Early old norse.

A forum for discussing linguistics or just languages in general.
Ephraim
sinic
sinic
Posts: 386
Joined: 15 Nov 2013 13:10
Location: Sweden

Re: Early old norse.

Post by Ephraim »

The Old Icelandic words á
In classic Old Norse, and specifically (Old) Icelandic, there are many homophonous words á. Many of these were probably distinguished in the early 9th century. I thought it might be interesting to examine the different words.

Below are all forms that appears as á on Wiktionary:
http://en.wiktionary.org/wiki/%C3%A1#Icelandic
http://en.wiktionary.org/wiki/%C3%A1#Old_Norse

*ā! interjection
Remained relatively unchanged throughout history, I think.

*āh v ‘own(s), have/has’, 1sg and 3sg prs.ind.act of *æiǥą
1sg: *āh
2sg: *āht
3sg: *āh
1pl: *æiǥum
2pl: *æiǥuð
3pl: *æiǥų

From PG *aih, inf *aiganą

*ą̄ adp ‘on’
From PG *ana.

*ɔ̄ f ‘ewe’, acc.sg of *ɔ̄ʀ
case: sg — pl
nom: ɔ̄ʀ — ɔ̄iʀ
acc: ɔ̄ — ɔ̄iʀ
dat: ɔ̄ — ɔ̄um
gen: ɔ̄aʀ — ɔ̄a

From PG *awiz. OIc has ær for the nom.sg, gen.sg, nom.pl and acc.pl. In other forms, the vowel is á. Early manuscripts have ǫ́, ǫ́um and similar forms, but they still have the æ-vowel before r. Here's my thinking:
  • W-umlaut affected the vowel while it was still short.
  • Medial *w was lost after short vowels, lengthening the preceding vowel (open stressed stems always have a long vowel).
  • The vowels in hiatus were not contracted.
  • The OIc vowel æ was caused by ʀ-umlaut and not i-umlaut. This is probably a relatively late change.
*ɔ̄w f ‘river’
case: sg — pl
nom: ɔ̄w — ɔ̄waʀ
acc: ɔ̄w — ɔ̄waʀ
dat: ɔ̄u — ɔ̄um
gen: ɔ̄waʀ — ɔ̄wa

From PG *ahwō > PN *ahwu. Early OIc manuscripts have ǫ́ (before the shift ǫ́ > á). The forms are based on these principles:
  • Medial and final *w remained after long vowels and diphthongs, as can be seen in early English loanwords. See E.V. Gordon, An Introduction to Old Norse, p. 328.
  • The semivowel *w occured after long vowels in tautosyllabic position for a time after the syncope and apocope, as it did in Old English. I guess it could be said to form overlong diphthongs. Loanwords perhaps indicate that this was still the case in the early 9th century.
  • *w was already lost before rounded vowels, however.
  • Short remained after monosyllabic stems ending in a long vowels, meaning that such stems counted as light.
  • There was less contraction of vowels in hiatus.
  • There was no contrast between a disyllabic sequence of a long vowel + and a monosyllabic sequence of a long vowel + *w. This is mostly speculation but it reduces the number of possible contrasts to uphold. Actually, *w and *ŭ were probably allophones in all positions. It appears that ultrashort developed as a vocalization (although the Runic inscriptions are ambiguous) of *w between consonants after vowel loss. Compare Rök <fiaru>, probably *feŭrŭ (although possibly *feŭrw), for OIc fjǫr, from PG *ferwą. Also Rök <karuʀ>, probably *gɔrŭʀ for OIc gǫrr.
  • *h was lost in most medial clusters (but not *ht) and between vowels, very much like in Old English.
  • *w caused mutation of the previous vowel in both East and West Nordic.
*ɔ̄w adv ‘always’
This is perhaps the most uncertain form. I can't find this OIc á in Cleasby/Vigfusson or Zoëga. The entry on Wiktionary appears to be yet another form of the word æ, which in OIc frequently appears as ei or ey. This is from a PG stem *aiwi–, with the interesting *aiw-sequence. According to Kock, OSw had a variant ā although SAOB writes that this form “kan ej anses säkert styrkt”.
User avatar
HinGambleGoth
sinic
sinic
Posts: 432
Joined: 01 Jul 2014 05:29
Location: gøtalandum

Re: Early old norse.

Post by HinGambleGoth »

Speaking of the dipthong shift, swedish has fiende with stress on the <i> whilst danish has fjende. Oswe "fiande", is the original verb *fia attested in Oswe? The noun "frände" from *fri-andi is smoothed out already in Oswe.

Interesting that alot of germanic nouns are present participles, like bō-andi = Oswe bōndi. friend and fiend are also old present participles as mentioned Above.
[:D] :se-og: :fi-al2: :swe:
[:)] :nor: :usa: :uk:
:wat: :dan: :se-sk2: :eng:
[B)] Image Image :deu:
Ephraim
sinic
sinic
Posts: 386
Joined: 15 Nov 2013 13:10
Location: Sweden

Re: Early old norse.

Post by Ephraim »

HinGambleGoth wrote:Speaking of the dipthong shift, swedish has fiende with stress on the <i> whilst danish has fjende.
Hjon discussed earlier may be another Sw example of the diphthong shift. There is also fräls and frälsa with smoothing after r. OIc has both frjálsa and frelsa.
HinGambleGoth wrote:Oswe "fiande", is the original verb *fia attested in Oswe?
I don't think it is. It's found in Ic, though as fjá ‘hate’.
HinGambleGoth wrote:The noun "frände" from *fri-andi is smoothed out already in Oswe.
In OIc as well, I think frændi is the most common form here. But I think older frjándi is attested, too.
User avatar
HinGambleGoth
sinic
sinic
Posts: 432
Joined: 01 Jul 2014 05:29
Location: gøtalandum

Re: Early old norse.

Post by HinGambleGoth »

I have been reading Eduard Prokosch's A comparative Germanic grammar and cam across this.
Image
Image

Germanic *z transcribed as *ʀ in Old Norse had a different influence on preceding vowels than *r , this is something of a theme in Germanic, namely "thick" and "thin" consonants, maybe medieval Germanic dialects distinguished between "thick" short [ɫ] and "thin" long [ʎ:], as stated Old-Germanic presumably had thick, almost retroflex coronal consonants that had a similar effect preceding vowels as velars.

OHG seems to have distinguished germanic *t <z> from *s <s> and also *k <ch> from *x <h>.
[:D] :se-og: :fi-al2: :swe:
[:)] :nor: :usa: :uk:
:wat: :dan: :se-sk2: :eng:
[B)] Image Image :deu:
Ephraim
sinic
sinic
Posts: 386
Joined: 15 Nov 2013 13:10
Location: Sweden

Re: Early old norse.

Post by Ephraim »

Many titles by Eduard Prokosch are available to read for free on-line, by the way. You need a login to download a pdf:

The sounds and history of the German language (1916)
http://babel.hathitrust.org/cgi/pt?id=n ... 3075922678

An outline of German historical grammar (1933)
http://babel.hathitrust.org/cgi/pt?id=m ... 5017683031

A comparative Germanic grammar (1939)
http://babel.hathitrust.org/cgi/pt?id=m ... 5002329137
HinGambleGoth wrote:Germanic *z transcribed as *ʀ in Old Norse had a different influence on preceding vowels than *r , this is something of a theme in Germanic, namely "thick" and "thin" consonants, maybe medieval Germanic dialects distinguished between "thick" short [ɫ] and "thin" long [ʎ:], as stated Old-Germanic presumably had thick, almost retroflex coronal consonants that had a similar effect preceding vowels as velars.
Yes, it has often been noticed that there are similarities between *r and the velars in the history of Germanic, and indeed the history of Indo-European.
https://www.academia.edu/3176734/Pharyn ... o-European
https://www.academia.edu/1525634/German ... i_Dialects

There is also an argument to be for ON ʀ being palatal in some way. Palatalized r may not just be a name. Apart from the ʀ-umlaut of preceding vowels, there is also iʀ-umlaut where *ir causes umlaut in environments where simple *i does not. This suggests that perhaps unstressed *i had a different quality before *ʀ. And conversely, *r often shifts to *ʀ after *i. Often *ʀ shifted to *r after dentals, which I'm not sure what to make of.

So perhaps *r and *ʀ were really /rˠ/ and /rʲ/, and the later thick l sound and palatalization of geminate *ll is perhaps an indication that a similar distinction was made for *l, if not fully phonemic. There is similarities in the treatment after /i/, too.
HinGambleGoth wrote:OHG seems to have distinguished germanic *t <z> from *s <s>
Not only that, it distinguished:
ȥ from PG final *t and medial non-geminate *t (the latter became geminated)
z from PG initial and geminated *t, and *t following a liquid or nasal.
s from PG *s.

The spellings <ȥ> (sometimes <ʒ>) and <z> are just academic conventions. But the interesting medieval distinction was between ȥ and s, which were presumably some kind of sibilants (with z being an affricate). I think the distinction may have persisted into MHG.
Ephraim
sinic
sinic
Posts: 386
Joined: 15 Nov 2013 13:10
Location: Sweden

Re: Early old norse.

Post by Ephraim »

All talks from the Third Conference on Elfdalian are now available as videos. Some are in English:
http://elvdalsk.inss.ku.dk/program/

Of special interest for Early Old Norse is Nikolai Brink Sandbeck's talk on The Nasal Vowels in Old Norse (in English).
User avatar
HinGambleGoth
sinic
sinic
Posts: 432
Joined: 01 Jul 2014 05:29
Location: gøtalandum

Re: Early old norse.

Post by HinGambleGoth »

Ephraim wrote:All talks from the Third Conference on Elfdalian are now available as videos. Some are in English:
http://elvdalsk.inss.ku.dk/program/
Never knew that gamall m and gamal f where still distinct, interesting. The loss of short n in fenemine adjectives/articles is common though, another thing that wasnt marked in most manuscripts, even though we can tell the distinction in many traditional dialects.

:elf: has a funny prosody, sounds a bit "whiny"
Ephraim wrote: Of special interest for Early Old Norse is Nikolai Brink Sandbeck's talk on The Nasal Vowels in Old Norse (in English).
It is interesting that nasalization was recorded and apparently fairly common, but was rarely or "optionaly" marked, It is interesting that you can gather much information on the sounds of ON from dialects, showing features that the medieval sources didnt bother with.

Image

as shown in this Oswe manuscript medial gemination is marked "annar" but not final as in "man" neither is vowel length, like long /a:/ in "mal" compared to /a/ in "bana" some west norse did mark long consonants with capital letters.
[:D] :se-og: :fi-al2: :swe:
[:)] :nor: :usa: :uk:
:wat: :dan: :se-sk2: :eng:
[B)] Image Image :deu:
User avatar
Aevas
admin
admin
Posts: 1445
Joined: 11 May 2010 05:46
Location: ꜱᴇ

Re: Early old norse.

Post by Aevas »

It doesn't say man, it says manne [;)]
User avatar
HinGambleGoth
sinic
sinic
Posts: 432
Joined: 01 Jul 2014 05:29
Location: gøtalandum

Re: Early old norse.

Post by HinGambleGoth »

hár byrias uaþa mal ok ſara mal. hor. ran ok ſtyld.
huggár maþár at manne ok uarþár annar firi.
ok far af bana. egh ma flere hug til uaþa taka án et.
af enu hugge kunnu tu sar uarþa. takár hand ok huvuþ.
þát ár høghſte uaþe. tiughu markár ok.


My attempt at normalization.

hǣr byrjas vāþa māl ok sāra māl. hōr. rān ok styld.
huggẹr maþẹr āt manni ok varþẹr annar fyri.
ok fār af bana. ¹ēgh mā flēri hugg til vāþa taka æn ētt.
af ēnu huggi kunnu tū sār varþa. takẹr hand ok huvuþ.
þæt ær hø̄ghsti vāþi. tiūghu markẹr ok.

¹ not sure, should maybe be short since it was often unstressed.
[:D] :se-og: :fi-al2: :swe:
[:)] :nor: :usa: :uk:
:wat: :dan: :se-sk2: :eng:
[B)] Image Image :deu:
Ephraim
sinic
sinic
Posts: 386
Joined: 15 Nov 2013 13:10
Location: Sweden

Re: Early old norse.

Post by Ephraim »

HinGambleGoth wrote:I have been reading Eduard Prokosch's A comparative Germanic grammar and cam across this.
Interestingly, Elfdalian did not lower *i before *ht as other Norse dialects. Guus Kroonen gives the following examples in his talk on the diachrony of Elfdalian:

PG *rehtaz > Early ON *rehtʀ > Elf rett, Sw rätt, OIc réttr
PG *linhtaz > Early ON *lį̄htʀ > Elf lit(t), Sw lätt, OIc léttr
PG *þinhtaz > Early ON *þį̄htʀ > Elf tit(t), Sw tät, OIc þéttr

http://elvdalsk.inss.ku.dk/program/
http://video.ku.dk/elvdalsk-introduktio ... lvdalsk-ii
http://elvdalsk.inss.ku.dk/program/hand ... alian.pptx
HinGambleGoth wrote:hǣr byrjas vāþa māl ok sāra māl. hōr. rān ok styld.
Perhaps you can normalize this with commas.
HinGambleGoth wrote:tiūghu
I think this should have a short diphthong as in siunga (I'm using something like Wesséns or your normalization for OSw in this post), not long as in diūr.

The origin is of course Early ON *twą̄ teŭǥų, accusative (plural) of *twai tiǥiʀ with ellipsis of the *twą̄ in Sw (but not in OIc which has tuttugu). The Early ON short diphthong *eŭ (from breaking of PG *e) of course normally developed to io in OSw, but I think the shift to iu may be connected to the u in the following syllable. This is also seen in the numeral fiughur, from Early ON feŭǥur (OIc fjǫgur). Compare the similar shift ɔ > u in hɔggwa > hugga (OIc hǫggva).
HinGambleGoth wrote:¹ēgh ¹ not sure, should maybe be short since it was often unstressed.
Not really sure either. It's a bit of an odd word. The word was obviously spelled in many ways in the manuscripts: <eigh>, <ey>, <eg> etc. I'm also not sure if the shift gh > j in this word is because of the following *i (in earlier *æiǥi) and if spellings like <egh> actually represented ē̆j from an early time, or if the shift is because of the preceding front vowel and <egh> really represented ē̆gh. The shift ghj > j took place in the early 14th cent. according to Wessén so it might not have taken place when this text was written (is this Östgötalagen?), whereas the final vowel seems to have been lost already. However, I don't think the shift gh > j after front vowels is a regular shift in OSw. The shift in unstressed words like mig and dig is much later, I think.

In any case, vowels were later shortened immediately before j.
User avatar
sestir
hieroglyphic
hieroglyphic
Posts: 61
Joined: 22 Mar 2015 16:05
Location: Markaryd
Contact:

Re: Early old norse.

Post by sestir »

Ephraim wrote:Yes, it has often been noticed that there are similarities between *r and the velars in the history of Germanic, and indeed the history of Indo-European.
https://www.academia.edu/3176734/Pharyn ... o-European
https://www.academia.edu/1525634/German ... i_Dialects

There is also an argument to be for ON ʀ being palatal in some way.
Does this mean to suggest that neither r, nor ʀ were alveolar? Wouldn't an alveolar sound have been closer to PG z?
User avatar
HinGambleGoth
sinic
sinic
Posts: 432
Joined: 01 Jul 2014 05:29
Location: gøtalandum

Re: Early old norse.

Post by HinGambleGoth »

sestir wrote: Does this mean to suggest that neither r, nor ʀ were alveolar? Wouldn't an alveolar sound have been closer to PG z?
There seems to be alot of circumstantial evidence to suggest that PIE *s was strongly retracted, if not downright post-alveolar, this would mean that the unstressed voiced allophone in PG would have been that as well, realized closer to [ʒ] than [z]. In IE langauges that didnt create "secondary" sibilants from affricates like /ts/ or /tʃ/ like Danish /s/ is more retracted. Medieval romance langauges and early high german contrasted "primary" and "secondary" /s/ during the middle ages.

http://www.academia.edu/6149816/Proto-Indo-Euro_s
[:D] :se-og: :fi-al2: :swe:
[:)] :nor: :usa: :uk:
:wat: :dan: :se-sk2: :eng:
[B)] Image Image :deu:
Ephraim
sinic
sinic
Posts: 386
Joined: 15 Nov 2013 13:10
Location: Sweden

Re: Early old norse.

Post by Ephraim »

sestir wrote:Does this mean to suggest that neither r, nor ʀ were alveolar? Wouldn't an alveolar sound have been closer to PG z?
The primary articulation could have been alveolar for both r and ʀ. Both were almost certainly coronal at least (dental, alveolar, retracted alveolar, post-alveolar etc.).

The suggestion is that both r and ʀ had some secondary articulation, r might have been velarized ([rˠ]) or, less likely, pharyngealized ([rˤ]) and ʀ palatalized ([rʲ]). The secondary articulation may not have been found in all positions, the contrast might have been neutralized to plain [r] in some positions.

It's very common for languages that have palatalized consonants to also velarize non-palatalized consonants in order to increase the contrast (and vice versa, I assume). This is certainly the case in Irish and Marshallese for example, and I think it also happens in Slavic languages at least to some extent. So these languages may lack some plain consonants.

As HinGambleGoth pointed out, PG *s and *z were likely somewhat retracted compared to English or Swedish /s/, as is common in languages that don't contrast /s/ and post-alveolars, and as it still is in Modern Icelandic.

It's possible for [rʲ] to become [ʒ], this happened in Polish (it might have passed through a stage of a partially fricative trill like in Czech). So perhaps the reverse is not to strange, that a retracted fricative [ʒ~z̠] rhotasized into palatalized [rʲ] rather than plain [r].

If PG *r was indeed already velarized, this might also have worked in favour of *z becoming palatal [rʲ] rather than plain [r].
User avatar
HinGambleGoth
sinic
sinic
Posts: 432
Joined: 01 Jul 2014 05:29
Location: gøtalandum

Re: Early old norse.

Post by HinGambleGoth »

Ephraim wrote: I will (for now at least) write <Ĭ ĭ Ŭ ŭ> for unstressed short i and u that would later be syncopated or apocopated. The unstressed vowel system of Early Old Norse, before the last syncope stage, is a topic for discussion but these must in some way have differed from the regular i and u that was not lost in the same environment.
I have been bugged by this since this thread started, for some reason u drops in sunu =sun after the 9th century, but Then you have fenemine short stems like wiku, still found in modern dialects, the pre-manuscript Viking era language must have distinguished more than the classical literary norse three unstressed vowels somehow, but once again the younger futhark is insufficient and the corpus to small to have direct evidence.

Maybe u~o and i~e but I am a layman to PG declension so I dont know where their origin or distribution would be in 9th century North germanic.
[:D] :se-og: :fi-al2: :swe:
[:)] :nor: :usa: :uk:
:wat: :dan: :se-sk2: :eng:
[B)] Image Image :deu:
User avatar
Xonen
moderator
moderator
Posts: 1080
Joined: 16 May 2010 00:25

Re: Early old norse.

Post by Xonen »

Ephraim wrote:It's possible for [rʲ] to become [ʒ], this happened in Polish (it might have passed through a stage of a partially fricative trill like in Czech). So perhaps the reverse is not to strange
It kind of is, though. Palatalized alveolar sounds turning into postalveolars (or perhaps raised alveolars in this case) makes sense, because it's a way of achieving a similar acoustic effect through a somewhat simpler articulation. The reverse would effectively be adding a secondary articulation to a sound for no particular reason, which seems a lot harder to explain. Not that it couldn't happen, of course - anything's possible - but I'm a bit skeptical. Germanic rhotics in general seem to be an area that could use some further study.
Ephraim
sinic
sinic
Posts: 386
Joined: 15 Nov 2013 13:10
Location: Sweden

Re: Early old norse.

Post by Ephraim »

HinGambleGoth wrote:I have been bugged by this since this thread started, for some reason u drops in sunu =sun after the 9th century, [...], the pre-manuscript Viking era language must have distinguished more than the classical literary norse three unstressed vowels somehow, but once again the younger futhark is insufficient and the corpus to small to have direct evidence.

Maybe u~o and i~e but I am a layman to PG declension so I dont know where their origin or distribution would be in 9th century North germanic.
It's etymologically predictable which *u:s and *i:s are dropped in the 9th century. They're dropped if they were short and oral in Proto-Norse. Vowels that were long and/or nasal in Proto-Norse remain (note that the PG nasalization in unstressed syllables must have been lost in PN). I think we have to conclude from this that 9th century Old Norse made some additional contrasts among the unstressed vowels, at least after short stems, because the language can't just remember which vowels to drop. Of course, the distinction is not made in the Younger Fuþąrk which had very few vowel letters.

Interestingly, it also seems that some *ŭ come from earlier *w:
Rök <fiaru>, probably *feŭrŭ, OIc fjǫr < PG *ferhwą
Rök <karuʀ>, probably gɔrŭʀ, OIc gǫrr < PG *garwaz

It is possible that the contrasts ŭ:u and ĭ:i remained when the vowels was nasalized but since the nasal vowels were never dropped, it's hard to know for certain. It's also entirely possible that the contrast was neutralized here (which is what I assume in my reconstruction).

The nature of the contrast is highly uncertain, though. The PN origin of the contrast between Early ON *u and *ŭ was one of length, which is why I mark the latter with a breve. It is tempting to think that *ŭ and *ĭ were ultrashort but it could also have been a contrast in quality.
HinGambleGoth wrote: but Then you have fenemine short stems like wiku, still found in modern dialects,
Do you mean forms like Icelandic viku, oblique singular of vika ‘week’? The reason this u remains is that it was both long and nasal in Proto-Norse. Early ON must have had nom *wika and obl *wikų.
Compare Old High German zunga:zungūn.
Xonen wrote:
Ephraim wrote:It's possible for [rʲ] to become [ʒ], this happened in Polish (it might have passed through a stage of a partially fricative trill like in Czech). So perhaps the reverse is not to strange
It kind of is, though. Palatalized alveolar sounds turning into postalveolars (or perhaps raised alveolars in this case) makes sense, because it's a way of achieving a similar acoustic effect through a somewhat simpler articulation. The reverse would effectively be adding a secondary articulation to a sound for no particular reason, which seems a lot harder to explain. Not that it couldn't happen, of course - anything's possible - but I'm a bit skeptical. Germanic rhotics in general seem to be an area that could use some further study.
Well, it's definitely not a common sound change. I tried to find an example of a sibilant rhotacizing into [rʲ] as opposed to plain [r] and I couldn't find any natlang examples (I did find one example from a conlang). Many South Slavic languages apparently have rhotacization of *ʒ (which is quite old) but from what I was able to tell, it became plain *r rather than *rʲ. Now, most South Slavic languages seem to have simply depalatalized older *rʲ (probably after the rhotacization), but from what I understand Slovene has both the rhotacization of *ʒ and different reflexes of older *r and *rʲ, and the result of rhotacization seems to have been plain *r.

I think you can make some sense of a shift of PG *z to [rʲ] if the normal *r was already velarized [rˠ] (or had some other secondary articulation). Perhaps *z simply rhotacized into plain [r] which did not merge with the velarized *r, and was later pushed into a more palatalized articulation in order to increase the contrast.

Still, the nature of the contrast r:ʀ is extremly uncertain and while the area could use some more research, it may be that there simply isn't enough evidence available to say anything remotely definitive. There are of course other proposals, and while I think some are very unlikely (I think Kortlandt proposed a voiced:voiceless contrast) some seem quite possible (ʀ being an approximant or voiced sibilant with a different point of articulation from /s/).
User avatar
HinGambleGoth
sinic
sinic
Posts: 432
Joined: 01 Jul 2014 05:29
Location: gøtalandum

Re: Early old norse.

Post by HinGambleGoth »

I have to agree with Xonen here, i dont see any motivation for a germanic language, that is charachterized for weakening and reduction to strenghten a voiced sibilant to a trilled and co-artikiculated rhotic, particulary in unstressed syllables.

If you think about it a bit, weakly articulated r:s are easily realized as something close to z, this is very common in for instance central swedish, and I also think this close to the situation in ON, that the r essentlially sounded the same and eventually merged with z/R in complimentary distribution, starting in weak final syllables ( where there are alot of mispellings like dottiR and æftiR).
[:D] :se-og: :fi-al2: :swe:
[:)] :nor: :usa: :uk:
:wat: :dan: :se-sk2: :eng:
[B)] Image Image :deu:
User avatar
Xonen
moderator
moderator
Posts: 1080
Joined: 16 May 2010 00:25

Re: Early old norse.

Post by Xonen »

HinGambleGoth wrote:I have to agree with Xonen here, i dont see any motivation for a germanic language, that is charachterized for weakening and reduction to strenghten a voiced sibilant to a trilled and co-artikiculated rhotic, particulary in unstressed syllables.
Thing is, sound change is dumb. [:P] It doesn't know if it's moving in the direction of weakening and reduction in general. Perhaps speakers are on some level aware of the kind of "feel" they identify as prestigious in their speech, and this can result in somewhat similar changes accumulating over time - but there's no real reason why some individual sound change couldn't move against the general trend.
If you think about it a bit, weakly articulated r:s are easily realized as something close to z
Well, yes. More scientifically, simple lenition is all it takes to change [z] into [ɹ], which I guess counts as a kind of "weakly articulated r"... And in a language which already has /r/, I suppose [ɹ] could then easily be reanalyzed as an allophone of it. Especially if the [ɹ] goes on to change into something else, like [ɾ]. (My pet hypothesis is that something like this sequence has taken place in most cases of rhotacism - since the simple change from [z] to [r] seems hard to justify phonetically - but of course it's probably impossible to prove in most cases.)

As for why /r/ would become confused with /ɹ/ when preceded by /i/, off the top of my head I'd suggest allophonic palatalization followed by a coalescense into some kind of postalveolar or alveolopalatal approximant or fricative. But I'm starting to stray outside my area of expertise here, so a standard-sized grain of salt is warranted.
Ephraim
sinic
sinic
Posts: 386
Joined: 15 Nov 2013 13:10
Location: Sweden

Re: Early old norse.

Post by Ephraim »

The two main hypotheses in this thread about the contrast between r and ʀ seem to be:
a. trilled r vs approximant ʀ
b. velarized r vs palatalized ʀ

Both solutions are quite plausible, I think.

I've also seen the argument that ʀ could have remained a voiced sibilant [z], contrasting with /s/ only in voicing.1 This I find unlikely for at least two reasons:
1. Proto-Norse went through a phase of final obstruent devoicing. But unlike Gothic, which also had final devocing, NG did not merge final PG *s and *z.
2. Old Norse typically had vocing assimilation in the direction of voicelessness. If s and ʀ only contrasted in voicing, we would have expected common ending –ʀ to devoice after many consonants (and merge with the genitive –s).

If ʀ was a voiced sibilant, but differed from /s/ in place of articulation, this could still work, though.

1 Among other places in this dissertation about rhotacism, which might otherwise be worth checking out:
http://gradworks.umi.com/34/95/3495217.html
HinGambleGoth wrote:I have to agree with Xonen here, i dont see any motivation for a germanic language, that is charachterized for weakening and reduction to strenghten a voiced sibilant to a trilled and co-artikiculated rhotic, particulary in unstressed syllables.
Rhotacization of a voiced sibilant to a trill or a tap is quite a common sound change known from many languages (Latin, South Slavic, Sanskrit, Spanish dialects, maybe West Germanic, probably Turkic). But rhotacization of a siblant directly to a palatalized rhotic is as far as I've been able to tell unattested, so it might make more sense as a two-step process:
1. z ⟶ r (distinct from rˠ)
2. r ⟶ rʲ (pushed to increase the distance from rˠ)

Both sound changes are quite common (although for 2, I think I've mostly seen the reverse: velarization to increase the distance from palatalized consonants). But this only works if /r/ was already [rˠ], of course. In North Germanic specifically, the evidence for this is actually not too strong, I think.

The rhotics in this setup don't necessarily have to have been trills, they point is that the contrast is based on a secondary articulation. I'm using the symbol r here more as a neutral rhotic-symbol. There may have been some variation between trills, taps or even approximants, just like there is a lot of variation today. A palatalized trill is apparently difficult to pronounce (see below).
Xonen wrote:As for why /r/ would become confused with /ɹ/ when preceded by /i/, off the top of my head I'd suggest allophonic palatalization followed by a coalescense into some kind of postalveolar or alveolopalatal approximant or fricative. But I'm starting to stray outside my area of expertise here, so a standard-sized grain of salt is warranted.
The shift r > ʀ after i is one of the main arguments for ʀ being palatalized, I think. But it might be possible to explain the shift if we suppose that r was a trill and ʀ an approximant as well.

I think trills are somewhat difficult to pronounce in palatal environments. From http://mproctor.net/docs/kavitskaya09_FASL.pdf
"We propose that conflicting physical constraints on the tongue dorsum can be held responsible for the sound changes that involve depalatalization of Proto-Slavic palatalized trilled /rʲ/. We show that palatalization, trilling, and different phonological environments impose conflicting demands on the dorsum, resulting in a physical instability that has phonological consequences.
[...]
In a study of alveolar taps and trills in Catalan, Recasens (1991) showed that trills have greater coarticulatory resistance to /i/ than do taps, suggesting an incompatibility between the palatal articulation of /i/ and trilling."


The West Norse ʀ-umlaut, where ʀ has the same effect on preceding vowels as i and j, is perhaps more puzzling if the contrast between ʀ and r was one between an approximant and a trill. There is also iʀ-umlaut, the phenomenon where unsyncopated seem to trigger umlaut while plain unsyncopated i does not.
User avatar
HinGambleGoth
sinic
sinic
Posts: 432
Joined: 01 Jul 2014 05:29
Location: gøtalandum

Re: Early old norse.

Post by HinGambleGoth »

ON names that appear in OE sources are interesting, for instance Oic Ívarr, shows up as <Inwær>.

Hāstæinn = Hǣstēn

GuðþormR = Godrum

Ōttarr = Ōhtere

Swæinn = Swegen

Āsbiarn = Ōsbearn

Frāni = Frǣna

HaraldR = Hareld

BāgsæggR = Bāchsecg
[:D] :se-og: :fi-al2: :swe:
[:)] :nor: :usa: :uk:
:wat: :dan: :se-sk2: :eng:
[B)] Image Image :deu:
Post Reply