners': 11 (12) ên dé tis Arkhibiádēs epikaloúmenos Lakōnistḗs, pṓgōná te katheiménos huperphuê megéthei, kaì tríbōna phorôn aeì kaì skuthrōpázōn 'There was a certain Archibiades, nicknamed Laconistes, because, in imitation of the Spartans, he let his beard grow to an extravagant size, always wore a short cloak, and had a scowl on his face' (Plutarch, Phocion 10.1) The occurrences of this "behaviour"-type are not very frequent and sometimes difficult to understand, because of the cultural distance and the lack of knowledge-sharing that made the interpretation arise. For instance, the noun akkismós 'vain attitude, prudery' and the related verb akkízesthai 'to affect indifference, to affect ignorance, to be prudish', lit. 'to behave like Akkó' make reference to the woman named Akkṓ, whose vanity was insomuch known and famous that her proper name was stably associated with this connotation and was used to refer to every vain attitude or behaviour. Sometimes, the connotations associated with proper names are explained by the same ancient authors, like in the case of the noun sardismós, which occurs in the work of the Roman rhetorician Quintilian: ica confundas. 'There is also a fault known as Σαρδισμὸς, which consists in the indiscriminate use of several different dialects, as, for instance, would result from mixing Doric, Ionic, and even Aeolic words with Attic' (Quintilian, Institutio oratoria 8.3.59.4) The lexical base Sárdis was the proper name of an important city of the ancient kingdom of Lydia, which was known for its multicultural and multilingual society. The fact that the noun sardismós is not attested in the Greek texts, but only in this passage by Quintilian is further evidence of the ephemeral existence of these nouns.

Italian
As discussed in detail in Necker & Tronci (2014), in our corpus and in the analysis of It. texts, already the 13 th century presents -ista forms with a proper name as morphological base denoting supporters, disciples and followers of religious, heretical movements. No formations of the "behaviour"-type occur for -ismo forms with proper names in our corpus of It. texts, forms with other types do not occur until the 14 th century, and the first "behaviour"-types with -ista formation are documented only in the 16 th century; this may be caused by a lack of documentation (as equally assumed for AG above), but at this point of our research we must keep to the facts. In this study we will concentrate on Modern Italian. As we have already shown, both -ismo/-ista are very productive in Modern Italian, as can be seen within the online Treccani Encyclopedia, where many neologisms (not yet accounted for in actual dictionaries, as e.g. merkelismo 'Merkelism') can be found. In the following some of these will be carefully discussed.
The first example for neologisms in Modern Italian is fazismo with the proper name Fabio Fazio (an Italian television presenter) as derivational base: (14) Il "fazismo" è una delle più bieche estetiche oggi in circolazione: un misto di politicamente corretto, di buoni sentimenti, di caramellose idee ricevute "giuste" e di piccoli consigli per gli acquisti. '"Fazism" is one of the most sullen aesthetics in circulation: a mix of politically correct, good intentions, sugar-coated accepted ideas and little tips for buying' In both examples we are facing an occurrence of the noun fazismo, which belongs to the "behaviour"-type in (14) and to the "ideology"-type in (15). In both cases, it is necessary to have a more detailed knowledge of both the behaviour and the ideology of persons bearing the proper names in order to understand what these proper names stand for. It is interesting that the corresponding form with -ista, fazista, designates 'a supporter of (Antonio) Fazio' and belongs to the "ideology"-type: (16) Siniscalco, il sottosegretario alla Presidenza, Gianni Letta e l'organizzatore dell'incontro, il senatore forzista e principe dei "fazisti", Luigi Grillo.  In order to understand these examples a high interest in and established knowledge of Italian politics and the media world is essential. The following example is less dependent on specific Italian knowledge: (19) L'"esproprio proletario" al simbolo della ricchezza è concluso: senza la frenesia delle razzie con pretese di "robinhoodismo" dei giorni scorsi nei supermercati romani. 'The "proletarian expropriation" of the symbol of richness is concluded: without the frenzy of the raids with claims of "Robinhoodism" that took place in the last days in the Roman supermarkets.' (La Stampa 13/11/2004) But what is the relevant type here? Is this behaviour like Robin Hood, or is take-it-from-the-rich-and-give-it-to-the-poor some kind of philosophy or political ideology that is of relevance in this case?

Similarities and differences between Ancient Greek and Italian
The choice to investigate two different languages and corpora has revealed itself to be useful to understand the processes of deriving nouns from proper names. First of all, the investigation of Modern Italian may shed a new light on AG data as well, whose scarcity in texts does not depend on the unproductiveness of the suffixes, but on the fact that many formations were created "online" and disappeared before they were registered in dictionaries. For their part, AG nouns in -ismós/-istḗs clearly show a morphological and semantic relationship with the verbs in -ízein and this fact is helpful to explain the semantic values of the nouns too. Even if in It. the morphological relationship with the verbs is lacking, the nouns in -ismo/-ista maintain the same semantic values as their AG counterparts. The productivity of the derivational processes investigated here is a specific convergence of the two languages and does not depend on their historical relationship. It is well known that many AG words were borrowed from Latin and were transmitted via Latin to Romance languages. In the case study here, Latin borrowed from Greek some words in -ismus/-ista, but these suffixes did not become productive in Latin and the derivations involving these suffixes were perceived as foreign-sounding by Latin authors (cf. Tronci 2015a and 2017).

Concluding remarks
The first result of our study is that deonymic and non-deonymic nouns in It.
-ismo/-ista and AG -ismós/-istḗs show no differences, as far as morphological processes are concerned. Once a noun has been formed with It. -ismo/-ista and AG -ismós/-istḗs, it behaves in the same way regardless of whether its lexical base is a proper name or another lexical category (for a more detailed analysis and examples for non-deonymic formations cf. Necker & Tronci 2012). What is indeed different is the relation with meaning, because it is necessary for the nouns derived from proper names to disclose the "meaning" of the proper name used in the derivation. Since the proper name per se has the simple value of a label identifying a determined referent, we need to investigate which are the connotative values associated to the proper name. Therefore the analysis does not concern only the linguistic aspects but also the cultural ones, which are always implied when it is a matter of proper names.
The second result of our investigation is that the interpretation of the derived deonymic nouns in It. -ismo/-ista and AG -ismós/-istḗs involves two types of meaning relationships, which appear to be systematic: the first relationship is what we call "ideology", the second is what we call "behaviour". It is well known that this fact is due to the complex relationship of proper names with "meaning" and, more specifically, to the connotations that are associated with proper names. For successful communication, the participants to the speech act generally need shared knowledge. In the case of proper names, this shared knowledge concerns the referent of the proper name and its typical or actually relevant properties and features. So, learnability of both proper names and deonymic nouns is more complicated than learnability of common nouns, because it involves actual real world knowledge and changing relevance aspects due to the current (more or less international) situation.
The two types of meaning relationships, i.e. the sub-patterns "ideology" and "behaviour", are not restricted to proper names as input category, because also common nouns (among other input categories) do occur in both types, e.g. abolizionismo 'abolitionism', capitalismo 'capitalism' for the "ideology" type, and divismo 'prima donna behaviour', esibizionismo 'exhibitionism' for the "behaviour" type. What is restricted to proper names is the systematic interpretative range of variation within the two types. Derivatives from common nouns normally belong to one of the groups, either "ideology" or "behaviour", while for derivatives from proper names there is always the possibility for both types. The input categories for the two types are not restricted to a pattern denoting a person, as we see from the above mentioned abolizionismo, capitalismo, esibizionismo and the derivatives formed with e.g. toponyms and ergonyms. As far as derivatives from proper names are concerned, no difference is found in derivational morphology with the various onomastic types, that is anthroponyms, ethnonyms and toponyms. All of them can be used in principle to derive nouns of both the "behaviour"-and the "ideology"-types. However, the most tokens of our corpora derive from anthroponyms and ethnonyms.
An additional result of this study arises from the comparative perspective adopted here. In both languages, It. and AG, the same word-formation processes develop a broad range of meanings which can be however reduced to two meaning patterns. The linguistic conditions in which these patterns develop are not comparable in the languages investigated here. In AG, the nouns in -ismós/-istḗs arose from the verbs in -ízein and became productive alongside with them. In It., on the other hand, the nouns in -ismo/-ista are isolated with respect to other derivational processes. Despite of this difference concerning the derivational system of the two languages, the same two derivational patterns involving proper names occur and they reveal themselves to be relevant both language-specifically and cross-linguistically.
As a challenge for further research, we would suggest a reflection concerning the role of word-formation processes in "actualizing" the social connotative meanings enclosed in the proper names. These meanings can be bounded in time and space and can arise in some communities and not in others, so they are important to understand the relationships within a (larger or smaller) social group and the relevance of some proper names (and consequently people named by them) within the group. So, investigation on deonymi-sation mechanisms appears to be a promising research field not only for linguistics but also for sociolinguistics and social sciences.
[Abstract: This paper analyses both Italian and Ancient Greek nouns that have proper names as lexical bases, seeking to describe the similarities and differences between the two languages in terms of the morphological processes involved and the lexical meanings of the derived nouns with respect to the proper names. In both languages proper names and common nouns can be combined with the suffixes It. -ismo/-ista, AG -ismós/-istḗs to form common nouns. Lexical borrowing via Latin explains the development of some Italian items in -ismo/-ista, but it cannot account for their extremely high productivity in Italian, which is comparable to that of AG but not at all to that of Latin, where these suffixes are much less productive. The general topics dealt with in this paper focus on the derivation from proper names and how the mean-ings of derived nouns arise from the morphological processes involved. The paper suggests a classification of the different meanings involved. In our work, neither deonymic nouns nor non-deonymic nouns show any differences, as far as morphology is concerned. The study is based on corpus analysis and also includes textual analysis integrating recent neologisms.]