Akerbeltz wiki: feedback on some historical details in a few articles

Ciamar a chanas mi.... / How do I say...
silmeth
Posts: 11
Joined: Tue Aug 11, 2020 12:47 pm
Language Level: lower intermediate, my Irish is better
Corrections: Please correct my grammar
Location: Poznań, A’ Phòlainn

Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by silmeth »

I don’t know how/where I could leave some feedback on historical info on Akerbeltz wiki, so decided to start a thread here.

First: huge thanks to the author as the whole wiki is a really great resource for information about Gaelic!

And I especially like that it often does give some historical background for the stuff discussed there – that’s very helpful! – there are some (mostly minor) details in a few articles which I think are not entirely accurate, though, and I hope my feedback can help to improve them. :)

Hope my criticism isn’t too harsh. ;-)

The Jesus is life? page

Not sure what it’s based on but I’ve recently got access to Brian Ó Cuív’s article The etymology of dia do bheatha (in Celtica, vol. 14, 1981) which presents very similar story but one detail is very different.
akerbeltz wiki wrote:The de is what it seems to be, the preposition "of" and bethu is, well, bheatha.
In Ó Cuív explanation the supposed original *rot·bia do bethu and *rot·bé do bethu would mean ‘thou shalt have thy life’ and ‘mayest thou have thy life’ – they contain do bethu in the sense your life. He even cites an argument against the preposition here – there are no instances of long dative bethaid attested in this phrase which strengthens the argument that bethu, later betha, should be interpreted as nominative (thus not following de).

This explanation was actually first suggested by an tAthair Peadar Ua Laoghaire in a 1897 letter in which he noticed a greeting rotfia do betha in “The adventure of Tadhg mac Céin”, and Ó Cuív mentions that “[t]he spelling rot fia is a fairly common variant of rot bia” (so the greeting in this form is attested in at least one late, 15th century, manuscript).

Ó Cuív also observes that Old Irish at·beir ‘(s)he says it’ evolved into later a-deir (and modern Irish deir(eann), Gaelic their) showing parallel development of internal /dv/, thus one can also reconstruct intermediate Middle Irish form ro·dia do betha (which later unsurprisingly lost the unstressed preverb, compare a-tátha, ad-chíchì, etc.).

This at·beir ‘says it’ → a·deir ‘says’ → deir parallel might be worth mentioning on the page (the -t·biadia development probably is the weirdest part of the explanation – at least was for me when I first read it years ago).

Ó Cuív also notices that in the earliest texts variants of this greeting are said to people who had recently been wounded or ill (so it’s a quite appropriate salutation) – and probably later spread during Christian times because of the association of initial dia with ‘god’ (so eg. in later texts it replaces older Dia latt ‘God with you’, eg. in Passion of Christ in Leabhar Breac, composed ~1200, ms. from early 15th c.: Aue rábii .i. Dia latt, a maigistir!, and in Smaointe Beatha Chríost, 15th c.: Aue rabi .i. Dia do betha, a Maighistir!; Passion of Christ: Dia latt, a rig na n-iúdaide!, Smaointe B. Chr.: De do betha, a ri!).

The Interrogatives or Who the what why? page

It claims that cò às? keeps old co· meaning ‘where’ – I don’t believe this to be true. Its main meaning was ‘how?’ and multiple sources don’t even list the meaning ‘where’, cf. eg. eDIL entry for 4 co. Thurneysen’s A Grammar of Old Irish gives it as part of cote, cate which “is used (sometimes also in the sense of ‘where is?’)” and notes “In ancient maxims, when co has the meaning ‘where?’ before other verbs, it is followed by -du- (probably ‘place’); e.g. codu·accobra creici cech dindba ‘where does every poor man seek to buy?’” – I don’t think Scottish Gaelic continues some usage from compounds of ancient maxims.

Instead it’s just regular meaning ‘which, who, what, where?’, just like ‘where’ which continues the exact same Old Irish interrogative, cía, which had all sorts of meanings depending on context. See explanations (c) and (o) in eDIL for cía used as ‘where?’.

This unstressed pretonic co· which merged with the following verb doesn’t seem relevant here to me, especially since ‘where’ wasn’t even its main meaning.

Later, about càite:
akerbeltz wiki wrote:Over time, this fused into one word càite, followed by a dependent verb because that's what this construction required back in the day, even though it no longer looks like it used to at all and even though it doesn't behave like the other question words.
Perhaps worth mentioning that today the same construction also requires the dependent form, eg. an taigh anns a bheil thu. Càite a is basically the same, except that the old short ia is used instead of newer longer anns a for ‘in which’ (also compare Irish an áit a bhfuil tú for ‘the place in which you are’ with the same old usage of a kept).

The Minding Your Ps and Qs or Why Porcom is a Headache page

Found this one recently by checking if there still is activity in the wiki – I clicked the Mùthaidhean ùra link. It does have a few more striking issues.
akerbeltz wiki wrote:Well, Indo-European, which is not recorded, but reconstructed based on what we know of its daughter languages, seems to have had an elaborate system of stops, 12 of them:
As a native speaker of a satem language I feel offended. Proto-Indo-European had 14–15 phonemic stops, the article is missing all the palatovelars (a.k.a. palatalized velars or palatal dorsals: *ḱ, *ǵ, *ǵʰ). There’s a reason why my native Polish has sto ‘hundred’ starting with an s and not a k.
akerbeltz wiki wrote:Late Common Celtic dropped p and said k wherever there was a p before.
Not sure how to understand it. In most positions *p was completely lost, eg.:
  • PIE *ph₂tḗr*ɸatīr → Late PCelt. *atīr → OIr. athair, cf. Latin pater, English father, Sanskrit pitā́,
  • PIE *pl̥h₂meh₂*ɸlāmā → Late PCelt. *lāmā → OIr. lám → Sc. Gaelic làmh, cf. Latin palma (→ English palm), Greek παλάμη (palámē), OEng. folm,
  • PIE *pl̥nós*ɸlānos → Late PCelt. *lānos → OIr. lán, cf. English full, Polish pełny.
But it got assimilated to /kʷ/ if the following syllable contained /kʷ/ (ie. *p…kʷ → *kʷ…kʷ). So Late Common Celtic mostly said nothing wherever there was a p before (at least between vowels), sometimes it said kʷ. (also *pC where C is any stop or /s/ changed to /xC/, eg. *septḿ̥*sextam → OIr. secht → Sc. Gaelic seachd, also other consonant clusters like *-sp- evolved a bit differently).
akerbeltz wiki wrote:Goidelic on the other hand would have none of it and did not redevelop the p. However, out of sheer spite, it merged the labial series with the plain stops so that kʷ merged with k and gʷ with g.
Might be worth mentioning that earlier in Celtic *gʷ merged with *b (and the *ɡʷ that remains continues older aspirated *ɡʷʰ) – but that might be too much detail (it’s seen in one pretty common word though, compare Gaelic bò and English cow from PIE *ɡʷṓu̯s).
akerbeltz wiki wrote:Well, it does explain it because in Manx because cóig is spelled queig.
Not sure if that’s just a joke about the spelling, or some suggestion that Manx continues the /kʷ/ as a separate phoneme.

If the latter – I don’t believe it does. Indeed in Manx five is queig /kweɡ´ ~ kwɛɡ/ – but I don’t believe it has anything to do with the labialized velar in Primitive Irish – the distinction between /k/ and /kʷ/ gets lost already in later ogam inscriptions, there’s no trace of it in Old Irish. I don’t know Manx history but I’d rather think that the vowel /oː/ got broken to a diphthong and gave /we/ later or something along those lines happened here.

Also it seems to me Manx for ‘head’ is typically spelt kione (and fockleyreen doesn’t know the spelling qione at all).

Later in the list of cognates the row with clòimh shouldn’t be there – it’s a borrowing from Latin plūma (and is rightfully listed as so in the table below), also the listed proto-form *petna makes no sense to me (and doesn’t look like an IE stem at all) – Wiktionary gives *plewk- (which would lose its *p in Celtic, cf. eg. làmh).

And the image at the end has quite a few issues:

The Old Irish forms cét, cóic (I guess the Goidelic in the image is transcription of them) should be transcribed as /kʲeːd/ and /koːɡʲ/ respectively (see my Guide to Old Irish spelling for an overview of basics of phonology and orthographic conventions in OIr.).

(Added later: Also I’d change the label to Old Irish – because Goidelic without further attribution just after Insular might suggest it’s Pritimitive Irish while it’s clearly post-apocope post-lost nasals OIr. forms (Prim.Ir. would have something like *kentan, later *kēdan for ‘hundred’, but that’s not attested).)

Scottish Gaelic koːkʲ doesn’t look right either, I’d expect either koːɡʲ or kʰoːkʲ.

Irish has cúig /kuːɡʲ/ (and on a side note, the word for ‘hundred’ in Munster, similarly to Sc. Gaelic, also has the long vowel broken: /kʲiad/, but /kʲeːd/ is good for other areas).

“Common Slavic” pjetĭ makes no sens at all, it should be pętь or pętĭ, or if using phonemic IPA, /pɛ̃tĭ/ – the PIE *-en- is continued as a nasal vowel in Proto-Slavic. Belarusian should either be the same as Russian (pʲatʲ) or perhaps pʲat͜sʲ (it most certainly doesn’t end in /c/). Slovak for ‘hundred’ should read stɔ – the same as Czech or Polish.

There probably are some other issues but those are the ones that caught my eye in this article.


akerbeltz
Rianaire
Posts: 1780
Joined: Mon Nov 17, 2008 1:26 am
Language Level: Barail am broinn baraille
Corrections: Please don't analyse my Gaelic
Location: Glaschu
Contact:

Re: Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by akerbeltz »

Thanks for the detailed feedback!
Hope my criticism isn’t too harsh
Heaven's no, you're become very nice about it and in the main, I concur with all points. A lot of that site, especially the older articles, are based on my notes from uni and that's a while back AND of course students don't always catch everything right :curam: and these days, I have few chances of fruitful exchanges at this level, so catching bloopers relies entirely on nice people like you dropping me a line.
The Jesus is life? page
Not sure what it’s based on but I’ve recently got access to Brian Ó Cuív’s article The etymology of dia do bheatha (in Celtica, vol. 14, 1981) which presents very similar story but one detail is very different.
Good question, I don't remember, most likely, study notes. But I fully agree with your points and have tweaked the page!
As a native speaker of a satem language I feel offended. Proto-Indo-European had 14–15 phonemic stops,
Haha true. I was taking a shortcut but forgot to explain the shortcut. Tweaked.
Not sure how to understand it. In most positions *p was completely lost, eg.:
Yeah, same problem, an unexplained shortcut and nobody to spank me for being naughty. Should be much better now! I'll think about whether to include ɡʷṓu̯s or not, I feel it's already quite complex.
If the latter – I don’t believe it does. Indeed in Manx five is queig
And you're right ... though it did say that in my notes but I can't remember which lecture is to blame! Tweaked, including kione and kicking pluma.
And the image at the end has quite a few issues:
I've added a note, the image uses a very dirty version of simplified IPA as it's just supposed to be a colourful family tree giving a general clue as to the changes along the nodes, not a perfect intro to IE comparative phonology :D I may revisit the image one day but I probably won't have time in a hurry.
The Interrogatives or Who the what why? page
Ok, you make valid points but I'm a little unsure about how to fix it - in the main because OI interrogatives were never my main focus. So if I get this straight, I need to kick the co· column completely (makes sense) and gloss cía/ce/ci as who/what/how/which/where and stick in a note that depending on what you combine this with, you get the remaining interrogatives? So cid > ciod é > gu dè > dè (what's the -d, do we know? ) still works and the noun combos (ce mét > co m(h)eud etc) are mostly obvious. But I'm lost now as to where the ò in cò comes from.
Perhaps worth mentioning that today the same construction also requires the dependent form,
It does say that higher up on the page.
silmeth
Posts: 11
Joined: Tue Aug 11, 2020 12:47 pm
Language Level: lower intermediate, my Irish is better
Corrections: Please correct my grammar
Location: Poznań, A’ Phòlainn

Re: Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by silmeth »

The pages do look better now. :)
if a word had an initial kʷ and if there was a second kʷ in the word somewhere, the p didn't disappear but became
Though I think you meant initial p, and then kʷ following later. ;-)
akerbeltz wrote: Sun Oct 31, 2021 7:36 pm Ok, you make valid points but I'm a little unsure about how to fix it - in the main because OI interrogatives were never my main focus. So if I get this straight, I need to kick the co· column completely (makes sense) and gloss cía/ce/ci as who/what/how/which/where and stick in a note that depending on what you combine this with, you get the remaining interrogatives?
Yup, at least that’s how I understand it – but bear in mind, I’m just an enthusiastic amateur, so asking someone actually competent ;-) (maybe David Stifter? he sometimes answers Old Irish stuff on his Twitter) might be a good idea too.
akerbeltz wrote: Sun Oct 31, 2021 7:36 pmSo cid > ciod é > gu dè > dè (what's the -d, do we know? ) still works and the noun combos (ce mét > co m(h)eud etc) are mostly obvious. (…)
Old Irish had unstressed interrogative (the same for all genders) ce·, ci·, cia·, and stressed cía ‘who, what, etc., etc.’ as feminine/masculine interrogative, cid /k´ið/ ‘what’ as its neuter counterpart¹ and plural citné (/k´id´N´eː/?). So cid was just stressed neuter form of cía.

And that neuter form is continued by later/modern forms like Sc. ciod or Munster Irish cad.

But it should regularly be *ciodh – I have no idea why the original /ð/ got delenited here.

It wasn’t in bardic Classical Gaelic – Irish Grammatical Tracts I² (§16 in Bergin’s edition, lines 271–274 in new Mac Cárthaigh’s edition) explicitly condems forms like cá, cad, cád, gad, gád… and instead gives as correct these: ciodh, ceadh, cía…, also créd (← O/MIr. crét from cía + rét ‘thing’, mod. rud).

But then modern Munster has cad, Scottish has remnants of ciod, and you get stuff like caidé, goidé or everywhere in between. Wildly guessing, but maybe influence from créad with non-lenited final /d/?
akerbeltz wrote: Sun Oct 31, 2021 7:36 pm(…) But I'm lost now as to where the ò in cò comes from.
And to be honest, I’m lost too – just like unlenited d above. You also get and in modern Irish dialects, and I have no idea what caused the vowel changes here. But evidently forms cá, cad, cé existed already in early modern times (since the tract explicitly condemns them), perhaps in Scotland too, then? The tract was written in Ireland so the author might not have been aware of it.

I’m also not sure why and have broad /k/ instead of slender (I guess just by influence from other merged forms like cairm, cáit(e)?).

____
¹ from PIE *kʷid, cf. Latin quid, and also English what from similar neuter *kʷod, see Wiktionary on *kʷis for more info
² the so-called Introductory tract, manuscripts from 17th and 18th centuries, but the text is likely much earlier, perhaps early 16th c.?
akerbeltz
Rianaire
Posts: 1780
Joined: Mon Nov 17, 2008 1:26 am
Language Level: Barail am broinn baraille
Corrections: Please don't analyse my Gaelic
Location: Glaschu
Contact:

Re: Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by akerbeltz »

Though I think you meant initial p, and then kʷ following later. ;-)
:mhoire: fixed
But it should regularly be *ciodh – I have no idea why the original /ð/ got delenited here.
Ciodh does appear as an obsolete form in some Gaelic dictionaries. But anyway, I agree there was probably some knock-on effect from a common form which somehow de-lenited the ð > d.
And to be honest, I’m lost too – just like unlenited d above. You also get cá and cé in modern Irish dialects, and I have no idea what caused the vowel changes here. But evidently forms cá, cad, cé existed already in early modern times (since the tract explicitly condemns them), perhaps có in Scotland too, then? The tract was written in Ireland so the author might not have been aware of it.
I suspect it's the back and roundness of the á that slipped into an ó. I mean, when I hear modern Irish á /ɑː/ it never sounds like an /ɑː/ to me but always like an /ɔ:/ sound. It's not a huge leap.

Anyway, you just beat me to posting that I've re-written the interrogatives page. I think all the booboos should be gone now
silmeth
Posts: 11
Joined: Tue Aug 11, 2020 12:47 pm
Language Level: lower intermediate, my Irish is better
Corrections: Please correct my grammar
Location: Poznań, A’ Phòlainn

Re: Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by silmeth »

akerbeltz wrote: Mon Nov 01, 2021 2:04 pmI suspect it's the back and roundness of the á that slipped into an ó. I mean, when I hear modern Irish á /ɑː/ it never sounds like an /ɑː/ to me but always like an /ɔ:/ sound. It's not a huge leap.
Hmm, you might be right. (West-South) Irish /ɑː/ is generally supposed to not be rounded (and it’s not in Ulster, but that’s different story, they say things like [æː] there) by default, but it definitely can be /ɒː/ or even /ɔː/ in contact with labials (and cá bhfuil…? ‘where is…?’ is a fairly common phrase, and can easily be /kɒːˈβˠilʲ ~ kɔːˈwilʲ/ or similar – Ó Sé in Gaeilge Chorca Dhuibhne gives boː [wɒː]n ɪ ʃuːl nə tr[ɑː] bó bhán ag siúl na trá as an example of such rounding, also noting that Bíonn an cháilíocht chruinn seo, ag druidim i dtreo [ɔː], ag roinnt cainteoirí go minic ‘Some speakers often have this rounded quality, closing towards [ɔː]’). Though it’s also often [aː] near slender consonants (eg. ɪ x´[aː] a Sheáin in Ó Sé’s).
Anyway, you just beat me to posting that I've re-written the interrogatives page. I think all the booboos should be gone now
Thanks! I do like the new overhauled page! :)

The last piece to fix would be the IE five and hundred infographic (some of the forms I mentioned are just wrong, even if you aim for simplified phonemic IPA) – but I understand editing the graphic might be a bit more involving (especially if you want to find someone to review forms in all the included languages).
GunChleoc
Rianaire
Posts: 4599
Joined: Mon Sep 17, 2007 11:26 am
Language Level: Mion-chùiseach
Corrections: Please correct my grammar
Location: Dùthaich mo chridhe
Contact:

Re: Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by GunChleoc »

Typo with closing tag on the interrogatives page:

Code: Select all

French qui/span>.
Oileanach chànan chuthachail
Na dealbhan agam
akerbeltz
Rianaire
Posts: 1780
Joined: Mon Nov 17, 2008 1:26 am
Language Level: Barail am broinn baraille
Corrections: Please don't analyse my Gaelic
Location: Glaschu
Contact:

Re: Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by akerbeltz »

Fixed :naire:
silmeth
Posts: 11
Joined: Tue Aug 11, 2020 12:47 pm
Language Level: lower intermediate, my Irish is better
Corrections: Please correct my grammar
Location: Poznań, A’ Phòlainn

Re: Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by silmeth »

akerbeltz wrote: Mon Nov 01, 2021 2:04 pm
But it should regularly be *ciodh – I have no idea why the original /ð/ got delenited here.
Ciodh does appear as an obsolete form in some Gaelic dictionaries. But anyway, I agree there was probably some knock-on effect from a common form which somehow de-lenited the ð > d.

Another mystery solved! :)

The answer hides in E. G. Quin’s article “Irish Cote in Ériu, vol. 20, 1966. Forms like ciod é, , Irish cad (é) do not derive from OIr. cid /k´ið/ → classical ciodh at all! (though maybe they were influenced by them and conflated to some extent?).

Old Irish had an interrogative cote, cate /kod´e, kad´e/ (with earlier forms like cati /kad´i/) meaning something like “what is the nature of it? of what kind it is?”, asking for definition, explanation. It itself was formed from co (this element which Thurneysen explains as “how, where?”, though probably originally it was just general interrogative “what?”) and the preposition di, de “from, of”. This gave the form caidhe in Classical Gaelic (the lenition of d is explained by Quin by fluctuations of de and dhe for the preposition, suggesting both lenited and unlenited forms could have been existing side by side) and later in various Gaelic dialects as caidhe, caidé, goidé and later and also cad (by analogy with copular sentences with the pronoun, since is é an fear… — is and an é an fear… — an, so also cad é an fear… — cad).

In Old Irish glosses and in Classical poetry this could also mean “where?” which I guess also explains phrases like Munster cad as duit? “where are you from?”. This meaning apparently developed from doubting rhetorical use of cote as “what is? where is?” expecting the negative answer “nowehere, it does not exist”, into an actual honest question about the physical location.
akerbeltz
Rianaire
Posts: 1780
Joined: Mon Nov 17, 2008 1:26 am
Language Level: Barail am broinn baraille
Corrections: Please don't analyse my Gaelic
Location: Glaschu
Contact:

Re: Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by akerbeltz »

Ah brilliant, that makes a lot of sense. I've tweaked the dé section, it now says:
dé "what?". Now the emphatic "fuller" form of this is gu dé and this itself is a re-interpreted form of ciod e (which is why Lewis dè with broad /d/ is actually the more conservative (i.e. historical) pronunciation), which Old Irish cote, cate meaning "what kind, of what nature", a combination of the interrogative plus the proposition de "of, from".
Mòran taing!
silmeth
Posts: 11
Joined: Tue Aug 11, 2020 12:47 pm
Language Level: lower intermediate, my Irish is better
Corrections: Please correct my grammar
Location: Poznań, A’ Phòlainn

Re: Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by silmeth »

akerbeltz wrote:Now the emphatic "fuller" form of this is gu dé and this itself is a re-interpreted form of ciod e (which is why Lewis dè with broad /d/ is actually the more conservative (i.e. historical) pronunciation)
Well, this part doesn’t seem to be true – OIr. form was /kod´e, kad´e/ with slender sound, so gu dé (~*guidé, like Irish goidé) is the more conservative one. But also note that in Munster Irish you get cad and cad é with broad /d/ (and the cad form used to be more popular outside of Munster historically too). So some kind of depalatalization there did happen.

I don’t fully understand why the Scottish form has slender /k/ in ciod (and not *cad, *cod) – but my guess is, this is exactly the conflation of cid /k´ið/ → ciodh (‘what’, neuter of cía) and cote, catecaidé; gu dé into ciod e.

Anyway, the idea that the e, é here is the masculine pronoun is a later reanalysis. Originally it was part of d(h)e ‘from him, off of him’ (Quin explicitly suggests the inflected 3rd person masc. form of the preposition).
akerbeltz
Rianaire
Posts: 1780
Joined: Mon Nov 17, 2008 1:26 am
Language Level: Barail am broinn baraille
Corrections: Please don't analyse my Gaelic
Location: Glaschu
Contact:

Re: Akerbeltz wiki: feedback on some historical details in a few articles

Unread post by akerbeltz »

dè < gu dè < ciod e I was taught at uni, MacBain also has it in his etymological dictionary. It may well be as you suggested, that there was an intermediate stage with depalatalization between cote and ciod e before it went back to being palatal in some areas with dè.
Post Reply