Chung wrote:Josquin wrote:You should try your luck with Finnish or Hungarian. Despite having numerous cases, the case systems themselves are supposed to be comparably straightforward and logical. As the Finno-Ugric languages are agglutinating rather than synthetic, the case markers consist of suffixes rather than endings and have probably evolved from postpositions (as Chung has pointed out, an example of case systems emerging from analytical structures).
The point is, each case has only one ending that may have different forms because of sound laws, but there are no declension classes with different endings as in the Indo-European languages, so there is no overlap between endings either. I'm not saying those languages are easier than Russian or Icelandic though.
Maybe Chung can offer some more information on this.
I could, but to be honest, it'd be an even longer-ass post, and with the way things have gone so far, I wouldn't be surprised if s_allard would do some articulate hand-waving and excuse himself thus leaving his FIGS-based POV intact.
Upon a bit more thought, I'll add a few qualifying points to the above about Finnish declension as it's not quite as straightforward as one might think when comparing it to declension in Latin or other Indo-European languages of central and eastern Europe. I do feel (this is very subjective, mind you) that it is less burdensome to learn than in those languages because of how little syncretism raises its head (see my comments below about Finnish "accusative") not to mention that Finnish doesn't draw on (or recycle) sets of case endings to signal gender or "humanness"/animacy (see some Slavonic languages for the latter).
It is true that the case endings themselves are unambiguous as seen
here (ignoring the questionable inclusion of "accusative" as discussed
here). Whenever you see
-lle, for example, you know that you're dealing only with allative (~ "for/to(ward) sb/sg") regardless of whether the noun is in singular or plural. The plural marker in case-marking is usually
-i- or
-j- (e.g.
kaverille ~ "for a buddy" - allative singular,
kavereille ~ "for buddies" - allative plural) but in the nominative plural is
-t (e.g.
kaverit ~ "(the) buddies") and in genitive plural can sometimes be
-(i)d- or
-(i)tt- in addition to "regular"
-i- ~ -j-. The
-(i)d- and
-(i)tt- in
-iden, and
-itten are related to the
-t- ending of nominative plural and thanks to Finnish standardization are deemed grammatical for some nouns as exemplified by
kavereiden and
kavereitten "of buddies" coexisting with "regular"
kaverien. Note how the
-n ending is the same one used for genitive singular. When put another way, the plural genitive ending (in line with case markers of any Finnish nominal in plural) is thus a compound of a number/plural marker and the case marker. That's not surprising, is it?
It is however possible to reclassify nouns (and verbs for that matter) into inflectional classes a bit like what you see in Latin. It's not that each class takes different case endings* but rather what happens to the stem during inflection (e.g. consonant gradation, change to the stem's first vowel) that justifies the subdivision into inflectional classes. Sometimes these subdivisions can be guessed when you look at the endings in their nominative singular forms (cf. the divisions in most Slavonic languages whereby a noun's declensional pattern can be gleaned by referring to the form (not just the ending) in nominative singular).
*you could make a certain argument that a word like
kaveri "buddy" belongs in its own class since it's declined in genitive plural as
kaverien,
kavereiden or
kavereitten, while
ystävä "friend" in genitive plural is only
ystävien with *
ystäveiden, *
ystävitten or similar being ungrammatical. If you count the accusative as a viable case in Finnish outside personal pronouns, then you see recycling of endings since the accusative singular can resemble either the nominative singular or the genitive singular (depending on whether it's an indicative or imperative) while the accusative plural resembles the nominative plural once it's satisfied criteria for telicity.
The Finnish government's research institute for languages, Kotus,
classifies nouns 51 ways based mainly on stem changes that occur in declension. It also classifies verbs 27 ways again based on stem changes that occur in conjugation. Finnish entries in Wiktionary make reference to Kotus' classification when identifying the inflectional pattern.
In practice, beginners end up learning in total about half of these classes since building a sufficiently useful stock of vocabulary such that they can start to express themselves grammatically and usefully doesn't require knowing all of the patterns. I estimate that by the time you reach B1, you've been exposed to about 25 models for declension and about a dozen for conjugation. Moreover, inflection draws on a relatively small set of rules for accommodating consonant gradation and changing vowels at the end of stems when declension creates duplicated vowels or strings of three or more vowels. There's also a small consideration for vowel harmony when choosing the correct case suffix, but that's relevant only when the suffix contains vowels other than
-e- or
-i-.
If someone were to try targeting his/her studies a bit, and assuming that the stock of Finnish entries on Wiktionary is representative of the distribution of inflectional types in Finnish, then he/she should definitely not be overwhelmed by Kotus' inflectional subdivisions. Most nouns or adjectives that someone will end up learning will fall into one of 15 classes. It follows that enough exposure will present mainly words drawn from the same subset of classes over and over again, and so a learner will start to develop native-like intuition of grammatical inflection in a large portion of the vocabulary more quickly than he/she might expect.
As of today, there are 31056 entries for nominals in Finnish Wikisanakirja (distinct from using English Wiktionary to look up Finnish words), and sorting them by membership by declensional class from most entries to fewest, I get:
No. 5 pattern
risti "cross": 5414 (17% of entries)
No. 38 pattern
nainen "woman": 4870 (16% of entries)
No. 1 pattern
valo "light": 3363 (11% of entries)
No. 40 pattern
kalleus "expensiveness": 2930 (9% of entries)
No. 10 pattern
koira "dog": 2270 (7% of entries)
No. 39 pattern
vastaus "answer": 2093 (7% of entries)
No. 9 pattern
kala "fish": 1626 (5% of entries)
No. 6 pattern
paperi "paper": 1563 (5% of entries)
No. 12 pattern
kulkija "traveller": 949 (3% of entries)
No. 34 pattern
onneton "unhappy" 857 (3% of entries)
No. 48 pattern
hame "skirt" 702 (2% of entries)
No. 18 pattern
maa "country" 695 (2% of entries)
No. 3 pattern
valtio "state" 493 (2% of entries)
No. 2 pattern
palvelu "service" 433 (1% of entries)
No. 41 pattern
vieras "stranger" 379 (1% of entries)
The remaining 36 classes/patterns comprise 8% or 2419 of the 31056 entries. If I've got the declension down cold for the preceding 15 nouns, then I know (or can quickly figure out) the declension of 92% of the Finnish nominals of Wikisanakirja.
This kind of distribution reflects to a good degree my experience with learning Finnish, and so a decently designed course will make it pointless to worry about this seemingly impenetrable or mind-boggling setup for declension. You'll basically learn enough nominals such that you'll have a good chance to decline unfamiliar ones as you encounter them later since their patterns will often be the same as that used in words that you've already used/learned/seen. As for myself, I guess that I have in my mind the patterns of about a couple of dozen nouns and adjectives, and can usually decline these words and any unfamiliar ones correctly on the first try about two-thirds of the time. However, I do get fouled up with partitive plural more than I'd like, and if I have the partitive plural wrong, then I'll probably foul up the remaining declensions of the plural, since they're based on a stem used in that case.
In the first half of Finnish for Foreigners I, these are the noun classes that I encountered (and this is with a lot of basic vocabulary - bolded numbers in red mean that they're not in the top 15 of Wikisanakirja by number of entries):
No. 1
valo and similar (2- and 3-syllable nouns ending in -o/-ö or -u/-y that use consonant gradation where applicable)
No. 2
palvelu and similar (Nouns with 3 or more syllables that end in -o/-ö or -u/-y that don't show consonant gradation)
No. 3
valtio and similar (Nouns ending in -ao, -eo, -io/-iö, -oe)
No. 5
risti and similar (2-syllable nouns ending in -i which are often seen in loanwords such as
bussi "bus",
tuoli "table" etc. - Cf. pattern of
paperi at no. 6)
No. 6
paperi and similar (nouns with at least 3 syllables ending in -i which are often seen in loanwords such as
hotelli "hotel" and
turisti "tourist" - Cf. pattern of
risti at no. 5)
No. 7 ovi and similar (2-syllable nouns ending in -i but whose stem ends in -e. This doesn't make the top 15, but these nouns often turn up in words for natural phenomena or "ancient" items such as
ovi "door",
kivi "rock",
pilvi "cloud" etc.)
No. 9
kala and similar (2-syllable nouns ending in -a which have -a-/-ä-, -e- or -i- in the first syllable)
No. 10
koira and similar (2-syllable nouns ending in -a/-ä that have -o-/-ö- or -u-/-y- in the first syllable)
No. 12
kulkija and similar (3-syllable nouns or longer with ending -a/-ä)
No. 15 vaikea and similar (Nouns ending in -ea/-eä or -oa - This isn't a large class of nouns (and doesn't make the top 15) but includes
vaikea "difficult", and
nopea "fast")
No. 16 vanhempi and similar (all forms which end in -mpi, which marks a comparative adjective. This class doesn't make the top 15 on Wikisanakirja but it's useful to know considering the utility of comparative adjectives in general)
No. 18
maa and similar (noun ending in duplicated vowels)
No. 19 suo and similar (1-syllable noun ending in -ie, -uo/-yö. This isn't a large class of nouns (and doesn't make the top 15) but includes the basic words
tie "road",
työ "work" and
yö "night")
No. 27 käsi and similar (2-syllable nouns ending in -si but whose (weak) stem ends in -de. This isn't a large class of nouns (and doesn't make the top 15) but includes basic concepts or terms such as
vesi "water",
uusi "new",
käsi "hand",
vuosi "year" etc.)
No. 33 kytkin and similar (nouns ending in -in, whose stem instead ends in -me. This class doesn't make the top 15 on Wikisanakirja but turns up often in nouns for machinery such as
puhelin "telephone")
No. 38
nainen and similar (nouns ending in -nen, whose stem ends in -se)
No. 39
vastaus and similar (nominals containing at least 3 syllables ending in -us/-ys but whose stem ends in -kse)
No. 41
vieras and similar (nominals containing at least 2 syllables and end in -as)
No. 43 ohut and similar (nominals ending in -ut/-yt. This class doesn't make the top 15 on Wikisanakirja but turns up in
lyhyt "short" and
olut "beer")
No. 48
hame and similar (nouns ending in -e, and contain at least 2 syllables.)
In other words, the first half of Finnish for Foreigners I (i.e. something like clearing A1) introduces vocabulary from 13 of the top 15 classes of nominals sorted by frequency on Wikisanakirja. Moreover, it also introduces nominals from 7 other classes that each comprises less than 1% of the entries on Wikisanakirja. While these 7 other classes comprise a very small share of the lexicon, they contain items that usually occur at a greater frequency than their puny class-size might suggest.
After doing a similar comparison with verbs (3755 on Wikisanakirja) and getting a sense for their distribution sorted by stem, you'd come to a similar conclusion by realizing that mastering about half of the 27 verb classes covers the patterns of the majority of the verbs you'll need to learn. Verbs such as
juostaa "run" or
saada "to get" belong to such small classes that one can pretty much learn them rather like exceptions. Because of their fairly high frequency, repeated exposure to them would ingrain their patterns anyway.
In brief:
27% of verbs conjugate like
muistaa "to remember" (infinitive ends in -aa/-ää excluding -taa/-tää, and has at least 2 syllables)
18% conjugate like
sanoa "to say" (infinitive ends in -oa/-öä or -ua/-yä)
15% conjugate like
salata "to steal" (infinitive ends in -ata/-ätä)
11% conjugate like
tulla "to come" (infinitive ends in -lla/-llä, -nna/-nnä or -rra/-rrä)
6% conjugate like
voida "to be able to" (infinitive ends in -da/-dä and is preceded by a diphthong with -i. See also
tupakoida below)
5% conjugate like
sallia "to permit" (infinitive ends in -ia/-iä with stem open to consonant gradation)
4% conjugate like
huutaa "to shout" (infinitive ends in -taa/-tää - different from the large class of verbs that conjugates like
muistaa)
3% conjugate like
rohkaista "to encourage" (infinitive ends in -sta/-stä)
3% conjugate like
katketa "to snap" (infinitive ends in -eta/-etä, -ota/-ötä or -uta/-ytä but stem adds -n- in potential mood only - this one's model can be confused with that used by verbs like
vanheta. See below)
2% conjugate like
tupakoida "to smoke" (infinitive ends in -oida/-öidä, but can use stem ending with -tse in addition to a stem identical to the model of
voida. See above)
1% conjugates like
vanheta "to grow old" (infinitive ends in -eta/-etä but stem adds -n- in present and past tenses, and conditional and potential moods - this one's model can be confused with that used by verbs like
katketa. See above)
1% conjugates like
valita "to choose" (infinitive ends in -ita/-itä)
The remaining classes of Finnish verbs on Wikisanakirja each make up less than 1% of the total.