Are some languages more "complex" than others?

It is a dogma of linguistics that "all languages are equally complex," but have different ways of being complex. For example, English lacks the inflections of French, but has complexity in the form of phrasal verbs ("look up" = research, "look out" = beware, "give up" = accept defeat, "come across" = encounter) and other idioms that must be memorized ("by yourself", "in person"). Some languages have weirdnesses like evidential markers (though one of those,  Eastern Pomo language, seems complex in other ways as well).

I am NOT a linguist (though some of the topics intrigue me enough that by now I've read dozens of books or papers), and only know three languages (English, French, Thai). But I just do NOT see how Thai can be called "complex." We have 2 or 3 linguists here and I hope they can show me what I'm missing. Or refer my question to a linguists' message-board.

(I know ZERO Chinese but have the impression its similarities to Thai may make it an exception also.)

I do feel that I know Thai well enough to point to what might be considered complexities:

• The written language. Thai has about 65 writing symbols compared with English's 26. And it has weirdnesses: 'TR' becomes an 'S' sound, and 'RR' becomes a schwa! But I think linguists and this complexity meme are focused solely on the spoken language, right?
• Classifiers. English also has special words to deal with "uncountable" nouns: heads of cattle or lettuce, etc. The fact that ALL nouns in Thai are treated, in a sense, as "uncountable" is almost a simplification.
• Pronouns. Thai uses a variety of pronouns depending on social status or personal relationship. But, like English, it also often uses ordinary words ("Sarge," "Doctor," "Dad") when a pronoun might be used.
• Consecutive verbs. I've previously posted an example Thai sentence of 13 words: 1 pronoun followed by 12 verbs! Again, this seems like simplification. One could construct such a sentence in English but connectors like "and," "to," etc. would be needed.
• Sentence-ending particles. Thais often end a sentence with a syllable or two showing mood or status. English does this also: "Sir", "hmmm."
• Tones. Again, English also uses tones, but in a more complex way. "Isn't she pretty" has different meanings depending on where the high-tone is placed.

Thai has no obligatory markers for tense, number, gender. Even the non-obligatory markers are ambiguous. A word for 'already' and a verb for 'to obtain' are perhaps the most common past-tense markers, but the latter is also used to mean 'to be able to.'

What am I missing? Surely professional linguists are well aware of Thai; in what way can it uphold the meme (which linguists seem to treat almost as an axiom) that "all languages are equally complex"?

To the contrary, the language is so simple that ordinary conversations are often full of ambiguity. I've witnessed examples of this as a bystander. Sometimes what would be a single statement in English necessitates extra turns. For example the English statement "We're going to see a movie" might become "Go see movie"; "Go already?"; "No, we go afternoon."

Some sorts of complexity are often less complex than they might seem at first sight. Consider noun cases of the older or more conservative Indo-European languages: multiple declensions, sets of their manifestations, that differ between singular and plural. This complexity goes back to Proto-Indo-European.

But many languages have much simpler sets of noun cases that are essentially attached postpositions, like prepositions but following noun phrases. To elicit some of them, I used these example sentences:

The dog is chasing me.
The dog's fur is black.
I gave the dog some water.
I am chasing the dog.

Repeated for "dogs", "horse", and "horses". These sentences elicit the nominative (subject), genitive (of-case), dative (to-case), and accusative (object) cases.

Spanish was much like English, with singular vs. plural only: perro, perros, caballo, caballos, but it has article-preposition combinations: de el > del, a el > al.

Spanish is descended from Latin, with canis, canis, canî, canem - canês, canum, canibus, canês - equus, equî, equô, equum - equî, equôrum, equîs, equôs.

But Latin did not have a a definite article, with the Romance ones come from its demonstrative ille "that".

Croatian has pas, psa, psu, psa - psi, pasa, psima, pse - konj, konja, konju, konja - konji, konja, konjima, konje.

Finnish has koira, koiran, koiralle, koiraa - koirat, koirien, koirille, koiria - hevonen, hevosen, hevoselle, hevosta - hevoset, hevosten, hevosille, hevosia

Instead of the dative, it's the "illative" case. But one can recognize some regularity: genitive -n, illative -lle, accusative -a, nominative plural -t, oblique (non-nominative) plural -i- (case ending).

Hungarian has kutya, kutya, kutyának, kutyát - kutyák, kutyák, kutyáknak, kutyákat - ló, ló, lónak, lóvat - lóvak, lóvak, lóvaknak, lóvakat.

Even more regular. No genitive case, but dative -nak, accusative -t, and plural -k.

Turkish has köpek, köpeğin, köpeğe, köpeği - köpekler, köpeklerin, köpeklere, köpekleri - at, atın, ata, atı - atlar, atların - atlara, atları

Genitive -in, dative -e, definite accusative -i, and plural -lar.

Tamil has nāy, nāyiṉ, nāykku, nāyai - nāykaḷ, nāykaḷiṉ, nāykaḷukku, nāykaḷait - kutirai, kutiraiyiṉ, kutiraikku, kutiraiyait - kutiraikaḷ, kutiraikaḷiṉ, kutiraikaḷukku, kutiraikaḷait

Genitive -in, dative -kku, accusative -ai(t), and plural -kal.

Quechua has genitive -p, dative -man, accusative -ta, and plural -kuna.

Etc.

"all languages are equally complex"

Click to expand...

Maybe a good starting point is to find out what linguists mean when they say this?

This article has some references to the idea of equal complexity:

‘All Languages Are Equally Complex’ The rise and fall of a consensus

This quote stood out to me:

There are no ‘primitive’ languages — all languages are equally complex and equally capable of expressing any idea in the universe. (Fromkin & Rodman 1983: 16)

Click to expand...

This reminded me of Turing completeness. If a programming language is Turing complete then it can execute any algorithm. Both C and C++ are Turing complete. C++ has more language features allowing the programmer to express algorithms with a more concise grammar, but there's no algorithm that C++ can run that C cannot.

So maybe you could ask: what meaning can I express in English that I can't express in Thai?

"Complexity" is not an inherently meaningful concept; whenever someone tries to reduce it to a quanitifiable metric, they accidentally betray some personal or cultural bias in what arbitrarily chosen feature might make a phenomenon "feel complex" to them in a way that someone else from a different background wouldn't naturally agree with.

Sounds to me as though the entire futile and pointless debate could have been avoided if only Fromkin, Rodman, et al.,* had chosen to co-opt the word "complete" (instead of "complex") as the one word summary of what they were trying to say.









* Professor Alan Smith, usually credited just as 'Al', is one of those unsung polymath geniuses who seem to have made significant contributions to pretty much everything. He and Ibid are the greatest authorities on the planet, and no research is complete without their contribution.

A possible measurement of degree of complexity is degree of modularity: the more modular, the less complex.

Thus, in my examples, Latin and Croatian have number-case combinations that cannot be analyzed into separate number and case markers, and more than one set of these combinations, while Finnish, Hungarian, Turkish, Tamil, and Quechua have (word root) - (number marking) - (case marking) at least partially separable. Most noun cases are syntactically comparable to prepositions, and English and Spanish have (preposition) - (word root) - (number marking), with a similar level of modularity.

So by that standard, Latin and Croatian have more complex noun inflections than the others that I'd mentioned.

A preposition can be moved to the other side of its noun phrase, making it a postposition. More generally, prepositions and postpositions are adpositions. If a postposition becomes suffixed to a word, it becomes a noun-case ending.

Quechua examples: allqu, allqupa, allquman, allquta - allqukuna, allqukunapa, allqukunaman, allqukunata - kawallu, kawallupa, kawalluman, kawalluta - kawallukuna, kawallukunapa, kawallukunaman, kawallukunata

Genitive -pa, dative -man, accusative -ta, and plural -kuna.

The Quechua word for horse was borrowed from Spanish.

If they're become part of the lexeme, they're affixes, not adpositions.

bigfield said:

"all languages are equally complex"

Click to expand...

Maybe a good starting point is to find out what linguists mean when they say this?

This article has some references to the idea of equal complexity:

‘All Languages Are Equally Complex’ The rise and fall of a consensus

https://sci-hub.st/https://doi.org/10.1075/hl.39.2-3.08jos

Click to expand...

Thanks for the lead! It will take me some time to read the whole paper, but the title ("The rise and fall of a consensus") shows the theme. Long ago, it was assumed that "savage" or "barbarian" cultures had primitive languages. Complex societies had complex languages.

This simplistic view was called into question when Chinese was studied. China was NOT barbarian but DID have a simple grammar:

19th century thought said:

Where matters came to a head was with regard to Chinese, especially Classical Chinese, seen as a language of maximum structural simplicity that was the vehicle of one of the most advanced cultures the world had known. Schlegel described Chinese grammar as “extremely poor, being in its basis extremely, not to say childishly, simple and quite ungrammatical”
. . .
The thesis is that Chinese has all the excellence Rémusat imputes to it, and that this excellence resides in the relation of language to pure logic, which stands above human thinking. The antithesis is that Chinese has the imperfections implied by Burnouf, and that these reside in the relation of language to style, in which human thinking is developed to its highest form. The synthesis — which, methodologically, represents the closest approximation to the truth — is that each of these views is correct within its particular realm: Chinese has the ideal structure for the formulation and expression of pure ideas, while the inflecting languages, particularly Sanskrit and Greek, have the ideal structure for the formulation and expression of human thought.

Click to expand...

Anyway, savage people with complex languages also turned up and by the early 20th century, the meme of equal complexity may have arisen as a counterfactual anti-racism, which today might be called "wokeism"!

Not all linguists bought into the meme: Sapir was a renowned linguist:

Edward Sapir said:

Both simple and complex types of languages of an indefinite number of varieties may be found spoken at any desired level of cultural advance.

Click to expand...

Nevertheless by the late 20th century the meme was very widely accepted.

More recently however, many linguists are rejecting the meme, for example:

Daniel Everett claims to have found such a language [i.e. lacking recursion], namely Pirahã, spoken in Brazilian Amazonia (see Everett 2005, 2008, 2009). Reinforcing his claim that Pirahã is less complex than other languages, Everett has also argued that this language lacks quantifiers, numbers, colour terms, and much more.

Click to expand...

The paper Bigfield cites has one paragraph mentioning Thai

Another example that has been cited to support the idea of complexity trade-offs is that Chinese and typologically similar languages, which have a simple (isolating) morphosyntax and individual morphemes that are multiply ambiguous, tend to have both constructions situated at the intersection between the lexicon and productive syntax like classifiers, reduplication, compounding, and verb serialization(see Riddle 2008 for Hmong, Mandarin, and Thai) and complex rules of inference and complex rules interfacing form and meaning (see Bisang 2009 for Khmer, Thai, and late Archaic Chinese).

Click to expand...

I am interested in the phrase I've reddened and will try to obtain Bisang 2009, “On the Evolution of Complexity: Sometimes less is more in East and mainland Southeast Asia”.

But what about reduplication? In Thai, (the equivalent of) 'child child' means 'children' and 'pretty pretty' means 'very pretty'. Is this really an example of "complexity"? I mentioned the sentence with 12 consecutive verbs as an example of verb serialization. Is that really an example of "complexity"?

As if my "homework" weren't already extensive I downloaded a paper on verb serialiation:

bigfield said:

This quote stood out to me:

There are no ‘primitive’ languages — all languages are equally complex and equally capable of expressing any idea in the universe. (Fromkin & Rodman 1983: 16)

Click to expand...

This reminded me of Turing completeness. If a programming language is Turing complete then it can execute any algorithm. Both C and C++ are Turing complete. C++ has more language features allowing the programmer to express algorithms with a more concise grammar, but there's no algorithm that C++ can run that C cannot.

So maybe you could ask: what meaning can I express in English that I can't express in Thai?

Click to expand...

Few of us will doubt that any language is "complete" in that trivial sense. A culture unfamiliar with smart-phones might need to replace that single word with a paragraph but it could be done. Recently I mentioned Toki Pona, a constructed language with less than 200 words and simple grammar. Recently someone has written a longish story in Toki Pona.

What WOULD be a way to quantify "completeness" is to quantify the effort needed to complete a conversation. I've already mentioned Thai's use of three turns instead of one, e.g. a Question/Answer exchange to clarify a "hidden" variable which a speaker of English (with obligatory tense markers) couldn't hide if he/she wanted to.

I was in a hurry this morning, and threw together my recent post quickly. Let me re-examine it.

Swammerdami said:

19th century thought said:

Where matters came to a head was with regard to Chinese, especially Classical Chinese, seen as a language of maximum structural simplicity that was the vehicle of one of the most advanced cultures the world had known. Schlegel described Chinese grammar as “extremely poor, being in its basis extremely, not to say childishly, simple and quite ungrammatical”
. . .
The thesis is that Chinese has all the excellence Rémusat imputes to it, and that this excellence resides in the relation of language to pure logic, which stands above human thinking. The antithesis is that Chinese has the imperfections implied by Burnouf, and that these reside in the relation of language to style, in which human thinking is developed to its highest form. The synthesis — which, methodologically, represents the closest approximation to the truth — is that each of these views is correct within its particular realm: Chinese has the ideal structure for the formulation and expression of pure ideas, while the inflecting languages, particularly Sanskrit and Greek, have the ideal structure for the formulation and expression of human thought.

Click to expand...

Click to expand...

I do NOT fully understand the sentence I've reddened. Any volunteers willing to express this idea with smaller words?

The MAIN conclusion I derived from the paper is that "Equal complexity" is a myth, perpetuated by a form of "political correctness"! This is supported by quotes from Edward Sapir and Daniel Everett. Sapir is a giant of linguistics and needs no introduction.

Swammerdami said:

Edward Sapir said:

Both simple and complex types of languages of an indefinite number of varieties may be found spoken at any desired level of cultural advance.

Click to expand...

Click to expand...

Wikipedia has a sentence which suggests the importance of Everett (author of several books):

Wikipedia said:

In 2016 Tom Wolfe published a book, The Kingdom of Speech, in which he discusses work of four major figures in the history of the sciences of evolution and language, Charles Darwin, Noam Chomsky, Alfred Wallace, and Daniel Everett.

Click to expand...

Placed on a pedestal with Darwin, Chomsky, and Wallace? High praise indeed! And, YES, that is the famous fiction writer Tom Wolfe. I herewith add another book to my must-buy list.

Swammerdami said:

Daniel Everett claims to have found such a language [i.e. lacking recursion], namely Pirahã, spoken in Brazilian Amazonia (see Everett 2005, 2008, 2009). Reinforcing his claim that Pirahã is less complex than other languages, Everett has also argued that this language lacks quantifiers, numbers, colour terms, and much more.

Click to expand...

Click to expand...

But I still want to understand what people mean when they say Thai is as complex as English or French.

Swammerdami said:

Another example that has been cited to support the idea of complexity trade-offs is that Chinese and typologically similar languages, which have a simple (isolating) morphosyntax and individual morphemes that are multiply ambiguous, tend to have both constructions situated at the intersection between the lexicon and productive syntax like classifiers, reduplication, compounding, and verb serialization(see Riddle 2008 for Hmong, Mandarin, and Thai) and complex rules of inference and complex rules interfacing form and meaning (see Bisang 2009 for Khmer, Thai, and late Archaic Chinese).

Click to expand...

I am interested in the phrase I've reddened and will try to obtain Bisang 2009, “On the Evolution of Complexity: Sometimes less is more in East and mainland Southeast Asia”.

But what about reduplication? In Thai, (the equivalent of) 'child child' means 'children' and 'pretty pretty' means 'very pretty'. Is this really an example of "complexity"? I mentioned the sentence with 12 consecutive verbs as an example of verb serialization. Is that really an example of "complexity"?

Click to expand...

I ordered eye-glasses in Bangkok and the lady reminded me to come "by yourself" to pick them up. I laughed and told her the idiom she wanted was "in person." Who could blame her? These are idioms without particular logic in the preposition choice that just need to be memorized. To claim that using "pretty pretty" to mean "very pretty" is a similar example of complexity reminds me of the title of the famous movie An Inference Too Far.

SUMMARY: Many top linguists agree that Thread question ("Are some languages more complex than others?") should be answered Yes. Those who say otherwise are largely parroting (out of "political correctness"?) a dogma from decades ago.

Swammerdami said:

Another example that has been cited to support the idea of complexity trade-offs is that Chinese and typologically similar languages, which have a simple (isolating) morphosyntax and individual morphemes that are multiply ambiguous, tend to have both constructions situated at the intersection between the lexicon and productive syntax like classifiers, reduplication, compounding, and verb serialization(see Riddle 2008 for Hmong, Mandarin, and Thai) and complex rules of inference and complex rules interfacing form and meaning (see Bisang 2009 for Khmer, Thai, and late Archaic Chinese).

Click to expand...

I am interested in the phrase I've reddened and will try to obtain Bisang 2009, “On the Evolution of Complexity: Sometimes less is more in East and mainland Southeast Asia”.

Click to expand...

The paper I seek is a chapter in a book whose abstract begins

This book presents a challenge to the widely-held assumption that human languages are both similar and constant in their degree of complexity. For a hundred years or more the universal equality of languages has been a tenet of faith among most anthropologists and linguists. It has been frequently advanced as a corrective to the idea that some languages are at a later stage of evolution than others. It also appears to be an inevitable outcome of one of the central axioms of generative linguistic theory: that the mental architecture of language is fixed and is thus identical in all languages and that whereas genes evolve languages do not. Language Complexity as an Evolving Variable reopens the debate.

Click to expand...

Bisang has a page listing his papers but my target is not available for Download. Instead I downloaded his "Grammaticalization and the areal factor: The perspective of East and mainland Southeast Asian languages" and make the following brief comments:

Bisang claims that Thai and other languages have a "high degree of grammaticalization" (with classifiers and rigid word order given as examples of grammaticalization). Does this imply "complexity"? Let's not go too far down this rabbit-hole, but here are comments on, and an example of, the use of classifiers:

The process of classification can be used to profile conceptual boundaries of concepts. Due to this function, classification is the basis of the two main functions involved with numeral classifiers, i.e., identification and individuation. On the one hand, classification helps identifying a certain sensory perception by using its conceptual boundaries to highlight that perception against other sensory perceptions. On the other hand, it can
establish a sensory perception as an individual item by actualising its salient inherent properties which constitute it as a conceptual unit.

Click to expand...

What does that even mean? After a while he gives an example Thai sentence translated into English as
The words translated as 'THIS' and 'THAT' are preceded by ... classifiers! The complexity! The complexity!

Does rigid word order represent "complexity"? Bisang offers an 8-word Thai example sentence and translates it into English:
The ordering of the 8 words admits no deviation. Does a rigid word order represent complexity??

I played with this example a little, discovering that Google Translate would produce the first 7 words of Bisang's 8-word Thai sentence when presented with
which translates word-for-word as
In brackets I've appended 'แล้ว/already', a tense marker Google omits.

SUMMARY: Many top linguists agree that Thread question ("Are some languages more complex than others?") should be answered Yes. Those who say otherwise are largely parroting (out of "political correctness"?) a dogma from decades ago.

Swammerdami said:

SUMMARY: Many top linguists agree that Thread question ("Are some languages more complex than others?") should be answered Yes. Those who say otherwise are largely parroting (out of "political correctness"?) a dogma from decades ago.

Click to expand...

Which "top linguists" are those? I'd never even heard of Walter Bisang before reading this thread. Had you heard of him before you before you started looking for evidence to justify your views? In what other avenue? And neither you nor Bisang have proposed any coherent, objective definition for "complexity", nor meaningfully addressed let alone disproved the classic studies that led to the death of Social Darwinism in linguistics in the first place.

Insisting that anyone who doesn't agree with you must be too "woke" or "PC" and therefore wrong is not the cunning argument you think it is.

Politesse said:

Swammerdami said:

SUMMARY: Many top linguists agree that Thread question ("Are some languages more complex than others?") should be answered Yes. Those who say otherwise are largely parroting (out of "political correctness"?) a dogma from decades ago.

Click to expand...

Which "top linguists" are those?

Click to expand...

You didn't notice that I mentioned Edward Sapir?

Politesse said:

I'd never even heard of Walter Bisang before reading this thread. Had you heard of him before you before you started looking for evidence to justify your views? In what other avenue? And neither you nor Bisang have proposed any coherent, objective definition for "complexity", nor meaningfully addressed let alone disproved the classic studies that led to the death of Social Darwinism in linguistics in the first place.

Click to expand...

I sought out Bisang's paper(s) because he appears to DISAGREE with me; i.e. I think he SUPPORTS the meme. Get it?
(a) OBVIOUSLY one should attempt to understand OPPOSING arguments; and
(b) If you really thought I agreed with Bisang you obviously skimmed my posts VERY briefly, hunting only for ways to insult me.

Politesse said:

Insisting that anyone who doesn't agree with you must be too "woke" or "PC" and therefore wrong is not the cunning argument you think it is.

Click to expand...

Blah, blah, Insult insult.

Frankly I have little interest in your opinions. I recall a thread where you confused "doctrine" with triviality until I pointed out that Daniel Wallace agreed with me, not you! That spun you around 180°.

Swammerdami said:

If what Professor Wallace writes is wrong, please take it up with him!

Click to expand...

Edward Sapir has been dead for nearly 85 years, and his academic career largely predated the debate in question, at least as it is now understood. Noam Chomsky was in grade school when Sapir retired from Yale. As for what Daniel B. Wallace has to do with this, being a fine Bible translator and theologian but not a linguist, I have no idea.

I am not "hunting for ways to attack you". I am not attacking you at all. It's your claim that is wrong, not you. If there is any dispute over the complexity of language, it's a debate within the field, certainly not a new consensus.

If you want to make a concrete argument in favor of linguistic inequality on the basis of complexity, you need to first define what you mean by "complexity" and make a strong case for the neutrality and utility of that definition. No one disagrees that languages differ from one another in syntax, that is non-controversial. But ranking languages from simple to complex was a habit we left in the academic past for good reason, and upturning that consensus isn't going to happen unless there were a very clear and unambiguous argument for why doing so is valuable.

Swammerdami said:

Thanks for the lead! It will take me some time to read the whole paper, but the title ("The rise and fall of a consensus") shows the theme. Long ago, it was assumed that "savage" or "barbarian" cultures had primitive languages. Complex societies had complex languages. ...

Anyway, savage people with complex languages also turned up and by the early 20th century, the meme of equal complexity may have arisen as a counterfactual anti-racism, which today might be called "wokeism"! ...

Click to expand...

While I don't doubt that the meme arose historically as a long-overdue reaction against the savages-had-primitive-languages meme, it seems to me at this point there's intellectually more to it than that. The negation of "Language complexities are equal." is "There exist languages X and Y such that Complexity(X) > Complexity(Y)."; the point of saying the former is to dispute the latter. And for any language L, Complexity(L) = Summation_{[i in language aspects]} complexity_i(L) * weight_i . So I'd take "Language complexities are equal." to be a TLDR executive summary of the view that:

Yes, I understand that complexity weights are ill-defined and subjective. We wouldn't want to posit a counterexample unless "complexity_F(X) ≤ complexity_F(Y)" for almost every factor F. But Everett seems to have achieved that with his study of Pirahã.

My question is VERY sincere. WHAT specifically about the Thai language provides its alleged complexity? Is reduplication -- the use of "pretty pretty" to mean "very pretty" or "child child" to mean "children" -- really a good example of Thai's alleged complexity?

This is why I'd like to read Bisang 2009, “On the Evolution of Complexity: Sometimes less is more in East and mainland Southeast Asia”.

Swammerdami said:

Yes, I understand that complexity weights are ill-defined and subjective. We wouldn't want to posit a counterexample unless "complexity_F(X) ≤ complexity_F(Y)" for almost every factor F. But Everett seems to have achieved that with his study of Pirahã.

Click to expand...

According to Wikipedia,

My question is VERY sincere. WHAT specifically about the Thai language provides its alleged complexity?

Click to expand...

Haven't a clue. Find a linguistics professor who speaks Thai.

Bomb#20 said:

Swammerdami said:

Yes, I understand that complexity weights are ill-defined and subjective. We wouldn't want to posit a counterexample unless "complexity_F(X) ≤ complexity_F(Y)" for almost every factor F. But Everett seems to have achieved that with his study of Pirahã.

Click to expand...

According to Wikipedia,

Their language is a unique living language (it is related to Mura, which is no longer spoken). John Colapinto explains, "Unrelated to any other extant tongue, and based on just eight consonants and three vowels, Pirahã has one of the simplest sound systems known. Yet it possesses such a complex array of tones, stresses, and syllable lengths that its speakers can dispense with their vowels and consonants altogether and sing, hum, or whistle conversations."[5] Peter Gordon writes that the language has a very complex verb structure: "To the verb stem are appended up to 15 potential slots for morphological markers that encode aspectual notions such as whether events were witnessed, whether the speaker is certain of its occurrence, whether it is desired, whether it was proximal or distal, and so on. None of the markers encode features such as person, number, tense or gender."​

Click to expand...

Thanks for the info! I'd read about evidential markers but didn't associate them with this language. Everett spent years with these people -- perhaps there's much controversy about "what is complexity?" as it applies to that language.

But Thai strikes me as simple phonetically, simple grammatically, with very few verb markers, etc. I am really puzzled about where its "complexity" lies.

Bomb#20 said:

My question is VERY sincere. WHAT specifically about the Thai language provides its alleged complexity?

Click to expand...

Haven't a clue. Find a linguistics professor who speaks Thai.

Click to expand...

Easier said than done! I'm reluctant to e-mail strangers, when I have nothing to offer. There are message-boards where linguists ask each other questions, but I'd guess laymen would face barriers at such a board. My hope is that some Infidel already has access and could pose the question for me. Or, since it seems like a routine question, perhaps someone adept at Google Search could find a message on the topic.

My impression is that Thai is similar to Chinese typologically, so an answer for Chinese MIGHT serve.

Complexity may be measured by how many individual forms one has to learn. For N numbers and C noun cases, a modular sort of noun inflection would have only N + C forms, while a non-modular sort would have N*C forms. In general, N*C > N + C, so modularity is lower in complexity.

A further problem is that a language can be simple in one thing and complex in another.

Chinese, Vietnamese, and Thai have highly modular and thus relatively simple grammar, but they have a kind of complexity that may seem odd to many of us:  Classifier (linguistics) using classifiers with counts. How do measure words work in Mandarin Chinese?
liǎngwèi lǎoshī - two person teacher
yīpǐ mǎ - one animal horse
sìbǎ yǐzi - four holdable chair
yībǎ huā - one holdable flower (bunch of flowers)
yīshuāng xié - one pair shoe
yītiáo yú - one long-and-narrow fish
...

Korean, Japanese, Khmer, and Burmese also have classifiers. The closest thing in English is "head of cattle", but that's very unusual. English distinguishes between countable and uncountable nouns, like countable "cup" vs. uncountable "water". Thus, "three cups" but not "three waters". One must specify something countable with an uncountable noun, like "three cups of water". But one can use "bottle" or "gallon" or "drop" or several other such words, words that can be used on their own.

Another sort of complexity is selection of pronouns by level of formality and social status:  T–V distinction and  T–V distinction in the world's languages

Many European languages have long used plural you (French vous /vu/) as a respectful singular you (French tu /tü/), and in English, the original singular you, thou, ended up dropping out.

Some languages have even more variants of pronouns, like Japanese:  Japanese pronouns with a big list of them. This extends to verbs, which have a suffix that implies respectfulness: -masu


Slavic languages distinguish verb aspects, imperfective (incomplete, continuing) and perfective (complete, momentary). These are formed in a variety of ways, usually from affixes (prefixes and suffixes), but sometimes from separate roots, like Russian impf govorit' pf skazat' "to speak".

Proto-Indo-European is also reconstructed as having verb aspects, with relics like Germanic strong-verb conjugations, with their vowel shifts. They are formed completely differently, usually from affixes and vowel shifts, but sometimes also from separate roots, like impf *es- and pf *bheuH- "to be".