http://vk.com/doc12620867_132484536?has ... ead2e50bd0

Logically posited declensions based on data (not Latin!)

_________________
__̴ı̴̴̡̡̡ ̡͌l̡̡̡ ̡͌l̡*̡̡ ̴̡ı̴̴̡ ̡̡͡|̲̲̲͡͡͡ ̲▫̲͡ ̲̲̲͡͡π̲̲͡͡ ̲̲͡▫̲̲͡͡ ̲|̡̡̡ ̡ ̴̡ı̴̡̡ ̡͌l̡̡̡̡.___

That's a mighty bit of work. Worth a look.

_________________
Bí cinnte de go nglacfaidh triúr le gach aistriúchán a thabharfar.
Be sure to get three in agreement with a translation given.

This book is all about the Caighdeán Oifigiúil - which is not a proper object of academic study, because it is not a native language. Serious academic works should be on the Gaeltacht dialects, rather than on made-up forms of a language. The reason for this is that academic works are often based on extensive field research among native speakers of a language. E.g., while it may be the case that "bought" is the past tense of "to buy" in "Standard" English, a linguist would take a non-judgmental view and survey a target population for use of forms such as "bought" and "brought". Once it is an artificial standard that academic monographs are describing, we are locked into a circular loop. The information in this book is exactly the same as that given in Foclóir Beag.

Carnie writes:



I opened the PDF with some excitement - a list of all the possible declined forms of so many nouns in all dialects (which is what I understood Jay Bee to mean by "declensions based on data"). Ceistreacha and ceisteanna, teanga, teangaidh and teangain - would all be there. But once I got to page 1, I saw - they weren't. There was no field research informing this book at all. Worse, this work "refers us to grammars of specific dialects" - grammar books that don't exist - and it seems wrong to claim to refer readers to non-existent works.

If there is anything of value in this book - of genuine academic value meriting publication by the Oxford University Press - it is the discussion of how the original 5 declension system gave way to a more diffuse system - a 10-class pattern. But once again, the flaw - a flaw that would have scuppered the book had it been properly peer-reviewed - is that the 10-class pattern is just the CO - and not linked to any observed usage in the Gaeltacht.

A deeply flawed book, but one from which it is possible to glean the odd interesting point from one or two pages.

Hold yer horses! (Or is that horsaii? lol).

I think it's a bit harsh to call it 'deeply flawed'. I know it's not based on real Gaeltacht data and for making the assumption that 'data' (meaning corpus data) was enough, I apologise. You're right that real, living forms would be great, but that's not to be (or perhaps ever).

It is, however, of value in getting rid of the dogma of there being 5 declensions, a number that was never true in Latin and certainly never true in Irish for the exact number was never without variance

_________________
__̴ı̴̴̡̡̡ ̡͌l̡̡̡ ̡͌l̡*̡̡ ̴̡ı̴̴̡ ̡̡͡|̲̲̲͡͡͡ ̲▫̲͡ ̲̲̲͡͡π̲̲͡͡ ̲̲͡▫̲̲͡͡ ̲|̡̡̡ ̡ ̴̡ı̴̡̡ ̡͌l̡̡̡̡.___

The five declensions system was basically that any noun obeyed one of the following rules when forming the genitive:

1. Slenderise (capall -> capaill)
2. Slenderise and add -e (fuinneog -> fuinneoige)
3. Broaden and add -a (feirmeoir -> feirmeora)
4. Do nothing. (cailín -> cailín)
5. Add a consonant, or broaden. (lacha -> lachan, athair -> athar)

Now this can split further. First of all under heading (5.), adding a consonant and broadening are two different declensions. Secondly, the consonant can either be -d, -n, -nn, -ach or -adh (the last being quite rare). So (5.) is really six different declensions.

Similarly under (1.), you're often given the rule that -(e)ach slenderises to -(a)igh, e.g. portach -> portaigh.
However this isn't really slenderising and so it's actually a separate declension. Hence (1.) splits into two declensions.

All Carney has done is basically gone through every declension in the Caighdeán and broken them into the "real declension" system.

However I'm not sure how useful this is, because:
(a) There are a few declensions not given in his book that are to be found in the Gaeltacht, the -s declensions in Iorras Aithneach, the -adh declension in Munster and certain genitive plurals still to be found only in Ulster Irish.
(b) Some of the declension patterns of the Caighdeán seem to be at variance with every dialect of Irish and are based on Classical Irish. For example for Dair, I've heard Daraí as the genitive in Munster, and Dair as the genitive in Connacht. However the Caighdeán lists it as Darach, is this the genitive in Ulster, perhaps?

Actually it can be difficult to know how to inflect a word. In Munster alone there is quite a bit of variation, some nouns being second declension in Cork, but fifth in Kerry.

EDIT: I read the entire book last night, I really don't think it's that necessary. Lar's Grammadach na Gaeilge, available for free online is better.

_________________
The dialect I use is Cork Irish.
Ar sgáth a chéile a mhairid na daoine, lag agus láidir, uasal is íseal

Jay Bee and a Luin Duibh,

I think Carnie's point that the singular and plural declensions need to be separated, as they are not necessarily linked today, is valid and useful.

The declension system of the Munster Irish I'm learning can often appear mixed up, with a lot older forms left in. Take anál, "breath":

N: anál
G: análach/anála
D: anáil
Npl: anála

These are all attested forms. The first thing you notice is that **logically** you would expect anáile as the genitive singular, but it is análach, which doesn't fit the rest of the paradigm. You could say that análach is the gs where the npl. is análacha, but in fact I have found anála attested in the Npl, so we have a genitive that is explicable by a variant Npl, and not the Npl used by the writer whose Irish I'm learning. Then you find that there are two forms of the gs attested as well, by the same writer, and the second form anála possibly fits the paradigm better. This sort of mix up is interesting, but difficult to shoe horn into a set declension pattern - or at least not one that doesn't have a lot of footnotes and exceptions.

Carnie's section on the vocative seems odd - he admits (on numbered page 17 - which is p30 of the PDF) that nearly all of the vocatives listed in the book are not attested and have probably never been used, but says you could say "a rothara!" as an exclamation of annoyance with bikes, which he says means "Oh Bicycles!" Apart from the fact that "oh" is an interjection in English, whereas "O!" is a vocative particle - so he means "O Bicycles!" - the vocation is not an exclamation of annoyance with something. It is a mode of direct address. Unless you are specifically **talking to the bicycles**, you can't say "a rothara!" I'm not sure what annoyance with the bikes as an exclamation would be [can anyone advise?] - "na rothair úd! don deamhan iad!"??? Probably wrong - any comments? - but in any case "these damn(ed) bikes" is *not* a vocative expression. He is right to say that vocatives of inanimate objects can be used - even if rare - but **only** if those objects are being addressed, as with "a rotha" addressed to the wheels in Sliabh na mBan bhFionn. "A rothara!" is a possible form - but is not an exclamation - but an address to bikes - and may be more likely to be "a rothair!" or "a rotharaibh!" in the current Gaeltacht.

It is altogether easier to write works on Standard Irish, because it requires no field research, but that does not meet academic standards for linguistic monographs.

Patrick, thanks for catching anál, I've edited my post, I even said to myself an anál as I wrote it, so that was a ridiculous error on my part.

Quote:
I think Carnie's point that the singular and plural declensions need to be separated, as they are not necessarily linked today, is valid and useful.

That's true actually.

I've quite a bit to say on the rest of your post, but I'll need to gather some material, later today perhaps.

_________________
The dialect I use is Cork Irish.
Ar sgáth a chéile a mhairid na daoine, lag agus láidir, uasal is íseal

Well... slenderising or not, that's a question.

Historically, this was two different things, but it's been lost as the case system has simplified.

Consider "istigh" and "isteach". Different cases in the Old Irish system (location vs direction of motion). -ach would go to -ich in the genitive, and to -igh in... something else. I'm not up on the details. (Old Irish isn't my thing... so take all this as a sort-of-kind-of-approximate general-rule-of-thumb guideline only

As cases were lost, Scottish Gaelic hung onto one form (slenderise -ach->-aich) and Irish kept the other, more complicated one (c->g + slenderise).

While it's technically more than simply slenderisation, it does seem to be something that has been generalised to a lot of situations where CH is slenderised (as opposed to merely declension of singular nouns), so in the native speaker's model, it must be considered part of the rule of "slenderisation".

_________________
A language belongs to its native speakers, and when you speak it, you are a guest in their homes.
If you are not a good guest, you have no right to complain about receiving poor hospitality.

Harsh? Not at all. No-one would publish a book on English with a similar basis, because it wouldn't be considered good enough. OK, the guy probably couldn't get hold of funding to do it properly, but methodologically, it's just way out of date.

_________________
A language belongs to its native speakers, and when you speak it, you are a guest in their homes.
If you are not a good guest, you have no right to complain about receiving poor hospitality.

It's just a list of patterns! What sort of methodology would you be looking for? Diachronic and synchronic analysis? Of course we'd all love that, but it's no harm to add such a work to the corpus



Functionally, it falls within that perch



Indeed. And you know, here is what I would do for the 'Caighdeán' in school -like how there are multiple plurals -aí, acha, achaí, anna, annaí, achannaí, slenderisation etc, just allow any genitive to be used as needs be, the only provisios being gender and word structure. What harm would that do?



And who is to do this research?



Maybe he's being a bit camp in his address?!



But were there ever really 5? leaba/leapa, bean/mrá/mraoi etc. The truth is, there are as many as their are discreet patterns. i'm talking about the real spoken language, not any particular grammar. Understand this is where I'm coming from. I only see the grammars as viewpoints or models to be used, modified or discarded as needs be



Can you give us some examples? I think I have a photocopy of an Erris book somewhere

_________________
__̴ı̴̴̡̡̡ ̡͌l̡̡̡ ̡͌l̡*̡̡ ̴̡ı̴̴̡ ̡̡͡|̲̲̲͡͡͡ ̲▫̲͡ ̲̲̲͡͡π̲̲͡͡ ̲̲͡▫̲̲͡͡ ̲|̡̡̡ ̡ ̴̡ı̴̡̡ ̡͌l̡̡̡̡.___