I'm sure most people here know, but for completeness, let's define what syntax is:

The arrangement of words and phrases to create well-formed sentences in a language. [NOAD]

N.B. It can also refer to the rules that govern it, or to the linguistics branch that studies such rules.

Anyway, this word-order can be more or less free, but it usually helps listeners to understand what are the roles of the single words or concepts in a sentence; we might say that you "expect" what is being told to you. Interlocutors expect the speaker to use pre-determined structures (pre-determined by the language rules). These rules simplify communication.

But I was wondering: Are there languages with a very strict word order or languages with a very free word order? Which are they?

  • I keep thinking of asking a question "what is syntax" so thanks for this. – hippietrail Sep 27 '11 at 13:14
  • You're welcome! I noticed there were almost no questions about syntax, and it's an important part of Linguistics. Not to mention it's something you always wonder when learning a new language. :D – Alenanno Sep 27 '11 at 13:16
  • Please go and write the tag wiki excerpt. It's one I've been scared to tackle due to the overlapping senses of "word order" and "linguistics branch studying rules". – hippietrail Sep 27 '11 at 13:19
  • I changed the title and some of the question wording to refer to word order rather than syntax. Saying a language has a "strict syntax" makes sense in the older, descriptivist use of the term, however in theoretical literature "syntax" can to very abstract rules that have little to do with surface word order. Using "word order" in the question makes it clearer what is being asked. – Aaron Sep 27 '11 at 17:12
  • @Aaron You mean the term syntax is getting "archaic"? – Alenanno Sep 27 '11 at 17:14

If you refer mostly to word order, then yes, some languages are syntactically stricter than others. It is often (as in: there is a correlation) related to the amount of morphological markers of function in the sentence. In other words, languages with a "rich" morphology (i.e. with expansive case systems, agreement of nouns with adjectives, rich verbal systems with person marking for all main participants, etc.) will tend to have a freer word order than languages that lack such markings. In the languages that lack such markings, word order will often take over the role of indicating grammatical roles in the sentence, and for this reason will tend to become fixed, as changing the word order in such languages would likely cause misunderstandings.

Of course, the correlation is not perfect, but as a rule of thumb it works very well.

Examples of languages with a strict syntax are simple to find: English is one of those. English is quite strictly SVO (alternative word orders are restricted to formal written prose and poetry), subjects are mandatory (even for verbs that logically don't take a subject, like "to rain"), adjectives must be put in front of the noun they complete (only heavy constituents can be put after a noun they complete), etc. About the only free constituents are adverbs and adverbial phrases, and even those are somewhat restricted in their movements in the sentence: fronting them, for instance, will sometimes result in a different meaning altogether. Contrast the following two sentences

(1) I only want to see her
(2) Only, I want to see her

I believe Mandarin Chinese is also quite strict in its syntax, but I don't know enough of the language to give examples.

German is also a language with some very strict syntactic rules, despite having declensions. For instance, it is strictly V2 in independent clauses (i.e. the finite verb must always be the second constituent of an independent clause), so when fronting another element of the sentence, the subject that normally comes in front of the verb must be put right after it (without anything in between). Word order is allowed some freedom, but the rules behind it are extremely strict. In the same way, the infinite verb part of a periphrastic conjugation must be placed at the end of the independent clause, and no other part of that clause can come after it.

Dutch, on the other hand, can be said to be somewhat less strict than German: it has the same ordering rules (V2, infinite verb at the end of the clause), but is somewhat less strict with the last one, allowing one to add some noun phrases after the infinite verb of a periphrastic conjugation. This is, AFAIK, allowed in some German dialects as well, but is still less common than in Dutch.

On the freer side of things, take for instance Modern Greek. Although the language is nominally SVO, any word order is allowed, depending on the kind of focus the speaker wants to give and the elements used. And like Spanish and Italian, it is pro-drop, and allows one not to show an overt subject, as it is already indicated in the verb. Word order in the noun phrase is slightly stricter, but even there it's freer than in English. For instance, as in English the article has to be put in front of the noun, but demonstratives can freely appear before or after the noun ("αυτό το αυτοκίνητο" and "το αυτοκίνητο αυτό" both mean "this car" and are completely interchangeable). Possessors normally appear after the noun, but they can be put in front of it, although this is considered somewhat formal. For example, both the following sentences can mean "the dog's toy", although the second one is rather marked and formal.

(3) το   παιχνίδι του      σκύλου
    the  toy      the-GEN  dog-GEN

(4) το   του      σκύλου  παιχνίδι
    the  the-GEN  dog-GEN toy

Another language with a relatively free syntax is Basque. Although it is typologically SOV, all word orders are allowed (even verb-initial clauses are allowed, although only if the verb has some particle in front of it, like the negative ez or the affirmative ba-), although which word order can be used in which situation is often pragmatically defined, especially by the rule of galdegaia, i.e. focus.

I believe there are even freer languages, even going as far as allowing the noun phrase to be separated and its constituents put in various places in the clause, but I don't have enough experience to comment much on them. I know Classical Latin allowed some quite outlandish things (having suffered through 5 years of Latin classes :) ), but it is unclear whether those things (found nearly exclusively in poetry) could be called grammatical at all. Poetic license was strong in Classical Latin.

For more info, you might want to take a look at WALS. Feature 81 "Order of Subject, Object and Verb" already lists 189 languages lacking a dominant word order. WALS lists quite a few other features related to "freedom" of syntax.

Note however that "free" syntax does not necessarily mean that all word orders can be equally used or are equivalent in meaning. Even in languages that allow re-orderings freely, word order often has a pragmatic meaning, indicating topic and/or focus.

| improve this answer | |
  • Thanks, your answer goes in depth. :) I took the liberty of bolding the languages name and changing the link to WALS. You are free to re-change and re-organize as you wish, of course. I just thought it would be helpful for others going through the answer. :) – Alenanno Sep 27 '11 at 15:15
  • You did well to make those changes, it makes my answer much easier to read. I'll try to remember how to make interlinears next time. The only thing I re-edited was the translation of the Greek article. – Tsela Sep 28 '11 at 9:09
  • 1
    Being a native speaker of German, I wasn't consciously aware of how strict the S/O word order rules are. (+1!) However, in some cases (not sure yet, seems to be 3rd person periphrastic conjugation forms) it is allowed (and sometimes even more natural) to have other phrases between the finite verb and the subject: „Gestern wurde mir im Hafen meine Brieftasche gestohlen.“ has object and adverbial between finite verb and subject, and is the word order I would naturally use for this sentence. – Anaphory Nov 22 '12 at 13:19
  • In addition to what Anaphory mentioned, sentences can have parts in the Nachfeld, i.e. behind the nonfinite verb. Example: Es hat angefangen zu schneien. – Jan Oct 10 '19 at 12:48

Tsela's answer is very thorough, but I wanted to point out another aspect of the question. There are some basically universal cognitive biases about linear order in language (basically, new information goes later in the sentence). Languages with a strict word order (in terms of argument positions) will have more resources for shuffling arguments to meet this property. An extreme example of this would be Malagasy, which has strict VOS word order, but has morphology that allows you to put almost anything in "subject" position.* "Free word order languages" on the other hand may lack some of this morphology, since they can satisfy the cognitive order requirements by reordering words in a sentence. However, in a given discourse context, only one word order may be grammatical. To use an example from Basque:

Q: Zein-ek ikusi du  Miren?
   Who-ERG see   AUX M.-DAT
   "Who saw Miren?"
A: Jon-ek ikusi du  Miren
   J.-ERG see   AUX M.
   "John saw Miren"
 * Jon-ek Miren ikusi du
   J-ERG  M.    see   AUX
 * Ikusi ba-du   Jon-ek Miren
   see   AFF-AUX J.-ERG M.

So describing a language as having "free word order" can be misleading, since speakers are often constrained by context to use only one order. When we discuss word order we should talk about which languages have more or less morphology for marking word order permutations.

*If this is unfamiliar to you, it can be understood by analogy to English passives, which work by putting the object in subject position (John kissed Mary -> Mary was kissed). Malagasy has "more passives" for putting things like an instrument or location into subject position.

| improve this answer | |
  • You're very right to point out that "free" word order doesn't necessarily mean that all word orders are allowed and equivalent in all contexts. I mentioned it somewhat in the last paragraph of my answer, but it's true that I could have been clearer. – Tsela Sep 28 '11 at 9:13

The question has essentially already been answered; I'll just add two more examples.

Another example for strict order: the Turkic languages. The order is SOV and the easiest way to describe is to say that the determiner always precedes the determined. This is an interesting case because unlike many (most?) languages with strict order, Turkic languages have an actually quite rich morphology.

Another example for free order: the Slavonic languages, possibly except Bulgarian, I'm not sure. I gave some examples from Polish in response to the question "What is word order used for in “free word order” languages?".

I just need to emphasize that unlike in Aaron's examples above, Polish speakers are not actually constrained by context. In most contexts a specific order is preferred (as in sounds more natural) but deviations are notorious in everyday communication so at least in this case I would say it is truly free.

| improve this answer | |

In French, the so-called "word order" is normally SVO :

"Il boit du lait" (He's drinking some milk)

Il (S) boit (V) du lait (O)

This word order roughly appears at the end of the Medieval Period when the mark of the subject (Latin nominative -> "cas sujet") and of the object (Latin accusative -> "cas objet") disappear progressively.

But there are exceptions. By example, French speakers use VSO when asking a question in a very formal way. Compare the three ways to say "Is he drinking some milk ?" :

"Il boit du lait ?" (very informal, SVO)

"Est-ce qu'il boit du lait" (the normal way to ask a question, est-ce que + SVO)

"Boit-il du lait ?" (very formal, VSO)

Same exception with imperative constructions (VO, like in English) :

"Donne-moi du lait !" (Give me some milk !)

By the way, the so called "t euphonique" placed between some verbs(*) and their subject in a VSO construction, like in :

Aime-t-il le lait ? (Does he like the milk ?) where a "t" is placed between "aime" and "il"

... is nothing but what remains from the Latin VSO corresponding words :

"Amat-ille" (Latin) > something like "aimet-il" in Old French > today written "aime-t-il"

notes :

(*) "verbes du premier groupe" : aime-t-il, pense-t-il + aller (va-t-il)

| improve this answer | |

English is a language with strich word order, Russian is a language with free word order.

In general this depends on how much a language is inflictive: the meaning is either provided by word order or by inflections.

| improve this answer | |

Add Persian as a free word-order language to your list. I was wondering Which languages are free word-order. I almost founded part of the answer in this post.

| improve this answer | |
  • 2
    Hello and welcome to Linguistics.SE. Perhaps, it would improve your answer if you added some backing references and/or examples proving your point about Persian. Without it, your answer rather looks like a comment. – bytebuster Jun 22 '14 at 5:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.