Naming in the Kashmiri Pandit Community

Socio-linguistics and Anthroponymy

Braj B. Kachru
University of Illinois, Urbana, Illinois

In earlier research on onomastics the concentration has been on two types of studies: One type focuses on the data on naming processes with reference to a well-defined social, regional, or religious group. The other type is essentially comparative, and has implications for defining and illustrating concepts such as "linguistic area" or "socio-linguistic area". There is a profusion of studies of the first type, but rather few on the second type. An example of a comparative study with typological and socio-linguistic interest is Emeneau 1978.

 In this preliminary investigation in anthroponomy of a small community of India, generally known as Kashmiri pandits and locally known as bati (Skt. bhartn), I have attempted to combine these two approaches. The pandit community forms an extremely small minority in India, not exceeding over one hundred thousand people. Of this total, a significant number live in Kashmir valley, mainly in Srinagar, and the rest have settled in the major cities of Allahabad, Agra, Delhi, Lucknow and Jaipur in northern India.

 During the last seven hundred years, this minority community has passed through a series of social, political, and cultural vicissitudes (1). Traces of these changes are naturally present in their language and in their naming system. The language of Kashmiri pandits is called kosur (or Kashmiri). This language has traditionally been treated as a member of the linguistic sub-group termed "the Dardic family", which is within the Indo-Aryan branch of languages. The Dardic family has been identified on the basis of several phonological, syntactic, and lexical characteristics of three language groups: the Kafin-group, the Khowar-group, and the Dard-group(2).

 However, the classification of Kashmiri continues to be controversial: There still is a need to draw evidence from those areas of language and culture which might throw more light on the controversy concerning the affiliation of Kashmiri language, and the types of convergence it has gone through. A study of names should provide evidence to show that Kashmiri forms a part of South Asian socio-cultural and linguistic area. In a stimulating paper "Toward an onomastics of South Asia" Emeneau (1978) has rightly suggested that, in the case of South Asia, onomastics provides interesting ethnographic, cultural, and historical evidence. In his paper Emeneau examines ". . . features of the naming system and their spread in the South Asian linguistic area" (128). There is evidence that the patterning of the naming system, the choices available within the system, and the functional use of such choices provide insightful data for characterizing South Asia as a "socio-linguistic area". I shall not go into any detail in explaining here the concept "socio-linguistic area" since this forms the topic of a study by D'sonza (1985).(3).

 What I propose to do is to discuss selected processes of naming among the pandits of Kashmir within a socio-linguistic context. These observations may hopefully, provide more data from an interesting linguistic and cultural area of India for our understanding of India as a "socio-linguistic area".

 The first issue one faces in a study of onomastics is to determine the proper approach. The body of literature on onomastics is awe-inspiring, with a long tradition of research from various disciplines: among others, anthropology, (e.g., EvansPritchard 1948), etymology (e. g., Bradsley 1915; see also for a bibliography Smith 1950, 277-282), linguistics, (e.g., Algeo 1973; Uhlenback 1969) and philosophy (e.g., Donnellan 1970). There are also several studies which provide useful theoretical discussions from various viewpoints (e.g., Gardiner 1940, Pulgram 1954, and S0renson 1963). In addition, there are numerous theoretical studies. However, the sifting through of this abundant body of material is not always rewarding. Zgusta (1974: 819) rightly warns us that "the study of names is plagued with imaginative and fantastic, but unfounded and unscholarly, publications; therefore, caution should be exercised in the choice of sources."

 In earlier literature very little attention has been paid to the linguistic aspects of onomastics and its contribution to linguistic and socio-linguistic theory and description, since linguists have considered onomastics "etymologically explanatory rather than systematically descriptive" (Allen 1968; quoted in Uhlenbeck 1969: 321). There are very that the synchronic study of proper names may be an interesting and rewarding subject for the linguist as well as for the ethnographer and that at least in some linguistic communities personal names may form a well-defined subsystem within the whole of the linguistic structure" (321-322).

 In his detailed study Utley (1963: 145-176) draws attention to the linguistic component of onomastics and the neglect of this field by linguists. He observes that ". . . the responsibility for correction lies not only on the historian, logician, and etymologist, but also on the modern linguist, structural or transformational, who has been slow to plow on onomastic pastures" (145).

 The Indian languages are no exception to this general neglect by linguists.4 The result is that the linguistic aspects of onomastics in these languages have not attracted serious attention of linguists, and no attempt has been made to provide typological statements which contribute to our understanding of India as a linguistic area or a socio-linguistic area (Emeneau 1978 is an exception). There are, however, several language-specific studies which provide useful data and descriptions (e.g., Dutta 1981; Jhungare 1975; Sjoberg 1968; and Temple 1883). There are not many observations in earlier literature on Kashmiri or Kashmiri paodits, except in Koul (1924: 18-23). This study, therefore, is one of the first attempts toward understanding anthroponymy of Kashmiri paudits in relation to socio-linguistics.

Naming and Gotra
It is claimed that traditionally Kashmiri pacdit last names indicated gotra (exogamous groups); these gotra-marking names were initially restricted to three names which Koul (1924: 20) terms the "three principal divisions", i. e., bhat, pandit and razdan. According to him, from these three "are derived the distinctive appellations of koul, sopor), paudit, and raina. >From these three families, [bhat, papdit and razdan] as each took to a particular occupation, or by adoption or intermarriage with other gotras, other gotras came into existence" (20). These gotras bear the names of rrszs ("sages"), and to some extent, as Koul says, they form a hierarchy. In this hierarchy the kul (Kaul or Koul) are considered very high (Koul 1924: 20). It is claimed that by intermarriage and intermixture "with other Brahmins the number of gotras multiplied to 199" (Koul 1924: 20). As time passed, the number of last names substantially increased. I will discuss the reasons for this later in this paper.

Naming - Ceremony and Nameability
Criteria The ritualistic aspects of naming among the pa4dits are not significantly different from the brahman groups of other parts of India. In a traditional paudit family the naming-ceremony (Hind), namkaran) takes place on the eleventh day after a child's birth. It is called kahnethir (kah "eleven" and nethir "wedding"). At this ceremony a child is expected to receive his or her given name.

 The names of the new-born may be chosen by various methods. In certain families the family priest pandith (Skt. pandi t) or gor (Skt. guru) might suggest the first letter of the name. The suggestion of the koligor (Skt. kulguru "family priest") is based on clues from the horoscope of the new-born child. Suggestions may be made by relatives from the mother's side or the father's side. Naming after parents, or grandparents, as is common in the Western world, is not practiced among the Kashmiri Pandits. The earlier practices of namkaran ('naming-ceremony') are, however, slowly changing as we will see in the following section.

The two productive processes for naming are Sanskritic and Persian. Muslim first names do not show the process of Sanskritization. However, a number of Sanskrit last names (Kashmir) zath) have been preserved by some Muslims after their conversion to Islam, for example, bath (Skt. bhatta), pandith (Skt. pand t).5 On the other hand, Hindu given names and last names show the influence of both Sanskritization and of Persianization. The result is, as Emeneau (1978:117) also attests in another context, "Hindu names of hybrid form" such as jawahar ("jewel"). In Kashmir, this hybridization of names still continues.

Dynamics of Kashmiri Pandit Names
The dynamics of the pandit names show mainly three types of changes. Generation-distance is the main factor in the changing pattern of the given names. As in other language areas, the changing social, political, and linguistic influences on the land and the people of Kashmir have left their impact on naming patterns. Consider, for example, given masculine names such as aftab 'tine sun', and toti 'parrot' (as in totikakh), or feminine names such as goni 'virtue', (as in gonivati ('the possessor of virtues'). These given names are now only of antiquarian interest and form part of the diachronic onomastics of the pandit community. These names suggest a distance of more than one generation. Names such as makhni (Hindi-Urdu, makEhan 'butter') and mah (Hindi-Urdu, mot) 'jewel') are much less frequent in the present generation.

Components of a Pandit Name

The naming system of the pandits follows the pauern of other Indo-Aryan languages. A pandit name has three components: The first name or given name, the middle name, and the last name (zath) or 'family' name. The first and last items are obligatory and the middle item is optional.6 (For example, madisodan (Skt. madhusudan ("the killer of the demon Madhu"), roginath (Skt. raghunath "master of Raghu dynasty"); omkar (Skt. omkar).

The given name has the structure of a noun phrase in which the family name occupies the position of head, unlike names in the Dravidian languages in general. In Telugu, as Sjoberg (1968: 314) shows the Family names appear to stand in an adjectival relationship to given names. That is, as an adjective in the Telugu language precedes the noun it qualifies, so too the family name precedes the given name, which is clearly a noun.

The types of compounds and modifiers which form a given name again fall into a typical Indo-Aryan pattern as does the constituent structure of names.

The lexical sets which occupy the modifier and head positions may be characterized in terms of semantic sets. These sets provide attitudinal, locational, and other clues. I shall discuss some of these sets below.
 

Given Name and the Kram

The given names are chosen out of a variety of lexical sets, the most common being theophoric names, or names which refer to various gods of the Hindu scriptures and mythology. As mentioned earlier, a new-born child is not necessarily given a name at its birth. A name may be given to the child at his kahnethir "naming ceremony" or at a convenient later time. In the meantime, a child may be called niki (fem. nich) 'baby' until it is given a name. In certain cases nik' (nic) may continue to be used as an alternate name, both for reference and address.

The term kram (a class name) is traditionally used to make a distinction between the nick-names and traditional last names or zath (e. g., Iast names such as Koul, Bhatt). The kram slowly gets institutionalized as the last name. The last name, therefore, need not be an indicator of gotra, but might develop out of the sobriquets (nick-names) acquired by the family for various reasons. Sobriquets are acquired in numerous ways: occupation, color, physical characteristics and so on.
 

Toward Structuring Kram

The kram or 'nicknames' acquired by each family may be listed in terms of lexical sets of sobriquets. Sobriquets were originally aptly called in English ekenames (an 'alsiname'; see Smith 19SO: 7492). These sets share semantic characteristics and provide semantic classes for the zath of Kashmi ris. The following kram-denoting lexical sets are illustrative.

1. Attitude-marking: The use of attitude-marking terms is a good example of making a generalization from an individual to the whole family. Such terms must have originally started as nicknames, and, then, slowly acquired the status of kram. There are, therefore, cases among Kashmiri Pandits where two brothers might use two distinct last names, one using the kram and the other the original last name. For example, one brother may use kol (Koul), and the other brother and his children use tut (Anglicized as Tutu). Kol is gotra-marking ancestral last name, and tut is the acquired kram-marker.

The attitude-marking terms have an underlying meaning which conveys an attitude and focuses on characteristics of certain types. Such terms can be subdivided into a number of classes on the basis of the shared characteristics of the members of a class of last names. Consider, for example, the following sub-classes.

• Animals: bror' 'cats'; gagar, 'mice'; hapath, 'bears'; host', 'elephants'; hagal, 'stage'; khar, 'donkeys'; ponz', 'monkeys'.
• Birds: bulbul, 'nightengale'; kav,'crows'; kokar,'roosters'; kotar, 'pigeons'
• Edibles: vagri, 'watery rice'
• Fruits: badam, 'almonds'
• Spices: ganhar,'poppy seeds'
• Utensils: dul', 'a huge brass or earthen pot'; notivj[, 'owners of pitchers'; tjk', 'earthen eating plates'; vokhal, 'mortar'
• Vegetables: hakh, 'collard leaves'; vagan 'egg-plants'; mu', 'radish'

3. Deformity: There is a number of words which show bodily deformity, for example, karihol, 'one who has a stiff neck'; khosh, 'left handed'; kob', 'hunchbacked'; mot', 'fat'; mak', 'snub-nosed'.

4. Location: The locational words are used either to specify one's place of residence or the place where one comes from, e. g., navs-ohoer', 'one who belongs to Navsahar' sogom', 'one who belongs to Sogam' sopor', 'one who belongs to Sopore'; k'alam ione who belongs to Kilam'.

5. Occupation: The use of occupation or profession-marking terms does not necessarily mean that the family is engaged in that particular occupation. One might mention here that last names such as vatal 'sweepers' or vaza 'cooks' semantically belong to this class, but do not always refer to the actual profession of the family. In this case the term vatal must have begun as an attitude marking kram. Professionally there is no Kashmiri pandit engaged in the occupation of a vatul 'sweeper'. These are, thus, terms which were originally used as a inick name' based on an act of a single member of a family and then slowly these got generalized as the last names of the family, for example, jotish (Hindi jyotisi 'astrologers'), sabin"soap makers'.

6. Physical characteristics: These mark the physical characteristics of a person with reference to his or her color, bodily features, etc., for example, katsur 'brown'; tut 'slim'.

7. Ownership: There is a small set of lexical items which take the suffix vol (plural vol'; Hindu-Urdu vala) to convey the idea of ownership, for example, hos'vol"the owners of elephants', tarivol' 'dealers in wire'.

 Note that the suffix vol' has other meanings in Kashmiri, for example, it is used in the sense of 'a dealer' too. Thus tarivol' may also mean a 'dealer in w*e' or 'seller of wire' as in Hindi cayvala, 'a dealer in tea' or 'a seller of tea'.

 The kram (or a nickname), until it gets established for a family, can be exasperating. Koul (1924: 20) has mentioned one such case

. . . a man, named Wa'sdev, had a mulberry tree growing in his courtyard and, therefore, he was called Wa'sdev Tul (mulberry). He, in order to get rid of his nick-name, cut down the tree. But a mund (trunk) remained and people began to call him Wa'sdev Mund. He then removed the trunk of the tree but by its removal a khud (depression) was caused and henceforth people called him Wa'sdev Khud. He then filled up the depression and the ground became teng a little elevated) and he began to be called Wa'sdev Teng. Thus exasperated, he left to do any further attempt to remove the cause of his nick-name and it continued to be Teng which is now attached to the names of his descendants.

Variant Forms of a Name: Hierarchy of Hypocoristic

The hypocoristic forms are determined on a deferential scale based on considerations of age, superiority, sex and the attitude one has toward the addressee. The term hypoconsticiorm refers to a variant form of a name which is used in intimate relations and familiar or friendly situations. A discussion of this aspect throws some light on the socio-linguistic aspects of Kashmiri pandit onomastics. This characteristic is shared with other Indo-Aryan and Dravidian languages (see, e.g., Emeneau 1978:118-124), and provides further evidence for India as a socio-linguistic area.

 Hypocoristic forms provide linguistic clues to the participant relationship and participant expectations, and mark the discourse as formal, informal, or intimate. These forms may thus be used to indicate the formal use of the name especially in the written style; to mark informal relationships; to express a difference in age and deference toward age; and to mark the sex and the marital status of a person.

 The variant forms are produced through various processes. One common device is by using the process of Kashmirization of Indic and Perso-Arabic proper names. The process is as follows: In a majority of cases the first names is reduced to the CVCV structure irrespective of the syllable structure of the first name in the written or formal style. Consider the following examples :
 

An identical process of reduction applies to Muslim first names such as gulam> gula (m); habib > hate' (m); phat'> fatima (f). In a majority Naming in the Kashmiri Pandit Community of cases, then, in a reduced name the final vowel is /i/. In the informal style a person may be addressed as dami, but the written form is damodar and sila, respectively.
 

Name Repertoire and Name Switching

There may be repertoire of names for a person used within a family; the selection of a particular name is determined by the relationship, status and age. The total range of the names used for a person is generally known only to those who are close to the family. First there is a formal given name. The formal name invariably preserves the Sanskrit, Perso-Arabic, or hybridized forms (e. g., pztambarnath, aftab, Abalnarayan). Second, in the informal situation a hypocoristic form is used by the peer group and elders. Third, in the intimate family circles a person might be given an entirely different name. Let me illustrate this point. A person may be formally called mohanlal (m), and informally, he may be called mohn`; his children may call him tathimahraj 'dear king'. He may also have a kinship denoting name such as boytoth ('d ear brother'). However, it is not necessary that this name be used only by those with whom he has such a relationship. The intimate family may thus use tathimahraj and boy.toth for intimacy and affection. A feminine name szla informally becomes szl' (f) and in the family circles she might be called benitath ('dear sister'), or benigash ('sister with light').

 It is also possible that after marriage a female might receive an additional name from her in-laws, such as p'anbabr (Hindu-Urdu pyarzbhabhz 'dear sister-in-law'), or mohanptarz, 'one who is d ear to Mohan' (if, for example, the name of her husband is Mohan). A son-in-law may also go through the same process of multi-identification markers.
 

Sources of First Names

The earliest source of Kashmiri first names is Sanskritic. Later after the Muslim conquest of Kashmir (around 1301-20), the Perso-Arabic influence provided the second major source. Before the Islamic influence, Kashmir had been the center of Sanskrit learning and had developed a long tradition of Sanskrit scholarship in the fields of philosophy, poetics and linguistics. During the Muslim period there was a major effort toward proselytization of Hindus to Islam. A small number of the pandits escaped by leaving the boundaries of the State, and others assimilated themselves in various ways with the new rulers. The Islamic influence is found naturally in the first names, though a number of last names may also be traced to this influence. It is evident that both the Sanskritic and Persian names have gone through various processes of nativization which are not discussed in this paper.

 The lexical sets for first names from the Sanskritic source are theophoric, primarily based on names, attributes and epithets of gods and goddesses (e. g. ramchand, radhenath, dayakisin, arjunnath, laksminarayan).

It is only later, due to the Persian irdluence, that first names such as aftab "the sun" were introduced. Consider, for example, aftab 'tine sun'; dilavaribrave'; iAbal 'good fortune'; mahtab 'the moon'. Smith's observation (1950:157) on the Hindu naming system applies to the Kashmiri Pandit community, too:

 In India the high caste Hindu believes that the more often the name of a deity passes his lips the more merit he stores up for the future life. Consequently he deliberately names his children after the gods. The Hindu Pantheon is crowded so the choice is not unduly restricted . . . The names of the dreaded spirits are not used for fear of inviting harm. If some of them bear human names, it is because they were probably originally human beings.

 The main categories of Sanskritic given names are listed below:

 (i) Epic-Sources: The Indian epics the Ramayana and the Mahabharata provide the main sources for names. The masculine names arjun, kisen, ram and the feminine names such as jankz, tara, rada can be traced to these sources.

(ii) Place-names: The first name may be an epithet for a character from the epic-sources, for example, ayodya or bade. These are followed by nath, e.g. ayodyanath, 'the master of Ayodhya' (where Rama, the hero of the epic Ramayana was loom), or brajnath 'tine master of braj bhumi' (where Krishna of the epic Mahabharata lived). These temms, then, stand as an attribute of the epic-heroes.

(iii) Abstract: The use of abstract terms as proper names is again shared with several other Indian languages. atma, 'soul'; anand, 'bliss'; parmanand,'etemal bliss'.

(iv) Attributes: Feminine names such as the following refer to various attributes and qualities, rits (mal), 'good', pious'; soba (vatz), 'grace'; dan (yatz), 'wealth'; usha, 'dawn'; lalita, 'attractive'. Masculine attributes are prem, 'love'; pran, 'soul'.

(v) Rivers: The given names of rivers are mainly feminine yam~na (Hind) yamuna 'river Yamuna'); sarisvati (Hind) 'sarasvati' 'a mythical river').

 A number of compounds with the first member referring to a goddess and the second to a god are used as masculine given names radhakishen (Radha and Krishna) (cf. sitaram (Site and Ram) used e. g., also in the Punjab).

 The following sers illustrate the range of the first name (FN) and its preferred second lexical item. Note that there are some given names which are shared by both men and women, e.g., mohan, janki and lakshmi. In the case of men, mohan will collocate with lal, and janki and lakshmi with nath. In the case of women mohan collocates with ran' end p'ar' and janki and laksmi with dev`. An elderly woman may be called jankid'ad (dyed 'granny').

Masculine First Names:

FN + cand: e. g., tard, ram
FN + kifen: e. g., teal, braj, daya, gopi
FN + lal: e. g., mohan, makhan, mot,-, jia, syam
FN + kamar: e. g., braj, raj
FN + narayan: e. g., chand, raj, rup, siv
FN+nath:e.g., amar, bhaskar, kashi, pran, janh, dina, niranjan
FN+ram: e.g., siv
FN + joo: e. g., anand

Feminine First Names:

 FN + vati: e. g., tara, soba, dan
FN + mal: e. g., ritsi
FN+kuman:e.g., raj, sita
FN+p'an:e.g., mohan
FN + rani: e. g., cand

The given names carry the number and gender markers and are inflected for the case. A number of these names are now disappearing in contemporary practices of naming. One does not normally encounter, for example, ram or teal as first names and jo and vah as suffixes in current Kashmiri pandit names. One also notices a tendency to drop the middle name.

Modification of First Names

First names in Kashmiri may be modified with venous lexical itans or auffixes. I will present a partial list of such modifiers with their distribution. The selection of a particular modifier with a given name is determined by various attitudinal considerations, as is the case with the selection of hypocoristic forms of a first name. The main considerations are the addressor/addresse relationship; the attitude of the addressor toward the addressee at the time the speech act takes place; the clues marking affection, anger, or indifference; the age and status of the addressor and addressee and their respective positions in the family and society.

 (a) kakh is used only with hypocoristic forms such as dayi (Skt. daya 'compassion'), hard (Skt. han), sarvi (Skt. sarva 'whole, entire'), sombi (Skt. sambhu 'auspicious' or Lord Siva), hradi (Skt. hriday 'heart'), treyi (Skt. triloh 'of three worlds' name of Siva), zani (Skt. janardan 'protector').

 (b) toth (fem. tith) means 'dear' and is used with a set of reduced names (e. g. sombitoth; brijitoth). It is not used with a formal names such as daya, hriday or triloki.

 (c) Ji (Hindi-UrduJ7) has roughly the same distribution in Kashmiri as in, for example, Hindi and Punjabi. It occurs with most hypocoristic forms and full forms.

 (d) sijb is a marker of affection and co-occurs with various lexical sets. Its preferred occurrence is with hypocoristic forms (e.g., brijisjb). Note also lali 'grandfather or father' or lalisab.

Collocability Constraints

The term 'collocation' is used here to refer to the tendency of cooccurrence of lexical items with other lexical items (for a discussion see Halliday 1966 and Kachru 1983). The elements which modify a Kashmiri given name in the hypocoristic or nonhypocoristic forms have constraints of collocability. Therefore the distribution of the items with reference to their potential for cooccurrence is important. Consider, for example, the modifying items, kakh, toth. The honorific suffix ,n may be used with the first teal, hari, upni but not normally with dayay`, hridayJu The co-occurrence of. i, with feminine given names is more frequent. The item toth ('dear') collocates with a set of first names, which, among others, may include dayi, hyadi, treyi. But the following do not normally collocate with toth.

 *haritoth, *upnitoth, *bab.toth

ji and toth have collocational constraints with certain types of given names. However, there are other items which do not have cooccur rence restrictions but the collocation may have underlying attitudinal implications; such items may be used in special contexts with a specific meaning in mind. The collocation of the item kakh with the following may be considered normal: briji kakh, dayikakh, hradikakh, sarvikakh, sombikakh, zanikakh. On the other hand, if kakh is used, for example, with balikakh, harikokh, upnikakh an attitudinal implication is involved, one of irony. This implication is not present if the person is elderly and the suffix is used as a mark of respect.

 The table presented below shows the preferential range of co-occurrence with various modifiers.
 
 

The co-occurrence of the members of the above sets is not mutually exclusive. That is, for example, sarvi, s-ombi and zani cooccur with kakh and do not normally co-occur with J`. On the other hand, sombi and zani may in certain situations be acceptable with sob, but sarvi will not normally occur with sob. Thus all the members of a set do not share all the constraints of occurrence with other items.

 There is another point which should be mentioned here. The Kashmirized reduced forms of Hindi forms teal and upan are bali and upni respectively. As is indicated in the lexical set III in the above Table, these do not co-occur with Ji.

Conclusion
It seems to me that there are several reasons for the study of Kashmiri pandit anthroponymy within the context of socio-linguistics. Let me present some of these reasons here. First, it is the user's attitude which determines the formal choices within the possible lexical sets. Second, the hierarchy of hypocoristic forms is directly related to the interactional network and the participant relationship. Third, the selection of a first name, out of the possible sources, is also indicative of the attitudes of the parents and the family. Fourth, at the lexical level there are several collocability constraints which provide interesting lexical relationships: these are useful in understanding the dynamics of Kashmiri pandit anthroponymy.

 A socio-linguistic perspective for anthroponymy might provide a deeper understanding of the functional aspects of the Kashmiri language, and its shared socio-cultural traditions with the neighboring linguistic communities. Onomastics may, then, give us supporting data for characterizing and defining India as a "socio-linguistic area".
 

Notes

1 A useful discussion of religious pluralism in Kashmir is presented in Ma 1972.

 2 For a brief discussion of and references on Kashmiri language and literature, see, Kachru 1973: 2-4, and 1981: 4-6.

 3. This preliminary paper and her dissertation (forthcoming) provide the first detailed treatment of this topic with appropriate references from Indian languages.

 4. For socio-cultural and other viewpoints on the naming patterns in India see Indic Names 1961; Karve 1947; Masani 1966; Sharma 1969; and Van Velze 1938. These include references to whatever linguistic research has been done on this topic.

 5 A number of sobriquets are also shared, e. g., mandal 'buttocks'; vokil 'attorney'; vazi 'cook'.

 6 If there is a middle name, it is not the name of one's grandfather (from mother's or father's side) as may be the case in the U. S. A. Cf., e. g., "middle names are often family names of mother's father or of the grandfather's from either side" (Vanburen 1974: 85n).

 7 In addition to the points discussed in D'souza 1985.

 8 Among other forms, note also treyi, premi, which are the reduced forms of treloki, and prem, respectively.
 

References

• Algeo, John. 1973. On defining the proper names. Gainesville: University of Florida Press.
• Allen, W.S. 1958. Proper names in onomastics and linguistics. Actes et memoires du 5e Congres international de Toponymie et d'Anthroponymie Salamanca. 1958. 195-201.
• Bardsle`. Charles Wareing.1915. Our English Surnames: Their sources and significance. London: Chatto and Windus.
• Datta, Bhabataran. 1981. A linguistic study of personal names and surnames in Bengali. Calcutta University of Calcutta.
• D'souza, Jean Forward defining the term 'socio-linguistic area'. Manuscript. Department of Linguistics, University of Illinois at Urbana
• Donnellan. Keith S.1970. Proper names and identifying descriptions. Synthese 21.3-4. 335-358.
• Emeneau, M. B. 1978. Towards an onomastics of South Asia. Journal of the Amencan OnentalSociety 98.2. 113-130.
• Evans-Pritchard, E. E. 1948. Nuer modes of address. The Uganda Journal l2: 166171.
• Gardiner, A. H. 1940. The theory of proper names. London: Oxford university Press.
• Halliday, M.A K. 1966. Lexis as a linguistic level. In C.E. Bazell et al, ea.: In memory of J. R. hrth. London: Longman.
• Indic Names. 1961. Indic names, including proceedings of the seminar on the rendering of Indic names held at Calcutta, December30, 1960-Januaryl, 1961.
• Calcutta: IASLIC [Indian Association of special libraries and information centres].
• Jhunghare, Indira Y. 1975. Socio-psychological aspects and linguistic analysis of Marathi names. Names 23. 31-43.
• Kachru, Braj B. 1973. An Introduction to spoken Kashmiri. A basic course and reference manualfor learning and teaching Kashmiri as a second language. Urbana, IL: Department of Linguistics, University of Illinois.
• Kachru, Braj B. 1981. Kashmiri literature. Wiesbaden: Otto Harassowitz.
• Kachru, Braj B. 1983. The Indianization of English: The English language in India. New Delhi: Oxford University Press.
• Karve, Iravati. 1947. Personal names in India. In G.S.Ghurye, ea.: Papers in sociology. Bombay: Popular Prakashan.
• Koul, Anand. 1924. The Kashmiri pandit. Calcutta: Thacker, Spink & Co.
• Madan, T. N. 1972. Religions ideology in a plural society: The Muslims and Hindus of Kashmir. Contributions to Indian sociology. Vl. 106-141.
• Masani, R. P. 1966. Folk culture reflected in names. Bombay: Popular Prakashan.
• Pulgram, Ernst. 1954. Theory of names. Berkeley: American Name Society.
• Sharma, Mohan Lal. 1969. Origin and meaning of some Indian names. Names 17:3. 208-213.
• Sjoberg, Andree F. 1968. Telugu personal names: a structural analysis. In
• Bh.Krishnamurti, ea.: Studies in Indian linguistics (Professor M.B.Emeneau Sastipurti volume) Poona and Annamalainagar: Centres in Advanced Study in Linguistics. 313-321.
• Smith, Elsdon C. 1950. The story of our names. New York: Harper.
• Sorensen, H. S. 1963. The meaning of proper names. Copenhagen: G. E.C. Gad.
• Temple, R. C. 1883. A dissertation on the proper names of Panjabis, with special reference to the proper names of villagers in the Eastern Panjab. Bombay: Education Socretes Press.
• Uhlenbeck, E. M. 1969. Systematic features of Japanese personal names. Word. 25. 321-335.