[A. G. Maisey]

David, this business of keris etymologies has been going on for years, both in the early western keris writings, and especially in Javanese writings and discussions. If you plod through all of it, you feel as if you are drowning in rubbish.

There seems to be a relationship between "iris", and "keris", but whether there is or not, and how it developed , I don't know, and no recognised authority on Old Javanese has ever written on this, to my knowledge. All you can do is go to the works of Zoetmulder, Pigeaud, and a few other notables, and glean what you can from their writings. My preference is for the iris root, simply because none of the suggested alternatives seem to make any sense.

There is a problem with Old Javanese and that is that by the time academics got around to working with it, it was already a dead language. Much of the lexicon of Old Javanese flowed into Modern Javanese, but Modern Javanese became a hierarchical language, which the authorities seem to think, Old Javanese was not, or at least not to the extent that the House of Mataram forced Modern Javanese to become.So, when the academics began to compile a lexicon of Old Javanese, sometimes the word meanings came down to logical analysis. They could not always say with certainty exactly what was being referred to. Similarly, Old Javanese was undoubtedly an "non-standard" language, as is Modern Javanese, which means that the same word can take different forms that depend on varying factors. This means that we don't necessarily know if we are looking at a different word, or the same word with a different pronunciation.

It also needs to be understood that there was not just one "Old Javanese" language. There was the spoken language of Old Javanese, but early inscriptions were in Sanscrit, the earliest inscription in Old Javanese is the Sukabumi inscription from about 805. Then we have Kawi, which is a literary language and that uses a script that developed from the Indian Pallava script, and Kawi also has an old form.About 20-25% of all Javanese words, Modern and Old Javanese, derive from Sanscrit, however, when they have come into Javanese they do not necessarily retain the same forms and meanings as they do in Sanscrit. To understand all of this well is the domain of a specialist linguist, and I do not pretend for one moment to be such.

However, we do need to have some grasp of this background to be able to understand the limits of dealing with the Old Javanese language.With this in mind, the words found in the various languages that can be considered as "Old Javanese" languages, and that can refer to the implement that we would regard as a keris include, but are not necessarily limited to:-

tewek, twek, panewek, tuhuk, curigo, curik, duhun, and of course, kris.

there are many words that come from, or are associated with these nouns, but I'm not going to go there.

Equally, there are many variant spellings in the romanised representation of these words.

In modern Javanese, keris is ngoko (low level), dhuwung is krama (usually referred to as "high" level, but in fact a ceremonial variation of ngoko), wangkingan is krama inggil (high level krama), curiga is a literary usage, and can also refer to a dagger.

The formality of using the Javanese language is that the person of higher status "talks down" to those under him, thus as an example, on a scale of one to ten, with the King at ten, #6 uses krama to #8, but #8 uses ngoko to #6.