Declensions and Cases and Plurals oh my!

[Ciarán12]

Hello all,

I'm trying to tackle what I see as my biggest problem with Irish - nouns. Specifically how they inflect for case depending on declension. I have read quite a few different approaches to this. One book (which seemed to be pretty on the ball) called Irish Nouns: A Reference Guide by Andrew Carnie postulates a 10 declension system, with a separate system for determining plurals where nouns are grouped into "plural-types". These plural types work for determining the inflections of plural nouns for cases just as his declensions do for singular nouns. There are 6 plural types. Now, 6 plural types and 10 declensions sounds a lot more complicated than 5 declensions (including plurals), or a different 6 declension system I remember hearing about, but I would imagine Carnie's 10 Declension system has fewer exceptions, given that there are more rules covering them. Now, I could get my head around this. If this were it, I would just have to learn off the rules for the different declensions, and plural-types, and then learn a method for determining which pigeon-holes to place nouns in (hopefully there would be one). But this is not all. In his book Carnie gives examples for one of his declensions:
Declension Class A (he listed them A~J)
Rule: All Singular cases are the same
(I don't know why he says "singular" here, he's already explained that all "declensions" relate only to singular nouns)

Examples:
"The Thousand"(m) Nom./Acc: an míle Gen.: an mhíle Dat.: an míle Voc.: a mhíle
"The Festival" (f) Nom./Acc: an fhéile Gen.: na féile Dat.: an bhféile Voc.: a fhéile

So the rule is that they are all the same. To me, however, "míle" and "mhíle" are different! And that's not even taking into account the differences between the masculine and feminine forms.
He mentioned something about "except for articles and mutations" in the explanation he gave in the book for this declension. I looked up his section on "Articles and Mutations" and it said (as far as I could make out) that there was a further set of inflectional changes that were independent of declension or plural type that are applied to all nouns based on case, gender, number and what the initial letter of the word is. There was 1 form per case (4 cases), per number (2- sing. and pl.), per gender (2- Masc. and Fem.) per initial letter group (he lists 5), so that means there are 80 inflections to learn off for the cases, and then all the further changes for each declension/ plural-type? I assume the changes for the declensions act in tandem with the innate case changes?

I'm confused, I don't mind if the system is complex or if I have to learn off a lot of rules to know how to inflect nouns correctly, I just want to be clear on what the system actually is, so that I can go about the grueling task of actually learning it. I am sick of having no idea how or why a noun is in the form it is (séamhú's and urú's and word-final vowels getting slenderised and all of the above all at once...).

[linguoboy]

When it comes to "plural types", Irish is not as strict as related language like Welsh or German. To quote Dillon and Ó Cróinín from the 1961 edition of Teach Yourself Irish:

In general it may be said that plural forms in Irish are much freer than in English. As pl. of óráid "speech", óráidí, óráideanna, óráideacha, and óráidíocha would all be understood, and none of them would be shocking to a native speaker.

A bit of exposure will give you an idea what you can get away with here and what you can't. The most important thing is to learn to recognise the plural endings, so you can pick up on when they're being used.

Now, about the initial mutations in the "dative", there's a reason these are treated separately, namely that they vary according to dialect, preposition, and whether the article is involved. In Munster Irish, for instance, most prepositions cause aspiration without the article and eclipsis with it. For instance, ar bhord "on a table" but ar an mbord "on the table". (Other dialects would have ar an bhord for this.) Then there is the added complication that nouns often resist mutation in certain fixed phrases. So ar bord means not "on a table" but "on board [a vessel]".

Given all of this, it's best not to think of the initial mutations as being part of the dative case inflection, but as a phenomenon to be learned separately. It's pretty misleading of Carnie to even give a separate entry for "dative case" since the dative pretty much exists only in fossilised forms in most dialects, including those which form the basis for the standard. You can happily forget he ever mentioned it and simply memorise the exceptions (e.g. tigh Sheáin "chez Seán", where tigh is historically the dative of teach "house") as you come across them.

And forget about the entry for vocative as well. You're not going to be going around saying "O house! O table!" anyway. It's mainly used for names and then it's more or less the same as the genitive, e.g. a Sheáin "yo, Seán!". So, really, you only have two forms to worry about: the direct case (a.k.a. "nominative-accusative") and the genitive.

Does this help at all?

[Ciarán12]

Well, as for the use of multiple forms for plurals, I suppose that's a good thing from the perspective that if I make a mistake, it won't be overly grave. But on the other hand, I'd still like some guidelines to work from. Just to clarify, my query relates to the standard language. I'm not really interested in the dialects spoken in the Gaelteachtaí (I'm from the Gallteacht, I'd like to learn the standard language and build a "leinster" dialect around that). I suppose your right about the vocative, but I'd like to avoid having to learn off a bunch of exceptions by rote if at all possible. Having a system, even a complex one, would help me more than trying to memorize every example I come by. Obviously, I am aware that natural languages are never devoid of exceptions to rules, but if I can get a system that covers most things, I'll have to rote learn less exceptions. I suppose the difficulty is in finding one single "This is right, that is wrong" grammar, given the disagreement among dialects. I was hoping An Caighdeán Oifigiúil would give me at least one version of the language that I could learn, and then embellish colloquialisms from how native speakers from Dublin speak and from phrasing things like they are in Dublin English to give it a Dublin-y feel. But like I said, I need at least one single version to base my learning on first. I was just trying to distill a system from this madness!

[linguoboy]

Even though it's firmly based on living dialects and not any artificial koiné, you should have a look at Ó Siadhail's Modern Irish; it's excellent. When it comes to noun morphology, he divides them into three categories based on the formation of the genitive singular, each with several subtypes. They are:

Category 1: Consonant quality changes (4 subtypes, three distinguished by slenderising the final consonant and one by broadening it)
Category 2: Vowel addition (6 subtypes)
Category 3: Consonant addition (5 subtypes)

So that's 15 declensions compared to Carnie's 10. On the plus side, several have only a couple nouns in them and would likely be treated as "exceptions" in Carnie's scheme. (E.g. the only member of subtype 3E is caora "sheep".)

As I said before, even in its standardised form, Irish allows for more variation than a number of other languages. The CO admits multiple plural formations of such everyday nouns as bliain and ubh. I wasn't suggesting you "memorize every example" you come by, but sometimes it's best to start off with a scheme that accounts for the most common formations and then learn the minor patterns by exposure.

[Ciarán12]

That sounds really good, thanks! I haven't seen the book on the shelves here. That does sound promising though, I'll definitely have a look at it. I've come across those different plurals before. I tend to favour whichever ones i've heard the most in school.

[Ciarán12]

I think I have a handle on at least one of the aspects I was confused about now, and it has nothing to do with declensions. This is all according to Irish Nouns: A Reference Guide by Andrew Carnie, which I mentioned in the original post. It seems that there are two types of change: those affecting at the beginning of a word and those that affect the middle/end of a word. There seem to be 4 "forces" (for lack of a better word) that act on the beginning of a word: Eclipsis (Urú), Lenition (Séimhiú), T-prefixing and H-prefixing. There are 3 sets of rules for the application of Eclipsis and 2 for Lenition:

Lenition

Lenition 1 (L1)This is the general form of lenition used, for example, in the
past tense of verbs, but is also found in the nominal system after the vocative
particle a, the possessive pronouns mo ‘‘my’’, do ‘‘your’’, a ‘‘his’’, after the
numerals aon, dhá, trí, ceithre, cúig, sé, and after certain prepositions without
an article (such as ar, de, do, faoi, mar, ó, ríomh, trí, thar, and um).
The sound s, when followed by a p, t, or c (i.e., sc, sp, st) is not lenited.
If the word triggering lenition 1 ends in an n, then words beginning
with d, t, and s do not lenite.
Lenites p, b, m, f, t, d, s, c, g.

Lenition 2 (L2) - is found after certain forms of the
definite article an (the feminine singular common form, the masculine
singular genitive). Lenition 2 differs from lenition 1 in that s, when followed
by a vowel or l, n, r is lenited by prefixing a t. In addition, d and t are not
lenited.
Lenites p, b, m, f, s, c, g.

(There are two further forms of lenition he mentions, but they are rarely used).

Eclipsis

Eclipsis 1 (E1) -affects verbs after certain particles (e.g., cá ‘‘where’’, go
‘‘that’’, nach ‘‘not that’’, mura ‘‘if ’’, dá ‘‘if ’’ etc.), and nouns after a ‘‘their’’,
bhur ‘‘your plural’’, ár ‘‘our’’, seacht ‘‘seven’’, ocht ‘‘eight’’, naoi ‘‘nine’’, deich
‘‘ten’’, i ‘‘in’’, and after the article na but only when it marks the genitive plural.

p – bp, b – mb, f – bhf, t -dt, d – nd, c – gc, g – ng, vowel – n-vowel

Eclipsis 2 (E2) - happens after the masculine article an, when it
follows certain prepositions (ag, ar, as, chuig, faoi, le, ó, roimh, thar, trí,
um). It differs from eclipsis 1 in that vowels, d, and t are never eclipsed.

p – bp, b – mb, f – bhf, c – gc, g – ng

Eclipsis 3 (E3) - is the same as eclipsis 2, except s, which is not
normally eclipsed, takes the t prefix (just like in lenition 2).This happens in
the same environments as eclipsis 2, but on feminine nouns, e.g., ag an sagart
(masc. – no eclipsis or lenition) vs. ag an tsochraid (fem. – shows eclipsis).

p – bp, b – mb, f – bhf, c – gc, g – ng, s - ts

T-prefixing (T)
The singular masculine common determiner (article) an affixes
a t- to any word beginning with a vowel that immediately follows it.
am ‘‘time’’ an t-am ‘‘the time’’
uisce ‘‘water’’ an t-uisce ‘‘the water’’

H-prefixing (H)
In words beginning with a vowel
an h is preWxed to the word. It is triggered by a ‘‘her’’, cá ‘‘where’’, Dé ‘‘day’’, go
‘‘to’’, le ‘‘with’’, na ‘‘the’’ (in the genitive singular feminine, the common plural,
and the prepositional plural), Ó ‘‘family marker’’, chomh ‘‘as’’, and ná ‘‘don’t’’.
na héin ‘‘the birds’’
a háit ‘‘her place’’

He says that one of the most important and useful applications of these rules is in changing the noun after the definite article. He gives a table as follows of the articles, their numbers, genders and cases and the above rules they invoke on the following word in each instance:
Masculine Singular
Common case: an(T) Genitive Case: an(L2) Prepositional/Dative Case:an(E2) Vocative Case: a(L1)
Feminine Singular
Common case: an(L2) Genitive Case: na(H) Prepositional/Dative Case: an(E3) Vocative Case: a(L1)
Plural (both Genders)
Common case: na(H) Genitive Case: na(E1) Prepositional/Dative Case: na(H) Vocative Case: a(L1)

Without an Article
With no article, Genitive nouns take (L1) rule after plural nouns ending in a slender consonant.

Is this right? I understand this, so if it's right, I can learn off this system and at least I will then be able to deal with the changes at the start of the word. As far as I can tell, the above system should not interfere with whatever declension system I subscribe to (there seems to be a few; 5 declensions, 6 declesions, Carnie's own 10 declensions + 6 plural types, and the 15 declension system Linguoboy gave above).

Ó mo Dhia, nach bhfuil sé like, ró-confusing?

[linguoboy]

Is this right? I understand this, so if it's right, I can learn off this system and at least I will then be able to deal with the changes at the start of the word. As far as I can tell, the above system should not interfere with whatever declension system I subscribe to (there seems to be a few; 5 declensions, 6 declesions, Carnie's own 10 declensions + 6 plural types, and the 15 declension system Linguoboy gave above).

Ó mo Dhia, nach bhfuil sé like, ró-confusing?

I have to say I think Carnie makes it all more confusing than it has to be. His distinction between "Lenition 1" and "Lenition 2" and "Eclipsis 1" and "Eclipsis 2", for instance. Other sources simply teach the DeNTaLS rule. That is, when two dental sounds come together, lenition does not take place. Practically speaking, the most common case is after the article an, but this also helps account for, e.g. the lack of lenition after as although this is characteristic of most other simple prepositions.

(I assume Carnie's rules work for the Standard, but in Munster the DeNTaLS rule applies to lenition but not eclipsis. So an dorais "of the door" but ag an ndoras "at the door". Personally, I find it makes more sense to describe the variation in this way rather than saying that there are two distinct types of eclipsis and one of them simply doesn't exist in Munster.)

T-prefixation can also be thought of as a special case of lenition. Historically, the definite article had a final t in some circumstances which was later lost. However, it's been retained before vowel sounds, though the convention here is write it as a prefix to the following word. Thus an t-ainm rather than *ant ainm, although the latter better reflects the historical development. Similarly, one can think of an tsaoil as representing a respelling of *ant shaoil. First you have lenition of the consonant, then the /t/ reappears. (This is how Ó Siadhail prefers to deal with it at least.)

Everyone's brain works differently, but I personally found it worked better for me to learn phrases which illustrated these rules rather than trying to memorise all the rules by themselves. For instance, rather than committing to memory the fact that Dé prefixes h, I just learned the names of the days of the week, one of which happens to be Dé hAoine "Friday". Learning the rule Carnie gives will ensure that you say Dé hAoine correctly. But simply memorising Dé hAoine gives you the same benefit, and adds to your vocabulary at the same time--something the rule is no help with.

[Ciarán12]

I have to say I think Carnie makes it all more confusing than it has to be. His distinction between "Lenition 1" and "Lenition 2" and "Eclipsis 1" and "Eclipsis 2", for instance. Other sources simply teach the DeNTaLS rule. That is, when two dental sounds come together, lenition does not take place. Practically speaking, the most common case is after the article an, but this also helps account for, e.g. the lack of lenition after as although this is characteristic of most other simple prepositions.

(I assume Carnie's rules work for the Standard, but in Munster the DeNTaLS rule applies to lenition but not eclipsis. So an dorais "of the door" but ag an ndoras "at the door". Personally, I find it makes more sense to describe the variation in this way rather than saying that there are two distinct types of eclipsis and one of them simply doesn't exist in Munster.)

T-prefixation can also be thought of as a special case of lenition. Historically, the definite article had a final t in some circumstances which was later lost. However, it's been retained before vowel sounds, though the convention here is write it as a prefix to the following word. Thus an t-ainm rather than *ant ainm, although the latter better reflects the historical development. Similarly, one can think of an tsaoil as representing a respelling of *ant shaoil. First you have lenition of the consonant, then the /t/ reappears. (This is how Ó Siadhail prefers to deal with it at least.)

Everyone's brain works differently, but I personally found it worked better for me to learn phrases which illustrated these rules rather than trying to memorise all the rules by themselves. For instance, rather than committing to memory the fact that Dé prefixes h, I just learned the names of the days of the week, one of which happens to be Dé hAoine "Friday". Learning the rule Carnie gives will ensure that you say Dé hAoine correctly. But simply memorising Dé hAoine gives you the same benefit, and adds to your vocabulary at the same time--something the rule is no help with.

I don't think you appreciate the depth of my ignorance here . I had never heard of a "DeNTaLS" rule before, so I looked it up and in a comment on a forum (here http://www.irishgaelictranslator.com/tr ... 48879.html) I found an explanation that says "The 'DeNTaLS- DoTS' rule is a mnemonic for the exceptions to the normal rules for lenition". It goes on to explain the mnemonic, but I unfortunately don't know what the "normal rules" are to which these are exceptions! Is the "Lenition 1" rule Carnie mentions the general rule for Lenition? And how does this account for Eclipsis?
You make a valid point regarding "Dé hAoine", and come to think of it I already knew that that was the way to say "friday", but there are only 7 days of the week, so that wasn't such a huge thing to memorize. I'm really more concerned about things that come after the article and after prepositions and such. I would, for example, say "ar an mbord", but I have no idea why that is the correct form. It is nice to be able to get that correct without learning the rules just because it "sounds right", but my innate knowledge of how to inflect words like that is extremely limited, so I will need rules to cover all the other words. I can work out (using Carnie's rules) that "on the chair" would be "ar an gcathaoir". I could learn off by heart that "on the chair" is "ar an gcathaoir" without knowing why, but wouldn't that approach mean that I would have to take every change in a noun as a new lexical unit to be learned by rote, thus multiplying the number of words to be learned off severalfold?
I'd happily use the DeNTaLS-DoTS method if it works, but I'll have to learn the normal rules first. Can you (or anyone) explain these or point me to somewhere where they are explained? I haven't come across them in my search to clarify the matter so far .

[linguoboy]

Is the "Lenition 1" rule Carnie mentions the general rule for Lenition?

Yup.

And how does this account for Eclipsis?

Same deal: Eclipsis 1 is the general rule for eclipsis. As you can see, version 2 of each of these rules is just the general rule combined with the DeNTaLS rule.

I just noticed an error in Carnie's rules: he says to use Lenition 1 after aon rather than Leinition 2. Of course, if you just combined Lenition 1 with the DeNTaLS rule, you'd get this right.

I'm really more concerned about things that come after the article and after prepositions and such. I would, for example, say "ar an mbord", but I have no idea why that is the correct form. It is nice to be able to get that correct without learning the rules just because it "sounds right", but my innate knowledge of how to inflect words like that is extremely limited, so I will need rules to cover all the other words. I can work out (using Carnie's rules) that "on the chair" would be "ar an gcathaoir". I could learn off by heart that "on the chair" is "ar an gcathaoir" without knowing why, but wouldn't that approach mean that I would have to take every change in a noun as a new lexical unit to be learned by rote, thus multiplying the number of words to be learned off severalfold?

If you know the general rule for eclipsis, than learning ar an mbord is enough to remind you that every singular noun takes eclipsis in that context. You don't have to learn ar an gcathaoir separately.

Find out what bits of rules or exceptions you're having trouble remembering and then find phrases to remember. For instance, if you're having trouble keeping in mind that ár is followed by eclipsis, then commit to memory that Ár nAthair is "Our Father".

I'd happily use the DeNTaLS-DoTS method if it works, but I'll have to learn the normal rules first. Can you (or anyone) explain these or point me to somewhere where they are explained? I haven't come across them in my search to clarify the matter so far .

Like I said before, I find Ó Siadhail's analysis very clear. If you're looking for an online source, I recommend http://nualeargais.ie/gnag/anlaut1.htm. The Christian Brothers' grammar is available in PDF format, but I think it's more comprehensive than what the average learner needs. Moreover, it's completely in Irish, which could be more than you're willing to tackle at this stage.