Page 1
Aspects
 

of Ethnicity r anong the of Sri Lanka

Page 2


Page 3

Aspects of Ethnicity and Gender among the Rodi of Sri Lanka

Page 4

Aspects of Ethnicity and Gender among the Rodi of Sri Lanka
Nireka Weeratunge
International Centre for Ethnic Studies

Page 5
International Centre for Ethnic Studies 2 kynsey Terrace, Colombo 8, Sri Lanka.
First Published, 1988 Second Edition, 2002
Copyright (C) 1988, 2002 by ICES
Printed by Unie Arts (Pvt.) Ltd. No.48 B, Bloemendhal Road, Colombo 13.

III
III
IV
V
Contents
Introduction
Historical Sources on the Rodi
From Gadi to Rodi: The Transformation from Tribe to Caste
The Research Area
The Rodi Identity: An Ethnic Identity?
The Status of Women in the Rodi Community
Conclusion
Asibliography
O
27
32
87
32
139

Page 6
NO te The text of this book has been reprinted intact from the first edition of the study published by ICES in 1988. All time-related references in the text must be understood On that basis.

Introduction
The Rodi or Gadi people, the only untouchables in the Sinhala caste hierarchy, are a marginalised group gradually making its way into the Sinhala Buddhist mainstream. A language and a mythical tradition of their own, as well as the remnants of a hunting tradition, indicate a tribal origin in the distant past. Raghavan, who up to now has done the most extensive study of this group, enumerated their population in 1956 at 3, 122, distributed in 72 villages. The majority of these villages were located in the Central, North-Western and Southern Provinces. The Rodi were not counted separately in the official Census since 1903, as the stated government policy was to make no distinction between the Rodi and the Sinhalese, since the Rodi were considered a caste of the Sinhalese.'
There is no doubt that the Rodi today are a far larger group than Raghavan's figure for 1956 indicates. The population of the principal village of my study, exceeds the 1956 figure for the entire North-Western Province that this village is located in. Rodi families seem to be larger than the national average. However, we are still considering a minority group which is less than 1% of the Sri Lankan population.
This is not an ethnographic study of a tribal or smallscale society whose survival is at stake during our historical period. If the Rodi were a tribe in the past, they have been incorporated at the margin of the Sinhala caste system for several centuries by now. Therefore, my interest has been
l Raghavan, 1957, pp. 5-9 2 Raghavan, 1957, p. 5 3 Ibid., pp. 6,9

Page 7
centered on the Rodi from an ethnic perspective; i.e. - to examine the relationship of a small culturally distinct group to the dominant cultural group, of which it is, and is not, a part.
In this study, I am focussing first, on the process of how a tribe may have got transformed into a marginalised caste of untouchables. The why and how of this transformation is clouded in obscurity and can be merely speculated upon. Secondly, I am attempting to find out how such a process of transformation from tribe to caste may be reflected in the selfidentity of this people. Do they interact with the dominant Sinhala Buddhist culture as a tribal group from without, or as a stigmatised caste group from within, or as both?. I am particularly interested in why this group has remained marginalised when other castes which originated as immigrant groups have succeeded in assimilating into the Sinhala mainstream. Thirdly, I am exploring the status of women in such a marginalised group vis-a-vis their men and the degree of control they exercise over their lives. Finally, I am placing the Rodi in a broader context of interaction with the dominant culture and examining the impact of political, social, economic and media transformations taking place at national level since Sri Lanka's independence in 1948, in order to identify the main trends affecting their marginalisation and assimilation.
This is a working paper which arose out of intermittent periods of field work. The information and data come mainly from one Rodi village. So this is a survey of sorts; it does not pretend to be anything more. I have refrained from arriving at any conclusions based on statistical evidence because my “sample” has been small. In fact, this paper offers nothing conclusive, and is to be regarded as being at the incipient stage of formulating questions.
I think that studying marginalised groups like the Rodi could help to understand and define the nature of the reality from which they are marginalised - i.e. the dominant Sinhala
2

Buddhist culture. I have always been fascinated and depressed by the manner in which the Dhammadipa (island of the Buddhist doctrine) ideology has succeeded in homogenising the heterogenous ethnic traditions of this society. So I hope that studying the Rodi would also help in defining some of the boundaries of inclusion/exclusion of the dominant culture.
Note: I have used the word Rodi rather than Rodiyas to denote the group as the suffix -ya carries with it a certain derogatory connotation. I have stuck to the use of Rodi rather than Gadi (which is the traditional term that the Rodi use to refer to themselves) except where I considered it appropriate, because of a lack of consensus among the Rodi on what they should be called, and the greater familiarity of the general population with this term.
4. Bechert, 1978, p. 7 s All scholars who wrote on the Rodi in the past with the sole exception of Nevill, referred to the group as Rodiyas.
3

Page 8
I
Historical Sources on the Rodi
The name Rodi receives no mention in any of the earlier Sinhala chronicles. There is however, the story of Duttagamini’s (101-77 B.C) son, Prince Saliya, who renounced the throne for a Candala woman of great beauty, Asokamala. The Candala, who are mentioned again twice as hired labourers in public works during the reigns of King Ila Naga (95 - 1 0 1 A.D) 7 and Vijayabahu IV (127-173 A.D.),o seem to have occupied a low status. However, there is no evidence to support the view that the Candala and Rodi are one and the same people, except that popular notions, generally of the nonRodi, seem to assume a link between the two groups of people. There is only one Rodi village near Anuradhapura where the inhabitants use the term Sadol (which is derived from Candala. and found often in Buddhist literature) of themselves. It is conceivable that the chronicle traditions would have some impact in the Anuradhapura region.
Another term found in the chronicles is Luddako or Luddo, a group of hunters in the time of King Duttagamini.' Luddi is the Tamil name for the Rodi. Ludha in Pali means “hunt”? The Rodi call cattle luddo in their language. There
6 Geiger, 1950, Ch. XXXIII, p.228, vv.2-4
7 Ibid., Ch.XXXV, p.247, vv. 16-18
B Geiger, 1930, Ch. LXXXVIII, p. 190, vv. 106-107
9 Raghavan, 1957, p.32
0 Geiger, 1950, Ch.XXX, pp. 187-88, vv. 9-12 and p. 190, v.341
Nevill, 1887, pp.87, 1 18
2 Raghavan, 1957, p.74

could be a connection between all these terms and the Rodi, indicating that they were either a hunting or a pastoral group of people. The popular association of Rodi is with the word rodda or “piece of dirt”. However, it has been suggested that both Luddako/Luddo and Rodi might be a derivative from the Sanscrit raudraka, a hunter. 'o
The first written mention of the term Rodi appears in the Sinhala chronicle, the Janawamsa, attributed to the 15th century or earlier.' The Janawamsa is not a historical chronicle but contains a set of theories or myths in some detail about the social structure of the island, specifically the castes. The Rodi are described as people who weave nooses and ropes, belonging to the Sudra division (the lowest caste in the Hindu caste structure; the influence of the Brahmanical tradition on the writer is obvious). The Janawamsa explained the origin of the Rodi thus: people who were afflicted with incurable diseases like leprosy were condemned at the beginning to live in caves and tree hollows away from their relatives; then their disease got a little better, their hair and beards grew longer and they ventured into the villages to beg; they lived unclean and had nothing to subsist on, so they survived by making nooses and ropes. The name Rodi, according to the Janawamsa, was a combination of roga (disease) and dadi stubborn.' Here we see the beginning of an ideology of uncleanliness associated with the Rodi by the dominant culture.
The Rodi began to be more extensively mentioned in the Kandyan period, particularly by British travellers and administrators. Knox, writing in the latter part of the 17th century, provided a story of their origins: the Rodi were
Ibid., p. 75 1“ Nevili, 1887, pp. 103, 1 13
Ibid., p. 103 Ibid., p. 103

Page 9
descended from the Dodda Vaddahs, hunters who were required to supply the King with venison; but one day instead of venison they brought human flesh to the King; the King's Barber revealed this to the King, whereupon the enraged King decreed that all members belong to this Rank or Tribe be expelled from the villages and lead a roving life, begging for their living.
And they are to this day so detestable to the people, that they are not permitted to fetch water out of their wells; but do take their water out of Holes or Rivers. Neither will any touch them lest they should be defiled.
Moreover, the wives and daughters of “Great and Noble Men” who lost favour with the King were delivered to the Rodi, as a punishment worse than death. The nobles themselves were executed and their womenfolk were cast off to the Rodi; the only other choice available to the women was to drown themselves in the river.
D'Oyly writing between 1815-1824 also mentioned “consignment to the Rhodiyas' as one form of punishment under Kandyan law.
There is still one species of punishment said to be sanctioned by ancient customs which can be inflicted only by the King’s order viz - the consignment of persons of superior caste to the caste of Rhodiyas, who are in general estimation the villest and most despised of the human race. The infamy of such a punishment cannot be
7 Knox, 1984, p. 70 辑然 Knox, 1984, p.71

equal led, and of course never be retrived (sic). It was awarded only for the most atrocious offences, and no more than two instances are known to have occurred in modern times.'
He gives the names of 7 koopaayam (hamlets) of the Rodi in the Kandyan Country under the control of two Adigars, 6 in Uva under a Dessawe (DisaWe, a feudal chief of a province), two in Matele (now Matale), adding that the Rodi were more numerous in the Provinces of Saffragam (Sabaragamuwa) and Seven Korles (North-Western Province). Also mentioned are two hamlets in Walapaney (Jaffna) which we know nothing about today. In addition he wrote that the Hiragey Kankanams (prison officials) appointed one of the Rekewal people (guards) to be Hoolawaliya (chief) over the Rhodiyas and he in turn appointed a Gas manda (supplier of rope for catching elephants) from among the Rhodiyas under his authority.
However, the first substantial ethnographical material about the Rodi came from Nevill in 1887. He wrote that the Rodi were divided into two sections, Nahalle or Udakatuwa Kandiya (Kandyan section) and Navaratna Valli (Vanni section), both of which “hold no intercourse with each other'.' He gave the names of four kuppayams in ancient Wanni and four in the Seven Korles, all belonging to the Wanni Kandiya. The Vanni Rodi, according to him, were divided into 12 exogamous clans (gotra), sons belonging to the father's clan and daughters to the mother's. The Kandyan Rodi did not have a clan structure; this he attributes to the Rodi social structures disintegrating under the influence of the Kandyan feudal system.?o
9 D"Oyly, 1975, p.90 20 Ibid., p. 113
2 Ibid., p. 114
22 Nevill, 1887, p.81 发3 Ibid., p.83

Page 10
Nevill studied the language of the Rodi and saw some affinities with Mug and Burmese in etymology, oriental Turkish in the structure of composite verbs and awkward compounds for what were single words in English or Sinhala. He provided one of the earliest versions of the Ratnavalli myth recorded from the Vanni Rodi, as their myth of origin. He suggested that underlying the myth were traces of an ancient religious tradition, practised by the Rodi. By his time, the Rodi claimed to be Buddhists and held the gods as inferior. He also mentioned Gara Yaka Dance and Bali, practised in the folk healing tradition of the dominant culture, as part of the Rodi religious repertoire as well.
According to him, the headman of each tribe was called a Hula valiya and a leading man of several tribes was called a Gas manda. ?? We can assume that by “tribe” he meant a “hamlet” since there were not so many tribes of Rodi, even according to his own information. He also did not explain how these institutions functioned. He described dress, customs related to birth, puberty, marriage and death; these were not very different from the customs of the Sinhalese, except for their greater simplicity. Men used to wear cruppers but by his time they were wearing a short cloth; they also wore ear jewels of silver or brass. Women wore a cloth up to their knees, ear jewels of silver or brass, brass bangles, finger rings; they were permitted to wear only small beads on their necks. Both men and women were bare-bodied.
The Rodi, according to his account, were beggars by hereditary custom, thieves, entertainers, makers of nooses and
24 Ibid., pp. 88-89 2S Ibid., p.82 26 Ibid., p.84 7 Ibid., p.83 28 Ibid., pp. 85-86
Ibid., p.84

rope, and scavengers (they removed carcasses of dead animals). All services rendered by the Rodi had to be paid for in either cash or kind; a Rodi would not budge from the compound of a high-caste person until paid, and his prolonged stay there was considered to be a disgrace to the household. A Rodi would beg for all her/his needs - rice from the farmer, pots from the potter, toddy from the toddy-tapper and so on. “While oppressed he is also an exacter,” Nevill commented.
The last and most extensive source on the Rodi is Raghavan who spent nine years studying this group. He listed all the Rodi villages in the island and enumerated the population at 3, 122 in 1956.3° He did a generalised ethnography of the Rodi for the whole island, giving details of regional variations. He was particularly interested in the changes taking place in Rodi villages in terms of social welfare measures, and provided very short case studies of a number of such villages. Raghavan's work is too long to summarise here, but I would be referring to it at appropriate places in this paper.
Ibid., p.85 Ibid., p.85 3. Raghavan, 1957, pp. 5-9

Page 11
I
From Gadi to Rodi: The Transformation from Tribe to Caste
From the historical sources, the evidence for a tribal origin comes from the vestiges of language, religion, hunting and aesthetic traditions that have survived into this century. The Rodi call themselves Gadi. Nevill has proposed that Gadi might mean “red', as preserved in the Tamil kadarm (purple); he pointed out that Kadavara (Gadavara) Deiyo (the name of a popular deity of the Vanni region) shares the same sense of "red'. He further suggested that the Gadi might have a link with Magadha (“Red Mother') region of India.
A tribe, for the purpose of this paper, is defined as a Society with a high degree of self-sufficiency at nearsubsistence level, politically autonomous, and possessing a distinctive language, religion and other cultural norms and traditions that provide members of this group with a cohesive sense of identity."
No one really knovs vhere the Gadi come from. Their language, according to Nevill, has similarities with Mug and Burmese. De Silva who provides the chapter on language in Raghavan's book, has suggested lexical affinities and structural similarities with the Munda group of languages, spoken by the hill tribes of Orissa, Bihar and Chittagong in India. He has proposed the inclusion of the Rodi “dialect”
33 Nevill, 1887, p. 1 18
34 This definition has been arrived at with the help of: Shorter,
1974, p. 2 and Southhall, 1970, pp. 28-48.
35 Nevill, 1887, pp.88-89.
6 Raghavan, 1957, pp. 1 02-107.
O

within the Austro-asiatic group of languages, so named by Schmidt who studied the Mon-Khmer people, and who included both the Munda and the Mon-Khmer languages as branches of this larger group." On the other hand, the grammatical structure of the Rodi language is very similar to Sinhala. Geiger, who divided Rodi words into four categories, has suggested that only one category of these words bore no resemblance to Sinhala. The other categories included words with clear Sinhala derivatives where syllables had been changed or compound words formed for an object that is referred to with a single word in Sinhala, so as to appear unintelligible to the non-Rodi, in a manner employed by other marginalised groups and subcultures. He also stated that the grammatical structure of the Rodi language was in no way diferent from that of Sinhala. It is possible that the Gadi language started out with a distinct etymology but has down the centuries incorporated Sinhala structures and words, and in its present form functions as a secret language of a marginalised group.
The Gadi social structure seems to have been relatively free of hierarchy. Of the kinship terminology that has survived in their own language, there are terms for mother, father, husband, wife, son, daughter, grandmother, grandfather, sister and brother; but there are no terms denoting hierarchy as in the Sinhala terminology. Of course it is not clear whether such terms existed in the past and were replaced by Sinhala terms. The Rodi rarely use their own kinship terms any longer. According to Nevill, the Gadi used to belong to exogamous clans in the past with a bilateral system of descent, but this was confined to the Vanni Rodi even during his time,'
7 Ibid., p. 106
38 Geiger, 1897, pp. 3-7, 25-32 39 Raghavan, 1957, pp.26-27 40 Nevill, 1887, p.83
11

Page 12
and had completely disintegrated by Raghavan's time.' Today they are a patrilineal and predominantly patrilocal people. The chief of a Rodi village until the abolition of the Kandyan feudal administration system, was called a Hulavaliya,' which is a distinct term of the Rodi, and is not found among any other caste or tribe. The word, according to Raghavan, is possibly derived from hula or sula meaning a “torch”; hulavaliya then means the “torch bearer', the man who leads by the light of the torch. This would have a relevance in the context of a hunting group. If this were originally a tribal institution, it had been appropriated by the administration system of the Kandyan Kingdom at the latest, and the Hulavaliyo were being appointed by higher authorities in the 19th century.“
Evidence of a hunting tradition is suggested by the suffix villiya attached to every Gadi male name in the past; this could be derived from the Tamil word vill or villu meaning “bow”, hence villiya denoting a bowman or hunter. Meanwhile, the traditional suffix for female names, valli, according to Nevill might mean “tuft”, carrying the connotation “tuffed”. He pointed out that tufts of hair were a common feature on arms, shields and ornaments of “primitive races” in the Indian subcontinent.
Nevill also suggested that the Gadi might have been a “race' which had a prejudice against ploughing and digging since the verb “to plough” in their language is bin talawwe hapa karanawa (injuring the earth).“
4. Raghavan, 1957, p. 12
42 Ibid., pp.12, 14, 15, 18, 33-34, 36, 87 43 Ibid., p. 16.
44 D"Oyly, 1975, p. 114
45 Raghavan, 1957, p.77
46 Nevill, 1887, p. 118
47 Ibid., p. 118
48 Ibid., p. 111
12

The Rodi have retained some of their skills as hunters to the present day. Until recently they supplied hide ropes for elephant kraals and spent several weeks in the jungle hunting sambhur and deer for these hides. Nevill wrote that the Vanni Rodi kept hunting dogs.“ According to Ferguson, the Rodi "hunt a great deal with trained dogs, and invariably use bows and arrows”. The Rodi still relish meat of the terrapin and iguana lizard, as well as fish, hunting these with no other tool than a knife and/or stick. Fish are also speared with a knife. This process of hunting and fishing is called kotanawa (chopping) in their colloquial Sinhala.
Crafts and music are important elements of Rodi socioeconomic life today. It is not clear how many of these traditions originated in a tribal context. The crafts they practice vary from region to region; there is cane work, drum-making, broom-making, hair-piece making, mat and basket weaving. The design of the cane objects produced in the villages visited was very distinct, different from cane objects produced by, non-Rodi craftspeople. Questioned about their designs, they informed me that these came down from generation to generation from the time of their ancestors. The musical tradition of the Rodi included invocatory songs sung by women while they spun brass plates, drumming on the bum maendiya (one-sided drum) by the men; their dance tradition included a communal round dance. They still perform some traditional music to make a living on the streets, but in the village itself this tradition has been replaced by the bajav session (a gathering among friends to sing and dance to popular music), with the accompaniment of a dola (two-sided drum). They have retained a high degree of skill at improvisation. It is possible
' Ibid., p.83 SO Ferguson, 1895, p.250 s Raghavan, 1957, pp. 54-59
13

Page 13
that knowledge of certain crafts, such as basket weaving, and elements of their musical tradition came from a tribal past, but were transformed and modified to meet the requirements of a market in the first case, and a caste occupation in the second case.
Up to this point I have outlined some of the elements that may be the remnants of the Gadi tribal culture. We have no idea at what point the Gadi were incorporated into the caste system. Neither historical sources of the Sinhalese, nor the oral tradition of the Rodi themselves, are very helpful.
The following observations made by Ryan on the Sinhala caste system, I think, are relevant to understand the position of the Rodi: that the Sinhala caste system was loosely structured at the beginning because the Sinhalese reached the island in periods of incipient caste organisation in North India; that the later South Indian influences reinforced this principle of social organisation, possibly during the medieval period, as it was given scant mention in the early Sinhala epics; that structurally, the Sinhala caste system is close to the South Indian system; that many Sinhalese castes represent immigrant groups who were Sinhalised as bodies rather than as individuals; that the historic basis of the castes might have been frequently tribal, but unlike in India there was no great borderland between tribe and caste.’
For the purpose of this paper, I will consider two working definitions of caste. Caste, as defined by Ryan, is
a social organization structure functioning through hierarchical birth status groups, they or their subunits being communalistic and usually endogamous, and possessing functional and ritual roles including expression of social distance and
S2 Ryan, 1953, pp.11-15, 17
14

privilege in reference to and in distinction from other social groups in the great society.
According to Bougle, as paraphrased by Dumont,
the caste system divides the whole society into a large number of hereditary groups, distinguished from one another and connected together by three characteristics: separation in matters of marriage and contact, whether direct or indirect (food); division of labour, each group having, in theory or by tradition, a profession from which their members can depart only within certain limits; and finally hierarchy, which ranks the groups as relatively superior or inferior to one another.
Even though division of labour and hierarchy are common to both definitions, I find the emphasis on separation, which is social distance carried to the extreme, a relevant distinction. Most castes in Sri Lanka have traditionally maintained a separation in marriage; but a taboo on shared food consumption is rigidly applied only to the Rodi. Thus the interaction between Rodi and non-Rodi include areas of social distance but it also includes areas of clear separation.
According to most authorities on the Sinhala caste system, the Rodi are placed at the bottom, below the Kinnara (a caste of mat and basket weavers). However, the Rodi and the Kinnara each claim superiority over the other. The rivalry between the two groups is legendary, but the Kinnara are an
S3 Ryan, 1953, p. 19
S4 Dumont, 1972, p. 57 and Bougle, 1971, pp. 9-27
ss Ryan, 1953, p. 225; Armour, 1842, pp. 103-104; Hayley,
1923, p. 36
s Raghavan, 1957, pp. 37-39
15

Page 14
even smaller group than the Rodi and encounter the Rodi only in the Kandyan areas.
The Rodi, from the 17th century (when Knox wrote about them) until the 1950s had to deal with an enormous number of prohibitions, some of which were socially sanctioned, others legally sanctioned under Kandyan law. According to Knox, they were not permitted to fetch water from the village wells; no one would touch a Rodi for fear of being defiled. Tennent observed that they were condemned to roam the countryside as they were prohibited to stay in one place for any considerable length of time; they would have to stand aside on the road if a member of a high caste passed by; they could not enter a swept yard of a villager or draw water from wells; both men and women were prohibited to cover their upper bodies, or to wear a cloth below their knees. In addition to these prohibitions, Nevill mentioned that they were required to address Vellala or Goligama (the cultivator castes) people as Hamuduruwo (a hierarchical term of respect), that wooden or metal clappers were prohibited on the necks of their bullocks and they had to use a stick and a coconut-shell instead, that Buddhist priests did not accept food offerings from them, that a touch of a Rod;1man to a woman rendered her an outcast.'
D'Oyly, writing on aspects of the Kandyan constitution, made the following observations regarding the Rodi: they were allowed to have only a slanting roof for each hut (similar to a lean-to) and a cadjan screen with a hide on it for a door; they were not permitted to cross a river in a boat, to travel through a Royal village or walk on the embankment of a canal in the
57 Ibid., p. 11 * Knox. 1984, p. 70 o Tennent, 1977, p.697 6o Nevill, 1887, pp.84-85
16

Royal fields, and as a result the Rodi of Dumbara had no communication with the Rodi ofUva. For criminal acts such as cattle stealing, a Rodi would be flogged and confined by members of another Rodi hamlet which was furnished with stocks for this purpose. For seizing Vellala/Goigama women and for plundering villages, they were put to death, the executioners being other Rodi. However, if a group of Rodi were culprits of such crimes, people of other castes surrounded the whole hamlet, and fired upon the entire population, regardless of sex or age, innocence or guilt. Such an occurrence took place in Paranagama in Uva, where 30 people were killed and the hamlet set on fire for seizing high-caste women and highway robbery, when a certain Angammena was Disawe of Uva.9
It is evident from the above historical sources that the Rodi were severely oppressed under the caste system, especially in the feudal regimes of Kandyan kings. However, it is not clear how far their jurisdiction extended. Nevill observed that the Vanni Rodi were relatively free of the constraints imposed by the Kandyan monarchy. Several old people from my research area, which comes within the Vanni category, told me that they had supplied nooses and ropes for the last elephant kraal of Nugavela Disawe in Kasipote; thus they were under the jurisdiction of this Disawe. From the information provided to me, I am inclined to agree with Nevill that the Vanni Rodi were not administered or policed to the same extent as the Kandyan Rodi; the caste sanctions confronted by the former were socially enforced.
The function of the Rodi in the caste system was to provide nooses and hide ropes. The institution of the Gasmanda (maker of nooses) probably originated in the caste
6 D’Oyly, 1975, pp. 114-115 62 Ibid., p. 115
17

Page 15
context. Ferguson noted that the ropes were made of Nepare fibres as well as of hides of beasts. It is possible that the Gadi made hide and fibre ropes for their own use in the tribal past. Williams, writing about the hunter/gatherer Birhor (who speak a Munda-related language) of South Bihar, has mentioned that the group made rope from inner-bark fibre of certain vines and bartered these, as well as small game, for rice in the neighbouring villages. Thus barter of this kind between Rodi and non-Rodi might have been institutionalised as an exchange based on rights and obligations within the caste system.The Gas manda who was a sort of deputy to the Hulavaliya was also appointed by the Disawe; his main role was to collect all the hide ropes from members of his hamlet and supply these to the Disawe for the elephant kraals. In the absence of the Hulavaliya, he also performed the functions of the former, i.e. to settle disputes in the community and maintain law and order.
Under the caste system, the Rodi also had the sanction to live off the surplus of other castes, especially the cultivators. This occupation was called illala kanawa (asking and eating), to be distinguished from hinga kanawa (begging and eating) which is used in relation to ordinary beggars.' Since begging as a caste occupation was their birthright, no one could refuse them, and Goigama people kept aside a portion of paddy or rice for them at the end of every harvest. How and why a tribe of hunters was reduced to a caste of beggars can only be a speculation. It is possible that the Gadi, who had survived as a hunter/gatherer tribe since their arrival from India at some point in time, were incorporated at the bottom of the caste system at a time when this system was
63 Nevill, 1887, p.83
64 Ferguson, 1895, pp.250-51 6S Williams, 1977, p. 128
66 D"Oyly, 1975, p. 114
18

being consolidated, perhaps during the medieval period. But why were they placed at the bottom and why was their status that of an outcast/untouchable group? In contrast, the Vedda (a group which has retained its hunter/gatherer existence into this century) have been left entirely out of the Sinhala caste system.
Some of the answers to this question might lie in the religious/mythical tradition of the Rodi which I have not touched upon so far. The Rodi have a myth of origin beginning from their ancestress Ratnavalli. This myth has been handed down in their oral tradition and is not found in either the written or the oral traditions of the Sinhalese. Nevill provided the first recorded version of the myth:
At Parakramabahu's court the venison was provided by a certain Vaedda archer, who, during a scarcity of game, substituted the flesh of a boy he met in the jungle, and provided it as venison for the Royal Household. Navaratna Valli, the beautiful daughter of the King, discovered the deception, and fascinated by a sudden longing for human flesh ordered the hunter to bring this flesh daily. The Vaedda accordingly waylaid youths in the woods, and disposed of their flesh to the royal kitchen. The whole country was terrified by the constant disappearance of youths and maidens. It happened that a barber who came to the palace to complain of the disappearance of his only son, while waiting was given, by the servants at the royal scullery, a leaf of rice and venison curry. Just as he was about to eat he noticed on his leaf the deformed knuckle of the little finger of a boy. Recognizing it by the deformity as that of his son, he fled from the palace and spread the alarm that the king was killing and eating the youths of the
19

Page 16
city. The facts then came to light, and the King, stripping his daughter of her ornaments, and calling up a scavenger then sweeping out a neighbouring yard, gave her to him as wife, and drove her out to earn her living in her husband's class. The princess and the scavenger fled from the town, and as night came on asked for shelter from a Kinnara, but were angrily repulsed.
The narrative is usually followed by a set of invocatory verses to Ratnavalli, who appears to be a deity rather than a princess, even though there is a line or two, which mention in passing that these verses are to King Perakum's daughter, Ratnavalli.
Nevill has suggested that the Ratnavalli of the verses is
the goddess Pattini in her fearsome manifestation of Kali.
Wrapped in wreaths of cool balmy sal flowers Oh! woman, At whose incantations diseases vanish, Who wears the fearsome strings of corals, Oh, Ratnavalli, respond to our call and descend.
By permission granted by Munindu, that day Your triumphal progress even gods dismayed, Beating the drum, you go from house to house, And Nagas rejoice at the dance of Ratnavalli.
The name Ratna-tilaka-Valli befits you; With rituals awe-inspiring I propitiate you;
And you whose twentieth year has passed,
You shall not go without the taste of flesh.'
67
68
69
Raghavan, 1957, p.42 Nevill, 1887, p.82 Ibid., p.87
20

Taking on from Nevill, Raghavan has focussed on the underlying meaning of these verses, connecting them to the ceremonial cult of the goddess Kali as well as her gentle manifestation as Bhagavati), which has roots in the early forms of worship practised by tribal groups in India.
The fearsome necklace of corals is the garland of human skulls round the neck of the awe-inspiring Kali. Munindu, who gives her permission, is here the great Lord Siva, and the reference is to thë permission which Siva vouchsafed to her, to betake herself to the mortals below, where she would be received and worshipped with proper rituals. The awe-inspiring offerings are the human bloody sacrifices."
The Rodi today do not practise worship of Ratnavalli. They insist that she was a princess, daughter of King Perakum and their ancestress, but not a deity. However, a cult of Ratnavalli existed in the Pihiti and Maya districts in the past, according to Barnett." She was worshipped by a telem bu tree in Anuradhapura, and King Dutugemunu (161-132 B.C.) had to appease her with blood sacrifices in order to cut down the tree and build Ruwanvalidagaba on that site. The dagaba was apparently named after her, ruwan having the same meaning as ratna (golden, precious). It is plausible that such a cult did exist and was submerged during a period of consolidation of Buddhism, as was characteristic of the Dutugemunu reign.
Raghavan pointed out that such a cult might have existed among the early peoples of the region, in the hunting stage of life, and that the outcast status of the Rodi might be connected to their membership in this cult.
70 Raghavan, 1957, pp. 62-63 " Ibid., p.64
21

Page 17
That a form of worship in which human offerings formed the essential ritual would have been anathema to the Buddhist way of life goes without saying; and it needs no stretch of imagination that any class of people in whom the cult prevailed or survived, even in an attenuated form, would have been pronounced by the Sangha as exiles from the social order.'
Nevill saw a link in the apparent contradiction between Ratnavalli, the princess of the myth, and Ratnavalli, the goddess of the invocatory verses, by interpreting the myth symbolically. He suggested that the cannibalism of Ratnavalli might symbolise human sacrifice to Ratnavalli, as a manifestation of Kali, and that religious cannibalism might have survived until the reign of Perakumba (1153-1186 A.D.)
I think it is important to keep the myth and the verses together in order to understand the Ratnavalli tradition and its relevance to Rodi culture and identity. Elaborating on Nevill's recognition that the myth has to be analysed symbolically, I would like to go back to the old question of transformation from tribe to caste in the Rodi dentity. It is clear that the Ratnavalli myth is not a “tribal myth” of origin, but rather a myth to explain the outcast status of the Rodi visa-vis the caste system. Thus, it can be termed a “caste myth". The verses, on the other hand, are clearly invocatory ones to a deity (modified only slightly to maintain a link with the myth), vestiges of a religious tradition that the Rodi do not practise any longer, but which could have been a part of their tribal way of life.
If a cult involving human sacrifice was outlawed during a period of consolidation of Buddhism, and if the members of
72 Barnett, 1917, pp. 86-87 73 Raghavan, 1957, pp. 65-66
22

such a cult were declared outcasts, how would this group explain their outcast status, without revealing the dimension of human sacrifice, which would offend the sensibility of a Buddhist culture? One way of doing it would be to conceal their deity Ratnavalli in the form of a princess, human sacrifice at her shrine in the form of human flesh consumption in the palace, the votary of the deity in the form of a dirt-sweeping servant who followed the princess into her exile, the exile itself standing for the persecution of the cult. In fusing the verses to the myth, the Rodi perhaps not only maintained a continuity with an older tribal myth of creation, but also continued to pay homage to their deity, as they sang her praises while begging from door to door. When the rituals were forgotten as the Rodi were forbidden to perform them, only the memory of the deity Ratnavalli was kept alive in metaphor from generation to generation. When I present the versions of this myth collected in the field, it would become evident that the Ratnavalli myth is a living myth, modified through time to suit the changing concepts of identity among the Rodi. However, I think it is important to stress that the Ratnavalli myth appears to be a “counter myth', constructed from within the group, rather than a myth frofi, the dominant culture imposed from without. The lack of a version of this myth in the Sinhala written and oral traditions, as well as the contents of the myth and the accompanying verses in the Rodi tradition, point to a construction of the myth from within. The myth will be discussed in detail in the section entitled "The Ratnavalli Myth”.
Myths involving human sacrifice to gods/goddesses exist among the hill tribes of Orissa and Bihar. In a Munda myth, the tribal male deity Sinbonga, appears in the guise of a boy and offers himself as a sacrifice to the supreme deity Haram.'
"* Nevill, 1887, p.82
23

Page 18
There are many versions of a myth among the Konds of a boy being offered as a sacrifice to the goddess Darni Pinnu (also called Tara Pinnu and Dala Pinnu, o including one version which retells how the government stopped this practice.' Dalton has described the fearsome rituals performed by the Konds to propitiate Tara Pinnu, including the sacrifice of youths and maidens, at a sacred grove." The sacrifices were offered to bring about fertility of the land and animals. The Konds hunt small game but are mainly shifting cultivators in the forest. The human sacrifices were stopped by the horrified British administration. The linguistic link of the Rodi to the tribes of Bihar and Orissa has to be re-examined to ascertain whether the Rodi myth has roots in the myth of the Konds. At any rate, it is possible that the Rodi, in their Gadi stage, were one of the groups who had a belief system based on the archaic “Great Goddess"/"Mother Goddess” concept."
In examining the stigmatization of the Rodi, in the context of the human sacrifice element in the Ratnavalli myth, one has to reconsider Ryan’s notion that there are no. "untouchables' in the Sinhala caste system, and the only “outcaste untouchable” status present (in relation to the Rodi) rests more on secular taboo rather than religious proscription. He further stated that in Buddhism, no man was “unclean' in the sacred meaning of the concept.'
I am suggesting that the untouchable status afforded to the Rodi has a religious basis, in addition to the social basis. Not only from the perspective of a “great tradition” (Buddhism) engulfing a "little tradition” (Ratnavalli cult) but
75 Verrier, 1954, pp.538-541, 545-551 76 Ibid, p.546
77 Dalton, 1973, pp.285-88, 296-98 7s Carmody, 1981, pp. 19-45
79 Ryan, 1953, pp. 16-17
24

from a central precept of Buddhism, ahimsa (refraining from violence). All groups of people who took life, whether as a means of livelihood (hunters/fishers) or for the purpose of religious sacrifice, seem to have had an ambiguous position vis-a-vis the Sinhala caste system. Inclusion in this caste system implied also being Buddhist. The Veddas who are hunters do not belong to the caste system at all. However, the Rodi, who also may have come from a hunting tradition and additionally propitiated a deity with human sacrifice, were incorporated into the caste system, but as outcasts. The fact that the Veddas did not practise human sacrifice may be the reason that they were left alone. The Rodi, on the other hand, were confronted with the prohibition of this ritual and a "punishment” that brought them under the control of the caste system, but as outcast, despised human beings.
Moreover, the popular notion of a Rodi as "unclean” is connected with his/her food consumption habits, i.e. eating flesh of a dubious nature, implying game not normally eaten by the peasants. But this is not merely a secular taboo. Perhaps in the Buddhist scriptures there is no concept of "unclean” as applied to the human being. But in popular Buddhism ("village Buddhism” of Southwold), a central precept is “not to kill animals”. As Southwold points out, this does not mean that
I am referring here to ahimsa as a philosophical concept and an ideal in the Buddhist belief system, not to its practice by Sinhala Buddhists; not only the recent events of the ethnic conflict between Sinhalese and Tamils, but the old chronicles, Mahawamsa and Culawamsa (which were written by Buddhist monks) reveal the extent of violence in the history of a Buddhist society, including the justification of violence in the name of Buddhism.
Southwold, 1983, pp. 65-69
25

Page 19
Sinhala Buddhists do not kill animals; but the notion functions . as an ideal in the Sinhala Buddhist world-view.
Buddhists, including village Buddhists, regularly explain that it is not so much what one does that has effect, but the intention, the state of mind and being that issues in the deed.
It is not very difficult to conceptualise how a label of "unclean' may be attached to a group of people who do not share such a worldview. Thus, contrary to Ryan, I would like to suggest that a notion of "unclean' exists in relation to groups who take life, which is implicit, rather than explicit, in a Sinhala/Buddhist caste syctem. In addition, it has to be remembered that the Sinhala caste system is borrowed (though modified) from the Indian system, and the Sinhala worldview carries within it the sacred concept of pure/impure, as vestiges of a Brahmanical tradition. The marginalisation of the Rodi, in contrast to other castes who arrived as immigrants but succeeded in assimilating into the Sinhala Buddhist dominant culture, has to be understood in terms of the Ratnavalli tradition within the group, and the notion of "unclean' without the group.
Ibid., p. 69
26

IV
The Research Area
Ranwewa is a large Rodi village in the Kurunegala District. It has 128 houses, according to a survey by the local Grama Sanvardana Samithiya (Rural Development Society). My estimate of the population, based on a figure of 5 individuals to a house, is roughly 600. It was settled under a state colonisation scheme in the early 1950s. The original settlers were 49 families who moved mainly from three nearby Rodi villages. They were allocated two acres of dry land and one acre of wet land each. They were also provided with assistance to build brick and concrete houses with tiled roofs. The increase in the number of houses is the result of children building in their parents' compounds, as well as newcomers coming from various Rodi villages in the country. These houses are of the wattle-and-daub type with cadjan roofs. The newcomers of course have no legal claim to the lands they occupy. Some of them have built houses on the hilly wet lands which have remained uncultivable ever since the lands were distributed to the original settlers. This is because no proper irrigation scheme was ever provided for the village. In this Yala season only 8 acres (out of the original 49) are being cultivated with rice. Around 25 acres are cultivated in the Maha season (as more fields receive water), I was informed. The rest of the wet lands are on hilly ground and have never been cultivated.
The village is situated near a tank; there is also a river
nearby which dries out during the dry season. There are two wells for potable water; these dry out too during the dry season. Ranwewa is surrounded by Goligama (cultivator caste)
27

Page 20
villages; there is also a household of the Rada (washer caste) and the Nakath (drummer caste) nearby. There is no explicit hostility among the different castes.
The Rodi in this village are largely dependent on their traditional craft, cane work, in which they are engaged at household level. The women also make wigs and hair-pieces. Some families who come from other areas of the country are engaged in broom-making and drum-making. Only five families were cultivating their land this season (all of whom were also dependent on cane work to survive); two had given the land out on badda (lease) to nearby Mudalalis (owners of village stores; they were non-Rodi). I was told that families sometimes also give out their land on ande (share-cropping) on the basis of 1/4 for the owner and 3/4 for the Mudalali who cultivated it by hiring labourers. Every house is surrounded by coconut trees; every household owns some animals - poultry, pigs, cows. Coconut trees and cows are also given out on ande sometimes. The older generation still engage in caste-sanctioned begging.
The village has an elementary school, Kindergarten through Grade 5. It had a pre-school, initiated by a state organisation dealing with community development, for about two years, but this has closed down for lack of funds for a teacher. There is a ruin of a coir rope workshop also established by the same organisation; the building is now occupied by a family. There is no place of religious worship. They have a Maranadara Samithiya (Funeral Assistance Society) and the Grama San vardana Samithiya (mentioned above), both with fluctuating membership. They meet in the pre-school building and the President of both societies is the same individual - a non-Rodi from a nearby village, said to be concerned about the welfare of the community. Many households own radios and cassette-recorders. I was told that the village had 15 TV sets; I observed four during the course
28

of my visit. These are powered by batteries, since the village has no electricity.
I also visited one of the ancestral villages of the settlers, Managala, from where 24 families (half of the settlers) came to Ranwewa. The two villages are closely linked by kinship ties and shared traditions. Managala, according to one of the village elders, dates from the 18th century, a Kuppayama (hamlet, settlement) where ancestors of the present community had been allowed to stay for free. They now hold legal ownership to the land. This village has about 30 houses on around 10 acres of land. This is a very crowded village, with houses almost entirely of the wattle-and-daub type with cadjan roofs.
This village is also near Goigama and Batgama (caste of palanquin-bearers) villages; there is one Nakath household in the surrounding area. The Rodi encounter no open hostility from the other castes. People make a living on the cane industry and begging. There is a Cane Centre, established in the 1950s under a state department dealing with crafts. The instructor is a young Rodi woman from the village, educated up to “O'” level. The centre offers a training course for 15 trainees. At present, there are 10 girls from the village and 5. from outside (non-Rodi) in the course. The centre also buys cane objects made by the people in the community at piece rates. There is no paddy cultivation in this village; there are some coconut trees and people keep a few animals.
Managala has an elementary school and a library built with the assistance of a Buddhist-oriented community development organisation, which was active in the area in the 1950s. There is no evidence of the presence of this organisation there any longer. There is a Bo-tree and a ruin of a small building which was supposed to have been an avasaya (dwelling) of a Buddhist priest in the 1950s. People own radios and cassette-recorders in this village too. There is one TV
29

Page 21
set. There is no electricity. One young Rodi man owns the
village store.
Both villages come under the area known historically
as the Seven Korale, thus, these are Nevill's Vanni Rodi.
Methods
This research was done in the course of about four months in 1986. I visited Ranwewa on four occasions for extended periods, Managala once. I stayed in the house of one Rodi family, introduced to me by a friend who had been involved in one of those failed projects but had nevertheless maintained cordial relations with members of this village.
Most of my data was gathered through participation observation, in-depth interviews, informal conversations while visiting homes of the people. Altogether, I visited 20 huts, i.e. 15% of the total. I had a questionnaire which I attempted to use but it proved to be too cumbersome, so I dropped it somewhere along the way; I was not getting anywhere with it. I interacted with members of this community of both sexes and all ages without much difficulty. The short nature of this research and the intermittent visits (I was occupied with the historical sources in Colombo during the rest of the time) have not contributed to making this an in-depth analysis. This is merely a scratching of the surface, and further research needs to be done.
I often make generational comparisons; this is partly due to the fact that ages given by people whom I interviewed were not always reliable. However, I have retained these approximate ages, as well as my generational categories. The four categories I have used are: gen. l - the parents of the original settlers; gen. 2 - the original settlers; gen. 3 - the children of the settlers; gen. 4 - the grandchildren of the settlers
30

did not encounter any hostility in the village. The only exception was a youth in his early 20s in the village of Managala. He was drunk and wanted to know what I was doing there. He complained that there were too many people who came to the village to find out about their life, took photographs, etc. and made money out of their poverty. The other people defended my presence there, and continued to abuse him soundly after he disappeared. I considered his question ing valid and did some self-reflection on the relevance of this research to their community; this is unfortunately beyond the scope of this paper. Names of the villages as well as individuals have been changed. All interviews were conducted in Sinhala. and the English translations are my own.
31

Page 22
V
The Rodi Identity: An Ethnic Identity?
In this section, I have attempted to grapple with the nature of the Rodi identity. I have assumed so far, that down the centuries a tribal identity was transformed into a caste identity. Or that a caste identity was superimposed on the tribal identity of the group. One can no longer sort out where the tribal identity ends and the caste identity begins. The main difference between the two is that a tribal identity would not necessarily assume a hierarchical relationship with other groups, whereas a caste identity would.
Ryan has pointed out that the historical basis of the castes in Sri Lanka may have been tribal; he has also observed that many Sinhala castes represent immigrant groups who were Sinhalised as bodies. What is special about the Rodi (whether they are an immigrant or an indigenous group) is that they kept a language, unlike other castes in the Sinhala hierarchy, and that they were outcast. It is possibly that there is a link between keeping the language and being outcasts; i.e. as they were not assimilated into the Sinhala mainstream, their discourse was largely within their own group, and a language that nobody el se could understand might have been useful for strategic reasons in a hostile environment.
As the Rodi express a self-awareness of their group, and their identity is culturally constructed with a basis in a language and mythical tradition of their own, I have decided to explore it from the perspective of an ethnic identity. The elements I have identified to construct an ethnic identity are: the self-making of the group, the Ratnavalli myth, language, names, hunting, caste-sanctioned begging and the aesthetic tradition.
32

According to Epstein, an ethnic identity could be expressed in a continuum marked by (+) and (-) poles: a positive identity is based on self-esteem, a sense of worthiness of one's own group's ways or values, manifested in one’s attachment to them; a negative identity is based on an image of self resting on internalised evaluation of others and behaviour promoted by the desire to avoid their slights or censure. I would like to consider whether a positive or negative identity, or both, are reflected within an individual, among different individuals and in the collective identity of the Rodi.
In my analysis, I am also taking into account that an ethnic identity is constituted of two components (1) a categorical aspect of ascription and identification, based on an interplay of internal and external factors; (2) an affective aspect of cognition supported by unconscious associations and identifications. It is the latter component that provides the emotional charge attached to an ethnic identity.
The self-naming of the group
A sense of group-consciousness is very much evident among the Rodi. The “we”/"they” boundary is used and reinforced in everyday life. The Rodi, in conversation, refer to themselves as ape minissu (our people); outsiders irrespective of their caste are rate minissu (people of the country); in Kandyan areas, this term is applied to the Goligama caste. Or they use ape kattiya (our group) as opposed to Oyalage kattiya (your group). There was one instance when a youth in his early 20s made the distinction ape minissu and honda minissu (good people); he came from a village in the Uva, where according to his own information, casteism is still very pronounced. An
82 Epstein, 1978, pp. 101-103
Ibid., pp.96, 108-112
33

Page 23
older man at Managala referred to ape rodi valliya (our Rodi clan or group; valli and valliya are old suffixes to Rodi names that have been mentioned before). The Rodi are casteendogamous and marry into distant Rodi villages; these they call ape gam (our villages).
So far I have dealt with the terms they use to refer to themselves, in contrast to others, in Sinhala, which is the language they generally communicate in within the village. In their own language, the Rodi call themselves Gadi. NonRodi are Baggiri. However, it is mostly the older generations (i.e. l and 2) who would declare themselves as Gadi. On two occasions when I had put the question, “What do you call yourselves?” to people of generation l, and they had answered Gadi, arguments broke out between those individuals and the youth who were listening, who wanted to make a case for Rodi. Here, I reproduce some of their conceptions.
Ridigira (58 yrs, genı.2, male)
“We don’t call ourselves Rodi. We're Gadi. It’s like the
way the Goigama people are known as Handuru Minissu. They
don't call themselves that. We and other people call them so.”
Utiya (86 yrs, gen. l., m)
“We call ourselves Gadi. Rodi comes from the story of the Rodda. Ve say gadiyange gadiya (the Gadi of the Gads), gadi bilanda (Gadi boy), gadi bilandi (Gadi girl). The word Rodi is not so good. Gadi? Nov ve say things according to the custom of the country. Now they say isay, thamuse, that's how some of our people speak. We don't say Rodiyo to ourselves.'
Sumane (early 20s, gen.3, m) adds:
‘’Gadi is our language. Ve used to say ara gadiya (that
gadiya), megadiya (this gadiya). Now we don't say those things
any more”.
34

Kakulumali (80, gen. 1, f)
“We call ourselves gadinge gadi (the Gadi of the Gadis). Our jati (caste) is Rodi. I call this boy, gadiya... ”
Dharmapala (early 20s, gen.3, m) joins in:
“We call ourselves Rodi. Our jati is Rodi. Like Goigama, Batgama, Rada..."
Somaratna (early 20s, gen. 3, m) adds:
“We don’t say Rodi to ourselves. It’s people from outside vho say that. Ve say Gadi”
Dharmapala
“We say Gadi also. But our paramparawa (lineage) is Rodi. But...”
Kakulumali
"In our language ve say Gadi, not Rodi.”
Somaratna
“If we see someone from one of our villages we say, there's one of our gadiya. That means, that’s one of our people.”
My question
“So in your language there's no word Rodi?"
K/D/S
“No. Ve say Gadi, not Rodi. ” Dharmapala
“We don't say, there goes a Rodi.”
As is evident in the different conversations, the term gadiya is also used in their colloquial Sinhala, in the sense of “chap',
35

Page 24
“fellow" or "guy"; gadi is used for “woman” or “girl". A similar meaning is attached to the terms angaya (male) and angi (female). The word gadin orgadiyan (plural) is also used to refer to people of their group. When domestic violence broke out once over food stamps in a particular household, on the day that the new food stamps were being issued, a teenage girl observed: “pen nadda, gadinta muddara hambu vela” (Can't you see, the Gadi have got their stamps).
From the Rodi's own self-definitions, I see an ambiguity. The Rodi, both older and younger generations, refer to themselves among their group as Gadi (the word has crept into their Sinhala usage as well). In identifying themselves in relation to other groups/castes, both generations would call themselves Rodi. However, there is a difference in the way generations l and 3 would answer if an outsider comes and asks, “What do you call yourselves? Who are you?' The older generation would tend to answer “Gadi”, and the younger “Rodi”. The difference has something to do with positive and negative identity. Generations 1 and 2 have retained, to some extent, a memory of a self-defined identity. Generations 3 and 4, on the whole, seem to have internalised a caste identity imposed from without.
How is the process of transformation from a Gdi identity to a Rodi identity reflected in the group? Is a Rodi identity constantly being imposed on this people from without? Or are there forces or a mechanism within that reinforces and perpetuates a Rodi identity? To attempt to answer these questions, we have to turn once again to the Ratnavalli myth.
The Ratnavalli Myth:
I have already outlined Nevill's version of the Ratnavalli myth, comments from the historical sources and some of my own speculations in the section entitled From Gadi to Rodi. The
36

Transformation from Tribe to Caste. In this section, I wish to explore how the myth has developed in a contemporary context and how it might define the present-day Rodi identity, in a conscious or unconscious manner. My understanding of the Ratnavalli myth and its impact on the Rodi identity is based on the versions of the myth retold to me, as well as comments made on the myth by people from the community. This section comprises:
(1) Background to the Ratnavalli myth
(a) Language of the myth (b) Occasion of retelling the myth (c) Possession of knowledge of the myth (d) Validity of the myth within the community
(2) Versions of the Ratnavalli myth and analysis
I am considering the Ratnavalli story as a “myth", based on the following definition of Eliade:
In general it can be said that myth, as experienced by archaic societies (1) constitutes the History of the Acts of the Supernaturals; (2) that this History is considered to be absolutely true (because it is concerned with realities) and sacred (because it is the work of the Supernaturals); (3) that myth is always related to a “creation', it tells how something came into existence, or how a pattern of behaviour, an institution, a manner of working were established; this is why myths constitute the paradigms for all significant human acts; (4) that by knowing the myth one knows the origin of things and hence can control and manipulate them
37

Page 25
at will; this is not an “external', 'abstract knowledge but a knowledge that one experiences ritually, either by ceremonially recounting the myth or by performing the ritual for which it is the justification; (5) that in one way or another one “lives the myth, in the sense that one is seized by the sacred, exalting power of the events recollected or re-enacted.'
In analysing the myth, I have used an eclectic approach. I have focussed both on the structure and the contents and tried to understand how these might correspond to the Rodi sociocultural realm as well as its relationship to the dominant Sinhalese realm. This analysis is concerned only with the "particular” as opposed to a “universal” meaning of the myth.
Earlier writers have kept the prose part of the myth (which they termed a “legend”) and the verses separate, even though their speculations were based on combining both. Whenever I use the term "myth” in this section, I am referring to the narration and verses together as a whole, except in the versions where the verses were not included in the retelling.
(1) Background to the Ratnavalli myth:
(a) Language of the myth:
One of the significant aspects of this myth is that it is always retold in Sinhala. None of my informants were able to provide
me with a version in their own language. I was repeatedly told that the myth was never recounted in their language.
84 Eliade, 1975, p.6
38

Kakulumali (80, gen. 1, f)
"We don't have any verses or stories in our language”. Ridigira (58, gen. 2, m)
“We have never told about our descent from Ratnavalli in our language'.
This is once again an indication that the Rodi myth in its present form does not come from a tribal past, i.e. it is not a tribal myth. However, it cannot be ruled out that the myth or parts of it could have been in their own language in the past and that surviving generations have retained no memory of such a version.
(b) Occasion of retelling the myth:
In many societies where a mythical tradition exists, myths are retold in a specific time and place, because myths re-enact a time of creation, a time of the ancestors, i.e. a sacred time. Myths have a regenerative function, being performed on occasions where the original creation is reactualised, as for example in a new year or as part of the healing process of an ailing member of the community.
The myth of the Rodi however, is not recited at a particular time or place. It is not performed on ceremonial occasions. It is not even passed down to the children as the history of their origins. The verses are still sung by the women when they go begging/plate-spinning and the narration is recounted whenever a non-Rodi person requests it. Only one person told me that he did teach his children the myth. It is not clear whether the myth or parts of it were recited or performed in the community in the past. At any rate, it seems that the myth is now reserved merely for non-Rodi consumption.
8S Ibid., pp. 43-47
39

Page 26
Kakulumali (80, gen. 1, f)
“When we go spinning plates, we first sing the verses of our lineage - before anything else. We sing the verses only when we go begging, not at home. We don't tell the story either at home. The children learn when they accompany us to go
begging.”
Ridigira (58, gen.2, m)
“We don’t have to tell the Ratnavallistory to get things when we go begging. It's an old tradition to give us new rice. We tell ourselves that story. We tell our children sometimes. It's important that they know something that they wouldn't have otherwise; I have told it to my children. Some of our people will forget this; others will remember it. It's just like teaching someone to read. Some will learn, others won't.”
It appears from these statements, then, that the Rodi are concerned with defining who they are not to themselves, but to the outside world. If their children know the myth, this is due not to an effort on the part of the adults but rather to circumstances. On the other hand, in many traditional societies clildren learn things not by being explicitly taught, but by observing and hearing adults as they go about their work. Thus, it is also possible that the myth is not recounted in the community because the adults already know it and the children would pick it up at some point from listening to adults when they do retell it.
(c) Possession of knowledge of the myth:
Those who knew the myth in the community were always willing to recount it. These individuals were all from
generations 1 and 2. Some of them had been “referred” to me by people in the community as having knowledge of their past;
40

others were people questioned on my own initiative in the course of my visits to houses. Members of the younger generations I encountered claimed they did not know the myth; not all older people claimed knowledge of the myth. Thus, there does not seem to be a specific criterion for knowing or not knowing the myth, except that those who did know the myth were older people. Both men and women were able to recount the myth, but women seemed to be less willing or did not know it as much as the men. Of five older women from whom I requested the myth, only two obliged; two claimed no knowledge of it and the other expressed general irritation at being disturbed. Of the older men I asked, all obliged. Thus, I was able to work with two female and four male versions (one of the latter from Managamia). Each version is different. Each seems to be an interpretation from the viewpoint of the particular individual, and knowing some personal background of the individuals concerned helped in making sense of their versions.
An important aspect was that while the women always incorporated the verses into the myth, the men, with one exception, did not. One of the male informants pointed out that the verses belonged to the women; one female informant indicated that the verses were passed down in the female line.
Ridigira (58, gen.2, m)
“Only the women sing the Ratnavalli verses. The men don’t have a right to those.” (my italics)
Kakulumali (80, gen. 1, f)
"When we go spinning plates, we first sing the verses of our lineage... The children learn when they accompany us to go begging. Only the girls go with us. It's very seldom that the boys go with us. Of course we don't take the girls anymore either. You can't take a girl on the road nowadays.”
41

Page 27
Significantly, I encountered many women in the community of generations 1,2 and 3 who did not know the narration part of the myth but were able to sing the verses. If the verses were or are a “right” of the women, why did one of the male informants include them in his version of the myth? The only reason I can think of is that this man considered himself and was respected by others as a man of knowledge in the village, and had a repertoire of verses and stories, including some knowledge of old Sinhala texts as well. Thus, he was in a sense the village “intellectual” and made it his business to know everything, including things he was not supposed to know. As no other men in the village were able to sing the verses, I am contending that the women possess the knowledge of the verses while both sexes can possess knowledge of the narration.
(d) Validity of the myth within the community
If the myth is somehow meant for the non-Rodi, does it have any validity at all within the Rodi community as a “history” of their origins and an explanation of their present status? Or do the Rodi consider the Ratnaval myth as some sort of story they have invented to persuade outsiders of their plight or/ and make a living? It is clear that the verses are now sung by the women when they go begging. However, as it has been previously pointed out, the Rodi do not have to recount the narration to get rice or money; whether they had to in the past remains unclear. To estimate to what extent the Rodi believe in the myth they retell to outsiders is always problematic as I am non-Rodi myself. I can never be sure whether people in the community actually believe in the myth when they say they do, or whether they are maintaining appearances for my benefit. At the least, I am assuming that the Ratnavalli myth portrays an identity that the Rodi wish to convey, for whatever reason, to the non-Rodi.
42

At the same time, my own feeling is that people in the community do believe in the Ratnavalli myth to a great extent, even though they have mixed attitudes and feelings towards it. No one told me that it was nonsense or false, with one exception. She was a GCE 'O' level educated women who had a state-sector job.
Kamala (30s, gen.3, f)
“I don’t think we're descended from Ratnavalli. Because there was a Rodda sweeping the palace of the king in the first place. Where did he come from? I think we have this position today because of a caste division devised by a king of the past.”
Thus, she expressed the belief that the caste division was primary and that the Rodi had always been a low caste within the feudal social system.
Several younger people did not like the myth being retold or discussed because it was associated with the Rodda, and this reminded them of the stigma attached to the Rodi people. However, they continued to believe in it, even though they were not inclined to pass it down.
Karunawathi (late 20s, gen.3, f)
“I don't know the Ratnavalli story at all. I believe in it when I hear it. But I’m not going to retell it to anyone.”
Some younger men of generation 3 also expressed a belief that they were descended from Asokamala and Prince Saliya (i.e. the Mahawamsa story which I have referred to in p.4 but apart from injecting a comment here or there when an older person was recounting the myth to me, they were unable to retell such a version through and were unsure whether it fitted into the Ratnavalli myth or not. For example, one youth thought that Asokamala was a daughter of Ratnavalli; another
43

Page 28
commented that some people said that they were descended from Saliya and Asokamala. Such notions associated with the Ratnavalli myth have to be understood in terms of the influence of the dominant culture through media, Smatterings of “history” lessons in elementary school and the filming of Candala (an action/romance movie based on the Mahawamsa account of Asokamala and Prince Saliya) in a nearby Rodi village in the 50s.
Others, especially women who knew the verses, claimed no knowledge of the Ratnavalli narration but expressed belief in the myth.
Leelawathie (30s, gen.3, f)
“I sing the Ratnavalli verses. They're important to me, these verses, because they go with the plate-spinning... I don't know how to tell the Ratnavalli story. But I believe that our lineage came from Ratnavalli.”
It is my contention that no one really challenges the Ratnavalli myth seriously enough to invalidate it within the community. The myth seems to “work” even today as an explanation of their social reality. For them, nothing else provides a better answer to why they are called Rodi or why they have to lead the existence they do. Moreover, the belief in the descent from Ratnavalli, the memory of a high status in the past, seems to somehow elevate them from their reality - it provides them with a psychological outlet to transcend their present low status and stigma.
Noni (55, gen.2, f)
“Though we’re like this now, we come from the royal lineage. From King Perakum's daughter - were even older than seven generations. Though we make things out of cane now, that’s not the kind of people we were. We're from a good lineage.”
44

It is interesting that the Rodi do not generally use Buddhist ideology to justify their caste status, i.e. I heard no one explain that their situation was due to bad karma. Not a single individual during my stay in the community uttered the classic Sinhalese phrase Katakiyannada? Ape karume thamai. (To whom can we complain? It's after all our karma.) Even though the Rodi expressed knowledge of the Buddhist concept of karma (and pin/pav) in other discussions, they did not use it to explain/justify their low caste status. In one version of the myth the concept of karma was present - Ratnavalli’s act was referred to as a papakarmaya (bad deed). But the narrator concluded that this bad deed had been absolved as time passed by and Ratnavalli became a goddess.
Thus, it is the Ratnavalli myth, rather than any myth/ ideology of the dominant culture, that still provides an explanation to the Rodi of who they are. To examine how this myth of the Rodi has changed in its interaction with the dominant culture and how it has borrowed/incorporated elements of the myths of the dominant culture, we have to turn to the actual versions of the Ratnavalli myth.
86 karma: the Buddhist law of cause and effect, based on pin (merit) and pav (demerit); i.e. doing evil results in sorrow, doing good results in happiness.
45

Page 29
(2) Versions of the Ratnavalli myth:
Version 1 (Kakulumali, 80, gen. 1, f)
"We are descended from Queen Ratnavalli. King Perakum was her father. Those are our people. When the king was away in another city, she started something with a man who was sweeping the dirt in the palace. Once there was a harvesting ceremony and meat had to be brought to feed all the people. The king left instructions with his daughter. She called a hunter' and told him that her father wanted him to hunt for some game in the forest. He spent the whole day in the forest and found nothing. As there was no meat they were afraid that the king would kill them. There was a poor man who had a son. The boy had no mother. It was decided to shoot the boy. The boy was killed by the hunter and the meat brought. When the meat was cut, the little finger of the boy fell into
87 Sin. devinvahanse is the term used for the wife of a king. The use of this word is misleading here as the narrator says that Ratnavalli is Perakumba's daughter, not his wife; this might be a conscious ambiguity and will be discussed in my analysis. In some other versions of the myth Ratnavalli is referred to as a kumari (princess).
88 Sin. rodu athugana minihek (a man sweeping the dirt); the link between rodu (dirt, pl.) and rodda (dirt, sing as wellas Rodi man) is clear.
9 Sin. kelaveddah (forest Vedda). The Veddahs are a hunter/ gatherer tribe; popular terminology differentiates between kela (forest) and gam (village) Veddahs, the former being associated with hunting/gathering and the latter with shifting cultivation. The Rodi are called gam vedaahs by some peasants, a term which they (the Rodi) themselves reject.
46

the pile of meat that was to be cooked. Those were the days of God Sakra. Such crimes didn't escape his notice. When the harvesting was over, people came back home and were served rice and curry on banana leaves. The meat was also served. The little finger fell onto the portion of the poor man. He recognised the finger and ran off home to look for his son. The boy wasn't there so he returned and asked the king to find out what kind of meat it was that they had eaten. The hunter and Queen Ratnavalli were brought. They said it was the meat of hare. But the poor man showed the little finger. There was no way out of it. Some people told the king to kill Queen Ratnavallion the stake. Others said that nothing would come out of killing her, and that she should be married to the Rodda and banished from the palace. The king used to do what his people asked him to. So they were banished. They left the palace and reached the house of the drummers' who were in the service of the king. They recognised her and said: “Why, isn't this the queen of our king for whom we dance thovil?' When they heard the story they were too frightened to keep them. Until dawn broke the old man and his wife wrote a set of verses and gave it to Queen Ratnavalli. The old man cut one side of his drum and gave it to the Rodda to play the onesided drums. So singing the Queen went away.
By permission granted by Munindu in the past, beating the drums you go from house to house, your triumphal progress even gods dismayed, and
90 Sakra, the chief of the gods in the Sinhala syncretic tradition.
9. Sin. nagitiminissu are a caste of drummers who perform thovil (ritual healing/dances/exorcisms, accompanied by invocatory verses).
92 Sin. thovil, see note l 0 above.
93 Munindu, another name for Siva the chief god in the Hindu .
pantheon.
47

Page 30
Nagas' rejoice at the dance of Ratnavalli. The name Ratnatilakavalli you bear, you who have reached your twentieth year without fail, with rituals awe-inspiring I propitiate you, sparing nothing, you shall not go without presents.
These are the verses that were given to Queen Ratnavalli.
Thy wrists with bracelets jingle and glisten, thy ankles with anklets jingle and glisten, the string of corals round thy neck is noise less, King Perakum's daughter Ratnavalli.
These are what we say when we go spinning plates.
Sitting by the water's edge from afar Ratnavalli, not glancing here nor there Ratnavalli, to hear Buddha's sermon Ratnavalli, descend at least to the Buddha's command Ratnavalli.
Wrapped in wreaths of cool, balmy sal flowers, oh, woman at whose incantations diseases vanish, who wears the fearsome strings of corals, oh, Ratnavalli, respond to our call and descend.'
Version 2 (Noni, 55, gen.2, f)
"I cannot tell you this as well as my parents. This is something
that comes down, after all, from generation to generation from the royal lineage. This is how we’ve heard it. Ratnavalli was
94 nagas, mythical snake deities; also a mythical tribe who are
believed to have propitiated the snake deities. 9s I have used and modified to some extent, Raghavan's
translation (pp.62-63) to approximate the verses I collected.
48

the daughter of King Perakumba. Queen Ratnavalli lived in the upper storey of the palace attended by her servants as was appropriate for someone of the high caste. One day a servant-cooked beef. The smell of the food reached the queen who demanded some of it. The king told his daughter that this food was unclean to the royal caste and that she should push the thought out of her mind. But she refused and descended one of the seven storeys. After she had descended, the king was helpless and told her that she would be banished if she ate the man's food. Thus, they were both sent away. The king gave the queen a golden plate, the man a one-sided drum. She sang and he played the drum from house to house and so survived. The descendants of their children live to this day. When that bad deed was absolved, she became a goddess and climbed the tellambu’ tree which used to be at the place where the Ruwan valisaya now stands in Anuradhapura.ooo Before it was built they entreated her to descend from the tree. She said that she would descend only if it were built in her name. This wasn't done. They say that the Ruwanvalisaya sank nine times. Finally she was told to pay heed to the command of the Buddha and descend. After that it was built. Her bad deed was now absolved. Now she is a goddess. I don't know
96 Sin. papakarmaya: papa = pav (demerit), karmaya (deed),
derived from karma (see note 5 above).
97 Sterculia Foetida, a dry zone tree with red flowers and black seeds. The tree is believed to have been the abode of the goddess Ratnavalli, and could have been sacred to the Rodi in the past.
9 The Ruwanvalisaya where relics of the Buddha are believed to be enshrined, was built by King Dutugemunu (161-137 B.C), according to the Mahawamsa. Saya (a shortened form for chaitya) dagaba and stupa all refer to a bubble-shaped shrine, crowned with a cone, typical of Buddhist architecture. See page 21 formeaning of ruwanvali.
49

Page 31
to tell you more. That's how our lineage came into being. Even though we're like this now, we come from a royal lineage.
The name Ratnatilakavalli you bear, with rituals awe-inspiring I propitiate you, who have reached your twentieth year without fail, the name of King Perakum's daughter is Ratnavalli.
Thy wrists with bracelets jingle, thy ankles with anklets jingle, the string of pearls round thy neck is noiseless, the name of King Perakum's daughter is Rtnavalli.
Wrapped in wreaths of cool, balmy sal flowers, oh, woman at whose incantations diseases vanish, who wears the fearsome strings of corals, descend at least to the Buddha's command.
Golden is your beauty, in gold are you adorned, golden is your body, descend from the golden teambu tree.'
Version 3 (Utiya, 86, gen. 1, m)
“Our lineage started during the time of King Mahasammat.' The princess grew up to be disobedient. The king banished his daughter straight away. He threatened to kill her but actually didn't. At this time there was a certain man sweeping. She ran away with him. Our people are descended from them.”
99 Mahasammata is the mythical primordial king who founded the Buddha's family lineage. See Smith, 1978, pp. 51-52 and Clifford, 1978, p.38
50

Version 4 (Huyiya, 73, gen. 1, m)
“I’m telling this the way I've heard it from my parents. There was a king called Perakumba. His wife, the queen died, when their daughter Princess Ratnavalli was small. She was brought up by wet nurses. She grew to be very pretty and when she was of age, she wa sought after by many princes, who visited the princess in a dandumonarayanthraya.' One day she became pregnant. This was discovered by the father, who questioned the princess, the wet nurses and the guards. But nobody could tell him what had happened, not even the princess. You understand? The king asked the princess to get out of the palace. There was a man who swept the palace; his general administrative title was Rodda. The king told him to catch the princess if he so desired. He did so. They both went away to a house in the neighbourhood. This was the house of the drummers who gave them food and lodging for the night. The princess requested some necessities from them - a pole, two bags, a brass plate and a drum. They obliged. He had only one drum so he cut it and gave them one half of it. When dawn broke Princess Ratnavalli covered herself with a handkerchief and a tattered cloth, and took a bag and the brass plate. Her man took the pole and they both went on the job. They went to big houses, asking for paddy, rice and money. The husband said that his wife knew how to dance and sing the verses of their lienage. She spun her plate and sang, he played the drum. She sang:
By permission granted by Munindu in the past, beating the drum you go from house to house, your
' Sin. dandumonarayanthraya: the mythical flying machine of Ravana, king of Lanka, who stole the princess Sita, away from her husband, Rama, in the classical Indian epic, the Ramayana.
51

Page 32
triumphal progress even gods dismayed, and Nagas rejoice at the dance of Ratnavalli.
Glistening you come from afar, gentle and sweet from near, the jingling anklets in your ankles making music, the name of King Perakumba's daughter is Ratnavalli. Your hair is decked high over your nape, adorned with five lotuses, seven days you have remained in the same place, Ratnavalli is written in the beginning of the ola. leaf book.
They went on doing this job. Years went by. The caste division continued. That’s what happened to the Gadi. From that time our people didn't own even a little piece of land. Now the ogvernment has given us land. Some people also have jobs from the government. Our situation is now a bit better than before. Not much attention is paid to the caste division in the country any more. That's very good. Our people have food and clothes. So there's nothing wrong.”
Version 5 (Simeon, 63, gen.2, m)
“Our origin is like this. There were servants who worked in the palace of King Perakumba. The following names were given to them. The man who swept the palace was called a Rodda. The man who rowed the king's boat was called Padda... In this manner the king had his servants organised. The king had a beautiful daughter called Ratnavalli. As time went by she got involved with the Rodda who swept the dirt and ran
0. Ola-leaf a kind of palm leaves, on which ancient manuscripts
of Sri Lanka were written.
52

away with him. The king asked his servants to catch the princess and the Rodda and bring them to him. A Kinnara' man, one of those who cut reeds, caught them and brought the couple to the palace. He told her: “Daughter, you have no other job now. He split a drum and gave one half to the Rodda and told him to play it. He gave his daughter a brass plate and asked her to spin it and sing and dance. The couple roamed from town to town, street to street, singing, dancing, spinning, drumming. The man who swept the dirt was then called a Rodda, but the name was lengthened to Rodiya. So to tell the truth, were not people who are that lowly. We’re from the royal lineage. We’re descended from an ancestor who was born in the womb of Queen Ratnavalli. We might now not possess enough to improve our lives but our caste is not low. King Mahasammata divided the castes. But we are not lowly... In the future, these things will disappear. Now you, I and everybody are the same. The past is gone. Now the times are good for everybody. There's no time to say anything more; there's also nothing more to say.”
Version 6 (Ridigira, 58, gen. 2, m)
“King Mahasam mata’s son was King Perakum. King Perakum's daughter was Queen Ratnavalli. That Queen Ratnavalli's lineage is today our lineage. It happened due to a tale of the servants who worked in the palace of King Perakum. The different tasks were entrusted to different people. Because
' The Kinnara are the traditional enemies of the Rodi. According to Raghavan, the Kinnara were probably a tribal group which had co-existed with the Rodi in the past but the two groups maintain a feud, related to a legendary romance between a Rodi girl and a Kinnara man (pр.37-39.)
53

Page 33
I was asked to sweep the dirt, I was called Rodda. There were other posts like Commander, Prime Minister, Minister in the service of the king. One day after sweeping the floor, this man filled his winnow with so much dirt that he couldn't lift it himself. He asked for help. But none of the other servants wanted to touch the dirt, so they didn't lend him a hand. The daughter of the king, the princess, who was in the upper storey, saw what had happened and felt sorry for him. She helped him to lift the winnow full of dirt. There was no other relationship between them. She felt sympathy toward him out of her own goodness. Since it was a matter of reputation, they talked and jumped to the wrong conclusion. The lineage was smeared. They made a big thing out of it and chased the princess out of her home. It was her royal father himself who told her to flee with the Rodda. They walked through seven cities, made and played a one-sided drum, spun plates from village to village. It is from that lineage that we’re descended. We have that wandering, that job even today. That's how we survive.'
Analysis
When the six versions are examined in the context of the background material, it appears that the Ratnavallistory meets broadly Eliade's criteria of "myth". It is about the history of acts of deities and mythical ancestors; it is held to be “absolutely true” by the tellers and has general validity in the community; it is about a creation, i.e. how the Rodi came into existence; those who know the myth interpret/manipulate the "origin” of things in recounting the myth; the tellers live the myth to the extent that they believe the events of their myth and take pride in their lineage. What is absent is a ritual enactment of the myth through which “one is seized by the sacred, exalting power of the events” (page 30).
S4

In my attempt to understand the meaning of the Ratnavalli story, I am treating individual characters and events as strictly “mythical” with no historical basis. Even though Nevill has considered the possibility that the King Perakumba mentioned in the myth was a historical king,' this notion, I believe, is not worlth pursuing, as there were several kings with the name Perakumba (also known as Parakramabahu), according to the old chronicle, the Culawamsa,' which itself, containing its own collection of myths, is not a reliable source to go by. When one adds to this fact that in two versions King Mahasammata (a primordial king and mythical ancestor of the Buddha) is cited as Perakumba's or Ratnavalli's father, it becomes futile to assume a relationship between mythical figures and events and historical figures and events. However, what might be of some relevance is that the Perakumbas were medieval kings (the reign of the first, also called the Great, is dated around 1153-1186) and it was during the medieval period that the Sinhala caste system probably emerged/was reinforced.
Let us examine the elements of the myth that are common to all the versions: there is a queen/princess, daughter of a king, who breaks a norm/taboo; she is punished by being given to a servant and exiled, by the order of the king. Excluding the raher scant Version 3, the other versions continue with the following main elements: both queen/princess and the servant are taught skills to survive in a new life; her descendants carry the stigma attached to her act and take on the name of the servant.
Three of the narrations are followed by verses. These are invocatory in form and are clearly addressed to a goddess
o Nevill, 1887, p.82 10“ Geiger, 1929/1930, Ch. 50-90
05 Smith, 1978, p.242
06 Ryan, 1953, p. 11
55

Page 34
9? ሩ&
in two of the cases ("awesome rituals”, “at whose incantations diseases vanish'); she is entreated to descend to the command of the Buddha; all three versions however, claim to be addressing a daughter of a king. The apparent link between the narration and the verses is the name Ratnavalli, cited as daughter of Perakumba. One version of the myth (2) states clearly that Queen Ratnavalli later became a goddess, thus creating an explicit link between narration and verses.
It seems obvious that we are dealing with a myth of creation. Two of the tellers begin the narration by saying that it comes down from the ancestors. There is a creator - the King; he “creates” the first ancestors - Ratnavalli and the Rodda. The first ancestors are instructed in new life skills by the King or the drummers. These life skills - singing, spinning, spinning plates, dancing and drumming in exchange for rice and/or money - are part of the Rodi way of life to a limited extent even today. The gender division of labour outlined in the myth (Ratnavalli, the first woman, is taught to sing, dance and spin plates; the Rodda, the first man, is taught to drum) has an actual basis in Rodi culture as well. In Version 4, the objects used for carrying the rice (pole and bags) and the caste attire for the women are also specified.
However, this is a creation myth of a particular kind. The first ancestors are not created with recourse to nature (sky, earth, water, trees, animals) or supernatural powers, as in many tribal myths. In contrast, this creation is entirely man-made; it is the Sinhalese king Perakumba, who is the creator. Thus, this myth is not about the creation of a tribe from nature or supernatural beings. It is about the creation of a caste by a “higher” human being.
Yet there are elements in the structure of the narration - first ancestors, violation of taboo, instruction in life skills, gender division of labour - that could belong to a tribal myth of creation. Moreover, not only the invocatory form of the verses but also their content (praising the power and glory of
56

Ratnavalli, entreating her to pay heed to the awe-inspiring rituals of her devotees) could have comprised an actual ritual involving sacrifice to a tribal goddess. Did the spinning which accompanies the verses also have a ritualistic significance in the past? Many symbolic meanings could be attached to the repeated motion of spinning a round object - for example, cosmic motion, cycle of creation, cyclical time, and so on.
It is interesting that the new life skills of Ratnavalli and the Rodda are taught by the king in two versions (2,5) and by the king's drummers (nagiti or nakath minissu) in two other versions (1,4). At the beginning of the narration there is a social order, decreed by the king, of the status and function of people under his authority. Ratnavalli’s action and her punishment destroys this order. In the first case, the king himself is instrumental in creating a new order. The new caste, Rodi, and its occupation of begging/singing/dancing/ drumming are legitimised by being royally crented/sanctioned. In the second case, the task of creating the new occupation is left to the drummers. To understand this version, one has to look at the function of the drummers in the caste system. At one level, being a musician caste, they would be the best instructors for imparting a musical training. At another level, the drummers who perform in various ceremonies of the king, including auspicious occasions and exorcism rituals, could be regarded as having a regenerative/healing function; that would make them more suitable than any other caste (apart from the king) to assist in the creation of the new order. Note how this process begins in the night; by dawn, the new order is already in place, as Ratnavalli and her Rodda leave the hosue of the drummers, practising their new occupation from village to village. Thus, the creation is engendered in the darkness (of the night) and completed by the appearance of light (of the dawn).
The main character in this creation myth is undoubtedly Ratnavalli, the ancestress. In all the versions, Ratnavalli is a
57

Page 35
daughter of a king; thus, logica liy she should be a princess (kumari). However, only two versions (3,4) refer to her consistently as Princess; in two (1,2) she is referred to consistently as Queen and in the remaining two (5,6) as both Queen and Princess. In one version (2) she is also explicitly a goddess; in another (1) this is implied in the verses. Note too that the Sinhala word for Queen, as wife of a king (devin vahanse) used in the narration, carries within it the root devi (goddess). Thus, Ratbnavalli assumes multiple forms within the same myth.
One way of reconciling these different forms is to examine the corresponding relationships: i.e. ancestress : ancestor (Rodda); queen : king (Perakumba); daughter/ princess; father/king (Perakumba); goddess: “god” (Buddha). In the first relationship, Ratnavalli has a higher status on account of her royal lineage - she has power over the Rodda and her descendants. In the second relationship, Ratnavalli has a parallel status - she has parallel power to the king. In the third relationship, she has lower status - she has lost power to the king. In the final relationship, she begins with having parallel or higher status and parallel power or power over the Buddha but she ends up with lower status and loss of power. Thus, the ambiguity of Ratnavalli's titles may indicate her changing status and power within and without the group, as conceived in the myth. For example, when in Version 2, Ratnavalli is referred to as the ancestress of the Rodi, daughter of King Perakumba, Queen and goddess at various points, what might be implied is that she was all of these things to her people at some time or other. The order of events does not matter; whether she was Queen first and then became a goddess or she was goddess first and then became Queen/ ancestress. What matters is that Ratnavalli, the goddess, who was confronted by the Buddha, was (became) Ratnavalli, the Queen, who had parallel power to a Sinhalese king, was (became) also Ratnavalli, the daughter of the king, who was
58

subjugated but under his paternalistic fold and was (became) Ratnavalli, the ancestress who was declared an outcast for not conforming to the custom of the king's people. In the transformation of Ratnavalli from one form to another, from one title to another, might be concealed the transformation of her people brought into confrontation with the Sinhala Buddhist socio-cultural realm.
If we characterise the verses referring to goddess Ratnavalli as concerned with a “sacred” domain (acts of a supernatural being), and the narration referring to Queen/ Princess/ancestress Ratnavalli as concerned with a "social' domain (acts of a “cultural heroine’) for analytical purposes, however arbitrary such a division may seem, certain parallel elements can be recognised. In both domains, Ratnavalli moves from high status to low status; this is illustrated symbolically by the descent from the tellambu tree in the sacred domain and from the upper storey in the social domain. The conflict between Ratnavalli and the Buddha in the sacred domain has its parallel in the conflict between princess/queen and king in the social domain; in both cases, Ratnavalli concedes her power/status. Male power triumphs over female power in both domains. Moreover, the paternal protection of god Siva in the sacred domain is paralleled by that of the king in the social domain - goddess Ratnavalli solicits offerings (sacrifices) from door to door, a sacred right sanctioned by Siva; the outcast princess/ancestress Ratnavalli begs from door to door, a social right sanctioned by the king. Thus, the narration and verses together reveal how changes in the sacred and social domains are interrelated.
This relationship can be explored further by focussing on the difference between the male and female versions of the myth. In the female versions, the reason for the queen's . exile is based on the consumption of flesh (human flesh in 1, beef in 2). This element is completely missing in the male versions (even though it was present in the male versions
59

Page 36
collected by Nevill and Raghavan.) At the same time, both of these female versions replace the explicit last line of the verse collected by Nevill:
“The name Ratna-tilaka-Valli befits you;
With rituals awe-inspiring I propitiate you; And you whose twentieth year has passed, You shall not go without taste of flesh.”
with “You shall not go without presents' in Version 1, and “The name of King Perakum's daughter is Ratnavalli” in Version 2; however, the subtler lines “who wears the fearsome string of corals' and “with rituals awe-inspiring I propitiate you’ are retained in both versions. To understand the "flesh element' in the narration, one has to examine its link to ritual sacrifice (whether human or animal) to the goddess Ratnavalli of the verses. Note how the ritual of the verses (the sacred domain) has been transformed into a taboo in the narration (the social domian).
Let us consider that the Gadi were a tribe propitiating a goddess with "rituals awe-inspiring”. Such a form of worship would invariably come into conflict with a Buddhist sacred and social domain at some point in time. This conflict is revealed most clearly in the narration and verses of Version 2. The goddess Ratnavalli is entreated to descend from her telambu tree (a sacred space/shrine of the Gadi?) in the name of the Buddha so that a chaitya could be built in its place. After much persuasion she relents but only on the condition that the chaitya is named after her. Thus, we have a "little tradition” forced to concede to a “great tradition”. By agreeing
07 Nevilll, 1887, p.82 and Raghavan, 1957, pp.2-3
Nevill, 1887, p. 87 with a translation rendered more appropriate by Raghavan, 1957, p.63.
60

to name the chaitya in honour of Ratnavalli, the “great tradition” also tries to placate and encompass, in a paternalistic manner, the “little tradition' and its adherents into its fold. However, the telambu tree has been cut and replaced by the chaitya. Is the telambu tree also a symbol for tribe? Without the tree, without the goddess and the rituals, how could there be a tribe any more? Moreover, if the tribe or members of it continued to propitiate the goddess, they could have been declared outcasts and forced into an existence of begting by royal decree - i.e. a dependent relationship with the Goligama (cultivator) caste who had to set aside a portion of their harvest for the Rodi. This also provides an explanation for the fact that the myth is recounted to the non-Rodi. The myth, in effect, might be saying that if a Buddhist prerogative was responsible for destroying a tribal besief system and with it a tribal way of life and means of subsistence, then the same prerogative had to come up with some way to deal with the consequences. Transforming the Gadi into a caste of sanctified beggars called Rodi and maintaining them in a feudal nexus of rights and obligations could have been one solution.
The question still remains why the women have maintained the “flesh element' of the myth, and not the men. The answer, I feel, is to be found somewhere in the fact that it is the women who have the right to the verses. These verses to a goddess are probably older than the narration; to the women the former is primary, the latter secondary. Thus, they have to approximate their narration to the verses, i.e. they have to maintain the link between veses and narration, whereas the men, lacking knowledge of the verses, can distance themselves from the verses when they recount the narration. The single male version of the verses significantly left out the references to “awe-inspiring rituals” and the fearsome goddess, replacing these with descriptions of a beautiful, dancing maiden. So the verse too can be changed if desired, but the female versions, with a few modifications, have chosen
6.

Page 37
to remain more faithful to a tradition. This applies even to Generation 4. I collected the verses from a teenage girl who did not know the narration and whose version differed from the versions given in this paper; but she left both “fearsome necklace of corals' and “with rituals awe-inspiring I propitiate you' intact. I am suggesting that in a ritual worship of a goddess, who has both fearsome and healing roiles, women might have had a special place and that they might have possessed power and status from the knowledge of the verses. The women still have an awareness of the significance of the verses, though they do not express it explicitly.
Kalulumali (80, gen. 1, f)
“When we go spinning plates we first sing the verses of our lineage, before anything else...”
Leelawathi (30s, gen.3, f)
“I sing the Ratnavalli verses. They’re importatn to me,
these verses, because they go with the plate-spinning.
Otherwise, I can sing other verses...”
The first statement reveais that the Ratnavalli verses are, in some way, the primary ones. From the importance attached to linking the verses to the spinning, one could speculate again whether these originated together as a ritual, later to degenerate into an income-earning activity. Going back to Eliade's definition of myth, the narration could then have been the "myth', and the verses, the "ritual for which it is the justification” sometime in the past.
At any rate, the verses are still being passed down from generation to generation with the references to the goddess and rituals intact. It is difficult to understand what sort of figure of identity Ratnavalli provides for Rodi women. But it is clear that it is her image that is primary; the Rodda, even the king, are secondary beside her. Most tellers emphasise that
62

they are descended from Ratnavalli, the daughter of Perakumba, rather than Perakumba himself.
At the same time, the reference to Ratnavalli, as "daughter of Perakumba” in the verses may be significant both as a description and as a form of address. If goddess Ratnavali was reduced to a “little tradition” under the protection of a “greater tradition", she would have been symbolically the “daughter of Perakumba' and could have been last worshipped by her adherents under that title, so that the verses are, in fact, still invocatory, not just appear to be so. Moreover, the verses are entreating Ratnavalli to descend from her tree to the Buddha's command and do not say that she has actually descended, so the possibility still remains that she is available for propitiation. And as long as Ratnavalli continues to be on her tree, she and her people have both status and power. The women, by retaining the verses containing such an image of Ratnavalli, may be expressing an ideal order which they still uphold in an un/subconscious manner.
In the male versions of the myth, the basis of the exile and stigma of Ratnavalli and her lineage lies no longer in the “flesh element'. The basis has shifted to non-compliance with social norms connected with marriage/sexual morality, which both the Rodi and non-Rodi experience and can identify with. This preoccupation of the male versions with lapses in sexual/ moral behaviour of their ancestress is significantly missing in the female versions.
Version 4 is interesting in that it focus ses on an “illegitimate” pregnancy. For a Rodi woman to have a child out of traditional wedlock is by itself not a misdeed, as long as the father is known and/or acknowledges the child. However, a woman who cannot or will not name the father of the child is ostracised by the community and would not be received back until the Hulavaliya decided to take her back, usually after she had paid a forfeit. Such a woman could be ostracised for as many as 15-20 years. So this version of the
63

Page 38
myth provides a reason for the outcast status of the group from the tradition of the group itself.
Version 3 is very bare in details; since being disobedient and running away is such a cliche sort of reason for the banishment of a whole people, I will not dwell on this version. Version 5 attributes the exile and stigma to the illicit relationship between a high-born princess and the low-born Rodda, and the problem of encountering social barriers. It also concentrates on details of the caste system and its unjust impact on the people. Version 6 gives a new turn to the story by making the Princess Ratnavalli into an innocent victim of gossip and tales of the palace servants. She herself has merely empathized with the downtrodden Rodda. Her act of kindness/ solidarity is regarded as disreputable to her lineage and she is punished together with the Rodda.
These last two versions have added the element of injustice to the myth. They show the punishment of Ratnavalli and her descendants as a consequence of the rigidity of the caste system and therefore unjust. Ratnavalli and the Rodi are victims of social circumstances. These versions offer a “modern' perspective on the creation of their caste. To make sense of thcse versions, one has to look at the worldview of the two men. The first is a man who is vocal in his support for a centre-left political party; the second has supported the work of a community development organisation (claiming to be based on Buddhist ethics) in his village. Both believe that the party/organisation they have supported has been responsible for giving the Rodi a better deal, and that the caste system is dissolving. Their versions reveal their individual experiences and beliefs to a greater extent than the other versions.
Note that all the male versions start with King Perakumba or King Mahasammata at the top whereas the female versions begin with Queen Ratnavalli. The reference to Mahasammata is interesting because he is the mythical ancestor of the Buddha's royal lineage, and of the Sinhala
64

people, according to the Mahawamsa. The insertion of Mahasammata (as well as Perakumba) into the myth could reflect on the one hand, a desire to trace their own lineage to that of the Buddha and the Sinhalese; i.e. to be recognised as Sinhalese Buddhists. On the other hand, Mahasammata is also considered by the Rodi to be the creator of the caste system and the low status afforded the Rodi within it; thus, he might be regarded negatively as the designer of an evil scheme.
The different versions of the myth reveal, above all, that the Rodi community is not a homogeneous entity. We cannot talk about a collective Rodi identity. We are dealing with a community of individuals who interpret their social reality in different ways, who have assimilated in varying degrees or desire assimilation into the dominant culture. This is not surprising since the Rodi, after centuries of incorporation within the Sinhala caste system, see themselves as a caste of the Sinhalese. Their identity is by and large, a Rodi identity, rather than a Gadi identity. Note that the word Gadi was missing in the different versions of the myth, except in one - and that as an aside. The Rodi identity with the stigma attached to it is something that they naturally wish to be rid of by now. If the myth reflects an image of themselves which the Rodi want to portray to the non-Rodi, an assimilating identity seems to be more desirable to the men than the women. The male myth-tellers have in many ways departed quite radically from the older version of the myth, as recorded by Nevill; they have transformed it to conform to their own views, experiences and interests. Most significantly they do not wish the "flesh element” to be associated with their identity any more.
' The old chronicle traced rather ingeniously the genealogy of the early Sinhala kings Vijaya and Pandukhabaya, to the family of the Buddha. See Gunawardana, 1978, pp. 101-102.
65

Page 39
The women, on the other hand, have also changed the myth (from taste of flesh” to “presents” in the verses; from “human flesh” to “beef' in the narration) but not quite so radically. I have suggested that this might, in some way, be related to their right to the verses, which they theoretically could change but for some reason, do not. They still cling to the goddess Ratnavalli, even when they claim that she is not a goddess or that they do not propitiate/worship her. By retaining the verses and narration with the “flesh element”, they perpetuate unwittingly a Rodi identity that coincides with the notion of “unclean' associated with it by the dominant culture. At the same time, concealed in their verses are what I assume to be remnants of a Gadi identity and its conflict with the dominant culture in the past - i.e. goddess Ratnavalli vs. Buddhism. Perhaps in a subconscious manner, the women desire to retain a memory of a source of power they had enjoyed in the past - the goddess Ratnavalli. At the least, Ratnavalli does provide a psychological means to transcend their present low status. زج
The Ratnavalli myth appears to act as a complex process of identity formation among the Rodi, offering them both positive and negative self-images, depending on their own identifications, associations, and interaction with the dominant culture. It could be a boundary maintenance mechanism'' (as in the female versions) or an assimilating mechanism (as in the male versions), varying with the people who recount it and use it. The function of this myth (which could have originated as an explanation of the Gadi of who they are to themselves) today seems to be to provide an explanation of who the Rodi are to the dominant culture. This is not surprising as the creation of the caste Rodi, despised by the non-Rodi, did not take place from within, but from without. The myth of Ratnavalli is a dynamic within their own culture that changes their identity, just as much as their identity changes the myth.
0. Epstein, 1978, p.96
66

Language
It is not my intention in this section to add to the speculations on the origins and classification of the Rodi language. I did not have sufficient time in the field to study this language from a linguistic perspective. My interest in it was focussed from an ethnic perspective - i.e. to what extent the Rodi know their language and use it in their daily lives, and the attitude they express toward it. It has to be remembered that the Rodi have been abilingual people for at least a century (when Nevill studied them), possibly longer.
There is no particular name for the Rodi language; some call it the Gadi language, others simply "our language”. As far as I could observe during my stay, this language is not in extensive use in the community. There is also evidence that the language is being passed down, either formally or informally, to children in their homes. Daily conversations are carried out in Sinhala, with a couple of words from their own language thrown in. Sinhala is spoken with a slightly different intonation, reflecting the lack of long vowels (with the exception of long “a”) in their own language. Their Sinhala is absolutely fluent; however, I am unable to judge to what degree of competence they speak their own language. There is a difference at least across generations. Members of generation 1 and 2 both claimed knowledge of their language and were able to demonstrate it extensively. Their vocabulary was consistent with what I knew from the historical sources; i.e. they were not making something up for my benefit. Younger people (gen. 3 and 4) could also speak a word or two, but they claimed they had no proper knowledte of this language.
None of the Rodi could provide me with an explanation of where their language came from or how old it was; there was no myth about their language. The community was divided over the way they felt about this language. Some thought it
67

Page 40
was useful in a matter-of-fact manner; others were extremely contemptuous toward it. Only a few expressed some sort of pride in it or attached a measure of importance to it. Both positive and negative attitudes towards their language were found across age and sex groups. Some expressions of these attitudes are quoted below.
Kakulumali (80, gen. 1, f)
“I don’t know where this language comes from - may be from King Perakum. It’s a good language because no one else can understand it. If you speak Tamil or English, people can understand. But not our language. All our children know the language. It's like Tamil people knowing Tamil. We don't teach our children - they just learn it without any effort. The young only speak it among themselves. Its not an important language. But it's somewhat important for our caste. Our children will always know it because they hear it at home. There isn't anything wrong with that, is there?'
Utiya (86, gen. l, m)
“Our language doesn't come from anywhere particularly. It comes from our people. We talk to each other and we lear: from each other. They don't speak it now like before. The ones who know it, know it; the others don’t. Why should it be bad? It’s good to know it. If the children don’t want to learn it, they don't have to. It won’t disappear. It'll always stay with us... It has no particular use. When we go out to the villages, we always talk it to exchange a piece of information, if there are two of us. It’s not such an important language.”
Ridigira (58, gen. 2, m)
“Our language originated in our Rodi community. Our language is not historical; it is only ours. It has been created from generation to generation; it is only ours. It is not something that was made yesterday or today... Our lineage
68

will disappear if our language disappears. We still have to keep our language even if one of us becomes President. We have to maintain our dignity. These are historical things. We should keep our craft, the language we speak, and our skills like spinning plates... This language comes from our lineage. But we can’t trace now where it comes from. The usefulness of it is that no one else understands it. We can conspire to kill you and you wouldn't know (laughs). We don’t speak our language in the village. Why should we? We speak Sinhala.”
Somaratna (early 20s, gen. 3, m)
“It’s good to keep our language. We can talk between the two of us secretly in your presence. I don’t speak it at home. But my children pick up a word here and a word there... It's useful to us. For example when the Police come in search of us, we can inform each other secretly and hide the kasippu (illicit liquor) under their noses.”
From these comments, it is apparent that those who feel positively about their language do so generally on pragmatic grounds; it is useful or important became no one else understands it. Rarely is it considered valuable on cultural grounds; i.e. because it comes from the lineage. Those who were positive about their language were also optimistic about its survival; that at least some of their people, if not all, would retain knowledge of it in the future.
Those who held negative attitudes toward their language did so mostly on the grounds of stigmatism and thus, their desire not to be identified as Rodi.
Huyiya 73, gen. 1, m)
“This language of our village is something the people have made up. They speak it to one another. Let's say you come here; without your knowledge, we can say, “Let’s ask this lady for some money. We use it for that kind of occasion.
69

Page 41
That's why this language has been made. Otherwise, it's not in books or anything. Nobody speaks that language now; to be frank, no one has any use for it. Everyone speaks Sinhala now; after all, we're all Sinhala people. It's all the same to me if it's forgotten or not. It's an old language but people don't want it any more. They are embarrassed to speak it ... You understand? When our people, our youth go anywhere now, nobody can make out that it’s one of our boys. Or girls - it's the same thing. Unless they know us. So our people are embarrassed to speak this language.”
Simeon (63, gen.2, m)
"This language is not in books. It's a language created by our people. Our ancestors used it. Even in our time, we had the ability to speak it to a lesser extent. The new generation, however, doesn't know it. When I say a word in it to my son, he asks me, 'What are you saying to me, father?'. He doesn't know. This is not an old language. It is something that has been created recently. It is being abandoned now. We shouldn't pay any attention to it. In a situation where there are 18 languages, this one is not even in a book, it's not historical, it's merely been made up by people. This language is not useful for any occasion. We only suffer on account of it. That's for sure. It was the language of our anscestors but in the new age our younger generation will not possess it, not see it, not even hear it.”
Noni (55, gen.2, f)
"I know the old language. I taught it to my children. The small ones don't understand it nowadays. It doesn't really matter what happens to it. It's the same whether they learn it or not. There's no use for it. We speak it on necessary occasions, we don't otherwise. We speak it in the presence of people like you but we don't speak it ourselves'.
70

Dharmapala (early 20s, gen.3, m)
“Don’t talk to me about that language. It is a stupid, useless language. What you can say in one word in Sinhala, you need ten in this language. We have always suffered because of this language. People know we are Rodi when we speak it. Those old people who still want to speak that language do so out of a false pride, which hasn’t got us anywhere.”
It is apparent that those who felt negatively about their language did not also foresee its future survival. The recognition of the stigma attached to them, and thus their language, was often accompanied by the assertion that they were Sinhala people and therefore should be speaking Sinhala. They also rejected their language on functional grounds, i.e. that it was anachronistic and served no purpose any longer. A more sophisticated reason was offered by several younger men; that the compound words in this language were too cumbersome (for example, uhalle rabot anga - the leaf thing of the tree - for “branch”), and the practice of using the same word for several different objects too impractical, compared with the Sinhala. Note how the inferiority or insignificance of their language is often attributed to its being not a written language. Consider also that Ridigira (p.38) contradicts himself by saying that their language is not "historical' (implying perhaps that it's not written) at the beginning and that language is a “historical thing” (because it comes from the lineage) later on.
Whether they reject or accept their language, the Rodi recognise that it comes from their ancestors, that it has been passed down and that it was spoken more frequently in the past than now. However, even those who find their language useful often think that it is insignificant because they do not acknowledge it in its own right; comparing it with the
71

Page 42
"historical”, written Sinhala, they regard their own language as inadequate.
Thus, we have a language that is spoken no longer within the community (it is not clear if it ever was), that is not passed "down to the children, that has no songs or stories sung or recounted in it, that is regarded as having no value and considered a sort of secret code coined by themselves and/or their ancestors. Again and again, I was told that they used their language only in the presence of the non-Rodi to communicate among themselves. However, the practical value they claim for their language (“we can conspire to kill you' or “let’s ask this lady for some money”) seems to be minimal. The Rodi do not normally go about killing people in reality; neither do they need to have a long discussion among themselves before they decide to ask somebody for money. The functional value of having a secret language to communicate in the presence of the police could be regarded as a more valid claim. Yet my own observations of such an incident do not confirm this claim. When the police appeared one evening at one end of the village, the message was passed down from house to house by both adults and children casually walking along the village streets. It did not really matter in which language they conveyed the news because the police were not yet there to hear it; in fact, I heard the news being communicated in Sinhala. By the time the police got to the house where the illicit liquor was, it had already been stacked away. I am not ruling out the possibility that in different circumstances (e.g. if they had been taken by surprise by the police) they might have used their language as a secret code. However, I doubt that this is the primary reason that the Rodi language is spoken in the presence of the non-Rodi.
Assuming that their assertions and my own observations about the Rodi language not being in use within their own community are correct (i.e. the Rodi do not start speaking their language as soon as I leave the village), I think that there are
72

significant reasons for the use of their language in the presence of strangers. Outside their own village, the Rodi confront a hostile environment and people who regard them as inferior beings; in such a situation, their language could be a means of maintaining their solidarity and cohesion as a group, giving them both security and self-confidence. It also provides them with a sense of distinctness, that they might take pride in as part of their Gadi past. At the same time, being distinct would be still useful to them as a caste because this was the only way their subsistence rights (rice and coins) within the caste structure would be guaranteed.
Thus, the language continues to provide the Rodi with a distinct identity. The attitudes they hold toward it differ among individuals and within individuals. On the one hand, they feel it stigmatises them; on the other hand they value it as something only they possess. Speaking in Sinhala within their own village, they seem to use their language as a boundary maintenance mechanism that defines their group (from the non-Rodi “other”) but not the “cultural stuff (which is no longer very different from the dominant culture) that it encloses. As there are members of the younger generations who still value this language as a “secret language” while scoffing at it, it is difficult to foresee whether the Rodi language would be retained by some of their group or disappear altogether.
Names
The Rodi generally have a single name. Sometimes children might carry the father’s name as well (eg. Hurata lage Kirimuthu - Hurutala’s Kirimuthu), resembling the traditional Sinhalege (house) name. Having an additional family name
Ibid., p. 96
73

Page 43
(similar to the western surname) is becoming common for generations 4 and 5. This is accomplished in several ways:
(l) take on a new name from the dominant culture;
(2) invent a name carrying a letter or syllable from the
father’s (mostly) or mother's name;
(3) invent a name carrying parts of both mother's and
father's names.
According to Raghavan, Rodi individual names used to have the suffix villiya for males and valli for females.' I encountered the name Ridivilliya but no female name with a valli ending.
Naming patterns seem to follow general trends in the country, particularly in regard to generations 3, 4 and 5.
Generations l and 2 carry names such as Rangira, Pemiya, Kira, Kotuwa, Huyiya, Utiya, Bayiya, Mayiya among males. Females of the same generations have names such as Kirimuthu, Noniya, Kalkulumali, Ehalamali, Punchamali, Dinga. Some of these names are resonant of old pre-Buddhist names encountered in the Vanni region in general; others are abbreviations of Sinhala names, indicating low caste. Generation 2 also carries some anglicised names such as Jamis and Simeon for men and Rosalin and Lucia for women.
Generation 3 has common Sinhala names such as Wijeratna, Gunapala, Saranapala, Somaratna among males and Kamalawathi, Karunawathi, Sriyani, Kanthi, Tilaka wathi, Nilmini, Anula among females.
Generations 4 and 5 carry more “fashionable” sanscritised names such as Chaminda, Asitha, Nilantha, Rohitha among boys and Sunanda, Deepani, Himali, Dilusha among girls.
2 Raghavan, 1957, pp.29-30
74

Rejection of old names as ugly and/or stigmatising is widespread, across the generations and sexes. Some of the prevalent views on names are reflected in the following COrnmentS:
Huyiya (73, gen. 1, m)
"It's good that the old names are disappearing. They are ugly, aren't they? The new names are much nicer.”
Dharmapala (early 20s, gen.3, m)
“When we have the old names people know where we come from. It's much better to have rate (Goigama, non-Rodi) names.”
A girl of about 8 years refused to tell me her name until her older cousin intervened and pointed out that she was embarrassed to tell me her name because it was Noniya and she thought that I would laugh at it.
Thus, it is apparent that a Rodi identity would not be preserved in their names after generations 1 and 2 are gone.
Hunting
The Rodi do not practise large-scale hunting any longer. I found neither bows and arrows nor hunting dogs in their villages. Boys and youth hunt iguana and fresh-water turtle with nothing more than a knife and their bare hands. They also go fishing - catfish and eel are favourites.
However, there is evidence that the Rodi have hunted bigger game in the past.
Utiya (86, gen. l, m)
“Those days we used to eat wild boar, venison, sambhur, iguana and terrapin. When there were elephant kraals our people went to make ropes. We used deer and sambhur hides to make those ropes.”
75

Page 44
Huyiya (73, gen. 1, m)
“Those days we went into the jungle, caught some animal or other and ate it...”
Now there are no jungles in this region; most of the land is under intensive cultivation. The Rodi would not be able to Survive as hunters, even if they desired to. The small game that the boys bring home provide merely a welcome supplement to their daily meals.
Thus, the Rodi do not claim an identity as hunters today. They are also sensitive about being regarded as meat-eaters by outsiders, because of the “unclean notion attached to it. As a consequence, they would often claim that vegetables are the main component of their diet, even though my own observations confirm otherwise. They relish and would consume fish, terrapin and beef at the same meal when these are available after a good day of hunting/earning; they would go without meat and eat only vegetables and rice at other times. Food supplies are not stored or divided into portions for future meals. They basically live from meal to meal, not very differently from a hunter/gatherer existence. However, the greater part of their food they obtain through the market. So the Rodi clearly do not desire to be regarded as a hunter/ gatherer group; nor are they in effect.
Caste-sanctified begging
The dominant culture has for a long time considered the Rodi as a caste of beggars. However, they are not ordinary beggars; their occupation has been sanctified by tradition. The Rodi themselves distinguish betwen illalakanawa (asking and eating) and hingakanawa (begging and eating), the latter being associated with common beggars. Illa la kan a wa carries with it a connotation of a right conferred on the Rodi under the caste system; the other castes then have an obligation
76

to fulfill this right. Traditionally this right has been defined in the form of a portion of rice or paddy from the new harvest of the cultivator caste. It has also included coins in exchange for the entertainment of singing/dancing/spinning/drumming they performed.
Caste-sanctified begging is still practised by older Rodi men and women, and infrequently by younger Rodi. When they go begging, they always wear their caste clothing - men are bare-bodied, women cover their breasts with a small cloth, and both sexes wear their cloth up to their knees.
None of the Rodi I spoke to thought that this sort of begging was a “right” they should continue to enjoy. For the most part, they viewed the occupation with disdain or shame, and said they did it because they had no other means of survival. Note that in the following excerpts, my translation "begging” is always equivalent to the term illalakanawa used by the Rodi.
Kakulumali (80, gen. 1, f)
“I go begging in the villages where there is sympathy. I usually get paddy at this time of the year. Now our people get wages but we used to live on begging. I used to do cane work too but stopped five years ago. I can't do any work now so I go walking around to beg. They always trust a grandmother like me, call me into the house, lay a mat and give me to eat. Nobody insults me. There's no one as old as me in all our five villages.”
Ridigira (58, gen. 2, m)
“I can't say that begging is bad work. I'm talking about our problem. The worm in the Kohomba tree doesn't talk about the bitterness of the Kohomba leaf. It eats that leaf as food, eats to satisfy its hunger. It’s all the same to it whether the leaf is bitter or sweet. That's how it is for us. Otherwise there is no other reason. It’s not a privilege; it’s not easy or even
77

Page 45
profitable for us. We’re downtrodden by everything. It's because we have no other way of making a living that we live like this.”
Leelawathie (30s, gen.3, f)
“We never thought of it as a shame to walk around. If you look at it now, it seems to be shameful. But we’ve done it for so long we’ve got used to it. But I have resolved that my children after me won’t have to do this, though.”
It is evident from these comments that the Rodi do not enjoy the caste occupation that has been imposed on them and do not wish to maintain an identity as sanctified beggars. As long as they are able to produce craft goods that they can sell to make a living, they refrain from begging. However, in the absence of a system of social security for the aged, older Rodi may very well continue to practise their caste occupation in the future.
The aesthetic tradition
(1) Music, Plate-spinning and Dance.
The caste occupation of the Rodi specified in their myth is singing, dancing, plate-spinning and drumming. This tradition still persists in the Rodi community though it has undergone many changes. Men do not play the one-sided drum any longer, women, however, do sing and spin plates. If they had distinct forms of dance in the past, these have disappeared by now.
It is mostly the older women (generations l and 2) who sing the traditional Ratnavalli verses and spin plates when they go begging or trading in nearby villages or towns, or at the Munneswaram temple (Hindu) festival. Some of them are excellent improvisors. This tradition, passed down from
78

mother to daughter, is still valued by the older women even while they voice the hardships that go along with it. At the same time, women of all generations decry the tradition, associating it with begging and wandering around the countryside, activities which they do not consider respectable, especially for young women.
Kakulumali (80, gen. l, f)
“I’m not all that sad about my daughter not being able to spin a plate or sing. My granddaughters can’t do these things either. My two older daughters know a little. The women these days don’t know how to spin; they’ve never learnt it... In the future, nobody is going to do that. Our people nowadays don’t walk around spinning plates, Singing, begging. Only the old people because it was an old tradition. It’s all right if they forget these things because its much better not to be walking around. It’s much better to have a job - a state job or day labour. There’s no dishonour in that.”
Noni (55, gen.2,f)
“I have spun plates at the (Town X) bus stand for 14 years. I taught all my daughters as well. When I go to the fair and don't have enough money to get back, I earn a couple of rupees by boarding the buses and spinning a plate. Or when I need a cloth and jacket... At Munneswaram, we all stand in a row, spinning our plates and singing the verses, so that someone’s heart will melt towasrds us. We can spin for hours on end with the blessings of the deities. When I get back home these two fingers are swollen. What can I do about that? Our life is always hard. Five of us, all relatives - two of my daughters, my sister, my daughter-in-law and I; we spin together and divide the money among us... I think it’s a good thing if all the girls knew how to spin plates. It’s good to learn something, isn’t it? Even though you have all these films and things now, you can’t find these songs anywhere, can you?”
79

Page 46
Leelawathie (30s, gen. 3, f)
“I still spin plates if I’m at a house and they ask me to. I don't go to Munneswaram. I don't carry a plate with me - they give me a brass plate sometimes; at other times I spin a ceramic plate... This spinning is something coming down from our lineage. We used to be proud of it, but not any more; it’s going to the bottom completely. It was big for our parents but not for the new generation. It doesn't really matter to me whether the tradition disappears or not.”
Karunawathi (late 20s, gen.3, f)
“We don't like to do the things our mothers did like plate spinning. We want to live better than they did. We stay at home. It’s better to do one’s work at home. I feel ashamed to be wandering around the countryside.”
Nilakanthi (late teens, gen.3, f)
“I don't know how to spin plates. I never thought it was important to learn it. I guess it would be useful when you need to earn something, after our parents are gone. My mother brings back quite a bit of money from her spinning sometimes.”
Evidently, most women consider spinning plates and singing a caste occupation which had to be practised out of necessity and which they would like to see disappear. There are some who value it still as an income-generating activity, fewer who value it as an aesthetic traditioin that has a special meaning for them.
However, music and dance continue to play an important role in the Rodi community. All generations possess repertoires of Songs and verses and sing with pleasure. These are mostly from the dominant culture, picked up from the radio, cassettes and television, ranging from folk, semiclassical to pop music. They like improvising their own words
80

to popular tunes. As a people, they seem to possess a high degree of musicality. The young men and boys get together in a group to sing to the accompaniment of a dola (two-sided drum), soemtimes with a soloist in the lead, very often in the evenings after their trading rounds or in between working on the cane. Boys as young as 5-8 years participate in these sessions and play the drum with an excellent sense of rhythm. At a marriage ceremony where taped music was blasted through the village over a loudspeaker system, this was constantly interrupted to make room for young men, and to a lesser extent, young women to provide the musical entertainment instead.
The only dancing I observed was at the same marriage ceremony. This was of course the inevitable baila (popular free dance form peculiar to Sri Lanka, influenced by Portuguese elements). The majority of dancers were young men; however, there were also children of both sexes, old men and women, and young women among the dancers. They danced with skill, vigour and obvious pleasure under the coconut trees of the compound where the marriage feast was taking place.
Two women in their 30s married to two brothers and living in adjacent huts confided that when their husbands were away in the evenings, they would collect a group of women together, close the doors of one of the huts and dance to the music of a cassette.
There are a few families in the village where the men know how to make drums. They originate from fairly distant Rodi villages. One of them considers himself to be a farmer but was proud of his musical abilities and said he had learnt the craft of making drums from non-Rodi craftsmen near his mother's village. He sang well but both songs he sang were semi-classical pieces he had picked up from the radio.
The Rodi retain a pride in their musicality. A teenage girl and her parents were excited about the name she was
81

Page 47
making for herself as a singer in the local maha vidyalaya (senior secondary school) where she was a student. A young man in his early 20s was likewise proud that he had sung in public at one of the state-sponsored Gam Udawa (Village Reawakening) programmes in the district.
Thus, there is a continuity in their musical tradition. But the continuity is in the musical ability of the people, not in the form or content of the music. As their music and dance forms are no longer distinct (with the exception of the Ratnavalli verses and plate-spinning, which appear to be disappearing), an identity based on a music/dance tradition would probably not be retained.
(2) Crafts:
Apart from Raghavan, we have no other historical source on the craft tradition of the Rodi. This could mean that as late as the middle of this century, the Rodi were not engaged in any kind of craft, except the caste occupation of making hide and fibre nooses. Ferguson wrote in 1895 that the Rodi were prohibited not only from cultivating land but also from carrying on any trade. This could mean that they were not allowed to sell any craft products that they might have made.
If the Gadi were a tribal people, it can be assumed that they had knowledge of some crafts which were useful to them in their daily lives. But it is difficult to establish when they started practising their crafts for a market. It is interesting that their myth does not refer to any craft. This might be an indication that it was not part of their caste status to be a craftsman/woman. The change from caste beggars to craftspeople and traders probably occured sometime during the first half of this century in response to the general economic transformations in the country, leading to a market
3 Ferguson, 1895, p. 251
82

structure in which caste rights and obligations would become anachronistic.
I did not encounter any production of nooses or rope in the two villages, even though several old men informed me that they had provided nooses for the last elephant kraal in Kasipote. Cane work, making of wigs/hairpieces, broommaking were the most widespread crafts in these villages; there were also several families who were engaged in making drums, as mentioned before.
Brooms, hairpieces/wigs and drums are all made by people of generation 2 downward, so these are evidently new crafts. However, many people claimed that the cane industry was an old tradition among them and that their designs were distinct ones, passed down from generation to generation. Even though very old people whom I met did not work with cane any longer, they claimed knowledge of the craft.
Kakulumali (80, gen. l, f)
“I used to do cane work too but stopped five years ago. I can’t do any work now so I go walking around to beg...'
Utiya (86, gen. 1, m)
“Cane weaving is an old craft. People have lived on that here, not by cultivating rice.”
Whether it is actually an old tradition or not, the Rodi seem to want to make a claim of tradition for it. Above all, they recognise its income-generating potential.
Huyiya (73, gen. l, m)
“Cane work is a developed craft. There's nothing wrong with it. The prices for our goods are high. We can give our things to the store. Everything is fine except our people don't save anything.'
83.

Page 48
Ivoni (50., gen. 2, f)
“Cane work is our main means of livelihood. If we could get cane from (district Y) without these permit complications we could live better; at least our little ones... We'd like to retain the traditions that we have that will improve the lives of our children. You can earn a living with cane work.”
The only person who belittled the cane craft was the drum-maker, who considered himself, first and foremost, a farmer.
Wijeratine (60, gen.2, m)
“There is no future in cane because we can't be sure how long people would want these objects. Farming is not like that. No one can stop it. It is something you can live on. It means prosperity for the future generation.”
The Rodi do not have a steady supply of cane. As cane cutting is an illegal activity under state forest conservation laws, only allowed with special permits, the Rodi undergo a lot of hardship to obtain a limited and fluctuating supply, unlike the big cane dealers. Most people in the village took pride in their craft and were interested in developing this craft rather than rice cultivation. Even though a state-sector job with a regular income was often cited as advantageous, an identity as a craftsperson was also much desired, if a regular supply of cane could be ensured.
At present, there is a market for the items they produce in the rural areas; seed baskets, winnows and strainers are needed in agricultural households. As long as their craft products do not become obsolete (e.g. through mechanisation of agriculture), they could continue to retain an identity as craftspeople.
84

The Rodi identity: some reflections
I began this section by trying to understand what constitutes a Rodi identity. I think it is clear that the Rodi have a distinct identity. However, it is not strictly an ethnic identity. Nor is it merely a caste identity. It appears to be a mixture of a caste identity, defined hierarchically from without and internalised by the Rodi to some extent, and a tribal identity which is selfdefined with a cultural basis and carrying within it a sense of distinction and self-worth. These dual identities could constitute something akin to an ethnic identity; the tribal and caste identities being comparable to the positive and negative identities respectively, defined in page 33.
I have examined several elements that might constitute a Rodi identity. Of these; it is apparent that neither the Rodi nor the dominant culture attributes to the Rodi a hunter/ gatherer identity. If the Rodi were a hunter/gatherer group in the past, as speculated by Raghavan and others, there is no memory of such an existence within the group.
The caste occupations of begging and performing defined the identity of the Rodi in a society structured on a caste system. These occupations and the identities they engendered have been internalised to a large extent by the Rodi, but are being gradually rejected today. These have been replaced by a new craftsperson identity, which is likewise a response to social and economic transformations from without. The elements that constitute a tribal or positive identity are the self-naming of the group as Gadi, the retention ofa language of their own and the tracing of their lineage to Ratnavalli.
The underlying tension between the tribal and caste or positive and negative identities is reflected in the Ratnavalli myth (both verses and narration). At the same time, perhaps the myth acts as a “meta-identity” which explains, incorporates and rejects old and new identities of the Rodi.
85

Page 49
It is evident that the Rodi identity has a categorical component - the self-definition of the group in relation to others, their caste occupations, the current usage of their language, the changing nature of their myth all point in this direction. However, the affective component of their identity is more elusive and harder to define. The figure of Ratnavalli may provide such an unconscious identification with emotional charge attached to it. Whether this is linked to an external conception of the Rodi as despicable, unclean, inferior beings is unclear.
I have not finished grappling with the nature of the Rodi identity. It seems inadequate to explain such an identity merely in terms of a positive/negative ethnic identity. I would like to compare this identity with tribal/caste identities in the Indian context in a continuation of this study. However, I think it has been useful to examine the Rodi identity from an ethnic perspective.
86

VI
The Status of Women in the Rodi Community
Rodi women have led a life of wandering, earning their living as performers and beggars, for at least three centuries. In this century many of them survive by trading their craft products. Underlying the air of independence they display is this considerable freedom of movement which has left them with some space outside of the domestic sphere. They are not meek, submissive women. I was impressed by their bearing, their strength and their ability to articulate themselves.
The dominant culture has objectified Rodi women to a great extent; Stereotypes about their beauty, sexuality and licentiousness abound, both in popular conceptions and in Sinhala literature. This is not very different from images held by other dominant cultures of female sexuality of oppressed minority groups (e.g. the notion of the American Black woman as a sensual sex object;''' It should be mentioned that Rodi men suffer likewise from stereotyping - as possessors of evil spells and charms they use to lure away unsuspecting high caste women into their clutches. Stereotypes connected with sexuality of oppressed groups can be understood only in sociopsychological terms - i.e. as projections of the fears and fantasies of the dominant group on the despised “other'. However, it is not my intention to dwell on the stereotypes of the dominant culture in a study concerned with Rodi women within their community. I would consider this aspect important
4 Steady, 1891, pp. 13-14
87

Page 50
to discuss only if I were to discover in the future that the Rodi, particularly the women, have internalised such projections of the dominant culture.
This section will explore the status of Rodi women in terms of marriage and family structures, rites of passage, the sexual division of labour, women’s role in production and reproduction, participation in political structures, religious beliefs, gender ideology and practice, and violence against women. This research is incomplete and needs to be continued before a comprehensive analysis could be made.
By "status of women' here is meant the extent to which women participate in the social, economic, political and religious life of their community and the degree of control they exercise over their daily lives vis-a-vis their men. This study will not measure Rodi women against an ideal standard of equality that should exist between women and men, or a comparative status scale (ranging from low to high) that could be applied to women of different ethnic groups. By “labour' is meant both productive and reproductive activities; by "production” is meant all activities which women engage in to secure the economic survival of the community; by “reproduction” is meant all activities related to sexuality, child bearing, child care, house work, food preparation and consumption (which taken together facilitate “production” in the community). “Ideology” is defined in Mannheimian terms as a set of ideas and beliefs (ranging from conscious lies to half-conscious and unwitting disguises) that are distorted by the interests of a particular group to maintain the status quo, "o i.e. its privileged position of power.
I have been influenced by the theoretical framework of Schrijvers, o whose work on Sinhala women in the North
s Mannheim, 1936, pp.55-59 6 Schrijvers, 1985
88

Central province of Sri Lanka I found invaluable; however, I have modified her framework to suit the needs of my study of women of a marginalised group.
Marriage and Family Structure
The Rodi are, for the most part, a patrilineal and virilocal group. Even though Nevill” mentioned parallel clan structures for men and women, there is no evidence to show that these exist today. Nevill also wrote that polyandry was considered “honourable” but not greatly practised and “is now considered a reproach”, and that polygamy was never practised.' I found no evidence of either polyandry or polygamy. Serial monogamy appears to be the norm at present; change of partners is common among young Rodi; older Rodi seem to form longer-term relationships with their partners.
That serial monogamy has been a tradition among the Rodi for a considerable time is evident from the folllowing COmment:
Huyiya (73, gen. l., m)
Those days there were no marriage or birth certificates - nothing at all. When a woman had four or five children, that woman took another man; when a man had four or five children, that man took another woman. That’s how they spent their time... I have two children from my first marriage, five from the next one. Seven.”
It has to be remembered that serial monogamy was also predominant in the peasant society of the dominant culture, before Victorian standards of marriage and sexuality, aped by
7 Neville, 1887, p. 83 8 Ibid., p. 85
89

Page 51
the local elites and middle classes during the colonial period, came to be held as ideal. Thus, it is not clear whether serial monogamy has been borrowed from the dominant culutre and whether the Rodi had another form of marriage in the past.
Marriages are very often not registered and are devoid of ceremony. By “marriage” is meant the Rodi concept of it; a female or male moving with a partner to the hut of her/his parents or to a separate hut in the compound of her/his parents or to her/his own hut with the approval of either or both sets of parents. Most marriages are made according to the choice of the couple concerned; a small percentage is arranged. Arranged marriages tend to be celebrated with large feasts in the homes of the bride and bridegrooms and registered. The Rodi marry young - girls are between 12-18 years, and boys 15-21 years when they move to the home of their first partner. Running away with a lover before an arranged marriage and being reconciled a few days later with the respective families (i.e. of the lovers), is a common feature. Both females and males initiate relationships, courtships and marriages.
At Ranwewa, entering a family compound, one observes the main hut of the parents with the huts of the married sons cluatered around it; sometimes, one also encounters a hut of a married daughter in the same compound. Teenage couples often stay in the hut of the boy's parents, building a hut for themselves only when they are somewhat older or expecting their first child. Most young people take partners within their village or from two nearby Rodi villages. Arranged marriages, on the other hand, are often made with distant villages. I also encountered couples who had got married after forming relationships while visiting relatives in distant villages.
According to Nevill, the Rodi practised clan exogamy, daughters belonging to the mother's clan and sons to the father’s '' However, as no clans exist today, restrictions on
Ibid., p. 83
90

marriage partners are few. There appears to be a restriction on parallel cousin marriage, at least as an ideal, probably borrowed from the dominant culture; in a parallel male/female clan structure based on exogamy of the clan, there need not necessarily be a parallel cousin restriction.
The preferred form of marriage at Ranwewa was diga (virilocal); binna (uxorilocal) marriages were fewer. The two main reasons expressed for preference of the diga form were: prevention of fights between sons and son-in-law (i.e. if the latter quarreled and/or beat his wife, her brothers would defend her against her husband) and lack of property to be given to daughter in order to bring her husband into the family. There was also a reluctance to take a son-in-law into the house because of a fear that he would take control of the household/ property. Together with the preference for the diga form was expressed a desire that the daughter should be married as far away as possible so that she would not be able to come home after a fight. Fights seem to be accepted as an inevitable part of marriage. Women who themselves were aware of the hardships of a diga marriage through experience continued to maintain it as an ideal. The contradictory notions held about diga by women are reflected below:
Kakulumali (80, gen. 1. f)
“Some people here go binna but most people go diga. In binna, the man has more disadvantages; in diga, the woman has more disadvantages from her mother-in-law. If there’s a daughter, one should give her away, not keep her in a binna situation. If the son-in-law hits the daughter, her brothers are not going to look away. So there’ll be fights between the brothers and brother-in-law. If she gets hit at her in-laws we can always bring her back-we used to those days. Not now. We don't break the marriage. We talk and advise them and send her back.'
91

Page 52
Noni (55, gen.2, f)
“We mostly give our daughters in diga. There's power to the daughter then from the man’s side. Once she's married and gets the name she has power there. Even if she gets a child and the man asks her to leave, she doesn’t have to go. Even if he goes, she has a right to stay in that house... If it's binna, the man has the right over that house, even if the woman leaves. The daughter has a place if she goes diga. If we take a man binna, he has control over us. We can't tell him to go unless he himself wants to... It’s better if the daughters go far away. All my children are around except for one daughter, who's also only a little further away. If we give daughters far away from us, they can't come home after a quarrel.”
Leelawathi (30s, gen. 3, f)
“My daughter won’t get any land. We'll give her money and presents when she gets married. I don’t have a preference over giving her to the village or outside. It all depends on what kind of man she gets. We won't give her in binna because we don't have dowry for her. To give binna we have to give her with a piece of land. It's better to give a daughter indiga. When you take a son-in-lal w into the house, you don't know when there's going to be a fight... Of course there can be a fight over at his place too.”
Observe the concern with land and its relationship to marriage in the third excerpt. In the traditional overcrowded village of Managala there is no land to give, and the problem is to find enough space to erect yet another hut in the compound. However, in Ranwewa where land has been distributed to each Rodi family recently with legal right over it, the issue of inheritance is already a matter of concern, and sons rather than daughters are being considered as heirs.
The preference for giving daughters in diga to distant villages and the justification for it is always expressed by
92

married women. Note that both Noni and Leelawathi are married to partners from nearby villages, and Noni has six of her eight children married within the village. Unmarried girls attempt to resist moving away from the village of their parents. Kanthi (18) told me that she would not want to get married to someone from another village because she wanted to stay near her parents. Six months later a marriage had been arranged for her and she had to move to a distant village; at that point, she said that she was sad but she would go because it was a good marriage. At her wedding, her cousin Suneetha (13) told me that she would never consent to a marriage that would take her so far away; she expressed fears that if something happened to Kanthi over there, none of her kinsfolk would be able to do anything.
It appears that the ideology of the diga marriage and the move to a distant village is imposed on the younger women by the older married women. Girls do not voice their opposition to the diga form per se, but rather to the move away from their parental village. Most would eventually marry within the village or move to nearby villages. But once married, they begin to express their preference in terms of the ideal rather than in terms of the actual practice.
Moreover, married daughters often come back home, with or without their husbands, for short or extended periods; each parental hut or compound has at least one married daughter, who had been given away in diga. The usual reason given for return with a husband is a quarrel with his family; without a husband, violence on the part of the husband and/or desertion. A daughter is generally accepted back by her parents.
Noni (55, gen.2, f)
“We do take our daughters back when they quarrel. They
leave and come home. Or I go and bring them. It happens
frequently. Then they go back. They move back and forth.
93

Page 53
Some men won't take them back if they leave; then they'd go on another marriage. If a man is constantly drunk and beating your daughter, you can’t blame her for separating from him. But even then, we go and give them advice and try to keep them together. We tell him not to drink, fight, keep bad company. If that doesn’t work, we accept our daughter back. How can she live amidst so much trouble?'
So mi (40s, gen. 2, f)
“There are many drunken husbands around. Women even go to the Police to complain. But the attempt is to unite the parties again. Only some women leave their men for good. Usually parents will take a daughter back when it is impossible for her to live with the man.”
Thus, even though there is family pressure to keep a marriage intact, it appears not that difficult for a woman to get out of an unhappy marriage and return to her parental home. If she has several children, she would eventually build herself a hut in the parental compound. It is also relatively easy for a woman to find a new partner; there is ro real Stigma attached to her as a woman separated from her former husband.
Women who have returned home after an unhappy marriage are critical of men in general, and marriage in particular. Both young women quoted below had been married in nearby Rodi villages. They both had a daughter each, of four years and one year respectively.
Nanda (20s, gen.3, f)
“I got married four years ago. I left my man a year ago because he drank too much. I don't want to go back to him. My parents and family are more important to me than him. I'm not going to burden myself with a marriage again. One can’t expect anything from these men.”
94

Nilakanthi (late teens, gen.3, f)
“I left my husband because he drank and smoked too much ganja. I don't want to go back. My mother accepted me back. I have no intention of taking another man., It's better to live by myself. When my child is bigger, I'm going on a job in the city, may be work in a house. The men here are useless. You have to support them; they don't support you... only a few do.'
There has been a traditionof ostracism of women who bore "illegitimate” children, i.e. where the family/community was unaware of the identity of the father. This ostracism was sometimes extended to the entire family of the woman and was removed only by a ruling of the Hulavaliya (see also page 34).
Kakulumali (80, gen. 1, f)
“The Hulavaliya used to decide when to take a woman who had a child without a father back into the community. It could take 10-12 years, until everyone forgets about it. It's an old tradition. Our people don’t even know about it now. These things happened more than 30 years ago. Such a person was ostracised by the community; she was not permitted to go to any shrine; nobody would want to be seen beside her.”
Leelawathi (30s, gen.3, f)
“Some people still follow the old kulabath (caste rice) tradition. This is how it is. If my daughter gets friendly with someone and runs away without telling me, some people would not invite us to their weddings and other ceremonies. If there's a wedding somewhere and they decide to invite us, they would let our daughter join the wedding table only after paying a fine. Some demand one or two bottles of arrack; others take money. They don't do that in this village, but it happens in Managala and Wel lawa.”
95

Page 54
I did not encounter such ostracism in Ran wewa. Considering the number of runaway daughters, it would mean that all the families would be ostracised at some time or other. But bearing a child without an acknowledgement of paternity by the father brings about the censure of the community.
Noni (55, gen.2, f) − “I’m very much against what my son has done. My daughter-in-law has been chased away. She's gone back home. My son stays in both places. When she was at home, she did a thing like that, got pregnant. It was then that mny son took that on his shoulder. I don't approve of that. She stayed about seven months with us though.”
Noni is implying that her daughter-in-law had got pregnant by an unknown man and had induced her son to marry her. When this was discovered, she had chased the young woman away. However, the son continued to have a relationship with his wife.
It is evident that there is a great deal of mobility among houses and among villages, with household composition fluctuating from time to time. In such a context, one could expect to find a significant number of female-headed households. A rough estimate, made with the help of my informants, is that nearly a third of the original main huts (15 out of 49) are female-headed. If daughters who have returned home and built a hut for themselves and their children in the parental compound, as well as women who have husbands working in a distant town, are added, the figure should be higher.
Thus, it appears that despite many oppressive elements in the ideals of marriage and family structures (which might be borrowed, to some extent, from the dominant culture) held among the Rodi, the actual practice is more flexible and benevolent. A Rodi girl is raised with the notion that she
96

would be "given” to others and would have to leave her childhood home. However, the natal home provides a basis of emotional and material support throughout her life. One can only wonder what kind of legacy the past clan structure where daughters belonged to the mothers clan could have left on the present realities of marriage and family.
Rites of Passage
The salient feature about the rites of passage of Rodi women is their essential simplicity. More elaborate rites, when practised, appear to be borrowed from the dominant culture. Rodi women consider the events that engender the rites, especially first menstruation and marriage, as significant and look forward to the ceremonies and feasts that accompany such events. There is neither solemnity nor secrecy attached to these rites. Males are excluded from certain parts of the rites related to first menstruation and marriage, but are otherwise participants in women's rites of passage. Some women, in the context of a past tradition of simplicity and lack of financial resources, adopt a matter-of-fact attitude towards these events; others, concerned with prestige value attached to customs borrowed from the middle classes of the dominant culture, desire to celebrate such events in a similarly elaborate fashion.
Birth
The child is delivered with the help of a traditional midwife in the village. It is bathed, clothed and placed in a cradle made out of a piece of cloth and a rope hung from a rafter. Some infants have a bit of sugar and milk rubbed on their lips (a custom called Rankiri kataganawa which is found in the dominant culture as well). Children are named whenever the parents come up with a good name. There is no differentiation
97

Page 55
in birth customs or the care of a girl child from that of a boy child.
First Menstruation
The simplest version of the first menstruation ritual in the Rodi community is to bathe the girl immediately and confine her to a room in the hut for a few days; at this time no male may see her or have any other contact with her. The more complicated ritual is similiar to that of the dominant culture. The girl is confined to a room and an astrologer is consulted on the auspicious hour to bathe her; this is done by two paternal aunts. A coconut is broken as she steps back into her house. A party where she receives presents from friends and relatives may follow. The event is usually celebrated according to the financial situation of the family. Families with more resources can not only afford to put up a feast but can also do without the daughter's contribution to the family economy for a longer period. The confinement of the girl is based on the belief that menstrual blood is potent and dangerous for the male members of the family.
Leelawathi (30s, gen. 3,f)
“The people who have money prolong the confinement, print “at-home” cards, bring loudspeakers and so on and celebrate in a big way. Almost everybody confines the girl for a few days in a room... Because it’s not good for the father and brothers; it’s all right for the mother.”
Marriage
The marriage ritual of the past consisted of the bride being taken away from her home to the home of the bridegroom by relatives on both sides; this was accompanied by feasting in both houses. No dowry was given.
98

Kakulumali (80, gen. 1. f)
“I was brought to the village of my man, wearing a Patalensuwa (kerchief) round my neck, a pataredda (cloth) round my waist and some bangles. They broke a coconut. I took off the clothes I was wearing and was given a new set sof clothes. There were no other ceremonies at that time.”
Leelawathi (30s, gen. 3, f)
“In those days, say the boy wants to marry the girl next door. There's no registration or anything. The parents talk it over, prepare food and bring the girl over from next door to the boy's house.”
Ridigira (58,gen.2,m)
“We never give dowries for marriage. No land is given even now. But Rs. 400-500 might be given to the bridegroom by the bride's parents these days...The bridegroom's party get together, take presents, go to the bride's, eat magul bath (marriage feast) there, bring back the bride, and there's a celebration at his home. Sometimes the bridegroom's party come back the next day, sometimes two or three days later; there's no limit on the time.”
The presents that were given by the family of the bridegroom to the bride’s family, as well as clothes for the bride (a custom also followed in the dominant culture together with the institution of dowry) might date from a time when brideprice was the norm among the Rodi,
One marriage that took place during my research period but unfortunately while I was not in the village, was between 17-year old Sena and 14-year old Sriyani. However, Sena, who was a son of the family with whom I lodged, related to me what had happened. He had been friends with Sriyani for a while; she had complained several times of problems at home, about her parents always picking on her, and had wanted to
99

Page 56
live with him. One evening he had gone over to her house and told her that she could come with him. Her parents were not at home and he had brought her to the house of his parents. He had explained that she did not want to live at home any longer. His parents had accepted her into the house. The couple was considered married from then on. Sena maintained that his parents had not reproached him in any way, and that there was no ill-will between the couple and either set of parents. Sriyani when I met her turned out to be a high-spirited young woman, who was shy to talk about her marriage but appeared quite happy about the move.
From my conversations with young people, it seems that this kind of marriage is the commonest in their community. This could be accompanied by feasting, depending on the family's resources, and if the parents had been informed beforehand.
Sena's sister Kanthi (18), on the other hand, was married off to a youth (19) from a distant Rodi village with more elaborate ritual. This was an arranged marriage negotiated by a mutual relative. Kanthi saw her prospective husband once when he came with his family to visit the bride's family, and consented to the marriage. I was present at both the ceremony and the discussions about preparations for it among the family earlier.
A month before the wedding. Kanthi's entire family in the village, including her brothers, sisters-in-law, married sisters and brothers-in-law met to discuss arrangements for the ceremony. The slaughtering of a pig, printing of invitations, studio photographs of the bridal couple and hiring of a vehicle to take the bridal party to the bridegroom’s village were the main items on the agenda. Kanthi herself was present and consulted. Her mother and father made the final decisions and delegated tasks to the younger members.
The ceremony itself which took place a month later was not very different from a rural wedding of the dominant
100

culture. The afternoon before, the bride was dressed by her sisters in a saree (she usually wore a blouse and a skirt or cloth) and kept in the inner room of the house where she could receive only her female friends and relatives. She was beginning to weep at the thought of leaving her home. Her friends tried their best to comfort her.
The bridgegroom's party arrived at an auspicious hour in the night and included his brother, brothers-in-law and sisters-in-law, as well as several friends. They all sat down to dinner at the wedding table with the bride, her parents, siblings and in-laws (as many as the table could accomodate) seated on the opposite side. After the dinner, the bride's friends and family members from the village came to the house. The bride herself was taken back to her room. There was music through a louspeaker system, dancing and drinking. The dancers were mainly young men and children of both sexes. One older woman who was very drunk and appeared at the compound partially clad also joined the dancers and was a source of amusement to everybody. A young woman in her twenties also danced for a while but came out when a brawl started between a Rodi youth and a drunken non-Rodi police constable in civil clothes who had been invited by one of the brothers-in-law of the bride. After the brawl was over and the policeman was escorted out, several older women admonished the young woman for dancing in the presence of non-Rodi males. The dancing stopped after the brawl and many guests left. An older woman relative who had been helping with the cooking expressed her regret that the women had not been able to dance that night. They had been talking about sellamkaranawa (playing) while they had been cooking and had planned on having some fun that night, she told me.
The bridegroom and his party slept at the bride's brother's house (the “best” in the family as it was made of brick rather than mud) next door. The next morning after a breakfast of traditional sweets at the wedding table, the bride
101

Page 57
was dressed in a saree brought by the bridegroom’s family and dressed by his sisters-in-law. Then the couple, accompanied by the male relatives of both, went to the nearby town to get officially registered. Meanwhile, the cooking for the main feast at midday was being done by the women; there were some men around cutting the meat and washing dishes. Guests from three nearby Rodi villages as well as two Kandyan villages began arriving. Presents of money were given to the bride's parents from the previous evening onward, and were collected by the bride's brother, who kept written accounts since the amounts would be reciprocated at forthcoming celebrations at the homes of the guests.
After returning from the registration, the bride and bridegroom sat on the same side of the table for the first time to partake of the marriage feast. The table was constantly filled with food and new rounds of guests were fed while the bride returned to her inner room to change her clothes once again. There was dancing going on in the compound, this time with more young women participating. The bride started weeping bitterly as the hour for her departure drew nearer. At an auspicious hour the bridegroom was permitted to enter the room and hold one end of the white cloth (that would be used to test her virginity) while his sisters-in-law wound the cloth round her waist. Then the bridegreom left and she was dressed in her proper wedding attire of white and gold, and decked with earrings and necklace, all brought by the bridegroom's family.
Then she and the bridegroom were escorted to a small table in the yard where two teenage relatives (female) sang verses (which had been written by an adult beforehand) welcoming their new “brother” to the village, entrusting their “sister” to his care, appealing to both to tolerate each other's weaknesses, bestowing blessings on both of them. Then came formal speeches - from the bride's family, apologising for shortcomings in their hospitality; from the bridegroom's side
102

thanking them for the hospitality received; from the guests who acted as representatives for the village they came from, giving advice on marital life as well as congratulating the bride's family for putting on such a fine wedding ceremony, which was an honour to “our people”. Then the bride began weeping again. this time joined by her female relatives. The bride and bridegroom knelt at the feet of her parents and older siblings, bidding farewell. The parents accompanied her to the hired van, into which both the bridegroom’s party and the bride's party (consisting of sisters-in-law, brothers-in-law and cousins) got in and set out to the home of the bridegroom, a 4-5 hour drive, after which more feasting and the virginity ritual would follow.
There are several significant features about this marriage ceremony. From the speeches and general comments heard at the wedding, I gathered that such a ceremony was a rare occurrence and that much prestige was attached to a family who could put on such an event. Missing from this wedding was the poruwa ceremony (where the bridal pair stand on a platform while their thumbs are tied together and water poured over them by the bride's uncle, thus pronouncing them united), which is in effect the essential ritual of the traditional Sinhala Buddhist wedding.
The bride's immediate family, with the exception of the bride herself. were dressed in their daily clothes (covered with soot from cooking, etc.) throughout the event, though they do possess “fancier” clothes. The bridegroom's party, on the other hand, were dressed in festive attire during both days; the bride's party were similarly attired when they accompanied her to the bridegroom's village. Thus, a clear separation between family and guests was maintained throughout.
No dowry was given. The bride carried with her a small suitcase, containing a few items of clothing and toilet articles. All her clothing and jewellery for the ceremony had been
103

Page 58
brought by the bridegroom. The money presents that were received from the guests went to meet the cost of the wedding and were not given to the couple.
During the time the bridegroom's party was in the bride's house, his sisters-in-law constantly accompanied her and dressed her up for the ceremonies. Her own mother, sisters and friends were nearby but everything was under the management of these sisters-in-law. Thus, it appears that in this transition period of being a "bride”, a young woman leaving her natal home for that of her husband, her future roles of being a wife, daughter-in-law, Sister-in-law in a strange house are being introduced to and reinforced on her.
It is noteworthy that while the Rodi leave out the poruwa ceremony of the dominant culture, they have borrowed the virginity ritual from it - the white cloth brought by the groom and wrapped round the bride's waist symbolically in his presence. From accounts I have heard so far, such a virginity ritual was not part of the Rodi tradition.
Finally, the night before the wedding could have been a time for fun and frolic for women of the family (i.e. Lucia's reference to “playing'), perhaps even for the bride herself, in the past; this, however, is changing with the presence of nonRodi men at their ceremonies.
Motherhood
There are no special rites or taboos connected with motherhood. After the birth of a child, the woman rests on her mat for a few days in order to overcome her exhaustion. Her husband and family members (both female nad male) are allowed to visit her, following the delivery. Cooking is done during this time by her female relatives or the husband. Children are breast-fed by the mother. There is no notion of pollution connected with birth.
104

Death
There is no differentiation between women and men in rituals connected with death. These have been simple in the past, but like marriage rites, are getting more elaborate at present.
Kakulumali (80, gen. 1, f)
“We took a bamboo pole, hung a mat on it, placed the body and tied it with three ropes. Two people carried the pole on their shoulders. There were no coffins. We went to our village graveyard. We buried our dead people. We never cremated them. Even now. I don't have to lie about it. We never cremated. Then we gave food to the people who came for the funeral. Now they bring coffins, they dress...even a small child is not taken away without a coffin.”
One man told me that three stones are laid on the grave as a protection against demons, and to prevent the deceased from harassing relatives or friends; another that a king coconut and a coconut flower are laid on the grave. Some families also follow Buddhist customs, i.e. call the monks for the funeral and give alms after three days.
Cremation is done only when people can afford to buy firewood, and therefore carries prestige with it. Poor people (which include both Rodi and non-Rodi) bury their dead. The graveyards/cemeteries of this region are caste-segregated.
Sexual Division of Labour
If the Gadi had been a hunter/gatherer group, the division of labour in such an existence cannot be ascertained today. As has been mentioned before, the boys and youth engage in small-scale hunting to supplement their diet. Gathering, which is mostly confined to greens and lotus seeds in the tank, is usually done by women and girls. The small vegetable patches
105

Page 59
in their compounds and animals (cows, pigs and poultry) are tended by both sexes.
The caste occupation of begging is practised by both men and women. They do their begging rounds separately: a man usually begs alone; women go alone, in pairs or in a small group. The women sing and spin plates; the men merely carry a pingo (pole with a basket or bag attached to it to collect the rice) which, together with their caste attire, identify them as Rodi. The men no longer play their traditional one-sided drum, but playing the dole (two-sided drum) in the village is a male activity.
Rice cultivating is done only by a handful of families. Both sexes participate in the activities connected with it. Making hairpieces and wigs is an essentially female occupation, mothers and daughters sharing the work among them. Broom-making is done by both sexes. Drum-making is a male activity.
The major subsistence activity in the community is cane work, which consists of three processes - cutting and cleaning the cane, weaving and selling. The first is always done by males. Young men go to the jungles of the North-Central province, on journeys that can take several weeks to obtain the cane, often illegally. Older men accompany their sons so that they can cook for them. The party of men first make offerings (money, flowers) at a shrine of Malalabandara deiyo. (one of the deities in the syncretic Sinhala Buddhist pantheon) not far from their village, requesting his permission to make the journey and his protection to return safely home with the cane. Once they reach their destination, they make similiar offerings to Pulleri deiyo (another name for Ganesh, the elephant god of the Hindu pantheon). Perhaps these cane cutting/cleaning trips with the rituals, dangers of the jungle and sense of adventure attached to them, act as a substitute for hunting trips of the past. At any rate, it is considered men's,
O6

rather than women’s, work to undergo the dangers and hardships of the cane-cutting expedition.
Weaving and selling are done by both sexes. However, weaving of the mouth of a paddy-sowing basket is considered a male task by many, even though there are some women who do this task just as well. In the selling of their products, there are some interesting generational differences. Both women and men in the age range 50-60 sell their crafts. But in the age range 30-50, men stay at home, making the objects, while their women go out to sell. In the age range 12-30, the males do the selling and the females stay at home, making the objects. In the 30-50 age range, several men said that they did not go out selling because they were ashamed to walk around in the countryside. They did not want to be identified as Rodi in the neighbouring villages. In the 13-30 age range, the women offered a similar explanation for not going out to sell. In addition, they considered that it was disreputable for them as women to be wandering around, away from their homes.
However, young women who had no husbands (because of separation or death), or mothers to go out selling for them, did trade on their own, staying several days or weeks away from the village, as they had no family responsibilities or could ask a relative to look after the children. They have regular rounds with houses to lodge in the night on these trading trips. The houses are of non-Rodi people with whom they have formed long-standing relationships. Some of these houses would be known to the young woman through her mother who had done the same round before; others she would find herself by making her own contacts. The great majority of women do not go on such long trading rounds; they usually go to the neighbouring villages, returning home at nightfall.
Household chores like cleaning, cooking, fetching water, gathering firewood are generally considered women’s activities, very often done by the teenage daughters of the
107

Page 60
family. Men do participate in them, usually when the women are not around to do them. Several women told me that a Son did chores on a regular basis; others that their husbands cooked if they came home late from their trading round.
Noni (55, gen.2, f)
“I do cane work and trading. My son does most of the house work. I als o help fetch water, collect fire wood sometimes. My son cooks most of the time. I don't like to cook. I never did it very well. He does it better than me... I feel lazy about cooking.”
So mi (40s, gen. 2, f)
“I have five boys and a little girl of 9 years who goes to
school. My third son does all the housework. I cook only if
I’m at home. Most of the time I go on the job.”
Leelawathi (30s, gen.3, f) V
“When I go trading and come back late my husband has usually already cooked and fed the children; otherwise, I do the cooking. My daughter fetches water and collects firewood.”
Childcare is done by both women and men. Infants are usually taken care of by the mother, as far as feeding, cleaning and dressing are concerned. However, fathers entertain and play with them. Toddlers are often looked after by older siblings of both sexes. Once children are able to walk, talk and feed themselves, they roam around in little groups, keeping themselves occupied and taking care of themselves.
Thus, in this community, one cannot draw a public domain for men and a private domain for women. Women’s lives span both domains, even though the younger women who are married or dependents of their parents, tend to engage in
108

a major part of the household chores. Men also span both domains, particularly because a considerable number of them stay at home making cane objects, while their wives go out selliling. There is also a variability along generation lines on which sex was occupying which domain at a given time.
One of the important factors for staying in the village or going out trading is prestige, which works in different ways. For the over 50 generation brought up in a caste-intact world, to go out begging was a matter of fulfilling their caste role; if they could now trade instead, it was a major improvement in their status. However, men in the 30-50 age range grew up in an environment where caste ism was being officially proclaimed as a violation of equality but nevertheless being practised. For them to be identified as Rodi was stigmatising and they preferred to stay at home, where they did not have to deal with the non-Rodi. For the women of this age range, on the other hand, the need to feed their families outweighed the cost of being stigmatised as Rodi. For the younger men who are living in a time where caste status is no longer as important as one's income, it is a matter of prestige that they are traders and income-earners, able to support their women who are "housewives”. At the same time, the younger women want to stay home because they consider it disreputable as women to be roaming the neighbourhood.
However, such age-range categories in the sexual division of production and sale of cane objects cannot be considered “watertight” as there are exceptions within the age ranges. Women and men of all ages may go trading because of individual attitudes toward prestige or in response to economic circumstances. For example, many young men cannot fulfill their own and wives' expectations of being the "provider' (alcoholism is a major reason), and young women would either have to go out selling their products themselves or be dependent on a female relative who does so.
109

Page 61
What is significant about the sexual division of labour within the Rodi community is that it appears to be not so rigid and is changing in response to economic factors and gender ideology from without.
Women’s Role in Production
All women in this community, apart from the very old, are engaged in production of some kind - making and selling cane objects, wigs/hairpieces and brooms. Often they make a living from two of these crafts, and additionally by tending a few animals (selling milk, eggs) or working on the family rice field in some cases. Most are craftswomen, many craftswsomen and traders, and some merely traders. Whether one activity or another was being undertaken could also depend on the mood or the immediate financial needs of the individual person. Thus, a woman can on one day take her own products to sell, another day stay at home making the objects and giving what she has already made to some other person to sell, and on yet another day take only the products of other people to sell. Karunawathi (late 20s), for example, would always give the objects she made to other people to sell. As her husband was an attendant at the local hospital and the couple had a regular income with no children, she would also act as a middlewoman, buying objects that other people had made and giving them to a third person to go out and sell.
They are all income-earners, independent of the income of their men. Even younger married women who are making cane objects at home would not always give these to their husbands to sell, but also to other relatives or friends, so that the money would come into their hands rather than their husbands. Producing and selling are two different activities, and the profit made on the sale is always the seller's even when the producer might have been the husband of the seller, or vice versa. So women and men maintain separate incomes, and it is the income of the woman that is crucial to the family.
110

Noni (55, gen.2, f)
“I do cane work and trading... It's about 18 years since I lost my man. The only child I have left at home is my youngest son. My daughter lost her man too, so she’s home until she builds a hut next to my place. We all do cane work. We don’t have any other way of survival. It’s difficult to live on it. We go around the countryside with a winnow or two; there are times when we sell these, and times we don’t. I'm the head of my household. All my 8 children survived because of me. It's not easy to carry such a burden... I have to even support the son I have at home. He and I eat together. My daughter and her four children manage for themselves.”
Leelawathi (30s, gen.3, f)
“I trade in both hairpieces and cane goods. I’m the one who goes out; he stays at home and makes cane objects. I make these things as well as trade them. His mother also takes some of his things to sell. I don't take everything - we give our goods to other people too. I get the money back after a thing is sold. I doesn’t matter how much I sell a winnow for - Rs.50 or 60. I have to bring him back Rs.35. Of course, the money I get is for household expenses. It's the same thing if we give to an outsider -that person has to bring us back Rs.35 for a winnow, Rs.40 for a food cover, Rs.5 for a strainer, Rs.35 for a seed basket... It's from this income that we send our children to school. They get books from school; but we give them clothes, other necessities. My man keeps his money for his own use. Sometimes he buys me something, like a cloth.”
Cane goods are generally sold for cash, but tometimes bartered. The condition of the barter is one full seed basket of rice for the basket itself. This is more profitable to the Rodi than getting cash for the basket. They bring the rice home or sell it on the market, depending on whether their immediate
111

Page 62
need is to have rice for their meal or cash to buy some other goods.
Women work all day and a good part of the night. I have not yet systematically measured the time they work, on both productive and reproductive tasks. Therefore, I cannot make a quantitative comparison with the workday of Rodi men. However, I have observed women continuing to work, while the men are watching television, drinking or singing and dancing elsewhere. At the same time, women may sometimes take time off work and take a nap when they do not feel like working. They may also gather at a house where the men are away, on the pretext of working together, but dance and chat instead. However, it seems to be not legitimate for women to be not working, while it is taken for granted for men.
Daughters are often reprimanded by fathers for procrastinating their household chores, especially cooking; but older women appear to have more flexibility with their time. Members of a household generally wake up around the same time; space restrictions and the noise level make it difficult for any member to continue sleeping when others are awake. Different households work on different schedules. I have observed households where women and men are up by six o'clock in the morning, others where both sexes are asleep till ten o'clock. But there is usually a corresponding difference in the times they go to bed. Some households work till two in the morning on their cane, by lamplight. The early risers consider the late risers as lazy and lacking pilivelak (a sense of order) in their lives.
Most Rodi women work for the day-to-day subsistence of their families.
Noni (55, gen.2, f)
"I go wandering around the countryside to sell a winnow
or strainer, sometimes three or four days at a stretch. Then when
we are short of money to eat, I go again. That's how it goes.”
12

There are also some women who have a concept of saving and maintain their own savings books at the local bank or post-office. However they do not always succeed in retaining control of their savings when confronted by their husbands. Only with ingenuity can they circumvent such confrontations.
Leelawathi (30s, gen.3, f)
“I had two savings books but my man found out and demanded the money. He wanted to drink with it. He threatened to beat me up so I had to get the money out of the books. Now I play sittu at the store. That mudalali (store owner) plays fair. The other one near the tank cheated me. And my sister-in-law too. With the sittu money I always buy something for the house. A metal water pot once. Then I did a bit of repair on the house. Another time I bought a cabinet...The sittu is done secretly and there's no proof like with the savings book. So my man never knows.”
Sittu is widespread in the rural areas and functions as a sort of savings scheme of the poor, under the control of a storeowner. Several participants visit the store-owner in a group or individually and deposit sums of money at intervals, and each participant receives an accumulated lump sum, corresponding to the deposits, over a period of time. Both store-owners that Leelawathi referred to are non-Rodi. Both men and women in the Rodi community participate in the sittu system.
Alcoholism of the men was often voiced by the women as the main reason for the inability to save.
Noni (55, gen.2, f)
“There are men in our village who drink and there are those who don't. There are lots of problems because of the drinking. It takes away money from the home. When the man comes home, there are quarrels with the woman. The women
13

Page 63
tell the men to save something. The men don't listen to you these days if you tell them not to drink. So you quarrel and don’t do anything much about it. People leave home, come back. There are lots of problems like that in the homes.”
It is evident that women have to constantly struggle not only to get some money out of the men for the household, but also to retain their own subsistence incomes. Sometimes, like Leelawathi, they find ingenious solutions to this problem, but most of the time the outcome of their struggles varies from day to day. Sometimes women manage to retain their share of the production for the family; at other times they do not. One reason for male-female conflict in the community, then, is the acquisition and control of their means of subsistence.
Women's Role in Reproduction
Motherhood is not an explicitly glorified concept among the Rodi. But its importance is implicit as children are valued as contributors to the family economy as well as a source of security in old age.
Ridigira (58, gen. 2, m)
“They (the non-Rodi) have lands, paddy, cattle...they
have the five wealths. We don’t have any of these. The only
thing we have is our child wealth. That's all we have”
Ridivilliya (60s, gen.2, m)
“We don't have wealth. We don't want it because it can
be taken away by thieves. We want more children; they’re our
wealth in both good times and bad times.”
Such expectations, of course, have to be fulfilled by the women, who are the producers of children. Rodi families tend to be large; 5-10 children in generations 1, 2 and 3, and 2-4 in
114

generation 4. Both younger men and women voice the view, perhaps influenced by the state family planning programme, that it is better to have fewer children. I met several young women and men who had got themselves sterilized after their second or third child. Sterilization is the only form of birth control that I encountered in the community. I avoided questions on attitudes towards birth control, who makes the decisions, and sexuality in general, because I did not wish to be suspected as a state family planning researcher in disguise at this stage of my research. However, several individuals volunteered the information that they had got themselves sterilized. One young woman, who was separated from her husband after their first child, had undergone the operation with her mother's consent. She wanted to work as a housemaid in the city after her child was big enough to be left with her mother.
Neelakanthi (late teens, gen.3,f)
“I got myself sterilized because I don't want another child. My brother-in-law, he comes to see me now. He was furious when he found out. He scolded me a lot. But it doesn't matter to me.”
Her ex-brother-in-law was courting her after her sister had left him for another man, and was angry about the sterilization. But she was not perturbed by his censure, and confided that she had no intention of marrying him.
Women's sexuality does not appear to be rigidly controlled in this community. Both men and women frequently leave their partners for other women and men, particularly among the young. There is a certain degree of tolerance and/ or resignation about such changes of partners. I have often heard comments of the sort, “That marriage didn’t work out. The other one didn't either. Now she's gone on another one.” However, a woman who changes partners very often is not
115

Page 64
respected within the community. It is common for a woman to have an "extra-marital” relationship for a time before it is resolved either way - i.e. she stays with the original man or leaves him for the new lover. All such transition periods, however, are accompanied by a lot of violence (with women usually at the receiving end) within the families concerned, and censure and gossip within the community. Thus, the ideals connected with women's sexuality and marriage are more rigid than the actual practice.
If a marriage fails to produce offspring, this is attributed to the infertility of the woman. However, there are no sanctions against infertile women. Infertility is considered an act of fate and the woman may be pitied; men do not appear to separate from infertile women on that ground. I met two sisters who had stayed married to the same men for 5-10 years without conceiving children.
Noni (55, gen.2, f)
“People don't do anything to a childless mother. There's nothing in her life; there's no point if there’s no child, isn't it? People don't do anything to her though. If she has no child wealth what can she do? The woman in the new house novs, she's been married for 10 years without a child. One can't do anything about that."
Rodi women carry the burden of bearing, and, to a large extent, nurturing children within their community. Men assist to some degree in child care, but the major share of work is done by women, especially young women. Taking care of infants requires a lot of time and effort by the women. However, once children start walking, older siblings would take over the task of watching them. Rodi children are independent and learn very early to look after themselves. The mother usually provides them with meals early morning and late evening. However, during the day the children often fend for themselves; they might
116

eat at a relative's house where a midday meal had been cooked or pick a king coconut off someone's tree, sell it and buy something to eat with that money. If the children go to school (less than half do), it is the mother who sees to it that they are supplied with school materials. Infants are bathed by both sexes. Children from about 3 years onward bathe and swim in the tank with no adult supervision. No large-scale washing of laundry is done by the Rodi; women, men and children (when they are old enough) wash their own clothes when they go for their daily baths to the tank.
Cooking, as has been mentioned before, is mostly done by the younger women. In the households where I was invited to eat, men and children were served first, and the women last. However, I, as a guest (even though female) was served before the family. As far as I could observe, there was enough food to go around for everybody. Rodi women did not seem to me to be particularly undernourished compared to their men, nor female children over male children. However, this would have to be established by further field work. Food items appear to be procured with the incomes of the women, and supplemented by the hunting/gathering activities of the children. But some men, particularly the younger ones, contribute to the food supplies of a household.
Other household chores like cleaning, fetching water and firewood are also done by the younger women and girls. Older women, on the other hand, neither expressed the desire nor saw the need to tie themselves to housework and children. They were primarily craftswomen and traders. However, an ideology of staying at home is prevalent among the young women, even though they bear the brunt of dealing with these chores.
Neelakanthi (late teens, gen.3, f)
“My mother goes on the job. I stay at home and do my
work. It's much better that way. I don’t want to be wandering
around the countryside selling things.”
117

Page 65
Kumari, who lives in the same house as her mother and is married with two children, considers her husband (even though he drinks regularly and contributes almost nothing to the household) as the “provider”. She disagrees with her brothers that she should get rid of her husband.
Kumari (early 20s, gen.3,f)
“I told my brother, Gune (the husband) can go, except he has to support me instead. How can bring up two children on my own, if he's asked to get out?”
This notion of being a “housewife' and being provided for by the husband probably comes from the dominant culture which promotes this as an ideal in the media (especially advertising in radio and TV). Older women, on the other hand, recognised for the most part that they were the providers of their families, instrumental in earning an income, as well as nurturing their children.
Whether Rodi women themselves recognise the role they play in production and reproduction, it is clear that they provide the basis for the economic and biological survival of their community.
Women's participation in political structures
There are no autonomous political institutions that exist within the community today. The traditional posts of Hula valiya and Gas manda were occupied by males in the past. The Hulavaliya (who was the chief of the village) and in his absence, the Gasmanda had the function of resolving conflicts within the village. These posts, which functioned similiarly to the Arachchi and Vidane of the peasant village of the past, were under the control of the Disawe (chief of province) within the Kandyan feudal system.
118

Kakulumali (80, gen. 1, f)
“The Hulavaliya and Gasmanda were appointed by Nugavela Disahamuduruwo as chiefs of the village. They were asked to settle disputes. They called the disputing parties, heard their differences, decided who was in the wrong and fined the offender. Rs.3.50 for beating up someone, for example. We didn't go to the courts or the police. These people solved everything. They beat up offenders - just the way the police does now...The Hula valiva used to decide when to take a woman who had a child without a father back into the community...”
Among the disputes settled by the Hula valiya were those related to marriage, illegitimate pregnancy and physical violence. Thus, matters concerning women were resolved through this male institution. Today the police have taken over Some of these functions. I would be discussing how the Women use the police as a means to punish their men later in this paper (see Violence Against Women).
The Rodi participate as voters in the electoral system of the country. Both women and men vote at elections, but I do not know whether the women cast their votes according to the wishes of the men. During my stay in the village. I did not hear women discuss politics or express partisan views, as the men did. However, women do have a knowledge of government policies that affect their daily lives and social seervices offered by the state. For example, all the women whom I met (including grandmothers) were aware of when, where and how to obtain food stamps. They also made use of the state health service in the area regularly.
There are two voluntary organizations that exist within the village, both led by a non-Rodi male from a nearby village. Membership in the Funeral Assistance Society consists of both men and women. I have not seen a single meeting of the Village Development Society yet, and I doubt whether this organization functions at all. At any rate, the impact of these organizations on the community is minimal.
119

Page 66
Thus, neither women nor men participate nor hold power in political institutions within the community. Both sexes have no access to power and decision-making within the political structure at national level.
Religious Reliefs and Practice
There is no evidence today that the Rodi practise any forms of belief specific to their community, such as the worship of the goddess Ratnavalli. They have incorporated the entire gamut of the syncretic tradition of the dominant culture including the Buddha, the gods and goddesses of the Hindu pantheon, various local and regional folk deities and demons, and the exorcism/healing rituals.
There is no Buddhist temple in either village, and there is no indication that Buddhism is practised on a regular basis. Many people said that they went to a temple in the neighbourhood on important holidays like Sinhala New Year, Vesak' and a few other full-moon days. Some families provide aims to the same temple. Buddhist temples in the neighbourhood were barred to the Rodi for a long time because of their outcast status. Thus, to be able to enter and worship at a temple or to give alms to the monks (who were mainly from the Goigama caste and rejected offerings made by the Rodi) is considered a sign of improved status. Many Rodi of both sexes worship at the Siva temple at Munneswaram.
20 Vesak is the full-mooon day in May, commemorated as the day of birth, enlightenment and passing into Nirvana of the Buddha by the Sinhalese.
2 To take sil (also called atas ill) is to observe for a whole day the eight precepts (refrain from killing, Stealing, sensual lust, lying, intoxication, eating after midday, reclining on luxurious seats and enjoying the pleasures of the senses such as music, dance, etc.) in Buddhism.
120

On the walls of their huts are images and pictures of the Buddha, as well as the deities of the Hindu pantheon. Younger women did not engage much in practices connected with Buddhism, except to visit the temple on an important holiday. Older women visit the temple more often, take si l'o" sometimes, and join on rare occasions pilgrimages organised by non-Rodi neighbours.
However, what is most important to both women and men in the community are the various local and regional folk deities, who are propitiated to ward off evil or to cure a sick family member or, in the case of women, to cure infertility. There is one male kattadiya (hea ler) in the village of Ranwewa. He is a specialist in Bali, a ritual widespread in rural areas in which a set of planetary deities are propitiated against evil influences. I encountered no female healers. A healer (male) from outside was once summonded to bind a thread and chant over an old woman who had a broken arm; this was a very simple ritual taking no more than half an hour. A popular deity/demon of the Vanni region is Kadavara, who is propitiated with elaborate ritual to bring prosperity to the village and to ward off evil influences. The Rodi follow this tradition too; however, such a ritual never took place during my stay at Ranwewa so I am unable to offer a description or to understand the role of Rodi women in the propitiation of Kadavara, a male deity. Many women light a little lamp in their compounds at dusk, seeking the protection of the gamdevathawa (village deity).
Ranmuthu (80s, gen. l., f)
“I like going to temple. But I haven't gone more than
three times this year. I take pansil' every evening. We also
worship the deities and go to Munneswaram sometimes.”
' Pansil: five precepts in Buddhism, excluding the last three
mentioned for atasil.
121

Page 67
Kakulumali (80, gen. 1, f)
“We worship the gamede vathawa, Kadavarade iyo and go to Munneswaram. We also go to Buddhist temples now. I even take sil. But not in the old days. We couldn't even go near a temple.”
Noni (55, gen.2, f)
“We are Buddhists so we go to temple. Usually on important Poya' days like Vesak. We also go on pilgrimages. I've gone on and off in buses organised by other villages... It’s good to do some pindaham (good deeds) for this world and the other world. We also go to Munneswaram to earn a few rupees by spinning; we always worship at the shrine before doing that...Every evening I light the lamp - there are blessings from that.'
The syncretic religious tradition of the island plays an important role in the life of Rodi women as well. They find some solace in it but do not derive any power from it.
Gender Ideology and Practice
In the high-born ancestress, goddess Ratnavalli, one might expect a strong female figure of identity for Rodi women. Or on the other hand, as the woman who committed the “sin’ and was thus responsible for the stigma, she could provide a weapon for the men to be used against their women. However, I did not encounter the Rodi, either men or women, using the figure of Ratnavalli ideologically. Ratnavalli appears to remain an ambiguous, even enigmatic, figure for the Rodi.
If female power related to a “mother' goddess has been replaced by male power through the incorporation into
23 Poya: the days of the four phases of the moon, considered
sacred by Buddhists.
22

patriarchal Buddhism, the corresponding gender ideologies and the process of transformation from one to the other are unclear.
However, there is no doubt that the gender ideology of the dominant culture, or at least, elements of it, prevail within the Rodi community. Kotuwa who had five married daughters, of whom at least one lives with her husband and children in Kotuwa’s compound, regretted that he had no sons.
Kotuwa (50s. gen. 2, m)
“Only sons are our own. The daughters have been raised for the country, to give away to other families. You can’t depend on daughters.”
Despite the fact that his daughter had not been actually "given away', and was within the compound of his own family, he continued to voice an ideology of daughters not being part of the family as sons are, and daughters being not dependable compared with Sons.
During a dispute over water usage for paddy cultivation between two families, the wife of one of the men involved joined the fray and began to shout at the other party. While the man was insulted for using his connection with a specific political party to get his own way in the matter, his wife was insulted for having a big mouth. Comments from both sexes. included “Look at that woman's mouth”, “Is that how a woman should speak?', and from a man, “It’s women who put men into trouble.”
Similar ideology was revealed in another dispute over a sale of a piece of land. Peter (m) and Karuna, his wife, had given money to her cousin Kumari (f) to acquire part of the land belonging to Kumari’s family to build a hut for themselves. Kumari’s brother, who was away from the village, returned and said that the land transaction was invalid because Peter and Karuna had paid too little. The money,
123

Page 68
meanwhile, had been spent on a cassette recorder by Kumari's husband (who is considered the main drunkard of the village). Much of the blame was placed on Kumari and her husband, Gune, for being so foolish as to sell the land. An older male relative of both Kumari and Karuna who was present during the dispute, used the proverb “A woman’s brain is no longer than a spoon handle” to refer to Kumari’s act of selling the land, and added, “The trouble is with women's heads.” Wimalaratna, a younger brother of Karuna, scolded her for letting Peter hand over the money to Kumari’s irresponsible husband, telling her that she should have known to advise her own husband, Peter, better, and that “women should be able to advise their men properly”. The mother of Karuna and Wimalaratna, Dinga, intervened then and said, “This should be a lesson to women; men are always creating problems', referring to the action of Peter handing over the money to Gune, who immediately goes and buys a cassette recorder, as well as Peter and Gune negotiating an unreasonably low price for a piece of land when Gune was drunk.
Thus, it is evident how quickly actions or behaviour, considered inappropriate or foolish, of individual women is extended in terms of the generalised category of “women” by both men and women. Such conflict situations reveal some aspects of the gender ideology present in the community, mostly expressed by men: women should not articulate themselves; women are responsible for men's problems; women cannot be expected to use their brains; it is women's responsibility to give proper advice to their men. Within such an ideology, women not only possess many shortcomings but are also responsible for the shortcomings of men.
However, Rodi women do not always accept being defined in terms of this kind of ideology. Dinga's advice to her daughter to learn from her experience and recognise problems that can be created by men indicates that there is some resistance by women to viricentric ideology. It remains to be discovered whether such resistance is widespread and
124

whether women possess an autonomous definition of womanhood that may be tapped by them.
Violence against Women
Violence is widespread in the two Rodi villages, both by men against women, and by men agaisnt men. Alcoholism is one of the immediate causes. But the roots of violence in this community can be understood only in terms of the stigmatisation and marginalisation that the Rodi have suffered from without for centuries. Alcoholism is an easy escape from the degradation of the past and the poverty of the present. However, Rodi women, unlike their men, have retained a strength to struggle and survive, not succumbing to alcoholism as easily as the men. It is the women, taking most of the responsibility for their families on their shoulders, who ensure that at least some of their subsistence incomes are retained to feed their children. It is significant that old women who no longer have any family obligations also indulge in drinking.
The commonest manifestations of violence are beating of women and destruction of dwellings. Here, I present three case histories of violence involving women.
Case 1
Kumari (20s, gen.3, f) is married with two small children and lives with her husband, Gune, in the house of her mother. Two of her brothers (one is married) and her sister who left her husband also live in the same house. One night Gune came home drunk and began beating Kumari. Her brothers intervened, and so did her sister, Nanda, who brought an axe to hit him. Kumari’s family asked Gune to leave. He moved to his younger brother's house nearby. Next morning the whole family went to the Police station. Kumari, her mother, brothers
125

Page 69
and her little son all testified against Gune. Gune admitted to the Police that he had got drunk and beaten Kumari. Kumari asked the Police not to beat him or remand him but to let him go, and to impress on him that he should not bother her or her family again. Kumaris brothers did not want Gune back in the house. But she herself did. She told her brothers that she was willing to let Gune go as long as they would support her instead, since she could not raise two children by herself. In the evening Guine was back in the house, and I saw him carry his infant daughter through the village, demonstrating his paternal rights over the child.
Case 2
Sriyani (20s, gen.3, f) lived with her parents and had three small children from three men. Her latest lover was a nonRodiman from a nearby village whose wife was in the Middle East. He did not live with Sriyani but visited her frequently in the nights; they also went on outings together. Sriyani was the subject of much gossip and disapproval on account of her “secret lover', as he was referred to by the community. One night Sriyani's younger brother Nimal came home drunk, waving a pole in his hand. Sriyani disappeared and hid herself in the scrub jungle behind the house. Nimal occupied himself for about two hours destroying the hut, hitting his parents and Screaming continuously, “I’m going to kill you, I'm going to eat you.” He then went away to a friend’s place. The next morning the hut was in shambles; Sriyani went trading and made herself scarce the whole day. Neighbours told me that Nimal disapproved of Sriyani's “secret lover” and always came home threatening her when he got drunk. He was also pressuring their parents to throw her out, but they would not do that.
126

Case 3.
Nilanthi (late teens, gen.3, f) had returned home to her parents because her husband Dasa drank too much. Nilanthi had two small children living with her. On the day new food stamps were being issued, Nilanthi collected stamps for herself as well as the children. Food stamps are considered easily convertible currency within the community, and Dasa came over to Nilanthi to demand the food stamps from her. He began a fight with her and hit her father who came to protect the daughter. Nilanthis old grandmother fainted. Her relatives in the neighbourhood intervened. Dasa took his seven-month old infant and disappeared. Meanwhile, Nilanthi's mother who had been out returned home and accompanied her daughter to the Police station; they made an entry, accusing Dasa of kidnapping the child. In the same evening Dasa returned with the baby, whom he could not handle because it was being breast-fed. Both Nilanthi and her mother refused to take the baby back, except in the presence of the Police. Dasa took the infant to a neighbouring house, fed it some tea and then disappeared with the baby. (I do not know how this incident ended as I left the village the next morning).
These cases illustrate the fact that men use violence against women
(a) merely because the men are drunk;
(b) to express anger when women do not adhere to sexual
norms held as ideal by them and/or the community:
(c) to obtain money or other means of subsistence from
WOCI).
However, these are not the only grounds on which men beat women. An older woman who had three married sons said that
127

Page 70
the two younger sons never beat their wives, but the older one did because his wife had “too big a mouth” and answered back if he scolded her. A young woman told me that her sister had been so badly beaten by the latter's husband for taking another man (non-Rodi) that she had to be hospitalised. She said that she didn’t blame her brother-in-law because “which man wouldn’t be angry when his woman goes with another man'. Thus, retorting to a man and jealousy on the part of men could also lead to violence against the women concerned. Women who are victims receive support from relatives or friends when the man is believed to have no ground to beat his wife except his own alcoholism or when the violence is connected with a struggle for the means of subsistence, i.e. when men attempt to take money away from women. However, violence related to female sexuality (whether over another partner by a jealous man, or over a “secret lover” by an enraged brother) is justified by both men and women, and the victim does not receive much sympathy.
Women are in a better position to resist violence against them when they live in their parents’ compounds or when their parents live in the village and are within easy reach. Mothers often provide support to daughters and encourage them to seek the assistance of the police to punish the offender, as well as to arrive at some sort of settlement to the conflict.
Leelawathi (30s, gen.3, f)
“Women go to the Police when men get drunk and beat them, because our men are frightened of the Police. So this is one way we can get them scared. The Police often beat them up and remand them for a night. It keeps them quiet for a while.'
Thus, wife-beating is not always accepted by the women as part of their lot as wives. They want some sort of justice to operate, so that the men receive punishment, when they think
28

that the men are in the wrong. They use the Police as a weapon to threaten their men. It is not clear why the Police (who are of course, mostly male) punish Rodi men for what they probably do to their own wives. Perhaps the low status of the Rodi in the caste system has something to do with it.
It is evident that resorting to an outside agency such as the Police provides only short-term relief to the women concerned. In the long term, the violence against that woman is likely to continue. Separation is effective when the man and woman come from different villages, but as Case 3 illustrates, separation is no guarantee for the end of violence when the man lives in the same village as the woman.
It is significant to examine the violence against women in terms of the ideology that women should marry far away from the village, particularly as all three cases I have outlined dealt with women who were living in their parents' houses. It seems that even though parents would rather have their daughters “out of sight, out of mind”, when they actually have to deal with a situation of the daughter being confronted by violence, they do offer their support to the daughter. However, instead of identifying the cause of the trouble in the violence of the males, they associate it with the victim - the daughter who has not been married far enough away from the village. Thus, the cycle of violence continues.
However, despite the ideology, Rodi women are able to resist violence from men to some extent by mobilizing the support of their natal families, especially the mother. But they receive very little or no support at all if the violence is connected with the expression of their sexuality.
The Status of Women: Some Reflections
Rodi women live in a patrilineal, patrilocal community defined by a gender ideology hostile to them. However, the norms and ideals are not so rigid that they cannot be flouted in practice.
129

Page 71
The discrepancy between the ideology and its practice reveals the success of the women in circumventing some of the constraints of their culture.
We have seen that women have space within their marriage and family structures to determine their partners, to change them and to return to their natal homes when a marriage fails. Thus, Rodi children can, in effect, be brought up in a matrilocal family, rather than a patrilocal one. Women are full participants in the economy of the community, maintaining an economic independence from their men. Their lives span both public and private domains, even though the “stay-athome” ideology prevalent among the young women can change this in the future. Women are also able with the help of their natal families to resist to some extent the violence directed against them by men. Moreover, they often seek the assistance of an outside agency to punish the perpetrators of the violence. Their rites of passage relating to birth, motherhood and death include no taboos or other discriminating features against them. However, the more elaborate rites of first menstruation and marriage (probably borrowed from the dominant culture) comprise several oppressive elements such as the confinement based on a notion of pollution in the former, and the virginity ritual in the latter.
Rodi women are restricted in the expression of their sexuality outside of their marriage and family structures, by the disapproval expressed by the community in general, and the violence of the men in particular. However, these factors do not necessarily stop them from actualising their wishes and desires.
The struggle of the women to retain control over their subsistence incomes as well as the expression of their sexuality reveals the very real conflict that exists between Rodi women and men. It is a visible conflict, relating to both productive and reproductive spheres, and accompanied by violence. The extent of the violence in this conflict shows the presence of a
130

considerable and continuous resistance by women; however, women resist individually rather than collectively, so that no long-term improvement in their status occurs.
The strength of the Rodi women lies in the persistence of their struggle to retain control over their day-to-day lives. These women cannot be so easily subdued by their men; they
are in many ways still the daughters of Ratnavalli.
131

Page 72
V
Conclusion
In the preceding sections I have emphasised the marginalisation of the Rodi as a group from the dominant culture and focussed on their community from within. in order to understand the specific experience and identity of a group in itself. However, this marginalisation does not mean that the Rodi have been isolated from the main currents of political, social. economic and media transformations occuring at a national level in Sri Lanka.
The marginalisation of the Rodi is, in fact, at a “microlevel - in the social intercourse between the Rodi and the non-Rodi in their own neighbourhoods. When the Rodi are not in their caste attire, it is almost impossible to distinguish a Rodi from a non-Rodi. However, within the neighbourhood, there has been a history of separation between Rodi and nonRodi, based on an identification by familiarity. Generations 1 and 2 still carry memories of being restricted to wear their caste attire, being refused entrance to temples, being served tea in coconut shells (disposable and thus non-polluting to other customers) at local tea shops, being insulted and beaten up by their peers if they attended school, being confronted by violence because they had dared to hire a vehicle for one of their weddings. Such a heritage of suffering and stigma has been passed down both directly and indirectly to generation 3, who themselves confronted more subtle forms of hostility and discrimination, particularly in schools, as a result of which most did not pursue their education. Some of these veiled forms of discrimination, even though more subdued than before, continue to operate for generations 4 and 5. Non-Rodi neighbours may exchange greetings and news with the Rodi
32

when they meet on the street, but very few would exchange visits or eat with them.
There are several Rodi/non-Rodi marriages in Ranwewa. The Rodi themselves have no objection to such marriages. However, the non-Rodi partner has forsaken most, if not all, ties to her/his natal home following such a marriage.
At the same time, the hostility that the Rodi face in Ranwewa and Managala (and the Vanni region in general) is less acute than in the Kandyan areas, where young women may still hide inside their homes when Rodi men pass by (fearing that they would be “charmed away”), and Rodi villages are popularly conceptualised as dens of thievery, prostitutioin and alcoholism, exerting a bad influence on the entire neighbourhood.
At the “macro-level' of the nation state on the other hand, changes contributing to the assimilation of the Rodi have taken place, particularly since Independence in 1948. Politically, the concept of equality ("one man, one vote") within parliamentary democracy led to both policy making and social welfare measures that sought to integrate the Rodi into the mainstream of Sinhala society. In the early 1950s, a Backward Communities Development Board' (of which Raghavan was an advisor), appointed to make recommendations to the government on several “backward” groups, visited many of the Rodi villages, reporting about the conditions under which the Rodi lived. One consequence of their work was the passing of the Prevention of Social Disabilities Act, No.21 of 1957, imposing penalties on extreme forms of caste discrimination such as restricting the Rodi to their caste attire. Several Rodi villages were also extended or newly created, on land given to them with legal titles under the Land Development Ordinance, during that decade.'
24 Raghavan, 1957, pp. 6, 38, 42, 73, 95 2S Ibid., pp. 13, 94, 96-98
133

Page 73
The sanction to discard their caste attire and the receipt of land allotments are identified by the Rodi with the figure of former Prime Minister S.W.R.D. Bandaranaike (1956-1959). Since then his Sri Lanka Freedom Party (SLFP) appears to have received a block vote from the Rodi community, confirming the caste basis of electoral politics in Sri Lanka, referred to by writers such as Ryan. However, this phenomenon seems to be changing, as the populist image of Prime Minister R. Premadasa of the United National Party (UNP) has captured the imagination of some Rodi, particularly the young, in Ranwewa. The left parties have made no impact among the Rodi in these two villages.
Social welfare measures have come both from the state and a quasi-Buddhist community development organisation, from the 50s up to the early 80s in an ad-hoc fashion. Socalled “development” projects such as a library and cane centre in Managala and a coir workshop and pre-school (both nonexistent by now) in Ranwewa, instituted without proper consultation with the Rodi themselves, have left no lasting impression on them, except pessimism and a lack of confidence in organisations and individuals who want to “develop them. However, these welfare efforts have brought about a consciousness among the Rodi decrying the legitimacy of the caste system and demanding equal recognitiion from the nonRodi. National welfare programmes such as food stamps and health services are utilised by the Rodi; however, they are not dependent on these programmes. Food stamps merely provide a supplementary income; traditiional healers within and without their community are consulted just as often as state health personnel.
Concern about the education of the Rodi was already expressed in 1892 by the Public Instruction Department of the colonial government, and a school was opened for the Rodi
26 Ryan, 1953, pp. 275-282
134

in Wellandora in the Ratnapura district. This was a failure, as was the efforts by Christian missionaries elsewhere. In 1904, the colonial government appointed a commission to report on the measures to be taken to initiate education of the Rodi. “” The commission recommended sex-segregated schools, with space for agriculture and crafts, within Rodi villages, making concessions to the casteism of a society which could not conceptua lise Rodi children atten ding neighbourhood schools with high-caste children. However, none of the se recommendations se em to have been implemented. Both Ranwewa and Managala have co-ed elementary schools within the village; some elementary-level and all secondary-level pupils attend the central school of the area. In Ranwewa, there is only one youth who has had 10 years of schooling and passed the GCE (O/L). In July 1986, there were 91 children from Ranwewa attending school; this was less than half of the school-age children in the community. Of these 91, more than two-thirds were in kindergarten through Grade 2. Grade 4 had less than half of the number of children in kindergarten (21). While boys outnumbered girls in kindergarten-Grade 2, girls reach parity with boys in grades 3 to 5. Only a handful of Ranwewa children (8) are receiving an education beyond elementary level (grades 6-9); however, when they do they are almost likely to be girls (7). Thus, the overall education level and the motivation for it are low within the Rodi community. The parents value the income-earning potential of the children more than sending them to school. The teachers lack any understanding of the special needs of the community and motivation to give a proper education to Rodi children, whom they are likely to describe as lacking discipline and “retaining power” and prone to the use of “bad
27 Raghavan, 1957, pp.70-71 2 Ibid., p.71 .69-73.Ibid., pp 29ا
135

Page 74
language”. However, even a limited immersion in the state education system contributes to the incorporation of the ideology and the value system of the dominant Sinhala Buddhist culture.
Economically, the market has had a significant impact on the Rodi. Caste-based occupations, based on a system of mutual obligations and rights, are easily eroded by a market economy based on a cash nexus. The Rodi, who were entitled to rice in exchange for services within the caste system, now have to buy rice in the market or barter their baskets for the rice. Thus, both rice and baskets have become commodities with value, mediated by the cash nexus. It is significant that the Rodi who still go begging would wear their caste attire while those who go trading wear regular clothes.
A minority of Rodi men in the villages have been absorbed into wage labour occupations within the state sector, and are employed as hospital attendants and sanitary cleaners, as well as in other menial positions. In the Kandyan areas more Rodi, both men and women, have found work in the public sector. It would be useful to examine whether presentday occupations, like electoral politics, have a caste basis. Regular incomes from such jobs are imaking an impact on the individuals and families concerned, giving them a sense of security and increasing their material resources in comparison with other members of the community who do not hold jobs. The market, particularly in the 80s in response to the economic policies of the present government, has provided the Rodi with consumer goods in the form of radios, cassette recorders, television sets, watches, synthetic clothing and plastic jewellery which have become status symbols and are much sought after, just as by people in the dominant culture. The market, then, has played a vital role in bestowing prestige and status on those who are capable of acquiring such goods, and the traditional ascriptive hierarchy of status and privilege has become less relevant, i.e. the caste identity is being
136

replaced by a "consumer” identity, particularly in a group like the Rodi, who remain for the most part marginalised from the class structure.
Media, in the form of picture papers, radio and television, have a profound influence on the Rodi. Weekly picture papers such as Siththara and Sethsiri are circulated within the village, particularly among the young. The radio is turned on to music programmes most of the day in their homes, providing entertainment to the whole family. Pre-recorded cassettes of popular music are cherished by both young and old. Even though the number of TV sets in the village is limited, most people have access to this medium through relatives and friends. The television is operated with batteries (as there is no electricity in the village) and turned on whenever Rupa vahini (the first channel of the state programme and the only one received in the village) is telecast; they watch anyth ing that appears on the screen, although Sinhala programmes (soaps and musical shows) are preferred. Children also like action programmes in English such as “Tarzan'. Television is mostly a medium of the men and children, as it usually coincides with the cooking time of the women. But younger women often attempt to get their cooking finished in time for their favourite programmes. All aspects of mass culture promoted by media, whether popular songs, film stars, television serials, cartoon strips or advertising images are absorbed by the Rodi and form a frame of reference for them in their daily lives.
The Rodi are aware and concerned about the SinhalaTamil ethnic war in the country, but do not have a grasp of the details. They identify themselves on the Sinhala side but maintain some distance to the conflict, as nobody from the community is directly involved. They do not encounter Tamil people in their daily lives because their village is located in a predominantly Sinhala area, with pockets of Muslims. The conflict has not affected their relations with the non-Rodi,
37

Page 75
i.e. the “Sinhaleseness” of the Rodi appears not to be under suspicion. a'
Thus, all the trends which I have outlined so far, seem to point to the assimilation of the Rodi into the dominant Sinhala Buddhist culture. The caste system may persist at a social level, particularly in the immediate neighbourhoods of the Rodi, imposing a Rodi identity on the group from without. However, the Rodi themselves are rejecting this identity, even though individuals who have knowledge of their traditions may continue to value some of these. The problem for the group is to separate their positive Gadi identity from the negative Rodi identity. Therefore, a revival of a positive Gadi identity from within the group seems unlikely. In the absence of a national effort to promote a multi-cultural society which recognises and respects the distinctness of its cultural groups, it is difficult to foresee the persistence of the Gadi as a distinct group with a sense of pride about themselves and their ways.
138

Bibliography
Armour, John, 1842:
Grammar of Kandyan Law. In The Ceylon Miscellany, Colombo.
Barnett, L.D., 1917:
Guide to Sinhalese Folklore, London
Bechert, Heinz, 1978:
The Beginnings of Buddhist Histriography: Mahawamsa
and Political Thinking. In Religion and Legitimation of Power
in Sri Lanka, ed. Bardwell Smith; pp.1-12. Chambersberg.
Bechert, Heinz, 1978:
Contradictions in Sinhalese Buddhism. In Religion and Legitimation of Power in Sri Lanka, ed. Smith; p. 188-198. Chambersberg.
Bougle, C., 1971:
Essays on the Caste System. Cambridge.
Carmody, Denise, 1981:
Women and World Religion. Nashville.
Clifford, Regina, 1978:
The Dhammadipa Tradition of Sri Lanka: Three Models within the Sinhalese Chronicles. In Religion and Legitimation of Power in Sri Lanka, ed. Bardwell Smith; pp. 36-47. Chambersberg.
Dalton, E.T., 1973:
Tribal History of Eastern India. New Delhi.
139

Page 76
D'Oyly, John, 1975:
A Sketch of the Constitution of the Kandyan Kingdom, Colombo.
Dumont, Louis, l972:
Home Hierarchicus. An Essay on the Caste System. London.
Eliade, Mircea, 1975:
Myths, Rites and Symbols - A Mircea Eliade Reader 1, ed. Wendell Beane and William Doty, New York.
Epstein, A.L., 1978: -
Ethos and Identity. Three Studies in Ethnicity, London.
Exem, A. Van, 1982:
The Religious System of the Muda Tribe, St. Augustin.
Ferguson, A.M., 1895:
The Rodiyas of Ceylon, In Monthly Literary Register, 3, (ll and 12): 247-25 l; 285-288. Colombo.
Geiger, Wilhelm, 1897:
Die Sprache der Rodiyas auf Ceylon. Berlin.
Geiger, Wilhelm, 1950:
Mahawamsa, Colombo.
Geiger, Wilhelm, 1929: Culawamsa, 1. London
1930: Culawamsa, 2. London
Gooneti like, Jane, 1890:
Sinha lese Folklore: The King of the Outcasts. In Orientalist, 4 (1 and 2): 30-31. London.
140

Gunawardena, R.A.L.H. 1978:
The Kinsmen of the Buddha: Myth as Political Charter in the Ancient and Early Medieval Kingdoms of Sri Lanka. In Religion and Legitimation of Power in Sri Lanka, ed. Bardwell Smith; pp. 96- 106. Chamabersberg.
Hayley, F.A., 1923:
A Treatise on the Laws and Customs of the Sinhalese Colombo.
Knox, Robert, 1984:
An Historical Relation of the Island Ceylon in the East Indies. New Delhi.
Leach, Edmund, 1967: ed.
The Structural Study of Myth and Totemism. London.
Levi-Strauss, C., 1955:
The Structrual Study of Myth. In Journal of American Folklore, 68: 428-444. Philadelphia.
Mannheim, Karl, 1936:
Ideology and Utopia. New York.
Nevill, Hugh, 1887:
The Gadi or Rodi race in Ceylon. In Taprobanian, 203): 81-96; 204); 108-121, Colombo.
Obeyesekere, Gananath, 1978:
Gajabahu and Gajabahu Synchronism: An Inquiry into the Relationship between Myth and History. In Religion and Legitimation of Power in Sri Lanka. ed. Bardwell Smith; pp. 155-176, Chambersberg.
141

Page 77
Raghavan, M.D., 1951:
Cultural Anthropology of the Rodiyas (Ethnological Survey of Ceylon, No. 1) In Spolia Zeylonica, 26(1): 77-116. Colombo.
Raghavan, M.D., 1951:
Some Aspects of Rodiya and Kinnaraya tribal cultures. In New Lanka, 204): 59-62. Colombo.
Raghavan, M.D., 1957:
Handsome Beggars. The Rodiya of Ceylon, Colombo.
Ryan, Bryce, 1953:
Caste in Modern Ceylon. The Sinhalese System in Transition. New Brunswick.
Schrijvers, Joke, 1985:
Mothers for Life, Motherhood and Marginalisation in the North Central Province of Sri Lanka. Leiden.
Shorter, A., 1974:
East African Societies, London.
Smith, Bardwell, 1978:
The ideal Social Order as Portrayed in the Chronicles
of Ceylon. In Religion and Legitimation of Power in Sri Lanka,
ed. B. Smith; pp.48-72. Chambersberg.
Smith, Bardwell, 1978:
Kingship, the Sangha., and the Process of Legitimation in Anuradhapura, Ceylon: An Interpretive Essay. In Religion and Legitimation of Power in Sri Lanka, ed. Smith; pp. 73-95. Chambersberg.
142

Page 78


Page 79
Nireka Weeratunge, who study for the International Ce mainly on field work in a Rod of Sri Lanka. Through the ora community regarding their considerable light on the proba people. Her study points to the as they appear to have orig hunting tribe with a distinct eth in a marginalised Way into the most interesting and illuminati is study consists of the differ origin the myth of the godde recorded in the Community W throw light not only on variou by which the Rodi were trans also on the changing perce regarding their identity and particular attention in her stud Rodi community, bringing out are affected by the patriarch dominant culture, Rodi wome anc Contro Overtheirives daughters of Ratnavali.
This edition is a reprint of
 

carried out this anthropological ntre for Ethnic Studies, based it village in the Kurunegala District traditions and the myths of this Origins she is able to throw ble iStoricaeVolution Off the ROC conclusion that the Rodi (or Gadi, nally called themselves) were a nic identity who were incorporated Sinhala caste system. Some of the
g material in Nireka Weeratunge ent versions of the Rodi myth of SSprincess Ratnavali which she
ere. She Worked. These versions S aspects of the Social processes ormed from a tribe to a caste but ptions of the contemporary Rodi SatUS. The aUOr haS devoted y to the position of Women in the he fact that even though their lives all structures and ideology of the en retain a certain independence "they are in many ways still the
he first, published in 1988.
ISBN 9.SS-S3-66-9