L* as a Secret Language: Social Functions of Early Coptic
The aim of the present chapter is to reconsider the use of Coptic as attested in the texts belonging to the Manichaean community in Kellis (Ismant al- Kharab, Dakhla Oasis). For this particular variety of Coptic, the siglum L* has been suggested by W-P. Funk, who thus qualified it as a variety of the so-called Lykopolitan dialect of Coptic, indicated by the siglum L (Gardner, Alcock, and Funk 1999:90—91). As Lykopolis is the Greek name of the city of Asyut, L* could also be called a variety of Asyutic Coptic (see the section “What Are L and L*?” below for discussion). The texts in which the variety in question is attested are dated to circa ad 355-80 and thus form one of the oldest Coptic text corpora. The majority of these texts have been edited by I. Gardner and his collaborators (Gardner 1996—2007; Gardner, Alcock, and Funk 1999).
The linguistic features of L* have been described expertly by Funk in Gardner, Alcock, and Funk 1999: 84—95; see also Shisha-Halevy 2002a. The purpose of this chapter is therefore not descriptive, but explanatory: I will put forward a number of hypotheses in order to reconstruct motivations that could have induced the members of the Manichaean community in Kellis to use Coptic for their writings. This was by no means a trivial matter: from the second to the fourth-century ad, it was Greek that functioned as the principal literary and administrative language in Egypt.
As R.S. Bagnall (1993: 238) boldly puts it: “There was no way to have an Egyptian sentence recorded except to translate it into Greek.” In the third century ad there were some attempts to write Egyptian texts with Greek characters. However, the idiom of these texts, conventionally called Old Coptic, differs considerably from fully developed Coptic as attested from the fourth century onward (Bagnall 2005). The emergence of Coptic, as it seems, was not preceded by recorded historical development.
Writing in Coptic implied, therefore, a conspicuous break with the existing conventions, a fact that calls for explanation. The main purpose of this chapter is to challenge some traditional explanations of this striking innovation and to present alternative ones. Specifically, I will argue that the Manichaean Coptic text corpus from Kellis provides a convincing piece of evidence to corroborate the hypotheses presented in my theoretically oriented articles on the rise of Coptic (Zakrzewska, forthcoming a; forthcoming b).
The structure of this chapter is as follows. In the first section, I present introductory information about the different varieties of Lykopolitan, L* among others.The next section summarily criticizes some traditional views on the rise of Coptic. This serves as a background for the third section, in which an alternative scenario is presented. The final section contains the summary and conclusions.
What Are L and L*?
L is a so-called siglum, the symbolic indication of a Coptic dialect group called Lykopolitan after the Greek name of Asyut, Lykopohs (see Kasser 1990: 148). In older studies, this dialect group is sometimes indicated by the siglum A2 (Asyutic), suggested by M. Chaine (1934). In spite of these designations, the geographical connection of Lykopolitan with Asyut is rather weak: the city could have been “the (more or less) northernmost borderline” of the vast area where L was in use (Gardner, Alcock, and Funk 1999: 91 n. 132). The southern border could have been the city of Akhmim, hence the alternative qualification of the variety in question as Sub-Akhmimic (see Nagel 1991:151-54).
It should be stressed that the so-called Lykopolitan is not a single dialect but a set of geographically dispersed, distinct varieties. Every corpus and even single texts that have been qualified as L has its own linguistic peculiarities. Most of the differences concern the divergent use of vowels, thus phonetics (or spelling), but there are also some morphological differences (Nagel 1991: 155—58). All these peculiarities have been meticulously described by scholars such as Chaine (1934), Vergote (1973—83), Nagel (1991), and Funk (1985; 1990), among others.
The main varieties of Lykopolitan are indicated by the siglum L with an index, in most cases a digit, but in our case an asterisk:
L4: Manichaean texts found in Medinet Madi
L5: fragments of the Gospel of John and the Acta Pauli from the Bibliotheca Bodmeriana
L6: part of the Nag‘ Hammadi texts and the Acta Pauli from Heidelberg
L*: the particular variety of Lykopolitan attested in the Manichaean texts from Kellis.
According to current views, the texts written in the diverse varieties of L originate for the greatest part in the fourth century ad, with a few exceptions dated to the end of the third century and the fifth century respectively (Nagel 1991: 152—53). They are thus among the earliest texts written in Coptic, which makes them an important subject of research not only as regards their content but also their specific idioms. Moreover, it is significant that these early varieties of Coptic seem to disappear in the fifth century. Both the abrupt emergence and the disappearance of these varieties can tell us a lot about the origins and rise of Coptic literacy. It is this process that I am going to illustrate with the material from Kellis.
Challenging Myths about Early Coptic
There are three persistent myths about early Coptic that I would like to question in this chapter:
- that early Coptic texts would show us how Egyptians of the fourth century spoke;
- that, due to contacts with Greek during the preceding centuries, Coptic became so heavily ‘Hellenized’ that one could qualify it as a “bilingual language variety” (Reintges 2001; Reintges 2004a: 36—43; Reintges 2004b);
- that Coptic would be used by the Egyptian Christians in order to convert simple people who lacked a command of Greek.
The last two statements are obviously mutually contradictory, as already observed in 1993 by R.S. Bagnall, who also points to the “deliberateness” with which Coptic was “invented” (Bagnall 1993: 238, 253). Still, many a study of Coptic is based on the above unquestioningly accepted assumptions, which can be paraphrased as follows: after the establishment of the Ptolemaic dynasty, which imposed Greek as the language of administration and cultural prestige, more and more Egyptians had to become bilingual and the Egyptian language they spoke became more and more influenced by Greek. When the need arose to translate Christian literature into their native tongue, they were thus compelled to use their heavily Hellenized vernacular, Coptic, for this purpose.
In what follows I would like to present an alternative view on early Coptic and to put forward arguments in favor of the view that it should be seen as a deliberately constructed literary idiom, akin to a secret language and intended for in-group use within particular ascetic communities of fourth-century Egypt. Furthermore, I will argue that the linguistic peculiarities of L* should be looked at and explained in that light.
The starting point for my argumentation is the above observation that the rise of Coptic as an alternative literary language, despite the dominant position of Greek as the language of culture, administration, and, importantly, Christian discourse, involved a dramatic change in the linguistic behavior of certain segments of the Egyptian population. As a change of linguistic behavior is basically a social phenomenon (Milroy 1992: 202), the approach adopted here is a sociolinguistic one: the main explanatory principles that I employ refer to social functions of language use. My approach is inspired by insights from the following disciplines:
- Social Network Analysis and its sociolinguistic applications, for example,Barabasi (2002),Bruggeman (2008),Kadushin (2012), Scott (2013), Ruffini (2008),Schor (2011), Collar (2013), L. Milroy (1987; 2002), J. Milroy (1992), J. and L. Milroy (1985)
- social dialectology and historical sociolinguistics, for example, Britain and Cheshire (2003), Hernandez-Campoy and Conde-Silvestre (2012)
- contact linguistics, for example, Muysken (2000; 2007; 2010), Myers- Scotton (2006), Matras (2009),Johanson (2013)
- literacy studies, for example, Gee (1996), Edwards and Sienkewicz (1990), Hornberger (2003), Street (2000), Wagner, Outhwaite, and Beinhoff (2013).
More relevant references are to be found in my articles (Zakrzewska, forthcoming a; forthcoming b), which form the basis of of the next section, below. In this section, theoretical argumentation from these two articles is illustrated with the data from Kellis. In this way I would like to achieve two aims: first, to illustrate my earlier theoretical studies on the rise of Coptic literacy with actual material and, second, to demonstrate that specific features of this material are best explained in the light of those studies, in other words to demonstrate the explanatory value of my argumentation.
An Alternative View on Early Coptic
Does Coptic Represent Spoken Egyptian?
As I argue in Zakrzewska, forthcoming a and forthcoming b, the above question has to be answered in the negative. Although a substantial percentage of the inhabitants of Egypt in the third to fourth centuries ad must have spoken a form of Egyptian, there is no evidence of what this spoken Egyptian could have been Eke. What can be studied is literary Coptic as represented by various scribal traditions. Instead of speculating about the supposedly high degree of bihngualism in late antique Egypt, we should thus better discuss well-attested biliteracy, that is “the use of two or more languages in or around writing” (Hornberger 2003: xii; see also Johanson 2013).
From our ethnocentric perspective we often take it for granted that people write down what they otherwise would say. In a miheu of restricted hteracy, however, literacy practices can have more symbohc than utiEtarian function. This phenomenon can be illustrated by the foUowing quotation from B. Street, one of the pioneers of Eteracy studies: “Literacy was not simply a set of functional skiffs, but rather it was a set of social practices deeply associated with identity and social position” (Street 2000: 22—23).
Even in modern highly standardized languages, considerable differences exist between spoken and written hnguistic varieties. This was even more so in antiquity when access to literacy was hmited. Being literate was the exclusive privilege of the upper strata of society, who could afford education and for whom practicing hteracy was a question not only of practical needs but also of social and cultural considerations, such as the expression of a certain lifestyle. As early Coptic texts are often texts of the highest status and the highest literary and intellectual quaEty (for example, translations of Holy Scripture and of theological and philosophical studies, including Manichaean and Gnostic treatises), it can safely be assumed that their writers tried to distance themselves from any vernacular.
Even in the case of letters, the utilitarian function of the text (conveying a message) was considered subordinate to the symbolic one (evoking the shared social and cultural background of the correspondents). This last function was facilitated by the ample use of appropriate formulary expressions that enhanced the symbolic function of texts, as they allowed both the author and the addressee mutual recognition as members of the educated milieu, sharing the intricate conventions of verbal etiquette. A characteristic example of the superiority of the symbolic function of written texts above the utilitarian one are some letters found in Egypt that consist almost entirely of polite formulae (Bagnall and Cribiore 2006: 12—14).The business-like part of the message, as socially less significant, was considered less worthy of being put down and possibly had to be delivered orally by the carrier of the letter. Another example is provided by some letters from Kellis in which formulary sections are in Greek, the language of higher prestige, while the business-like body of the letter is in Coptic (P.Kell. Copt. 22,24,35,43).
This last phenomenon, a situation in which one text contains chunks in two languages, chunks that are larger than one word or phrase, is a conscious communicative strategy called alternation^ code mixing (Muysken 2000: 96-121; Muysken 2007: 321-22, 332-34; see also Johanson 2013: 299-306) .The alternating text chunks are not random but are differentiated with respect to their discourse functions, as, in our case, participant-oriented (formulary) versus business-like. This is a nice piece of evidence which points to conscious use of linguistic means by the members of the Manichaean community, well known for their respect for writing and the written word (Gardner, Alcock, and Funk 1999: 77,163 n. 19).
Were Coptic Writings Used for Converting Egyptians without a Command of Greek?
My answer is again in the negative. In a culture characterized by restricted literacy, as in late antique Egypt (Goody 1968,11—20; Bagnall 1993: 259- 60), conversions, even of individuals, were primarily a social process more likely to be brought about by the influence of social networks (see “Social Networks and Innovative Language Use,” below, andVan derVliet 2013).
As for the use of Coptic, it is remarkable that from the very beginning texts written in this language show a mature literary idiom in which even the most complicated subject matter can be expressed.This functional versatility of Coptic came at a price. Coptic writers had to adopt not only the Greek alphabet but also literary genres, argumentative structures, and rhetorical figures. Moreover, in order to handle new intellectual notions they had to borrow numerous Greek words. Paraphrasing P. Muysken (2000: 275), one could say that they wanted to express in Egyptian the Christian Greek conceptual universe. For example, in the Kellis letters, 25 percent of the vocabulary is of Greek origin (see “The Use of Greek Loanwords,” below).Therefore, early Coptic writings were simply incomprehensible to people without a thorough Hellenistic education. Nevertheless, the fact that literary Coptic was constructed on the basis of native Egyptian mor- phosyntax enriched with Greek borrowings, and not the other way around, suggests that Egyptian was the first language of the early Coptic writers.
An obvious question presents itself at this moment: who were these writers and for whom were their texts intended?
Social Networks and Innovative Language Use
As argued above, the use of Coptic as a literary language must have been a matter of choice rather than of necessity. It was also a radical change affecting the long-established linguistic behavior of certain groups of literate inhabitants of Egypt. In order to explain this change, I will resort to the applications of Social Network Analysis (SNA) in the field of sociolinguistics. Specifically, I will refer to the groundbreaking research of Lesley Milroy and James Milroy.
Linguistic changes follow scenarios known from other changes in human behavior, scenarios that have been the subject of much research within the framework of SNA. For a linguistic change to take place, at least two crucial initial stages are required: innovation and diffusion. The present considerations are limited to the first of these stages, innovation (for a reconstruction of the following stages, see Zakrzewska forthcoming a). As argued by J. Milroy, linguistic innovators are typically peripheral members of their community who “are often perceived as underconforming to the point of deviance” (Milroy 1992:184).
For potential innovators it is important not to have particularly strong ties within one particular community but to have multiple weak ties with members of various communities. Thanks to their multiple and rather loose contacts, or ties, with people from many communities, these peripheral members become aware of new ideas and can become inspired by them. Consequently, they can function as “bridges through which information and influence are diffused” (Milroy 1992: 184; see also Bruggeman 2008: 68—77; Kadushin 2012: 31—31, 42—43). In contrast, having strong ties with other members favors norm-enforcing behavior and quite disfavors the bridging function.
Now, let us return to late antique Egypt. The end of the third century and the first half of the fourth saw the rise of diverse ascetic communities: Catholic, Manichaean, Mehtian, Gnostic (Emmel 2008; Goehring 2007; Wipszycka 2009; Wipszycka 2013b). It is within these communities that the locus of the sociolinguistic change that led to the rise of Coptic should be situated. As is well known, these pioneers of the monastic movement consciously distanced themselves from their original social milieu and networks in order to create an alternative lifestyle that can only be characterized as innovative in the extreme. Part of this alternative lifestyle, in line with the revolutionary intellectual dynamism of these religious movements, may have been divergent linguistic behavior, in this case the use of Coptic in writing. To put it somewhat sharply: literary Coptic was thus constructed not in order to convert Egyptian peasants but to discuss new ideas with like-minded, well-educated ‘counter-culturists.’ One of these counter-culturist groups was the Manichaeans from Kellis.
How to Express Exclusion or Solidarity?
The life in the Manichaean community in Kellis, reconstructed on the basis of both textual and archaeological sources, has been summarily described by I. Gardner (Gardner 1996—2007, part 1: vi—xvii; part 2: 3—8; Gardner, Alcock, and Funk 1999: 5—58, 72—83). It is important to emphasize the ambivalent social situation of the members of such an ‘alternative’ community. On the one hand, they had to loosen their social ties with broad layers of the population and even their networks in the Nile Valley/‘probably years might pass before a friend or family member in the valley would be able once again to visit the oasis” (Gardner, Alcock, and Funk 1999: 12).
On the other hand, they had to forge strong ties with other members of their own community, a fact revealed by the Kellis letters, which “indicate a particularly close-knit set of family groups” resembling “sectarian movements, particularly in their earlier, world-denying stages” (Gardner 1996—2007, part 1: viii; see also Dal Fiore 2007, Bruggeman 2008: 37—61, and Kadushin 2012: 108—34 for communities vis-a-vis social networks in the context of innovative behavior). This interplay of weak versus strong social ties had clear correlates in the linguistic behavior of the members of the Manichaean community in question, at least as far as the use of Coptic in texts is concerned.
First and foremost, these early Coptic texts written with Greek characters and with a considerable quantity of Greek borrowings were inaccessible to both speakers of Egyptian lacking Hellenistic education and to people literate in Greek without a command of Egyptian. Writing in Coptic thus implied the exclusion of potential readers who belonged to either of these two groups, probably a greater part of the potential reading public.This can be interpreted as a so-called negative act of identity, underlying the construction of secret languages (see Croft 2003: 65-66).
Another important function of a secret language, beside exclusion of outsiders, is to confirm the shared social identity of the group members and their mutual solidarity (“positive acts of identity,” Croft 2003: 65). In this context it is significant that potential adepts of the new way of writing could not rely on the traditional system of education, limited as it was to teaching Greek Hteracy. The education in Coptic at this initial stage took place within one’s own community, as illustrated by one of the Kellis letters in which a father instructs his son to do writing exercises every day in order to be able to copy books eventually (P.Kell. Copt. 19, 13—19). Another example is a syllabary with specifically the ‘new’ (non-Greek) characters of the Coptic alphabet (P. Kell. Copt. 10).
In this way, one acquired linguistic norms that were accepted in this particular community and not outside it. At least two conspicuous linguistic properties of early Coptic, also observable in the Kellis texts, are generally acknowledged as facilitating the expression of the shared social identity: the cultivation of particular spelling conventions and the use of words borrowed from another, more prestigious language, in our case Greek.
A well-known feature of linguistic behavior in close-knit communities is the cultivation of local standards of pronunciation, since even slight phonetic differences allow one to distinguish the members of the in-group from outsiders. For example, Milroy (1987:116-38) distinguished six phonetic variants of the vowel /a/ in the speech of the inhabitants of certain close-knit Belfast neighborhoods. Milroy (1992: 98) observes that an occasional use of a more standard variant by one of the informants was “loudly mocked by his companions” and concludes that this extensive variation was maintained due to “community pressures.”
This phenomenon seems to be reflected, at least to some extent, in a considerable ‘dialectal’ variation between the early Coptic text corpora. Recall that the differences pertain first of all to the divergent use of vowels (“What Are L and L*?” above). To what degree this variation reflects the actual speaking habits and to what degree the local scribal traditions and spelling conventions is of secondary importance. What counts is the very fact that these particular traditions were cherished within the particular communities. Thanks to such linguistic strategies, as a rule meaningless for outsiders, the members of a given community could give expression to their shared experience and solidarity.
Seen in that light, early Coptic “dialect boundaries” could probably better be considered “social boundaries” or community boundaries (Myers- Scotton 2006: 31). This reconsideration of early Coptic dialects explains two things: first, that letters of the members of the Kellis community, written in several regions of Upper Egypt, are linguistically so conspicuously uniform among themselves (Gardner, Alcock, and Funk 1999: 84) and, secondly, that the Kellis texts are at the same time significantly different from texts written in other varieties of so-called Lykopolitan (see “What Are L and L*?” above). These other texts originate, according to my line of reasoning, from different communities, which could also be geographically dispersed. The lack of geographical contingency renders traditional dialectological explanations inadequate.
In contrast, according to social network theory, geographical considerations are of less importance than social and relational ones (Bruggeman 2008: 23).
The Use of Greek Loanwords
Another well-known feature that points to the ‘constructed’ character of early Coptic is the ample but arbitrary use of Greek lexical borrowings. Much research on language contact has conclusively demonstrated that the ‘gap theory,’ according to which the function of borrowed words is to fill lexical ‘gaps’ in the recipient language, is only one part of the story. At least equally important is the symbolic function of borrowings to express one’s social identity (Muysken 2000: 69; see also Myers-Scotton 2006: 143, 167—69; Matras 2009: 149—53). An interesting property of a ‘constructed’ code, as opposed to one which arises as a result of organic’ or spontaneous borrowing over an extended period of time, is different restrictions on lexical borrowings (Muysken 2007: 316-20). In the case of historical borrowing, the use of certain well-established loanwords becomes obligatory, especially as regards the so-called cultural vocabulary, lexical items that designate concepts valid in a specific cultural environment (such as the loanwords in English ‘altar,’‘chapel,’‘Mass,’ which are all of Latin origin and belong to the semantic domain of Christian worship). In contrast, a ‘constructed’ code typically displays various lexical idiosyncrasies: stylistic variation or words of native stock where one expects borrowed ones.
The arbitrary character of Greek lexical borrowing in Coptic was already signaled by Steindorff (1950: 201—202) and Kasser (1991: 217—18). As regards the Coptic texts from Kellis, the majority of Greek loanwords are indeed cultural vocabulary items, but some stylistic variation in their use can nevertheless be noticed; for example, the Greek legal term kategorein, ‘to accuse,’ is used interchangeably with its Coptic equivalent smme (Demaria 2005: 111-13). The arbitrary character of Greek borrowings in these texts can be illustrated by the use of two words: sah, ‘teacher,’ versus kathekoumenos, ‘catechumen.’ Both these words are technical terms that designate a particular position in the hierarchical structure of Manichaean social organization and thus belong to the cultural vocabulary.
The term sah is used to designate ‘the Teacher,’ a dignitary holding one of the highest offices in the transnational Manichaean hierarchy and operating outside the Kellis community (Gardner, Alcock, and Funk 1999: 75—76). This dignitary is nevertheless referred to by a ‘native Egyptian’ word. In contrast, fellow villagers of a low status within the Manichaean hierarchy are referred to with the word kathekoumenos, borrowed from Greek. This points to a deliberate construction of certain terminological conventions rather than to organic development.
As for the proportion of words of Greek origin, these form 25 percent of the lexical entries listed in the Indices of the edited documentary Coptic texts from Kellis (Gardner, Alcock, and Funk 1999: 282—315) and 18 percent of the entries of the literary Coptic texts (Gardner 1996—2007, part 1:145—69; part 2:139—56).These percentages may seem high in comparison with pre-Coptic Egyptian, but are not especially impressive from a cross-linguistic perspective. Compared with the results of the Leipzigbased Loanword Typology Project (LWT) reported by Tadmor (2009: 56-57), this particular variety of Coptic could qualify as an ‘average borrower,’ comparable to Dutch (with 19 percent loanwords) rather than a ‘high borrower’ like English (41 percent).
This is clearly at odds with some rather ‘impressionistic’ assumptions of an exceptionally high degree of Greek lexical influence on Coptic mentioned in the section above. The actual influence, at least as attested in L* appears to be limited to cultural vocabulary adopted in a rather arbitrary fashion and in an average quantity. It is thus much less spectacular than what Reintges’ dramatic expression “bilingual language variety” might suggest.
As suggested above, the well-established statements about the excessively high proportion of Greek loanwords in Coptic are influenced by comparisons with the earlier stages of Egyptian, which do not show any significant foreign lexical influence. In this respect, writing in Coptic involved, again, a break with old (and by that time obsolete) literary conventions. From a cross-linguistic and cross-cultural perspective, however, it is these conventions of pre-Coptic Egyptian that should be considered exceptional. In other literary traditions, of which Coptic appears to be an instance, a strategy of adopting diverse elements from a prestige language is quite common and, as Johanson concludes, “the codes representing this type do not seem to deserve characteristics such as ‘intertwined,’ ‘mixed’ or ‘hybrid,’ even if they, as English, manifest excessive copying” (Johanson 2013:323).
L* as an In-group Variety: More Indications
The Kellis corpus also includes text fragments written in Sahidic. Significantly, these Sahidic texts are not Manichaean, but Catholic (Gardner 1996—2007, part 1: vii; part 2: 5). This indicates once again that the selection of a particular variety of early Coptic was determined by the wish of its users to express a certain social meaning, to mark their allegiance to a particular community.
The important role of expressive function of early Coptic may explain the disappearance of most Upper Egyptian varieties toward the end of the fourth century when they were superseded by Sahidic. These socio- linguistic processes coincide with the decline of some of the early ascetic communities and the growth, in both size and importance, of others, in particular the Catholic Pachomian ones. As is well known, Sahidic, which became the most important supra-regional variety of Coptic in the centuries that followed, can hardly be pinpointed as a dialect of a specific locality or region in Upper Egypt. It is tempting to hypothesize that the variety of Coptic conventionally called Sahidic was the one that the successful Catholic communities adopted for their purposes and passed on to their followers. Such a hypothetical scenario is entirely consistent with Social Network Analysis, which ascribes successful diffusion of an innovation not to the original nonconformist inventors, but to well-connected and influential opinion leaders (Kadushin 2012: 141—46). Without the slightest doubt, Pachomius and his successors can be considered prominent opinion leaders of their times.
Summary and Conclusions
In this chapter, I have argued that early Coptic was not simply an Egyptian vernacular but a deliberately constructed alternative literary language and a prestige variety. In order to stress this point, I have consistently referred to the writers or users, instead of speakers, of Coptic. I have also refrained from sticking anachronistic ‘nationalistic’ labels on members of supposedly antagonistic ethnic groups (Egyptians versus Greeks) and from simplistic equations between language use and ethnic identity: “Markos speaks Coptic, so Markos is a Copt” (see also Wipszycka 1992 and Fournet 2009: 432—34). Incidentally, I have carefully avoided using the term ‘Copts,’ anachronistic for the time before the Arab conquest.
The association of Coptic with Christianity is not due to its alleged function of converting ‘indigenous’ Egyptians, but to its rise and development within early ascetic communities that were the locus of innovative and highly regarded social practices in late antiquity. I have argued that early Coptic was an in-group variety akin to secret languages, which was used as a linguistic correlate of innovative social practices in these communities. Specifically, one of these early communities comprised the writers and readers of the Manichaean texts from Kellis who, by constructing the variety of Coptic conventionally called L* creatively used the Egyptian and Greek linguistic resources at their disposal in order to strengthen their social ties and in-group solidarity. Accordingly, I have argued that the traditional dialect boundaries should be reinterpreted as social boundaries, rather than geographical ones.
As regards the role of Greek lexical borrowings, without denying their rather obvious communicative function of filling semantic ‘gaps,’ I have emphasized their symbolic function of signaling one’s social identity. Moreover, I have argued that the use of Greek loanwords in Coptic, besides being less excessive than is sometimes assumed, was more likely an expedient for constructing novel literary conventions than a reflection of actual bilingual speech.
The interpretation offered here has allowed me to explain in a natural and coherent way several apparent oddities that were left unexplained by traditional accounts, for example:
- The linguistic differentiation of the particular corpora and the proliferation of distinct varieties of early Coptic. This can be explained by referring to particular norms maintained and taught within a particular community.
- The geographical dispersion of these varieties, as social networks are based on relations between their members rather than geographical contingency.
- The arbitrary character of lexical borrowings from Greek as characteristic of deliberate borrowing, in contrast to organic or spontaneous borrowing over a prolonged period of time.
The sudden appearance and disappearance of most Upper Egyptian varieties of Coptic in the fourth century as corresponding to the rise and decline of diverse ascetic communities in that century, as well as the fact that these varieties were eventually ousted by Sahidic, the idiom of the Catholic communities that came to dominate the Christian landscape of Upper Egypt from the fifth century onward.
Ewa D. Zakrzewska