Славянские языки | Филологический аспект №04 (96) Апрель 2023

Перцептивные характеристики оппозиции непалатализованых и палатализованных согласных в родственных языках

Тананайко Светлана Олеговна
кандидат филологических наук, доцент кафедры фонетики, Санкт-Петербургский государственный университет, stananaiko@mail.ru

Аннотация: Работа посвящена изучению того, как типологически и генетически сходные дифференциальные признаки фонем проявляются при контакте двух родственных языков и как эти признаки распознаются в родной и в интерферированной речи на этих языках. Исследуется действие фонетической интерференции в области твердости-мягкости согласных при контакте польского и русского языков. Выясняется, как носители русского и польского языков опознают твердые и мягкие согласные в родном языке и в неродном языке (русские аудиторы в польском и польские в русском), а также на какие фонетические признаки аудиторы опираются в процессе опознания.
Ключевые слова: межъязыковая интерференция, фонетическая интерференция, твердые и мягкие согласные, контакты родственных языков, дифференциальные признаки фонем, акцентная речь

Tananaiko Svetlana Olegovna
PhD in Linguistics, Associate Professor at the Department of Phonetics, Saint-Petersburg State University

Abstract: The work concerns the study of how typologically and genetically related distinctive features of phonemes manifest themselves when two related languages come into contact and how these features are recognized in native and accented speech in these languages. The article deals with the phonetic interference and how it affects non-palatalized and palatalized consonants in Polish-Russian language contact. Russian and Polish speakers are being studied to identify non-palatalized and palatalized consonants in their native language and in the non-native language (Russian auditors in Polish and Polish in Russian), as well as what phonetic characteristics auditors rely on in the identification process.
Keywords: language interference, phonetic interference, non-palatalized and palatalized consonants, contacts of related languages, differential phoneme features, accented speech

The matter of the correlation and interaction of genetically and typologically similar phonemic distinctive features in case of the related languages contacts has long attracted the attention of scientists. Thus, U. Weinreich pays considerable attention to this issue in his classical work “Languages in Contact” [4]. This problem is relevant both from a theoretical point of view, since it directly concerns the mechanisms of interlingual phonetic interference, and in the applied aspect it allows you to pay special attention to the relevant aspect of phonetics when teaching foreign languages. Russian and Polish auditors' perception of the presence or absence of consonant palatalization in Russian and Polish languages is considered in this study as a special case of such interaction, specifically, the peculiarities of perception by auditors (Russians and Poles) of the presence or absence of consonant palatalization in Russian and Polish languages. These Slavonic languages are related, both have an opposition of non–palatalized and palatalized (in Polish - partially palatal) consonants [3, p. 46; 1, p. 55] which goes back to the opposition of non-palatal and palatal consonants in the Proto-Slavic language [2, p. 44]. Accordingly, it is necessary to establish whether auditors in both languages will use similar phonetic characteristics to recognize non-palatalized and palatalized consonants, or whether the set of characteristics will be different for each language, and also whether auditors will use the perceptual mechanisms of the native language to recognize distinctive features in the non-native.

The material for the study was Russian and Polish words containing stressed syllables C+V, C’+V, C+/j/+V, C’+/j/+V types. All consonants of the corresponding language were present in the material in combination with all possible vowels for this language. These words were included in the sentences read by three groups of speakers:

Standard Russian speakers (three men and two women, 25-30 years old, higher philological education), who read Russian sentences;

Standard Polish speakers (two men and three women, 20-25 years old, incomplete higher humanitarian education), who read Polish sentences;

Polish speakers studying Russian (two men and three women, age 20-25 years, incomplete higher philological education), who read Russian sentences.

To limit the positional influences of neighboring sounds on a given syllable in Russian and Polish, words were selected, if possible, where this syllable was in similar phonetic contexts, for example, related words in both languages: карьера– kariera, первый – pierwszy, etc. The syllables studied were located closer to the beginning of the sentence and did not correlate with the phrasal stress, since the duration of the vowel was not taken into account. Syllables were segmented from the speakers' speech and then randomly presented for listening to two groups of auditors of 10 people each (age 20-30 years, higher humanitarian education): Russian listeners who did not know Polish, and Poles who spoke Russian fluently. In total, 9 series of syllables for listening were compiled; each series included syllables from different languages and different types. Each series consisted of 95-96 syllables; each stimulus was repeated 3 times with an interval of 1 second. The task for the auditors was to identify the syllable and write it down in any way convenient for the auditor.

When processing the results of the listening analysis, it was necessary to give a generalizing quantitative assessment of the general and specific features of the non-palatalized and palatalized consonants perception in Russian and Polish by two groups of auditors. The methodology was required that would allow, firstly, to compare the results of perception of the same consonants by different groups of auditors and, secondly, to consider the peculiarities of different languages consonants perception by each group of auditors. To do this, Spearman's rank correlation method was used according to the formula,

where d is the difference between the ranks of conjugate observations, and n is the number of pairs of these conjugate observations. The indicators (in %) of the correct identification of the non-palatalization or palatalization of a given sound and the correct identification of the phonemic quality of this sound were considered conjugate. Since the values of the rank correlation quotient /ρ/ modulo turned out to be in the span 0.086 </ρ/< 0.975, i.e. /ρ/ took almost all possible values for it, the Guilford scale was used to assess the significance of the correlation, according to which the correlation is considered very high if 0.8</ρ/< 1; high if 0.6</ρ/<0.8; medium if 0.4</ρ/< 0.6; weak if 0.2 </ρ/<0.4 and absent if /ρ/<0.2. This method makes it possible to establish the conjugation of the recognition of sounds of different languages by the same auditors and the conjugation of the recognition of the feature of non-palatalization or palatalization of consonants of the same language by different groups of auditors.

Considering the data of the correlation study of the results of the listening analysis one can make some generalizations.

Figure 1. Correlation quotient /ρ/ distribution in the span 0 to 1 (on the abscissa)

Firstly, all the rank correlation quotients were analyzed and it was found that the value of the rank correlation quotient /ρ/ modulo is within 0.086 </ρ/< 0.975, i.e. practically ρ can take any values from -1 to +1. All values of /ρ/ modulo were plotted, and then the histogram was made (Figure 1) showing how many points fall on each interval of the histogram corresponding to 0.1; all this made it possible to determine the areas of concentration that account for the largest number of values of

/ρ/. These are the ranges around 0.3 and around 0.8. One can conclude that there is a tendency to divide the correlation quotients into two groups corresponding to a strong and weak dependence between, on the one side, the recognition of the consonant non-palatalization-palatalization and on the other, some other indicator. Obviously, the studied correlation dependence turned out to be stronger when the identification of the differential feature of non-palatalization-palatalization presented a certain difficulty for the auditor, and this forced him to look for other additional criteria for classifying the consonant as non-palatalized or palatalized. The weak correlation showed that the identification of the consonant non-palatalization-palatalization was not difficult for the auditors.

In some cases, the attitude of auditors to consonants was ambiguous, and when identifying consonants, sounds were divided into two groups with a different type of correlation dependence within each of them. Most often, these groups could be characterized as the main and additional. Moreover, the distribution of sounds into groups was largely regular. Thus, when studying the relationship between the recognition of the consonant non-palatalization-palatalization and the recognition of the phonemic quality of this consonant, (as well as in one case, when studying the relationship between the recognition of the same consonants by Russians and Poles), systematically an additional group (whose sounds were recognized worse than the sounds of the main group) included anterior-lingual sonorants, affricates and Polish palatals. Apparently, the patterns of perception of these sounds differ from the perception patterns of most consonants; this may be explained by the known difficulty of identifying these sounds themselves, that is, their quality.

Summarizing the data obtained during the study of the results of the listening analysis, it should be remembered that when perceiving speech sounds, native speakers use two different levels of description – universal phonetic, associated with the general manifestation of language competence, and phonological, associated with phonological relations in their native language [6, p. 147]. The involvement of native speakers of languages with different phonological systems as auditors gives a chance to try to identify both the universal perceptual characteristics of the non-palatalization-palatalization opposition, and the perception specifics of the phonetic implementation of this phonological feature caused by the system of the auditors' native language.

When describing the results of the listening analysis, the following patterns were revealed in the perception of the presence or absence of the consonant palatalization. In the syllables of both languages and by both groups of auditors, the non-palatalization of consonants was recognized much better than palatalization. The current research also revealed the perceptual pattern explaining this fact. Based on the results of correlation analysis (in particular, the analysis of the correlation between the recognition of the consonant non-palatalization feature and the recognition of a consonant phonemic quality, as well as the analysis of the correlation between the perception of a certain group of consonants by Russian auditors and the perception of the same consonants by Polish auditors), we can say that when identifying non-palatalization of consonants, both groups of auditors were guided by the similar criteria, while in the perception of palatalization, auditors of different groups mainly relied on different criteria. This is probably due to the fact that in both languages the feature of palatalization of a consonant is marked as opposed to the feature of non-palatalization, and the fact that the phonetic correlates of palatalization are different in Russian and Polish, while the phonetic correlates of non-palatalization are very similar.

Another feature of the consonant perception, common to both groups of auditors, turned out to be that the feature of the presence or absence of palatalization was always best recognized in labial consonants, somewhat worse in posterior-lingual ones. This is probably due to the nature of palatalization of labial consonants: the most essential characteristic of palatalization for this group of consonants is the subsequent [i]-shaped transition to the next vowel and the consonant itself (because of its labial, not lingual articulation) has only very weak signs of palatalization or none at all. As for the anterior-lingual group, it, depending on the auditors' assessment of the quality of the consonant, fell into 2 subgroups: the main one, whose members were identified as palatalized or non-palatalized quite regularly, and an additional one, into which sonorants and affricates systematically fell. According to the listening analysis, Polish palatals can also be included in this subgroup. The sounds of this subgroup were recognized much worse than all the others, and the identification of their phonemic quality, as well as the presence or absence of palatalization in them, was quite irregular and inconsistent. Obviously, we can say that these consonants have some perceptual characteristics that make it difficult to identify them as palatalized or non-palatalized. With a high degree of probability, it can be assumed that these perceptual characteristics have become, in the case of Polish palatals, the result of the fact that palatal articulation, which is additional for all other palatalized consonants, turns out to be the main one for palatals.

The result of the study of the perception dependence of a consonant non-palatalization or palatalization on the subsequent vowel turned out to be unexpected. This dependence proved to be statistically significant only for 18 percent of consonants – these are mainly palatalized consonants of the non-native language. Russians and Poles did not rely on the characteristics of the vowel following the consonant, and also, as the results of correlation analysis show, they did not rely on the phonemic quality of the consonant (this quality did not affect the identification of the consonant as palatalized by either Russian or Polish auditors).

Obviously, the decisive criterion for classifying a consonant as non–palatalized or palatalized, as well as when deciding on the presence of /j/ between a consonant and a vowel, were not the characteristics of the consonant itself or the transitional part of the vowel, but the presence (or absence – for non-palatalized), as well as the very nature of the [i]-shaped transition to the next vowel, i.e. the specifics transients between consonant and vowel. Moreover, the [i]-shaped transition played a significant role in the perception of palatalized labials. When referring to non-palatalized or palatalized non-labial consonants, this transition was less significant. It should be added, however, that before the vowels of the front row, the palatalization of consonants was recognized by both groups of auditors better than before the vowels of the non-front row. Obviously, it is caused by a greater degree of palatalization of consonants in this position and mainly by the laws of phonemic distribution of both languages: in both Russian and Polish standard pronunciation, only a palatalized consonant can be combined with the vowel /i /, therefore its presence in the syllable automatically meant the palatalization of the preceding consonant. In addition, asynchronous palatalization was much weaker in syllables with vowels of the front row (and before the vowel /i/ didn’t take place at all), which made it easier for Russian auditors to identify the palatalization of consonants in these syllables. Thus, at the level of tendency, we can say that in the position before the vowels of the front row, the palatalization of consonants is recognized better than before the vowels of the non-front row.

An important feature of syllable recognition was that the difference between Russian syllables of type C+/j/+V, C’+/j/+V and C’+V was recognized regularly only by Russian auditors. Poles, though they spoke Russian fluently, in most cases did not perceive this difference. As for Russian auditors, they did not recognize the difference between the Polish syllables C’+V and C’(or C)+/j/+V, often perceiving the syllable C’+V as C’(or C)+/j/+V. So one can surmise that Russian has a different way of implementing palatalization from Polish language: in Polish, the transitional [i]-shaped section is longer than in Russian, and often transforms into non-syllabic [i] [6, p. 75].  Russian phonological hearing is unaccustomed to such palatalization, so the [i]-shaped section was regularly perceived by Russians as an independent phoneme /i/ or /j/. The main difference of Polish palatalized consonants from Russian is a lower degree of their palatalization and of the asynchronous implementation of this palatalization (elongation of the [i]-shaped transition). This is what caused completely different perception by Russian and Polish auditors of syllables C+/j/+V, C’+/j/+V type, of Standard Polish palatalized syllables and of Russian syllables pronounced by Poles with palatalized consonants.

In general, the results of the audit analysis allow not only to identify some of the most common perceptual features of the palatalized-non-palatalized consonants opposition, but also make it possible to determine the specific characteristics of non-palatalized and palatalized consonants in Russian and Polish languages, as well as to identify similarities and differences in the perceptual base of Russian and Polish auditors and explain this similarity and difference, based on the phonological systems of both languages. The data obtained allow us to draw the following conclusions.

1. Both Russian and Polish auditors best recognize the palatalization of consonants before the vowels of the front row, which is easily explained: there are simply no non-palatalized consonants before the vowel /i/ in both languages, and before /e/ the consonant is palatalized in most cases.

2. Recognition of syllables containing non-palatalized consonants with vowels of both native and non-native languages does not cause significant difficulties, which indicates similar articulatory and perceptual mechanisms involved in this process.

3. As for the syllables like C’+V and C’(or C)+/j/+V, it is obvious that the mechanisms of their implementation and, consequently, perception differ in Russian and Polish. Russian has a transitional [i]-shaped section between a consonant and a subsequent vowel that is somewhat longer than in Russian and is more autonomous, which leads to interesting results in the perception of Russians and Poles. Polish auditors, although speaking Russian fluently, do not distinguish between the syllables C’+V and C’(or C)+/j/+V, perceiving both as C’+V. Russian auditors, who also do not distinguish between these two types of Polish syllables, perceive both as C’(or C)+/j/+V.

1. Колесов В.В. Историческая фонетика русского языка. 2-е изд., испр. и доп. М, «Юрайт», 2023. 217 с. Kolesov V.V. Istoricheskaya fonetika russkogo yazyka [Historical phonetics of Russian, 2nd ed.]. Moscow, Youright Publishers, 2023. (In Russ.)
2. Попов М.Б. Фонетика современного русского языка. СПб, Изд-во СПбГУ, 2014. Popov M.B. Fonetika sovremennogo russkogo yazyka [Phonetics of modern Russian]. Saint-Petersburg, Saint-Petersburg Univ. Press, 2014. (In Russ.).
3. Jaskuła K. Average or unique – Polish place-names and word-initial consonant groups // Poznan Studies in Contemporary Linguistics. Volume 59 Issue 1 March 2023; https://doi.org/10.1515/psicl-2022-1016. (Application date 09.04.2023).
4. Weinreich, U. Languages in contact. The Hague, Paris, NY, Mouton Publishers, 1968
5. Wierzchovska B. Fonetyka i fonologia jezyka polskiego. Warszawa, Wroclaw, Krakow, Gdansk, Zakład Narodowy im. Ossolińskich, 1980. 186 s.
6. Yizhou W. et al. Same vowels but different contrasts: Mandarin listeners' perception of English /ei/-/iː/ in unfamiliar phonotactic contexts. // Journal of Phonetics Volume 97, March 2023; https://doi.org/10.1016/j.wocn.2023.101221. (Application date 09.04.2023).