Fòram na Gàidhlig

IainDonnchaidh wrote:

The do dh' / a dh' is not the result of this elusive infinitive that people keep trying to conjur up but simply a directional preposition when there is a verb of motion involved (a' dol, thàinig etc) i.e. I am doing towards drinking (in Gaelic anyway).

Hmmm ... I was simply following the regular verb conjugation guide in my dictionary

Mar eisimpleir:

Tha mi a' dol dham taigh-osda uisge-beatha a dh'òl.

no

Tha mi a' dol dhan bùth biadh a cheannich.

No no, that's quite right.

The difference here is "dol".

As Akerbeltz says, this "a" is really "do" -- a preposition of motion.

"iarr"/"iarraidh" is not a verb of motion, so it would be strange to use a preposition of motion, would it not?

Let's wind it back.

Gaelic is all about nouns, right? There's a reason Akerbeltz gets all het up about the word "infinitive" -- it hides the truth (and I finally understand why!)

So let's start with just plain nouns in English:
I want the house.
I'm going to the house.

Notice how with "want" there is no "to", but with "going" there is? Well that's obvious -- it's English and we all speak it well enough here!

Let's put them into Gaelic.
Tha mi ag iarraidh an taighe.
Tha mi a' dol dhan taigh. (= Tha mi a' dol do an taigh)

But a verbal noun is just like any other noun, so we can just replace "taigh" with any verbal noun:
Tha mi ag iarraidh òl.
Tha mi a' dol a dh'òl. (= Tha mi a' dol do dh'òl)

or
Tha mi ag iarraidh tighinn.
Tha mi a' dol a thighinn.

Gaelic goes to the event described by the verbal noun:
They're coming to take me away = They're coming to the taking away of me
but
They want to take me away = They want the taking away of me (no to)

The confusion here is that English seems to have taken the "to" from "going to" and generalised it to be a marker of the infinitive, so we as English speakers think about "wanting to" and "going to" as being the same construction, when they're not*; but if you stop thinking of verbal nouns as verbs and keep thinking of them as nouns, then obviously there's no "to" after "want"


*(The distinction occurs in other languages, too. Did you do any Spanish at school? Compare "quiero hacer algo" with "voy a hacer algo". Teachers might tell you that "a" isn't "to", but it looks like a duck and it quacks like a duck...!)

(Note correction to n/m below.)

IainDhonnachaidh wrote:Tha mi a' dol dhan taigh-osda uisge-beatha a dh'òl.

Tha mi a' dol dham bùth biadh a cheannich.

And yes, that means that "to" occurs twice in these sentences, but that's just like English, so no problem there:

I'm going to the hotel to drink whisky.
I'm going to the shop to buy food.

I guess I finally understood today why my 1971 TYG talks freely about infinitives but my 1993 TYG never uses the word once...

Except now I'm confused again.

Tha iad air a' clach a thogail. (From the appendix to Lamb 2008)

That "a thogail" -- what is the "a" this time? If they'd ate the rock rather than lifted it, would they be "air a' clach ithe", or "air a' clach a dh'ithe"? I was leaning towards the first, but the poor man's corpus isn't very helpful today....


Edit:
Found half my answer in Marks's verb book:
Bhiodh duine sam bith air a' cheist cheudna fhaighneachd (p158).

Unfortunately, Mark doesn't say what this is, instead using the nasty swearie word (*nf*n*t*v*) and on page 144 says "However, the a, the a dh' or even the lenition may be omitted according to the construction used."
Which of course obscures entirely the fact that These Are Different Things. Lumping them together as "the" infinitive just confuses everyone. (Including me, hence this thread.)

Michel Byrne, d. 127:

Tha mi air ròn fhaicinn

He groups it with examples like:

Feumaidh tu do chòta fhàgail

(Note correction to n/m below.)

I must have edited the wrong n to an m


Chan eil mi a' tuigsinn gu math fhathast

As in the above, a' tuigsinn is not a verbal noun, is it?

Tha mi ag iarraidh a thuigsinn (I am wanting to understand).

A bheil sin ceart?

Mar eisimpleir:

Tha mi a' dol dham taigh-osda uisge-beatha a dh'òl.

no

Tha mi a' dol dhan bùth biadh a cheannich.

Tha an t-òrdagh ceàrr agaibh ann.

"Tha mi a' dol dhan taigh-òsta a dh'òl uisge-beatha."

Tha mi a' dol dhan bhùth a cheannach biadh.

Ach, chanainn sa chumantas-

Tha mi a' dol dhan taigh-òsta airson uisge-beatha òl.

Tha mi a' dol dhan bhùth airson biadh a cheannach.

GunChleoc wrote:Michel Byrne, d. 127:

Tha mi air ròn fhaicinn

He groups it with examples like:

Feumaidh tu do chòta fhàgail

Which Byrne? I've only got the all-Gaelic one (Facal air an Fhacal) and that page number doesn't match it.

Iain,
Once I stop being confused myself I'll try to write a clearer explanation.

Chan eil mi a' tuigsinn gu math fhathast

"Chan eil mi aig tuigsinn..." => "Chan eil mi ag tuigsinn" => "Chan eil mi a' tuigsinn". 'S e ainmear gnìomhaireach a th' ann an "tuigsinn", mar a lorgar sa Chuimris (Nid wyf yn deall...), agus 's e "I'm not at understanding..." sa Bheurla gu litricheil. Ma dh'arras tu rudeigin a ràdh mar "I'm not understanding IT", feuch "Chan eil mi aig a thuigsinn" (I'm not at its understanding) => "Chan eil mi ga thuigsinn"

Tha mi ag iarraidh a thuigsinn

Seo "I'm wanting its understanding", .i. "I want to understand IT" agus tha sin ceart. Ach ma bhios tu airson "I'm wanting understanding" (I want to understand) a ràdh, feuch "Tha mi ag iarraidh tuigsinn"

Níall Beag wrote:

GunChleoc wrote:Michel Byrne, d. 127:

Tha mi air ròn fhaicinn

He groups it with examples like:

Feumaidh tu do chòta fhàgail

Which Byrne? I've only got the all-Gaelic one (Facal air an Fhacal) and that page number doesn't match it.

'S e am fear sa Bheurla a th' agam.

'S e "6.12.1 Complementing with Noun Objects" an cuspair.

Except now I'm confused again.

Tha iad air a' clach a thogail. (From the appendix to Lamb 2008)

LOL I remember being compuzzled by the same thing in uni till I made a list. Let me see if I can still rattle it off:

a 1: vocative particle (a Mhàiri.../Mary!)
a 2: masc. possessive (a chù.../his dog)
a 3: fem. possessive (a cù.../her dog)
a 4: reduced preposition do + dh' (do dh'Iain/a dh'Ìle.../to Iain, to Inverness)
a 5: reduced preposition de + dh' (de dh'uisge/a dh'uisge.../of water)
a 6: def. article a' (a' chlach.../the stone)
a 7: question particle (a bheil?.../is?)
a 8: relative pronoun (an rud a chunnaic mi.../the thing that I saw)
a 9: reduced form of ag (a' dèanamh.../at doing)
a 10: Niall's Duck (i.e. something that in some very limited types of sentence looks like it's marking an infinitive)

The reason I usually bring out a horsewhip to stop people from mentioning the infinitive is not because I don't like it but because people tend to misuse it. In many IE language the infinitive is a "lazy verb form" because it allows you to say many things with limited grammar and/or occurs VERY frequently - and of course it's the citation form of verbs.

In Gaelic, the form that occurs MOST often is the verbal noun in some shape or form, so telling people there is an infinitive gives them the wrong expectations. For starters, they suddenly think that the Gaelic citation form for a verb is the infinitive because annoyingly it is glossed as such: seall! to show ith! to eat, dèan! to do... when it's really an imperative.

Howsomeever.

Let me re-iterate that Gaelic does not have an infinitive, i.e. there is no special verb form that looks different from other verb forms in Gaelic that is what a linguist would call an infinitive. Not like e.g. Italian marking all infinitives in -are: mangiare, stuzzicare, malmenare...

Where other IE languages expect an infinitive, Gaelic does something totally different. The CLOSEST thing Gaelic has is an infinitive-like CONSTRUCTION:

Feumaidh mi ubhal a [a10] cheannach

As to the historic origins of a10, I'm not 100% sure. Either way, this occurs in a sentence of the type:

MODAL CONSTRUCTION + DIRECT OBJECT + VN
Feumaidh mi càr a bhriseadh
Bu toigh leam do leabhar a pheantadh
Bha agam ri do dhoras a mholadh

Lenition of the VN with a10 happens irrespective of gender, tense etc.

The reason I don't think this is an infinitive construction though in the IE sense is because a particle with the same behaviour crops up elsewhere, namely in perfective constructions:

Bha mi air clach a thogail
Bha Donnchadh air Màiri (a) fhaicinn

We either have to introduce a11, a perfective particle or, as I tend to, lump a 10 together with this one and call it a "leniting particle" because they behave the same. They just lenite, no dh' or anything.

So, building on that:
Tha mi ag iarraidh clann a bhualadh
Modal Construction + Direct Object + a10 + VN
Bha mi air clach a thogail
PAST + air + a10 + VN

Tha mi ag iarraidh bualadh
Modal Construction + VN (no particle required)

Bu toigh leam clann ri bhualadh
Bu toigh leam uisge ri òl
Modal Construction + Direct Object + ri + a2/a10 + VN
Hard to say here if it's fossilised a2 or a10

Bu toigh leam a h-òl
Feumaidh sinn a bhualadh
Modal Construction + a2/a3
This occurs when the direct object is a pronoun (which is also why I suspect we don't have a11 at all but fossilised a2).

Better now?

Oi! -- my duck was Spanish -- cuac cuac!

Anyway, this brings me back to what I used to call a10 a few years ago: a pseudo-article, something which makes the verbal noun definite.

Think about it:

• It only appears when the verbal noun occurs at the end of the noun phrase.
  It can be replaced by a possessive.

Sounds almost like the double-definite rule to me.
Except of course that it normally breaks the double-definite rule -- an doras (a) fhosgladh -- but that's why it's only pseudo. As I see it, the VN just wants to be definite, one way or another.


So, a10 occurs in inverted structures (direct object before VN), and only when that direct object is present.

Aside from "ag", the only non-inverting preposition is "do" (used after verbs of motion-as-intent -- eg they're coming to take me away).

Any other uninverted form would therefore have to be a complete nominalisation of the verbal noun, right?

So "fosgladh dorais" is "door opening" if it doesn't follow ag or do, but "opening/to open a door" after ag or do. On the other hand, "opening a door" without ag or do would be "doras fhosgladh" (a10 elided).

Crikey, I think I may finally understand the verbal noun.
The next step is to work out how to explain it....

I have the TYG dictionary (Robertson & MacDonald) and there are verb conjugation tables in the gramar section at the back.

Root/Verbal Noun/Infinitive/Subjunctive(conditional)

So basically this is wrong

IainDonnchaidh wrote:I have the TYG dictionary (Robertson & MacDonald) and there are verb conjugation tables in the gramar section at the back.

Root/Verbal Noun/Infinitive/Subjunctive(conditional)

So basically this is wrong

Yes, it's wrong, and for a mere learner to have the confidence to say that about the work of the Direct of the SMO, it must be very wrong.

A noun is a word. A verbal noun is a noun. Therefore a verbal noun must be a word. A word -- ie one, not two.

"ag ràdh" is two words.

"ràdh" is the verbal noun, and ràdh occurs in both the "verbal noun" and "infinitive" columns.

If you look at the whole list, you'll see that the same word appears in both these columns all the way down -- no matter whether the verb is regular or irregular, the two are identical, with the exception of the initial lenition.

What do we know about lenition? Lenite is caused by the preceding word, so tells us nothing about the word we are looking at.

Therefore, that word is the same in both columns.

The difference in the two columns is the prepositions. What do we know about prepositions? They are positioned directly pre(=before)... nouns. In this case verbal nouns.

Now, the two prepositions are slightly obscured by modern spelling.

"ag" (ag ràdh etc) is really "aig" -- "at". "aig" does not lenite nouns, so "ag" logically doesn't either. A verbal noun is just a noun after all.

"a" (a ràdh, a bhréith, a dh'fhaicinn) is really "do" -- "to". "do" does lenite nouns, so "a" logically does too. A verbal noun is just a noun after all.

Incidentally, in case someone has spotted that:

do dh'Ùistean
do dh'Fhlòra
do Dhòmhnall

and

a dh'Alba
a dh'Fhìobha
a Dhùn Éideann

stand vs

(a) òl
a dh'fhaicinn
a dhèanamh

I don't think this is a problem. The reduplicated dh' after all is a recent-ish innovation (cf Dàin do Eimhir) and the incongruence with a dh'fhaicinn may well be avoidance of leniting to zero.