Ilankai Tamil Sangam

18th Year on the Web

Association of Tamils of Sri Lanka in the USA

The Sinhalese of Ceylon and The Aryan Theory

Letters of a Tamil father to his son – Section 4

by Samuel Livingstone

Let me make a resume of the story of the origin of the Sinhalese people which the Sinhalese want us to believe. It is as follows. A certain Vijaya, whose grandfather is a lion, is driven out of his country (somewhere in North India) by his father and accompanied by 700 followers comes to Ceylon. He and his followers call themselves Sinhalese because of this origin. On arrival, they find the island occupied by a race of demons called the Yakkas and the Nagas. They settle down near Puttalam and, after driving out these demons to the mountains or to the see or otherwise exterminating them altogether, marry Tamil wives and bring Tamil skilled workmen and thereafter colonise the whole of Ceylon...

The above is the version of the history of early Ceylon in a nutshell as propounded by the Sinhalese historians. Now if this version is true, it must be supported by or fit into all the facts of history already known to and accepted by us as true. In other words it must be verified and proved beyond any shadow of doubt in the light of our existing knowledge of men and things. Until it is thus verified or proved, it is only theory or a hypothesis.

Section 1

Section 2

Section 3

Front Note by Sachi Sri Kantha

This constitutes the fourth and final section of Samuel Livingstone’s 1971 book on the orign of  Sinhalese in Ceylon. After the publication of Samuel Livingstone’s book, quite a few research papers by non-Sri Lankan authors had appeared in peer-reviewed journals. I cite three of them here, and provide scans of their first pages.

Saha N: Blood genetic markers in Sri Lankan populations – Reappraisal of the legend of Prince Vijaya. American Journal of Physical Anthropology, 1988; 76: 217-225.

Coningham R and Lewer N: The Vijayan colonization and the archaeology of identity in Sri Lanka. Antiquity, 2000; 74: 707-712.

Strathern A: The Vijaya origin myth of Sri Lanka and the strangeness of kingship. Past and Present, 2009; no.203: 3-28.

Saha 198 Blood Genetic Markers in Sri Lankan Populations Reappraisal of the Legend of Prince VijayaUnfortunately, none of these reports cite Samuel Livingstone’s research or his book. So much for the contemporary scholarship on ancient Sri Lankan history by non-Sri Lankan authors! At least one (N. Saha in 1988) had confirmed one of the seven pillars of the ‘alternate hypothesis’ proposed by Samuel Livingstone in 1971 that, “genetic evidence linking the legendary origin of the Sinhalese population to East India (Prince Vijaya) is lacking.” 

In letters 25 to 30, Samuel Livingstone reiterates the message: “In spite of all the evidence, if the Sinhalese still persist in saying that Sinhalese is an Aryan language, or an Indo-Aryan language, as they prefer to call it, just for the honor of it, they are at liberty to do so. They may thereby be able to hoodwink some Europeans who do not know Tamil, but certainly not those who know both the Sinhalese and the Tamil languages. Theirs was a Dravidian language and continues to be so. ”

He proposed an ‘alternate hypothesis’, the pillars of which are as follows:

  • No Vijaya came to Ceylon at any time, nor any Aryans from any part of India.
  • The present Sinhalese population of Ceylon are the direct descendants of the Yakkas and the Nagas, who were Dravidians, occupying Ceylon in ancient times and of a few immigrants who might have come from India from time to time.
  • The Yakkas and the Nagas, who did not become Buddhist or who after becoming Buddhist became Saivites again, spoke the Tamil language throughout and the present Ceylon Tamils are their descendants.
  • The Sinhalese language of today is the product of evolution from Elu, the early Dravidian dialect spoken in Ceylon, when the latter came into contact with Pali.
  • The word ‘Sinhalam’ was derived from the Tamil word ‘Sri Ilam’, which became in course of time Sihalam and later Sinhalam.
  • The knowledge of the scientific system of irrigation found in Ceylon was not imported from abroad but was already in the possession of the people of the island from very early times.
  • The civilization of Ceylon, except for the introduction of Buddhism, is a branch of the old Dravidian civilization of India in almost all its aspects.

As I have indicated previously, for convenience and stylistic reasons, I have revised the settings of footnotes (wherever they appear), at the end of each letter, in a numbered sequence. Footnotes 21 to 23 appear between Letters 25 and 30. The dots wherever they appear in the letters are, as in the original.

Letter 25

The Legend of Vijaya

My dear Son,

I think we have wasted our time with the Aryans of Greece and Rome and even India, owing to the indiscretions of Dr.Mendis and the imperfections of Professor Pakeman. I have also spent much of your time in taking up cudgels on behalf of the indigenous Dravidians, whose contribution to the civilization of the world, nobody seems to know or cares to know. However, let us now come to Mother Lanka and examine once more the period prior to the introduction of Buddhism, in order only to find out whether the so-called Aryans ever came to Ceylon. The history of Ceylon really commences with the reign of Devanambiya Tissa and the conversion of Ceylon to Buddhist faith, and one of the event preceding this is the legend of Vijaya’s arrival to Ceylon. Whether this legend is purely mythical or has a kernel of truth in it must be carefully examined. For this purpose, I think we should consult Mr. Codrington, whose ‘Short History of Ceylon’ is an excellent book, for he clearly shows signs, unlike Dr. Mendis and Prof. Pakeman, of being an impartial historian without any axes to grind. He himself doubts the history of Vijaya and raises the following points to indicate its improbability. (I am quoting his words):-

“There is no mention of the Kuweni legend in the oldest chronicle, the Dipawansa.”

“If there is any truth in the account of Vijaya’s ancestry at all, it is difficult to admit the probability of any connection between the petty kings of Bengal and Gujarat on the opposite sides of the Indian continent. (Subdadevi, the grandmother of Vijaya flees from Bengal to Gujarat, a distance of over 2,000 miles. Her son Sinhabahu, father of Vijaya, runs back to Bengal where he slays his father, the lion, and returns to Gujarat).”

“Vijaya is made to land in Ceylon on the very day of Buddha’s death, manifestly due to the desire to connect the early history of Ceylon with that of Buddha.”

“Panduvasa, Vijaya’s successor, is made to marry a Sakya princess, again doubtless due to the desire to connect the royal family of Ceylon with the race of Buddha himself” (for as you know the Buddha belonged to kshatriya clan called the Sakyas)”.

In regard to all these points, you will admit that the probabilities are practically nil. The truth is that many of these things are purely mythical and had been drawn from imagination and hearsay when they came to be written centuries after their supposed occurrence.

By the way, let me remind you that the Mahavansa was written for the first time only in the first quarter of the 6th century AD, i.e. nearly 1,000 years after the supposed arrival of Vijaya. The point is that from Devanampiya Tissa onwards, that is, from the time of the introduction o Buddhism to Anuradhapura and the establishment of the Buddhist clergy, we are all prepared to admit the story of the Mahavansa on the ground that it was related by eyewitnesses. It is in regard to the account of the kings, who preceded him, that we have our difficulties, for it was not related by those who lived at the time. Let us, however, go deep into the matter and try to discover further argument in support of Mr. Codrington’s doubts about the legend of Vijaya. Without resorting to any external evidence let us resort to the Mahavansa itself and from the internal evidence therein deduce who Vijaya was or was not.

In any case, from the facts stated in the Mahavansa, we already see that the account is highly exaggerated and made to conform to the writer’s predilections. Many a wish is father to the thought. Who would believe that Vijaya’s progenitor was a lion, or that he landed in Ceylon on the very day of Buddha’s death or that his nephew Panduwasa married a Sakya princess of the clan of Buddha himself? So Codrington is right in pointing out that these are so highly improbable that they are unacceptable as facts for the purposes of history. It would therefore look as if there was no one by the name of Vijaya who ever came to Ceylon and that the whole story is a fabrication.



Coningham & Lewer 2000 The Vijayan Colonization & the Archaeology of Identity in Sri Lanka


Letter 26

The Language spoken by Vijaya and his Successors

My dear Son,

But although the story of Vijaya is to be doubted, still let us assume that a Vijaya came from Bengal and that he spoke a foreign tongue, Sinhalese. Even then from what we read in the Mahavansa, it is clear that Vijaya knew Tamil too, for he was able to manage with Kuweni, the Yakka princess, who did not know a word of Sinhalese. Later, he and his 700 followers marry Tamil girls from Madura. This also supports my contention that all these 701 young men knew Tamil to be able to speak with their Tamil wives, who knew no word of Sinhalese, the Aryan language which was supposed to have originated in Bengal or in Gujarat.

Take their children, and what language do you think they would have spoken? Their children by these Tamil wives would have spoken their mother tongue Tamil, and so Tamil would have been the language used by their descendants too. In the light of these facts, there was no chance whatsoever for Sinhalese, the language supposed to have been spoken by Vijaya, to have been used as the language of his descendants, at least up to the time a change takes place with the arrival of Buddhism in the reign of Devanampiya Tissa.

You may ask me how it is that I have stated that in all other countries people used the language of the conquerors and that it is only in Ceylon that the reverse is true. The circumstances here are entirely different. Here there is no wholesale conquest. Only a few colonists arrive, 700 in all. They marry Tamil-speaking wives, as soon as they come and before they settle down. There is no sudden change of government. Had these 700 men brought their wives from Bengal or Gujarat and confined themselves as a separate community, without intermarrying from among the native population, (as, say, the English colonists did in America in the 17th century), and had they been recouped from time to time by fresh colonists coming from the mother country to join the predecessors, there was some chance for their language, Sinhalese, to have been used without interruption, as the language of the ruling classes, and continued to be used by the subject population too gradually. Here, there was no such possibility whatsoever, in view of these 700 colonists, all of them marrying Tamil-speaking wives and, not only marrying Tamil-speaking wives, but also importing the Tamil-speaking workmen with their families from the Tamil country, Madura. You will also see that no further Sinhalese ever came from Gujarat or from Bengal, whichever was the original home of the Sinhalese immigrants. In the absence of fresh immigrants who spoke Sinhalese, and in the presence of such a large number of people who spoke Tamil, the Tamil wives of Vijaya and his associates, the Tamil workmen who accompanied them, and the Tamil-speaking Yakkas who formed a large percentage of the subject population, there was every chance only for the Tamil language to have been the principal language of communication and for the Sinhalese language to have been neglected by everybody. In fact the Sinhalese language of Vijaya had absolutely no chance of survival from the very start. Under the circumstances, I have every right to presume that Tamil was the language used by the Vijaya’s successors, if not by Vijaya himself, for a considerable time.

Look at some further facts in support of my theory that Vijaya’s mother tongue had been forgotten and that Tamil was the language of the successors of Vijaya from what happened during the reign of Devanampiya Tissa. See what Devanampiya Tissa himself did. He personally went to Jambukola in the Jaffna peninsula to receive the bo-tree saplings sent by Asoka. He took the gift first to Pacina Vihare in Jaffna, before he removed them to Anuradhapura. At Pacina Vihare he had a celebration in which sermons were preached, evidently in the language of the people there. The people there cooperated with him to such an extent that there is clear indication that he was no foreigner but was one with the people of Jaffna. Jaffna at that time was ruled by Naga chieftains who must have been friendly with him or related to him. How was he so friendly, if he was not one of them but was a Sinhalese-speaking stranger? These facts clearly prove that he was conversant with the Tamil language.

As I have already told you, I am assuming that the Sinhalese language was the mother tongue of Vijaya and it is on that assumption that I say that his mother tongue was smothered at the very start for the reasons stated above. As to whether Vijaya’s mother tongue could have been Tamil is a different question. In any case, I shall lead evidence later to show that the language spoken by Vijaya was not Sinhalese, for I shall prove to you shortly that Sinhalese as a language was created only after the arrival of Buddhism in the island.

By the way, of Jambukola, you know the prefix Jambu is the Tamil Naval. (I have already commented on this in my letter No.7, while tracing the meaning of the word Yalpanam.) Jambukola of the Mahavansa would thus appear to correspond to the present Navatkuli of the Jaffna peninsula. In the ancient past the Jaffna peninsula was known as Jambudivu (vide, Manimekalai epic) and the present Pannathurai, being the Jaffna fort, was one of its ports, probably called Jambupannai, later becoming corrupted to Jappanam. This leads one to the consideration whether Tambapanni, where according to the Mahavansa, Vijaya is supposed to have landed in Ceylon, could not have been an incorrect spelling of Jambupannai. Mudaliyar Rasanayakam, in his ‘Ancient Jaffna’ (pp. 45-55) adduces very cogent reasons to urge that Vijaya’s landing place was in the Jaffna peninsula. Of course the story of a Yalpadi as the first colonist of the peninsula is nothing but a figment of someone’s imagination.




Letter 27

Vijaya’s successors were Nagas

My dear Son,

There is another reason how Devanampiya Tissa would have spoken the Tamil language, which he would not have, had he been a blue-blooded descendant of the Sinhalese-speaking Vijaya of Bengal and a pure Aryan. I suspect that he was related to the Nagas, and his knowledge of Tamil. Look at some relevant facts. His brother was Maha Naga. Very strange. How was it possible for his brother to have been called a Naga, if Devanampiya Tissa and his parents were not related to the Nagas, or if they had not intermarried among them earlier? It is quite obvious that the descendants of Vijaya, and the predecessors of Devanampiya Tissa, and perhaps Vijaya himself, had made marriage alliances with the Naga chieftains, who were ruling the country in the North and the West and the Southeast of Ceylon, and who had all become Buddhist before Devanampiya Tissa, as will be seen by the visits of Buddha himself earlier to Naga Dipa and Kelaniya where Nagas were ruling.

That the Sinhalese descendants of Vijaya had mixed up with the Nagas of Ceylon and had become one people with them will be clear from some further facts. We notice that, while every endeavor was being made to placate the Yakka population by their being treated with equality on important occasions, no mention whatsoever is made in the Mahavansa of the Nagas, an element of the population of Ceylon no less important than the Yakkas. Why was this? There must have been some reason. We find nowhere any such step taken to pacify the Naga population; not that they were negligible, for they too were an important section, of the population of Ceylon. The reason is not far to seek. The reason (and there is no other), is that the Naga chieftains and the Sinhalese kings were one and the same people. There could have been no other reason. There was therefore no necessity for any special steps to have been taken to please them, as was done in the case of the Yakkas.

It is clear that the Nagas were the ruling classes at that time and were more advanced than the Yakkas. In the North they had also become Buddhist earlier than Devanampiya Tissa. They were also the ruling classes with whom Devanampiya Tissa and his ancestors had intermarried and become relations, and that was why that they cooperated with Devanampiya Tissa to take the Bo-tree from Jambukola to Anuradhapura.

There is also another reason. The fact that Devanampiya Tissa was a relative of the Nagas and that the latter had become identified with the Sinhalese descendants of Vijaya could be easily judged not only from Naga, the brother of Devanampiya Tissa, but from the names of several kings who were successors of Devanampiya Tissa, whose names ended in Naga, as I had already pointed out to you in one of my previous letters (vide, letter 11).

I realise that against this proposition of mine, my critics may have two points to urge. It is stated, after discarding Kuweni, Vijaya married a Pandian princess. Only God knows whether it is true. Against this you will notice that not a single marriage alliance took place at any time thereafter between the kings of Ceylon and the kings of Pandiya. If it is true that Vijaya’s descendants were related to the Pandian royalty, they would have normally taken at lease occasionally Pandian princesses as their wives. According to the Mahavansa this never took place. That is one point.

The other point is that, Panduwasa, Vijaya’s nephew married Buddakachchana, a daughter of Pandu, who is a cousin of Buddha, from ‘beyond the Ganges’. Here too, be it noted, no marriage alliances from among this royal family took place subsequently at any time.

What do these two things indicate? They only indicate that no reliance can be placed on these two isolated statements of the authors of the Mahavansa. These are two instances of wish being father to the thought. I am wondering whether the writer of this portion of the Mahavansa had not been drawing too much from his imagination. Codrington would therefore appear to be right in his criticism that the desire of the writer was to connect the Buddha with the Sinhalese kings of Ceylon, a very bold attempt indeed.

Let me advance some further facts in support of my contention that the Sinhalese kings were Nagas. Not only did several kings, whose names end in Naga, rule Ceylon but the image of the cobra, which was their tutelary god, could be seen carved on every imaginable rock or stone considered important. The mark of the lion to denote the race of the Sinhalese was nowhere to be seen. I shall give you one example of what king Maha Naga, the last of the kings whose names ended in Naga, did as late as  AD 620. In building a bath for the priesthood at Mihintale, called Nagapokuna out of solid rock, he had a five-headed cobra, measuring 7 feet high and 6 feet across the head, carved in high relief out of the rock. (page 93 of D.G.Obeyesekera’s History of Ceylon). This is only one example. But numerous such examples of cobra images carved in stone could be seen in almost every tank constructed in ancient times. (vide, Ancient Irrigation Works, by Brohier, part II, p.11). Even today these carved stones are commonly found near the ancient sluices of tanks. For a specimen of a Naga-kal [cobra stone], please see page 24 of the Ancient Irrigation Works, part III.

I am citing these examples merely to show that it was the Nagas who were the leading community of the society of Lanka in ancient times. Imagine what their contribution to civilization could have been when we see it was they who gave the alphabet to the Aryans, the Deva Nagari script, music in Nagasuram, built cities and towns in Nagaram, brother refinement and culture as in Nagarikam (civilization).

If Vijaya had taken his wife from among such civilizaed people as the Nagas, after discarding Kuweni, there was nothing derogatory in that. It was all the more honorable than seeking alliance from the distant Pandiyan kingdom. So I am led to surmise, considering all the foregoing facts, that Vijaya had to discard Kuweni not to marry a Pandyan princess but to please the Nagas, who were of a superior status than the Yakkas, and that the second wife he married was not a Pandyan girl but was a Naga princess. If this were not so, it is impossible to account how Maha Naga became the brother of Devanampiya Tissa; it is impossible to account how Vijaya’s successors became related to the Nagas, for the Nagas to become kings of Ceylon on several occasions; and it is impossible to account why no Pandyans or the ‘beyond the Ganges’ relations of Buddhakachchana ever ascended the throne of Lanka even once. But the authors  of the Mahavansa would have us believe thing which we cannot reconcile one with another.



Stratton 2009 The Vijaya Origin Myth of Sri Lanka & the Strangeness of Kingship


Letter 28

The Aryan Myth

My dear Son,

I have discussed the matter so far on the assumption that the coming of Vijaya is true, but after examining all the relevant facts I am led to believe that the portion of the history of Ceylon prior to Devanampiya Tissa, is so uncertain that I think even the story of Vijaya is a myth and cannot be relied upon, for in fact there is no shred of independent evidence to prove its veracity or even its probability.

The story of Vijaya, it looks, was invented clearly for the sole purpose of attributing to the Sinhalese kings an Aryan origin. It serves no other purpose. This Aryan origin was necessary in order to keep the kings of Ceylon away from the kings of South India who were the enemies of Buddhism and who had proved their enmity by invading the island on more than one occasion, as I have pointed out to you in one of my previous letters. In fact the Mahavansa came to be written only just after the invasion of Pandu and his successors in  AD 468-512, when the hatred of the Hindu Dravidians was at its highest. Mind you, this was nearly 1,000 years after the supposed arrival of Vijaya and nearly 750 years after the introduction of Buddhism. There was at this time nobody to question the authenticity of this concocted story, and so it became current. What was there to vouch for the truth of the history of the period prior to the advent of Buddhism when it came to be written? There was no record whatsoever of either priest or poet or religious reformer for purposes of verification, no written or unwritten memorial or testimony of any kind for purposes of corroboration.

Apart from its Aryan pretensions, the story of Vijaya also appears to be a crude attempt, at some later date, by some person unacquainted with the original Dravidian language of the people of Ceylon, at giving a plausible meaning, in his own language, to the word ‘Sinhalam’ which had come into common use then. It was the work of someone who knew only Pali and not Tamil. Hence the emergence for the first time of the lion to denote ‘sinha’ and the consequent creation of a suitable story to support the lion theory simultaneously with the Aryan theory. Clearly the story of Vijaya is seething with inconsistencies and exaggerations and is lacking in naturalness to such an extent that no reliance can be placed on it at all.

On the contrary, whether there was a Vijaya or no, there is unmistakable evidence to show that Devanampiya Tissa, from whom the history of Ceylon really commences, was a Naga prince, whoever Vijaya might have been, and that all his successors had Naga blood in their veins. In other words, I only want to stress the fact that the Sinhalese kings of Ceylon were Ceylonese from the very start and that they owed no allegiance to outsiders, be they Aryans or Pandyans, and that they were Nagas who had been in Sri Ilam from time immemorial, long before the Aryans were heard of. I want also to stress the fact that the people of the North and the people down South were one people at one time and their separation is due to the different languages they speak at present and the different religions they now profess. Racially, both the Sinhalese and the Tamils of Ceylon have been Dravidians from the beginning of time. I do not see why the Sinhalese should not be proud of their Dravidian heritage and why they should seek any evidence to prove that they are Aryans or any race that came from Central Asia or even from ‘beyond the Ganges’, or celestials that descended from the Sun or the Moon or any planet, or sub-humans sprung from a wile beast like the lion. Even when they realise that the Indian civilization of today is traceable to the Dravidians and not to the Aryans, I am afraid the Sinhalese are not prepared to change their opinions.

Of course I sympathise with them in the difficult situation in which they find themselves for after having called themselves Aryans all these years, it would require super-human effort to call themselves Dravidians now. But still there are some amongst them who would not hesitate to obliterate any available evidence which might militate against the claim that the Sinhalese are Aryans. As a significant example of the mentality of this category of people, I would refer you to the last Census Report of the Government of Ceylon where we learn for the first time, after so many years of its publication, that Madura of the Pandyans, where Vijaya is supposed to have taken his second wife, is situated in Madya Pradesh in the basin of the Ganges; and this mind you, in spite of the mature opinions to the contrary previously expressed by scholars of such eminence as Messers Turnour, Wijesinghe, Dr. Geiger, Mudaliar Gunawardene, Farther S.G. Perera and others.

No wonder that Dr. Slater observes as follows in his, ‘Dravidian Element in Indian Culture’:

“The great obstacle to a right appreciation of the Dravidian influence in the evolution of Indian culture is the widespread currency and established position of what may be called the Aryan myth….Indians cling to the theory that they are Aryans [footnote 21] and that their religion and culture are Aryan. The word Aryan is legitimate enough provided the definite meaning is attached to it, as a name for the invaders from the North-West who introduced the Sanskrit language into India. It is illegitimate if used to imply the theory popularized by Max Muller than the ancient ‘Aryan’ race, superior to other races, spread from the original ‘Aryan home’ somewhere in Europe or Asia, over India, Persia and Europe, displacing the previous occupants all regarded as inferior mentally, physically and culturally and bequeathing to their descendants the various languages of the Indo-Germanic family. All attempts to harmonise that theory with the facts have broken down hopelessly and Max Muller himself was brought to admit that language is no test of race.”

It was Max Muller who started the myth of the existence of an ‘Aryan race’. But later he was convinced of the enormity of his error and wrote as follows:

“Aryans are those who speak Aryan languages, whatever their colour, whatever their blood. In calling them Aryans we predicate nothing of them except that the grammar of their languages is Aryan… I have declared again and again that if I say Aryans, I mean neither blood nor bones nor hair nor skull; I mean simply those who speak an Aryan language…When I speak of them I commit myself to no anatomical characteristics. The blue-eyed and fair-haired Scandinavians may have adopted the language of their darker Lords or vice versa…To me an ethnologist who speaks of Aryan race, Aryan blood, Aryan eyes and hair, is as great a sinner as a linguist who speaks of a dolichocephalic dictionary or a brachycephalic grammar.”

As another writer puts it:

“If the term Aryan is given a racial meaning at all, it should be applied to that ethnic unit, whatever it was, that first spoke a language distinguishable as Aryan. Of the character of that hypothetical unit it is the simple truth to say that we know nothing whatever.”

I must say that, even though the discovery of the Indus Valley civilization has upset all our conjectures and prepossessions, and even though the ethnological survey of India reveals that the people of India have very little in them of the features of the original Aryan invaders, who came through the North West frontier of India, the Aryan myth still persists, much to the consternation of historians who wish to unearth the truth.



Footnote 21: I include Ceylonese too here.


  Letter 29

The Evolution of the Sinhalese Language

My dear Son,

You might probably began to wonder by now how, if the Sinhalese do not belong to the so-called Aryan race, they came to possess the Sinhalese language, which they, as well as certain European scholars, consider is an Aryan, or rather an Indo-Aryan tongue.

Let me tell you at once that these European scholars, beginning from the late Prof. Max Muller downwards, were not acquainted with the Tamil language. They were taken up with the large vocabulary of Pali and Sanskrit words found in the Sinhalese language, without an examination of its syntax or the construction of its sentences (which are essentially of Tamil origin) and came to this conclusion. But it is those scholars who know both the Sinhalese and the Tamil languages who could say with authority whether Sinhalese is an Aryan or a Dravidian language. Let me mention the names of two such scholars in Ceylon itself; viz. the late Mudaliyar H.F. Gunawardene, and the late Father T.C. Closset S.J. I propose to make extensive quotations from their writings for your information as I proceed.

But before I do so, let us try to find out what was the language spoken in Ceylon before the arrival of Buddhism and even before the advent of Vijaya, for the Sinhalese language, as such, made its first appearance only subsequently. In one of my earliest letters [foot note 22], I had mentioned that Elu was that language and that the formation of the present Sinhalese language was due to the subsequent addition of copious Pali and Sanskrit words to Elu. In this connection, let me quote what the late Mudaliyar Rasanayakam wrote in his ‘Ancient Jaffna’ (pp.176-180):

“What was the earliest language spoken by the people of Ceylon? A language is invariably known from the name of the country in which it is spoken or from the name of the people who inhabit it…The name Lanka, applied to Ceylon, had not the remotest connection with its people or with the language spoken by them. The name Ilam, which was also given to Ceylon, has some affinity with its earlier language Elu. The island must have been called Ilam because Elu was spoken there, or perhaps the language was called Elu because it was spoken in Ilam. The name Ilam was undoubtedly given to Ceylon by the Tamils, her neighbours. Elu was only a spoken dialect and had not reached a state of development sufficient to produce any literature in that language.”

If Elu was the original language of Ceylon, then the question arises how the Sinhalese language made its appearance in the island. As Mudaliyar Rasanayakam says:

“Vijaya and his followers could not have introduced into the island a new language and imposed it upon the people. They and their descendants would have adopted the language previously spoken there.”

Besides, let it be noted, that if the Sinhalese language was brought to Ceylon by Vijaya from India, there is no trace whatsoever of this language in India, either now or in the past. So it is clear it was created in Ceylon and not brought from India. Then let us see how it was possible.

Mudaliyar Rasanayakam further says:

“Upon the introduction of copious Pali and Sanskrit words (after the arrival of Buddhism), a new language came to existence, with the ground work of Elu and Tamil and a superstructure of Pali and Sanskrit. Pali and Sanskrit were dead languages; they, therefore, could not furnish the foundation of a living language, but were only instrumental in furnishing a voluminous vocabulary to the new language. In a similar manner were formed Malayalam and Telugu; from their copious vocabulary of Sanskritic words it is now almost impossible to trace their origin to Dravidian dialects….The Sinhalese language, which was in an infantile stage in the 3rd century BC, as will be seen from the undeveloped phraseology used in the Cave inscription of that period, took about 1,500 years to reach that degree of development necessary for the composition of literary works in that language, for the first work was composed in the reign of Lilavati…Thus it will be seen that the mixed population from Point Pedro to Dondra Head must have, during the early centuries of the Christian era, spoken one language…This proposition is further supported by most of the place names in Jaffna which have an Elu or semi-Sinhalese origin.”

What is meant by the phrase ‘ground-work of Elu’ in the above passage is that the syntax and the construction of sentences in Sinhalese are the same as in the original Elu, a Dravidian dialect, so that even today if you take a Sinhalese sentence and substitute Tamil words, noun for noun, verb for verb, participle for participle, mind you, in the same order, you get the Tamil sentence intact. In regard to this aspect of the matter, let me give below a few extracts from the writings of the late Mudaliyar W.F. Gunawardene.

“Language is the medium for communication of our thoughts and thoughts are communicated not in isolated words but in sentences. Language is essentially the sentence, and grammar is the science which analyses and explains the sentence. Therefore, scientifically the determining factor of a language is not its vocabulary, but its structure, i.e., that aspect which is concerned with the arrangement and adjustment of words in the sentence for the expression of thought. In this respect Sinhalese is essentially a Dravidian language. That is not all. Its evolution too seems to have been on a Tamil basis.

And so we seem safe in saying that while Sinhalese is the child of Pali and Sanskrit it is in regard to its physical features and physical structure, essentially the daughter of the Tamil language.

Sinhalese is the standing monument attesting the language of the original inhabitants of Ceylon. That language was Tamil, and Sinhalese which came from nowhere had its origin in Ceylon, and was built up with Tamil as its framework.”

From the point of view of syntax, it has even been said that Sinhalese has retained the purity of the primitive Dravidian language from which both Sinhalese and Tamil were derived. Father Closset in his Dravidian Origin of Human Speech, [notes]: Apart from the syntax, even in regard to the vocabulary, in spite of the large influx of Pali and Sanskrit words, what I would call the basic vocabulary of the Sinhalese language is wholly Dravidian. See what this writer [Father Closset] has to say on this subject:

“The origin of the human speech and its construction can be traced in the Dravidian languages, among which I say at once is included Sinhalese. First, because the construction of the sentences in Sinhalese is essentially Dravidian. Secondly, because many of its words, even the most elementary, as inta, wanta, ganta, are of Dravidian origin.

When a language is said to be derived from Dravidian, it does not mean that the whole vocabulary is derived from Tamil. The derivation of a language begins with the formation of its vocabulary from the nominal or verbal stems of another language, or by agglutination of elements belonging to it, since the mother tongue of the same word existing in different languages is that which possesses its constitutive elements. In this sense we will see that the Sinhalese is derived from Tamil, for it is in the Tamil language that you find these constructive elements…

The derivation of Sinhalese, like that of all the other languages, from Tamil, can be proved not only by the universal law of Tamil, whose stems are found in the Sinhalese vocabulary, but also, and more convincingly, by the Sinhalese suffixes, which are Tamil verbal stems or nouns, as in other languages.

The law is that the mother tongue of words is the one which possess their constitutive elements. The law is still more stringent in the case of words into whose compositions inseparable suffixes, i.e., those suffixed which by themselves have no meanings, enter.”

To give you a clearer idea of what is meant by the words ‘constitutive elements’ and ‘inseparable suffixes’ etc., let me select a few examples out of the list given by Father Closset in his book.

Sinhalese verbs derived from Tamil Stems

Tamil root or infinitive            Present relative participle        Sinhalese

Talla (to push)                         tallukira                                   tallukara (to push)

Oppa (to offer)                       oppukira                                  oppukara (to offer)

Wisa (to throw)                       wisukira                                   wisikara (to throw away)

Parra (to kindle)                     parrukira                                 pattukara (to kindle)

Niyami (to order)                    niyamikkira                             niyamakara (to order)

[Past relative Participle]

Pila (to split)                           pinta                                        binda (to split)

Seyya (to make)                       seyta                                        sada (to make)

Aniya (to dress)                       aninta                                      andinta (to dress)

Ta (to give)                              tanta                                        denta (to give)

Wa (to come)                           wanta                                      wanta/wenta (to become)

Pongka (to swell/boil)             pongkina                                 ponggana-wa (to swell)

Kolla (to take)                         konda                                      ganta (to take), gatta

Erra (to lift up)                       erru (verbal participle)            ettuganta (to receive)

Sinhalese nouns derived from Tamil past sentence

Al (to govern)                          anda                                        anduwa (government)

Tal (to be low)                         talnta                                       dilinda (a poor man)

It will be seen that in Tamil the roots of verbs can be made to form all the tenses, relative and verbal participles, verbal nouns etc. These latter are what are called the constitutive elements of the roots. In the case of Sinhalese, the words are formed from the verbal or nominal stems of Tamil and these stems are rigid and cannot form any tenses or other variations as in Tamil. So, according to Father Closset, the mother tongue of the above Sinhalese words is Tamil, for it is in Tamil that you find the constitutive elements of these words.

In addition to the above list of verbs and nouns, Father Closset gives two more lists. One is a list of Sinhalese-Tamil compound words. There are Sinhalese words in which all the constitutive elements are Tamil or some of the constitutive elements are Tamil and some are Sinhalese. In regard to the former, let me take only one example among the many given by him. The Sinhalese word kandulu means tears, which is a Tamil composition kan and tuli. In Tamil, kan is eye, and tuli means drops. In Sinhalese, kan has no meaning nor has tuli. So the mother tongue of kandulu is Tamil. In other words, Sinhalese kandulu was derived from Tamil.

The list of words is what he calls, inseparable Sinhalese suffixes, which are Tamil verb forms, adjectives etc. Here too let me give you only one example. The Sinhalese word kurulla means a bird, literally meaning having claws or points. It consists of two Tamil words, kur and ulla. The prefix kur means a sharp point and the suffix ulla means having or being. In Tamil, the suffix ulla, which is a participle adjective, is a separable suffix and can be used similarly with other words to denote having or being: examples: kannulla (having eyes), kathulla (having ears) etc. It is what the author calls as inseparable suffix, which by itself has no meaning in Sinhalese. Therefore the Sinhalese word kurulla is a Tamil derivative.

As regards Sinhalese words taken from Tamil, Father Closset seems to think that there are thousands of such words in the Sinhalese language and not merely four hundred as sated by Mudaliyar A.M.Gunasekera, in his Sinhalese grammar.

Passing on to other matters, let us now examine such a simple matter as the Sinhalese alphabet. You will see how closely they follow the Tamil script in their formations. Besides, the consonants N and L has two forms each in both languages, which are not found in other languages, barring Sanskrit, which borrowed them from Tamil. In Tamil, the two forms are, GNa + la and Na+la. The former GNa + la is pronounced by the tip of the tongue touching the middle of the palate and the latter Na+la by the tip of the tongue touching the end of the palate where it meets the front (upper) row of teeth.

In addition to these two consonants, let us examine the long vowel EE. See what Dr. A.S.Rajasingam writes in his article, ‘Are the Sinhalese Aryans?’, published in the Daily Mirror of 24 Dec. 1968. He says:

“What is more remarkable and sets the seal on the borrowing of the whole Tamil alphabet by the Sinhalese as the letter EE. There is not much similarity, one might say, excepting in the two eyes. But why use the two eyes for the same letter? If there is no answer to that, would it not be still more surprising to note that the letter used with the two eyes is in each case the letter R. Hence the letter R rightly reveals the reality of the oneness of the two alphabets and the common parentage of the two languages, i.e. Dravidian.”

In regard to the script, Mudaliyar Gunawardene says: “I must say, in fairness, that the Sinhalese script is derived immediately from the Tamil to any reasonable mind, though remotely from the same source.”

As regards the claim that Sinhalese is an Aryan or an Indo-Aryan language, such a claim cannot, in the light of the facts mentioned above, be maintained. This is applicable not only to Sinhalese but also to the re-called Aryan vernaculars of India like Hindi, Gujarati, Marathi, Bengali etc. As I have previously stated in my letters to you regarding the ‘Aryanisation of the Dravidians’ and the ‘Dravidisation of the Aryans’, these Aryan vernaculars came into existence when the Dravidian dialects then existing throughout the whole of India came into contact with Sanskrit, the language of the Aryan ruling class. (foot note 23) The result was that even when the Dravidian dialects were superimposed with Sanskritic words, there structure remained the same. Let us hear what Mudaliyar Gunawardene has to say on this matter:

“In the so-called Aryan vernaculars of India and Sinhala on one side and Tamil on the other, the vocabulary is different. But if in the Indian sentence you substitute Tamil, word for word, each word keeping its own inflection, one will be surprised to find how the composition stands perfectly Dravidian.

Not only Sinhalese but all the so-called Aryan vernaculars of India are founded on a structural basis mainly Dravidian and are indebted for much of the fundamentals of their structure to Dravidian influence… In the early Prakrit the disappearance of the dual number already shows the Dravidian influence. When the Aryan vernaculars had further evolved under such influence we find them bearning on almost every feature the Dravidian stamp imprinted unmistakably.”

Apart from the structure, even in regard to inflection, he says:

“Sanskrit has three numbers for a noun. Tamil has only one set of terminals for both singular and plural. Likewise, the Aryan vernaculars:


Hindi: Ek Ghere (one horse), Tamil: Oru kudirai (one horse)


Hindi: Panch Ghere (five horses), Tamil: Eindu kudirai (five horses)

In Sanskrit, the adjective is inflected for gender, number and case. In Tamil, the adjective is indeclinable. In Aryan vernaculars, the adjective conforms to Dravidian sometimes.

Ko-dative suffix of India, and ka, ki suffixed of the genitive are of Dravidian origin from the Tamil Ku.

The dative case is prominent in Aryan vernacular as in Dravidian. E.g:

English: go to the house

Sanskrit:  House go (accusative of place)

Tamil: veetuku po (to the house go)

Sinhalese: geta palayang (to the house go)

In Sanskrit, the case of possession is genitive. In Tamil, Hindi and Sinhalese, it is dative. E.g:

Sanskrit: Mama pustakam asti (Of me there is a book.)

Tamil: Enakku pusthakam irukkirathu (To me there is a book.)

Hindi: Mungke pothi hai (To me there is a book.)

Sinhalese: Mata pothak thibe (To me there is a book.)

It seems to me that the Aryan vernacular is a misnomer…Aryan vernacular, I say that name was suggested by external appearance at a time when light on the true internal character was wanting and that in the present day of light, that name is a mere convention, with next to nothing of probative value as evidence.”

In spite of all the evidence cited above, if the Sinhalese still persist in saying that Sinhalese is an Aryan language, or an Indo-Aryan language, as they prefer to call it, just for the honor of it, they are at liberty to do so. They may thereby be able to hoodwink some Europeans who do not know Tamil, but certainly not those who know both the Sinhalese and the Tamil languages. Theirs was a Dravidian language and continues to be so. The Pali and the Sanskrit words used therein, even though they may form a very high percentage of the modern Sinhalese vocabulary, cannot alter the basic fact that the structure and the skeleton of the Sinhalese language is Dravidian, apart from the fact that almost all the fundamentals and elementary words and suffixes are, as stated by Father Closset, also of Dravidian origin. It is thus clear that in every respect Sinhalese is essentially a Dravidian language. As Mudaliyar Gunawardene says, “Its evolution too appears to have been on a Tamil basis, for it is dominated to a considerable extent by the principles of Tamil grammar and by Tamil idioms.”

Clearly, the classification applicable to the Sinhalese language (as also to the Indian vernaculars referred to above), is not Indo-Aryan, but Indo-Dravidian.



Footnote 22: vide letter, No.7; the concluding paragraph.

Footnote 23: vide letter, No.19.

Saha 1988 Map of Vijaya Legend Locations


  Letter 30

A Good Hypothesis for the Origin of the Sinhalese Race

My dear Son,

Let me make a resume of the story of the origin of the Sinhalese people which the Sinhalese want us to believe. It is as follows. A certain Vijaya, whose grandfather is a lion, is driven out of his country (somewhere in North India) by his father and accompanied by 700 followers comes to Ceylon. He and his followers call themselves Sinhalese because of this origin. On arrival, they find the island occupied by a race of demons called the Yakkas and the Nagas. They settle down near Puttalam and, after driving out these demons to the mountains or to the see or otherwise exterminating them altogether, marry Tamil wives and bring Tamil skilled workmen and thereafter colonise the whole of Ceylon.

And after they colonise, they introduce the art of irrigation, which it is claimed is unknown in Ceylon before them; and it is they who build all the tanks and reservoirs now found in Ceylon. Further, they speak an Aryan language, now called the Sinhalese language, which originated in North India and which has no connection whatsoever with the Dravidian languages of South India. Any Dravidians or non-Aryans, who may have come to Ceylon either as invaders or as merchants come only after the Aryan Sinhalese arrive.

The above is the version of the history of early Ceylon in a nutshell as propounded by the Sinhalese historians. Now if this version is true, it must be supported by or fit into all the facts of history already known to and accepted by us as true. In other words it must be verified and proved beyond any shadow of doubt in the light of our existing knowledge of men and things. Until it is thus verified or proved, it is only theory or a hypothesis.

Now, you know logic (for you offered it as one of your subjects for your London matriculation) and in inductive logic, you have to read all about a hypothesis. What are the requirements of a good hypothesis?

The first condition or characteristic of a ‘good’ or ‘legitimate’hypothesis is that it must be based upon facts or events, actually occurring in nature. The second requirement of a good hypothesis is that it must not conflict with received or accepted knowledge or with any of the established laws of Nature; i.e. it must be capable of being brought into accord with existing knowledge. The third requirement is that it must be capable of use as a principle from which consequences may be derived by deduction; i.e. to assume something utterly unlike all that we are previously acquainted with is to assume what can neither be proved nor disproved for we cannot draw any conclusions from it. Such an assumption is called a barren hypothesis.

The term ‘theory’ is used for hypothesis which is suggested but not yet sufficiently verified. Some or more of our historical legends, traditions and tales are theories. I hope you will keep these things in mind and see whether the theory or hypothesis put forward in regard to the origin of the Sinhalese people satisfies all the requirements of a good hypothesis referred to above.

In the first place, we are told that Vijaya’s grandfather was a lion and this cannot be in accord with any of the established laws of nature.

Secondly, in verifying the claim that the Yakkas and the Nagas were demons and were completely exterminated by the incoming Sinhalese we find a discrepancy, for we see Yakka chieftains holding important positions during Pandukabhaya’s reign and vast numbers of common Yakka people embracing Buddhism during Dutugemunu’s reign 300 years later and we find too that several of the kings of Ceylon bore Naga names.

Thirdly, the claim that the swarthy natives of Ceylon are of Aryan origin is in conflict with our accepted knowledge of the complexion and physical features of the genuine Aryans.

Fourthly, the claim that Vijaya and his followers while coming from North India spoke an Aryan language, which is the present Sinhalese language, is in conflict with the fact that this language was or is not spoken anywhere in India at any time by any people.

Fifthly, the claim that Sinhalese is an Aryan language is in conflict with our knowledge that in etymology and syntax, Sinhalese shows signs of its having been a Dravidian dialect at the very start.

Sixthly, the possibility that the children of Vijaya and his followers would have spoken Sinhalese, the language of their fathers, instead of Tamil, the language of their mothers, who were all Tamil-speaking girls from South India, is in conflict with our experience of what happens everywhere as regards the mother-tongue spoken by children.

Seventhly, the fact that the terms used for tanks and other irrigation works are Tamil derivatives is in conflict with the claim that the system of irrigation was introduced into Ceylon for the first time by the Aryans and was unknown to the people then occupying the island and South India.

Eightly, from our knowledge of the people of South India, who were very highly civilized having intercourse and trade with neighbouring countries at the time Vijaya came to Ceylon, the surmise that Ceylon, which was only 30 miles across the Palk Strait, was left unoccupied by them is in conflict with our accepted knowledge of the facts associated with the movements of people and the spread of civilization all over the world.

In view of these discrepancies and conflicting facts, the present story about the origin of the Sinhalese, that they were Aryan settlers who came to Ceylon from North India speaking an Aryan language entirely foreign to Ceylon and, after displacing the original population, introduced civilization for the first time into the island in the form of irrigation works unknown before, cannot be a fact and cannot be regarded as a legitimate hypothesis or even a good hypothesis. I am therefore putting forward an alternative hypothesis which is as follows:

  • No Vijaya came to Ceylon at any time, nor any Aryans from any part of India. The legend of Vijaya was invented to give the Sinhalese, who were Buddhists, an Aryan origin in order to alienate them from the Hindu Dravidians who were surrounding them so that thereby Buddhism might be safeguarded from the onslaughts of Hinduism.

  • The present Sinhalese population of Ceylon are the direct descendants of the Yakkas and the Nagas, who were Dravidians, occupying Ceylon in ancient times and of a few immigrants who might have come from India from time to time. Of the Yakkas and the Nagas, the former formed the bulk of the common people and were (probably with the assistance of Tamil artisans) the builders of the magnificent irrigation works both earlier and in subsequent times, and the latter, a more refined and cultured community, were the upper classes and became the Sinhalese kings referred to in the Mahavansa after the advent of Buddhism.

  • The Yakkas and the Nagas, who did not become Buddhist or who after becoming Buddhist became Saivites again, spoke the Tamil language throughout and the present Ceylon Tamils are their descendants. They did not come to Ceylon recently excepting for a few immigrants who might have come along with the various Tamil invasions but remain here from pre-historic times. The present Tamil inhabitants of the North and the coastal areas in the East and the Northwest are these people.

  • The Sinhalese language of today is the product of evolution from Elu, the early Dravidian dialect spoken in Ceylon, when the latter came into contact with Pali in the same way as the present languages of North India came into existence after the Dravidian languages spoken there came into contact with Sanskrit, and as the Romance languages in Europe after the contact of the indigenous languages there with Latin.

  • The word ‘Sinhalam’ was derived from the Tamil word ‘Sri Ilam’, which became in course of time Sihalam and later Sinhalam, with the elision of the consonant r and the addition of the euphonic n.

  • The knowledge of the scientific system of irrigation found in Ceylon was not imported from abroad but was already in the possession of the people of the island from very early times, as will be seen from the names of numerous tanks ending in kulam as well as from the terms used for irrigation works which are Tamil derivatives.

  • The construction of larger and more numerous tanks and elas than in earlier times was due to the fillip given by the subsequent Sinhalese kings, after the arrival of Buddhism, owing to the necessity to feed vast numbers of the Buddhist clergy, to whom lands were donated by them, and also owing to their desire to acquire merit thereby.

  • The civilization of Ceylon, except for the introduction of Buddhism, is a branch of the old Dravidian civilization of India in almost all its aspects, as will be seen; for example, in the village panchayat system and the caste system. Even in the matter of the New Year celebrations, the Sinhalese follow the Tamil practice in regard to date and other details, and not the North Indian practice which is entirely different.

The above is my hypothesis and it appears to be the only legitimate hypothesis possible, when we examine the subject in all its aspects, linguistic, ethnological, social and cultural. In other words my submission is that the Sinhalese were Tamils up to the time Buddhism came to Ceylon, and thereafter these very Tamils became Sinhalese gradually. To put it somewhat mildly, the Sinhalese are not Aryans, but have been Dravidians right from the very beginning to the present day, and their language too has been Dravidian.

In writing these letters, let me assure you that I do not wish to offend anyone in the least, your Sinhalese friends and college mates included. My object is to pursue truth for its own sake for, according to Gandhiji, Truth is God. I hope that you too, and also your Sinhalese friends, will pursue it similarly, without any bias for politics or a desire for the mere glorification of one’s country or community or greed.