An archive is only as good as its willingness to be corrected. Over the past week this one was read closely by someone who knows the subject better than we do, and a number of things turned out to be wrong. This is what they were, and what has been done about them.
The corrections come from Anuru de Silva, whose knowledge of Sri Lanka's caste communities has already shaped this site once — the dark green and gold it is built in follows the community's documented colour convention on his advice. He has since reviewed the content itself. We are grateful for both, and the errors below were ours, not his.
Five people were listed who should not have been
The notable-people directory was inherited from the previous version of this site as a flat list of names, presented without qualification as Govigama. Expert review found five of them belonged to other communities: Mangala Samaraweera, Martin Wickramasinghe, Munidasa Kumaratunga, Gongalegoda Banda and W.A. de Silva.
All five entries have been removed. This site does not publish which communities they did belong to, for the same reason the original entries were wrong: asserting a named person's caste without a published source is not something an archive should do. Two entries in the same section — Sir Paul Pieris and Dr E.W. Adikaram — were reviewed and confirmed, and remain.
It is worth being plain about the arithmetic. Five wrong entries out of a list of thirty-three is not a rounding error, and the remaining names carry no more evidence behind them than the removed ones did. The directory now opens with a note saying so: these biographies describe public record — offices held, dates, what each person did — and do not assert anyone's caste.
The Radala are listed under their own heading
The Kandyan aristocratic families were previously named in a single run with low-country independence activists, under a heading that read "Aristocracy & Independence". Presented that way, the whole list read as one group. On Mr. de Silva's advice the two are now separated: an Independence Movement section, and a Radala section with an introduction of its own, which sets out what the Radala were and notes that authorities differ on whether they are best described as a community in their own right or as a stratum within the Govigama. The section does not settle that question.
The Durava article was telling the story backwards
The Durava page opened with toddy tapping, as did its summary everywhere else on the site. That is the association the community is usually given, and the order of events is the wrong way round.
Before the Portuguese arrived, the Durava occupation carried ten obligations to the Sinhala royal court — dasa dura, the ten duras — and it is from these that the community's name is said to come. Coconut planting expanded sharply after the Portuguese arrival, the coconut industry was loaded onto the community, and the label that followed dates from that imposition rather than from what the community had been. The page now says so, and leads with the court service instead.
The article gained a good deal more besides: the community's own account of its descent, set beside the standard colonial-era account rather than replaced by it; Sitinamaluwe Dhammajoti Thera, the Durava monk who worked alongside Welivita Saranankara in the eighteenth-century revival of Buddhist learning and on whose proposal the Siyam Nikaya was brought from Siam — the same fraternity that excluded his community in 1764; Andare and Gajaman Nona; and the settlement geography from Gampaha through Colombo and Panadura to Matara and Hambantota.
Two published works entered the bibliography with it: E. W. Perera's Sinhala Banners and Standards (1916), on the red standard bearing the lion having long served as a Durava emblem, and Sir Paul E. Pieris on the Portuguese period, who endorsed the point.
The communities are no longer a ladder
Separately from the expert review, a reader pointed out that the Communities index listed the seven communities as a numbered column running top to bottom — which reads as a ranking, with the first name highest and the last lowest. The page's own text spends three paragraphs arguing against exactly that.
The list is now a set of equal cards, ordered alphabetically in whichever language you are reading. Because English and Sinhala alphabetise differently, the order changes between them, which is itself a useful signal that nobody chose it. The same ordering now governs the menu, the footer and the links at the foot of each community page.
Every change is on the record
None of the above would be worth much if you had to take our word for it. There is now an updates and corrections log listing every published change to this archive's content — what changed, when, and on whose guidance. Each article also carries the date it was last updated, and its own revision history where one exists.
The community pages remain under review while the rest of the material is checked against publications. If you find something here that is wrong, incomplete, or put badly, we would rather hear it than keep it. Get in touch.