What Russian Speakers Find Hard About Greek (According to Their Teachers)

“It’s hard to tell which language is more ‘mine’” said one of the teachers we interviewed in a recent study, when my co-author asked how she felt about her linguistic repertoire as a Russian speaker learning Greek. Another remarked that she lived “in two languages at the same time”. Both remarks struck a chord with me, because I am no longer sure whether ‘my language’ is Modern Greek or English. I have been thinking about that observation for some time now, and although none of its implications really seemed to fit the focus and word length of the chapter, this seems like the right place to revisit the question.

The chapter in question, “When Modern Greek and Russian meet: Teachers’ perspectives on learning challenges”, has just appeared in Analyse contrastive et didactique des langues étrangères [Contrastive analysis and Foreign Languages Teaching], a volume edited by Freiderikos Valetopoulos (University of Poitiers).¹ Eleni Motsiou and I have been working together for some years now on questions of language contact, from family language policy in mixed-language households to translanguaging and, more recently, language policy in Greek universities. The focus of this particular chapter is somewhat more granular: what actually goes wrong, and where, when a speaker of one language sets about learning another.

Why ask teachers?

To gain insights into the challenges Russian-speaking learners face when learning Modern Greek, we turned to language teachers who—in most cases—had also learnt Greek in instructed settings themselves, so they could speak from both sides of the desk.

The most direct way to learn about learner difficulty would, of course, be to study learners: i.e., collect their output, identify deviations and count them (if you’re the sort of researcher who is into that thing), sort them into categories, and try to make meaning out of the patterns that emerge. We took a somewhat different approach in this study, and I would like to say a few words about our methodological thinking because it has some implications for what this kind of research can usefully deliver and what it cannot.

Counterintuitively, learners are not always good witnesses to their own learning. They may lack the metalanguage that would help them accurately describe their experience. When talking about their learning, they might assign great weight to a difficulty that is communicatively trivial, and do not always notice if and how output differs from the target at all. Language teachers, on the other hand, can retrieve pattern-level knowledge from insights accumulated over years of marking, correcting and explaining; and they normally have the vocabulary to articulate it.

I have written elsewhere about how much of the professional knowledge teachers construct as they teach never travels beyond the confines of the classroom(s) where it was learnt. This chapter is, in a modest way, an attempt to make one small piece of that knowledge travel: to take what experienced teachers of Greek know about their Russian-speaking students and place it in a format where interested colleagues might find it.

A weak form of an old theory

Readers with an applied linguistics background will likely recognise the intellectual lineage of this chapter. Contrastive Analysis,² in its original formulations at least, was based on the belief that systematic comparisons between pairs of languages could predict where learners would stumble. That promise, we know by now, never fully materialised: similarities do not always make learning easier and differences do not always obstruct it.³ A subsequent intellectual step, Error Analysis, shifted attention to learner output and reframed mistakes as evidence of hypothesis-testing. Interlanguage Theory went further, treating the learner’s language as a system in its own right, shaped by the first language but not reducible to it.

Our theoretical position in the chapter is a deliberately modest one. We take as our starting point that cross-linguistic influence exists and that one can make sense of what appears in learners’ output by drawing on knowledge of both languages. What we do not assume is that any of this is strongly predictive. We expect to find similarities in the interlanguages of learners who share an L1. But we treat this expectation as a hypothesis about tendencies, not a law about outcomes. Those who follow my interest in complex dynamic systems will probably recognise the shape of the argument: we are, again, in “things that are never fully predictable, but not entirely random either” territory.

What the teachers reported

The interviews, conducted by Eleni in Modern Greek with frequent switches into Russian,⁴ were analysed thematically and organised by level of linguistic description. I will not reproduce the tables here, but the broad picture about the difficulties Russian speakers learning Greek encounter is as follows.

Phonological challenges

At the level of sound, the difficulties were roughly what one would predict. Greek fricatives that do not appear in Russian, notably [θ] and [ð], get assimilated to the nearest available Russian category. This makes [a’θina] ‘Athens’ something like [a’fina]. A less predictable change, which I found quite fascinating, concerned the role of vowels: Russian reduces unstressed vowels in ways that Modern Greek does not. What happens, then, is that learners are likely to carry into Greek a rule that Greek has no use for, and this results in forms that are morphologically correct but sound wrong and occasionally mean something else.⁵ Stress placement was also a recurring problem, even though Greek marks it in writing.

Spelling challenges

Spelling, perhaps surprisingly, caused little trouble; the Cyrillic and Greek scripts are close enough that the transition is gentle. What did cause trouble was the etymological baggage of Greek orthography. It’s not easy learning how to spell in a language that has six or seven different ways of writing out the [i] sound.

Morphological challenges

Persistent difficulties also seem to cluster in morphology. Gender assignment in Greek is often arbitrary, and the neuter, so common in Greek and so marginal in Russian, caused one participant to suggest (I believe only half in jest) that we should simply abolish it. Verb aspect is a persistent challenge, too, as are the absence of an infinitive and the existence of verbs that do not have the passive or active forms Russian speakers expect.

And then there is register: Modern Greek somewhat distinctively indexes formality through morphology. Formal language is often marked by using morphology that echoes ancient forms of the language. Knowing when to use an archaic genitive or a high-register verb form is a matter of feel more than anything else, and even the most experienced participants admitted they had not fully acquired that.

Syntactical challenges

Unsurprisingly, syntax generated fewer complaints (languages with complex morphology, like Modern Greek and Russian, tend to be fairly flexible when it comes to structuring sentences). The only two challenges that came up, however, were quite telling: one was the tendency of Russian speakers to use redundant subject pronouns (Greek is what we call a pro-drop language, where subject pronouns are not strictly necessary). The other one was their tendency to omit copular verbs, which are redundant in Russian, but necessary in Greek.

Lexical challenges

The lexical difficulties were few, but they tended to be unusually stubborn. Some of them involved verbs of motion, the distinction between ‘where’ and ‘where to’, and the idiosyncratic Greek use of going up and going down to describe travel between city and village. Added to these were also the hazards of dialectal variation: What counts as «κασέρι» [kasseri], a type of yellow cheese, depends on where in Greece you are. More problematically, βίλα [‘vila], a country house in mainland Greece, means something quite different in Cyprus.⁶

The observations to which I keep returning

I began this post by describing my own unease about how thinking about identity complicates our efforts to tidy up language. I would now like to return to this thought, and its implications.

Three aspects of language education

The participants in the study were highly proficient speakers with a strong sense of dual linguistic identity and considerable metalinguistic knowledge. That the three do not fully overlap should not be surprising, but it is still something that we need to bear in mind: knowing about a language, being able to use it, and feeling that it is yours are different aspects of language education, and they develop along different trajectories.

Just like the participants, I feel equally competent in the language into which I was born and in the language of my profession. I am conscious of the tensions between the three aspects in my linguistic identity, and I am left wondering whether all of them (should) have a place in the way we structure our lessons.

Separate languages or repertoires?

There is a second, related observation, and it comes from reading the transcripts. All the participants treated translanguaging as normal for multilingual students, and they all described it as a sign of progress rather than deficit. Interestingly, not one of them mentioned translanguaging as something a teacher might do for pedagogical effect. They spoke of their own switching between languages as deliberate identity choices, although the transcripts left me wondering how many of these choices were conscious.

What this suggests to me is that a language separation ideology (which I will provisionally define as the tacit view that each language is a closed system) can coexist with generous attitudes towards learners who mix. The attitude is pedagogical; the ideology is structural; and the second does not shift just because the first is kind. And this has made me think of Maria Cioé-Peña’s remarks about how much of what teachers might do with learners’ repertoires goes unimagined not for lack of goodwill but for lack of a frame.

What this means for teaching Greek

We conclude the chapter with teaching implications, and I will be brief about them because the argument we make is uncomplicated.

A teacher who can identify areas of heightened difficulty for a particular group of learners can also manipulate the salience of those features in input and task design (in the spirit of Schmidt’s noticing hypothesis). That, in turn, is an argument for supplementing published materials, often designed in a ‘one-size-fits-all’ way, with content tailored to learners who share a first language, and there is some emergent scope here for using Artificial Intelligence in a pedagogically useful way. More importantly, there is an argument to be made here for more agentic teachers and less textbook-led instruction, a position I have argued from other directions before, not least in connection with the Greek textbook I helped to write.

The last implication is the one I would most like language teachers to take away. Practices that stigmatise deviation from the norm tend to obscure where the deviation comes from, or unhelpfully attribute it to lack of intelligence, aptitude or effort among students. If, instead, we choose to view learner output as a set of traces left by a developmental trajectory, we might learn new things about the learner, and the learners might learn new things about their languages, both L1 and target. That is, I think, a better bargain for everyone.

Where this leaves us

Eight interviews do not make a description of an interlanguage, and we do not pretend that it does. What they do make is an inventory, generated by people who know what they are looking at, of where a particular group of learners tend to struggle and why. The next step, which I hope someone takes, is to put that inventory against classroom data and see how it holds up: how frequent these forms actually are, at which stages they appear, which of them fade and which of them stay.

Until then, I will keep thinking about my own repertoire and what it says about my identity. Sometimes we direct our empirical gaze outwards, just because it’s easier than introspection.

---

Notes

1. The volume’s editorial introduction, by Fredereikos Valetopoulos and Anabela Fernandes is kind enough to direct readers to our chapter for the history of contrastive analysis; I will reciprocate by sending interested readers to the introduction for a much fuller account of Lado, Weinreich and the Contrastive Structure Studies of the 1960s than we had room for. ↩︎
2. Its ironic that the very first thing I ever wrote publicly, as an MA student, was a criticism on someone’s work on Contrastive Analysis. At the time, my former wife had somewhat correctly pointed out that it would be a better use of my time and intellect if I developed ideas of my own instead of criticising others. She was, of course, right but being right is not always enough to get things done. ↩︎
3. The standard reference for this is Whitman and Jackson (1972), who tested several contrastive analyses of Japanese and English against learner performance and found the predictions did not hold. The point was made with some force at the time and has, I think, been under-remembered since. It is one thing to say that the first language matters; it is another to say that a linguist with two grammars in hand can tell you in advance what a learner will get wrong. ↩︎
4. Eleni is herself a Russian-Greek bilingual, and the switching in the interviews served two functions: to supply examples in the language in which they were most naturally available, and to signal shared identity with the interviewer. We have written about these functions of translanguaging in an earlier chapter (Motsiou & Kostoulas, 2023, in Greek). ↩︎
5. The culprit here is Russian аканье (yakanye), the merger of unstressed [o] with [a]. A learner who applies it to Greek produces [ska’tono] for [sko’tono], both of which describe spectacularly failing at something, the former being markedly vulgar; likewise, [xor’topita], a pie with greens, can become [xar’topita], a pie that tastes like paper. These errors seem to persist well into advanced proficiency, which is exactly what one might expect of phonological rules operating below the level of conscious attention. ↩︎
6. It means ‘penis’, which can have mortifying consequences when students to practice comparative and superlative forms. ↩︎

---