5

I am native Swedish speaker and I have a problem that the language seems to have no grammar in some cases. For instance there is both "en lag" and "ett lag" meaning completely different things but the word "lag" is the same sound and same spelling, but the difference is in the meaning of the word. So you can't know from just the word "lag" whether it is "ett lag" or "en lag".

It seems completely arbitrary whether you should say "en lag" or "ett lag" but the meanings are different.

It seems like there is no logic there or at least something I never saw in other languages i.e. the meaning of the word determines the case. It is the same difference in English in "an" or "a" and there it is the sound of the word which determines the case.

In all other cases besides Swedish, you can know from the word itself what the gender or case should be e.g. English has "a" or "an", German has "der", "die", "das" with very distinct clear rules so that you know, given a word, how to write the definite article.

6
  • 2
    I don't think the rules for German "der", "die", "das" are very clear at all. They are actually complicated enough that German textbooks or courses that I've seen recommend memorizing every German noun with an article. My understanding is that native speakers disagree about the gender of some German words. May 7, 2018 at 1:58
  • 1
    @sumelic But compared to Swedish, German is very regular. May 7, 2018 at 1:59
  • 4
    I don't know enough about each language to say. I just wanted to point out that Swedish is far from the only European language where the grammatical gender of a noun can be relatively hard to predict just from the form of the noun. May 7, 2018 at 2:00
  • 2
    German (and most languages with gender) has similar situations where the essentially two lemmata with different genders have the same nominative singular surface form. eg en.wiktionary.org/wiki/Teil#German, en.wiktionary.org/wiki/mar#Spanish. May 7, 2018 at 8:41
  • 4
    Your last paragraph is completely wrong. In English, the choice of a or an is purely phonetic, and has no grammatical implications whatever. In languages with grammatical gender, in some (eg Latin, Italian, Spanish, Russian, Swahili) you can usually guess the gender from the form of the word (but not always). In others (eg German, Swedish, Welsh, Hebrew) the correlation with form is much weaker and you generally can't guess the gender from the form.
    – Colin Fine
    May 11, 2018 at 12:59

3 Answers 3

11

Many languages have such an ambiguity built in. It's very common that you can't tell the morphosyntactic properties of a word or even its phonological interactions just from its form, even given certain generalizations.

  • In German if you encounter the word Schlüssel on its own you don't know whether you have one key (der) or multiple (die) keys.

  • In French if you encounter the word grammaire you don't know whether it's the set of rules governing a language (la) or a book describing those rules (le).

  • In English if you encounter the name Taylor without context you don't know know whether to make a subsequent pronoun agree as him or her.

  • In Hebrew if you encounter the word אוכל /oχel/ you don't know whether it's the participle "eating" or the noun "food".

  • In Spanish if you encounter the word flor for the first time you can't guess the gender. If you encounter the word mano for the first time and apply the usual rules you will guess wrong.

  • In German if you apply a generalization about words ending in el to Bibel you will be wrong.

  • In English if you try to decide between "a" and "an" using the same rule for hour and heel you will get one of them wrong.

  • In French if you encounter the word haricot for the first time you won't know whether the definite article is "le haricot" or "l'haricot".

Each of these is one example of large sets of ambiguous cases!

In short, the entry in the lexicon presumably carries some of the information needed to settle these cases — not just the meaning.

9
  • "In English if you try to decide between "a" and "an" using the same rule for universal and unusual you will get one of them wrong" - this is the only one I disagree with. You can't tell from the spelling, sure, but the rule is based on the pronunciation ("universal" starts with a consonant sound /j/, while "unusual" has a vowel).
    – Draconis
    May 12, 2018 at 2:38
  • @Draconis Hmm, true. See if this revision is any better... May 12, 2018 at 2:57
  • In my (British) English we say "a history" and "a heel". What am I not understanding?
    – Lorraine
    Apr 8, 2019 at 11:17
  • @Wilson I should probably not have generalized across a country. Can you help me identify a dialect where they say "an history"? Apr 8, 2019 at 11:41
  • 1
    How about "an honor" or "an hour" for something most dialects probably agree on? Or are you looking for words where the /h/ is actually pronounced, but they still take "an"? (But in that case, I would venture to guess that in dialects that use "an" in front of those, the /h/ is not actually pronounced...)
    – LjL
    Apr 18, 2019 at 14:46
5

I have a problem that the language seems to have no grammar in some cases. For instance there is both "en lag" and "ett lag" meaning completely different things but the word "lag" is the same sound and same spelling, but the difference is in the meaning of the word.

The grammar is there, even if it's confused by something else.

As you probably know, in Swedish every noun is either "ett" or "en". It's unpredictable sometimes, but this feature is inherent in the noun.

Why is "lag" sometimes "en" and sometimes "ett"? It's because actually there are two/three separate words here.

  • One is laget, "the team".
  • One is lagen, "the law".
  • Another is lagen, the juice/brine.

Each has its gender; they are never mixed up. Swedish people never say "en lag" to mean "a team".

It seems like there is no logic there or at least something I never saw in other languages i.e. the meaning of the word determines the case.

It's mostly arbitrary, like I said. But there are certain generalisations. For example,

  • living beings are mostly "en". "en häst", "en fluga", "en pojka", "en främling", "en svamp".

  • Masses, like water and slime are usually "det": "vattnet", "slemmet", "trädet". This is what explains "laget", "the team". I guess you'd expect "juice/brine" to fall under this category but for whatever reason it doesn't.

  • Nouns for humans/humanoids that are both genders are usually "ett": "ett barn", "ett troll", alternative explanation for "ett lag".

  • Nouns which are historically derived from verbs are almost always "en". "lag", "the law", comes from "ligga". That's why it's "en lag".

You see, there is method to the madness. Not always a perfectly reliable method, but it's undeniably there.

4
  • Just realised, the word for "juice/brine" probably also comes from ligga (in the sense of "steeping something in water"), which would be why it's "en".
    – Lorraine
    May 8, 2018 at 14:29
  • 1
    Could you explain whether skit (shit, dirt, damn) is a single word or several, when they say ”hela skiten” but ”inte ett skit” ?
    – J-mster
    May 10, 2018 at 11:17
  • @J-mster that's a tricky one; skit is derived from the verb skita so it's "en", on the other hand it's a mass as in my second bulletpoint, so it's "ett". Whether it's one word or several though is up for debate. I don't think there's a clear answer.
    – Lorraine
    May 10, 2018 at 12:40
  • 2
    En other helpful rule is that if the words plural is the same as singular, it is an "ett" word e.g. ett träd/flera träd. Of course that means you need to know the words plural.
    – Midas
    May 11, 2018 at 6:02
3

Two, three homonyms lag (different meanings, identically written/sounding words), here with two genders for the articles: en/et.

A German example with the same phenomenon:

  • die Steuer = the tax,
  • das Steuer = the steering wheel.

It is a helpful coincidence that the articles differ.

Languages with several lingual origins/cultures, like English (romance+germanic), or a larger reuse of a word for different contexts (both Steuer's have their origin in steer), may have more homonyms, whereas an artificial language like Esperanto tries to avoid homonyms.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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