Comprehension of Ukrainian by Estonians via Russian: Structural and extra-linguistic aspects

Full Text

1. Introduction

In the contemporary world, people often need to communicate across linguistic and cultural boundaries without having a perfect command of a foreign language. Very often English as a lingua franca or any local lingua franca are not an obvious choice in many regions and communicative situations. Therefore, interlocutors employ different language modes in order to make communication happen. One of these is receptive multilingualism (RM) a mode of communication where passive understanding of an L2 suffices: all participants use their L1 while speaking to each other (Rehbein et al. 2012). This mode is mostly employed (and investigated) in the case of related languages (inherent RM, e. g. Estonian-Finnish) but also in communication between speakers of unrelated languages where the participants have at least a passive command of each other's language (acquired RM, e. g. Estonian-Russian).

The subject of this study is the comprehension of Ukrainian without previous exposure to it among Estonians with some proficiency in Russian. Estonian and Ukrainian are not related and no significant bilingual community speaking these languages exists (although there are indeed a few individual cases of Estonian-Ukrainian bilingualism); yet speakers of Estonian may be able to comprehend Ukrainian through the knowledge of Russian. Knowing Russian as L2 should help to cope with Ukrainian as L3: they belong to the same language family (East-Slavic), have a lot of typological and lexical similarity (62% similarity in lexical composition, Tyshchenko 2010: 66). This mode of communication was termed “mediated receptive multilingualism,” where understanding of L3 can be achieved through the medium of L2 closely related to L3 (Branets et al. 2020).

The comprehension of Ukrainian among speakers of Estonian via their varying levels of proficiency in Russian was first examined by Branets, Bahtina & Verschik (2020). They found that Estonians were quite successful in reading comprehension of Ukrainian without previous exposure to it. It was attested that, in addition to structural and material similarities between Russian and Ukrainian, there are a number of extralinguistic factors that affected understanding, such as metalinguistic awareness, previous exposure to Russian and to various registers thereof, experience in multilingual communication, learnability, and attitudes towards Ukrainian (Branets et al. 2020).

The role of material and structural similarity in comprehension between closely related languages has enjoyed a lot of scholarly attention in the field of RM (Gooskens 2007a, Gooskens et al. 2008). Although similarity is highly relevant, there are other factors that may play a role, including experience in multilingual (or RM) communication, exposure to different varieties and registers (slang, regional dialects, colloquial speech; see Kaivapalu 2015), general cognitive skills (posing a hypothesis, making the comparison), and individual linguistic trajectories (personal experience, communicative needs, repertoire, Blommaert & Backus 2011). We agree with the view that language skills and language learning are shaped by use (meaning both active usage and passive exposure, see Barlow & Kemmer 2000, Blommaert & Backus 2011) and experience (Backus 2014, Bybee 2010, Croft 2001, Langacker 1987, Quick & Verschik 2019). In this study, we will analyse the participants' debriefing data where they explained their decisions. On the basis of these data, we were able to detect the participant's comprehension strategies that helped them to complete reading comprehension tasks in Ukrainian.

The aim of this paper is twofold. First, we investigate what role similarity (objective or perceived) played in the definition of Ukrainian words by speakers of Estonian in the reading comprehension experiment of Ukrainian. In addition, we focus on the participants' perspective of employing different linguistic resources that they may already have from the prior experience of communication in complex multilingual settings. This allows us to see the mechanism of how available linguistic resources are activated from a participant's perspective in a difficult multilingual situation.

Secondly, we explore what other factors, in addition to objective and perceived similarity, played a role. In our previous research on mediated RM, we found that proficiency in Russian in itself did not determine successful comprehension and provided a list of extra-linguistics factors that facilitate comprehension (Branets et al. 2020). In contrast to the previous study, here we examine only Estonians with Russian as L2 and do not include other groups such as Russian-dominant bilinguals, balanced Russian-Estonian bilinguals, etc. The number of Estonian as L1 speakers was increased from 20 to 30.

The paper is organized in the following way: first, we discuss theoretical premises of receptive multilingualism research with a focus on mediated receptive multilingualism. We will also provide a background on objective and perceived linguistic similarities and extra-linguistic factors. Then we describe the experimental design and the participants. After that, we proceed with our findings and data analysis. Finally, we complete the article with the main conclusions.

2. Theoretical considerations

The phenomenon of RM is covered by a variety of synonymous or near-synonymous terms in the literature: mutual intelligibility (Voegelin & Harris 1951), semicommunication (Haugen 1953, 1966, 1981), plurilingual communication (Lüdi 2007), intercomprehension (Berthele 2007), receptive multilingualism (Braunmüller 2007, Zeevaert 2004, ten Thije & Zeevaert 2007), lingua receptiva (LaRa) (Lingua Receptiva 2021¹, Rehbein et al. 2012, ten Thije et al. 2017). The main objective of RM is to activate linguistic, mental, interactional, and intercultural competencies of the interlocutor's passive language during RM interactions (Rehbein et al. 2012: 249).

Nowadays, many studies in RM theory as well as in language acquisition in general have shifted from "ideal bilingual,” perfect command and productive skills towards receptive skills, not necessarily perfect command, and to the purpose-based focus of reaching communicative goals in complex multilingual situations (Branets et al. 2020, Braunmüller 2007, ten Thije & Zeevaert 2007, Zeevaert 2004). Since successful communication is possible without “perfect” language use, communicative aspects of RM become central instead of formal aspects of language (Bahtina & ten Thije 2012).

The asymmetry between comprehension and production skills in receptive bilinguals has also been brought up in RM literature (Sherkina-Lieber 2015). However, RM has the potential for interlocutors with asymmetrical competencies to be effective by using suitable communicative strategies in exolingual interactions (Lüdi 2013). The potential of RM can be developed over time. Making full use of RM and of the resources that come with knowing another language takes time; continuing practice with the same interlocutors increases your common ground with them, and this makes you better at using effective communication strategies. The evidence of such processes was attested in the research of Czech-Croatian (Golubovic 2016) and Estonian-Russian-Ukrainian language constellations (Branets et al. 2020). In both studies, the respondents were divided into two groups: those who received instruction and those who did not. The results demonstrated a significant improvement in comprehension of the trained group. In addition, in the Estonian-Russian-Ukrainian constellation, the comprehension of three Ukrainian texts was tested while the texts were provided to participants in a different order. The participants’ comprehension of the last text was always higher as they learned from one text to another and consequently applied more advanced strategies (Ibid). This suggests the language learning trajectory of RM or learnability.

Some researchers have paid particular attention to linguistic facilitators of comprehension in RM by controlling extra-linguistic factors (Härmävaara & Gooskens 2019, Gooskens et al. 2015, Salehi & Neysani 2017). The notion of objective and perceived linguistic similarity was brought up a number of times (Gooskens et al. 2008, Kaivapalu & Martin 2017). Objective similarity (and difference) is the actual degree of correspondence between languages (Jarvis & Pavlenko 2008: 177). In turn, the perceived similarity is defined as “what language learners perceive to be similar between languages” (Ringbom 2007: 7). Perceived similarity does not always function in a positive way, but there also might be negative cases of understanding or misinterpretation.

Perceived similarity by language learners with a limited command of the target language is based on their L1 or other acquired languages, “especially if they are related to the target language” (Ringbom & Jarvis 2009: 106). In our case, L1 Estonian (Finnic, Uralic) is non-related to L2 Russian and L3 Ukrainian (East-Slavic, Indo-European); however, interlocutors could rely on their knowledge of L2 Russian that could positively affect the comprehension of L3 Ukrainian and facilitate a possibility of the acquisition of Ukrainian. Our previous study has shown that the comprehension of Ukrainian by the participants with Russian as L1 differs from the participants with Estonian as L1: namely, the participants with Russian as L1 were better at understanding Ukrainian than the participants with Estonian as L1. Yet, Russian-Estonian balanced bilinguals performed better than dominant Russian-speakers from Estonia, probably because of their higher metalinguistic awareness (Branets et al. 2020: 13–14).

As for extra-linguistic factors (social, individual, communicative, etc.), several authors mention attitudes, geographical distance, exposure, metalinguistic awareness, etc. (Gooskens 2006, 2007b, Gooskens & Schneider 2019, Kaivapalu 2015, Schüppert & Gooskens 2011, Gooskens & van Heuven 2019). The difference between perceived and objective similarities in comprehension experiments was also explained by the role of various non-linguistic factors. Kaivapalu (2015: 69) proposed a descriptive model of RM that, in addition to the degree of similarity between languages, includes such notions as various registers of L1 (colloquial usage, dialects, familiarity with slang and archaisms), metalinguistic awareness, general knowledge, random knowledge of some language items of the target language from the past, and the context. Several studies emphasized the important role of language variation that equips interlocutors with more advanced strategies of finding similarities between languages (Berthele 2008, Gooskens & Heeringa 2014, Kaivapalu & Maisa 2017). For instance, in inter-Scandinavian communication, Norwegians understand Danish and Swedish better than Danes, and Swedes understand Norwegian due to exposure to Norwegian dialects. It was suggested that exposure to a vast range of varieties raised language awareness among Norwegians and consequently helped them to establish linguistic cues and find similarities between closely related languages (Gooskens & Heeringa 2014). In addition to linguistic distance, Gooskens (2007a) highlighted the role of language attitudes (see also in Gooskens 2006, Schüppert & Gooskens 2011), contacts, and language experience with the language towards comprehension.

Various communication strategies in RM towards reaching comprehension have been attested, such as accommodation or reducing linguistic differences (Giles et al.1991, Hlavac 2014) or hearer’s and speaker’s metacommunicative practices that are provided naturally by the assistance of interlocutors to each other during a conversation in complex multilingual situations (Bahtina-Jantsikene & Backus 2016). In RM, the context and multimodal elements of interaction play an important role (Härmävaara & Gooskens 2019: 19, Muikku-Werner 2014). In reading comprehension, participants mostly rely on linguistic similarities; however, when they cannot find them, they turn to the context. In such cases, the context functions as a so-called filter that helps participants to confirm or refute their assumptions (Kaivapalu 2015, Kaivapalu & Muikku-Werner 2010). Also, according to Grosjean (1998), the conversation topic within the context affects the language mode and the comprehension process.

In a narrow sense, a context may mean the plot, the topic, preceding and following words and sentences. Another type is a wider cultural context, for instance, accidental familiarity with Russian or Ukrainian songs, culture, traditions, and so forth. In a broader sense, a context may mean knowledge about the world, including specialized knowledge in a certain field, for instance, how social networks function.

Thus, even though linguistic factors play an important role in comprehension, extra-linguistic factors such as cognitive, sociolinguistic, and individual should not be disregarded as material, and structural similarity itself does not guarantee intelligibility (Bahtina-Jantsikene 2013, Branets et al. 2020, Härmävaara 2014, Kaivapalu 2015, Muikku-Werner 2013, Verschik 2012).

3. Method and participants

A written comprehension experiment was carried out with 30 Estonian participants and consisted of a socio-linguistic questionnaire, a C-test in Russian (Grotjahn 1987), several tasks for individual Ukrainian words (Shumarova 2000), and a Ukrainian text as a whole (Gooskens 2013). The experiments were followed by debriefing interviews. Each experiment lasted approximately two hours and was conducted individually with every participant with pen and paper.

3.1. Participants

30 Estonian speakers with language proficiency in Russian on a B1 or B2 level participated in a reading comprehension experiment. The experiment was conducted in 2017 and 2018 in Tallinn. All respondents were living in Tallinn at the time the experiment was carried out. The group comprised ten males and 20 females, aged from 22 to 59 years. In comparison to the data presented in (Branets et al. 2020), we have increased the number of Estonian as L1 speakers from 20 to 30 in order to provide more precise findings, and we are not taking into consideration the results obtained from other groups of participants.

The participants of the experiment were chosen based on their language proficiency in Russian (Branets et al. 2020). B1 and B2 proficiency in Russian was determined to be enough to be able to complete the Ukrainian test based on the pilot study and was tested with a C-test in Russian (Grotjahn 1987).

Seven participants already have higher education, but most of them were university students at the moment of conducting the experiment. They study sociology, architecture, youth work, business administration, craft technologies, and design, recreation arrangement, dance and choreography, pedagogy, audiovisual media, social work, linguistics, administrative management, teaching, European languages, pharmacy, graphic design, anthropology, Asian studies, communication, physics, editing, music, and IT. It is evident that linguistics students have a higher degree of linguistic awareness than others, but there were only four such students among the participants, so we do not think they influenced the results.

3.2 Testing material and procedure

The testing material consisted of a questionnaire, C-test, three Ukrainian texts with tasks, and a debriefing. The questionnaire was used to establish the sociolinguistic background of participants and their exposure to Russian and Ukrainian. It consisted of 16 questions and was modeled on the questionnaire used in a previous study by Bahtina-Jantsikene (2013) on the acquired Russian-Estonian receptive multilingualism (see more in Branets et al. 2020).

The C-test was indicated as an optimal cross-language test for measuring comprehension in the European language area (Gooskens & van Heuven 2017). In our study, the C-test was used to test the participants' proficiency in Russian. It was developed according to the instructions presented by Grotjahn (1987) and evaluated on the basis of the scoring system proposed by Bahtina-Jantsikene (2013). The C-test comprised four short texts that were selected from different magazines. Every word was divided into two approximately equal parts, and the second part of every second word starting from the second sentence was deleted (see more in Branets et al. 2020). The participants’ task was to fill in the gaps using the correct word based on the context and the required grammatical form. The participants were given 20 minutes to complete the task (5 minutes per each small text).

The main part of the experiment explored comprehension of Ukrainian texts at the B1 level. The texts were selected from the collection of texts for B1 learners of Ukrainian and belonged to different genres (artistic and media texts). The respondents received three Ukrainian texts arranged in a different order. They were requested first to read the text and then to complete the tasks which were the same for each text. The tasks for Ukrainian texts consisted of two parts: definition of individual words from the text (Shumarova 2000) and tasks for the context comprehension (Gooskens 2013, as we do not focus on this group of tasks in this article, see more in Branets et al. 2020). In this paper, we will focus on the first task (definition of individual words from the text). For this task, we selected 55 words (based on the classification below). The participants were asked to translate or to explain them in their own words. They were also able to rely on the context, as all the words from the definition task were highlighted in the text.

The words belong to three groups: (1) 36 words have Russian cognates with the same meaning (Ukrainian знання (znannya) ‘knowledge’, cf. Russian знания (znaniya) ‘knowledge’); (2) 12 words that have Russian cognates with different meanings (Ukrainian чоловік (cholovik) ‘man, husband’, cf. Russian человек (chelovek) ‘human’) or cognates that belong to different registers, i.e., stylistically neutral in Ukrainian vs. colloquialisms, archaisms, regionalisms, etc. in Russian (Ukrainian очі (ochi) ‘eyes’, cf. Russian глаза (glaza) ‘eyes’ and Russian archaic/poetic очи (ochi) ‘eyes’); (3) seven words that do not have Russian cognates (Ukrainian цікавий (tsikavyi) ‘interesting’ cf. Russian интересный (interesnyi) ‘interesting’). Word recognition tasks included nouns, verbs, adjectives, adverbs, prepositions, and numerals. The same scoring system was applied as for the Russian C-test (more details are outlined in Branets et al. 2020):

• 1 point: an entirely correct answer (e.g., when a participant recognizes that Ukrainian казка (kazka) ‘fairytale’ as Estonian muinasjutt ‘fairytale’ etc.)
• 75 points: a correct definition presented in an incorrect grammatical form (e.g., Ukrainian любляче (lyublyache) ‘loving’ cf. Estonian armastus ‘love’ etc.)
• 5 points: almost correct meaning (e.g., щодня (schodnya)‘every day’ as Estonian päev ‘day’ instead of correct iga päev ‘every day’)
• 25 points: a semantically related lexeme that fits the context but is incorrect (e.g., Ukrainian cторінка (storinka) ‘page’ as Estonian sein ‘wall (on Facebook)’
• 0 points: a completely wrong answer (e.g., Ukrainian розлучень (rozluchen’) ‘divorce’, genitive plural as Estonian suhe ‘relationship’) or no answer.

The last stage of our experiment was debriefing in order to collect the participants' comments and explanations and to detect the strategies they used. First, the participants were asked to describe their level of Ukrainian texts comprehension in their own words. Five participants decided to use percentages in order to describe their level of comprehension, i.e., “I understood 60% of the meaning of the texts”. Then the tasks for each text were discussed separately. The participants were asked to explain why they gave their definition for each word and to retell the story of each text. In the end, they were asked which text and which group of tasks (for individual words or meanings) was easier for them to understand. It allowed us to check the learnability effect since we randomized the order of the texts. The duration of the debriefing varied from 10 to 20 minutes, depending on each participant.

4. Results

4.1. Self-evaluated comprehension

After completing the tasks, all the participants were asked to describe their understanding of the Ukrainian texts in their own words^[2]. They reported a level of comprehension averaged at 62% (SD = 10.65). In general, the respondents did not expect to understand Ukrainian without previous exposure to it and were surprised by their results. The participants reported that they needed to read the text several times in order to understand it. One of the participants made a comment: “After the first reading, the level of understanding was 10-20%, and after the second time the comprehension grew up to 60-70%”. However, another participant said: “The understanding depended on how many times I read the text. The first sentence was clear from the beginning. After the first reading, I already understood 50% of the text's meaning”.

4.2. Measured actual comprehension

The actual level of comprehension of Ukrainian separate words and context was established to be 70.55% (SD = 11.19), with averages for context understanding reaching higher than averages for the understanding of separate words (83.98% (SD = 4.08) and 61.76% (SD = 8.01), understanding of context and separate words respectively). More specifically, success in the word recognition task was calculated separately for each group of words that participants received for definition (see section 3.2) and is presented in Table 1.

Table 1. Level of success of different groups of words in the word recognition task

Name of the group of words	Number of words	Maximum number of points for 30 participants	Success score in points	Success rate in %	SD range
Cognates with the same meaning	36	1080 (36 x 30)	760.5	70.4%	7.57
Cognates with different meanings	12	360 (12 x 30)	193.75	53.82%	4.60
Unrelated words	7	210 (7 x 30)	64.75	30.83%	7.02
Mean score of understanding of separate words				61.76%	8.01

5. Analysis

The results show that the respondents with L1 Estonian were quite successful in understanding Ukrainian via their knowledge of Russian. Based on average percentages for self-reported text comprehension (62%) and measured success (70.55%), there was no significant discrepancy; however, the participants provided a slightly lower percentage for self-comprehension than the actual results showed. Furthermore, we will look more closely into the performance results of each separate group of words using the participants' comments and explanations. The last subsection will be dedicated to extra-linguistic factors.

5.1. Cognates with the same meaning

As expected, the success level of recognition of the words that are cognates and have the same meaning is the highest among other groups of words. In general, the comprehension of cognates was constructed on the objective similarity between Russian and Ukrainian. The participants' main strategy within this group of words was to find similarities with Russian and then to confirm their hypothesis with the context. Most of the results dealing with this group of words (see Table 1) were positive (70.4 %, see Table 1) and depended on the participants' proficiency in Russian, context, and other factors, according to the information provided by the participants during the debriefing (see Branets & Backus 2020 for a more detailed discussion of individual proficiency and test results).

Similarity ignored (with both positive and negative effects)

The following examples present the cases when the participants ignored the similarity even if it was obvious and instead turned to the context that in some cases was not helpful. For instance, when we review the answers on the Ukrainian word життя (zhyttya) ‘life’, we observe the following:

Table 2. Example 1. Similarity ignored between cognates with the same meaning

Ukrainian	Russian	answers	Correct Estonian
життя zhyttya ‘life’	жизнь zhyzn’ ‘life’	ühiskond ‘society’	elu ‘life’
		elanike ‘of residents’

Even though the Ukrainian word is very similar to the Russian жизнь (zhyzn’) ‘life’, in the first case, the participant's explanation was as follows: “I did not look into similarities with Russian here and decided to get the meaning from the context and the word ühiskond ‘society’ perfectly fits the context”. In another case, the participant took into consideration only the similarity with Russian word жители (zhiteli) ‘residents’ and interpreted it as elanike ‘of residents’, yet failed to provide the correct definition. Concerning the recognition of this particular word in general, only one participant left a blank space, and twenty gave the correct definition elu ‘life’. The other seven participants used different grammatical forms of elu ‘life’: eludes ‘in the lives’, elama ‘to live’, eludele ‘to the lives’, eludesse ‘into the lives’, в жизни (v zhizni) ‘in life’, elus ‘alive’ (used twice).

The same tendency when the participants relied more on the context was observed with other words but with a positive effect. For instance, for the definition of the Ukrainian word вчитель (vchytel’) ‘teacher’, two participants chose close but not entirely correct answers based on the context. Instead of giving a definition as ‘teacher’, one of the participants wrote õpetatud mees ‘learned men’ which basically corresponds to the meaning of ‘teacher’. The same happened with the Ukrainian lexeme казка (kazka) ‘fairytale’ in seven participants: it is very similar to the Russian сказка (skazka) ‘fairytale’ but was interpreted as lugu ‘story’ or jutuke ‘short story’. This word was recognised correctly by 27 participants. In both examples вчитель (vchytel’) ‘teacher’ and казка (kazka) ‘fairytale’, the lexical meanings of the definitions were very close to the target meanings.

The following definitions were given based on the context rather than similarity by two participants who provided similar answers in Table 3.

Table 3. Example 2. Similarity ignored between cognates with the same meaning

Ukrainian	Russian	answers	correct Estonian
донька don’ka ‘daughter’	дочка dochka ‘daughter’	daam ‘lady’	tütar ‘daughter’
		tütar ‘daughter’

In general, 26 participants provided the correct answer tütar ‘daughter’, two left an empty space, and two provided a totally incorrect meaning. Interestingly, out of 26 participants, two participants wrote two answers: daam ‘lady’ and tütar ‘daughter’. The word daam ‘lady’ has a similar sound and meaning with the Russian дама (dama) ‘lady’ but has nothing to do with the Russian дочка (dochka) ‘daughter’. These two participants explained in example 3 that, based on the context, they assumed that it should be a female and then arrived at the conclusion that it was ‘daughter’.

Table 4. Example 3. Similarity ignored between cognates with the same meaning

“The Ukrainian донька (don’ka) ‘daughter’ is similar to the word дочка (dochka) ‘daughter’ in Russian but there is a possibility that it might mean something else, so I used the context to recognise it”.

The level of exposure to Russian was indicated by the participants as one of the factors that helped them to understand the lexical items:

Table 5. Example 4. Similarity ignored between cognates with the same meaning

Ukrainian	Russian	answer	correct Estonia
тривога tryvoga ‘anxiety’, ‘alarm’	тревога trevoga ‘anxiety’, ‘alarm’	hoiatus ‘warning’	ärevus, rahutus ‘anxiety’

The respondent provided a definition to the word based on the Russian song about the war Тревога, тревога (Trevoga, trevoga) ‘Alarm, alarm’ where the word тревога (trevoga) had the meaning ‘alarm’. However, in this particular context, the correct meaning was ‘anxiety’. Five more participants interpreted this word as häire ‘alarm’. In total, based on both similarities with the Russian word and the context, the lexeme was interpreted correctly only ten times (two times mure ‘concern’; two times ärevus ‘anxiety’ and three times тревогa^[3] (trevoga) ‘anxiety’) by the respondents from the older group that had more exposure to Russian during the Soviet time.

The confusion caused by different inflections

When participants relied only on similarities, perceived or objective, between Russian and Ukrainian and could not understand the meaning of the words, did not implement any other strategies to identify the words, they often were not able to recognise the meaning of the words correctly. We observed that in most cases, the participants were challenged by the cognates in Russian and Ukrainian that have the same stem but different inflections. In such cases, these words became either unrecognisable for some participants (see Table 6) or were interpreted by words with other morphemes in Russian that have different meanings (see Table 7).

The Ukrainian item щовечора (schovechora) has the component що- (scho-) that means ‘every’ and stem вечора (vechora) that corresponds to the Russian вечер (vecher) ‘evening’. This word was reported by 10 participants as unknown and defined five times with completely wrong meanings, for instance, nõuanne ‘advice’, südametunnistus ‘conscience’, täiesti ‘completely’, совершенное (sovershennoe) ‘perfect’, pesema ‘to wash’. However, in nine cases, this word was recognised correctly by the participants, and in six cases partially (only the meaning of the stem: Ukrainian вечора (vechora) 'evening’ cf. Russian вечера (vechora) ‘of evening’, for instance õhtuti ‘in the evenings’, õhtu ‘evening’, õhtul ‘in the evening’).

Table 6. Example 1. The confusion caused by different inflections between cognates with the same meaning

Ukrainian	Russian	answers	correct Estonian
щовечора schovechora ‘every evening’	каждый вечер kazhdyi vecher ‘every evening’	каждый вечерь kazhdyi vecher’ ‘every evening’	iga(l) õhtu(l) ‘every evening’
		каждый вечер kazhdyi vecher ‘every evening’
		igal õhtul ‘every evening’
		õhtuti ‘in the evenings’
		õhtu ‘evening’
		iga õhtu ‘every evening’
		õhtul ‘in the evening’
		nõuanne ‘advice’
		südametunnistus ‘conscience’
		täiesti ‘completely’
		совершенное sovershennoe ‘perfect’
		pesema ‘to wash’

The next example (Table 7) presents the case when the Ukrainian word йшли (jshly)‘went’ that has a cognate in Russian шли (shli) ‘went’ was misinterpreted because of a slightly different form in Russian. It was confused with a similar sounding Russian word, derived from the same stem but with a different prefix: нашли (nashli) ‘found’. It was reported that this definition was given due to the similarities with Russian.

Table 7. Example 2. The confusion caused by different inflections between cognates with the same meaning

Ukrainian	Russian	answer	correct Estonian
йшли jshly ‘went’	шли shli ‘went’	otsisid ‘looked for’	läksid ‘went’

The Ukrainian word сторінка (storinka) ‘page’ appeared to be challenging for definition. Some participants that did not find similarities with the Russian страница (stranitsa) ‘page’, quite successfully used the context to derive the meaning. 

Table 8. Example 3. The confusion caused by different inflections between cognates with the same meaning

Ukrainian	Russian	answers	correct Estonian
cторінка storinka ‘page’	страничка stranichka ‘page’	lehekülg, leht ‘page’	lehekülg, leht ‘site, page’
		sein ‘wall (on Facebook)’	konto ‘account’
		külg ‘side’
		lugu ‘story’

Even though the following versions of interpretation are not exactly correct, they would fit the context. More specifically, two participants defined this word as sein ‘wall’ and two as konto ‘account’ and explained that they were not able to find similarities with Russian and used the context. Both meanings suited well in the context (see examples 4 and 5 of Table 9).

Table 9. Examples 4–5. The confusion caused by different inflections between cognates with
the same meaning

Example 4 “I used the word sein ‘wall’ because in the next paragraph the statistics about Facebook was mentioned”.

Example 5 “The sentence started with ‘80% users’, and I assumed that the word means konto ‘account”.

On the contrary, two respondents defined it as külg ‘side’ and two as lugu ‘story’ by looking into similarities with the Russian сторона (storona) ‘side’ and история (istoriya) ‘story’. However, both suggestions were not correct, which consequently affected the general understanding of the text in a negative way. In total, only six respondents answered as lehekülg, leht ‘page’.

Inability to recognize cognates

When the participants were not aware of a cognate in Russian and were not able to use the context, they experienced problems with providing a correct definition:

Table 10. Example 1. Inability to recognise cognates with the same meaning

Ukrainian	Russian	answers	correct Estonian
ніяк niyak ‘by no means’	никак nikak ‘by no means’	mitte kuidagi ‘by no means’	mitte kuidagi ‘by no means’
		kuidagi ‘somehow’
		mitte ‘no’, ‘not’
		mitte ükski ‘no one’
		kunagi ‘once’
		kuidagi ‘somehow’
		mitte midagi ‘nothing’
		mitte kedagi ‘nobody’

Twelve participants provided the correct answer mitte kuidagi ‘by no means’, three participants defined it as kuidagi ‘somehow’. The rest were challenged to find similarities with Russian as well as support from the context and derived different answers based on the assumptions as listed in Table 8, which are not correct.

One more example of such occurrence is the Ukrainian lexeme протягом (protyagom) ‘during’ that turned out to be the most difficult to define. Although it is a cognate with the Russian на протяжении (na protyazhenii) ‘during’, it is rare in everyday colloquial speech and mostly used in written genres. Our participants did not have much exposure to written genres, i. e. to media, fiction, Russian internet sites, etc. Some assumptions were made that this word could mean tõmme ‘draw’ (noun) or протягивать (protyagivat’) ‘to stretch (out)’, based on the similarities with the Russian тянуть (tyanut’) ‘to pull’. Apparently, the participants recognised the stem (cf. tõmbama ‘to draw, to pull’), but here we deal with a conventionalized, grammaticalized metaphor in Russian/Ukrainian, the meaning of which is difficult to derive because the Estonian ‘during’ has a different underlying metaphor. The postposition jooksul, literally ‘in the run’, is derived from jooks ‘run’ (the allative case); similarly, ajal ‘at the time’ is derived from aeg ‘time’ (the allative case). One participant conveyed that his/her definition was based on the assonance with Russian противно (protivno) ‘disgusting’. Another respondent suggested the English protect because it sounds similar, but neither of these meanings was correct.

Table 11. Example 2. Inability to recognise cognates with the same meaning

Ukrainian	Russian	answers	correct Estonian
протягом protyagom ‘during’	на протяжении na protjazhenii ‘during’	tõmme ‘draw’	ajal, jooksul ‘during’
		vaenlane ‘enemy’
		протягивать protyagivat’ ‘stretch’
		противный protivnyi ‘disgusting’
		protect

Table 12 presents the case where the impact of similarity together with the context was positive.

Table 12. Example 3. Inability to recognise cognates with the same meaning

Ukrainian	Russian	answers	correct Estonian
важливого vazhlyvogo ‘important’	важного vazhnogo ‘important’	tähtis, oluline ‘important’	tähtis, oluline ‘important’
		olulisemat ‘more important’
		kõige tähtsam ‘most important’

The Ukrainian word важливий (vazhlyvyi) ‘important’ was interpreted 20 times correctly. One participant recognised the word ‘important’ in a comparative form olulisemat ‘more important’ (partitive). The participant used partitive, an object case that corresponds to the accusative in Ukrainian, i.e., the grammatical form in which the word was presented in the text. Two more respondents identified it in the superlative form kõige tähtsam ‘most important’ due to the unfamiliar ending of важливий (vazhlyvyi) ‘important’. Interestingly, in our previous study, the participants with L1 Russian and Russian-Estonian simultaneous bilinguals confused the meaning of this word with the paronym in Russian вежливый (vezhlivyi) ‘polite’ because the words look and sound alike. However, only one participant with L1 Estonian first wrote вежливый (vezhlivyi) ‘polite’ and then crossed it out and gave a definition важный (vazhnyi) ‘important’ due to the confirmation from the context. It shows the difference between cognitive processes and strategies that are applied by L1 and L2 language speakers.

In some cases the participants provided false answers due to the so-called false friends with Russian, as in Tables 13 and 14.

Table 13. Example 4. Inability to recognise cognates with the same meaning

Ukrainian	Russian	answers	correct Estonian
перевірити pereviryty ‘to check’	проверить proverit’ ‘to check’	tõlkima ‘to translate’	kontrollima ‘to check’
		pöörduda ‘to turn to’
		ette valmistama ‘to prepare’
		ümber pöörata ‘to turn around’
		proovile panema ‘to test’

Only seven participants identified the word перевірити (pereviryty) ‘to check’ correctly. Based on the perceived similarities with several Russian words, three respondents confused this word with the Russian перевести (perevesti) ‘to translate’; one respondent with the Russian приготовить (prigotovit’) ‘to prepare’; two participants suggested the Russian повернуться (povernutsya) ‘to turn around’. Two participants recognised it as pöörduda ‘to turn to’ that is not correct but fits the context, and two more participants as proovile panema ‘to test, to challenge’ (correct definition).

In the same vein, the lexeme віддати (viddaty) ‘to give away’ was in many cases confused with the Russian видеть (videt’) ‘to see’.

Table 14. Example 5. Inability to recognise cognates with the same meaning

Ukrainian	Russian	answers	correct Estonian
віддам viddam ‘to give away’	отдам otdam      ‘to give away’	nägin ‘(I) saw’	annan ära   ‘(I) give away’
		vaatama ‘to look’

The Ukrainian віддам (viddam) ‘(I) will give away’ was defined by seven participants as nägema ‘to see’ or vaatama ‘to look’ due to the perceived similarity with the Russian видать (vidat’), видеть (videt’) ‘to see’. At the same time, 16 respondents provided the correct answer as annan ära ‘(I will) give away’ based on the context.

5.2. Cognates with different meanings

This group presents less correct answers, as it includes cognates with different meanings or cognates that in Russian belong to different registers and are used with a different frequency than in Ukrainian. Within this group of words, more various strategies and factors came into play.

The positive role of context

Table 15 presents the Ukrainian word мережа (merezha) ‘network’ that has a cognate in Russian мережка (merezhka) ‘a technique used in embroidery’ with quite a different and rather specific meaning unknown even to many native speakers of Russian (unless they know something about embroidery). The chances that a B1 learner/user of Russian would have encountered this item are slim, so the respondents were unable to draw parallels with Russian:

Table 15. Example 1. The positive role of the context between cognates with different meanings

Ukrainian	Russian	Estonian
мережа merezha ‘network’	сеть set’ ‘network’	võrgustik ‘network’

20 participants understood the meaning correctly, based on the general knowledge about social media. Their explanations were as follows:

Table 16. Examples 2-5. The positive role of the context between cognates with different meanings

Example 2 “I understood мережа (merezha) ‘network’ as it reminded me the word мир (mir) ‘world’ and then since it was used together with Ukrainian word соціальний (sotsial’nyj) ‘social’ that is similar to Russian социальный (sotsial’nyj) ‘social’, I figured out that it is võrgustik ‘network”.

Example 3 “I did not understand мережа (merezha) ‘network’ from the beginning, but somewhere at the end of the first paragraph because of the context I understood that it means võrgustik ‘network”.

Example 4 “I did not know this word at first, but then I found some information in the text about an account and FB, and I assumed that it might be võrgustik ‘network”.

Example 5 “I heard this word somewhere. I cannot remember where but I knew that it was võrgustik ‘network”.

Table 17 demonstrates how the context outweighs perceived similarity.

Table 17. Example 6. The positive role of the context between cognates with different meanings

Ukrainian	Russian	answers	correct Estonian
увійти uvijty ‘to enter’	войти vojti ‘to enter’	väljuma ‘to leave’	sissenema ‘to enter’, ‘to log in’
		ära minema ‘to leave’
		sissenema ‘to enter’, to log in’
		vaatama ‘to look’

Twelve participants confused увійти (uvijty) ‘to enter’, ‘to log in’ with the Russian увидеть (uvidet’), видеть (videt’) ‘to see’ because it sounded similar. Eight participants defined it as väljuma ‘to leave’, ‘to log out’ because of the Russian выйти (vyjti) ‘to leave’, ‘to exit’, ‘to log out’, and three participants gave a definition as ära minna ‘to go out, to leave’ due to the Russian уйти (ujti) ‘to leave’. In this example, the perceived similarity with Russian had a negative effect as only two participants provided the correct answer and were asked to explain their decision:

Table 18. Examples 7–8. The positive role of the context between cognates with different meanings

Example 7 “I wrote first ära minema ‘to go away, to leave’ because it was similar to the Russian уйди (ujdi) ‘to go out, to leave’ but then I changed it to sisenema ‘to log in’ according to the context”.

Example 8 “I derived the meaning from the context as the next words were в свій аккаунт (v svij akkaunt) ‘into your account’”.

In both examples 7 and 8 (Table 18), the context outweighed the perceived similarity with Russian. These two factors could be considered as competing. This requires more research because we cannot say in which case exactly the context and general knowledge appear more relevant than similarity.

The same process was observed with the Ukrainian word очі (ochi) ‘eyes’ that has a cognate in the archaic Russian очи (ochi) ‘eyes’ that is used only in limited contexts (poetic, high style etc.). A stylistically neutral lexeme is глаза (glaza) ‘eyes’ (see also the discussion in Branets et al. 2020: 19).

Table 19. Example 9. The positive role of the context between cognates with different meanings

Ukrainian	Russian	answer	correct Estonian
oчі ochi ‘eyes’	глаза glaza ‘eyes’	silmad ‘eyes’	silmad ‘eyes’

17 participants provided the correct definition. One participant provided the definition очки (ochki) ‘glasses’ based on linguistic similarity. Three participants mentioned that they knew this word from the well-known Russian song Очи черные (Ochi chernye) ‘black eyes’ and 14 mentioned that they turned to the Russian word очки (ochki) ‘glasses’ that has the same stem as the Ukrainian очi (ochi) ‘eyes’.

Table 20. Examples 10-11. The positive role of the context between cognates with different meanings

Example 10  “At first I wrote prillid ‘glasses’ but then I figured out that these are silmad ‘eyes”.

Example 11 “I wrote prillid ‘glasses’ and it did not match the context, so I wrote silmad ‘eyes”.

In both examples 10 and 11 (Table 20), the participants were searching for confirmation from the context instead of relying on similarity.

In Table 21, the Ukrainian lexeme чоботи (choboty) ‘boots’ has a Russian cognate чоботы (choboty) that means a certain kind of boots and is used in regional varieties. Thus, the range of meanings and connotations in the two languages differ:

Table 21. Example 12. The positive role of the context between cognates with different meanings

Ukrainian	Russian	answers	correct Estonian
чоботи choboty ‘boots’	ботинки botinki ‘boots’	saapad  ‘boots’	saapad ‘boots’

This word was defined correctly by 21 participants. Most of them derived the meaning from the context. Some explained their choice with the similarity to the Russian stem бот- (bot-) in the word ботинки (botinki) ‘boots’ that appeared similar but is not a cognate. However, this accidental similarity helped the participants to find the correct meaning.

Difficult instances where the context does not help

The next Table 22 represents the definition of the superlative from the Ukrainian word великий (velykyj) ‘big’ that has a Russian cognate великий (velikii) ‘outstanding, great, famous’:

Table 22. Example 1. The difficult instance where the context did not help to recognise cognates with different meanings

Ukrainian	Russian	answers	correct Estonian
величезна velychezna ‘huge’	oгромная ogromnaya ‘huge’	suur ‘big’	tohutu ‘huge’
		suurendama ‘to increase’
		ületahtsustatud ‘overrated’
		võimsus ‘power’
		suurenenud ‘increased’, ‘augmented’
		suursugune ‘majestic’
		palju ‘many’
		enamus ‘majority’
		suurus ‘greatness’, ‘size’

No one provided a correct definition for this word. Three participants recognised it as suur ‘big’, one as suurendama ‘to increase’ and one more as ületahtsustatud ‘overrated’ due to the similarity with the stem in the Russian великий (velikii) ‘outstanding, great, famous’, увеличивать (uvelichivat’) ‘to increase’, преувеличивать (preuvelichivat’) ‘to exaggerate’ respectively. One participant interpreted it as võimsus ‘power’, two as suurenenud ‘increased’, ‘augmented’ and one as suursugune ‘majestic’. The participant commented: “I was familiar with this word from Russian fiction”. Apparently, these respondents were more exposed to Russian and were likely to have encountered this word. Three respondents assumed that it could mean palju ‘many’; one suggested enamus ‘majority’. They explained their choice as the assumption that it could be a part of a measurement component. Three more participants suggested suurus ‘greatness’, ‘size’, so the suggestion in its first meaning ‘greatness’ is not entirely wrong (but the part of speech is incorrect). The participants mentioned that they did not use the context to define this particular word.

In Table 23, the meaning of the Ukrainian одержати (oderzhaty) ‘receive’ was derived from the Russian cognate держать (derzhat’) ‘to keep, to hold’ with a slightly different meaning. However, there is also a similar Russian lexeme одержать (oderzhat’) ‘to receive’ (derived from the same stem), but it is used only in fixed expressions like одержать верх (oderzhat’ verh), одержать победу (oderzhat’ pobedu) ‘to win’, ‘to overcome’ that are more typical of written genres. Apparently, the participants had not been exposed to this false friend.

Table 23. Example 2. The difficult instance where the context did not help to recognise cognates with different meanings

Ukrainian	Russian	answers	correct Estonian
одержати oderzhaty ‘receive’	получить poluchit’ ‘receive’	saada ‘to receive’	saada ‘receive’
		omandama ‘to acquire’
		võtta ‘to take’
		hoidma ‘to keep’, ‘to hold’

Ten respondents understood this word correctly. Four participants identified it as omandama ‘to acquire’ and two participants as võtta ‘to take’ that is somewhat similar to saada ‘to receive’. Four participants gave a definition as hoidma ‘to keep’, ‘to hold’ because of similarities with the Russian держать (derzhat’) ‘to keep’ that is not correct.

5.3. Unrelated words

When similarities with Russian were not available, participants applied different strategies in order to recognise the meanings of the words.

Context and knowledge of the world

In most cases, they were trying to understand the meaning from the context by using general knowledge of the world or assumptions. For instance, our next case presents the case when all the aforementioned strategies were implemented.

Most of the respondents found the meaning from the context: four respondents defined the word as uurijad ‘researchers’, eight participants as teadlased ‘scientists’, and one as uurimus ‘research’. In general, they explained that since this word was followed in the text by the verb провели (provely) ‘conducted’ that was easily recognisable due to similarities with the Russian провели (proveli) ‘conducted’, they assumed that it should be either researchers or scientists who conduct the research or the research itself that could be conducted. One participant recognised it as psühholoogid ‘psychologists’ because the next paragraph was about relationships.

Table 24. Example 1. Context, knowledge of the world and unrelated words

Ukrainian	Russian	answers	correct Estonian
фахівці fachivtsi ‘experts’	эксперты eksperty ‘experts’	psühholoogid ‘psycologists’	eksperdid ‘experts’
		faktid ‘facts’	asjatundjad ‘experts’
		uurijad ‘researchers’
		teadlased ‘scientists’
		uurimus ‘research’
		ametniku ‘official’ (noun)

One participant defined it as ametnik ‘official’ (noun) because the Ukrainian word фахівці (fachivtsi) ‘experts’ resembled the German das Fach ‘speciality’ which is etymologically correct because фах (fach) ‘speciality’, ‘profession’ is a German borrowing in Ukrainian. So, according to this logic, an official is someone who deals with a particular specialty. Of course, the respondents would not necessarily know this, but here the parallel is correct. One more participant did not write an answer but, during the debriefing, shared the following (see Table 25).

Table 25. Example 2. Context, knowledge of the world and unrelated words

“It reminded me of the word Fach ‘specialty’ from German, but I was not sure if I could use it in this case”.

Two more participants relied on the similarities with the Russian факты (fakty) ‘facts’ or maybe also with the Estonian faktid ‘facts’, and consequently identified this word as faktid ‘facts’ that is incorrect.

Knowledge of other languages and meta-linguistic awareness

The previous example 2 in Table 22 demonstrates, in addition to other things, how knowledge of other languages may be useful, at least to some extent. According to the concept of foreign language mode (Selinker & Baumgartner-Cohen 1995), language learners of L3 rely more on their knowledge of L2 rather than on L1 when they have high proficiency in L2. In the Estonian-Russian-Ukrainian comprehension experiment, the direction of lexical transfer was L2 to L3 in most cases, as expected. There are rare cases of transfer from L1 to L3. When the participants were not familiar with the word in Russian and were unable to establish connections from the context, they turned to search for help in their L1:

Table 26. Example 1. Knowledge of other languages, meta-linguistic awareness and unrelated words

Ukrainian	Russian	answer	correct Estonian
у колі u koli ‘among’	в кругу, среди      v krugu, sredi ‘among’	koolis, в школе v shkole ‘in school’	hulgas, seas, keskel, vahel ‘among’

Three participants answered that they found у колі (u koli) ‘among’ similar to Estonian koolis ‘in school’ and two participants indicated that it was similar to both Estonian koolis ‘in school’ and Russian в школе (v shkole) ‘in school’. In total, only eight participants provided the correct definition to this word from the context and structure of the sentence:

Table 27. Examples 2-3. Knowledge of other languages, meta-linguistic awareness and unrelated words

Example 2 “I thought that it is seas ‘among’, like among the community of psychologists”.

Example 3 “It fitted the context, as seas ‘among’ was the first word in the sentence and the next word was ‘psychologists’”.

Two participants recognised this word as vahel ‘between’, ‘among’ that is also correct. One of them reported in Table 28 below.

Table 28. Example 4. Knowledge of other languages, meta-linguistic awareness and unrelated words

“I understood it as vahel ‘between’, ‘among’, as it was applicable to the context”.

Table 29 represents the case when the meaning of the word was interpreted correctly only three times due to unrelated lexemes; however, due to their meta-linguistic awareness, all participants listed the correct part of speech, e. g. verb:

Table 29. Example 5. Knowledge of other languages, meta-linguistic awareness and unrelated words

Ukrainian	Russian	answers	correct Estonian
запитав zapytav ‘(s/he) asked’	спросил sprosil ‘(s/he) asked’	meelde tuletama ‘remind’	küsis ‘(s/he) asked’
		mõtleb üle ‘thinks over’
		vastas ‘(s/he) replied’
		otsustas ‘(s/he) decided’
		meenutas ‘(s/he)recalled’
		mõtlema ‘to think’
		ütles ‘(s/he) said’
		andis nõu ‘(s/he) gave advice’
		lisan ‘(I) add’

The word запитав (zapytav) ‘asked’ was interpreted 12 times as vastas ‘replied’ and three times as ütles ‘said’. Even though the answer is not correct, it perfectly fits into the context, as well as the rest of the answers listed above.

One more example 6 (Table 30) is in line with the previous case:

Table 30. Example 6. Knowledge of other languages, meta-linguistic awareness and unrelated words

Ukrainian	Russian	answers	correct Estonian
цікавий tsikavyi ‘interesting’	интересный interesnyi ‘interesting’	tähtis ‘important’	huvitav ‘interesting’
		oluline ‘important’
		osaline ‘partial’
		uus ‘new’

Only one participant defined this word correctly. Five respondents recognised that it should be an adjective and provided definitions according to their assumptions: tähtis, oluline ‘important’; osaline ‘partial’; uus ‘new’ that are incorrect. One participant commented: “I thought that it should be an adjective, and I found one that fits the context”. Another participant did not provide any definition but instead wrote ‘adjective’. In this case, the participants' strategy was first to establish which part of speech the word represented, and only then they formed their assumption about the meaning.

The participants were asked to define one lexical item that is an established common borrowing from English in Ukrainian, Russian, and Estonian and specific to social media.

Table 31. Example 7. Knowledge of other languages, meta-linguistic awareness and unrelated words

Ukrainian	English	Estonian
лайкнути laiknuty ‘to like (on social media)’	to like	meeldima, laikima (colloquial) ‘to like’, ‘to like (on social media)’

Only five participants did not recognise this word and commented: “I knew this word as it is international but maybe because it is written in Cyrillic, I did not recognize it”. However, when this word was read out loud, the listener’s perception skills were activated, and the word was recognised immediately. Naturally, the perception of items in another alphabet is slower. Thus, it might also be caused by the level of meta-linguistic awareness of the language structures.

5.4. The role of non-linguistic factors

Different extra-linguistic factors affected the success of comprehension. We have found numerous evidence from the participants' comments about the factors that helped them to cope with the task.

Exposure to Russian

In our previous study, we emphasized the importance of exposure to Russian based on the environment, professional activities, and individual level (Branets et al. 2020: 17–18, Branets & Bahtina accepted). During the feedback session, the participants reported that such factors enhance their comprehension of Ukrainian (see Table 32 below). 

Table 32. Examples 1–2. Exposure to Russian

Example 1 “Because I use Russian at work, it was easy for me to understand the text”.

Example 2 “I understood the texts very well because I use Russian quite often. I have many Russian friends”.

Exposure to registers in Russian

Exposure to different registers such as colloquial and regional registers as well as to high language style is beneficial in comprehending Ukrainian texts. Examples in Tables 19 and 23 above belong to the cases when the exposure to archaisms and regional registers respectively foster the comprehension process. See below
Table 33 with some more comments from the participants.

Table 33. Examples 1–3. Exposure to registers in Russian

Example 1 “I recognised хата (chata) ‘house’ because I heard a poem and a song in Russian with this word”.

Example 2 “I understood батько (bat’ko) ‘father’ because of the word батя (batya) ‘father’”.

Example 3 “The word батько (bat’ko) ‘father’ is similar to батюшка (batyushka) ‘priest’”.

Example 2 in Table 33 presents the case of the colloquialism батя (batya) ‘father’ that has different connotations (characteristic of uneducated speech or regional colloquial use, etc.) than the stylistically neutral отец (otets) ‘father’ (see also Branets et al. 2020: 18). In example 3 (Table 33), it resembles the colloquial name for orthodox priest батюшка (batyushka, could also have an archaic meaning of a father); common Standard Russian священник (svyaschennyk) ‘priest’.

General knowledge

Different types of familiarity with the texts were detected depending on the field of occupation and background, general knowledge of the topic, or some individual factors. For example, the text about social media was easier for some participants that knew this topic well than other texts the topic of which was less familiar. Likewise, some participants reported that fairytales were more predictable for them than social media.

Table 34. Examples 1-3. General knowledge

Example 1 “It is more like a standard text that you can find on the Internet, so when you read about social networks, you can predict what might be said there”.

Example 2 “In other texts, I used more similarities with Russian, but in the social media text I used more context that was closer to daily life like in everyday use”.

Example 3 “Fairytales were more predictable for me: a standard beginning of the story, typical characters like an old man and his daughter here, the traditional development of the story and a happy ending made it easy to understand”.

Learnability

The emergent nature of language acquisition was taken into account for our experiment. According to the usage-based approach, the participants learn about form and meaning “in use” on a daily basis (Tomasello 2003). In our experiment, we have tested learnability by randomising the order of the texts and providing instructions about similarities and differences between Ukrainian and Russian (see more in Branets et al. 2020). We consider learnability as a general cognitive process of the development of explicit and implicit skills by participants. The respondents reported that they learned from one text to another, and in most cases, every next text was easier to understand (see Table 35).

Table 35. Examples 1-3. Learnability

Example 1 “Repetitiveness of the words helped me to understand the third text best of all. Such words as по-перше (po-pershe) ‘first’ etc. were repetitive. I got used to Ukrainian and understood how I need to work to understand it”.

Example 2 “I understood the third text best of all because I learned from the two previous ones”.

Example 3 “If I read a few more texts in Ukrainian, I will be able to understand Ukrainian perfectly”.

M-factor

Every learned language affects the understanding of another language and the mechanism of comprehension in general. Thus, M-factor was distinguished as one of the predictors of comprehension (Jessner 2014, Verschik 2017). In addition, studies on crosslinguistic influence (CLI) have shown that every interlocutor's learned language has an impact on each other and might result in further language acquisition (Cenoz et al. 2001, 2003, Dewaele 1998). All our participants were multilingual and spoke at least three languages. The respondents provided the following comments in Table 36 below. 

Table 36. Examples 1-2. M-factor

Example 1 “I am quite good at languages, and since I have experience with different languages (for instance, I also speak Finnish), it is easier to find similarities between languages and in every new language that I know. More of these connections are available especially if the languages are similar or belong to the same language family”.

Example 2 “Finding similarities between Estonian and Finnish helped me to be creative in this task”.

Metalinguistic awareness

Metalinguistic awareness presents the ability of participants to grasp language categories and grammatical forms (Blees & ten Thije 2016). Examples in Tables 29 and 30 present the cases of raised metalinguistic awareness and understanding of the language systems. Below is the comment from one participant in line with developed metalinguistic awareness:

Table 37. Example 1. Metalinguistic awareness

“My main strategy was to find what part of speech the word belongs to by using my linguistic knowledge and context. Then I proceeded with the definitions”.

Context

A study on the comprehension of Danish by Dutch speakers via their knowledge of German without previous exposure (Swarte et al. 2013: 153) has shown that the foreign language mode is smaller when words for the definition are placed in the context. In our study, we observed a tendency in the participants' strategies, namely, to turn more to the context when there are fewer similarities between Russian and Ukrainian. Generally speaking, the context played a key role and was a strong supporting factor to confirm the assumptions.

Language attitudes

Since 28 participants expressed positive attitudes and two participants were neutral towards Ukrainian, we were not able to test the role of language attitudes in our experiment.

6. Conclusions

The participants' comments in the debriefing interviews shed light on the comprehension process that is behind success results from the participants' perspective. We collected qualitative data on how the participants evaluate various factors and strategies that helped them to understand Ukrainian. Without the participants' explanations, we would not be able to determine how exactly objective and perceived similarity worked, nor to outline extra-linguistic predictors of success.

As expected, the similarity between various items in Ukrainian and Russian was both objective and perceived. In some cases, the participants were able to recognise the meaning of the words based only on similarity; however, when they were challenged by different inflections, false friends, cognates with a different meaning, unfamiliar words in Russian, etc., it turned out not to be enough to rely only on similarities. It became clear from the debriefing interviews that those who verified their assumptions on the basis of the context reached better results than those who did not. Also, in some cases, the context turned out to be more important than similarity.

At the same time, various extra-linguistic factors came into play. Exposure to Russian and frequency of use of Russian foster the comprehension of Ukrainian. Exposure to different registers and access to written registers, for instance, Russian fiction, colloquial Russian, significantly affected the comprehension success rate. General knowledge about specific domains or topics positively affected the performance results. The M-factor supported the participants in recognizing similarities between two languages via already existing RM experience in other language constellations. Raised metalinguistic awareness, or understanding a language system as such, contributed to the comprehension. Finally, the participants reported about their learning process when moving from one text to another by picking up different language items and developing more advanced strategies of understanding from one text to another. This is in line with our previous study (see Branets et al. 2020: 24) that demonstrated that the comprehension level of the last text was always higher, even though Ukrainian texts were presented in a different order among the participants.

 

^{1 http://www.luistertaal.nl/en/ (accessed 15 November 2021).}

^{2 25 participants provided no comprehension estimates, and all the calculations in this subsection are based on responses by five participants.}

^{3The participants were free to provide answers in the language they were comfortable with. Most of the participants (24) provided answers in Estonian, one in Russian, one participant provided answers in both English and Russian, three participants in Estonian and Russian, and one in Estonian and English.}

About the authors

Anna Branets

University of Tartu

Author for correspondence.
Email: anna.branets@ut.ee
ORCID iD: 0000-0002-7599-2169

Linguistics, PhD Student

Jakobi 2, 50090 Tartu, Estonia

Anna Verschik

Tallinn University

Email: annave@tlu.ee
ORCID iD: 0000-0003-3989-0146

PhD, Professor of General Linguistics

Narva mnt 29, 10120 Tallinn, Estonia

References