loanwords as social history
I went to a fascinating seminar some time back at the Asia Research Institute in Singapore, given by Dr Uri Tadmor on the subject of Malay and Indonesian loanwords. Loanwords, by their nature, can often be strong evidence of sustained cultural interaction -- ephemeral contact is often not enough to stimulate widespread borrowing -- but they can do more than allow us the rather facile conclusion that, for example, Persians and Indians were, once upon a time, in contact with Malays. What Tadmor was really concerned with was to showcase the potential of linguistics, in particular the study of loanwords, for investigating the social history of a region, sometimes even against the grain of received (or convenient) wisdom.
One could look at the type of words that are borrowed: in particular, what Tadmor called their 'semantic fields', which is to say, the general intended function of these words, such as 'religion', 'clothing' or 'law'. This can say a lot about what exactly Indians or Persians were doing in Malaysia; for example, one might draw conclusions about the fact that many Persian words in Malay are to do with important cultural practices, like marriage, which is kahwin in Malay and kawin in Persian. Tadmor points out that in most languages, just as you would expect, words least likely to be borrowed are those from semantic fields governing qualities common to all human beings: words of spatial relations, movement, sense perceptions, quantity. Yet even at this basic level, social factors can prevail.
Take quantity. The Malay-Indonesian word for 'three', tiga, is from the Sanskrit trika, which means 'trinity'. One might rightly, Tadmor says, wonder why Malay-Indonesian would need to borrow a word for such a basic number, when there already existed telu, the Javanese and Old Malay word for 'three'. His rather amusing conjecture, which I find entertaining enough to pass on with a pinch of salt, is that the Sanskrit word was poached in cultural embarrassment. The word telu sounds uncomfortably similar to the Malay word telur, which means 'egg', but which Tadmor points out was also a common euphemism for 'testicles'. One, two, testicle! Better to bung in the more lofty Sanskrit when speaking to important people. And apparently, it stuck. There's a huge amount of Sanskrit in Malay, including such basic words as those for 'head', 'light', 'because', 'all' and 'when'.
But one could also look at how a word gets imported, and what morphological changes occur along the way. Note how the unvoiced consonant in the Sanskrit trika, the hard 'k', has become the voiced consonant, the 'g', in Malay tiga. Tadmor says that this form of morphological change very commonly takes place when Indo-Aryan words (Sanskrit) are imported into Dravidian languages; for example, Tamil. Can it be, then, that Sanskrit words are being relayed into Malay-Indonesian through South Indians? Sure enough, there's an astounding number of Dravidian-origin loanwords in Malay and Indonesian. To name just a few: Tamil's kappal for Malay's kapal, or 'ship'; Tamil's taman for Malay's teman, or 'friend'; Tamil's katai for Malay's kedai, or 'shop/market'. Most interestingly, the word for 'shop' in Indonesian is not kedai but toko, which is a word of Hokkien Chinese derivation -- perhaps giving some indication of the early social landscape in Malayan and Indonesian towns and cities, and in particular, who were the ones keeping shops. Finally, through some pretty intense morphological analysis, Tadmor shows that most words of Arabic origin in Malay and Indonesian were ultimately borrowed via Persian -- even, I believe, aspects of Jawi, the Arabic written script adapted for Malay.
One can also, metaphorically speaking, carbon date loanwords; that is to say, get some sense of approximately when a given foreign word entered a language. For example, there seems to be good evidence suggesting that Dravidian loanwords in the Malay Archipelago can be traced back to Old Tamil, which scholars date variously between 300 BC and 700 CE. An example would be the Malay word nelayan, for 'fisherman'; according to Tadmor, this comes from a Tamil word ulayyan [sic?] that no Tamil speaker today would recognize, because it's been out of use for centuries and is only found today in ancient Sangam period literature. This suggests that there must have been Dravidian, specifically Tamil seafarers in the Malay Archipelago more than a thousand years ago.
And it seems it wasn't only Tamils living on the peninsula a thousand years ago. Tadmor intimated that this aspect of Austronesian linguistics is almost entirely understudied: Malay contains an astounding number of Khmer loanwords. Some of these words have been borrowed through Thai; for example, the word for 'census', which is bonci in Khmer, banchii in Thai, and banci in Malay. Some of these words are the same in all three languages: the word for 'recognize' or 'remember' is cam in Khmer, Thai and Malay. But some Malay words are different enough from their Thai or Khmer versions that we might wonder whether they were really so related: the word for 'candle', for example, is thien in Khmer and thian in Thai, but dian in Malay. And here's where the carbon dating comes in: the present Khmer form of 'candle' is thien, but its form in Old Angkorian Khmer (roughly, 9th century to 13th century) is exactly the same as that of present day Malay: dian.
I do think all this, conjectural and thin in the way only archaeology can be and still maintain its academic reputation, is nonetheless rather interesting in particular for Malaysia, whose official history maintains a studious, almost desperate disinterest in pre-Islamic history of the peninsula. This is, I think, a product of some deep insecurities about the place of Malay and the Malays in history, and the cultural defensiveness that comes with such complexes; it's a set of concerns that I think Indonesia doesn't share to quite the same extent. But in the face of the unmistakable traces of Ancient Khmer, Old Tamil, Urdu, Sanskrit and a distinctly Persianized Arabic in the very language that you utter your denials of pre-Islamic-Malay history in -- this is masterful hypocrisy. Also, failing to acknowledge the deeply porous character of the Malay language (nearly a third of the Malay-Indonesian language is loaned) might blind us to the way in which two of the most salient 'semantic field' vacuums in Malaysia, over the last century, have been and are still today being filled by non-Malay terms: science and politics. No prizes for guessing what the words for those in Malay are: sains and, yes, politik.