The ordering of adjectives


In a recent comment, CaitieCat said something that really surprised me. She pointed out that in English, whenever a noun is prefaced by multiple adjectives, those adjectives follow a specific order. We all instinctively use that order though, as far as I can tell, no one is ever actually taught to do so but pick it up as they learn language.

For example, we would say “the big red ball”, and not “the red big ball”. Clearly there is some classifying of categories going on as well as some rule of precedence. She later gave a link to an article that explained the ordering rule.

  1. Determiners — articles, adverbs, and other limiters.
  2. Observation — postdeterminers and limiter adjectives (e.g., a real hero, a perfect idiot) and adjectives subject to subjective measure (e.g., beautiful, interesting), or objects with a value (e.g., best, cheapest, costly)
  3. Size and Shape — adjectives subject to objective measure (e.g., wealthy, large, round), and physical properties such as speed.
  4. Age — adjectives denoting age (e.g., young, old, new, ancient, six-year-old).
  5. Color — adjectives denoting color (e.g., red, black, pale).
  6. Origin — denominal adjectives denoting source of noun (e.g., French, American, Canadian).
  7. Material — denominal adjectives denoting what something is made of (e.g., woolen, metallic, wooden).
  8. Qualifier — final limiter, often regarded as part of the noun (e.g., rocking chair, hunting cabin, passenger car, book cover).

So, in English, adjectives pertaining to size precede adjectives pertaining to age (“little old”, not “old little”), which in turn generally precede adjectives pertaining to color (“old white”, not “white old”). So, we would say “One (quantity) nice (opinion) little (size) round (shape) old (age) white (color) brick (material) house.”

Are those rules only for English? Are they restricted to the Indo-Aryan group? Or are they universal?

This is where speakers of other languages come in. How would you say “the big red ball” and “the big red shiny ball”? Is there a fixed order too? And is it the same order as in English?

I had never thought about this before.

Comments

  1. leni says

    Oh wow, interesting. I don’t recall ever learning that. I do recall learning a lot of other basic things, but not that.

  2. arno says

    In German (my mother tongue) the same order feels “best”, but for some reason, deviations in German seem less odd to me than reshuffling the words in English.

  3. Kimpatsu says

    Just to nit-pick, Japanese doesn’t have any adjectives; they are descriptive verbs.

    In French, some adjectives precede the noun and some follow the noun (e.g., le gros ballon rouge, not le gros rouge ballon), and the order of the adjectives maybe reversed for harmony.

  4. maudell says

    Interesting. As has been pointed out, French (my first language) has similar intuitive rules, but adjectives may be both before and after the noun. I’m trying to teach this to my English partner, and it’s not easy.

    Another interesting point: I learnt English as an adult, and picked this ordering up automatically, without realizing it existed. I know I still make ESL mistakes, but overall, it is intuitive. I wonder if it is due to historical proximity of French and English, or if it is a universal element of understanding language…

  5. Brian E says

    I’m not a native speaker of Spanish, but self-taught and lived in Spain a long time ago. You’d say la gran pelota roja’ which is ‘the big ball red’. Adjectives like big precede normally, but most adjectives follow, so pelota roja, ball red. This is the same in Latin as far as I can tell, but you do have some stylistic leeway that could only occur in English if in poetry or trying to sound archaic. For example, un gran hombre = a big or great man, but un hombre grande which is similar in basic meaning, but different emphasis.

  6. says

    The languages I’ve learnt (all Indo-European) all have rules, insofar as it’s not random how adjectives are ordered.

    I’d say mostly there’s a default order, which can be changed in some cases for emphasis or subtle changes of meaning. In general my impression is that the adjectives closer to (preceding) the verb seem to be the ones we think are more important, or more closely tied to the nature of the noun. For those following the verb it may be the other way round.

    In German “Eine schöne junge Frau” (default) is a pretty young woman, “eine junge, schöne Frau” is young and beautiful. I’d say it’s similar in English where “a young, intelligent woman” is subtly stronger than “an intelligent young woman”. I’d say that the default order becomes meaningless through overuse. [My instinct is to put no comma in the default order, but add one in the other -- probably it’s right in one of the two languages 🙂 ]

    Sometimes the adjective changes its meaning completely when the word order is changed, especially in the Romance languages where the adjective can come before or after the noun. In French “Un homme brave” is brave man, “un brave homme” is law-abiding or harmless; In Portuguese “uma mulher pobre” is without means, “uma pobre mulher” is pitiable.

  7. uri says

    it’s even more pronounced with adverbs. the syntactician guglielmo cinque wrote a book showing regular cross-linguistic ordering in adverbs. if i remember correctly, he divided adverbs into something like 20-30 semantic categories and showed that there is a consistent ordering across languages.

    as for adjective ordering, i would anticipate (on analogy with joseph greenberg’s universal no. 20) that all language in which adjectives come before the noun would work like english, whereas languages in which adjectives follow the noun, they would either have the english order or the mirror image of the english order.

    definitely look up greenberg’s typological universals if you find this sort of thing interesting.

  8. Jeffrey Johnson says

    This is very interesting. In Mongolian, which is not an Indo-European language, the same ordering is followed for “big red shiny ball” == tom ulan gelgar bombog (Mongolian has no articles). My wife is a native speaker and she said it just doesn’t work to change the order.

  9. Ethan Myerson says

    When we were younger, my friends and I noticed the near constancy of adjective ordering, and set about trying to figure out the rules. We never did come up with anything as systematic as the rules above. What we did come up with was a simple heuristic -- the farther from the noun, the less germane the descriptor. Or, put another way, adjectives closer to the noun are more a part of the identity of that noun (from the POV of the speaker, I suppose). So we’d have “little old lady”, because the lady is more defined as being old than by being little. We’d have “red rubber ball” because the material is very much germane to the ball’s identity; far more than its color.

  10. Emory K. says

    Some exceptions immediately occur to me, but maybe that’s because I’m a protoplasmic American white middle-aged skinny nerdy guy.

  11. left0ver1under says

    And as I’m sure you can explain better, French adjectives aren’t limited to being in front of the noun as in English, some can go after the noun. “Ancienne femme” and “femme ancienne” have different meanings. Learning masculine and feminine adjectives in French annoyed/annoys me no end.

    English also partakes in multiple locations of words, but with adverbs. Most adverbs go after verbs (with similar rules for order as with adjectives), but adverbs of time and frequency go before the verb. Weird.

  12. says

    Well, and the exceptions themselves, Emory K, tell a story. It’s one of those weird little things that people tend not to know unless they study linguistics or translation.

    Another fun one, particularly in English, French, and German (I expect it’s also true in the other Germanic languages, not as sure about the Romance), are the rules on when to use “a” and when to use “the”. It’s a lot more complex than we were taught in grade school.

    Or knowing how many ways we pronounce in English the sound represented by the letter L (hint, it’s more than one), or the sound represented by T (I can think of at least three regular variants, depending on where in the word it is and what sound comes next -- aspirated much of the time, unaspirated after “s-“, and palatalized, think Liverpool and some Irish accents for the last one), never mind how often a T or D or R is in fact represented by a glottal stop or a flap/tap.

    Granted that I’m a linguist and multi-language translator, I’m also a huge language nerd, so it’s a good profession for me. Glad to have given you something to think about, Prof. 🙂

  13. Jeffrey Johnson says

    Another fun one, particularly in English, French, and German (I expect it’s also true in the other Germanic languages, not as sure about the Romance), are the rules on when to use “a” and when to use “the”. It’s a lot more complex than we were taught in grade school.

    Since you brought up definite and indefinite articles, I thought I’d mention something interesting I learned about Mongolian, a language that has no articles. So how do you think they handle the conceptual difference between “I’m reading a book” and “I’m reading the book” if there are no articles?

    It turns out that this distinction is clear to Mongolian speakers without the use of articles by using case instead. Nouns are inflected by 7 different cases.

    The indefinite, “a book” is handled using the nominative case for book (“nom”) and the definite, “the book” is expressed in the accusative case for book (“nomig”). At first I just thought that concept must be absent from the language because articles are absent. When I figured out how they represent the definite/indefinite distinction I thought it was pretty darn cool. Maybe a language nerd will appreciate that…

    They do also have a way to say “this” and “that”, so sometimes definiteness is represented that way, especially if they are using the noun with, say, the instrumental case: “I killed the fly using a book” would be said something like “I fly(+accusative) book(+instrumental) killed” (bi yalaag nomoor xoroocon), and “I killed a fly using the book” would add “that” = “ter” (bi yalaa ter nomoor xoroocon). The instrumental ending “oor” depends on vowel harmony with the noun it modifies.

  14. drd says

    Word order can also depend on what you want to emphasize or for comparison. In your example, “the big red shiny ball” can be changed to “The shiny big red ball is preferable to play with than the dull big red ball.”

    In English, when using multiple adjectives, the order in which they are written is not as specific as when only two adjectives are used. In addition, further emphasis can be added through the use of commas. The shiny, big red ball fell.

    The sentence, “The shiny big ball…” does not read as well as, “The big shiny ball….”, but both forms are still correct but the choice would depend on the rest of the sentence.

  15. lpetrich says

    Another interpretation is that in Japanese, adjectives can be treated as verbs, meaning “to be “. Thus, “Let’s active” for “Let’s be active”.

    But does Japanese have the adjective order that our host mentioned?

  16. lpetrich says

    As to word order, another interesting feature is a tendency to choose between head-modifier and modifier-head. HM vs. MH:

    verb-object vs. object-verb
    auxverb-mainverb vs. mainverb-auxverb
    noun-adjective vs. adjective-noun
    noun-genitive vs. genitive-noun
    noun-relative vs. relative-noun
    adposition-noun vs. noun-adposition
    (adposition = preposition + postposition)

    English is mostly HM except for adjective-noun and possessive-noun. Other Germanic languages are much like English, but the Romance languages go even further in being HM. Latin, however, tended to default to MH, and the Indic Indo-European languages, like our host’s native one, are mostly MH. Japanese and Mongolian are also mostly MH.

  17. rq says

    Latvian is also a non-article language, and it’s difficult teaching people when to use them and when not to (and they looooove going the excess route because they think it makes them sound more authentic). It’s especially difficult in front of abbreviations, but I figured that one out -- if you can say an abbreviation like a word (like UNESCO), it takes no article and acts as a proper noun; if you can’t say it as a word (such as the UN), it needs the article, usually the definite.
    Latvian defines definite/non-definite nouns within the word ending, plus different endings for gender (for red -- sarkanā (fem.def.) and sarkanais (masc.def.) vs. sarkana (fem.indef.) and sarkans (masc.def.)). There is no neutral / non-gendered option for any noun, which makes gender-neutral language impossible, at best.

    Latvian adjective order, though, seems to be similar to English. I hadn’t thought about it before, but saying liela sarkana gumijas bumba feels far more natural than saying sarkana gumijas liela bumba. I think it does have something to do with which aspects of a ball are more important to its ‘being’, or maybe it’s just ingrained cultural habit rearing its head.

    My experience in French is that there are certain kinds of adjectives that tend to get placed after the noun more than others (such as colour: la balle rouge seems more correct that la rouge balle), and if there are two adjectives, they get split according to some criteria unknown to me ** (la belle balle rouge vs. la rouge balle belle). I actually learned my French adjectives better by defaulting to adjective-after-noun, but I’m fuzzily recalling that this was a taught rule at one point.

    ** In English, adjective phrases come after the noun being described -- saying “criteria unknown to me” seems more correct than “unknown to me criteria”, and far more correct than splitting the phrase “unknown criteria to me” (common mistakes in translations I see). Then again, one can merely say “unknown criteria”, but then it loses the specific aspect of “to me”. Anyway.

  18. lpetrich says

    Another aspect of word order: the order of the subject, verb, and object in a simple sentence or clause.

    The most common ones worldwide are SVO and SOV, with VSO and VOS less common. OSV and OVS are very rare.

    With auxiliary and main verbs, there are variations like:
    English: S AV MV O
    German: S AV O MV

  19. Mano Singham says

    I’ve been curious about the SVO order for some time but was under the impression that all languages were either SVO or SOV. Could you please tell me what languages have other orders?

  20. Mano Singham says

    Can you give some examples of these cases for us non-linguists? What is an example of a head-modifier, for example?

  21. Mano Singham says

    That is true, that when one is aiming for a comparison, one often uses the compared adjective first.

  22. uri says

    maybe most famously, the hebrew bible is largely written in VSO order. e.g. the first sentence:

    be-reishit bara elohim et ha-shamayim ve-et ha-aretz.
    in-beginning created god the-skies and-the-earth

    modern hebrew is SVO. i don’t know if ancient hebrew speakers would use VSO in everyday conversation, or if it is a feature of an elevated, literary register.

    modern arabic is said to be VSO as well, but i’m not sure if that’s just a feature of the formal arabic that is taught in schools, or if it’s true of some or all of the dialects of arabic that are actually spoken.

    malagasy (the malayo-polynesian language spoen in madagascar) is known for its unusual VOS order, with subjects last.

  23. Jeffrey Johnson says

    In Mongolian post-postional suffixes are used to represent noun cases, rather than prepositions as in English, so it is noun (head) + post-position (modifier).

    So instead of “with milk” as in tea, you have “suutai” where “suu” means milk, and “tai” is the communitive case postpositional suffix meaning “with”. English transliteration would be “tea with milk” = “milk-with tea”, which is not to be interpreted as adding tea to milk, it is tea consumed “milk-with”.

    Another is instrumental, in which the noun is used for something, indicated with a suffix -oor, -eer, -aar, or another requiring the cyrillic “theta” char as the vowel. The vowels in the endings depend on vowel harmony with the noun being modified. English represents this with prepositions before the noun (to go by bus, to eat with a fork, to type using a keyboard). “I’m going by bus” would be like “I bus-eer go”.

    Also there is the ablative (for coming from or going to a place) which would be the location + suffix “ruu” meaning “to the location” and location + suffix “aas” (or other vowels) meaning “from the location”. So you might say what transliterated into English sounds like “This fruit California-from is”, or “That Ulaanbataar-from train is” or “I’m tommorow Chicago-to going”.

  24. lpetrich says

    English has both orders:

    The red ball
    The, red = modifiers
    ball = head
    Modifier-head

    The ball’s color
    The ball’s = modifier
    color = head
    Modifier-head

    The color of the ball
    The color = head
    of the ball = modifier
    Head-modifier

    of the ball
    of = head
    the ball = modifier
    Head-modifier

    The ball hit the wall
    Hit = head
    the wall = modifier
    Head-modifier

    The ball that I have
    The ball = head
    that I have = modifier
    Head-modifier

  25. lpetrich says

    From a quick trip to Wikipedia, I found lots of them.

    Gaelic and Welsh have VSO order, but with an auxiliary verb, it becomes AV S MV O

    Biblical Hebrew and Classical Arabic also have VSO, as do many Austronesian languages, like Maori and Hawaiian. Some Austronesian languages, like Malagasy, have VOS order.

    OSV and OVS are very rare, and occur mostly in some Amazon-Basin languages.

  26. says

    I can’t speak to it directly, but given the number of times I’ve had to correct Arabic speakers’ sentences to move the verb to the post-subject position it usually takes in English, I’m going to say that the interference pattern I see suggests that colloquial Arabic does, indeed, start with the verb, and go VSO.

  27. says

    And some languages don’t really have a very set order. Russian, for instance, being highly inflected, has a very flexible word order, where the ordering of major elements usually represents emphasis or focus or order of importance, so “Big was the man I saw!” is a different meaning than “I saw a big man!” or “A man I saw was big!”. Inflections -- changes to the nouns and other words to show case (which asserts purpose/use of the word) make it so that, like Latin, the relation of words to one another is clear despite the very flexible order.

    This is one of the reasons Russian poetry can be so dense and multi-layered in meaning, that the purpose information of a word is encoded in the construction of that word, and order can be used for emphasis as well as easier rhyming.

    We maintain only tiny vestiges of English’s former case-marking system, most notable in the array of pronouns. He becomes him or his, I becomes me or my or mine, you becomes your. Otherwise, our case system is pretty much gone, so English word order has become much more important, to allow encoding of the information previously (and still in German, to some extent) encoded in the case markers.

  28. Mano Singham says

    Thanks for the examples. The trouble with never having learned English grammar in any formal way is that these subtleties are unknown to me.

  29. says

    Don’t feel bad about that -- honestly, this stuff is unknown to the vast majority of English speakers. I would guess that both lpetrich and Jeffrey Johnson have some linguistics courses in their backgrounds, or are highly motivated language nerds indeed.

    One of the things I find coolest about it is exactly that: that we can manage this amazingly complex task, of deciphering impacts of air molecules against our eardrums into concepts, then formulating answers to those concepts and vibrating a column of air in a specifically-encoded way to make someone else’s eardrums buzz, all while almost never consciously thinking about how the process is accomplished. The human brain is an astonishing organ.

  30. lpetrich says

    As to linguistics, my interest is a purely amateur one — I’m not a professional in that field.

    Another one about head-modifier order:

    The ball has hit the wall
    has = head
    hit = modifier
    Head-modifier

    From Jeffrey Johnson’s Mongolian examples:
    bi yalaag nomoor xoroocon
    I fly-acc book-instr killed

    SOV order

    yalaag (fly-acc) = modifier
    xoroocon (killed) = head
    Modifier-head

    yal- (fly) = modifier
    -aag (-acc) = head
    nom- (book) = modifier
    -oor (-instr) = head
    Modifier-head

    Some other terms:
    Head-modifier order = right-branching
    Modifier-head order = left-branching

  31. Jeffrey Johnson says

    @Ipetrich

    yal- (fly) = modifier
    -aag (-acc) = head
    nom- (book) = modifier
    -oor (-instr) = head
    Modifier-head

    I’m confused by this. First, just a minor point, “yalaa” is nominative for “fly”, and the accusative in this instance is formed by adding -g.

    But I was offering the Mongolian usages of noun + post-position as an example of HM ordering, where the noun as head comes first, and the post-position as case suffix comes second. In English these same ideas are expressed as a preposition modifying a noun (MH) as in “with milk”, “by bus”, or “to Chicago”.

    I’m confused because you labeled the noun as modifier and the case suffix as head. Was this just a mistake, or am I missing something?

  32. lpetrich says

    I’m interpreting a case suffix as a postposition, and there’s a strong correlation between verb-object vs. object-verb and a preferences for prepositions or postpositions. Generically: adpositions.

    WALS -- Chapter 95: Relationship between the Order of Object and Verb and the Order of Adposition and Noun Phrase -- http://wals.info/chapter/95
    Common:
    VO Adp-N
    OV N-Adp
    Rare:
    VO N-Adp
    OV Adp-N

    Adpositions as heads instead of modifiers may seem odd, but an adposition is syntactically an adverb or an adjective, and its noun phrase adds additional info to it. In some cases, that’s very evident, like the case of Chez -- French Preposition -- http://french.about.com/od/grammar/a/preposition_chez.htm -- “at the home of” and metaphorical extensions.

  33. lpetrich says

    http://www.sfs.uni-tuebingen.de/~gjaeger/lehre/ws0910/languagesOfTheWorld/wordOrderTypology.pdf

    contains a lot of detail:

    Head-final
    NP-P, AN, Gen-N, Adv-V, V-Aux, clause
    final question particles, clause final complementizer,
    standard-marker-adjective

    Head-initial:
    P-NP, N-relClause, V-Adv, Aux-V,
    clause initial question particle, clause initial
    complementizer, adjective-marker-standard

    What’s what:
    NP = noun phrase, P = preposition or postposition, generically adposition
    A = adjective, N = noun, Gen = genitive, Adv = adverb, V = verb, Aux = auxiliary verb

    Question particle: a word like “huh?”

    complementizer: a conjunction that makes an adverbial phrase, like “when” in ‘when I was reading”

    adjective-marker-standard is for comparisons:
    cats are bigger (adjective) than (marker) mice (standard)
    standard-marker-adjective would be:
    cats mice than bigger are

  34. lpetrich says

    Pre/post/adpositions as heads may indeed seem odd, but it makes syntactic sense. To see that, let’s expand some common prepositions:

    I am going to the store
    I am going in the direction of the store
    to = in the direction of

    The ball is in the yard
    The ball is at the interior of the yard
    in = at the interior of

    I cleaned the ball with a paper towel
    I cleaned the ball using a paper towel
    with = using (object of the verb: a paper towel)

  35. rq says

    Maybe someone on this comment thread (which is exceptionally educational and reminds me I haven’t read any linguistics texts in a good long while) can help me out: back in university, I read a book about linguistics and neurobiology by two scientists, and it had a whole section of exercises/examples at the end with empty categories. And I can’t remember the authors or the title and the book appears to be lost (several moves in the interim). Does this ring a bell for anyone, by any chance? Thanks in advance!

  36. Ghazala says

    In my mother tongue ,Pashto we are variety of adjectival orders and surprisingly there are no articles as well due to which we use determiner of number ( one) as article. Like a hospital ( yu hospital= one hospital).
    Example:
    naray tora danga Pakistanay jinay. ( Smart, damsel, tall Pakistani girl)
    danga nary tora Pakistanay junay. ( Tall, smart, damsel Pakistani girl)
    Tora nary danga Pakistanay jinay. ( damsel, smart, tall Pakistani Girl)
    Pakistanay, tora nary danga jinay. ( Pakistani damsel,smart, tall girl)
    The same sentence can be intrepreted in multiple ways.
    Though in Pashto we have direct modification and according to linguists , direct modification are order restricted but stil we have these violation of the oerders.

Leave a Reply

Your email address will not be published. Required fields are marked *