Restoration

[ObsequiousNewt]

I recently found myself wondering: what language has restored the most features of its ancestor?

For the purposes of this question I've left the exact definition vague, but in general I'm looking for e.g.: a phoneme, element of morphosyntax, or particular construction that was lost and then regained; a sound change that was reverted (either in whole or in part); etc. I'm less interested in words that were restored by analogy, or words borrowed from the parent language, unless these are particularly interesting/relevant to the question.

For example, here are some things that English has done:

* Lost and then restored five out of six (late) PIE diphthongs: /ei/, /ou/ (lost in PGmc, restored by the GVS); /ai/, /au/ (lost in OE, restored by the GVS), /oi/ (lost in PGmc, restored in OE by palatalization)
* Lost the distinction between /a/ and /o/, then restored it.
* Lost /æ/ in Middle English, then restored it.
* Lost /z/ in OE, then restored it in ME.
* Raised /a/ to /æ/ in Proto-OE, then restored it before a back vowel.
* Raised /eu/ to /iu/ in PGmc, then lowered it in OE (io > eo).

There are probably others which I can't remember off the top of my head.

[Travis B.]

OHG onset f > MHG v > NHG f

[zompist]

More for English:

* (with the rest of Germanic) Lost separate IE perfect & aorist; redeveloped the perfect using an auxiliary
* lost singular 'thou'; many dialects have redeveloped a s/pl distinction

[Pole, the]

Polish:
· lost /kʲ ɡʲ/ (PIE *ḱ *ǵ → PBSl *ś *ź), then regained them (from PSl *k *g before front vowels)
· lost /ɕ ʑ/ (PBSl *ś *ź → PSl *s *z), then regained them (from PSl *s *z before front vowels)
· lost /ʂ/ (from the ruki law; → PSl *x), then regained it (from PSl *š)
· lost the distinction between /a/ and /o/ (late PIE), then regained it (PSl *a ← *ā *ō; PSl. *o ← *a *o)
· lost the vowel length (PSl), then regained it (Old Polish), then lost it again (Modern Polish)
· merged the two nasal vowels (PSl. *ǫ *ę → /ã/), then regained the distinction (/ã ãː/ → /ɛ̃ ɔ̃/)
· palatalized /l/ before front vowels, then depalatalized it
· lost /w/ (PIE *w → PSl *v), then regained it (from PSl *l)
· lost closed syllables (PSl), then regained them
· lost the /ar/ sequence (PSl *ar → *ro), then regained it (*ŭr → ar)

[Zju]

Middle Bulgarian and other south Slavic languages had ʒ > r in certain environments, but in Bulgarian this has been largely reversed due to changes by analogy. Infact the only word with the sound change still in place I can think of is the archaic дорде 'until' < до же де. Other south Slavic languages also had analogical levelings, but to lesser extents.

[Zaarin]

According to Ehret, Proto-Afroasiatic *f > Proto-Semitic *p > Arabic f.

[Porphyrogenitos]

There's the super-long shift of t > θ > ð > d > t from PIE to High German, as in *ph₂tḗr > *fadēr > Vater

[Frislander]

There also the crazy Hawai'ian/Samoan t > k > ' > 0

[Pole, the]

There also the crazy Hawai'ian/Samoan t > k > ' > 0

What is restored here?

[Frislander]

There also the crazy Hawai'ian/Samoan t > k > ' > 0

What is restored here?

Glottal stop is lost, so k debuccalises to k and restores it. Then t backs and restores k.

[KathTheDragon]

Oh, you mean as a chain shift?

[Astraios]

Proto-Semitic *θ *ð became š z in Hebrew, but θ ð were restored in Biblical Hebrew by spirantization of post-vocalic t d.

PS *p *b *t *d *k *g were all replaced in post-vocalic position in Biblical Hebrew by fricatives through the same process, but p b have been restored in Modern Hebrew by borrowings (and occasionally by analogical levelling); t k have been restored by depharyngealization of ṭ q; and t d g have been restored by losing all traces of spirantization.

Earlier w was lost to v in Modern Hebrew, but w is restored in borrowings.

The pharyngeal /ʕ/ is replaced in the standard register by a glottal stop or zero-consonant, but is occasionally restored in colloquial registers when it occurs in despectives (such as /magˈʕil/ “Gross!”, /maˈʕafan/ “Laaame”, and a few others).

[Alon]

Not sure whether it counts, but in American English, prestige Eastern accents dropped non-prevocalic r's under British influence, but later restored them under inland US influence.

In Middle Chinese, there was a palatal series, /tɕ tɕʰ dʑ ɕ ʑ/. It merged into the retroflex series, but then modern Mandarin created new palatal phones out of alveolar and velar phonemes before /i y/.

[Zaarin]

Proto-Semitic *θ *ð became š z in Hebrew, but θ ð were restored in Biblical Hebrew by spirantization of post-vocalic t d.

PS *p *b *t *d *k *g were all replaced in post-vocalic position in Biblical Hebrew by fricatives through the same process, but p b have been restored in Modern Hebrew by borrowings (and occasionally by analogical levelling); t k have been restored by depharyngealization of ṭ q; and t d g have been restored by losing all traces of spirantization.

Earlier w was lost to v in Modern Hebrew, but w is restored in borrowings.

The pharyngeal /ʕ/ is replaced in the standard register by a glottal stop or zero-consonant, but is occasionally restored in colloquial registers when it occurs in despectives (such as /magˈʕil/ “Gross!”, /maˈʕafan/ “Laaame”, and a few others).

Wait, has Modern Hebrew lost all emphatics? I knew it had the ejective-to-pharyngealized shift under influence from Aramaic and Arabic and that it lost its pharyngeal fricatives under influence from European, but I didn't realize it had lost its emphatics altogether. Ashkenazi and Yemeni Jews still pronounce emphatics, don't they?

[Vijay]

FWIR, Ashkenazim are the ones who don't have any of the emphatics.

[Astraios]

Wait, has Modern Hebrew lost all emphatics? I knew it had the ejective-to-pharyngealized shift under influence from Aramaic and Arabic and that it lost its pharyngeal fricatives under influence from European, but I didn't realize it had lost its emphatics altogether. Ashkenazi and Yemeni Jews still pronounce emphatics, don't they?

No, Modern Hebrew has no emphatics at all. All European variants of Hebrew lost emphatics quickly, from Denmark to Spain to Persia (obviously, because they adopted IE languages which have no such creatures). The immigrant generations in Israel who were/are kicked out of Arabic- and Aramaic-speaking lands might use emphatic stops sometimes, if their Modern Hebrew never became very fluent and they're using a Hebrew word which was preserved with an emphatic in their Arabic/Aramaic dialect, and of course the best of the cream of the Yemenite/Moroccan/Iraqi singers do try to use emphatics in traditional singing or ritual language, but even they don't always, and certainly not in ordinary speech. I've heard Ethiopians using ejective stops in MH, but I think that's more due to poor language ability than anything. And even the queens of pharyngealization the Arabs themselves don't bother to pharyngealize stops in MH. So no, the Hebrew that is being passed on to the next generation has no emphatic stops, and only has post-uvular consonants at all as a stylistic phenomenon.


EDIT: One has to be careful of is the distinction between Biblical Hebrew (spoken ~2500 years ago), Modern Hebrew (spoken today as the native language of Israelis), and "Liturgical Hebrew" (the written language of the Bible and other scriptures, as pronounced by modern religious communities of different linguistic backgrounds). Liturgical Hebrew certainly still has emphatics for communities with Semitic-speaking backgrounds.

[Zaarin]

Wait, has Modern Hebrew lost all emphatics? I knew it had the ejective-to-pharyngealized shift under influence from Aramaic and Arabic and that it lost its pharyngeal fricatives under influence from European, but I didn't realize it had lost its emphatics altogether. Ashkenazi and Yemeni Jews still pronounce emphatics, don't they?

No, Modern Hebrew has no emphatics at all. All European variants of Hebrew lost emphatics quickly, from Denmark to Spain to Persia (obviously, because they adopted IE languages which have no such creatures). The immigrant generations in Israel who were/are kicked out of Arabic- and Aramaic-speaking lands might use emphatic stops sometimes, if their Modern Hebrew never became very fluent and they're using a Hebrew word which was preserved with an emphatic in their Arabic/Aramaic dialect, and of course the best of the cream of the Yemenite/Moroccan/Iraqi singers do try to use emphatics in traditional singing or ritual language, but even they don't always, and certainly not in ordinary speech. I've heard Ethiopians using ejective stops in MH, but I think that's more due to poor language ability than anything. And even the queens of pharyngealization the Arabs themselves don't bother to pharyngealize stops in MH. So no, the Hebrew that is being passed on to the next generation has no emphatic stops, and only has post-uvular consonants at all as a stylistic phenomenon.


EDIT: One has to be careful of is the distinction between Biblical Hebrew (spoken ~2500 years ago), Modern Hebrew (spoken today as the native language of Israelis), and "Liturgical Hebrew" (the written language of the Bible and other scriptures, as pronounced by modern religious communities of different linguistic backgrounds). Liturgical Hebrew certainly still has emphatics for communities with Semitic-speaking backgrounds.

Thanks for summarizing that; I was unaware that Modern Hebrew had entirely lost its emphatics even among Asiatic Jews (my chief--actually, only--exposure to Modern Hebrew is Ofra Haza). My interest in Semitic languages has chiefly centered around Biblical Hebrew, Akkadian, and Syriac (the writing system more than the language itself), so I'm not very familiar with the modern vernaculars.

[Yng]

Wait, has Modern Hebrew lost all emphatics? I knew it had the ejective-to-pharyngealized shift under influence from Aramaic and Arabic and that it lost its pharyngeal fricatives under influence from European, but I didn't realize it had lost its emphatics altogether. Ashkenazi and Yemeni Jews still pronounce emphatics, don't they?

No, Modern Hebrew has no emphatics at all. All European variants of Hebrew lost emphatics quickly, from Denmark to Spain to Persia (obviously, because they adopted IE languages which have no such creatures). The immigrant generations in Israel who were/are kicked out of Arabic- and Aramaic-speaking lands might use emphatic stops sometimes, if their Modern Hebrew never became very fluent and they're using a Hebrew word which was preserved with an emphatic in their Arabic/Aramaic dialect, and of course the best of the cream of the Yemenite/Moroccan/Iraqi singers do try to use emphatics in traditional singing or ritual language, but even they don't always, and certainly not in ordinary speech. I've heard Ethiopians using ejective stops in MH, but I think that's more due to poor language ability than anything. And even the queens of pharyngealization the Arabs themselves don't bother to pharyngealize stops in MH. So no, the Hebrew that is being passed on to the next generation has no emphatic stops, and only has post-uvular consonants at all as a stylistic phenomenon.


EDIT: One has to be careful of is the distinction between Biblical Hebrew (spoken ~2500 years ago), Modern Hebrew (spoken today as the native language of Israelis), and "Liturgical Hebrew" (the written language of the Bible and other scriptures, as pronounced by modern religious communities of different linguistic backgrounds). Liturgical Hebrew certainly still has emphatics for communities with Semitic-speaking backgrounds.

I know I've read in a few places that there is a distinct (low-prestige) and more classically Semitic-sounding Sephardi accent - is this a thing which is now disappearing, or is it characterised by other features? I wasn't under the impression it had emphatics (though I guess some would make sense if people knew the liturgical pronunciation of liturgical Hebrew before they learnt Modern Hebrew and carried features over), but I swear it had some Semitic-y features like perhaps having 3ayn or something... Can you enlighten further??

Incidentally emphatic stops in the Arabic dialects of the Levant at least are generally analysed as velarised and not pharyngeal.

Edit: I'd guess ma3afan is an Arabic borrowing or influenced by the Arabic word (< m3affen 'rotten') which probably explains the 3ayn there.

[Astraios]

my chief--actually, only--exposure to Modern Hebrew is Ofra Haza

If you like her, I could recommend also Gila Bashari, for example.

I know I've read in a few places that there is a distinct (low-prestige) and more classically Semitic-sounding Sephardi accent - is this a thing which is now disappearing, or is it characterised by other features? I wasn't under the impression it had emphatics (though I guess some would make sense if people knew the liturgical pronunciation of liturgical Hebrew before they learnt Modern Hebrew and carried features over), but I swear it had some Semitic-y features like perhaps having 3ayn or something... Can you enlighten further??

Oy. I have so many things to say, but this is getting out of hand, so I’m making this as concise as I can.

Yes, the ‘Mizrahi’( not Sephardi)-accented variant of Modern Hebrew is low-prestige, is disappearing from normal speech registers, and doesn’t have emphatics, unless perhaps, as I noted above and you say, the speaker has such poor Hebrew that they prefer their own native language’s pronunciation of a particular Hebrew word which had been preserved in their native language (Judeo-Arabic, Jewish Neo-Aramaic) with an emphatic.

The Sephardi variant of liturgical Hebrew is kind of a non-entity, as ‘Sephardic liturgy’ covers both sides of the Mediterranean from Portugal to Persia and has influences from tens of native languages. But yes, you are right, the liturgical variant which was influenced by Judeo-Spanish does not have emphatics, though it does preserve both /ʕ/ and /ħ/, and also has a flapped [ɾ] for MH /ʁ/. This was also the variant chosen for revival as the national language in the 1880s, but due to complex societal reasons (in short, association with refugees from the Middle East), it has lost some of its prestige, such that it’s now low-prestige in ordinary speech registers (it’s mostly spoken by older generations). However, it is still considered “more beautiful” than standard Israeli MH, and is still the only appropriate variant for certain genres of music, for example, where it is very much alive. Extremely unhelpfully, this variant of Israeli MH is today called a ‘Mizrahi’ accent, and its relation to Judeo-Spanish is forgotten.

The actual Mizrahi (Arabic- and Aramaic-influenced) variants of liturgical Hebrew may still have emphatics, at least for those very few Jews who still speak better Arabic/Aramaic than Modern Hebrew, but the vast majority of Mizrahim don’t have emphatics even in liturgy, since the vast majority of them were born in Israel and speak the ‘Mizrahi’ variant of MH described in the preceding paragraphs, not any variant related to their own liturgical tradition. In addition, consider this song in the Yemenite liturgical variant, by a native Arabic-speaker; his emphatics are very much more purely pharyngeal than what you find in Arabic, and don’t much affect the vowels, which is part of a general tendency in liturgical Hebrew (having many more distinct vowels than Arabic), and probably contributed to their current loss. The Mizrahim tend to be very proud of their pharyngeals though, and like to use them on purpose to make a statement about their non-Ashkenaziness, because there is naturally a certain classism against Israelis of poorer, less-educated (Middle Eastern) backgrounds and in favor those of wealthier (European) backgrounds, and equally naturally a great deal of pride among Mizrahim in not being European - cf. this fun song, This Isn’t Europe, Honey, which scolds/mocks gay Israelis for becoming over-Europeanized, while (hilariously) being performed in very standard, non-‘Mizrahi’, Israeli MH* (by a singer born in Yemen).


tl;dr
1. Emphatics are not found in MH, but may exist in Mizrahi liturgical variants (especially Yemenite).
2. If they still exist, they’re harder to spot because they don’t affect vowels.
3. Pharyngeals are found in ‘Mizrahi’-accented MH, which is actually the Sephardic liturgical variant.
4. This ‘Mizrahi’-accented MH is dying with the older generations, but still very alive in traditional music.
5. Standard Israeli MH is identical to the above, but the pharyngeals are lost due to IE-language influence and classism.


*Incidentally, and more on-topic, it contains the slang word wej “facepic” (< Arabic wajh “face”), which demonstrates MH’s restored /w/.

Edit: I'd guess ma3afan is an Arabic borrowing or influenced by the Arabic word (< m3affen 'rotten') which probably explains the 3ayn there.

Neat, I didn’t know the source. That also explains its un-Hebrew maCaCaC shape.


ETA: An example of ‘Mizrahi’-accented MH. Note the pharyngeals, but everything else is standard Israeli (uvular rhotic, deletion of glottals, voicing assimilation in consonant clusters) - shows how far the pharyngeals still are from actual death.

[Zaarin]

my chief--actually, only--exposure to Modern Hebrew is Ofra Haza

If you like her, I could recommend also Gila Bashari, for example.

I'll check her out.

[Yng]

Thanks astraios, interesting stuff!

[Zaarin]

Thanks astraios, interesting stuff!

Indeed.

On the topic of Semitic, how do Semitic languages pronounce the imperative of resonant-initial roots like LQṬ? Epenthetic schwa, i.e. lqaṭ [ləqatˁ]?

[Yng]

I don't really understand your question. Different Semitic languages form the imperative differently. In Classical Arabic it's formed from the bare root with a vocalic prefix to prevent impossible consonant clusters. So the imperative would be ilqaT. Some colloquials have lqaT or lqaaT instead.

[Zaarin]

I don't really understand your question. Different Semitic languages form the imperative differently. In Classical Arabic it's formed from the bare root with a vocalic prefix to prevent impossible consonant clusters. So the imperative would be ilqaT. Some colloquials have lqaT or lqaaT instead.

I'm reading Lipinski (who admittedly has some non-standard views on Semitic, including a Moroccan homeland for Semitic that I've heard nowhere else and that PS emphatics were pharyngealized rather than ejective), and he says the basic form of the Semitic imperative is CCvC. According to Lipinski, the prefix-conjugation of the imperative forms the jussive, which is used as the basis for optatives, volitives, vetitives, etc. So do all Semitic dialects have the vocalic prefix to avoid illegal clusters? If not, how do those other dialects pronounce imperatives with illegal clusters?

[Richard W]

So do all Semitic dialects have the vocalic prefix to avoid illegal clusters? If not, how do those other dialects pronounce imperatives with illegal clusters?

Underlying BCvD works quite well. In Biblical Hebrew, schwa is generally inserted between B and C, but if B is a weak consonant, it may be dropped instead.

The regular Akkadian imperative (m.s.) is <BuCuD>, but I suppose that could hide a phonetic [BCuD].