A Comparative Study: Astori Dialect of Shina Language vs. English Language | Academic Blog


 “A Comparative Study/Analysis Between the English Language and the Astori Dialect of the Shina Language based on Affixation

(Prefixation and Suffixation)”.

 


 

 

ABSTRACT

The fundamental principle of this project is to examine or analyze the similarity in the process of affixation (prefixation and suffixation) between the English language and the Astori dialect of the Shina language. Mainly the native vocabulary of the Astori dialect is adopted for this research. All the borrowed and loan words are being neglected because of this we can get awareness about the flexibility of the respective dialect. This similarity looks for the ‘formation of new words’ due to affixation, ‘modification in meaning’ because of affixation, and eventually ‘variation in the class of words’ as a result of affixation. So, all the needed data is arranged from the related texts, articles, books, websites, and native population and actual speakers of this dialect. Thus, the results of this research proved that because of prefixes and suffixes (either in the English language or in the Astori dialect of the Shina language), new words can be formed, the class of words can be changed and meaning also changes in each case. At last, this research has nominated that the flexibility of affixation of the Astori dialect is similar to that of the English language under the case study of the conversion of the ‘noun to noun’, ‘verb to the noun’, noun to the adjective’ and ‘adjective to the noun’ with the help of affixation. Hence, in the future, further studies can be conducted on the remaining parts of speech on the basis of affixation to evaluate the similarity between both of them.

 

TABLE OF CONTENT

 

TITLE OF RESEARCH

ABSTRACT

TABLE OF CONTENT   

 

CHAPTER 1

INTRODUCTION

1.1    Introduction

1.2    Research Problem

1.3    Research Objectives

1.4    Research Questions

1.5    Delimitation of Study

 

CHAPTER 2

LITERATURE REVIEW

 

CHAPTER 3

METHODOLOGY 

 

CHAPTER 4

DISCUSSION/FINDINGS

 

CHAPTER 5

CONCLUSION 

 

REFERENCES

 

 

CHAPTER 1 

INTRODUCTION

 

1.1    Introduction

In this very passage, we will be looking for a comparative study of the suffixation and prefixation in the English language and the Astori dialect of the Shina language. As Shina (/ʃinãã/ or /ʃiɳãã/) speaking areas are several in number in both India and Pakistan. Therefore, on the basis of Kohistani and Schmidt (2006: 137), The Shina-speaking territories under Pakistani rule include the following: Gilgit, Tangir-Darel, lower Hunza, Chilas, Astor, and the region of Indus Kohistan, while the Shina-speaking ranges in the Neelam (Kishenganga) drainage, the Gurez and Tiliel valleys, the Drass plain, and Ladakh are controlled by India. Shina is spoken in the Gurez Valley of Kashmir Division's District Bandipora and the Drass Area of Kargil District in Ladakh Division in India's northern state of Jammu and Kashmir. Radloff (1992: 122-150) says that the ‘Astori dialect is most familiar (81%) with Gurezi dialect of Shina language’, which is supposed to be famous in one prominent part of Northern Areas of Pakistan. i.e. Astore and people living in this region are called /astorĩje/

Bailey (1924: 13-14), was the first to present an in-depth study of the grammar and phonology of Shina. He proposed three main types (dialects) of Shina based on his studies: Kohistani, Astori, and Gilgiti. i.e.

a) Kohistani (Chilasi and Kohistani)

b) Astori (Gurezi, Astori and Drasi)

c) Gilgiti

So, Bailey believes that 'Gilgit' is the true heartland of the Shina language and that Shina-speaking places include 'Gurez', 'Tiliel’, 'Burzil Valley’, 'valleys of upper Kishenganga’, 'the districts of Astor,' ‘Chilas’, ‘Kohistan’, and 'Gilgit’, among others. Thus, “Astori, Gurezi, and Drasi make up the Astori group” (Radloff. 1992: 122). 

We have a total of 31 consonants (sometimes 33, as two are repeated according to the situation) in the Gurezi (Astori) dialect of the Shina language. i.e. p, pʰ, b, t, tʰ, d, k, kʰ, g, ʈ, ʈʰ, ɖ, s, z, ʃ, ʦ, ʦʰ, h, ʧ, ʧʰ, ʒ, m, n, w, j, l, r, ɽ, ʂ, ɳ, ŋ. In addition, there are 5 primary vowels in the Gurezi Shina. Considering length and nasalization, the total number of distinctive vowel phonemes in Shina is 20. Following are the graphemes for the non-nasalized 5 short vowels and non-nasalized 5 long vowels of Gurezi Shina: /i//ii//e//ee//a//aa//u//uu//o//oo/. Whereas, following is given the 10 nasalized vowels: /ĩ//ĩĩ////ẽẽ, /ã//ãã//ũ//ũũ//õ//õõ/ (Ahmed, 2020, 105-109). So, henceforth we will pin down all the necessary words of the Astori dialect according to the above distribution of phonetic transcription.

The most notable thing is that we are going to focus on just similarities between the selected languages only. Furthermore, it is worthful to mention that both under-examined languages have the process of affixation (prefixation and suffixation), which is very influential for new word formation and meaning changing and sometimes word class changing. Thus, our research will draw the attention of the readers toward these three topics that will cover an equal portion of both languages.

Thus, in the English language, we can find many examples of prefixes and suffixes that become cause to create new words.

For instance:

Treat – retreat

Fortune – misfortune    

Fix – prefix

Thus, in the above examples ‘re’, ‘mis’, and ‘pre’ are prefixes. So, we can see there is a variation based on meaning and identity. 

Similarly, the Astori dialect has also a concept of prefixes and suffixes which are very beneficial for new word formation.

/bό/ - /nẻbό/ means ‘don’t go’ 

/tʰee/ - /kamtʰee/ means ‘to reduce’

So, in the above-mentioned examples, ‘nẻ’ and ‘kam’ are supposed to be prefixed. That changes the meaning and word class as well.  

In addition, we can also highlight suffixes from the English language as well. i.e.

Help – helpful

Change – changeable

Sad – sadness

Thus, words including the above examples ‘ful’, ‘able’, and ‘ness’ are suffixes. Due to this, there is variation in meaning and word class also.

Moreover, there are so many suffixes in the Astori dialect also, these are

/pijoo̗/ - /pijoo̗nããw/ means ‘drinkable’

/a̗j/ - /a̗jlό/ means ‘goat family’

/aʐó/ - /aʐónó/ means ‘inside’

Hence, the above examples have example suffixes, that are ‘nããw’, ‘lό’, and ‘nó’. Thus, the meaning and class of words changed, and eventually, new words formed.

Jones, W. (1786). Furthermore, says that the comparative technique was established and successfully used to reconstruct the parent language, Proto-Indo-European and has been used in the study of other language families. So, without going too deeply into the subject, let us try to lay down some basic and broad concepts in comparative linguistics. Comparative linguistics (previously Comparative Grammar or Comparative Philology- the humanistic study of language and literature) is the study of the relationships or correspondences between two or more languages, as well as the methodologies used to determine whether they share a common ancestor. Eventually, in Europe during the nineteenth century, comparative grammar was the most prominent branch of linguistics. The discovery by Sir William Jones that Sanskrit was related to Latin, Greek, and German sparked the study, also known as comparative philology.

 

1.2    Research Problem

This effort will explore the prefixation and suffixation process in the English language and the Astori dialect of the Shina language. Moreover, this study would be influential for the investigation of similarities in the affixation processes of both languages. Eventually, it will propose the meaning-changing phenomena with the help of suffixation and prefixation in both selected languages and the changing of word class as well. 

 

1.3    Research Objectives

The main objectives of this research are as follows:

1. To describe the suffixation and prefixation phenomena in the English language and Astori dialect of the Shina language.

2. To determine the mutual resemblance of the suffixation and prefixation process among both languages.

3. To verify the affixation process is influential for meaning changing of the base form of the targeted word.

4. It will also explain whether the variation in the word class of base form occurs or not. 

 

1.4    Research Questions

According to the supremacy of the topic, our study will be confined to these basic questions so that is:        

1. Do both the English language and the Astori dialect of the Shina language have an affixation process?

2. Is there any similarity between the suffixes and prefixes of both languages?

3. Does affixation help to change the meaning of the base form (of the word)?

4. Does the word class remain the same or change after (affixation) formation of a new word and the meaning change?

 

1.5    Delimitation of the Study

This little effort of comparison for affixation among both languages focuses on the similarities only. So, it has no relation to the differences between the two targeted languages for their suffixation and prefixation.

Moreover, this study also highlights the prefixation and suffixation process in native words or vocabulary of the Astori dialect of the Shina language only. Thus, all the borrowed words or loan vocabulary have been omitted during this lexical study. In addition, it focuses on the class changing words, meaning changing, and new word formation.

At last, our study is confined to some conversions including ‘noun to the noun’, ‘verb to the noun’, ‘noun to the adjective’, and ‘adjective to the noun’ only. While the rest of the parts of speech are not included in this study.       


CHAPTER 2 

LITERATURE REVIEW


Literature Review

Ahmed, M. (2020: 77) made an effort to show the purpose of ‘able’, ‘like’, and ‘-ish’, etc. The derived adjectives in the Gurezi dialect are generated by adding suffixes /-ããj/ and /-ããw/ to verbs, nouns, adverbs, and adjectives. These derivative adjectives have a lot of different meanings. For instance, when the adjectives are derived from nouns, these generally convey the meaning of ‘like’ as exemplified in:

Noun           Gloss              Adjective                        Gloss

/ʐe̗el/           forest           /ʐee.lã̗ãw/, /ʐee.lã̗ãj/         forest like

 /ɡo̗oʂ/         home            /ɡoo.ʐã̗ãw/                       home like

 /mu̗u.ʐu/      rat               /muu.ʐã̗ãw/                       rat like

 /mu.ɳi/       tuber             /mu.ɳaã̗̃w/                        tuber like

Again (Ahmed, M. 2020: 78) says that the scope of verb formation from adjectives is very limited in Gurezi. So, we tried our best to collect some of them. i.e.  

 Adjective         Root             Suffix                Verb                    Gloss

 /kri.du̗/           /krid/ʒ/           /-oonu/            /kri.ʒoo̗.nu/        bitter/ to become bitter

 /ʐi.gu̗/              /ʐig/              /-oonu/            /ʐi.gjoo̗.nu/        dry/ to become dry

 /waa̗.zu/          /was/             /-oonu/            /wa.ʒoo̗.nu/        descending/ to descend

 /ʈʃii.̗mu/          /ʈʃiim/            /-oonu/            /ʈʃi.mjoo̗.nu/       thick/ to become thick

 /miʂ.tu̗/           /miʂt/             /-oonu/            /miʂ.tjoo.nu/      well/ to become well

 /ʈʃi.ʈu̗/             /ʈʃiʈ/               /-oonu/            /ʈʃi.ʈjoo̗.nu/        bitter/ to become bitter

/ʃu.ku̗/              /ʃuk/              /-oonu/            /ʃuk.ʒjoo̗.nu/     dry/ to become dry


Plag (2003: 98-99) has divided prefixes among four broad categories, with an additional miscellanea category for prefixes that could never be characterized adequately. These would be (a) quantitative prefixes such as poly-, semi-, hyper-, uni-, di-, bi-, multi-, etc. (e.g. polysyllabic, semi-conscious, hyperactive, unification, ditransitive, bifurcation, multilateral); (b) locative prefixes such as endo-, counter-, circum-, trans- retro- inter-, etc. (e.g. endocentric, counterbalance, circumscribe, transmigrate, retroflex, intergalactic); (c) temporal prefix like pre-, fore-, ante-, neo-, post-, mis-, mal- etc. (e.g. preconcert, forsee, antedate, neoclassical, postmodify, mis-trail, malfunction); (d) negative prefixes like un-, in-, de-, a-, non-, dis-, etc. (e.g. unwrap, inactive, dethrone, asymmetrical, non-commercial, disagree); and others like vice-, pseudo-, etc. (e.g. vice-regal, pseudo-archaic).

 

According to Quirk (1985: 1546), the affix a- (together with be- and en-) serves primarily as a class-changing prefix along with little discrete semantic value. Its meaning is similar to that of the progressive (like aglow = glowing). Moreover, asleep, atop, abroad, ablaze, and apart are some other examples.

 

The varieties on the boundaries of the Kashmiri-speaking zones, Gurezi as well as Drasi, also developed possession suffix that expresses things for sexual identity, coinciding only with the gender of a possessive noun, according to Schmidt (2008: 21). It's most likely due to contact with Kashmiri, where the possessive suffix inflects to agree with possessive nouns. The following are some examples of possessive singular cases:

Gilgiti: muliay-ey nom  -----  (the girl's name)

Kohistani: Gozo-ee ʃeron ---- (the roof of the house)gozo-ee tiki ---- (home-made bread/food)

 

Rajapurohit, B. (2012: 41). Singular nouns are given the plural suffix -e to render each other plural without altering their noun. Take note of the singular and plural versions that follow. 

Singular forms                                 Plural forms

[sin] `River’                                       [sine] `Rivers’

[pon] `Road’                                      [pone] `Roads

[ʧhúp] `Bank of river’                       [ʧhúpe] `Banks of river’

[ó:ʃ] `Air’                                          [ó:ʃe] `Different kinds of Air’

[udú:] `Dust’                                     [udú:e] `Dusts’

[so:r] `Ice’                                         [so:re] `Ices’

[kha:y] `Pebble’                                [kha:ye] `Pebbles’

[ró:ŋs] `Deer’                                    [ró:ŋse] `Deer’

 

Ahmed, S. (2020: 4). There are following categories of verbs that remain verbs by adding the prefixes to the verbs as follows:

Prefix                     Meaning                        Verb                               Verb

re-                            ‘again’                          organize                        reorganize

un-                    ‘the opposite of’                    cover                            uncover

mis-                  ‘the opposite of’                    apply                            misapply

dis-                 ‘not, the opposite of’               please                           displease

 

  

CHAPTER 3 

METHODOLOGY

 

Methodology 

To conduct our research, we have gone through the technique of qualitative method of data collection. Therefore, we employed three primary methodologies to examine the similarities between the English language and the Astori dialect of the Shina language on the basis of prefixation and suffixation. Thus, we reviewed articles and related documents, arranged word lists, and attempted interviews.  

Moreover, after a micro-level study in affixation of both languages in articles, books and related documents, we assembled a word list of both languages on the basis of similarity in their characteristics. Apart from them, we divided affixes with their respective words to form the combination of affixation and word.

As a result, we came up with new words, meaning changing process and eventually word class variation as well. For instance, the conversion of nouns into nouns, verbs into nouns, nouns into adjectives and adjectives into nouns etc. Hence, this method is adopted to mark the similarity among the selected languages. 

At last, several interviews have been taken with the native speakers of the Astori dialect of the Shina language and they were almost 100 in number, in which most of the participants were literate and also were familiar with the concept of prefixation and suffixation in both targeted languages. Also, the location of this sampling was the Astori speakers around the centre area of Gilgit. In addition, for the affixation in the English language we strived in different books and websites as well as in articles also. 

 

  

CHAPTER 4 

DISCUSSION/FINDINGS

 

Discussion 

As previously mentioned, all the selected words for the following discussion are not borrowed from any other language, rather these are the native vocabulary of the Astori dialect of Shina language only. So that the similarities can be extracted purely and appropriately.

Moreover, this study primarily concentrates on the meaning-changing phenomena undergone by affixation and also will be beneficial for observing the changing word class and new word formation during affixation. In addition, it would also demonstrate that sometimes only meaning changes and new words form but the word class remains the same, while rarely affixation changes the part of speech as well. Thus, we will try to present the previously mentioned qualities as a similarity between both languages. Therefore, our primary target is to reveal all these words that vary by class on the basis of prefixation and suffixation. Let’s discuss them one by one;     

 

Prefixes

A prefix is an affix that is adjoined to the left of the base of a word or comes before the base form and plays a lexical role – which means it allows for the construction of a large number of new words. It could be a letter or a group of letters. Prefixes usually do not change the class of the base word but apart from new word formation and meaning changing, this effort really focuses on the class-changing cases also. For instance;

 

From Nouns to Nouns

Following are a few examples of the English language from noun to noun conversion with the help of prefixes. So, firstly we have nouns, then we have prefixes and last, there are nouns (which are a combination of both prefixes and nouns). As a result, new words are formed but the class of words remain the same. Thus, ‘tele’, ‘up’, ‘pre’, ‘tri’, ‘mis’ and ‘micro’ are prefixes. i.e. 

Nouns                                Prefixes                        Nouns

Communication                   tele                           Telecommunication    

Grade                                    up                            Upgrade

Position                                pre                            Preposition 

Angle                                    tri                             Triangle

Deed                                     mis                           Misdeed

Scope                                  micro                         Microscope

 

Similarly, some examples of the Astori dialect of the Shina language are from noun to noun conversion by using prefixes. So, initially, we have nouns, then there are prefixes, and eventually, we have nouns (a combination of prefixes and nouns). Consequently, new words are formed but the class of words remains the same. Therefore, all sounds like ‘/k/’, ‘/b/’, ‘/dá/’, ‘/hĩ/’, ‘/la/’ and ‘/tr/’ are prefixes of this dialect. i.e.

Nouns                             Prefixes                   Nouns

(Female goat)                                                   (Field)

 /ãj/                                   /k/                              /kãj/

 

(F. goat)                                                        (Cooked rice)

/ãj/                                  /b/                               /bãj/

 

(Daughter)                                                   (Grandmother) 

/dĩh/                                  //                         /dádĩh

 

(Barley)                                                           (Heart)  

/joo̗/                                  //                          /hĩjoo̗/

 

(Mud)                                                              (Tail)

/mõʈĩ/                                /la/                        /lamõʈĩ/

 

(F. goat)                                                        (Window)  

/ãj/                                      /tr/                          /trãj/

 

From Nouns to Adjectives

In this study, we will see how prefixes alter the meaning as well as the word class and form new words by converting nouns into adjectives. So, just observe the below English language examples where we initially have nouns. There are prefixes and eventually, there are adjectives (which are a combination of both prefixes and nouns). So, as a result, meaning changes, new words are formed and the class of words also differs. Moreover, ‘in’, ‘pre’, ‘post’ and ‘im’ are prefixes. i.e.

Nouns                    Prefixes                 Adjectives

Land                         in                         Inland  

War                          pre                        Pre-war 

Meridian                  post                      Postmeridian 

Patient                      im                        Impatient  

Similarly, the Astori dialect of the Shina language also has the same example meaning from noun to adjective conversion by using prefixes. So, initially, we have nouns, then there are prefixes and eventually, there are adjectives (which are a combination of both prefixes and nouns). Therefore, new words are formed but the class of words also changes. Thus, ‘/sĩ/’, ‘/ʃej/’, ‘/ʃe/’, ‘/p/’ and ‘/l/’ are prefixes of this dialect. i.e.

Nouns                      Prefixes                     Adjectives

(Barley)                                                     (Handsome/Beautiful)   

/joo̗/                            //                                /sĩjoo̗

 

(sawdust)                                                     (Unmarried boy)

/kʰo/                            /ʃej/                              /ʃejkʰo/

 

(Cap)                                                           (Unmarried girl)

/kʰo̗j/                            /ʃe/                               /ʃekʰo̗j 

 

(Cry)                                                               (Old)

/roono/                         /p/                              /proono/

 

(Today)                                                           (Shy)

//                               /l/                                 /laʃ/  

 

From Verbs to Nouns

As following are some examples of the English language in which we can use prefixes to convert verbs to nouns easily. So, at first, we may counter verbs, then we can see prefixes and at the end, there are nouns (which are the combination of both prefixes and verbs). Hence, new words are formed and the class of words also changes. Thus, ‘in’, ‘over’ and ‘anti’ are prefixes of this language. i.e. 

Verbs                     Prefixes                Nouns 

Come                         in                       Income 

Lay                            in                         Inlay

Flow                         over                    overflow 

dote                           anti                     antidote

In the same study, we may notice that the Astori dialect of the Shina language also have several examples in which we can use prefixes to convert verb to noun. So, with the beginning we may counter verbs, then we can see prefixes and at the end, there are nouns (which are the combination of both prefixes and verbs). As a result, new words are formed and the class of words also changes. Thus, ‘///ba//ʐa/’, ‘/d/ and ‘/mi/’ are prefixes. i.e. 

Verbs                      Prefixes               Nouns

(Spray)                                                  Oxen

/sãr/                           //                      /sãr/

 

(Putting/to put)                                    Echo/Beating drum

/ʃõno/                        /ba/                       /baʃõno/

 

(To cry)                                                 (Orphan)

/ró:/                           /ʐa/                         /ʐaró:/

 

(To come)                                                (Bull)

õõno/                       /d/                            /dõõno/

 

(To gather)                                             (Bone marrow)

/yó:/                        /mi/                          /miyoo/

 

From Adjectives to Nouns

In this part, we can see how prefixes are altering the meaning as well as the word class and form new words by converting adjectives into nouns. So, let's study the below English language examples in which initially we have adjectives. There are prefixes and finally, we have nouns (which are a combination of both adjectives and suffixes). In addition, ‘sub’, ‘dis’ and ‘trans’ are prefixes of this language. i.e.

Adjectives           Prefixes              Nouns 

Marine                    sub                  Submarine 

Conscious               sub                  Subconscious

Lighted                    dis                  Dislighted 

National                  trans               Transnational (someone operating in several countries)

Similarly, the Astori dialect of the Shina language also has noun-to-adjective conversion by using prefixes. So, at the beginning we have adjectives, then there are prefixes and eventually, there are nouns (which are a combination of both prefixes and adjectives). Thus, the following examples have ‘/ʐãʂ/’, ‘/a/’ and ‘/ʈu/’ which are supposed to be prefixes. i.e.

Adjectives            Prefixes                  Nouns

(Curve/bend)                                        (Head of an area)

/ʈẽrõ/                           /ʐãʂ/                    /ʐãʂʈẽrõ/

 

(White)                                                  (Eyes)

/ʃʰee̗/                            /a/                       /aʃʰee̗/

 

(Strict)                                                   (Basket)

/ku̗ri/                          /ʈu/                       /ʈuku̗ri/

 

(Blind)                                                   (Walnut)

/tsʰoo/                         /a/                        /atsʰoo/

 

Suffixes

A suffix is an affix that is adjoined to the right of the base of a word and changes the meaning the of base form. The most popular suffixes of the English language are –ness, -ed, -er, -est, -ity, -ly, -al, -ous, -ary, -ic, -ish, -less, -like and -y etc.

Suffixes can also be a letter or group of letters. They may change the meaning of a base word, the class of the word and form a new word in the language. So, our primary step is to highlight these facts.

In the following examples, we will encounter the influence of suffixes on the words of the English language and the Astori dialect of the Shina language, which basically changes the meaning, class and formation of new words in the language. i.e.   

 

From Nouns to Nouns

Here are several examples of the English language, in which suffixes modify nouns to nouns, that have the same class but different meanings and form new words. So, firstly we have nouns, then we have prefixes and last, there are nouns (which are a combination of both nouns and suffixes). Thus, ‘ess’, ‘er’, ‘yer’ and ‘ness’, are suffixes of the English language. i.e.

Nouns                Suffixes                   Nouns

Actor                     ess                         Actress 

Paint                      er                          Painter

Law                       yer                        Lawyer

Poor                      ness                      Poorness

Tiger                     ess                        Tigress

Waiter                   ess                        Waitress

Similarly, there are many examples of the Astori dialect of the Shina language from noun to noun conversion with the help of suffixes. So, initially, we have nouns, then there are suffixes and eventually, we have nouns (a combination of nouns and suffixes). Therefore, ‘/ki/’, /ʂek/’, //’, /ro/’, /í/ and ‘/ŋu/’ are suffixes of this dialect. i.e.

Nouns                              Suffixes                 Nouns

(Female servant)                                             (Bubble)

/bó:i/                                    /ki/                       /bó:iki/ 

 

(Cow)                                                               (House) 

/go:/                                    /ʂek/                      /go:ʂek/

 

(Cloud)                                                             (Abdomen)

/áʐo/                                     //                         /áʐoló/

 

(Butter)                                                             (Rock/Stone) 

/gí:/                                       /ro/                          /gí:ro/

 

(Mint)                                                                   (Ant)

/phílí:l/                                   /í/                          /phílí:lí/

 

(Fire)                                                                  (Maize stakes)

/pʰu̗u/                                    /ŋu/                         /pʰu̗uŋu/

 

From Nouns to Adjectives

In this examination portion, we will be familiar with the concept that how suffixes modify the meaning as well as the word class and form new words by converting nouns into adjectives. Also, this phenomenon can be noticed in both selected languages. So, just observe the below English language examples in which initially we have nouns. There are suffixes and eventually, there are adjectives (which are a combination of both nouns and suffixes). Moreover, ‘full’, ‘able’, ‘ing’ and ‘ous’ are suffixes. i.e.

Nouns                        Suffixes               Adjectives

Cheer                            full                    Cheerful

Fashion                         able                   Fashionable

Interest                          ing                    Interesting

Danger                          ous                    Dangerous

Profession                      al                     Professional 

Like the English language, the Astori dialect of the Shina language also has many examples of conversion from noun to adjective by using suffixes. So, firstly we have nouns, then there are suffixes and last, there are adjectives (which are a combination of both nouns and suffixes). Thus, ‘/bo/’, ‘/ʂu/’, ‘/ʈo/’ and ‘’ are suffixes. i.e.

Nouns                     Suffixes                   Adjectives

(One)                                                       (Alone/Single) 

/ek/                            /bo/                            /ekbo/

 

(Fire)                                                           (Empty) 

/pʰu̗u/                        /ʂu/                              /pʰu̗uʂu/

 

(Harm/Damage)                                          (Short)

/khu/                          /ʈo/                              /khuʈo/

 

From Verbs to Nouns

The following are a few examples of the English language in which we can use suffixes to convert verbs into nouns very easily. So, at first, we may find verbs, then we can see suffixes and at the end, there are nouns (which are a combination of both verbs and suffixes). Thus, ‘ment’, ‘tion’, ‘ist’ ‘er’ and ‘ar’ all are suffixes. i.e. 

Verbs                             Suffixes                 Nouns

Establish                          ment                    Establishment

Investigate                        tion                     Investigation

Accompany                       ist                       Accompanist

Rob                                    er                       Robber

Beg                                    ar                       Beggar

Just like the English language, our Astori dialect of the Shina language has several examples of conversion from verb to adjective by using suffixes. So, first, we have verbs, then there are suffixes and last, there are nouns (which are a combination of both verbs and suffixes). Thus, ‘/ji/’, ‘/méh/’, ‘/kúr/’, ‘/j/’, ‘/kuɳ//khá/’ and ‘/mo/’ are suffixes. i.e.

Verbs                          Suffixes                      Nouns

(Eat)                                                                 (Itch)

/khá/                                /ji/                            /khá:ji/ 

 

(Snore)                                                           (Date palm)

/khór/                              /méh/                      /khórméh/

 

(Cough)                                                           (Puppy)

/khu/                                 /kúr/                       /khukúr/

 

(Give)                                                             (Beard)

//                                   /j/                             /dãj/

 

(Cough)                                                       (Pigeon pea)

/khú/                                 /kuɳ/                      /khúkuɳ/

 

(Eat)                                                      (Support/Partialities)

/khá/                                 /khá/                       /khári/

 

(Eat)                                                              (Eater) 

/khá/                                  /mo/                      /khámo/

 

From Adjectives to Nouns

In this area, we can observe how suffixes are changing the meaning as well as the word class and form new words by exchanging adjectives for nouns. So, here we will study the English language examples in which initially we have adjectives. There are suffixes and finally, we have nouns (a combination of both adjectives and suffixes). In addition, ‘ness’, ‘dom’, ‘ity’ and ‘y’ are suffixes. i.e.

Adjectives                   Suffixes                      Nouns

Sick                                 ness                        Sickness

Free                                 dom                        Freedom

Stupid                               ity                         Stupidity

Difficult                            y                           Difficulty

Silly                                ness                        Silliness

Like others, the Astori dialect of the Shina language also has several examples of adjective-to-noun conversion with the help of suffixes. So, at the beginning we have adjectives, then there are suffixes and eventually, there are nouns (which are a combination of both adjectives and suffixes). Thus, the following examples have ‘/jar/’, ‘/alo/’ and ‘/jaa̗r/’, which can be considered suffixes of this regional language. i.e.

Adjectives                  Suffixes                      Nouns

(Small)                                                          (Poverty)   

/ʃuni/                               /jar/                         /ʃunijar/   

 

(Less)                                                            (Blanket)

/kam/                              /alo/                        /kamalo/

 

(Bitter)                                                          (Bitterness)

/ʈʃi.ʈ/                           /jaa̗r/                          /ʈʃi.ʈjaa̗r/

 

(Well)                                                            (Wellbeing)  

/miʂtu̗/                             /jaa̗r/                      /miʂtjaa̗r/

 

(Light)                                                           (lightness)

/loo̗ku/                             /jaa̗r/                       /lok.jaa̗r/

 

 

CHAPTER 5

CONCLUSION

 

Conclusion

The flexibility of every language allows their different aspects to be compared. That’s why we have made a comparison between the English language and the Astori dialect of the Shina language on the basis of prefixation and suffixation. Thus, this little effort was about similarities found in both languages according to the selected topic. As we have highlighted previously the dialects of Astori have three subparts named Astori, Gurezi and Drasi. So, we collected our primary data from pre-existing articles, books, grammar and websites etc. While other required data is gained from the native speakers of these dialects living around the centre of Gilgit City. Moreover, we introduced the concept of prefixation and suffixation, then mentioned their examples from both languages and then elaborated on all the variations that occur due to the application of affixation. In addition, we tried our best to portray the results coming from the suffixation and prefixation phenomenon. This proved that the meaning of words can be changed with the help of affixation, similarly, new words can be formed because of prefixation or suffixation and eventually, we can modify the class of base words that undergo the affixation process. Finally, both languages have an affixation process and the same results can be extracted from them after adding prefixes and suffixes. Thus, this could be the essential similarity between both languages and hence, this is the basic goal of conducting our research. i.e. the consequences of applying prefixation and suffixation have the same impact on both languages which is the greatest similarity among them.

 


References

Bailey, T. G. 1924. “The grammar of the Shina (ṢiṆā) language”. London: The

Royal Asiatic Society.

B. B. Rajapurohit. 2012. “Grammar of Shina Language and Vocabulary”, p. 41.

Kohistani, R. - Schmidt, R. L. (2006): Shina in Contemporary Pakistan. In: Saxena,

Anju – Borin, Lars (eds): Trends in Linguistics: Lesser-Known Languages

of South Asia. Berlin, Mouton de Gruyter, p. 137.

Musavir Ahmed, 2020. “A Descriptive Grammar of Gurezi Shina”. P. 77-78.

Peter C. Backstrom and Carla Radloff (eds.). 1992. “The Dialects of Shina”.

Sociolinguistic Survey of Northern Pakistan. Volume 2. 122-150.

Plag, Ingo (2003), “Word-Formation in English”, Cambridge University Press,

Cambridge. 

Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech, Jan Svartvik (1985), “A

Comprehensive Grammar of the English Language”, Pearson Education,

Harlow.

Ruth Laila Schmidt and Vijay Kumar Kaul, 2008, pa. “A Comparative Analysis of

Shina and Kashmiri Vocabularies", p. 21.

Saza Ahmed, 2020. "A Comparative Morphological Approach to Class

Maintaining Derivational Affixes in English and Kurdish", Vol: 6. “Journal

of University of Human Development”, p: 4.

Sir William Jones, 1786. “Language Families, and Indo-European”.

            https://doi.org/10.1080/00437956.1992.12098279. Garland Cannon (1992).


Comments