A Linguistic Study of the Tamil Language

Co-Authored with Meghna C. N

Index

  1. Introduction
  2. Phonemes
  3. Minimal Pairs in Tamil
  4. Allophones and Free Variation
  5. Syllable Structures
  6. Morphology
  7. Adaptation of Borrowed English Words in Tamil
  8. Conclusion
  9. References


Chapter 1: Introduction 

One of the oldest surviving classical languages in the world, Tamil ([t̪amiɻ]) belongs to the South Dravidian branch of what the 19th century linguist Robert Caldwell termed the Dravidian language family, alongside Malayalam, Telugu, Kannada and Tulu among others.  (Krishnamurti, 2003).
With over 75 million speakers worldwide of which approximately 60 million are native speakers in India, Tamil is the 17th largest spoken tongue in the world. Colonialism and migration in the 20th century has imparted a Tamil diaspora in Indonesia, the Philippines, Malaysia, Mauritius, South Africa, Fiji, the United States, the United Kingdom, Canada, South Africa, Germany and the Netherlands. In India, Sri Lanka and Singapore it is recognized as an official language and in Malaysia and Mauritius as a minority language. (Tamil, 2009)

Tamil is said to have evolved from the common ancestor, Proto-Dravidian, which is considered indigenous to our subcontinent and spoken around 3rd millennium BC. Tamil’s earliest recorded inscriptions date back to the middle of the 3rd century BC and the earliest literature belongs to the Sangam Age no less than over 2000 years ago. Tolkappiyam, the oldest existing text on Tamil language, grammar and linguistics, belongs to the same era. (Steever, 2015).

Tamil has evolved through three stages corresponding to three periods of history: Old Tamil between 300 BC and AD 700, Middle Tamil between AD 700 and AD 1600 and Modern Tamil, from AD 1600 to the present. (Steever, 2015). Modern Tamil, which is analyzed in this project, is vastly different from the earlier forms in terms of phonology, morphology, syntax and script and has witnessed greater influence from other languages such as English and Portuguese.

The modern Tamil script has evolved over the centuries from early Tamil Brahmi, to early Vattezhuthu (rounded alphabets) in the 5th and 6th centuries,  to which in the 7th century features from the Grantha Script (derived from Southern Brahmi) were added, to form the script currently in use. In the 19th and 20th century, for the convenience of printing, many alphabets were further simplified in form. An interesting feature of the Tamil script is that it is an abugida writing system, which means that a single unit is a combination of a vowel and a consonant. (Steever, 2015).

Like pre-modern Greek and Arabic, Tamil is a diaglossic language, implying a dual system varying by social status, consisting of a ‘high’ and a ‘low’ variety of spoken and written Tamil. The classical version generally considered ‘superior’ and purer is known as Centamiz or beautiful Tamil, while the dialectal or colloquial version is called Kotumtamiz or crooked/vulgar Tamil which is not used in formal contexts. (Krishnamurti, 2003).

Besides this diaglossia, the sociolinguistics of Tamil is marked by a large variety of caste-based and regional dialects. These dialects have evolved in phonologically distinct ways from old Tamil. In regions close to the linguistic boundaries demonstrate influence from the neighbouring language, such as the Tamil used in Palakkad with lexicon and syntax borrowed from Malayalam. Further, caste-based sociolects such as Brahmin and non-Brahmin Tamil distinguish Tamil speakers from each other.

In this project, we have studied not a specific dialect, but Standard Tamil, or rather, a variety of Tamil standardized through radios, films, televisions and so on, working with the understanding that this is the Tamil used in common spaces by speakers of different dialects of the language. In the following chapters, we explore the phonological and morphological features of this standard Tamil.

Chapter 2: Phonemes 

A phoneme is a minimal, distinctive unit of sound in a language system. It is an abstract or theoretical idea – it is almost impossible to produce a pure phoneme. In practice, we are dealing with speech sounds. Phonemes must essentially distinguish one word from another in a language. It functions contrastively – if we substitute one sound for another in a word, it brings about a change in meaning.

Sounds may be vowels or consonants. A consonant is a speech sound occurring when the vocal tract is partially or wholly blocked. These can be described according to the place of articulation (where in the mouth the air is obstructed), manner of articulation (how the air is obstructed), and voicing (whether vocal cords are open or closed).

A vowel, on the other hand, is produced when there is no obstruction of air in the mouth. They do however require some tongue movement. Vowels are described according according to the height, backness (of the tongue), and tenseness, and sometimes lip rounding. Vowels which require a single articulatory configuration are called monothongs, as opposed to diphthongs, which is when two monothongs join to form one sound sequence. In English, there exist twelve monothongs and eight diphthongs.

Every language has a limited number of these speech sounds. A phoneme that appears in one may not exist as contrastive units of sound, or as sounds at all. English has 44 phonemes, while Tamil has 34.

Tamil phonology is characterised by a lack of aspirated sounds, presence of retroflex consonants, and multiple rhotics and nasals. In speech, word final nasals are cut, and the vowel before it is nasalised. It also does not distinguish between voiced and voiceless plosives; voiced stops exist only as allophones, governed by strict rules (discussed in Section 4). In addition to these, there exist borrowed sounds from the grantha script, initially added in order to write Sanskrit loanwords, now used to write words from other languages as well (these may often be replaced by native sounds). These borrowed consonants make up for the lack of fricatives in the language.

Consonants 

Labial Dental Labio-dental Alveolar Retroflex Palatal Palato-alveolar Velar
Plosives/Stops P T k
Nasals m n N ɲ ŋ
Trill ɾ
Approximant ʋ r ɻ j
Lateral l L
Affricate č ʤ*
Fricatives* F s ʂ H

 *Borrowed consonants (grantha) 

Vowels

Monothongs

Front Central Back
Long Short Long Short Long Short
Close  i: i u: u
Mid e: e o: o
Open a: a

 

Diphthongs

/ai/       /au/


Chapter 3: Minimal Pairs in Tamil 

In the previous chapter, we have seen that Modern Tamil uses 34 distinctive phonemes.
This distinctiveness derives from the fact that these phonemes are in contrastive distribution. In other words, if two sounds are in contrastive distribution their substitution will result in a change of meaning and they will make a minimal pair.  For example in English, the words /tɔi/ and /bɔi/ demonstrate that the phonemes /t/ and /b/ are minimal pairs, since the environment remains the same while a single phoneme substitution alters the meaning. This section looks at some of the minimal pairs in Tamil.

 

No. Minimal Pair Phoneme 1 Phoneme 2
1 /a/ and /aː/ /paDə/  ‘lie down’ /pa:Də/ ‘sing’
2 /a/ and /u:/ /paTTə/  ‘silk’ /pu:TTə/ ‘lock’
3 /a/ and /i/ /aDuppə/ ‘stove’ /iDuppə/  ‘hip’
4 /i/ and /u/ /kiLi/ ‘parrot’ /kuLi/ ‘bath’
5 /i/ and /i:/ iDə/  ‘leave/let go’ i:Də/ ‘house’
6 /i/ and /o/ /inDrə/  ‘today’ /onDrə/ ‘one’
7 /r/ and /ɻ/ /irə/ ‘be’ /iɻə/ ‘pull’
8 /N/ and /ɻ/ /paNəm/  ‘money’ / paɻəm/ ‘fruit/
9 /k/ and /p/ /kallə/ ‘stone’ /pallə/ ‘tooth’
10 /č/ and /p/ /ča:d̯am/ ‘rice’ /pa:dəm/ ‘foot’
11 /t̯/ and /p/ /uLi/ ‘drop’ /puLi/ ‘tarmarind’
12 /p/ and /ʋ/ /pa:l/  ‘milk’ /ʋa:l/ ‘tail’
13 /t̯/ and /ʋ/ /a:ji/ ‘mother’ /ʋa:ji/ ‘mouth’
14 /k/ and /m/ /ka:lai/  ‘morning’ /ma:lai/ ‘evening’
15 /ʋ/ and /m/ /ʋe:gam/  ‘fast’ /me:gam/ ‘cloud’
16 /k/ and /t̯/ /kaLLə/  ‘toddy’ /aLLə/ ‘shove’
17 /n̯/ and /t̯/ /i:/ ‘fire’ /i:/ ‘you’
18 /ñ/ and /ʋ/ /a:nam/ ‘knowledge’ /ʋa:nam/ ‘sky’
19 /R/ and /r/ /KaRai/ ‘stain’ /karai/ ‘river bank’
20 /L/ and /l/ /ʋaLai/  ‘bangle’ /ʋalai/ ‘net’

 

Chapter 4: Allophones and Free variations 

Allophones

Phonemes, as discussed in Section 2, are abstract and theoretical. They may be realized differently depending on the context. The individual sounds produced are called phones, and realizations of the same phoneme are called allophones of that phoneme. In other words, allophones are different ways of pronouncing a sound depending on its context, never causing a change in meaning. Whether a phone is a phoneme in itself or an allophone varies from language to language.

Two sounds are said to be allophones if they occur in mutually exclusive environments, meaning individual allophones can occur only in specific contexts, never in the same one. Occurrence of allophonic variants is predictable. This basic criterion of mutual exclusivity in environments is called complementary distribution.

A common allophonic variation in English is aspiration. The rule states that when voiceless plosives, [p] [t] [k], occur at the beginning of stressed syllable, they become aspirated.

Examples:        /pen/  [‘phen]                /əpa:(r)t/  [ə.’pha:(r)t]                /spit/       [‘spit]

/tæp/  [‘thæp]                /ətæk/                  [ə.’thæk]                   /sta:(r)t/  [‘sta:(r)t]

 

Another allophonic variation is of clear [l] and dark [ɫ]. The rule states that when /l/ is followed by a vowel or occurs word initially, it is palatal, clear [l]. However when /l/ is followed by a consonant or occurs word finally, it is velarised, dark [ɫ].

Examples:        /slip/     [lip]                  /lisp/     [lisp]

/pil/      [phiɫ]                /silk/     [siɫk]

/lʌl/      [lʌɫ]

 

In Tamil, allophones occur among plosives. Tamil plosive phonemes are all unvoiced. Voicing of these unvoiced plosives are complementary, and are thus allophones. The general rule is that plosives are unvoiced if they occur word initially, or are doubled (geminated), elsewhere, when they appear intervocalically (between vowels) or after a nasal, they are voiced. There are a limited number of exceptions to this rule.

The allophones are:     /k/        [k]        [g]

/t̪/         [t̪]        [d̪]

/T/        [T]       [D]

/p/        [p]        [b]

 

(#_ : Word initial, V: Vowel, N: Nasal, _ _ : Geminated consonant)

Example 1: /t̪/

[t̪] [d̪]
Word Gloss Environment Word Gloss Environment
[t̪erə] ‘street’ #_e [ka:d̪al] ‘love’ a:_a
[t̪akka:Li] ‘tomato’ #_a [mad̪ja:nam] ‘afternoon’ a_j
[t̪uNi]] ‘cloth/clothes’ #_u [pa:d̪i] ‘half’ a:_i
[kat̪t̪ə] ‘shout’ a_ _ə [pan̪d̪ə] ‘ball’ n̪_ə
[ʋit̪t̪jasam] ‘difference’ i_ _j [ʋan̪d̪a:n] ‘he came’ n̪_a:
[rat̪t̪am] ‘blood’ a_ _a [t̪an̪d̪ai] ‘father’ n̪_ai
[ʋa:rt̪t̪ai] ‘word’ r_ _ai

 

*The only nasal [d̪] can follow is [n̪].

The rule:          /t̪/ → [t̪] when occurring #_ or  _ _

/t̪/ → [d̪] when occuring V_V or N_V

There are a few exceptions, however, to this rule. Rarely, [d̪] may occur word initially.

Examples:        [d̪u:ram]          ‘distance’         #_V

[do:sai]            ‘dosa’               #_V

Example 2: /u/

Another allophonic variation is that of /u/. When occurring word finally, it is realized as [ə], in all other positions as [u].

[u] [ə]
Word Gloss Environment Word Gloss Environment
[ukkarə] ‘sit’ #_k [ukkarə] ‘sit’ r_#
[uLLe] ‘inside/within’ #_L [t̪appə] ‘mistake/fault’ p_#
[kuTTi] ‘small’ k_T [ka:lə] ‘leg’ l_#
[porume] ‘patience’ r_m [pa:TTə] ‘song’ T_#

 

The rule:          /u/ → [ə] when occurring word finally

 Free variation 

A single sound may also have different possible pronunciations in the same word without causing a change in meaning, and this is called free variation. These generally occur across geographic (dialects) or socio-economic divisions.

In English for example, there are two ways of pronouncing plosives at the end of a word, both of which are correct – they may be exploded or unexploded. For example, /kæt/ can be [khæt˺] or [khæt], /bæg/ can be [gæp˺] or [gæp]. Free variation in English is also common across dialects, like American and British English. For example, the word ‘dance’ would be pronounced /dans/ in British English, but /dæns/ in American.

Free variation in Tamil occurs across geographic region and caste. Variation also exists between the ‘pure’ or formal Tamil (/sen̪t̪amiɻ/) and the colloquially spoken Tamil (Tamil has a high degree of diaglossia).

Brahmin Tamil is one form of Tamil that exhibits a high degree of variation. It is a combination of Classical Tamil with Sanskrit loanwords, while non-Brahmin Tamil doesn’t show such a degree of borrowing, but is at the level of phonemic and morphological changes. Brahmin Tamil may thus even have a slightly different vocabulary. Some examples of differences in vocabulary include /a:t̪t̪ə/ (Brahmin) and /ʋi:Də/ (non-Brahmin) for ‘house, /t̪i:rt̪am/ or /ʤalam/ (Brahmin) and /t̪aNNir/ (non-Brahmin) for ‘water’. Brahmin Tamil itself varies according to geographic region.

At a phonemic level, free variation exists in the pronunciation of /ɻ/. Brahmin Tamil emphasizes the use of the retroflex approximant /ɻ/, while some non-Brahmin dialects, and colloquial language increasingly substitutes it for /l/ or /L/.

(The phoneme /ɻ/ cannot occur word initially, or geminate)

 

/ɻ/ (B) (NB) Gloss
/paɻam/ /palam/ ‘fruit’
/ki:ɻe/ /ki:le/ or /ki:ye/ ‘down’
[aɻagə] [alagə] ‘beauty’
[eɻud̪ə] [elud̪ə] ‘write’
/ʋaɻi/ /ʋali/ ‘donkey’
[magiɻčči] [magiLčči] ‘happiness’
/t̪amiɻ/ /t̪amil/ ‘Tamil’

 

Chapter 5: Syllable Structures 

Having looked at the phonemes, minimal pairs, allophones and free variations in Tamil, let us move on to analyze the possible syllable structures in the language. A syllable is a series of sounds in a word, typically organized around a vowel or a semi-vowel. For example, in English, the word ‘lovely’ contains two syllables and cal be split as ‘love• ly’ or /ləv.li/. Unlike English, Tamil syllables do not have phonemic stresses and stress usually falls on the first syllable. (Krishnamurti, 2003).

A syllable contains three parts: the Onset, the Nucleus and the Coda (the last two being part of the Rhyme). The first part or the Onset need not be present in a syllable, though when present it is usually a consonant or a consonant cluster. English allows up to three consonants in the Onset. In Tamil, however, consonant clusters are not allowed in this position. The second component of a syllable is the Rhyme. The Rhyme has two parts, the Nucleus and the Coda. The Coda is the last component of the syllable and like the Onset, may or may not be filled and when full, is always occupied by a consonant or a consonant cluster. While English allows up to four consonants in the Coda, Tamil allows only a maximum of two, and only rarely, as part of a borrowed word (like bank, from English) or as the first half of a geminate (as in /t̯i:rp.pə/ meaning ‘verdict’). Further, word end consonants are more prevalent in writing since a short vowel is usually added to the word in such a situation. It is therefore possible to surmise that native Tamil prefers simpler syllable structures and avoids consonant clusters in general.
However, in recent times with the influences from other languages such as English, consonant clusters are introduced in the language through borrowed words. The middle part, the Nucleus is the most important part of a syllable since this is the position which contains the vowel and without a vowel, a syllable cannot exist. It is the vowel that defines a syllable. This structure is represented in the diagram below.  (Thangarajan, 2012).

Considering both short and long vowels as V and the consonants as C, Tamil allows fundamentally five syllables structures: V, CV, VC, CVC and CVCC. (Narayana & Ramakrishnan, 2007).These are represented as follows:

 

  1. V

Examples:        /i:/ ‘fly’

/a:Də/  ‘goat’   /a:•Də/

/e:lam/ ‘elaichi’ /e:•lam/

/aDimai/ ‘slave’  /a•Dimai/

Let us analyze the word /a:Də/, which can be split into two syllables /a:•Də/. These two syllables can be represented as follows:

σ1 : V /a:/

σ2  : CV/Də/

 

  1. VC

Examples: /anBə/  ‘love’ /an•Bə/
/aɻagə/ ‘beauty’ /aɻ•ag•ə/
/iravə/ ‘night’ /ir•av•ə/
/uNmai/ ‘truth’ /uN•mai/

As you can see, /uNmai/ has two syllables as demonstrated below.

 

σ1 : VC /uN/

σ2  : CV/Də/

 

  1. CV

Examples:  /kəDi/ ‘bite’   /kə•Di/

/poDi/ ‘powder’ /po•Di/

/ka:Də/ ‘forest’ /ka:•Də/

/maram/ ‘tree’ /ma•ram/

Let us look at the word /podi/ which can be split into two syllables as /po•Di/.

 

σ1 : CV /po/

σ2  : CV/Di/

 

  1. CVC

Examples: /ka:l/ ‘leg’

/t̯ambi/ ‘brother’ /t̯am•bi/

/t̯ummal/ ‘sneezing’ /t̯um•mal/

/mi:n/ ‘fish’

 

The monosyllabic word /ka:l/ is in the CVC form.

σ1 : CVC /ka:l/

The next structure is not acceptable in spoken Tamil and is merely a theoretical representation. Therefore, the first four must be considered the primary popular syllable structures of spoken Tamil.

  1. CVCC

This syllable structure CVCC, is a theoretical representation since in actual utterance, phonotactic syllabification adds a V to the end. (Savaranan, 2009). Otherwise this structure usually represents the splitting of a geminate when it is preceded by the retroflex approximant /ɻ/or the alveolar trill /ɾ/. (Immanuel, 2008). In borrowed words such as /je•ra:ks/ (Xerox), consonant clusters are used. But sometimes, consonant clusters are syllabified separately, such as ‘form’ which is separated as ‘fa:•ram’

Examples :       /kamb/ ‘stick’
/t̯i:ɾppə/ ‘verdict’ /t̯i:ɾp•pə/
/art̯t̯am/ ‘meaning’ /art̯•t̯am/
/viɻčča/ ‘while falling’ /viɻč•ča/

 

Let us look at /vi:ɻčča/.

σ1:  /vi:ɻč/

  • σ2: CV / ča/

 

Thus, we have seen that Tamil has four major syllable structures V, VC, CV and CVC and the theoretical structure CVCC.

 

Chapter 6: Morphology 

Morphology is the study of internal structures of words and their meaningful parts. It is not only the study of existing words, but also of possible words and the rules and processes that govern their formation. Each minimal, distinctive unit of meaning or grammatical function is called a morpheme. A word, made up of morphemes, is the smallest grammatically independent unit of language, and the words of one’s language make up the lexicon.

There are two kinds of morphemes: free and bound. Free morphemes can exist on their own as words, while bound morphemes must necessarily attach themselves to another form. Another way to classify morphemes is into root and affixes – the root is the core of the word, which conveys the main meaning of the word, while affixes are what attach themselves to roots to add or modify meaning.

Free morphemes are of two kinds: lexical and functional. Lexical morphemes are those which have an inherent meaning of their own. These are usually nouns, verbs and adjectives. Examples of these from English include man, jump, honesty, table, etc. Functional morphemes are those which have no meaning by themselves, but add meaning in a given context and serve some grammatical function. These are usually articles, conjunctions, prepositions, and pronouns. Some examples are they, he, since, the, near, above, because, etc.

Bound morphemes are divided into derivational and inflectional. Derivational morphemes are those which change the form or grammatical category of the stem. This is a very productive process, and allows new words to enter the language, and to express ideas more succinctly. Some common derivational morphemes in English are prefixes dis-, mis-, pre-, un- and suffixes -able, -ness, -ment,  -ify, -ate, -y, -er.

Inflectional morphemes are used to add grammatical information to the stem. They indicate a property or feature within the grammatical framework. They often help contrast between singular/plural, gender, tenses, person, number and case. English has only eight inflectional morphemes: a plural marker, a possessive marker, four attached to verbs, and two to adjectives – the comparative and the superlative.

Tamil is said to be an agglutinating language, meaning that complex words are formed by adding affixes to the lexical base. It allows words combine without phonologically altering their structure. Almost all affixes in the language are suffixes. Prefixation is not a productive process in Tamil. They are often found on words borrowed from Sanskrit. Tamil is very rich in inflectional morphemes, which mark case, number, gender, tense, amongst others.

Case is what distinguishes the role played by a participant in an event. More specifically, it defines a noun’s relation to some other element in a clause or phrase. Tamil employs postpositions, they may be free or bound as a case marker.

Inflectional marking in Tamil:

Cases marked in Tamil include the:

  • accusative (indicating the object of a verb) (suffix: -e),
  • instrumental (means by which action is done) (suffix: -a:le, ‘by/with’),
  • associative (means by which something is done, association) (suffix: -o:Da, ‘along with’),
  • dative (indicates the direct object of a verb) (suffix: -(ə)kkə or -kki, ‘to’),
  • ablative (indicates movement from something) (suffix: -lernd̪ə/-kiTTernd̪ə, ‘from’),
  • genitive (possessive) (suffix: -o:Da, ‘of’),
  • locative (indicates a location) (suffix: -le/-kiTTe, ‘in/with’) and
  • vocative (addressing or calling).

*for nouns ending in -am and -ru, word final -m and -ru changes to -t̪t̪ when markers are added, for nouns ending in -Də, -Də changes to -TTə when markers are added.

Example 1: Dative Case

This is used when a verb has a noun toward which motion is expressed; to indicate possession, or the indirect benefactor of some act. The suffix -(ə)kkə is employed, except after front vowels, where it becomes -kki.

(The employment of different suffixes, as in this case, as realizations of the same marker (here, dative case) is known allomorphs.)

  1. /maram/ ‘tree’ +  -(ə)kkə         →  /marat̪t̪əkkə/  ‘to the tree’
  2. [ʋi:Də] ‘house’ +  -(ə)kkə         →  /ʋi:TTəkkə/  ‘to the house’
  3. /ko:jil/ ‘temple’       +  -(ə)kkə         →  /ko:jiləkkə/  ‘to the temple’
  4. /nari/ ‘fox’              +  -kki              →  /narikki/  ‘to the fox’
  5. [t̪ambi] ‘younger brother’ + -kki    →  [t̪ambikki] ‘to the younger brother’

 

Example 2: Locative case

It is used to express location with respect to some object. It may indicate lack of motion, containment in, or a means of transportation. In inanimate nouns, the marker -le is used, whereas in animates, -kiTTe is used. The animate marker means ‘in the possession of’.

 

  1. /maram/ ‘tree’ +  -le                →  /marat̪t̪əle/ ‘in the tree’
  2. [ʋi:Də] ‘house’ +  -le                →  /ʋi:TTule/  ‘in the house’
  3. /ko:jil/ ‘temple’       +  -le                →  /ko:jille/  ‘in the temple’
  4. /nari/ ‘fox’              +  -kiTTe          →  /narikkiTTe/  ‘in the possession of the fox’
  5. [t̪ambi] ‘younger-brother’ +  -kiTTe  →  [t̪ambikkiTTe] ‘in the possession of the younger brother’.

 

Another operation on nouns includes that of pluralization. The suffix -gaL (or -ŋgaL) is the most commonly used. The plural suffix is used before the case ending (if any).

 

  1. [maram] ‘tree’ +  -ŋgaL          →  [maraŋgaL]  ‘trees’
  2. [ʋi:Də] ‘house’ +  -gaL                        →  [ʋi:DəgaL]  ‘houses’
  3. /ko:jil/ ‘temple’       +  -gaL                        →  [ko:jilgaL] ‘temples’
  4. /nari/ ‘fox’              +  -gaL                        →  [narigaL]  ‘foxes’

 

Other than marking nouns, Tamil also marks number, person, gender and tense in verbs. What is unique is that the three are marked together on verbs. Second and third person markers also include a separate conjugation to show respect/honour.  Marking persons include gender, singular/plural, honour. They are marked in the format

Verb + Tense + Person (can include gender and singular/plural)

In the following examples, we have chosen a combination of tense and person marker and tried to show their recurrence in many verbs. Due to the highly diglossic nature of Tamil, inflectional markers are separable only in in the written or literary form (/sent̪amiɻ/). In speech, they are altered/shortened to the extent that individual morphemes cannot be identified. For the convenience of analysis, we have chosen to illustrate conjugation in the written form.

 

Example 1:      Verb + Present Continuous (-giɾə) + 1st Person Singular (-e:n)

  1. ‘run’ [o:Də] +  -giɾə  +  -e:n                        →  [o:Də]giɾ]e:n]  ‘I am running’
  2. ‘play’ [ʋiLaija:Də] +  -giɾə  +  -e:n  →  [ʋiLaija:Də]giɾ]e:n]  ‘I am playing’
  3. ‘see’ [pa:rə] +  -giɾə  +  -e:n                        →  [pa:r]kkiɾ]e:n]  ‘I am seeing’
  4. ‘talk/speak’ [pe:sə] +  -giɾə  +  -e:n                        →  [pe:sə]giɾ]e:n] ‘I am talking/speaking’

 

Example 2:      Verb + Future (-ʋ/-p)  +  2nd Person Singular (-a:i)

  1. ‘run’ [o:Də] +  -ʋ  +  -a:i     →  [o:Də]ʋ]a:i]  ‘you will run’
  2. ‘talk/speak’ [pe:sə] +  -ʋ  +  -a:i     →  [pe:sə]ʋ]a:i]  ‘you will talk/speak’
  3. ‘study/read’ [paDi] +  -p  +  -a:i     →  [paDi]pp]a:i]  ‘you will study/read’
  4. ‘buy’ [ʋa:ŋgə] +  -ʋ  +  -a:i  →  [ʋa:ŋgə]ʋ]a:i]  ‘you will buy’

 

Example 3: Verb +  Past perfect (-in/-t̪/-n̪)  +  3rd Person Female Singular (-a:L)

  1. ‘run’ [o:Də] + -in  +  -a:L                →  [o:D]in]a:L]  ‘she ran’
  2. ‘play’ [ʋiLaija:Də] +  -in  +  -a:L      →  [ʋiLaija:D]in]a:L] ‘she played’
  3. ‘see’ [pa:rə] +  -t̪  +  -a:L                →  [pa:r]t̪t̪]a:L]  ‘she saw’
  4. ‘study/read’ [paDi] +  -t̪  +  -a:L                 →  [paDi]t̪t̪]a:L]  ‘she read/studied’

 

Example 4:  Verb  +  Past perfect (-in/-t̪/-n̪)  +  3rd Person Neutral Singular (-ad̪ə)

  1. ‘run’ [o:Də] + -in  +  -ad̪ə               →   [o:D]in]ad̪ə]  ‘it ran’
  2. ‘play’ [ʋiLaija:Də] +  -in  +  -ad̪ə     →   [ʋiLaija:D]in]ad̪ə] ‘it played’
  3. ‘see’ [pa:rə] +  -t̪  +  -ad̪ə                →   [pa:r]t̪t̪]ad̪ə]  ‘it saw’
  4. ‘be located, is’ [irə] +  n̪  +  -ad̪ə                    →   [irə]n̪]ad̪ə]  ‘it was’

 

Some examples of derivation in Tamil:

 

Example 1:      Noun + masculine -ka:ran → [Noun]N ka:ran]N

-ka:ran is thus a bound, derivational suffix. It does not cause change in grammatical category. Similar derivations can be made from the female suffix -ka:ri

 

  1. ‘work’ /ve:lai/ + masc. -ka:ran → [ve:la]N kka:ran]N      ‘male worker’
  2. ‘kill’ /kolai/ + masc. -ka:ran → [kola]N kka:ran]N     ‘assassin’ (male)
  3. ‘drink’ [kuDi] + masc. -ka:ran → [kuDi]N kka:ran]N   ‘drunkard’ (male)
  4. ‘white’* [ʋeLLai] + masc. -ka:ran → [ʋeLLa]N kka:ran]N ‘white man’
  5. ‘garden’ [t̪o:TTam] + masc. -ka:ran → [t̪o:TTa]N kka:ran]N                 ‘gardener’
  6. ‘house’ [ʋi:Də] + masc. -ka:ran           → [ʋi:TTə]N kka:ran]N                 ‘man of the house/husband’

(*’white’ as in the race)

Example 2:      Adjective + -t̪anam → [Adjective] t̪anam]N

-t̪anam is a bound, derivational suffix. It changes the word category from an adjective to a noun.

  1. ‘good’ [n̪alla] + -t̪anam → [n̪alla]Adj t̪t̪anam]N                   ‘goodness’
  2. ‘stealth’ [kaLLam] + -t̪anam → [kaLLa]Adj t̪t̪anam]N               ‘stealthiness’
  3. ‘miserly’ [kanʤa] + -t̪anam → [kanʤa]Adj t̪t̪anam]N              ‘miserliness’
  4. ‘rowdy’ [po:kkiri] + -t̪anam → [po:kkiri]Adj t̪t̪anam]N          ‘rowdiness’
  5. ‘coward’ [ko:ɻai] + -t̪anam → [ko:ɻai]Adj t̪t̪anam]N             ‘cowardice’

 

Example 3:      Noun + -a:na → [Noun]a:na]Adj

-a:na is a bound, derivational suffix. It changes the word category from a noun to an adjective.

 

  1. ‘height’ [ujaram] + -a:na → [ujaram]N a:na]Adj                     ‘high’
  2. ‘beauty’ [aɻagə] + -a:na → [aɻagə]N a:na]Adj                        ‘beautiful’
  3. ‘honesty’ [ne:rmai] + -a:na → [ne:rmai]N a:na]Adj                    ‘honest’
  4. ‘speed’ [ʋe:gam] + -a:na → [ʋe:gam]N a:na]Adj                    ‘speedy’
  5. ‘truth’ [uNmai] + -a:na → [uNmai]N a:na]Adj                     ‘true’

 

Compounding

We have seen how Tamil morphology is rich in derivational and inflectional word formation processes. While the most productive processes in the language are suffixation and compounding, we see the working of borrowing, prefixation, clipping and even a few acronyms. Unlike English, where new words can be coined (coinage) without any phonological, morphological or semantic motivation, like ‘Xerox’, Tamil does not coin words in this manner. (Rajendran, 2005).

Compounding is a process by which two words are put together to form another word of a more complex structure with possible change in word category. The constituent elements of a compound cannot be rearranged. In Tamil, compounding is an extraordinarily productive process and in this section we will merely look at three kinds of compound categories: Compound Nouns, Compound Verbs and Compound Adjectives, with examples for three kinds of the processes involved within these. It must be added that there are many more categories of compounding than those tabulated below (Rajendran, 2005).

 

 

 

 

 

 

 

 

 

Examples

Compound Nouns
No. Noun+ Noun Verb+ Noun Adjective +Noun
1 /t̯irai/ ‘curtain’+
/paDam/ ‘picture’
→/t̯iraippaDam/ ‘movie’
/kuDi/ ‘drink’ + /n̯i:ɾ/ ‘water’

→ /kuDini:ɾ/ ‘drinkinng water’

/činna/ ‘small’+/ʋi:Də/ ‘house’

→/činnaʋi:Də/’small house’

2 /pakal/ ‘day’ + /kanaʋə/ ‘dream’

→ /pakalkanaʋə/ ‘daydream’

/t̯u:ngə/ ‘sleep’+ /mu:nǰi/ ‘face’

→/t̯u:ngumu:nǰi/ ‘lethargic person’

/kiLi/ ‘parrot’+ /paččai/ ‘green’

→/kiLippaččai/ ‘parrot green’

 

3 /maram/ ‘wood’+ /maNDai/’head’

→ /maramaNDai/ ‘fool’

/t̯a:li/ ‘wedding garland’+ /keTTə/ ‘tie’

→ /t̯a:likeTTə/ ‘wedding’

/konǰam/ ‘some’+ /ne:ram/ ‘time’

→/konǰane:ram/ ‘sometime’

 

 

 

 

 

 

 

 

 

 

Examples

Compound Verbs
No. Noun+ Verb Verb+ Verb
1 /po:ɾ/ ‘war’ + /a:Də/ ‘perform’

→/po:ɾa:Də/ ‘struggle’

/čolli/ ‘having said’ + /koDə/ ‘give’

→ /čollikkiDə/ ‘teach’

 

2 /ko:bam/ ‘anger’+ /paTTə/ ‘experience’

→ /ko:bappaTTə/ ‘in anger’

/kanDə/ ‘having seen’ + /piDi/ ‘catch’

→ /kanDupiDi/ ‘find out’

 

3 /kuɻal/ ‘cylinder’+ /u:tt̯ə/ ‘blow’

→/kuɻalu:tt̯ə/ ‘play flute’

/o:Di/ ‘run’+ /po:/ ‘go’

→/o:Dippo:/ ‘flee’

 

 

 

 

 

Examples

Compound Adjectives
No. Adjective+ Adjective Verb + Noun
1 /paɻam/ ‘old’ + /pe:ɾum/ ‘large’

→ /paɻampe:ɾum/ ‘seasoned’

/aɻu/ ‘cry’+ /mu:njd͡ʒi/ ‘face’

→/aɻumu:njd͡ʒi/ ‘sulky’

 

 

2  

/iLam/ ‘light’+ /čivappə/ ‘red’

→/iLančivappə/ ‘light red’

 

/t̯iruTTə/ ‘mischief’ + /mu:njd͡ʒi/ ‘face’

→/t̯iruTTumu:nd͡ʒi/ ‘mischievous’

 

3 /kaɾu/ ‘black’+ /ni:lam/ ‘blue’

→/kaɾuni:lam/ ‘blackish blue’

 

/aruL/ ‘mercy’ + /mika/ ‘increase’

→ /aruLmika/ ‘merciful’

 


Chapter 7: Adaptation of Borrowed English Words in Tamil
 

As mentioned earlier in this study, over centuries many English words have entered the Tamil lexicon. When these words entered the language, its own phonological rules have been imposed on the words. In a process of adaptive borrowing, phonemes of the original word might be retained, modified or even deleted. However, deletion is never witnessed in Tamil and we see extensive adaptation even when there are no corresponding phonemes. This chapter looks at the adaptation of English intervocalic plosives (between two vowels), fricatives and the two diphthongs /ei/ , /eə/ and /ɔi/.

Intervocalic Plosive Adaptation

English Phoneme English Tamil
/k/ /ə’kaunt/ /’akkauNT/
/g/ /ə’geinst/ /’agge:nsT/
/t/ /ə’tein/ /’aTTe:n/
/d/ /’bidiŋ/ /’piDDiŋ/
/p/ /ə’pauz/ /’appo:s/
/b/ /’rʌbər/ /’ɾabbaɾ/

 From this data, it is possible to arrive at the conclusion that intervocalic plosives in Tamil can be geminated in Tamil. English alveolar plosives do not have counterparts, and are represented as retroflex /T/ and /D/. Since voicing is a prevalent feature in spoken Tamil allophonic variations, there is no issue in uttering the voiced plosives.

 

Fricatives

English Phoneme English Tamil Manifestations
/f/ /fi:z/ /pi:s/ or /fi:s/ /p/ or /f/
/v/ /væn/ /ʋan/ /ʋ/
/θ/ /θi:f/ / t̯i:f/ / t̯/
/ð/ / ðei/ / d̯e:/ / d̯/
/z/ /zu:/ / ǰu:/ /ǰ/
/ʃ/ / ʃaut/ / ʂauT/ / ʂ/
/ʒ/ / pleʒə(r)/ /pleʂaɾ/ or /plesaɾ/ / ʂ/ or /s/

 

Tamil only has four as opposed to the nine fricatives of English, which were themselves drawn from Sanskritic roots. Therefore, while /f/, /s/ and /h/ map directly, other fricatives will be adapted at borrowing as demonstrated. Since /f/ is a modern addition to the phonology of Tamil, there are situations where it is mapped instead to /p/. Since they share a similar place of articulation with /θ/ and /ð/, /t̯/ and /d̯/ are used as replacements. (Yang, 2011). Labiodental fricative /v/ is adapted as a labiodental approximant and the paletoalveolar fricative / ʃ/ is changed to a retroflex fricative. The voiced /z/ and /ʒ/, due to an absence of corresponding voiced phonemes, are changed to voiceless / ʂ/ or /s/. In sum, where parallel fricatives do not exist in Tamil, voiceless phonemes closest to the place of articulation and in concordance with the general manner of articulation in Tamil are used.

 

Diphthongs

Diphthong English Tamil Manifestation
/ei/ /steipəl/ /sTe:ppiL/ /e:/
/peint/ /pejinT/ /eji/
/əveil/ /aveejl/ /e:j/
/weit/ /ʋe:jiT/ /e:ji/
/eə/ /eər/ /e: ɾ/ /e:/
/eə.ri.ə/ /eɾija:/ /e/
/ skweər/ /skoja/ /oja/
/ɔi/ /ɔil/ /a:jil/ /a:ji/

 

All diphthongs except /ai/ and /au/ which are present in Tamil are altered. We have looked at three such diphthongs. Like in Hindi, some environments change /ei/ into the long /e:/, though we notice here, the abundant use of glides (the semi-vowel /j/) in the adaptation process of the vowel /i/, a semi-vowel bridging the gap between two vowels which do not exist as diphthongs in the language. Adaptation is also influenced by the presence of /w/ in the onset, which becomes /o/, for example in the word ‘square’. Since Tamil does not possess the open mid-back rounded vowel /ɔ/, it replaces this with the open mid-back unrounded vowel /a:/.

The studying of adapted borrowing has immense implications for linguistics. Various viewpoints exist on loanword adaptations in current linguistic studies. For example, some believe that perception plays an important role, that is when a Tamil speaker hears the word ‘oil’ they actually perceive it as /a:jil/. The theory of ‘production’ on the other hand states that it is done by bilinguals who possess the knowledge of both phonologies and are trying to bridge the gaps. The third view combines both these ideas. Whichever the viewpoint, the analysis of loanwords can give us insight into the productivity of the existing phonological rules of the language. (Yang, 2011)

Chapter 8: Conclusion 

Tamil is an ancient language, with a rich history and literature. It is said to have come from a Proto-Dravidian source, along with the other Dravidian languages. It has undergone several transitions, and is deeply influenced by other language systems.

The Tamil diaspora is spread across vast geographic regions, which has enabled the language to develop and branch into several dialects. It also exhibits a high degree of diaglossia, meaning there is a sharp distinction between the pure, ‘high’ variety, and the colloquial, ‘low’ variety. In this project, we have examined a kind of standardized variety, an amalgam of the different varieties that is popularized in common parlance and media.

We began by examining the very sounds that make up the language – the phonemes. Tamil has 34 phonemes. The language lacks aspirated sounds, but is rich in retroflexes, rhotics and nasals. Phonemically, the voicing of plosives are not distinguished. Most vowels are either long or short, along with two dipthongs. The lack of fricatives in the language is made up for by the addition of letters from grantha, initially used to write Sanskrit loanwords. Another feature of the language is gemination of certain consonants.

With this knowledge of phonemes, we identified 20 sets of minimal pairs, of both vowels and consonants, which helped us understand the contrastive nature of phonemes.

Realizations of the same phoneme in speech are allophones. Tamil has well documented allophonic variations in the voicing of plosives. It was found that word initially or geminated, plosives are unvoiced, but when occurring intervocalically or post-nasally, they are voiced. We demonstrated this with the example of /t̪/. another allophone that was explored was of the vowel /u/, which provides a unique sound /ə/ when occurring word finally.

We also explored free variation, with respect to Brahminic Tamil, which makes a conscious effort not to adapt, and to cling on to their language this Tamil is close to the ‘high’ variety of Tamil. In the attempt to retain their language, the retroflex rhotic is more prominent, whereas in non-Brahmin Tamil, its usage is fading into /l/ or /L/.

We then moved to syllable structure. We demonstrated how Tamil prefers simpler syllable structures, allowing no consonant clusters in the Onset, and a maximum of two in the Coda (theoretically). Five syllable structures were proposed: V, CV, VC, CVC and CVCC (which does not occur in practice). We then gave examples for each of these possible syllable structures, and for syllabification of words.

In morphology, we described the agglutinating nature of the Tamil language. It employs multiple word formation processes. While prefixation is not a productive process, suffixation is highly productive. Inflectional markers include case, number for nouns and a unique blend of tense, person, gender and number in verbs. Some examples of derivational markers were also shown.

In addition, we focused on compounding, another highly productive process in Tamil, and focused on specific kinds of compounding.

In the chapter on borrowing of English words. In borrowing, Tamil has imposed its phonological rules onto these words. Tamil and English have vastly different phonemes, English consonants take on the closest one in Tamil based on manner and place of articulation. The process of adaptation by deleting of phonemes is absent. We focused on Intervocalic plosive adaptation (intervocalic plosives in Tamil are usually voiced, this is not necessarily the case in loanwords), fricatives (Tamil’s limited number of fricatives themselves are borrowed) and diphthongs /ei/, /eə/ and /ɔi/. In loanwords, it was found that voiced plosives may be geminated, and many were converted to the Tamil retroflex/dental consonants. Diphthong alteration occurs much like in Hindi. There is also increased use of /j/, and unique adaptation of /w/ to /oi/.

In studying borrowing and loanwords, we learnt about the immense productivity of a process such as this, not just for Tamil, but all other language. It speaks to the openness of languages to borrow and adapt and has incredible implications for linguistics. It is not only English from which words are borrowed, today Tamil is greatly influenced by Sanskrit, Hindi, and neighbouring languages like Malayalam. It is more the spoken Tamil which is adaptive. The written, ‘pure’ form gets further and further separated from the spoken, colloquial due to its resistance to change. This will someday cause a break between the two.

In this project, we have only briefly covered topics such as inflections and derivations, word formation processes, free variation and sociolinguistics. There is immense scope in covering morphology – Tamil theoretically has thousands of possible inflectional combinations with respect to both case markers and tense-person combination. We have also not gone into irregularities and exceptions to these general rules, suppletives, or even allomorphs and undertaken an analysis of these. Properties of the language such as gemination of certain consonants have been mentioned, but not described in detail in terms of morpho-phonemic rules. Sociolinguistics is yet another rich area, and the sociolects and dialects are in themselves complete language systems. We have only briefly touched upon Brahmin Tamil as a caste-based dialect, which is in itself not a uniform category.

It is also important to note that since the written variety and the spoken variety are so different, we have utilised the written in some sections to show the processes in action. Tamil also has a high degree of elision in speech, words are merged slightly together. A detailed study of spoken Tamil has not been conducted.

In this project, we have thus arrived at a comprehensive morphological, phonological understanding on the Tamil language. It has enabled us to use our knowledge of linguistics and apply it, and take a more critical as well as conscious approach to the use of the language.

References

Asher, R. E., & Annamalai, E. (2002). Colloquial Tamil: The complete course for beginners (The Colloquial Series). London: Routledge.

Bright, W. (1976). Social dialect and language history. In Variation and change in language: Essays. Stanford: Stanford University Press.

Fasold, R., & Connor-Linton, J. (Eds.). (2006). An introduction to language and linguistics. New York, NY: Cambridge University Press.

Genetti, C. (Ed.). (2014). How languages work: An introduction to language and linguistics. New York, NY: Cambridge University Press.

Immanuel, X. (2008, 9 12). Syllable Structure in Tamil:A Case Study of the Eastern Perspective.  Retrieved 03 29, 2016, from Tamil Linguistics: https://xavieremmanuel.org/syllable-structure-of-tamil-dialects-ma-dissertation/

Keane, E. (2004), “Tamil”, Journal of the International Phonetic Association 34 (1): 111–116, doi:10.1017/S0025100304001549

Krishnamurti, B. (2003). The Dravidian Languages. United Kingdom: The Press Syndicate of the University of Cambridge.

Rajendran, S. (2005, 10 10). A Comprehensive Study of Word Formation in Tamil. Retrieved 4 15, 2016, from Language in India: http://www.languageinindia.com/oct2005/wordformationrajendranbook2.html#chapter3

Ramaswami, N. (2001). Lexical formatives and word formation rules in Tamil. Language in India, 1. Retrieved April 20, 2016, from http://www.languageinindia.com/index.html

Savaranan, B. (2009, 9 28). Monosyllables: From Phonology To Typology. Retrieved 3 28, 2016, from Festival of Lanuages: http://www.fb10.uni-bremen.de/monosyllables/abstracts.aspx

Schiffman, H. F. (1998). Standardization or restandardization: The case for “standard” spoken Tamil. Language in Society27(03), 359-385.

Schiffman, H. F. (1999). A reference grammar of spoken Tamil. New York, NY: Cambridge University Press.

Schiffman, H. F. (2004). The Tamil case system. South Indian horizons: felicitation volume for Francois Gros on the occasion of his 70th birthday, 293-322.

Steever, S. B. (2015). The Dravidian Languages. Routledge.

Tamil. (n.d.). Retrieved April 16, 2016, from http://www.ancientscripts.com/tamil.html

Tamil. (2009). Retrieved 4 10, 2016, from Ethnologue: http://archive.ethnologue.com/16/show_language.asp?code=tam

Thangarajan, R. (2012). Speech Recognition for Agglutinative Languages. In S. Ramakrishnan (Ed.), Modern Speech Recognition Approaches with Case Studies. doi:10.5772/50140

Yang, N. H. (2011). Loanword Adaptation in Tamil. Department of English Language and iterature, National University Singapore .

Yule, G. (2010). The study of language (4th ed.). New York, NY: Cambridge University Press.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s