September 24, 2007

Why is it hard to learn Thai?

Learning another language is difficult. I often hear people classify Mandarin, Cantonese, Thai, and others as "the hardest language" or "one of the hardest languages." But if you were a native speaker of, say, Lao, and attempted to learn Thai, it would be vastly easier than if you were a native English speaker. In my mind, it's not really about how inherently difficult a certain language is, but rather about how different the language you would like to learn is from the language(s) you already know. Relative similarity or difference in the phonology, morphology, and syntax of a language is what counts. Are they typologically similar? That is, do they share many features--word order, for example?

Thai and English have the same basic word order: what we call SVO (subject+verb+object). That means that where in English we say "I eat rice," Thais say ฉันกินข้าว. The subject, followed by the verb, followed by the object (when there is one). This is a huge leg up for English speakers! Sure, you have to remember to say อาหารอร่อย [food+delicious] instead of "delicious food", but if you're an American, where it seems like everyone is exposed to Spanish, you may already know that it shares this word order of noun+modifier with Thai, so it doesn't seem so strange.

Then compare Thai word order with Japanese, which has SOV word order: "I rice eat", or Tagalog, which has VSO word order: "eat I rice". See how you've already got a head start?

Apart from that, however, English and Thai diverge pretty quickly. They are both generally classified as analytic languages, meaning that words tend to stand alone and that word order is usually significant for conveying a given meaning, but English has synthetic features, like attaching /s/ for plurals, or other various prefixes and suffixes.

The phonology is also quite different in many ways, which I plan to tackle in the (long overdue) continuation of my Improve Your Accent series. And I probably needn't even mention tones. If you want to boggle the minds of unsuspecting English speakers, just bust out with a minimal quintet of the five tones: มา หม่า ม่า ม้า หมา (or the classic tongue twister about whether the new wood is burning: ไม้ใหม่ไหม้ไหม ไม่ไม้ใหม่ไม่ไหม้). The flip side of this is that English has crazy spelling, some verb conjugation, many sounds that are foreign to the Thai speaker, including many difficult consonant clusters, etc.

Mastering Thai requires a commitment to learn to speak in new ways. Moving your tongue in new ways, making sounds in different ways, internalizing new word orders, and stepping outside of the way you're used to thinking about the world. It's a difficult task that requires significant self-awareness and dedication. It's not for the faint of heart, but the rewards of knowledge for those who persevere are immense.


  1. Hi Rikker,

    Interesting post, but I think sometimes Thai can be just as confusing as Japanese/Tagalog examples you gave. How about, say, the song lyric "แฟนมีก็คงต้องมา ที่แฟนไม่มาก็เพราะว่าแฟนไม่มี" which isn't SVO, but seemingly OV with S just implied, another victim of the subject pro-drop.

    The serial verb construction presents challenges in this respect too I think. A sentence like รู้สึกอยากวิ่งกลับไปบอกเขาว่า... has 6 six verbs in a row and still sounds natural enough in Thai, but I think it would challenging to get anything like that in English!

    My first comment here by the way but I read regularly, keep up the good work! :)

  2. Mike, you're right, of course. The post is at best a superficial treatment of the issue, but it's mostly meant to give my view on why learning language Thai is difficult (i.e. because English are Thai are different in so many ways), but also pointing out that new learners have a leg up as far as basic word order.

    Once you get past the surface, Thai has myriad complexities that make it not easy at all. Those are the differences I alluded to (but didn't explicate). So I don't know if I accomplished my goal very well, but so it goes.

    That's a great sentence, by the way. How about รู้สึกอยากวิ่งกลับไปบอกให้รู้ว่า... that's 8. :)

  3. Rikker, I disagree that your "conceptual distance" theory is all there is to it. All languages are more or less equally complex (too simple, and they're not useful; too complex, and they're not comprehensible). But I think some people have a propensity for some language features, just as they do for, say, academic subjects or sports.

    I had French in school from third grade until eleventh grade. In college, I took Spanish. Through all of that, I was probably about average compared to my classmates. Five years later, I don't remember a thing.

    However, after four months of living in Thailand, I was already having conversations and had surpassed a good portion of foreign residents. By one year, taxi drivers started asking me if I grew up here or if I was half-Thai. (My answer was always ใช่ครับ เป็นลูกครึ่งจริงด้วย คุณพ่อเป็นคนฝรั่งเศส คุณแม่เป็นคน เยอรมัน but the humor was usually lost on them.) For years now, I have been placing phone calls without revealing that I am not Thai. And sometimes, when everything works out just right, I have frightened people by speaking from behind them (they get the sick feeling of an unknown and previously unaccounted-for presence, like when your clock radio kicks on unexpectedly and a talk show fills the room.

    I'm comparing apples to oranges, but even given the immersion in Thailand and motivation to communicate, I'm sure there's something about the way I think and the way Thai works that helped me pick it up. Really, I just notice how people say things and then emulate that later on. But in Thai, that will get you pretty far, with its extremely regular structure.

    Also, I feel like I constantly notice coincidental similarities between the two languages --similarities that don't get enough exposure.

    Finally, I am fighting to debunk the myth that tones are hard. People are afraid of them, and too few foreigners are willing to sit down and drill the fundamentals. But the reality is that tonal meaning is built in to our brains. In English, tone makes a statement, or a question, or a sarcastic joke, or a flippant insult. A misplaced English tone can make you look uneducated (stressing "police" on the first syllable as in the rural South) or ridiculed (as in the nearly-naughty Southern version of "cement" -- yes I am from the South).

    In Thai, a tone spells a syllable and granted, it is comparatively inflexible there. But that's all there is! There are five of them and then we're done. Distributed over the two syllable lengths and the stop vs. sonorant finals, that's 5x2x2=20 combinations we can drill to master practically every sound in the Language. To finish, throw in a handful of consonants, none of which are completely foreign to us. Say "apple." Congratulations! You just learned "ป"; say, "Wrong. Ah ha! Met?" -- you just said, "I am waiting for five sesame seeds."

    Well, to summarize, I believe some people can pick up Thai or Asian languages more easily than they think because once you get past pronunciation, there are few obstacles to being conversant. Hell! Once you get past pronunciation, you'll speak Thai better than half the Thais in Thailand. The answer? First, foreigners have to drill the fundamentals of pronunciation systematically (Thais learn it "for free" as toddlers). Immersion won't cut it if you don't have the building blocks, and the truth is, many foreigners learn Thai from... well... their co-workers if they're lucky. Second, after you have a foundation, it's just a matter of listening to how people phrase things and developing the same usage.

  4. Jason, great comment. On the contrary, though, I wasn't arguing for "conceptual distance," but rather "typological distance." And I didn't mean to give the impression that I think that's all there is to it.

    What do I mean by typological distance instead of conceptual distance? Well, I'm not claiming that Thais "think differently" than English speakers do, or that concepts in Thai are inherently "harder" than English. Sure, there are different concepts that you have to get used to--like distinguishing light blue and dark blue systematically, with no true general term for blue. But in general, I don't dogmatically accept the Sapir-Whorf hypothesis, which claims that the language we learn to speak governs how we think. Though I do generally think it influences it, as a matter of habit.

    The idea of "distance" between two languages is something that independently occurred to me, but I've since learned (in the last few days since I wrote this post, actually) that it's a common idea in language learning theory (although sometimes it is termed "cultural distance," which may lean more towards the "conceptual distance," as you've called it).

    The study of linguistic typology is another thing entirely. Rather than make any claims for language being related to one another, as is done in drawing "family" connections, typology is simply about categorizing the breadth of language variation, without respect to relationship. Many languages share features by coincidence, that do not necessarily require any sort of contact or genetic relation between them (genetic in the linguistic sense, meaning derived from a common language).

    My point in bringing typological features into the argument was to claim that if the typological features to one's native language (L1) are very similar to the learning language (L2), then the language will be easier to learn. It's kind of like saying if you know how to drive an automatic Chevy, it'll be easier to learn to drive an automatic Ford, since most of the gauges will look the same, and the basic motion is the same, even if some of the switches and buttons are moved around. Compare this to learning to drive a manual transmission, or learning to drive one of those fully hand-controlled car, like for paraplegics. You begin to have to learn to drive all over again. (This is an imperfect comparison that should not be taken too far. I may have taken it too far already. So it goes.)

    Much of typology is about word ordering. Things like basic word order (SVO, SOV, etc), which I mentioned above, but also grammatical case, verbal agreement, its sound system. Sometimes there is "conceptual" classification, like the common distinction between "grue" and "non-grue" languages (that is, languages which always distinguish between blue and green, like English, and those that don't, like Khmer--grue comes from combining green+blue).

    And as I've learned, the idea of language distance is pretty well documented as having a real influence. Here's an interesting paper (in DOC format, or here's a Google HTML cached version) about Portuguese speakers learning English, for example.

    This needn't have come as any surprise to me, since it's really quite a natural application of what we've been observing for a long time: languages can be grouped into "families" based on certain characteristics.

    The paper above is particularly fascinating, because it talks about perceived language distance, and how Portuguese students of English didn't take full advantage of the similarities between L1 and L2 if they perceived the languages as being very different from one another. This is not constant, however. During the first stages of learning, a lot of parallels between L1 and L2 are drawn by the learner, and similarities are taken advantage of. But as time goes on, students start to notice more and more differences, and in fact start making silly mistakes.

    Re-reading my post, I think that this is what I was getting at--that "perceived distance" makes learning harder. If we can learn to draw the connections where they exist, then learning gets that much easier when we see the various "legs up" that we have. I pointed this out in my first paragraph by saying that Westerners like to classify Asian languages as among the "hardest" languages, but of course that's relative to what your L1 is. And even if you are a native English speaker, learning the similarities will lighten the "learning burden" of the language.

    All of this may or may not be relevant to the argument at hand. Mostly, I want to clarify that I'm not making a claim about conceptual distance, but rather about distance between relatively concrete features of the language--word order, pronunciation, grammar, and other prosodic features (tone, stress, intonation, etc).

    I'd like to make a few more comments on your statements about tones. I don't believe tones are inherently difficult, just that they are different from how speakers of non-tonal languages are used to conveying meaning. Pitch and contour in English usually expresses connotative meaning, while in Thai tone (its pitch and contour) generally express denotative meaning. This is an extremely simplistic explanation, but allow it for the sake of argument. Learning to control these things requires learning to communicate in a different way (both in producing the correct tone and suppressing the learned tendency to express mood, truth-value, etc. via tone of voice). It's difficult if you've never done it. But it's not "more difficult" than learning to communicate in the way that English speakers do. Do you agree with that?

    Jason, it sounds like you and I have had similar experiences both in the way we learned the language (through immersion among and conscientious imitation of natives, in addition to intensive personal study--is that right?), and in that we both seem to have a knack for it (I was also getting asked if I was born in Thailand or am a ลูกครึ่ง after a short time, and still regularly do, and often get mistaken for a Thai on the telephone).

    However, I have to disagree with the sentiment in the last paragraph, "Hell! Once you get past pronunciation, you'll speak Thai better than half the Thais in Thailand." That's only maybe true if you want to speak like a broadcaster. I'm a descriptivist, so I believe that the Thai Language exists as the people speak it, not as it's taught in a textbook or prescribed by the Royal Institute or anyone. So if you ask me, any five year old Thai kid knows a lot that I don't--even if I, in turn, know a lot he doesn't about his own language, too.

    There's just so much cultural and lexical content that is absorbed over the years (and makes up the native speaker's receptive knowledge of the language, eve if it's not productive--think idioms and allusions and register) that leaves us ever playing catch up. But what a fun game it is. That's not to say that folks like us don't have a more systematic understanding of the language in many ways, but in real life practice, a native speaker pretty much always wins. Or so I think.

    Thanks for the comments, and for making me think harder about the topic. Please pardon the longwinded and probably over-esoteric response.

  5. I found what looks like a good article about the "language distance" (in PDF format):

    Performance on ESL Examinations: Is There a Language Distance Effect?

  6. Rikker, you might as well have turned this into a supplementary post :)

  7. Well, what started out as a simple comment sort of devolved into a rambling rant from me. (Oh no! Twice!) Of course, generally speaking, Spaniards learn Portuguese faster than they learn Russian, and so on.

    Strangely, what I was really thinking when I said "conceptual distance" was hamming distance, which is an engineering concept and only vaguely related. The idea is to compute the distance between two messages based on the number of components that differ.

    I didn't know about grue; that's pretty cool. I got really excited when I learned that Thai considers dark blue and light blue to be both first-class colors. I get a satisfaction when I get to use one or the other in conversation. I heard that there's another language out there that does the same thing for some shade of red or purple or something.

    About perceived distance vs. actual distance, that is really the source of my strong reaction against the "Thai is hard" mindset. I think great strides could be made in teaching Thai to English-speakers if they could be shown just how similar they can be. This is why krav maga is successful: everything builds from your in-born behavior.

    Also, I think that many foreigners would benefit from learning Thai from other foreigners because many Thais can't explain how they work unless they are very observant and good at introspection. (The same is true learning English from Americans in my experience.) Other foreigners can explain things in terms of things you already know.

    About tones, yes we're in agreement. My main point is that while they are foreign and require effort, tones are not this insurmountable challenge that many people assume they are.

    The way I always thought of Thai vs. English is that Thai is a very pronunciation-critical language, and English is a very syntax-critical language. Thai pronunciation is like a gymnast's balance beam: one slight deviation and you're done. This is exacerbated because Thais are only accustomed to hearing the native accent, whereas English speakers are more used to foreign accents from the melting pot and mass media. Anyway, while English gives a lot of pronunciation leeway, particularly with vowels, you can't screw up the articles, helping verbs, word order, and the whole syntax generally. If you get your word order wrong, you might turn a statement into a question. Well, that's my own pet theory at least.

    Unfortunately, while I enjoy picking up Thai where I can, my personal study could be best described as casual. I'm do computers here, and that takes pretty much all of my time.

    While I'm on the subject, I'll come out and say my contentious and disputed theory of Thai discouragement. Here goes. Keep in mind that I am explaining perceptions, especially during first impressions. Not realities.

    The sad fact about I.T. in Thailand (and I suspect most professional settings) is that speaking Thai is culturally discouraged, for two reasons. First, my customers are managers who have subordinates and personal pride. They speak to foreigners in the modern Latin, period. They wouldn't be caught dead having to resort to the poor man's language with a foreigner! Are you kidding? Indeed, most of my conversations with customers are slower and more painful than they could be in a perfect world.

    Second, foreigners in Thailand tend to fall into two categories: the cream of the crop, and the bottom of the barrel. Guess which group tends to speak Thai more. The cream of the crop is usually consultants, experts, or executives flown in at great expense to do good things for the organization. The bottom of the barrel hangs around for years because he's not competitive back home and he skates by from the low cost of living.

    You may argue that there's a huge difference between foreigner who speaks business Thai vs. one who speaks bar Thai. Well, my third point is that foreigners who speak proper business Thai are suspicious. A talented foreigner should be working with Thais who are smart and educated and who can speak English well. If a foreigner has "been forced" to learn Thai because his peers and subordinates don't speak English, what does that say about him?

    Fourth, some Thais (not all) unknowingly consider the Thai language to be the exclusive domain of Thais: "unlearnable" by foreigners, the secret handshake between brothers. You never know which individuals feel this way, but by speaking Thai in a professional setting you run the risk of setting off the silent internal intruder alerts.

    As much as I hate it and wish it were not so, that is my assessment of using Thai in a professional setting. That is my opinion and I am confident of it, but plenty of people have strongly disagreed with me here. But I have to admit, when I made that realization, it was a very discouraging feeling and it caused my Thai progress to plateau out for a couple of years.

    Finally, about speaking Thai better than the Thais, I'm making the same point as you, in joke form. The "Thai" language means different things in different contexts.

  8. wow, a whole lot of arguing about a completely subjective topic. maybe we need a a different, micro perspective. like what does the nature of the thai language do to us? i'm a firm believer that because i didn't have to variate thai words due to the complete absence of conjugation i became predisposed to eat phad-gra-phow every day, every meal.

  9. And the nature of the Thai language led me to be officially too lazy to learn any language with conjugation. I can see myself delving deeper in Khmer, Lao, or Burmese, but when I think back to my high school Spanish, my head still hurts. Which totally conflicts with my desire to learn more Pali and Sanskrit. Yeargh.

    Which explains why for now I'm just sticking to Thai.

  10. I think the difficulty - even for Thais to understand one another at times - is because Thai language is a topic-prominent language (not necessarily "subject" of the sentence, not necessarily "theme" - as in theme-rheme) and we use a lot of zero-substitution (ellipsis) and ergative verbs. The deitic turn belongs to the speaker. Context is everything.
    In Mikes'example: "แฟนมีก็คงต้องมา ที่แฟนไม่มาก็เพราะว่าแฟนไม่มี" - subject is zero-substituted (omitted) - we do not know whom the speaker is referring to - himself/herself, he, she, they? - "who" has, or have or had แฟน (singular or plural?), while แฟน is the sentence topic (while the discourse topic might be about the zero-substituted person - that I, he, she,or they has/have no แฟน/แฟนs).

    About serial verbs
    Since one word can serve several functions and Thai language does not have the infinitive form.
    In Mike's example: "รู้สึกอยากวิ่งกลับไปบอกเขาว่า..." I should think that "กลับไป" in this case/context are not verbs but adverbs - a doublet perhaps - to make it clearer that กลับ in this case is not a verb (which means พลิก เปลี่ยน เช่น กลับตัว กลับใจ) but an adverb (back)and that is the function that "ไป" serves here. วิ่งกลับไป - run back to where he came from.

    An example: "เขากลับบอกว่าฉันพูดผิด" - กลับ here is not a verb - "instead [or on the contrary], he told [someone]that I lied," or it can be considered an equivalent to "ดัน" - เขาดันบอก (ใครก็ไม่รู้) ว่าฉันพูดผิด Compare with "เขากลับไปบอกว่าฉันโกหก" กลับ here is a verb - "to return [to someone. somewhere and teld that person/s that" " or it might not be a verb as in "on the contrary [or instead] he [went up to someone, somewhere to] tell [that person] that I lied" - it depends on the pause/stress or in written from depends on the context).

    By the way ไป and มา are interesting words that often confuse me. It's a deitic word. เสฐียรโกเศศ and พระสารประเสริฐ used them in different ways. One used จดหมายมายัง... to begin his letter, while the other replied with จดหมายไปยัง...

    Thank you for your great blog! It is wonderful to have well-informed foreigner scrutinizing Thai language and contributing to its development.

  11. Bancha, thanks for your very good comments. You clearly know your stuff.

    On ไป and มา: the deictic* markers, aren't acting as verbs in any real sense, and as you rightly point out, กลับ, etc. aren' either.

    That doesn't change the fact that it's confusing for foreigners who lack the training in formal linguistics to understand why there are so many of what appear to be verbs, and what are called verbs in the dictionary. So I think his point stands.

    ไป/มา are indeed very complex and subtle. Consider talking on the phone. If I make reference to the fact that I'm calling them, do I say ไป or มา? Well, from my perspective, I โทรไปหา; from their perspective, I โทรมาหา. Seems simple. But if I say "I'm calling because", I'd say ผมโทรมาหาคุณเพราะว่า...

    But if the next day I say "I called you because", wouldn't I say (เมื่อวาน)ผมโทรไปหาคุณเพราะว่า....

    My intuition for the reason behind this is that มา is used when the conversation is still ongoing, because you're both in the same conceptual location--part of the same phone call.

    When you talk about it the next day, ไป is appropriate because it's over, and neither of you are in that phone call anymore. You're both in a different conceptual place now.

    Now, if I tell them I'm going to call back later, should I use โทรกลับไปหา or โทรกลับมาหา? Either seems possible. But I don't trust own intuition here.

    I once wrote a (not very good) paper on the deixis of Thai ไป/มา for a class on semantics in college. I wonder if I still have it...

    Anyhow, Bancha, your comments are an excellent addition to the conversation here. Please keep commenting!

    (*Not to make a fuss, but deitic is a misspelling. It's the adjective form of the noun deixis, and is pronounced dike-tick.)

  12. It is really impressing to hear foreign people speak Thai because it is a tonal language and unless you "sing" the word in exactly the right way you probably won't be understood.

  13. Rikker is right. The degree of difficulty in learning Thai is related to what you are already familiar with. If you speak Lao, Thai is a piece of cake. It's easy to understand เข้า ใจ ง่าย kôw jai ngâi. Especially if you are from Vientiane.
    But if you come from Luang Prabang, I suspect you would have some trouble using 5 tones instead of 6.
    I used to teach Thai to foreigners. Those who picked up "the tone thing" the quickest often had a musical background. They were used to hearing distinctions in pitch.
    I've made a web page that focuses on Thai tones. It's got audio and video. Check it out.

  14. Rikker,

    Thanks very much for the comprehensive and inspirational comments!

    Gene Panasenko

  15. Jason,

    Nice comments!

    Gene Panasenko