K and G in Thai

[Kanga Japan]

Why are some Thai words, which begin with a G, written in English as beginning with a K?

For example: gradon (กระโดน) is almost always written as kradon. Such words appear to be pronounced as if they start with a G.

Can anyone tell me why? Thank you.

[rgs2001uk]

The difference between written and spoken Thai.

gradon (กระโดน) is almost always written as kradon.

It has been written in English as it would be pronounced (spoken).

Example,

กุ้ง

Could be written as a direct translation as gung, but written as kung how it is pronounced.

กฺระ-เทย

Is another example, direct translation gatoy (written), pronunced with a k not a g when spoken

[eric67]

I don't think it's a difference between written and spoken Thai.

It probably has more to do with different transcription systems.

Some like to use g and k for ก and ข

Others like to use k and kh for ก and ข

It can be a bit confusing at times, but it is how it is.

Better to learn how to read Thai, then you won't have to bother with that anymore

[Kanga Japan]

Dear rgs2001uk,

Thank you for your reply.

I am not sure exactly what you mean. Do you mean that even though it begins with ก, it is pronounced as K?

[Kanga Japan]

Dear Eric67,

Thank you for the information.

I am learning Thai, but I was told that if ก was at the beginning, it was pronounced as a G.

I want to write it down for non-Thai speakers, so that they can pronounce it as close as possible to the Thai pronunciation.

Would กระโดน be pronounced with the first sound as a G or K? Google Translate sounds like a G.

Sorry for this very basic question.

[eric67]

The ก doesn't really have an equivalent as an initial consonant in English.

Technically it's an unaspirated k, that exists in English only as a second consonant in consonant clusters like "sky"

But for the average English speaking person the "g" is the one that comes closest, so I would use that one.

[Kanga Japan]

Thank you Eric67,

I understand now. Thanks for the easy to understand explanation.

I will continue my Thai studies and maybe one day I will finally be able to read Thai with ease.

[Johpa]

The ก doesn't really have an equivalent as an initial consonant in English.

Technically it's an unaspirated k, that exists in English only as a second consonant in consonant clusters like "sky"

But for the average English speaking person the "g" is the one that comes closest, so I would use that one.

Since the unaspirated stop does exist, and is used in English (although not phonemically), the average English speaking person would be well advised to make the attempt to produce the sound in the syllable initial position. In my admittedly less than humble opinion, it is only the below average native English speaker who is content with substituting the voiced aspirated stop /g/ for the unvoiced unaspirated stop /k/ (ก). So best to use the letter 'k' for the unaspirated stop and 'kh' for the apsirated stop rather than reinforcing the erroneous use of /g/ for /k/.

Yes, sorry, it is a pet linguistic peeve of mine.

[Kanga Japan]

Dear Johpa,

Thank you for your detailed explanation.

I quite understand what you mean and am in agreement with this. The problem, however. is that I am making a dictionary of Thai food-related ingredients, such as vegetables, fruit, different types of noodles, fish, flours, etc.

It will be mainly designed with the intention of assisting non-Thai speaking people interested in Thai cuisine and cooking. They will therefore be using Thai cooking books, which have been written in English. Those books employ many different methods of transliteration, so I have to (in my opinion) include common spellings used in those books and also perhaps, spellings close to the actual Thai pronunciation. I am sure that a large majority of those people have never studied Thai at all and are therefore unfamiliar with the spellings common to Thai, such as the H following a consonant, such as PH, KH, etc. I am worried that they would mispronounce them. Thai script will also be included (as well as Japanese).

If it was a dictionary for people who have an understanding of the Thai language, there would be no problem.

This is a subject which I will have to deal with later on (with the assistance of a Thai person, or someone who understands Thai). At the moment, I am still in the research/checking stage. Most of the words are there already. Also, I myself am still in the process of studying Thai, so I have to rely heavily on many different resources - particularly of late from the many kind members of this Thai Language forum. Although my questions are not 100% language-related, I am unable to post my queries in the Thai Cooking section, as the Thai script cannot be used there.

Again, thank you for your help.

[Richard W]

Consider the word กุ้ง 'shrimp'. The Royal Thai General System (RTGS) will transcribe it as 'kung', while an English-based transcription will render it 'goong'. Neither system will tell you the length of the vowel or the tone. Most non-Roman script words transcribed into the Roman script use the system of English consonants and Thai vowels, so 'kung' should not be so strange. You probably need three ways of recording Thai names - Thai script (which will make your book more useful if communicating with Thais), RTGS (to stop many of us from sneering at it) and an extreme transcription as a warning of how it may appear. While the problem you raise - of the contrast g/k or k/kʰ, there is a similar problem with representing the contrast d/t/tʰ. When you use the RTGS system, you can of course use the reminder that 'th' is pronounced as in 'Thai', 'thyme' and 'Thames'.

[Johpa]

Dear Johpa,

Thank you for your detailed explanation.

I quite understand what you mean and am in agreement with this. The problem, however. is that I am making a dictionary of Thai food-related ingredients, such as vegetables, fruit, different types of noodles, fish, flours, etc.

It will be mainly designed with the intention of assisting non-Thai speaking people interested in Thai cuisine and cooking. They will therefore be using Thai cooking books, which have been written in English. Those books employ many different methods of transliteration, so I have to (in my opinion) include common spellings used in those books and also perhaps, spellings close to the actual Thai pronunciation. I am sure that a large majority of those people have never studied Thai at all and are therefore unfamiliar with the spellings common to Thai, such as the H following a consonant, such as PH, KH, etc. I am worried that they would mispronounce them. Thai script will also be included (as well as Japanese).

As they say down south, Phuket if they mispronouce a word here or there. Pick a transliteration system and stick with it. Regardless of which system you use, you will need to describe it in your book. There will always be other books in the hands of your future readers that will use alternative systems. As long as you are consistent your readers will gert the hang of it.

[Kanga Japan]

Dear Richard W,

Thank you for your recommendations.

My initial intention was to eventually publish something in book format, however I thought that once a book is published, it cannot be revised that easily. I therefore decide to put the dictionary out in application format (perhaps separated into 3 separate ones) first. That way, any errors could be revised and additions could be made.

Using 2 transliteration systems is a good idea. It would be easy to do in book format, however the apps must be in column form. It currently has 7 horizontal columns, which really makes it wide, for an app to appear on a screen:

English

Scientific name

Thai transliteration

Thai script

Thai pronunciation in Japanese (katakana)

Japanese (kanji)

Japanese (romanized form for those who cannot read Japanese)

Perhaps I can adjust the positioning someway, to include another transliteration method.

Forgive me if I ask some stupid questions, but I am still a beginner when it comes to the Thai language, although I am doing the best I can, considering I don't live in Thailand. You have my permission to sneer (liked that). You suggested using the RTGS system and I spent some time this morning trying to find out about various transliteration systems. It seems that there are so many and there were some comments that the RTGS system was ambiguous and not widely used. Is that correct?

I do not really want to go to the trouble of adding tone marks, characters such as ɛɛ, etc. I already have close to 4,000 words in the list.

Looking forward to your further comments, when you have time. Thanks again.

[Kanga Japan]

Thanks Johpa.

I appreciate your comments. Sounds easy-going down there in Phuket.

Just in the process of considering a suitable transliteration system.

[rak sa_ngop]

Personally I think the k/kh transliteration is preferable to the g/k version.

Why?

Because I hear so many farangs who have learnt Thai, probably mainly from books, and consequently have never learnt to say the ก sound correctly. They pronounce it as they read it i.e with a hard g, a sound which immediately identifies them as non Thai.

To my ear ก sounds closer to k (unaspirated)

[Richard W]

You suggested using the RTGS system and I spent some time this morning trying to find out about various transliteration systems. It seems that there are so many and there were some comments that the RTGS system was ambiguous and not widely used. Is that correct?

Thai is an Indic script, and most of the letters can be equated with corresponding Devanagari letters. Devanagari letters have long-established transliterations into English. This gives us the graphic method of transliteration, which is widely used for names composed of Indian elements. It works very badly for native Thai words, and I think we can therefore rule it out for food names.

The usual method is more phonetic. An elaborate scheme was promulgated (I hesitate to say 'worked out') which has two variants - a 'precise' system which showed the pronunciation and also the spelling, and a 'general' system. The precise system used marks above the vowels to show vowel length (more precisely, shortness) and tone. Typographically, it was too complex for widespread use. The general system simplified the 'precise' system in two ways. Firstly, it removed the indication of the spelling, and secondly it omitted the shortness marks and tone marks. The orthography of the general system has been simplified, with the unfortunate loss of the distinction between two vowels and the loss of the distinction of two consonants, /tɕ/ and /tɕʰ/, originally written <čh> and <ch>, but now regrettably both written <ch>. However, these simplifications made it very easy to type.

Of the two systematic losses, the most important to a Thai ear is tone. Vowel length is less important. The RTGS is widely used in signposts and is the system for atlases. On signposts it competes with other systems, especially for well known names. There is also a tendency for the final consonant written in Thai to be echoed, so one can see 'Cholburi' instead of the phonetic 'Chonburi'.

There are various adaptations of the RTGS that attempt to address the ambiguity of vowel length. They are bedevilled by the interference of English spelling patterns. There are 9 basic vowel qualities in Thai, which the RTGS would write, if it hadn't dropped diacritics, as <i>, <ue>, <u> (3 high vowels), <e>, <oe>, <o> (3 mid vowels) and <ae>, <a>, <ǫ> (3 low vowels). The length discriminating systems one sees are:

For <a>, the long vowel is often written as <aa>, but <ar> is popular in closed syllables, especially when the Thai spelling has the Thai <r>.

For <i>, the long vowel is often written <ii>, but under English influence <ee> may be seen.

For <u>, the long vowels is often written <uu>, but under English influence <oo> may be seen.

For long <ǫ>, <or> is common, especially in closed syllables. Short <ǫ> might be written this way, for the the short vowel is uncommon. Note that the diacritic is not used, so the actual RTGS symbol is the ambiguous <o>.

There is no good way of writing long <e>, for 'ee' may be taken to be long <i>.

For long <o>, a popular method in monosyllables and final syllables is to add a silent 'e' as in English. One may use <oo>, but this is ambiguous.

Short <ae> and <oe> are rare enough that one does not bother to indicate the length.

The vowel quality <ue> is most typically found in a long vowel, and I have seen the convention of contrastively using <eu> to indicate the short vowel.

Unfortunately, what one usually sees is not the RTGS, but a mishmash of systems. There are, for example, at least two karaoke systems of transliteration in use, including one that makes some attempt at indicating tone.

For vowel length, is your intended audience familiar with the use of macrons? They seem to be quite common in the transliterations of Japanese names. It is possible to specify a single macron over two letters, as in a͞e, or alternatively a breve for a short vowel, as in a͝e.

[Kanga Japan]

Dear Richard W,

What can I say? Wow! You are obviously someone who has a complete grasp on the intricacies of the Thai language and has studied it intensely in a formal situation.

I cannot thank you enough for the time and effort you have put into this, on my behalf. You have explained this so well that even an 'amateur' such as myself can understand (although I did have to look up a couple of words). Thank you.

You have covered many areas which have always perplexed me, such as if I should wrote 'ii' or 'ee'. Having a background in the Japanese language, sometimes I think my way of thinking, when it comes to romanization differs from others. That is one reason I decided to add the 'katakana', which, to explain simply, is a system for Japanese to understand the pronunciation of English words. It has many downfalls, as it is based on Japanese, in which there are no TH, V, distinct L or R sounds, etc.

Can I assume that what you are saying is that I should use an adaptation of the RTGS system? Sorry, but at this stage I am not very familiar with that system. I am certain that most people who will (hopefully) be using this dictionary will not be familiar with macrons, however I feel that it would be very practical to employ this, to cover pronunciation misconceptions (for example the 'ii' or 'ee' situation). Regarding the brave for a short vowel, perhaps if I just showed one singular vowel, this would be understood. I will include an explanation of this in the dictionary.

I hope I have fully understood your recommendations regarding adopting the transliteration system.

Thank you so much.

[Mole]

I for one would recommend you to use RTGS despite its quirkiness and shortcomings, because it's used in most dictionaries. Most people who are familiar with other systems will also be familiar with RTGS while those familiar with RTGS may not necessary be familiar with other systems.

Despite RTGS don't cover tones and long/short vowels, most of the time, you will be understood even if you pronounce with wrong vowel length or tone.

You are after all not making some kind of Thai language teaching tool, but just a guide. People who really want to pursue their Thai pronunciation can look for additional comprehensive resources elsewhere.

Remember that whatever "standard" you chose to transcribe, there will always be some quirkiness involved simply because certain Latin characters will be pronounced differently by different languages. You might as well stick with one which is backed by the Thai authorities.

I also had (and still have) problems with many of RTGS' shortcomings, but in the end I'm getting used to it and is able to read things "correctly" when RTGS is used.

However, since your audience will mainly be Japanese it's also a very good idea to include the katakana as an additional guide.

For example, I'm sure it took you some time to get used to the Japanese romaji standards, and I'm sure each of them have their own shortcomings, but in the end you just have to stick to one standard across the board.

[Kanga Japan]

Thanks Mole, for your encouragement.

I think I will be using the RTGS system, perhaps with slight adjustments (macrons to indicate a long vowel), as well as a system close to the Thai pronunciation which non-Thai speakers (cooking enthusiasts, etc.) can easily read.

I think the katakana representing the Thai pronunciation will also be handy for Japanese (still not as close to the actual Thai pronunciation as it should be).

Your comments and, of course Richard W's comments have provided me with a guideline which I much needed.

[kokesaat]

In today's age of 'googling' for answers, if you're looking for ต้มยำกุ้ง recipes, but can't write Thai, do you search for tom yam kong, tom yum kong, tom yum koong, tom yum gong, tom yum goong, etc?

Recently, I helped a Thai family apply for tourist visas to visit a relative in the US. The variety of spellings for the same last names, in several cases, was mind-boggling. Some names were spelled differently on their Thai ID and passports. The relative living in the US had a Thai passport and a US passport with different spellings for the same last name. Th...... on one, T.....on another. Th.......g on one, Th.........k on another. Ch........ on one, J.......on another.

When I studied Chinese years ago, I first learned the Wades Giles system of romanization.....then after a few years, the Pinyin. If I heard a word in Chinese, I could write it in either system comfortably. I can't say the same about Thai.....despite playing with the language (on an amateur level) for 10 years, some of the signs I see around town (government produced) leave me shaking my head.

Ultimately, learning to read Thai was the best move I made in learning the language....should have never ever learned to read a Thai word except as written in Thai.

[Kanga Japan]

Dear Kokesaat,

Thank you for your comments. I know what you mean. Thai words in cookbooks and on the internet are romanized in so many ways. Even on my dictionary, I when I see a word in a book, or on the internet which think I may not have included yet, I have to go back and forward to find out where it could possibly be - sometimes starting with a G, sometimes starting with a K, etc., etc.

I remember one case here in Japan, where a Japanese woman who was married to a foreigner named Smith had her passport issued with the name Sumisu (direct translation from katakana). She had so many problems overseas.

I wish the Thai government would put together a revised, easy to understand system and have everyone stick to that. It would involve a lot of changes though.

With my luck, probably just after I have put out my dictionary, they will do so.

Thanks again. Very interesting story.

[desi]

Perhaps this'll help explain: Thai Bites - There's No Such Sound as G in Thai by Stuart Jay Raj

[Kanga Japan]

Dear Desi,

Thank you for that. I have seen it before and it is quite a controversial subject. My Thai teacher does not agree with it, but as I am still a beginner, I am not in a position to express an opinion. I would be opening a can of worms if I did.

Thank you anyway for that.

[kokesaat]

Thanks for the video.....any chance he offers the same for the ด ต d dt sounds?

[Johpa]

Thanks for the video.....any chance he offers the same for the ด ต d dt sounds?

It won't work for the other "stop" consonants because Thai does include the voiced variant of both /t/ and /p/, which are of course /d/ and /b/. So there is little chance of using the letters used for those voiced consonants to represent the unaspirated and unvoiced consonant from the same point of articulation. Using the letter h combined with the consonsant to represent the aspirated vs unaspirated is the next best solution to learning to read Thai.

And I admit find the use of "dt" rather baffling and best used for languages with even more complex consonant systems than Thai or English.

[Richard W]

An even more controversial observation in this matter is that English /g/ (and /b/ and /d/) aren't necessarily voiced either. Peter Ladefoged observed that in some native speakers' English, they have every sign of being voiced except the voicing itself. Polish speakers have more voicing than English speakers, and Polish teachers of English spend time getting them to adjust their voicing to English levels.

There are quite a few languages, especially in the USA, where <d> etc. are used for the unaspirated stops and <t> etc. are use for the aspirated stops. Chinese pinyin is the best known example of this system. The system can be extended by using <dd> etc. to represent the voiced sounds.

[Mole]

I also don't quite agree on that video.

Because in his aspirated kh sound, he does pronounce the aspiration a bit too much. In Thai, this would not exist.

Also, since I grew up in Norway, I'm also used to Norwegian, and the "g" in Norwegian corresponds exactly to the ก in Thai.

I also believe when the "g" in English is also pronounced the same as in ก. It may be due to dialect differences that the "g" may not exactly sound like a ก.

I disagree with RTGS to use "kh" for the ค ข and "k" for ก. If you look at how for example romanisation Malay, you will see that they use "k" for the aspirated k and "g" for the ก and it works fine.

It's one of the nuances with RTGS. I wonder exactly how the people behind this could reason and decide it to be like it is right now. There are many things there which I find strange how they could come to their conclusion.

I have a feeling it's based on how the letters in the Thai alphabet corresponds to the letters in the Indic alphabets. However, the problem is that many sounds in Thai has undergone a sound shift, so words in Sanskrit/Pali which begins with "g" are pronounced with "k" sound and vice versa.

So by doing things they way they do, it'll just cause confusion in modern Thai for foreigners to be able to pronounce Thai properly.

[Mole]

This probably also explains why จ and ช are both transcribed as "ch" because จ corresponds to the "ch" character in the Indic alphabets, but they're no longer pronounced like that in modern Thai.

Examples are such chandra -> จันทร์

RTGS really need to be totally revamped and the rules regarding the conversion be totally revised.

You can not base your romanization on how the Thai characters corresponds to their Indic equivalent because they're no longer pronounced the same as their origins anymore.

[desi]

Thanks for the video.....any chance he offers the same for the ด ต d dt sounds?

The videos are a part of Stu Jay Raj's Thai Bites programme at jcademy.com. Stu does have a video about the ด ต sounds, but his programme is set up to only release 1 out of every 15 or so videos to the public for free.

I could talk to him about releasing the 't' 'd' video in the future as one of the public ones. Otherwise, you can just check them out on his site.

The Thai Bites are seriously good. Here's a few more on YouTube: http://www.youtube.com/playlist?list=PLTL02WBRCaWTsClEmpagRJRjmErmRq7jd

[Kanga Japan]

Dear Mole,

Even though (or perhaps because) I am a beginner in reading Thai, I totally agree with you.

[Richard W]

One should review the contrasts /b ~ p ~ pʰ/ and /d ~ t ~ tʰ/ and then consider 'k and g'.

Also, since I grew up in Norway, I'm also used to Norwegian, and the "g" in Norwegian corresponds exactly to the ก in Thai.

I also believe when the "g" in English is also pronounced the same as in ก. It may be due to dialect differences that the "g" may not exactly sound like a ก.

I think the Germanic languages are following the path of High German as far as the initial consonants of monosyllables are concerned. <k>, <p> and <t> are aspirated in English, Low German and (most) Scandinavian languages. If you look at the chart of Danish consonants, you'll see that Danish <b>, <d> and <g> have lost voicing. Affrication of initial English <t> to [tˢ] also occurs sporadically. Icelandic <b>, <d> and <g> are unambiguously described as voiceless.

I disagree with RTGS to use "kh" for the ค ข and "k" for ก. If you look at how for example romanisation Malay, you will see that they use "k" for the aspirated k and "g" for the ก and it works fine.

That's not what the Wikipedia article says about Malay. Also the Narathiwat dialect of Thai is unusual in that it has /g/ (distinct from ก), most commonly used for Malay names with <g>.

I have a feeling it's based on how the letters in the Thai alphabet corresponds to the letters in the Indic alphabets.

Actually, ก ข ต ถ ป ผ are the anchor points, having kept their Indic sounds. The transcription probably has more to do with the perceptions of European phoneticists.