TRANS Nr. 16: Marilena Karyolemou (University of Cyprus): Reproduction and innovation of communicative patterns in a former-"diglossic" community

Researchers’ reluctance to count the Greek Cypriot community among diglossic communities pertains less to the ideological content of the term put forward by the "linguistes natifs" - of which local scholars were least than aware before the 90’s - than to the ideological load of a term already used to describe the complicated and, at times, extremely conflict-bound relation in post-19th century Greece, between two varieties of Greek, δημοτική and καθαρεύουσα. The social and ideological apparatus of the "language question" being well known to the hellenophone Cypriot scholars - mostly philologists -, they thought it unconceivable that a regional variety could be said to hold with its mother tongue the same conflictual relation as the two Greek varieties.

The idea that the relationship between the Cypriot variety (hereafter CV) and standard Greek (hereafter SG) is a naturally ordered, harmonious relationship of the kind we find between a mother and her daughter seems to have pertained until the ’90’s. Ioannou (1988) for instance, visibly ignoring the four prototypal cases of diglossia according to Ferguson (1959), claims that it is not possible to apply a diglossic scheme designed to describe colonial situations, where an exogenous language was imposed upon local languages, to a community (the Greek Cypriot community) that has freely chosen to retain a language (Modern Greek) it considered as part of its identity. At the same time, in the light of developments in sociolinguistics regarding the ideological impact of the term, and independently from each other, a new generation of researchers took up to reconsider the relation between the local variety and standard Greek. Karyolemou (1992) for instance insists on the fact that, when it comes to describing this relationship, language attitudes are as important as language practices. According to Karyolemou, one has to take into consideration both how speakers perceive or feel about this relation and how they actually use their dual knowledge to communicate. Moschonas, in two lectures given in 1996 and 1997 (published in 1996 and 2002 respectively), underlines that, in reality, many Cypriot speakers do not consider this relationship as a natural outcome or a historical accident, nor do they regard the complementary distribution of the two varieties within the Cypriot society as an ideologically neutral act. They rather perceive it as an externally imposed norm of behaviour. Both authors seem to be in agreement as to the possibility to treat the Greek Cypriot community in diglossic terms. However, whereas Moschonas admits for Cyprus a rather prototypic diglossic pattern, that is the existence of a high and a low variety used by Cypriot speakers in formal and informal communication respectively, Karyolemou (1992, 1994), following on research by the "linguistes natifs" and especially on work by Félix-Lambert Prudent (1980, 1981, 1982) in the Petites Antilles affirms that, despite the fact that speakers clearly perceive the distinction between CV and SG in dichotomic terms, language practices do not reveal the usual complementary pattern of distribution, but rather a mixing of the two varieties in a range of intermediary uses. As a result, Karyolemou suggests the existence of diglossia in perception and continuum in usage (Karyolemou 1992, 2000). The existence of this kind of perceptual diglossia - according to Karyolemou (1992:264) "une conceptualisation dichotomique nette des codes ou variétés linguistiques et de leurs domaines d’utilisation avec des valeurs différenciées attachées à chacun" - is a remnant of an older stage in the sociolinguistic history of the island when CV and SG represented relatively closed and distant linguistic systems. As perceptions tend to persist long after reality has evolved, the diglossic perception has not only been maintained, but it has also retained a classificatory value, since it is used by Cypriot speakers as a means to categorize linguistic performances (theirs or others’).

2. Καθαρεύουσα and Δημοτική as high varieties

Given that the sociolinguistic history of the Greek language in Greece is also marked by the existence of a diglossia between καθαρεύουσα ("purist form of Greek", hereafter K) and δημοτική ("popular form of Greek", hereafter D), we do need to bear in mind that the variety that fulfilled the place of the dominant high member in the diglossic pair CV vs SG has shifted from K before 1976 to D or a modified version of D we call "κοινή νέα ελληνική" (common Modern Greek) after 1976. This is not to say that D suddenly appeared at the very moment that K lost its official credit: both varieties coexisted and D was admitted in some prestigious domains of use, for instance as the language of instruction in primary education, for short periods of time prior to 1976 (in 1917, 1929 then 1964). However, it did not then have the necessary official support it acquired in the mid seventies to play the role of a high variety.

The shift from K to D is anyway a reality we should acknowledge when discussing diglossia in Cyprus by speaking of the existence of high varieties vs. a low variety, although one must also acknowledge that meanwhile the low variety had also changed. Even so, most researchers rarely pay attention to this shift and even more rarely do they question or recognize its consequences for the evolution of diglossia in Cyprus or/and of CV itself (with the exception of Moschonas 1996). Although, in their quality as conservative varieties, CV and K share several features(1), it could be the case that the considerable structural distance between the Cypriot variety and the purist form of Greek acted as a shield against the erosion of the former. At the opposite, the coming into power of D in post-junta Greece has played a role in the acceleration of structural changes - mainly the loss of local variety features - that bring today CV closer to SG(2).

The role of the structurally distant K as a high variety in the diglossic pair CV vs. SG might have also contributed to the persisting impression of a clear-cut boundary between the low and high variety (successively K and D or standard Modern Greek), long after this boundary became less straightforward because of the raise of standard (Demotic) Modern Greek, closer to the Cypriot variety, to the status of the official Greek language. As a consequence, it is not clear any more which variety is used (CV or SG) because speakers tend to combine CV and SG features in many areas of communication, where the diglossic scheme would have predicted use of only one of them. These mixed varieties are mostly remarked and negatively judged when they are interpreted as uses of CV features in non CV contexts. When CV features occur in such areas of communication as political or public discourse, the media, education etc., where, according to the diglossic scheme, it should/could not appear, speakers tend to treat such a use as a sociolinguistic misbehaviour. At the opposite, the use of SG features in non SG contexts is not generally speaking negatively commented upon, unless it takes the form of a complete shift towards SG at all levels, including the thorniest prosodic level. Καλαμαρίζω from καλαμαράς (the person who holds a feather pen, that is the educated and, more commonly, "a Greek-speaking person from Greece"), is used to designate the act of completely accommodating to SG by Cypriots.

3. Competence in SG and accommodation to the standard

A common attitude as regards this mismatch between domain and code is either to ignore it ("This does not happen") or to consider it as a kind of communicative abnormality (Christodoulou 1997). One should however try to lay an explanation for this shift. One such explanation could be that speakers do not have the ability to switch between CV and SG because they lack the necessary knowledge to do it, i.e. they do not know well enough either SG or the differences between CV and SG in order to be able to accommodate to the prestigious norm, and therefore fill the gaps by using what they know best, i.e CV elements.

I will cite in this respect the following - favourite to me - anecdote of how images of what is correct language use may have a totally reverse effect from the one aimed at. One of the characteristics that CV shares with other north east Greek varieties (Dodecanesian, Cretan) concerns the lenition and drop of strident consonants intervocalically. As a consequence, words like SG "λαγός" [la γ ós]," έφυγε" [éfi γ e], "αδελφός" [a δ elfós], όχι [óçi] are pronounced in CV as [laós], [éfien], [aerfós], [ói]. Back in 1985, we were visiting with a friend from Greece a quite remote area of Cyprus in the wider region of Paphos district, in the north-west coast of Cyprus, called Akamas. Despite its natural beauty Akamas was, and still remains, a touristically under-developed area, and, at that time, it was quite a trip to get there. As we were passing by the village of Pomos, we stopped at a small kafeneio (coffee shop) where an old lady was sweeping the floor and our Greek friend asked some water for his "παγούρι" [pa γ úri] "flask". The lady gazed at him then corrected him: "Οϊ παγούρι, παούρι, που το πάος" [oi pa γ úri, paúri, pu to páos] "Not pa γ úri, paúri from páos". She was obviously taking her own regional form as a norm, correcting a (correct) SG form to a CV one. Her conviction that "paúri" was the correct form was backed up by "etymological" knowledge: "paúri from páos", where the neutral noun "to páos", although similar to SG "o pa γ os", the ice, of masculine gender, means specifically "extremely cold weather" and could be considered as a synonym of SG "παγωνιά" "extremely cold weather" and not of SG "πάγος" "ice".

This example illustrates quite well that the lack of competence in SG may at times be responsible for forms of speech where elements of variety are used in the place of standard elements. But while incomplete knowledge of SG could be a valid explanation for some cases of CV use in formal and semi formal contexts, it cannot be a general statement about Cypriot speakers’ knowledge of SG, and it surely cannot be a valid argument for speakers’ use of standard elements in informal contexts that normally call for the use of the CV.

4. Language use in education

To begin with, such an explanation is contradicted by the high level of school enrolment figures which show that at the beginning of the 21 st century, 94% of Cypriots above the age of 15 are literate as against 55% of literates in the middle of the 20 th century. As a consequence, one could assert that there is a more prolonged contact with SG, which is the language of instruction than fifty or sixty years ago. Moreover, Greece continues to be one of the two top destinations of Cypriot youth for studying abroad: almost 45% of students studying abroad chose Greece as their country of studies.

Generally speaking, Cypriots have an overall greater contact with SG than before through the media and frequent travels to Greece, and more opportunities for contact with Greek speakers permanently living on the island. Admitting, as we do, that education is the main channel through which contact with SG is secured, another possible answer could be that the form of Greek transmitted through education is not in fact SG but an approximation to the standard used by Cypriot teachers. As a result, educated speakers get to use in formal and semi-formal settings an approximation SG they are acquainted with at school.

During the past few years, we have had quite a number of studies dealing with pupils’ and students’ (poor) performance in SG, revealing CV interference both orally and in writing and seeking ways to amend to these interferences (Papapavlou & Yiakoumetti 2000, 2003, Yiakoumetti 2003, 2005, Yiakoumetti, Evans & Esch 2005, Yiakoumetti, Papapavlou & Pavlou in press) by more specifically proposing to raise awareness of differences between CV and SG. We also have several attitudinal studies on what teachers think about CV use in the classroom and dispose of several self-reports on language use by teachers in the framework of wider studies on language use in education (Ioannidou 2002, Pavlou & Foussias, 2005). Strangely enough, however, we have scarcely any studies as far as teachers’ performance in the classroom is concerned. Pavlou and Fousias (2005), using both questionnaires and guided interviews with secondary education instructors have shown that, whereas teachers hesitate to negatively evaluate pupils who use CV in the classroom, they generally correct them when using the CV, especially in writing. As far as their own use of the Cypriotic variety, they claim not to play special attention to their performance but to overall avoid using CV. They however also state that they use CV on various occasions in order to: focus their pupils’ attention, explain or encourage them to express their opinion.

Recorded observations in the classroom (Pavlou & Fousias 2005) have shown however that teachers use on average more CV features than their pupils. Interestingly enough, the greatest number of CV elements is used both by teachers and their pupils during History and Literature lessons, an average of 76% for teachers and 84.25% for pupils. Of course, the question that these observations raise is "Why is it that lessons that are supposed to cultivate the use of the standard language are those where the CV seems to be more extensively used?" The answer to this question has probably to do with the production of different kinds of speech for each lesson: Mathematics is a largely technical subject where pupils are asked to think by numbers and learn a lot of specific terminology. Generally speaking, the Mathematics lesson does not favour the production of continuous speech. At the opposite, History and Greek lessons aim specifically at the development of language skills. As a consequence, the longer the production of speech, the more occasions there are for CV elements to pop in.

This observation may bear part of the truth in so far that recent research by Ioannidou (2002) and Jakoumetti (2003) has shown that SG and CV do not fit the traditional pattern of language complementary distribution by sharing the (semi-) formal communication inside the classroom and the informal communication outside the classroom respectively. CV use in the classroom - from both teachers and students - seems to be a reality not only in the pre-primary and primary levels of education where such an introduction is expected, but also in secondary - and probably tertiary education - where such a use is less predictable.

Teachers represent of course a special category of speakers within the Cypriot society - or any society as a matter of fact: they are supposed to know SG better than other Cypriots, because of their role as transmitters of the norm and because, as a result of higher level studies in universities in Greece, they have been exposed - in their vast majority - to SG long enough to have a sufficient knowledge of the standard they use as a medium of instruction.

5. Language attitudes

Many people have the same kind of attitudes as teachers: although they use CV features widely themselves, they overtly state their preference for SG which they find more "correct", "polite", "sound" or "educated". It is not, thus, surprising that people who tend to use the CV in formal and semi-formal domains deny they do it, thereby adopting the same avoiding strategy ("stratégie d´évitement") as speakers of regional varieties who regularly use the variety but point to somebody else, when asked, to designate a speaker of that variety.

Karyolemou (1994a, 1994b) analyzing Cypriots’ metalinguistic discourse, has shown that speakers of a regional variety value CV less than SG by overtly evaluating it as a less correct, less expressive, and less educated form of speech. They often accuse the CV of disloyalty vis-à-vis SG since, as they claim, weaknesses in expression lead Cypriot speakers to have recourse to English loanwords, the CV thus becoming a Trojan horse for the Greek language itself (Karyolémou 1994a). As Pavlou has shown in his work with pre-primary children (Pavlou 1999), the devaluation of CV is transmitted in early childhood, mostly through the family, and therefore difficult to prescribe. It is a socially approved and sustained attitude both reproduced and enforced through social institutions and practices. Language uses in the media, for instance, tend to downgrade the CV by confining it to a limited and well specific set of (humorous) functions, thereby both reflecting the overall law status of the CV within the society and contributing to a further loss of social status (Pavlou 2004). Unpublished research (Aristeidou & Neophytou 200(3)) looking into the use of some CV features in talk shows and news bulletins (considered as (semi-) formal contexts of communication) suggests the percentage of CV use to be strongly determined by the nature of the programme, its range of audience, its subject, in other words its overall "seriousness": popular talk shows call for a much greater percentage of CV features - when these features are not subject to phonotactic constraints - than political discussions or news bulletins.

This is not however the whole story: Cypriot speakers usually rate speakers of SG as more attractive, ambitious, intelligent, interesting, modern, dependable, pleasant and educated than speakers of CV, but they do not consider them more sincere or humorous, friendlier or kinder (Papapavlou 1998), thus confirming once more that the evaluation of non prestigious varieties involves both detachment and attachment processes along a competence vs affection judgemental scale. It is clear that speakers prefer to follow a safe course of action by overtly expressing their dislike for a variety that is already overall stigmatized than contravening the social verdict and overtly supporting it, thereby risking a negative social judgement destructive for their image. Karyolemou (2001) also pointed to the fact that positive evaluation of the CV is not often encountered, but it does occur on occasion, if speakers feel that local identity and values are threatened. One such occasion has been the standardization of place names in Cyprus. Starting as early as the 60’s, under the aegis of Unesco, the standardisation process has lead to a much controversial proposal elaborated by the Cyprus Permanent Committee on the Standardisation of Place-names that eliminated all elements in the written form of place names that could reveal a CV pronunciation and adopted a written version that brought them closer to a standard pronunciation (Karyolemou 2001, Papapavlou 2002). Although these changes were quantitatively limited - they mostly concerned orthographic correction or etymological revision - they generated vehement opposition because they operated on salient dialectal features (Karyolemou 2001) clearly marked as expressions of a local linguistic identity. Their adoption was taken to mean disrespect of local values and was therefore strongly rejected.

To summarize: Cypriot speakers value SG more than the CV on intellectual grounds but not on affective grounds; they overtly devalue the CV and very often claim not to use it; they do, however, make extensive use of the CV even in formal and semi-formal domains. These evaluations prove the existence of the kind of "perceptual diglossia" I was talking about earlier on. Of course, researchers who are used to working with devalued and prestigious varieties (be it minority languages, regional languages or dialects) would not be surprised with the existence of such a dual attitude.

6. Cutting across the continuum: systematicity, focussing and unlocalizability

Until recently, scholarly work on CV has failed to recognize the changes that affected traditional patterns of behaviour within the Cypriot society and the way diglossia had evolved, especially since the second half of the 20th century. Failure to recognize the sociolinguistic situation in its complex reality - and especially the existence of a dialect continuum - had also consequences on the description of the CV itself. It resulted in a kind of uncertainty as to the structural proximity of CV and SG: is CV a, maximally different from SG, variety or a, minimally different from SG, idiom? The situation becomes more puzzling for those who try to get an idea of the relation between CV and SG, when one and the same researcher within one and the same work alternatively holds both positions. Thus, for instance, while in the first pages of his seminal work on the Modern Greek varieties, Nikolaos Contossopoulos (1980) describes CV as an idiom - "a variety close to SG only improperly called a dialect" (1980:3) -, a few pages later he asserts that the CV is the only Greek dialect to be still alive (1980:20). The only rational explanation for the above-mentioned hesitation one can think is that there is not one CV but a variability of, again, minimally or maximally diverging from SG, CV varieties, thus unconsciously recognizing the structural variability that constitutes the reality of actual linguistic behaviour within the Cypriot society: linguistic choices bring speakers sometimes closer to SG and sometimes closer to CV.

Generally speaking, researchers fail to distinguish between linguistic practices and linguistic attitudes, focussing exclusively on one or the other of the two levels. Thus, they either fail to recognize the existence of a dialect continuum that stretches from localized varieties of CV to SG or a regional form of SG (Moschonas 1996, Arvaniti 2006) or - when they do recognize the existence of a variety continuum - they disregard the value laden perception of the two varieties by Cypriot speakers (e.g. Davy & al. 1996, Pavlou & Papapavlou 2004, Sivas 2001), sometimes qualifying it as a bi-dialectal community. In both cases, this can only lead to a partial description of sociolinguistic reality. Furthermore, variability towards the local dialect pole of the continuum is usually more readily acknowledged than variability towards the SG pole. In other words, researchers do recognize the existence of a multiplicity of varieties of CV but, with the exception of Arvaniti (2006), they treat SG as a homogenous variety, ignoring both SG internal variation and the coexistence of a form of SG that is specifically Cypriot (hereafter CSG). Internal variation to SG is of course rather irrelevant for our purpose here, because Cypriot speakers tend to ignore such differences, identifying all linguistic uses by mainland Greek speakers indistinctively as SG. But CSG should be accounted for in an overall description of the sociolinguistic situation in Cyprus, since it forms part of the active linguistic repertoire of many Cypriot speakers.

Such difficulties in accurately describing the sociolinguistic situation are due to the paucity of empirical research and, at the same time, contribute further to the lack of empirical research. Therefore, both the nature of the continuum and its structural properties as well as the management of dialect continuum resources by speakers is largely ignored.

Recent discussion ha s mostly concentrated on the possibility to delineate distinct varieties within the continuity of uses, to layer the continuum. Karyolemou (1994), Davy & al. (1996), Karyolemou & Pavlou (2001), Panayotou (1996) consider that it can, at best, be divided into three bits: local Cypriot varieties, urban Cypriot(4) and SG(5). Researchers have recognized, though not empirically proved, the existence of an urban variety of CV since the end of the 19 th century, at least, (according to Newton (1973) "general Cypriot"). Papapavlou (2004) and Katsoyannou & al (in press), on the other hand, consider that one can identify five and six distinct layers (varieties) within this continuum respectively. In Katsoyannou and al. (in press), the authors discuss in extenso practical issues as regards the segmentation of the continuum and illustrate each of the five levels with examples where they combine features from CV with features from SG.

From a sociolinguistic point of view, to prove the existence of distinct varieties within a variety continuum, one has to establish both systematicity and focussing, in other words, demonstrate that a set of features is systematically used by an important number of speakers. Such systematicity and focussing have never been proven for any of the above-mentioned levels, since there has been very little empirical work.

In addition to the two criteria of systematicity and focussing, research concentrating on the urban variety of Cyprus should also document unlocalizability of features, i.e. show that speakers who use this variety of Cyprus cannot be located within a specific geographic area. The criterion of unlocalizability has been used in Scandinavian dialectology by Brink and Lund (1975:763, cited in Jorgensen & Kristensen 2001), Brink (1988: 24) and more recently by Lars (1998) and Jørgensen & Kristensen (2001) to refer to the emergence of a national Danish standard variety spoken by people all over the country, as compared to the local (classical) Danish varieties used in specific areas. Though used to define a national koiné, no localizability could also, in my opinion, serve to refer to the existence of a CV koiné such as urban Cypriot: the Cypriot urban variety should, then, be described as that variety of Cyprus that does not allow to guess the speaker’s geographical origin. In this perspective, the use of the label "urban", which anchors speakers in a special socio-geographic environment, is misleading, in the sense that speakers of this variety can be located anywhere in the country. Brink and Lund (cited in Jørgensen & Kristensen 2001: 154) affirm that unlocalizable standard features of Danish may be used by many speakers all over a country but standard speakers who use only unlocalizable features are themselves localizable within a specific area of the country: the Copenhagen area. Although the authors do not clearly establish a link between unlocalizability and unmarkedness, these two notions should be linked together as the geographical and social dimensions of one and the same phenomenon: unlocalizable features are marked as non-local and are not usually associated with such negative values as lack of education, rudeness, incorrectness, ugliness, ignorance. Following Brink and Lund (2001, op. cit.) we could say, mutadis mutandis, that users of unlocalizable linguistic elements are found all around the island but concentrate mostly in urban centres. As Jørgensen & Kristensen (2001: 155) justly put it: "researchers must go into the real world to find the "unlocalizable features". In this quest for the unlocalizable, evaluation from the speakers could be useful. Karyolemou and Pavlou (2001) questioning 45 young Cypriot speakers between the ages of 18 and 22 in order to establish the degree of awareness as regards (a) CV differences from SG, (b) CV differences from other Greek varieties and (c) the most distinctive features of CV, have established that Cypriot speakers are well aware of the differences both between CV and SG and between CV and other Greek varieties. More importantly, when asked to identify the five more salient features of CV, speakers chose unlocalizable features, all of them present in urban Cypriot, dropping all together localizable and strongly stigmatized features. On the basis of these results, Karyolemou and Pavlou (2001:119) conclude: "speakers identify [urban Cypriot] as the Cypriot variety par excellence" (emphasis theirs). Though this conclusion might seem circular - the urban variety of Cypriot not having been described yet - it is a fact that, in their answers, speakers rule out all localizable variety characteristics.

Urban Cypriot, linguistic and sociolinguistic evidence

The emergence of an urban variety of CV is, for one thing, supported by socio-economic and demographic data we dispose concerning the Cypriot society. These data show an important trend towards socio-professional and geographic mobility from the 50’s to the 70’s (Ragkou 1983), which transformed the Cypriot society from an agricultural society to a society of tertiary economy, despite the severe economic upheavals imposed upon the island’s economy after the invasion of the Turkish army that took possession of almost 70% of its economic resources. The massive displacement of population as a result of the Turkish invasion - with almost 200,000 Greek Cypriots (28% of the total Greek Cypriot population) moving from the north to the south and approximately 51,000 Turkish Cypriots moving in the opposite direction - had also important consequences for demography and geolinguistics, as it brought more people into the towns and in settlements around them. The critical human mass thus accumulated lead to an inflation of the urban centre where the surrounding area was gradually incorporated. Socio-economic and demographic factors as these usually contribute to the emergence of urban varieties and favor the homogenization of linguistic practices. In this respect, the Cypriot variety of late 19 th and early 20 th century found mostly in the capital Nicosia, and with some differences also in other urban centers (whatever the term "urban" means for a small sociopolitical space such as Cyprus in the late 19 th and early 20 th century), cannot be considered as a true urban variety, not only because of the absence of socioeconomic conditions that usually preside over the emergence of an urban variety, but more importantly because the three criteria of systematicity, focusing and no localizability are not at all present.

From a structural point of view, we should neither take for granted that a continuum of uses is made of combinations of elements that speakers pick up from the pool of CV and SG features already present in their linguistic repertoire to set up any of the above mentioned levels, nor think that combination means putting together CV and SG features that belong to the same level. If it is true that combination is a process that we often encounter in cases of contact between a standard and its varieties, this is only part of the truth: combination is only one process among others.

Let’s take for instance the lexical level. Repeated short scale studies in various communities all over Cyprus, seem to suggest that this is the level that has most dramatically been transformed in recent years (Karyolemou 2000) by showing an important degree of lexical levelling, then readjustment (or convergence), towards standard Greek. As a result, very often we come across cases of CV lexical items combined with SG lexical items: "ti kamneis" "what do you do?/how are you?" for instance where the CV word for "what is that" [indambu] has been replaced by its SG equivalent [ti], whereas the verb "to do" "kamneis" retains its dialectal form (SG would be "kaneis") (see also Katsoyannou & al. in press). There is of course a plurality of possible combinations at the lexical level but not all of them are used in practice. Lexical convergence may have another consequence: since dialect words can no longer act as agents of difference, they pass on their function to phonology and phonetics. As a consequence, a word such as αδειάζω "to empty" is marked as dialectal not because of its fundamental difference with a corresponding SG word - the dialect word would be φκερώννω [fcerónno] - but because it is pronounced [a θ cázo] and not [a δ jázo], which is the standard pronunciation. Examples like [a θ cázo] can be treated as cases of combination of elements from CV (pronunciation) with elements of SG (lexicon). These elements, because of their intermediate position, can be considered as non localizable when compared to such local lexical items as φκερώννω [fcerónno] and are good candidates for inclusion into the urban Cypriot variety. We could thus conclude that combination does not necessarily involve elements of the same level - as in the case of the phrase "ti kamneis" - but elements from different levels as well.

In a recent research on language use among Cypriot speakers, Elisavet Sivas (2002) has made some interesting observations about the way speakers handle variety continuum resources. Using Karyolemou and Pavlou (2001) results, Sivas set up a dialect-standard index that comprises 9 variables, concerning unlocalizable features, from various linguistic levels. She then checked the presence of each of these variants in 40 different oral texts with variable degree of formality - ranging from radio interviews to informal conversations between friends(6), all of which were produced by urban Cypriot speakers. The main aim of Sivas’s study was to provide empirical evidence of the nature of the variety continuum and, more specifically, to verify whether the speakers use distinct codes according to the formality of communication as predicted by the classical scheme of diglossia or, on the contrary, make use of a set of intermediary varieties combining elements from both CV and SG. The presence of each variant (be it standard or dialect) was noted 1, its absence 0. Each text was thus given a sixteen-digit number where the first eight numbers represented the standard variants and the eight last numbers the dialect variants. A speaker that would ideally use all standard variants would appear as number 1111111100000000; 000000001111111 on the contrary would represent an ideal dialect speaker making exclusive use of dialect variants. Although the method used is qualitative(7), since it only accounts for the presence or absence of a standard or dialect variant and not for its frequency, Sivas was able to place each speaker on a continuum axis, closer to the regional variety or standard pole according to his/her sixteen-digital number. Her results show that, with the exception of one speaker who made an exclusive use of the standard variants, all speakers used, for some of the variables, both the dialect and the standard variants of the variables under investigation, in other words they made use of mixed forms.

There are two conclusions we can draw from Sivas’ work. The first one concerns the nature of linguistic practices and brings an answer to the question "diglossia or continuum?" by showing an overwhelming use of mixed elements by all speakers, irrespective of the formal, semi-formal or casual nature of the communicative context. It thus confirms that the classical pattern of diglossia that predicts a complementary distribution of varieties to different domains of use cannot be used to account for speakers’ performance. The second one concerns the way speakers agree in their use of these mixed forms and brings an answer to the question of systematicity and focusing: within this community of mixed usages, each speaker combines regional and standard variants differently. In other words, whereas all speakers range within the continuum - in the middle ground of the continuum rather than at its edges - there is but little evidence that they handle linguistic resources the same way. One could therefore argue that although there is evidence of the existence of systematicity in the sense that all speakers use both regional and standard variants, there is no evidence of focusing among speakers, and therefore there is yet no decisive proof of the emergence of a common Cypriot variety among urban speakers. This conclusion is at odds with previous observation underlining the emergence of such a variety since the late 70’s and should be investigated further. It might be the case that the method used by Sivas in her research, though appropriate if we want to investigate the nature of the relation between CV and SG, it is not adequate when we want to establish systematicity among speakers. But this is not the only possible explanation.

Existence of systematicity and absence of focusing do not necessarily mean that speakers act chaotically by variably mixing together CV and SG variants in a fortuitous way. In subsequent works, Sivas (2003) affirms that speakers are aware of the amount and kind of dialectal and standard features they are "allowed" to mix within a specific context, and can regulate them accordingly by proceeding to modifications within an acceptable range of possible combinations. Sivas (2002) defines this acceptable range of possible combinations as "sliding norms". We could thus consider that sliding norms provide evidence of a kind of focussing, in the sense that speakers - although they do not use exactly the same combinations of variants - they do know what the limit for possible combinations is. If this is indeed the case, we can argue that both systematicity and focussing are at work and this can be used as an argument in favour of the emergence of an urban Cypriot variety.

Conclusion

The use of variants in non regional admitting contexts of communication may be a sign that sociolinguistic norms have shifted away from a previous unanimously accepted and practiced distinction between formal and semi formal contexts of communication allotted to SG, on the one hand, and informal domains allotted to CV, on the other hand. Although law degree of linguistic awareness may play a role in regional features being maintained by Cypriot speakers in semi-formal and formal domains of communication, alternative use of CV and SG variants shows that speakers have opted for an intermediate variety that - from a sociopsychological point of view - is neither too close to SG nor too close to the traditional rural varieties. Though more research remains to be done to confirm the existence of an urban variety of Cypriot and give an account of its main characteristics, especially in the area of systematicity and focussing, observations seem to suggest that these sociolinguistic norms are socially approved since they cut across the socio-economic spectrum and affect the majority of speakers without distinction of education or class.

(1) For instance in phonetics the prenasalised pronunciation of sonant obstruents, in morphology the preference for the contracted forms of oxyton verbs like - αγαπώ [a γ apó], χαιρετώ [çeretó] instead of δημοτική [a γ apáo], χαιρετώ [çeretáo] or the use of archaic verbal forms συνελήφθην [sinelif θ in] instead of δημοτικη [sinelif θ ice], in syntax the use the conjunction « όπως » instead of " να " e. g. « παρακαλώ όπως …» instead of « παρακαλώ να »

(3) Aristeidou N & Neophytou M 2000. "The use of the Cypriot dialect in television programmes". Unpublished text.

(4) According to Newton (1973) "general Cypriot", according to Panayotou (1996) "common Cypriot".

(5) According to Karyolemou (1994) a "regional standard"; according to Arvaniti (2006) "a Cypriot Standard Greek".

(6) For a detailed account of the variables chosen and the methodology used cf. Sivas 2002, 2003.

Arvaniti, A. 2006 "Linguistic practices in Cyprus and the emergence of Cypriot Standard Greek". Mediterranean Language Review.

Christodoulou, M. 1997. "The standard and the dialect" [in Greek]. Conflict: The Greek language and the Cypriot dialect. Nicosia: Nicosia Municipality/Popular Bank. p. 247-275.

Davy, J., Ioannou, Y. & Panayiotou, A. 1996. "French & English loans in Cypriot Diglossia". Chypre hier et aujourd’hui, Entre orient et Occident. Echanges et relations en Méditerranée orientale. Maison de l’Orient Méditerranéen & Diffusion de Boccard, p. 127-136.

Heeringa, W. & Nerbonne, J. 2001. "Dialect areas and dialect continua". Language variation and change. 13/3: 375-400.

Ioannidou, E. 2002. This ain’t my real language, miss: on language and ethnic identity among Greek Cypriot students. University of Southampton. Ph.D. Thesis.

Karyolemou, M. 1994a: CLinguistic attitudes and metalinguistic discourse: an investigation in the Cypriot press" στο Philippaki-Warburton I., Nikolaidis K. and Sifianou M. (επιμ.) Themes in Greek Linguistics: Papers from the First International Conference on the Greek Language. Amsterdam/Philadelphia: John Benjamins, σσ. 253-259.

Karyolemou, M. 1994b: La communauté sociolinguistique chypriote: analyse des discours métalinguistiques parus dans la presse écrite (1985-1992), Unpublished PhD, Sorbonne.

Karyolémou, M. 2000a. « Le dialecte chypriote ». Christidi A-Ph. (ed.) La langue grecque et ses dialectes. Republic of Greece/Centre of the Greek Language. p 111-115.

Karyolémou, M. 2000b. « Ne touchez pas à mon dialecte »: Normalisation des noms géographiques et saillance de variables à Chypre ». Journal de l’Association canadienne de linguistique appliquée 2/1: 1001-1013.

Kretzschmar Willian, 1996. "Quantitative areal analysis of dialect features", Language variation and Change 8/1: 13-39.

Moschonas Sp. 2002. "Standard Greek and the dialect: diglossia in Cyprus". Estia 1745: 899-928.

Paolilo, J. 1997. "Singala diglossia: Discrete or continuous variation?". Language in Society 26: 269-296.

Papapavlou, A. 1998. "Attitudes toward the Greek Cypriot dialect: sociocultural implications". International Journal of the Sociology of Language 134: 15-28.

Papapavlou, A. 2002. "Mind your speech: Language attitudes in Cyprus". Journal of Multilingual and Multicultural Development 22 (6): 491-501.

Pavlou, P. & Ch. Fousias 2005. "The influence of school subject on dialect use". Paper presented at the Conference on Dialects in education. University of Cyprus, 18-20 November 2005.

Pavlou, P. & A. Papapavlou 2004. "Issues of dialect use in education from the Greek Cypriot perspective". International Journal of Applied Linguistics. 14/: 243-258

Pavlou, P. 1999. "Children’s language attitudes in a bi-dialectal setting". Proceedings of the 16th Annual Meeting of the Department of Linguistics, Faculty of Philosophy, Aristotle University of Thessaloniki, 617-627.

Sivas, E. 2002. The language situation in Cyprus: diglossia or dialect continuum? [in Greek]. Unpublished Master’s thesis, University of Cyprus.

Sivas, E. 2003. "Sociolinguistic description of the urban community of Cyprus. Consequential hierarchy of the linguistic features" [in Greek]. Proceedings of the 24th Meeting of the Department of Linguistics, Faculty of Philosophy, Aristotle University of Thessaloniki.

Lars V., 1998: "The position of standardized vs. dialectal speech in Norway". International Journal of the Sociology of Language 80: 41-59.

Yiakoumetti, A. 2003. Bidialectism and linguistic performance in the standard: the case of Cyprus. University of Cambridge. Ph.D. Thesis.

Yiakoumetti, A., M. Evans and E. Esch 2005. "Language Awareness in a bidialectal setting: the oral performance and language attitudes of urban and rural students in Cyprus." Language Awareness 14/4: 254-260.

Papastavrou, A . 2000. Dialect interference in the speech of contemporary Cypriot Greeks [in Greek]. Nicosia: Cultural Services-Ministry of Education and Culture.

For quotation purposes:
Marilena Karyolemou (University of Cyprus): Reproduction and innovation of communicative patterns in a former-"diglossic" community. In: TRANS. Internet-Zeitschrift für Kulturwissenschaften. No. 16/2005. WWW: http://www.inst.at/trans/16Nr/01_4/karyolemou16.htm

Reproduction and innovation of communicative patterns in a former-"diglossic" community