Questions about Welsh

[Jonlang]

While I was working yesterday I glanced up at my copy of Y Geiriadur Mawr and wondered what the etymology of geiriadur is. I know the word itself means "dictionary" and I quickly found that it's from gair word + -iadur with -adur being some sort of suffix. The Wiktionary entry for -adur isn't very helpful but it does link to words with the -adur suffix. Many of them, but not all, appear to be something to do with writing/recording information, like geiriadur word + adur = dictionary; dyddiadur day + iadur = diary; holiadur to question + adur = questionnaire. But others are less informative, like amseriadur as "pacemaker" or cysgadur as "lazy person". Is there some sort of rule for what -adur does?

[Yng]

Now I'm sure when I was first learning Welsh, I learnt that using nagw/nacw was a thing too, alongside nac ydw (and meaning the same thing), but I only ever seem to see nac ydw around anymore.

Was I imagining it, or is it actually a thing? And if so, where in Wales?

That's a southern thing. The nac particle is pronounced nag but written (formally) as nac. I think nagw is probably the negative form of the Southern yw.

[Znex]

Now I'm sure when I was first learning Welsh, I learnt that using nagw/nacw was a thing too, alongside nac ydw (and meaning the same thing), but I only ever seem to see nac ydw around anymore.

Was I imagining it, or is it actually a thing? And if so, where in Wales?

That's a southern thing. The nac particle is pronounced nag but written (formally) as nac. I think nagw is probably the negative form of the Southern yw.

Then what are the northern forms? Just nac ydw, nac wyt, nac ydy, etc. with no shortenings?

[Yng]

Nac ydy. Yw and ydy are descended from (yd)yw, which has an optional prefix.

I imagine nacw is supposed to just be the negative of yw (which itself sounds weird to me) but who knows how people down South use it.

[linguoboy]

While I was working yesterday I glanced up at my copy of Y Geiriadur Mawr and wondered what the etymology of geiriadur is. I know the word itself means "dictionary" and I quickly found that it's from gair word + -iadur with -adur being some sort of suffix. The Wiktionary entry for -adur isn't very helpful but it does link to words with the -adur suffix. Many of them, but not all, appear to be something to do with writing/recording information, like geiriadur word + adur = dictionary; dyddiadur day + iadur = diary; holiadur to question + adur = questionnaire. But others are less informative, like amseriadur as "pacemaker" or cysgadur as "lazy person". Is there some sort of rule for what -adur does?

-adur...trf. enwol gwthr....hefyd i ddynodi math o lyfr...a math o ddyfais.

Translation: "agent noun suffix...also for indicating a type of book...and a type of device."

[Jonlang]

While I was working yesterday I glanced up at my copy of Y Geiriadur Mawr and wondered what the etymology of geiriadur is. I know the word itself means "dictionary" and I quickly found that it's from gair word + -iadur with -adur being some sort of suffix. The Wiktionary entry for -adur isn't very helpful but it does link to words with the -adur suffix. Many of them, but not all, appear to be something to do with writing/recording information, like geiriadur word + adur = dictionary; dyddiadur day + iadur = diary; holiadur to question + adur = questionnaire. But others are less informative, like amseriadur as "pacemaker" or cysgadur as "lazy person". Is there some sort of rule for what -adur does?

-adur...trf. enwol gwthr....hefyd i ddynodi math o lyfr...a math o ddyfais.

Translation: "agent noun suffix...also for indicating a type of book...and a type of device."

Ah! So I was on the right track with it being to do with writing/recording information! Thanks!

[Jonlang]

Could someone explain to me why "I am going for a walk" is Dw i'n mynd am dro and not Dw i'n mynd am cerdded?

[Yng]

mynd am dro is a set phrase meaning 'go for a walk'. tro here means something like 'turn' or 'tour' (troi = to turn). mynd am gerdded would not be grammatical - you could say mynd i gerdded 'going to walk' but it would mean something else.

[Jonlang]

mynd am gerdded would not be grammatical

But why? Is it just because it isn't used and everyone uses the phrase mynd am dro and mynd am gerdded actually works grammatically but just isn't used, or is it because cerdded actually can't be used as a noun? Because I thought the whole point of verb-nouns was that they can be both I'm assuming phrases like I'm going for a bath / wash / shower / shave / run / jog / beer don't encounter the same problem as walk.

[linguoboy]

Why is "I'm going for a walk" grammatical in English but not *"I'm going for walking"?

Just because two words are the same part of speech doesn't mean that one can be used in all the same constructions as the other.

[Yng]

But why? Is it just because it isn't used and everyone uses the phrase mynd am dro and mynd am gerdded actually works grammatically but just isn't used, or is it because cerdded actually can't be used as a noun? Because I thought the whole point of verb-nouns was that they can be both I'm assuming phrases like I'm going for a bath / wash / shower / shave / run / jog / beer don't encounter the same problem as walk.

Well, 'verbnoun' and 'noun' are not the same thing. A verbnoun is similar to an infinitive or a gerund in English (and covers most of their uses) in that it's a nominal form of a verb - a form that acts, syntactically, much like a noun. This doesn't mean that it will always act exactly like every other noun. You'll notice that in your examples all the nouns are preceded by 'a'. You can't say, in English, 'I'm going for a running'. The verbnoun is most similar to this 'running' form in English.

That said, mynd am redeg does exist as a calque on 'go for a run' in English. Geiriadur yr Academi prefers mynd am dro bach, though.

[Jonlang]

But why? Is it just because it isn't used and everyone uses the phrase mynd am dro and mynd am gerdded actually works grammatically but just isn't used, or is it because cerdded actually can't be used as a noun? Because I thought the whole point of verb-nouns was that they can be both I'm assuming phrases like I'm going for a bath / wash / shower / shave / run / jog / beer don't encounter the same problem as walk.

Well, 'verbnoun' and 'noun' are not the same thing. A verbnoun is similar to an infinitive or a gerund in English (and covers most of their uses) in that it's a nominal form of a verb - a form that acts, syntactically, much like a noun. This doesn't mean that it will always act exactly like every other noun. You'll notice that in your examples all the nouns are preceded by 'a'. You can't say, in English, 'I'm going for a running'. The verbnoun is most similar to this 'running' form in English.

That said, mynd am redeg does exist as a calque on 'go for a run' in English. Geiriadur yr Academi prefers mynd am dro bach, though.

I thought that the "yn" was doing the job of the English "-ing" in the present tense, and "wedi" in the perfect etc. So there's no actual noun specifically for "walk"?

[Yng]

You're getting confused between different types of 'ing' and also trying to map English directly onto Welsh.

There are two -ing forms in English, the gerund or verbal noun (I like running, running is good for you - notice how it is acting like a noun) and the continuous active/present participle (I am running, the running man, I saw him running down the street). The Welsh verbnoun, in and of itself, is like the former and not like the latter:

rydw i'n hoffi rhedeg
mae rhedeg yn dda i ti

*dw i rhedeg

Welsh's equivalent of the English 'I am running' construction is not structurally the same. It's not 'I am' + an adjective. It's I am, plus a particle (or preposition), plus a verbal noun:

'I am in running' (where running is a gerund, a verbal noun - NOT an adjective) - rydw i'n rhedeg

There are loads of prepositions that can form constructions like this -

rydw i am siarad gyda fe - I'm for talking to him (this exists in my Welsh English and probably does in yours too)
rydw i ar siarad gyda fe - I'm on talking (about to talk) to him (dialectal)
rydw i wedi siarad gyda fe - I'm after talking (have talked) to him
rydw i heb siarad gyda fe - I'm without talking (haven't talked) to him

There is a noun for 'a walk' in the sense 'a wander, a hike, a trip around, a walk around the village, a walk up the hill' etc etc etc etc and it is tro. This is the idiomatic way of saying 'go for a walk' and also e.g. 'go for a drive' - es i am dro rownd y bentref 'I went for a drive round the village' etc, etc etc. We just don't generally say mynd am gerdded.

[Jonlang]

You're getting confused between different types of 'ing' and also trying to map English directly onto Welsh.

There are two -ing forms in English, the gerund or verbal noun (I like running, running is good for you - notice how it is acting like a noun) and the continuous active/present participle (I am running, the running man, I saw him running down the street). The Welsh verbnoun, in and of itself, is like the former and not like the latter:

rydw i'n hoffi rhedeg
mae rhedeg yn dda i ti

*dw i rhedeg

Welsh's equivalent of the English 'I am running' construction is not structurally the same. It's not 'I am' + an adjective. It's I am, plus a particle (or preposition), plus a verbal noun:

'I am in running' (where running is a gerund, a verbal noun - NOT an adjective) - rydw i'n rhedeg

There are loads of prepositions that can form constructions like this -

rydw i am siarad gyda fe - I'm for talking to him (this exists in my Welsh English and probably does in yours too)
rydw i ar siarad gyda fe - I'm on talking (about to talk) to him (dialectal)
rydw i wedi siarad gyda fe - I'm after talking (have talked) to him
rydw i heb siarad gyda fe - I'm without talking (haven't talked) to him

There is a noun for 'a walk' in the sense 'a wander, a hike, a trip around, a walk around the village, a walk up the hill' etc etc etc etc and it is tro. This is the idiomatic way of saying 'go for a walk' and also e.g. 'go for a drive' - es i am dro rownd y bentref 'I went for a drive round the village' etc, etc etc. We just don't generally say mynd am gerdded.

Sorry, I forgot to reply to this, and thanks for the help on that. It's very much appreciated.

Just a quickie: Does Welsh have a different way of expressing or asking about habituals? Like dych chi'n yfed coffi? can be translated as Are you drinking coffee? or Do you drink coffee?, yes? But in English the second question is usually meant as a question of habit, "Do you drink coffee?", "Does she eat meat?" we don't use it to ask if it's actually happening. But does Welsh use the same question for both and you just have to know which question is being asked, or does it express habitual questions differently?

[Yng]

Good question! The simple answer is that the dych chi construction covers both English present tenses (habitual and continuous) and it is perfectly fine to use it for both: ydych chi'n yfed coffi? - are you drinking coffee/do you drink coffee? This also goes in literary Welsh for the synthetic present: a yfwch goffi?

The slightly more complex answer is that in some dialects at least, there is a separate construction for the habitual which is formed with bydd, thus introducing a slightly different ambiguity: bydda i'n mynd i'r ysgol = I go to school (every day). But to me at least this doesn't line up exactly with all 'simple present' forms in English - I would still say dach chi'n yfed coffi? for 'do you drink coffee?' or fedri di'r Gymraeg? for 'can you [speak] Welsh?' because these are sort of dispositional, rather than habitual - they're not so much questions about repeated, habitual action as they are about overall ability or disposition, if you see what I mean?

If you want to avoid getting into that mess, though, I'd just stick with the general present for everything until you get an intuition otherwise.

Incidentally, dych is one of these horrible Cymraeg Byw forms which appear only in the messed-up Welsh of learners, proponents of ~standard Welsh~ and the victims of the utter failure of Welsh-medium schooling (outside areas which already had a majority Welsh-speaking population). Whilst there's probably more regional variation than just this, the normal forms are dw i, wyt ti, ma o, dan ni, dach chi, dan nhw in the North and w i/fi, wt ti, ma e, yn ni, ych chi, yn nhw in the South (I think there's more regional variation in these forms down there, though).

[Jonlang]

Good question! The simple answer is that the dych chi construction covers both English present tenses (habitual and continuous) and it is perfectly fine to use it for both: ydych chi'n yfed coffi? - are you drinking coffee/do you drink coffee? This also goes in literary Welsh for the synthetic present: a yfwch goffi?

The slightly more complex answer is that in some dialects at least, there is a separate construction for the habitual which is formed with bydd, thus introducing a slightly different ambiguity: bydda i'n mynd i'r ysgol = I go to school (every day). But to me at least this doesn't line up exactly with all 'simple present' forms in English - I would still say dach chi'n yfed coffi? for 'do you drink coffee?' or fedri di'r Gymraeg? for 'can you [speak] Welsh?' because these are sort of dispositional, rather than habitual - they're not so much questions about repeated, habitual action as they are about overall ability or disposition, if you see what I mean?

If you want to avoid getting into that mess, though, I'd just stick with the general present for everything until you get an intuition otherwise.

Incidentally, dych is one of these horrible Cymraeg Byw forms which appear only in the messed-up Welsh of learners, proponents of ~standard Welsh~ and the victims of the utter failure of Welsh-medium schooling (outside areas which already had a majority Welsh-speaking population). Whilst there's probably more regional variation than just this, the normal forms are dw i, wyt ti, ma o, dan ni, dach chi, dan nhw in the North and w i/fi, wt ti, ma e, yn ni, ych chi, yn nhw in the South (I think there's more regional variation in these forms down there, though).

Thanks! I asked the same question on Facebook and got this answer:

" To indicate something habitual we use the word "arfer" - so, "Do you normally drink coffee?" would be "Ydych chi'n arfer yfed coffi?" To say that you used to do something habitually or repeatedly over a period of time you would say: "Roeddwn i'n arfer ...". The Welsh for 'habits' is 'arferion'."

But I'm assuming that this is more "I am in the habit of drinking coffee", which sounds odd in English but for all I know it could be perfectly good everyday speech in Welsh.

[Yng]

Yeah, that's also used. Yn arfer means usually. But I'm only really used to seeing it as a calque on 'used to'.

[Jonlang]

When Welsh borrows English words there appear to be predictable patterns. Like Ps often become Bs (hospice -> hosbis); Ts often become Ds (Tavern -> Dafarn). But this doesn't always happen, like Policeman -> Plîsmon. Are there any rules for how Welsh accommodates borrowings? It's not particularly relevant to anything, it's just something I've been pondering.

[Yng]

Generally speaking borrowings are adapted to Welsh phonology and then spelt phonetically. hospis could exist, but is spelt hosbis because voicing is neutralised in stops after s (e.g. in ystafell which is pronounced, generally, [sdavɛɬ]). Dafarn is actually just a mutated form of tafarn (there's no t > d, you've just only encountered the mutated form).

There is a lot of regional and individual variation on how to deal with certain impermissible combinations of sounds, like final syllables with a schwa which do not appear in any Welsh dialects; as a result you get the standard plismon (I think /o/ is the most common way of dealing with it) but also p(o)lisman, etc. Likewise for 'fucking' you get ffwcan, ffycan, ffwcin, ffwcen etc etc etc. There is no clear consensus on how to spell certain other sounds like /dZ tS S/ which do not appear or only appear in restricted contexts in Welsh - they are borrowed sounds-wise perfectly happily, but historically /S/ could not appear word-finally and so there is no 'native' spelling convention for it; likewise /tS dZ/ (all of these sounds are present in native words in some southern dialects but spelling conventions were never formalised for them). <si sh tsh tsi dj j> among others are all sometimes used, but <si-> is the only one with proper credibility in official spelling and that can only be used word-initially.

I'm not entirely sure how those dialects that don't accept /e/ in final syllables treat loans with a final /e/, although it's worth pointing out that tafarn despite being the standard form looks like a straightforward borrowing of 'tavern' other than the final syllable having an /a/. But it might be an older borrowing from Latin taberna in which case the sound changes will be more long-term and may explain the /a/.

Note as well that even though Welsh is borrowing from a largely non-rhotic English variety (BrE), standardised loans and most common loans are rhotic - parcio etc.

Morphologically, nouns typically have an -(i)au plural (or borrow their English plural, as in plismyn/plismen and watsis). Verbs are usually formed with -(i)o, although lots of older English loans borrowed the OE infinitive suffix (lician for example).

Loans also act weirdly with regard to mutations and so on. Many old English borrowings (and the occasional native word like wyneb) that begin with <w> like wal from 'wall' have been reanalysed as mutated forms of an underlying form beginning with /g/ and thus become gwal, gwyneb etc). Borrowed forms beginning with g in English, though, almost never take the soft mutation (y gem fawr 'the big game') even where they should. In the northwest at least though almost all English loans can be mutated normally: dy blaysuit 'your playsuit', dy djips 'your chips', etc.

Gender-wise, English nominal borrowings were typically feminine historically I think.

[Jonlang]

As I understand it, Welsh compounds have a soft mutation in the second word: llysfam = llys + mam. So why isn't the town Pencoed not Pengoed? Or is there a different set of rules and circumstances surrounding place names?

[Yng]

As I understand it there are two types of compounds which I think are usually termed 'true' and 'pseudo-' compounds or whatever. True compounds are head final and take soft mutation, as llysfam (court-mum). Pseudo-compounds were originally possessive constructions or noun-adjective combinations which eventually ended up taking stress like normal words (pencoed < pen y coed?). Some of these - such as place names - retain now-archaic patterns of mutation, as in for example Llangollen (mutation of a possessor after a feminine noun, which was presumably originally productive).

[Jonlang]

As I understand it there are two types of compounds which I think are usually termed 'true' and 'pseudo-' compounds or whatever. True compounds are head final and take soft mutation, as llysfam (court-mum). Pseudo-compounds were originally possessive constructions or noun-adjective combinations which eventually ended up taking stress like normal words (pencoed < pen y coed?). Some of these - such as place names - retain now-archaic patterns of mutation, as in for example Llangollen (mutation of a possessor after a feminine noun, which was presumably originally productive).

So Llangollen was originally Llan collen?

[Yng]

No - what I'm saying is that there are two types of compounds. (I'm not 100% sure on this but this is how I recollect it). One kind, which is head-final and involves if you like the 'prefixation' of other nouns and adjectives, causes soft mutation by its own nature. Llysfam is an example of this (mam is the head and llys is modifying it). Another kind, which is head final, does not as part of the compounding process in itself have mutation. So pencoed for example does not have mutation. However, this DOESN'T mean that they can't have mutation for other reasons. As you may be aware, proper names in modern Welsh do not generally mutate. Likewise, possessors after feminine nouns (unlike adjectives) do not generally mutate. This was not always the case, however, and so Llangollen and similar compounds exhibit patterns of mutation which we wouldn't expect from modern Welsh. But this mutation is not because they're compounds but because of an old mutation rule which is now archaic.

[Jonlang]

I've just been doing a few exercises on Duolingo, and when topping up the Money section I gave an answer of Mae siwt yn costio pedwar deg tri phunt and it corrected phunt to punt, but I'm guessing this is because the aspirate mutation hasn't been introduced to the course at this stage. And this got me thinking: I know that un causes SM in feminine nouns, dau/dwy causes SM, tri and chwe(ch) cause AM, but I have no idea why. Are the mutations triggered by the phonetics or something else? I mean "tri phunt" rolls off the tongue easier than "tri punt" and "dwy gath" is a little easier than "dwy cath" but that could just be because I'm used to it being that way.

[Yng]

Well, they're all kinds of lenition - weakening of sounds. That's why it 'runs off the tongue' easier. But I would never use the aspirate mutation after tri.

The reason they exist where they do is largely etymological - historical processes of lenition (soft, aspirate and mixed mutations) and assimilation (nasal mutation) of different kinds in different environments. The results of these sound changes are also visible within words - the same sound change which produced the soft mutation (postvocalic lenition after final vowels which were later lost) also produced e.g. gwyrdd from Latin viridium.

'Mae siwt yn costio pedwar deg tri phunt' is wrong, anyway. You need to say Mae'r siwt yn costio pedwar deg tri phunt.