In the study of a language's phonemic structure, establishing distributional patterns through statistical evidence is of paramount importance. This issue has not been sufficiently explored in the Abkhaz language, as demonstrated by the following review of specialized literature.

K. Lomtatidze in the monograph “Historical-Comparative Analysis of Abkhaz and Abaza Languages” discusses the occurrence of only certain phonemes in specific positions [Lomtatidze, 1976:24-61, 139-210].

G. Hewitt also examines the issue of Abkhaz phoneme distribution. According to his observations, initial consonants in Abkhaz are not subject to restrictions; however, tʷ and dz do not occur in word-final position, while the phonemes dʷ, ts, g, k, gʷ, and ʁʷ are rare [Hewitt, 1979:260]. The scholar does not address the inlaut position at all.

V. Chirikba conducted a statistical analysis of consonants based on a ten-page text from the Abkhaz literary journal “Alashara”. He established that sonorants, semivowels, and labiodental stops are dominant. Specifically, the most frequent are: j, r, n, l, w, z, d, m, x, and t [Chirikba, 1996:27]. This study was carried out without accounting for the specific positions of phoneme distribution.

In my own article, “Verbal Roots of CVC Structure in Abkhaz-Abaza Languages”, attention is also given to the issues of phoneme distribution [Poniava, 2012:125-142]. However, the analysis is restricted to verbal material of only one structural model, which, understandably, cannot provide a comprehensive picture of consonant distribution.

N. Machavariani also touches upon the occurrence of phonemes in Abkhaz affixes and modal elements in the book “Affixes and Modal Elements of the Abkhaz Language with a Grammatical Overview” [Machavariani, 2020:38].

As evidenced by the review above, previous studies remain fragmentary. There is an urgent need for a detailed analysis of representative material covering the entire linguistic system through a modern approach. Consequently, a corpus-based study was initiated to investigate the distribution of Abkhaz phonemes.  I have conducted research of this type on Abkhaz fricatives [Poniava, 2020] and vowels [Poniava, 2021:87-90], as well as stops, though the latter was restricted to the anlaut position. The latter was presented as a paper at the 2021 scientific session dedicated to Arnold Chikobava. Although that report included a distribution table, it lacked depth, as it did not distinguish whether a given phoneme was realized within the native Abkhaz lexical core or in loanwords. Furthermore, phonemes were represented by only a single example, and the factors determining the frequency of specific phonemes were not analyzed [Poniava, 2021:39-40]. The present article reflects the results of a more profound investigation of the anlaut position, addressing the aforementioned shortcomings. Additionally, the inlaut and auslaut positions have been thoroughly examined. Accordingly, this article provides a distributional-statistical analysis of Abkhaz stops across all three positions.

The Abkhaz literary language comprises 58 consonants and 6 vowels. More than half of the consonants – 33 phonemes – are stops. The stop inventory of literary Abkhaz is as follows:

Bilabial: b p pʼ 

Dental: d t tʼ 

Dental labialized: dʷ tʷ tʷʼ 

Alveolar: dz ts tsʼ 

Alveolar labialized: dzʷ tsʷ tsʷʼ 

Post-alveolar: dʒ tʃ tʃʼ 

Post-alveolar fortis: dʒ¾  tʃ¾ tʃ¾ʼ 

Velar: g k kʼ 

Velar palatalized: gʲ kʲ kʲʼ 

Velar labialized: gʷ kʷ kʷʼ

Uvular: qʼ

Uvular palatalized: qʲʼ

Uvular labialized: qʷʼ

The research was conducted using the National Corpus of the Abkhaz Language , considering the anlaut, inlaut, and auslaut positions. The development of the National Corpus of the Abkhaz Language began in 2015 within the framework of the “Palag" project. It comprises over 10 million units, of which 7,596,355 are word-forms. The corpus includes diverse texts from both fiction and mass media [ANC]. 

As noted, the present study is corpus-based; therefore, relevant examples are cited as they appear in the corpus rather than in their dictionary forms. For instance, the dictionary form a-kʲamsa (“stomach”) is attested in the corpus as kʲamsa and akʲamsa. In the former, the stop is in the anlaut position, while in the latter, it is in the inlaut. It should also be emphasized that the corpus does not distinguish the root morpheme or morpheme boundaries; consequently, the distribution of consonants is analyzed based on word-forms. Within each position, phonemes are arranged in groups of five according to their frequency of occurrence. Two or three examples are provided for each phoneme. In the case of borrowed vocabulary, the source language – from which Abkhaz borrowed the word either directly or indirectly – is indicated.

 

The Distribution of Abkhaz Stop Phonemes in Anlaut

 

The study based on the National Corpus of the Abkhaz Language establishes that the most frequent group of five stops in the word-initial (anlaut) position consists of d, b, p, q’, and t:

d – Highly productive as a class-personal prefix: dgəlojt’ “he/she stands”, dara “they”, darban “who?”; 

b – bargʲə “you (fem.) too”, bʁʲə “leaf”. b is frequent as a class-personal prefix: ban “your (fem.) mother”;

p – psə “soul”, pʃ¾dza “beautiful”;  

q’ – q’alap “will be”, q’aħʦ’ejt’ “we did it”, q’ama “dagger” < Turk. kama “id”;

t – tbaa “wide”, tra “nest”, təp "place”.

The second group of five realized in the anlaut consists of k’, gʷ, tʃ¾ʼ, dʒ, and tsʼ:

k’ – k’ak’atʃʼ “flower”, k’nə “at”, k’urs “course” < Russ. kurs “id”;

gʷ – gʷə “heart”, gʷəp “group”;

tʃ¾ʼ – tʃ¾ʼar “youth”, tʃ¾ʼəts “new”;

dʒ – dʒma “goat”, dʒanat “paradise” < Turk. cennet “id”;

ts’ – tsʼaq’a “below”, tsʼəs “bird”, tsʼla “tree”.

The third series of five stops found in anlaut in Abkhaz comprises the phonemes tsʷ, ts, p’, g, and tʃ¾: 

tsʷ – tsʷəbbra “September”, tsʷaɥw a “skin”, tsʷa “quality”; 

ts – tsəra “grain”, tsentr “center” < Russ. tsentr “id”; 

p’ – Realized only in borrowed vocabulary: p’lan “plan” < Russ. plan “id”, p’ark “park” < Russ. park “id”; 

g – gxa “defect”, gara “cradle”;

tʃ¾ – tʃ¾ənla “by day”, tʃ¾ə “horse”, tʃ¾ada “donkey”.

The fourth group of five stops in the word-initial position of the language under study appears as dz, t’, tʃ, tʃ’, and qʷ’:

dz – dzəχʲ “spring ”, dzia “lake”, dzias “river”;

t’ – t’ə “owl”, t’at’a “soft”, t’aksi “taxi” < Russ. taksi “id”;

tʃ – tʃara “wedding”, tʃəs “food”, tʃgara “fasting”;

tʃ’ – tʃ’ədala “especially”, tʃ’ərtʃ’ərra “chirping”, tʃ’kʷʼən “boy”;

qʷ’ – qʷ’əd “beans”, qʷ’əʃ¾ “wise”, qʷ’aqʷ’a “back”.

The fifth group of five stops in Abkhaz is attested with the composition kʷ’, kʷ, k, kʲ’, and tsʷʼ:

kʷ’ – kʷ’ara “stream”, kʷ’ats “meat”, kʷ’at’a “duck” < Megr. k’vat’a “id”;

kʷ – kʷaʃ “luck”, kʷtʃ¾ʼəjarala “successfully”, kʷab “copper cauldron” < Georg. kvabi “id”;

k – kdə “stump”, kəta “village”, kalakʲ “city” < Georg. kalaki “id”;

kʲ’ – kʲ’akʲ’a “hard”, kʲ’atʃ¾ʼ “short”, kʲ’əba “stairs” < Georg. k’ibe “id”;

tsʷʼ – tsʷʼa “apple”, tsʷʼəwara “crying”, tsʷʼtsa “glass”.

The rarest, sixth group of five stops in the anlaut position is represented by the phonemes tʷ’, kʲ, qʲ’, dʷ, and gʲ:

tʷ’ – tʷ’əm “stranger”, tʷ’əla “country”, tʷ’əɥwa “horn”;

kʲ – kʲamsa “stomach”, kʲaf “feast” < Turk. keyif “id”;

qʲ’ – qʲ’aqʲ’a “flat”, qʲ’ant’az “naked”; 

dʷ – dʷəʁba “train”, dʷə “field”. dʷ is frequent as a structural component of a local preverb: dʷəltsʼra “going out”; 

gʲ – gʲagʲa “ring”, gʲeʒ “wheel”, gʲama “taste” < Georg. gemo “id”.

In the anlaut position of the Abkhaz language, the stops dzʷ, tʷ, and dʒ¾ are least frequently manifested:

dzʷ – dzʷərɥw “many”, dzʷdzʷara “to wash”; 

tʷ – tʷə “full”, tʷa “hay”;

dʒ¾ – dʒ¾ər “steel”, dʒ¾əʃ¾ “garlic”.

The research establishes that the distribution of stop phonemes in the anlaut of the Abkhaz language, according to their frequency sequence, is as follows: d, b, p, q’, t, k’, gʷ, tʃ¾ʼ, dʒ, ts’, tsʷ, ts, p’, g, tʃ¾, dz, t’, tʃ, tʃ’, qʷ’, kʷ’, kʷ, k, kʲ’, tsʷʼ, tʷ’, kʲ, qʲ’, dʷ, gʲ, dzʷ, tʷ, dʒ¾. The phonemes d and b share the first and second positions; this can be explained by the fact that both serve as class-personal markers and are more heavily loaded with morphemic functions. In the anlaut, p’ is attested only in borrowings established through the Russian language (p’oet’ “poet”, p’oezia “poetry”, p’roza “prose”), whereas the majority of other phonemes occur in both native Abkhaz and borrowed lexical units.

 

The Distribution of Abkhaz Stop Phonemes in Inlaut

 

In the Abkhaz language, similar to the anlaut, all stops are realized in the inlaut position as well. 

According to the corpus-based research, the first group of five stops in the inlaut includes the phonemes t, p, b, kʷ, and g:

t – apta “cloud”, akəta “village”, aʃtaχʲ “afterward”;

p – arpəs “youth”, aapsara “merit”, atəpantʷʼi “local”;

b – asabi “child”, abas “this way”, jarban “who?”;

kʷ – akʷa “rain”, aʃ¾əkʷs “year”. kʷ is frequent as a component of the plural suffix: amɥʷakʷa “roads”;

g – alagamta “beginning”, aga “coast”, aagara “bringing”.

In the inlaut of the Abkhaz language, the second group of five consists of the stops k’, ts’, d, gʲ, and q’:

k’ – ak’ə “one”, ajlk’aara “understanding”, aresp’ublik’a “republic” < Russ. respublika “id”;

ts’ – apənts’a “nose”, axats’a “man”, q’asts’ejt’ “I did it”;

d – xada “leader”, adamra “tomb”;

gʲ – atsʷgʲa “bad”. Suffixal gʲ, which functions as the particle “also,” is very frequent: sabgʲə “my father too”;

q’ – abaq’a “monument”, usq’an “then”, ʃ¾aq’a “how much?”.

In the inlaut position of the Abkhaz language, the third group of five is formed by the phonemes dz, ts, tsʷ, t’, and gʷ:

dz – madza “secret”, aadzara “raising”. The suffixal dz of intensity is highly prevalent: duddza “very big”;

ts – mtsa “fire”, atatsa “bride”, jatsə “yesterday”;

tsʷ – atsʷa “skin”, antsʷa “God”. tsʷ is highly frequent as a segment of the productive plural suffix: asastsʷa “guests”;

t’ – aart’ra “opening”. t’ is common in borrowings: at’eat’r “theater” < Russ. teatr “id”;

gʷ – sgʷə “my heart”, amgʷa “stomach”, aajgʷa “nearby”.

In the inlaut position of the Abkhaz language, the fourth group of five realized phonemes comprises tʷ’, kʷ’, tʃ¾ʼ, p’, and tʃ¾:

tʷ’ – səntʷ’a “this year”, dtʷ’owp’ “he/she sits”. The destinative suffixal tʷ’ is very widespread: pasatʷ’i “former”;

kʷ’ – akʷ’t’aʁʲ “egg”, aħkʷ’aʒʷ “queen”, atʃʼkʷ’əngʲə “the boy too”;

tʃ¾ʼ – axatʃ¾ʼə “mouth”, adatʃ¾ʼə “outside”;

p’ – amp’əl “ball”. p’ is very common in borrowed vocabulary: asp’ort’ “sport” < Russ. sport “id”;

tʃ¾ – atʃ¾tra “stable”, atʃ¾aɥwra “harvest”.

In the inlaut of the Abkhaz language, the fifth group of five is represented by the phonemes tʃ, tsʷʼ, qʷ’, tʃ’, and dʒ:

tʃ – atʃa “bread”, amtʃra “power”;

tsʷʼ – ajatsʷʼa “star”, watsʷʼə “tomorrow”, amtsʷʼəʒʷɥʷa “wing”;

qʷ’ – anəqʷ’aɥʷ “traveler”,  at’qʷ’a “captive” < Georg. t’q’ve “id”;

tʃ’ – akʷ’tʃ’əʃ “chicken”, amatʃ’xara “reduction”, atʃ’ək “nearly”;

dʒ – adʒtsʼla “oak tree”. dʒ is also realized in borrowed vocabulary: abdʒar “armor” < Georg. abdʒari “id”.

In the inlaut of the Abkhaz language, the rarest sixth group of five is formed by the phonemes qʲ’, k, kʲ, kʲ’, and dzʷ:

qʲ’ – adaqʲ’a “side”, atsʷqʲ’a “trap”;

k – akətatʃ¾ʼə “in the village”. k is also found in borrowed material: arχit’ekt’or “architect” < Russ. arχitektor “id”; 

kʲ – akʲamsa “stomach”, adʷkʲan “shop” < Turk. dükkân “id”;

kʲ’ – iʒ¾akʲ’a “his beard”, sarkʲ’a “mirror” < Georg. sark’e “id”;

dzʷ – adzʷə “someone”, adzʷdzʷara “to wash”.

In the inlaut position of the language under analysis, the stops dʷ, tʷ, and dʒ¾ are least frequently realized:

dʷ – adʷaχʲ “outside”, adʷəltsʼra “to go out”;

tʷ – amatʷa “clothes”, amatʷar “thing”, atʷan “ceiling” < Turk. tavan “id”; 

dʒ¾ – adʒ¾əʃ¾ “garlic”, adʒ¾ər “steel”.

Research has shown that the distribution of stop consonants in the inlaut of the Abkhaz language, according to their frequency sequence, is as follows: t, p, b, kʷ, g, k’, ts’, d, gʲ, q’, dz, ts, tsʷ, t’, gʷ, tʷ’, kʷ’, tʃ¾ʼ, p’, tʃ¾, tʃ, tsʷʼ, qʷ’, tʃ’, dʒ, qʲ’, k, kʲ, kʲ’, dzʷ, dʷ, tʷ, dʒ¾. Unlike in the anlaut, the phonemes t and p occupy the first and second positions in the inlaut. Another distinction from the word-initial position is that every phoneme without exception is realized within native Abkhaz material, and most of them also appear in loanwords.

 

The Distribution of Abkhaz Stop Phonemes in Auslaut

 

In the Abkhaz language, similar to the anlaut and inlaut positions, all stops appear in the auslaut as well. 

The corpus-based study I conducted revealed that the first group of five stops, characterized by the highest distribution in the auslaut, consists of t’, p’, k’, tʷ’, and t:

t’ – Most frequent as a component of the finiteness suffix: itsejt’ “went (inanim)”. It is also highly prevalent in borrowed lexical units: ap’oet’ “poet” < Russ. poet “id”, mart’ “March” < Russ. mart “id”;

p’ – map’ “no”. p’ is mostly a static suffixal element: dəq’owp’ “he/she is”;

k’ – atak’ “answer”, znək’ “once”. -k’ primarily functions as the indefinite suffix: aʒʷabʒk’ “one story”;

tʷ’ – ak’ərəntʷ’ “many times”. tʷ’ primarily functions as a destinative suffix:  afatʷ’ “food”;

t – urt “they”, milat “nation” < Turk. millet “id”. 

The second most common group of five stops in the word-final position of the Abkhaz language is represented by the consonants ts, gʲ, b, p, and tʃ¾ʼ:

ts – aʃ¾əts “boxwood”, amts “fire”, arəts “grain”;

gʲ – esnagʲ “always”, abragʲ “bandit” < Georg. abragi “id”;

b – ajħab “elder”, aχəb “roof”, adzʁab “girl” < Megr. dzʁabi “id”;

p – atəp “place”, agʷəp “group”, anasəp “fate” < Turk. nasip “id”;

tʃ¾ʼ – kʲ’atʃ¾ʼ “short”. tʃ¾ʼ is highly productive as a postposition: abnatʃ¾ʼ “in the forest”. 

In the Abkhaz language, the third group of five in the auslaut position is revealed in the form of the stops d, kʲ, tʃ, q’, and dz:

d – aqʷ’əd “beans”. d is more characteristic of borrowed vocabulary: asud “court” < Russ. sud “id”, afond “fund” < Russ. fond “id”;

kʲ – atʃəkʲ “bush”, atʃotkʲ “brush” < Russ. ɕːotka “id”;

tʃ – aʁʲətʃ “thief”, q’amtʃ (the word-final ə is apocopated) “whip” < Turk. kamtʃə “id”, amat’tʃ “match” < Russ. mattʃ “id”;

q’ – usq’aq’ “so much”, artmaq’ “bag” < Turk. artmak “id”, iwadʒaq’ “his fireplace” < Turk. odʒak “id”;

dz – apsədz “fish”, aχʲədz “name”.

In the Abkhaz language, the fourth group of five presented in the auslaut contains the stops gw, g, ts’, dzʷ, and tsʷ:

gw – ləgw “her heart”, imagw “his boot” < Megr. magu “bad shoe”;

g – ats’abərg “truth”. g is very frequent in borrowings: aek’olog “ecologist” < Russ. ekolog “id”, alozung “slogan” < Russ. lozung “id”;

ts’ – amts’ “fly”, aħʷaanχəts’ “abroad”;

dzʷ – adzʷ “someone”, datʃ¾adzʷ “another”. In both cases, the word-final ə is apocopated;

tsʷ – abatsʷ “medlar”, nərtsʷ “beyond”, abnatsʷ “deer”.

The fifth group of five realized in the word-final position of the Abkhaz language is represented by the stops tʃ’, dʒ, kʲ’, kʷ’, and tsʷʼ:

tʃ’ – amatʃ’ “little”, ak’atʃ’ətʃ’ “Kachichi (grape variety)” < Megr. k’atʃ’itʃ’i “id”;

dʒ – adʒ “oak”, ak’olledʒ “college” < Russ. kolledʒ “id”;

kʲ’ – abəkʲ’ “signal”, akʷ’akʲ’ “corner”, asakʲ’ “sack” < Megr. sak’i “id”;

kʷ’ – alakʷ’ “fairy tale”, zakʷ’ “what?”;

tsʷʼ – amatsʷʼ “branch”, zatsʷʼ “alone”.

In the auslaut, the sixth group of five is formed by the phonemes kʷ, qʷ’, dʷ, k, and tʃ¾:

kʷ – atʃ¾ʼəkʷ “edge”, aʁaʃ¾akʷ “hostile”;

qʷ’ – bʃʷqʷ’ (the word-final ə is apocopated) “your (fem.) book”, rqʷ’aqʷ’ (the word-final a is apocopated) “their back”;

dʷ – adʷ “field” (the final ə is lost), iqʷ’ardʷ “his chair”;

k – atʃ'ək “nearly”, aq’abak “zucchini” < Turk. kabak “id”;

tʃ¾ – amaχtʃ¾ “camel”, atʃ¾ (the final ə is lost) “horse”.

The most infrequent stops in the word-final position are dʒ¾, tʷ, and qʲʼ:

dʒ¾ – aχbədʒ¾ “adolescent”; 

tʷ – atʷ (the final ə is apocopated) “full”. In other cases, it is mostly a typographical error − tʷ is written instead of tʷʼ: milattʷ, which should be milattʷʼ “national” < Turk. millet “nation”;

qʲʼ – akʷʼt’əqʲʼ “droppings”,  abətsaqʲʼ “thatched wattle hut”. 

Corpus-based research shows that stops in the word-final position of the Abkhaz language appear in the following sequence: t’, p’, k’, tʷ’, t, ts, gʲ, b, p, tʃ¾ʼ, d, kʲ, tʃ, q’, dz, gʷ, g, ts’, dzʷ, tsʷ’, tʃ’, dʒ, kʲ’, kʷ’, tsʷʼ, kʷ, qʷ’, dʷ, k, tʃ¾, dʒ¾, tʷ, qʲ’. Unlike in the anlaut and inlaut, t’ and p’ occupy the first and second positions in the auslaut, as both are productive as formative elements of affixal morphemes. A large portion of stops is attested both in native Abkhaz lexical units and in those established from other languages. Contrary to G. Hewitt’s conclusion, the stops tʷ and dzʷ are manifested in the word-final position in Abkhaz, albeit as a result of word-final vowel apocope; furthermore, it is not only the phonemes dʷ, ts, g, k, gʷ, and ʁʷ that are rare [Hewitt, 1979:260], but others as well (see table).

The results of the corpus research can be perceived more clearly in the table provided below, which reflects the quantitative and percentage indicators of the occurrence of each phoneme in the anlaut, inlaut, and auslaut positions:

 

Table of the Distribution of Abkhaz Stops

In summary, the corpus-based research has demonstrated that in Abkhaz, as anticipated, all stops are realized across all three positions with varying frequency indicators.

In Abkhaz, the most widespread stops are those that function as affixes or are components of affixes. In the anlaut and inlaut, these are primarily class-personal prefixes, whereas in the auslaut, they serve as formative elements of suffixes for dynamic or static forms.

In all three positions, the majority of stops are attested in both the native lexical fund and in borrowings. However, certain phonemes occur largely in borrowed material, and p’ in the anlaut is realized exclusively in lexical units adopted from other languages. In all three positions, tʷ and dʒ¾ are rare.

In the anlaut of the Abkhaz language, only one stop – d − is characterized by a frequency higher than 1%. In the inlaut, 22 stops exceed this threshold: t, p, b, kʷ, g, k’, ts’, d, gʲ, q’, dz, ts, tsʷ, t’, gʷ, tʷ’, kʷ’, tʃ¾ʼ, p’, tʃ¾, tʃ, and tsʷʼ. In the auslaut, 5 stops have a frequency higher than 1%: t’, p’, k’, tʷ’, and t.

In the word-initial position (anlaut), the frequency of occurrence is lower than 1% for 32 phonemes: b, p, q’, t, k’, gʷ, tʃ¾ʼ, dʒ, ts’, tsʷ, ts, p’, g, tʃ¾, dz, t’, tʃ, tʃ’, qʷ’, kʷ’, kʷ, k, kʲ’, tsʷʼ, tʷ’, kʲ, qʲ’, dʷ, gʲ, dzʷ, tʷ, and dʒ¾. In the inlaut, 11 phonemes fall below 1%: qʷ’, tʃ’, dʒ, qʲ’, k, kʲ, kʲ’, dzʷ, dʷ, tʷ, and dʒ¾. In the auslaut, the frequency is lower than 1% for 28 phonemes: ts, gʲ, b, p, tʃ¾ʼ, d, kʲ, tʃ, q’, dz, gʷ, g, ts’, dzʷ’, tsʷ, tʃ’, dʒ, kʲ’, kʷ’, tsʷʼ, kʷ, qʷ’, dʷ, k, tʃ¾, dʒ¾, tʷ, and qʲ’.

The research establishes that in Abkhaz, stops exhibit the highest distribution in the inlaut, followed by the auslaut, and are most rarely represented in the anlaut. The results obtained are significant for conducting further comparative and typological research