Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Do the geminate spellings in present day English indicate English had the corresponding geminate sounds in the past?

193 views
Skip to first unread message

S K

unread,
Sep 10, 2022, 6:07:35 PM9/10/22
to
not talking about artifical geminates like "night-train".

Also excluding indian English which has them profusely.

Dingbat

unread,
Sep 10, 2022, 9:43:21 PM9/10/22
to
On Saturday, September 10, 2022 at 3:07:35 PM UTC-7, S K wrote:
> not talking about artifical geminates like "night-train".
>
> Also excluding indian English which has them profusely.

I'm reminded that SUNNY and PULLEY have
geminates in Malayali English. But Indian
Englishes don't always geminate a doubled
consonant; they have no geminate in APPLE.
Old English had [p:] in æppel, pronouncing it
like app-pell.
https://en.wikipedia.org/wiki/West_Germanic_gemination

As per that page, a sound shift produced geminates in Old
English lost later to to further sound shifts. But the word for
APPLE doesn't seem to have the right context for the sound
shift described on the page. Could this mean that English
had a few geminates even before the sound shift?

1) Modern English needs the <pp> in order to retain the
[æ]; dropping one <p> to spell it <apel> would change
the initial vowel to [eI].
2) Degeminating the p also changed the ending from
[El] in Old English to syllabic l.in modern English.

Dingbat

unread,
Sep 10, 2022, 9:55:45 PM9/10/22
to
On Saturday, September 10, 2022 at 6:43:21 PM UTC-7, Dingbat wrote:
> On Saturday, September 10, 2022 at 3:07:35 PM UTC-7, S K wrote:
> > not talking about artifical geminates like "night-train".
> >
> > Also excluding indian English which has them profusely.
> I'm reminded that SUNNY and PULLEY have
> geminates in Malayali English. But Indian
> Englishes don't always geminate a doubled
> consonant; they have no geminate in APPLE.
> Old English had [p:] in æppel, pronouncing it
> like app-pell.
> https://en.wikipedia.org/wiki/West_Germanic_gemination
>
> As per that page, a sound shift produced geminates in Old
> English lost later to to further sound shifts. But the word for
> APPLE doesn't seem to have the right context for the sound
> shift described on the page. Could this mean that English
> had a few geminates even before the sound shift?
>
Oops; I hadn't read the whole page. Halfway down the page,
is verbiage that suggests that the geminate in æppel was
triggered by the l.

Christian Weisgerber

unread,
Sep 11, 2022, 11:30:06 AM9/11/22
to
On 2022-09-10, S K <skpf...@gmail.com> wrote:

> Subject: Do the geminate spellings in present day English indicate
> English had the corresponding geminate sounds in the past?

Only some instances of double consonant spellings in Modern English
can be traced back to geminates in Old English, e.g. in "will".

Old English had phonemic geminate consonants. IIRC, the vowel
before those was always short. When the geminates were lost and
short vowels in open syllables lengthened, the double consonant
spelling became associated with short vowels and has been inconsistently
used to mark them, even when there was no etymological geminate,
as for instance in the morphological spelling rule "trap - trapped".

(More or less the same happened in German, where the principle was
applied more consistently across the orthography.)

There were also Old English geminates whose modern reflexes are not
spelled with a doubled consonant, at least -dg- as in "bridge",
"hedge".

Latin had phonemic geminates, but I think double consonant spellings
in English loans from Latin have always been purely orthographic.

Western Romance lost the Latin geminates early on, so double
consonants in French words--including those borrowed into English--are
purely in imitation of Latin spelling.

--
Christian "naddy" Weisgerber na...@mips.inka.de

S K

unread,
Sep 11, 2022, 5:45:46 PM9/11/22
to
On Saturday, September 10, 2022 at 9:43:21 PM UTC-4, Dingbat wrote:
> On Saturday, September 10, 2022 at 3:07:35 PM UTC-7, S K wrote:
> > not talking about artifical geminates like "night-train".
> >
> > Also excluding indian English which has them profusely.
> I'm reminded that SUNNY and PULLEY have
> geminates in Malayali English.

Way before I knew the remotest thing discussed here I heard an Indian say "Cutty Sark" and though "something is off with that"

When Raj Brits heard "pukka Sahib" guess they didn't hear the gemination in Hindi.

How times have changed - Goras connected with Chess these days pronounce Rameshbabu Praggnanandhaa more or less close to the original.

Wonder if Standard Indian English has rules for gemination.

Ruud Harmsen

unread,
Sep 12, 2022, 3:05:30 AM9/12/22
to
Sun, 11 Sep 2022 14:45:45 -0700 (PDT): S K <skpf...@gmail.com>
scribeva:
>Way before I knew the remotest thing discussed here I heard an Indian say "Cutty Sark" and though "something is off with that"
>
>When Raj Brits heard "pukka Sahib" guess they didn't hear the gemination in Hindi.
>
>How times have changed - Goras connected with Chess these days pronounce Rameshbabu Praggnanandhaa more or less close to the original.
>
>Wonder if Standard Indian English has rules for gemination.

If you want to hear real and consistent gemination, listen to
Hungarian.

wugi

unread,
Sep 12, 2022, 6:01:01 AM9/12/22
to
Op 12/09/2022 om 9:05 schreef Ruud Harmsen:
Italian will do.

--
guido wugi

Ruud Harmsen

unread,
Sep 12, 2022, 6:46:19 AM9/12/22
to
Mon, 12 Sep 2022 12:00:47 +0200: wugi <wu...@scrlt.com> scribeva:
Yes, but never at the end of a word.

All Hungarian comparatives and superlatives end in -bb, e.g. szép =
beautiful, szebb = more beautiful, legszebb = most beautiful. And that
doubleness, or in fact length, is clearly audible.

Itt = here, ott = there, also quite noticeable, even to my non-native
ears.
--
Ruud Harmsen, http://rudhar.com

Christian Weisgerber

unread,
Sep 12, 2022, 9:30:06 AM9/12/22
to
On 2022-09-12, Ruud Harmsen <r...@rudhar.com> wrote:

>>> If you want to hear real and consistent gemination, listen to
>>> Hungarian.
>>
>>Italian will do.
>
> Yes, but never at the end of a word.

If we stay in Italy, Neapolitan has surface final gemination. I
understand that phonemically there are final schwas, but they are
frequently silent. The TV show _Gomorra_ (widely available on Sky
or Netflix in some countries) is a great showcase for Neapolitan.

Tim Lang

unread,
Sep 12, 2022, 10:10:59 AM9/12/22
to
On 12.09.2022 12:46, Ruud Harmsen wrote:

>>>If you want to hear real and consistent gemination, listen to
>>>Hungarian.
>>
>>Italian will do.
>
>Yes, but never at the end of a word.
>
>All Hungarian comparatives and superlatives end in -bb, e.g. szép =
>beautiful, szebb = more beautiful, legszebb = most beautiful. And that
>doubleness, or in fact length, is clearly audible.

Yet not as clearly as in Italian words. In Hungarian they are as
clearly audible only when spoken with exaggerated emphasis (esp.
in acting or in reading certain types of poetry). Which is to
be seen in the spelling: not every well educated Hungarian knows
all correct spelling cases.

>Itt = here, ott = there, also quite noticeable, even to my non-native
>ears.

Yes, but the Italian ones (in several dialects as well as standard
Italian) non-natives really hear them. IMHO -tt- in attenzione! and
aspetta! have more "intensity" & "stress" than Hungarian one in itt,
ott. Though, bit intenser are the ones followed by a vowel: e.g. látta
"(s/he) saw", ette "(s/he) ate", i.e. stronger than -tt- in látott
"seen" and evett "eaten"; and in rettenetes "terrible". All this
can be heared by non-natives when hearing Hungarian spoken by common
native-speakers in colloquial Hungarian. Thus the Hungarian ones
are a bit "weaker" in usual/common speech. Very intense are rather
the Hungarian long vowels - so much so that in colloquial and
dialectal speech some long vowels sound rather like diphtongs, e.g.
long ó => almost /ou/ or even /au/. Another example: in állat "animal",
the -ll- is pronounced by most people without the appropriate geminate
emphasis; therefore, one has to learn the spelling rule by heart. But,
OTOH, that initial á (long a) is "exaggerated" by most native
speakers that such a long /a/ (uttered with a maximal opening of the
jaw; other peoples would get ... joint problems in the aftermath
thereof! :-)). It'll sound longer/intenser than the equivalent vowel
in English, Dutch, German, Italian, French. (NB: in Hungarian every
long a is long; there are no short a'. OTOH, a in written is to be
pronounced: /ɔ; /ɒ/. No exception.)

Tim

Ruud Harmsen

unread,
Sep 12, 2022, 2:14:29 PM9/12/22
to
Mon, 12 Sep 2022 16:10:51 +0200: Tim Lang <m...@privacy.net> scribeva:

>On 12.09.2022 12:46, Ruud Harmsen wrote:
>
>>>>If you want to hear real and consistent gemination, listen to
>>>>Hungarian.
>>>
>>>Italian will do.
>>
>>Yes, but never at the end of a word.
>>
>>All Hungarian comparatives and superlatives end in -bb, e.g. szép =
>>beautiful, szebb = more beautiful, legszebb = most beautiful. And that
>>doubleness, or in fact length, is clearly audible.
>
>Yet not as clearly as in Italian words. In Hungarian they are as
>clearly audible only when spoken with exaggerated emphasis (esp.
>in acting or in reading certain types of poetry). Which is to
>be seen in the spelling: not every well educated Hungarian knows
>all correct spelling cases.
>
>>Itt = here, ott = there, also quite noticeable, even to my non-native
>>ears.
>
>Yes, but the Italian ones (in several dialects as well as standard
>Italian) non-natives really hear them. IMHO -tt- in attenzione! and
>aspetta! have more "intensity" & "stress" than Hungarian one in itt,
>ott.

The Hungarian ones I think are a matter of length: the closure and
release are wider apart in time, in other words, the pressure build-up
phase lasts longer. That is, in case of stop consonants. <ssz> and
<nn> etc. simply last longer. Simply said, in Hungarian all phonemes
can be long and short, not only the vowels, also all consonants. Cf.
Finnish.

>Though, bit intenser are the ones followed by a vowel: e.g. látta
>"(s/he) saw", ette "(s/he) ate", i.e. stronger than -tt- in látott
>"seen" and evett "eaten"; and in rettenetes "terrible". All this
>can be heared by non-natives when hearing Hungarian spoken by common
>native-speakers in colloquial Hungarian. Thus the Hungarian ones
>are a bit "weaker" in usual/common speech. Very intense are rather
>the Hungarian long vowels - so much so that in colloquial and
>dialectal speech some long vowels sound rather like diphtongs, e.g.
>long ó => almost /ou/ or even /au/.

I think I've heard that, yes, in old songs.

Long vowels aren't always that long: I find <u> and <ú>, <i> and <í>
sometimes hard to distinguish.

>Another example: in állat "animal",
>the -ll- is pronounced by most people without the appropriate geminate
>emphasis; therefore, one has to learn the spelling rule by heart. But,
>OTOH, that initial á (long a) is "exaggerated" by most native
>speakers that such a long /a/ (uttered with a maximal opening of the
>jaw; other peoples would get ... joint problems in the aftermath
>thereof! :-)). It'll sound longer/intenser than the equivalent vowel
>in English, Dutch, German, Italian, French. (NB: in Hungarian every
>long a is long; there are no short a'. OTOH, a in written is to be
>pronounced: /?; /?/. No exception.)

Phonologically, I still see <a> and <á>, and <e> and <é> as pairs,
even though they are phonetically quite different. This can be see in
the short final -e and -a, which become long when a suffix follows.
E.g. nominative singular Béla, accusative singular Bélát.

wugi

unread,
Sep 12, 2022, 3:30:32 PM9/12/22
to
Op 12/09/2022 om 20:14 schreef Ruud Harmsen:

> Phonologically, I still see <a> and <á>, and <e> and <é> as pairs,
> even though they are phonetically quite different. This can be see in
> the short final -e and -a, which become long when a suffix follows.
> E.g. nominative singular Béla, accusative singular Bélát.

I find them altogether not that different from Dutch pairs "a, aa" and
"e, ee". Eg Dutch "a" and "aa" will sound a little less rounded viz.
more open than their Hungarian equivalents. In careful Dutch speech I
mean, not in our casual Flemish of course.

--
guido wugi

Tim Lang

unread,
Sep 12, 2022, 4:29:34 PM9/12/22
to
On 12.09.2022 20:14, Ruud Harmsen wrote:

>The Hungarian ones I think are a matter of length:

That's right.

[snip]

>That is, in case of stop consonants. <ssz> and <nn> etc. simply last
>longer.

In theory as well as in emphatic, correct pronunciation. But in
normal colloquial (rapid!) speech the difference between <ssz> and
sz as well as <n> and <nn> is difficult to hear. (E.g. very frequent
in building the subjunctive and imperative.)

>Simply said, in Hungarian all phonemes can be long and short,
not only the vowels, also all consonants.

As in ... Italian!

>I think I've heard that, yes, in old songs.

And it's an every day's occurrence in some regional Hungarian varieties.

> Long vowels aren't always that long: I find <u> and <ú>, <i> and <í>
> sometimes hard to distinguish.

Of course, since these vowels don't have that "extreme" position of the
jas as especially in uttering -e- /æ/, -é- /e:/, -á- /a:/ and even -ó-
/ɔː/ and -ő- /ø:/.

>Phonologically, I still see <a> and <á>, and <e> and <é> as pairs,
>even though they are phonetically quite different. This can be see in
>the short final -e and -a, which become long when a suffix follows.
>E.g. nominative singular Béla, accusative singular Bélát.

Of course. Moreover, -e-, -é-, -í- are in a relationship with -ö- and
-ő- as well, esp. in dialects. In the standard language note fel
("up(wards)" as well as "up"/"high"/"super" as a prefix): this
/fæl/ coexist with /föl/ and even the long -ö- /ő/. In some dialectal
and archaic cases these are rendered by /ü/ and long ü /ű/, e.g.
fésű /fe:ʃü:/ "comb" - fűsű /fü:ʃü:/. (BTW: the Hungarian "comb"
word has an Ossetian-Sarmatian origin; i.e., from a small group of
words of this Scythian-Persian origin but which is of utmost importance
and frequency in the main vocabulary; even the Hung. words for "God,
devil, (some) spirits" are old Scythian-Sarmatian and Persian loanies.)

These are phonetical difficulties for all neighbors (except the German-
speaking ones): Slovakians, Croats/Serbians, Romanians, Ukrainians,
since in their languages there is almost no longer vowels and
consonants. Which then causes some peculiar, "foreign", pronunciations
whenever they speak Hungarian. A similar phenomenon is there in Italy
for Romanians, even if they otherwise speak a very good standard
Italian: they virtually cannot utter longer vowels and esp. consonants
in Italian.

Tim

Ruud Harmsen

unread,
Sep 12, 2022, 9:08:07 PM9/12/22
to
Mon, 12 Sep 2022 21:30:26 +0200: wugi <wu...@scrlt.com> scribeva:

>Op 12/09/2022 om 20:14 schreef Ruud Harmsen:
>
>> Phonologically, I still see <a> and <á>, and <e> and <é> as pairs,
>> even though they are phonetically quite different. This can be see in
>> the short final -e and -a, which become long when a suffix follows.
>> E.g. nominative singular Béla, accusative singular Bélát.
>
>I find them altogether not that different from Dutch pairs "a, aa" and
>"e, ee". Eg Dutch "a" and "aa" will sound a little less rounded viz.
>more open than their Hungarian equivalents.

Dutch short a isn't rounded, the Hungarian one always is. In both
languages, the long ones are unrounded.

It may be that the Hungarian short a is similar to a dialectal one in
the area where I grew up. But I'm not sure if it's the same.

Anyway, that doesn't make Hungarian easier to pronounce, because the
short <a> is quite frequent, and frequently occurs in positions, like
final, where in Dutch it can never occur. (With the exception of
"bah".)

Tim Lang

unread,
Sep 13, 2022, 3:42:49 AM9/13/22
to
On 13.09.2022 03:08, Ruud Harmsen wrote:

>It may be that the Hungarian short a is similar to a dialectal one in
>the area where I grew up. But I'm not sure if it's the same.

The "short" <a> in Hungarian is pronounced /ɔ/. Quite as in British
English in "hot, not, lot, pot" or in Bavarian and Austrian German
in "da hat er gesagt" /dɔ hɔddə ksɔkt/. (Kind of a very "open" /o/.)
(Caution: the comparison is a bit weak, since there are lots of
Austrians and Bavarians who don't pronounce /ɔ/ but rather
/do hoddə/ or /hodda ksokt/, i.e. with real /o/, sounding almost
as /o/ in Oostende.)

>Anyway, that doesn't make Hungarian easier to pronounce, because the
>short <a>

The term "short" is inappropriate, since the sound for the letter
<a> in Hungarian (whenever without the accent on it) is not an /a/,
but is a vowel closer to an /o/. It is perceived as a vowel on
its own, without dependence on /a/ and /o/. Illustrated by the
fact that in Hungarian spelling, it has its own letter: <a>.
(As the Danish and Swedish convention <å>!)

The other vowel, written <á> is very different: a very "intense"
and long /a/ of the best kind - i.e. never pronounced in
variations, i.e. unlike in German where for the same <a> the
native-speakers has the "libertarian" freedom to pronounce it
as they like (with varieties from /o/ to the most "shrill" types
of /ʌ/ or the Northern/Hamburg kind of /æ/).

Thus, in Hungarian there is an awareness that these are
different/separate vowels. Contrasting with other European
languages esp. German, English and Dutch, where there is no
"fix standard" to "rivet" the relationship 'certain letter
for a certain sound', i.e. no "multi-usability".

BTW, the Hungarian spelling is almost kind of a ... phonetic
transcription. Almost like the received API/IPA phonetic
transcription for English. Something which in many European
languages doesn't exist (hence the difficulties in school
and concerning certain types of ... illiteracy). In these languages
there isn't a standard pronunciation /ɔ/ by everybody in the
standard received pronunciation. Unlike the case in standard
Hungarian in whichever province in Hungary and in the lost
territories: the standard language is always the same, the
pronunciation of <a> and of <á> being constant (ie, having
no variations/oscillations).

Something which for example in the German-speaking area is hard
to obtain at least in the received pronunciation
by professional actors (e.g. "Bühnendeutsch": "actors' German"),
especially whenever their initial regional dialect goes on exerting
a too strong influence on their wanna-be standard "Bühnendeutsch"
pronunciation.

Moreover, native-speakers of some languages pronounce <o> like
the Hungarian <a>. Example: e.g. standard German speakers (radio &
TV) pronounce president Macron's name always /makrɔ/ and not
in the French way, with the French /o/, because they aren't
aware that they apply the German phon. custom of the /ɔ/
usage esp. when followed by consonants such as <r>, <n>, <t>.
Something that in other languages cannot be applied.

>is quite frequent, and frequently occurs in positions, like
>final, where in Dutch it can never occur. (With the exception of
>"bah".)

This is a good example: it reminds me Ottawa, in which the last a
is pronounced at least by some Canadians almost as /ɔ/ as in
Hungarian <a>: /'ʌ-tə-wɔ, 'ʌ-tə-wɒ/ (where /t/ rather sounds
like ... /d/ :)), and not only /'ɔ-tə-wə/ as taught in BE classes
in Europe.

Hungarian consonants aren't "behaucht"/aspired (unlike those in Germanic
languages) at all, but rather as their equivalents in Slavic and
Romance languages and Turkish (except ending /r/). So, I assume
this might be a difficulty for Dutch, Germans, Danes, Brits and
Americans when learning Hungarian. The Hungarian /r/ is "gerollt"
(apical), but the uvula /r/ is heard quite frequently.

A strange aspect: in some regional Hungarian /t/ and /d/ are
pronounced as in English: alveolar not dental. But this is a
mere exception, the vast majority have the dental variants only.

Tim

Christian Weisgerber

unread,
Sep 13, 2022, 8:30:06 AM9/13/22
to
On 2022-09-13, Tim Lang <m...@privacy.net> wrote:

> This is a good example: it reminds me Ottawa, in which the last a
> is pronounced at least by some Canadians almost as /ɔ/ as in
> Hungarian <a>: /'ʌ-tə-wɔ, 'ʌ-tə-wɒ/ (where /t/ rather sounds
> like ... /d/ :)), and not only /'ɔ-tə-wə/ as taught in BE classes
> in Europe.

A more salient Canadian comparison is Quebec French, where <â> /ɑ/
is realized as rounded [ɒ] or even [ɔ]. Historically, that was a
long vowel /ɑː/, though.

wugi

unread,
Sep 13, 2022, 10:20:10 AM9/13/22
to
Op 13/09/2022 om 14:02 schreef Christian Weisgerber:
C'est pɔ: vrai!

--
guido wugi

Ruud Harmsen

unread,
Sep 13, 2022, 10:22:30 AM9/13/22
to
Tue, 13 Sep 2022 09:42:43 +0200: Tim Lang <m...@privacy.net> scribeva:

>On 13.09.2022 03:08, Ruud Harmsen wrote:
>
>>It may be that the Hungarian short a is similar to a dialectal one in
>>the area where I grew up. But I'm not sure if it's the same.
>
>The "short" <a> in Hungarian is pronounced /?/. Quite as in British
>English in "hot, not, lot, pot" or in Bavarian and Austrian German
>in "da hat er gesagt" /d? h?dd? ks?kt/. (Kind of a very "open" /o/.)
>(Caution: the comparison is a bit weak, since there are lots of
>Austrians and Bavarians who don't pronounce /?/ but rather
>/do hodd?/ or /hodda ksokt/, i.e. with real /o/, sounding almost
>as /o/ in Oostende.)
>
>>Anyway, that doesn't make Hungarian easier to pronounce, because the
>>short <a>
>
>The term "short" is inappropriate, since the sound for the letter
><a> in Hungarian (whenever without the accent on it) is not an /a/,
>but is a vowel closer to an /o/.

Phonologically, historically, and morphologically, it is short a. The
actual phonetic value, somewhat shifted, probably for extra contrast,
does not change that. The spelling is appropriate.

<a> is shorter, backer, higher and rounder than <á>.
<e> is shorter and lower than <é>.

So be it, and so it is.

>but is a vowel closer to an /o/.

<o> and <ó> are already /o/ and /o:/.

>BTW, the Hungarian spelling is almost kind of a ... phonetic
>transcription.

Phonemic.
It correctly doesn't show assimilations.

>Hungarian consonants aren't "behaucht"/aspired (unlike those in Germanic
>languages) at all, but rather as their equivalents in Slavic and
>Romance languages and Turkish (except ending /r/). So, I assume
>this might be a difficulty for Dutch, Germans, Danes, Brits and
>Americans when learning Hungarian.

Dutch doesn't have aspiration in /t/, /p/, /k/. That makes English
difficult, but not Hungarian.

>The Hungarian /r/ is "gerollt" (apical), but the uvula /r/ is heard quite frequently.

So rarely that I can't remember ever having heard it. I happened to
hear it in Russian yesterday, in a fragment of a fascist TV discussion
over Ukranian military advances.

Tim Lang

unread,
Sep 13, 2022, 3:09:08 PM9/13/22
to
On 13.09.2022 16:22, Ruud Harmsen wrote:

>Phonologically, historically, and morphologically, it is short a. The >actual phonetic value, somewhat shifted, probably for extra contrast,
>does not change that. The spelling is appropriate.

The spelling is as it is only because Hungarians once chose the <a>
rendition. As the Danes adopted the <å>.

><a> is shorter, backer, higher and rounder than <á>.

Yes, but in the native Hungarian perception and pronunciation the
<a> is never an /a/ or /ʌ/, it is always, i.e., without any
exception whatsoever, an /ɔ; ɒ/. So, in the strict sense of
the wording "short a" the usage of it is ... wrong: because the
Hungarian vowel simply written <a> is no /a/ at all. It might
be perceived as an /a/ in other languages, incl. in German and
Dutch.

This is why, e.g. words such as "Hans, hat, Tanne, Katze" & the like in
_standard_ German (genuine Hochdeutsch), cannot be read with /a, ʌ/
by a Hungarian unless s/he is _taught_ the correct pronunciation
by a teacher who's aware of these differences and nuances. Otherwise,
any Hungarian will automatically read <a> as /ɔ/ in any foreign
spelling. Which happens zillion times in Hungary and in its
former provinces whenever the German names of Hungarians ending
in -mann (Neumann, Baumann, Kaufmann) are pronunced ... /mɔn/.
Only trained people learn that -mann has to be pronounced /mʌn/.
I once heard one native-speaker reading the English name spelling
Alack ... /'ɔlɔtsk/, because that's the way one reads it in
Hungarian; and this illustrates how they automatically see in
every <a> their /ɔ/, unless they learn, say, English.

Whereas words such as "Maat, Lahn, Wahn" & the like are easy for
them, since Hungarian has the wovel <á> /ɑː/, which is a genuine
/a/ and it is long at that, as it is in the standard German
pronunciation. But as soon as the same /a/ gets short, the
Hungarian mind doesn't grasp it (unless the person has some
goog knowledge of a foreign language that makes the distinction
short /ɔ/ versus short /ʌ/. Esp. Hung. natives living in former
territories of Hungary, now in Slovakia, Ukraine, Romania, Serbia
can much better be aware of the phenomenon than Hungarians living
only in Hungary. Since those living outside are usually in
perfect command of Slovak, Ruthenian/Russian, Rumanian, Serbo-
Croatian. And in East Austria of High German.

Several German dialects are also full of <a> pronounced /ɔ/ or
even /o/, but there is no common spelling rendition for that.
For instance, in spelling attempts used in Germany (Bavaria)
one writes <o>, Austrians tend even to use <aung> whenever
<a> is followed by <n>, as in the prefix <an->.

This is why, the Austrian esp. Viennese authors of Wiki-articles
written in the Bavarian dialect use for that the Danish-Swedish
letter <å> (which has been the best idea, IMHO), whereas the Bavarian
native-speakers from Germany do not use this letter, although
the sound is almost ubiquitous in most German dialects as well,
esp. in Bavarian, only that non-linguist Germans are not aware
of this vowel, that it is in fact neither an /a/ nor an /o/.
Besides, because there isn't any special letter for this vowel,
only native-speakers of the relevant dialects (e.g. Bavarian,
Saxonian, Swiss Alemanian & al.) know when to utter /ɔ/ and when /ʌ/.
Even German linguists with much "Übung" won't be able to
make the correct disctinctions betw. <a> /a/ and <a> /ɔ/ and /o/.
Hungarians don't have this problem, since their <a> and <á>
have only one type of vowel, not several or various ones.

><e> is shorter and lower than <é>.

Something similar here: <e> is always short and has the quality of
/æ, ɛ/ or like the French è. There is no long version of it.
OTOH, <é> is like the French <é> /e:/, and always has to be long.
There is no short variety of it (as the /e/ in French et, c'est,
Italian che, Engl. best, bet(ter), Ken, Span. tambien. Hung.
linguists sometimes use the spelling ë for a historic /e/ I
suppose, which no longer exists in contemporary Hungarian
dialects & standard language. In regional dialects in some
words <é> can turn <i> /i/ or <í> /i:/; and <e> tends to be
replaced by <ö> (e.g. fel-föl, szeg-szög).

><o> and <ó> are already /o/ and /o:/.

That's right. (And in many words in specific regional dialects
these tend to be replaced by <u> /u/ and <ú> /u:/. In a similar
way <ö> /ö/ and <ő> /ö:/ tend to turn <ü> /ü/ and <ű> /ü:/.
E.g. jön "comes" > dial. gyün "comes"; töröl "wipes" > türül.
But whichever the regional "deviation", the simple but fixed
Hungarian spelling is very helpful to anyone, incl. newbie
foreign studens as well as natives who learn the alphabet and
spellings.)

>>BTW, the Hungarian spelling is almost kind of a ... phonetic
>>transcription.
>
>Phonemic.

"Tetracapillotomy" :-)

>It correctly doesn't show assimilations.

To what avail would be the information regarding assimilations?

The pragmatic use of such "lucky" (light) spelling is that the
reader (including the "bloody" newbie) automatically knows
how the frigging written word has to be read/pronounced. Do
that in English, French, Portuguese, Danish, Dutch, Flemisch,
even in German! :-)

>So rarely that I can't remember ever having heard it.

It (the "raccsolás", with a geminate /č/ in it) is rare but not as rare
as one might assume majority having the apical /r/.

Tim

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 14, 2022, 12:46:01 AM9/14/22
to
On Tuesday, September 13, 2022 at 9:09:08 PM UTC+2, Tim Lang wrote:
> On 13.09.2022 16:22, Ruud Harmsen wrote:
>
> >Phonologically, historically, and morphologically, it is short a. The >actual phonetic value, somewhat shifted, probably for extra contrast,
> >does not change that. The spelling is appropriate.
> The spelling is as it is only because Hungarians once chose the <a>
> rendition. As the Danes adopted the <å>.
> ><a> is shorter, backer, higher and rounder than <á>.
> Yes, but in the native Hungarian perception and pronunciation the
> <a> is never an /a/ or /ʌ/, it is always, i.e., without any
> exception whatsoever, an /ɔ; ɒ/. So, in the strict sense of
> the wording "short a" the usage of it is ... wrong: because the
> Hungarian vowel simply written <a> is no /a/ at all. It might
> be perceived as an /a/ in other languages, incl. in German and
> Dutch.

One way to put it (just a model) is that
<a> = /a/ = [ɔ; ɒ]
<á> = /a:/ = [a:]
<e> = /e/ = [æ, ɛ]
<é> = /e:/ = [e:]

> This is why, e.g. words such as "Hans, hat, Tanne, Katze" & the like in
> _standard_ German (genuine Hochdeutsch), cannot be read with /a, ʌ/
> by a Hungarian unless s/he is _taught_ the correct pronunciation
> by a teacher who's aware of these differences and nuances.

I know one Hungarian who speaks Interlingua quite well, without interference from Hungarian vowels. Interlingua has the five vowel scheme common to Spanish, Esperanto, Yiddish and modern Hebrew.

> Otherwise,
> any Hungarian will automatically read <a> as /ɔ/ in any foreign
> spelling. Which happens zillion times in Hungary and in its
> former provinces whenever the German names of Hungarians ending
> in -mann (Neumann, Baumann, Kaufmann) are pronunced ... /mɔn/.
> Only trained people learn that -mann has to be pronounced /mʌn/.
> I once heard one native-speaker reading the English name spelling
> Alack ... /'ɔlɔtsk/, because that's the way one reads it in
> Hungarian; and this illustrates how they automatically see in
> every <a> their /ɔ/, unless they learn, say, English.

https://rudhar.com/toerisme/alcercth/en.htm
Quote from myself:
==
I am silly and weird

All this reminded me of silly mind games I sometimes play by myself, now and already as long ago as the 1980s or 1990s. For example, I used to reverse the name of the ancient and modern Dutch city of Utrecht, and got Thcertu. You could read that as the Portuguese word for ‘certain’, said with a Castilian accent:

Th as in English ‘think’.
C as in Spanish ‘cerca’. In large parts of Spain, that is pronounced with that same sound of English ‘think’, so in this hypothetical word, it is written twice.
The u at the end indicates how in Portuguese, the unstressed letter o is pronounced with a high vowel, [u].

I find such things funny, amusing and fascinating. If you don’t, I can quite understand that. I am silly and weird. But I do no harm to anyone by being like that, so I won’t try to be somebody I’m not.
More silliness

Another such silly game is to interpret the Dutch name for Egypt, ‘Egypte’, as if it were a Hungarian name, but with a Dutch schwa sound (which the Hungarian language does not have) at the end. So you’d get as a pronunciation, in the IPA alphabet: [ɛɟptɘ], or in Kirshenbaum’s ASCIIisation: [EJpt@], and in X-Sampa: [Ed'pt@].

Unpronounceable in pretty much any language. But fun. For me.
== end of quote ==

> Several German dialects are also full of <a> pronounced /ɔ/ or
> even /o/, but there is no common spelling rendition for that.

Yiddish is an example. Doz hot doz oykh.

> Something similar here: <e> is always short and has the quality of
> /æ, ɛ/ or like the French è. There is no long version of it.

There is a long version of /a/ = [ɔ]: when an Hungarian spells out
an internet address that contains a name with á, but all the accents
were stripped for the URL. Heard on Dankó Rádió. Unfortunately
I can't remember the example. I think the accouncer's name is
Popovics László.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 14, 2022, 12:50:05 AM9/14/22
to
On Wednesday, September 14, 2022 at 6:46:01 AM UTC+2, Ruud Harmsen via Google Groups <goo...@rudhar.com> wrote:
> There is a long version of /a/ = [ɔ]: when an Hungarian spells out
> an internet address that contains a name with á, but all the accents
> were stripped for the URL. Heard on Dankó Rádió. Unfortunately
> I can't remember the example. I think the accouncer's name is
> Popovics László.

Or maybe that was just dankoradio.hu, pronounced as written,
according to Hungarian rules.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 14, 2022, 12:59:10 AM9/14/22
to
>>> BTW, the Hungarian spelling is almost kind of a ... phonetic
>>> transcription.

On Tuesday, September 13, 2022 at 9:09:08 PM UTC+2, Tim Lang wrote:
> To what avail would be the information regarding assimilations?

To no avail. That was my point. And that's I wrote "correctly", correct
because a phoneme-based spelling does not need to show automatic
assimilations. Because they are automatic.

wugi

unread,
Sep 14, 2022, 4:56:31 AM9/14/22
to
Op 14/09/2022 om 6:59 schreef Ruud Harmsen via Google Groups
<goo...@rudhar.com>:
If they were automatic the Dutch wouldn't be busy nowadays corrupting
the whole system of Dutch assimiliation with their crazy
hypercorrections (together with crossover corruptions of strong/soft
consonants).

--
guido wugi

Tim Lang

unread,
Sep 14, 2022, 7:04:17 AM9/14/22
to
On 14.09.2022 06:46, Ruud Harmsen via Google Groups <goo...@rudhar.com>
wrote:

>One way to put it (just a model)

Exactly showing what's going on. "In a nutshell"

>is that

><a> = /a/ = [ɔ; ɒ]

Without any long variety of it whatsoever (non extant in Hungarian.
What I'm underscoring here is that this ist not a "short a" in the
sense of "short /a/", but it is a different vowel, namely /ɔ/, also
IPa-written /ɒ/. In Hungarian speech replacing this one with a
... short /a/ means only this: the speaker is a foreigner; or
the speaker speaks some Hungarian small dialect from the plains
between the rivers of Tisza in the East and the Danube in the
center of "little" Hungary. Everybody would understand what such
people'd talk about, though.

><á> = /a:/ = [a:]

Here too: there is no short variant of this /a/ (which is the genuine
/a/, in contrast with the vowel written <a> in Hungarian, which is
no /a/ at all.

In Hungarian classes of any kind nobody teaches the stuff this
way, foreigners understanding what's going on only based on the
IPA transcriptions for these vowels - as well as by listening
to the way native pronounce them.

What's important too is the fact that these Hungarian vowels keep
staying constant, they do not vary according to individuals - i.e.,
unlike the case in all German dialects and High German; in French;
in English, both BE and AE etc, and I dare say in Flemish and
Dutch.

><e> = /e/ = [æ, ɛ]
><é> = /e:/ = [e:]

Yes. And here my comment too: /e/ has no long "sibling", and /e:/
has no short variant in Hungarian (this in contrast to other
European languages that have both kinds of /e/, even 1-2 more
variants of it). With the aspect worth mentioning that in small
regions (approx. in the center of today's Hungary, people speaking
that dialect tend to pronounce that "short" <e> as it were an
... <á> /ʌ/, but a ... short one, which otherwise doesn't exist.
e.g. edd meg ezt (colloquially and dialectally also "eztet")
(which means "eat this") is pronounced /æd mæg æzt(æt)/, but
that tiny regional Hungarian has it ... ʌdd mʌg ʌzt(ʌt). Who-
ever doesn't know those native exist, automatically concludes
such people are foreigners unable to correctly pronounce that
kind of /e/, ie, /æ ɛ/ (which in French is written <è>).

>I know one Hungarian who speaks Interlingua quite well, without
>interference from Hungarian vowels. Interlingua has the five vowel
>scheme common to Spanish, Esperanto, Yiddish and modern Hebrew.

Of course, there are whole lotte talented people: ie, with
best auditive acuity combined with the ability to ... imitate
the original sounds. But most people aren't "endowed" with
such skills (even after four decades of "willy-nilly" training
surrounded by natives only :-)).

>https://rudhar.com/toerisme/alcercth/en.htm

The Hungarian spelling convention <gy> has nothing in common with
the <gy> usage in other spellings (for other languages). The
sound /ɟ/, which is kind of /d/ pronounced with the tongue
kept in the maximum kind of relaxation (and almost impossible
to be done by most Europeans except for: Slovaks, Poles,
Ukrainians, Russians and that part of Romanians who live in
Transylvania) and pushed against the palate also in an
extreme way. The "devoiced" counterpart or variant of it
is <ty>. These hardly can be compared with /dʒ/ and /tʃ/,
although there are - indeed - two Hungarian areas where
the Hungarian natives themselves cannot pronounce these
two "soft" plosive consonants properly, so that they replace
them with /dʒ/ and /tʃ/, ie, in Croat spelling: <dž> and <č>
(or <ć>). But this dialectal phenomenon is within the entire
Hungarian frame an "exotic" one. Those who aren't aware of
that dialect think those people pronouncing that way were ...
foreigners. And there is a logic in it: All those who don't
have these 2 sounds in their languages, tend to help them-
selves by using these "alternatives". But moreover: the
few Slavic equivalents of them aren't always the same:
methinks, that the Russian, Belorussian and Ukrainian
sound a bit different in the narrowest transcription than
the Hungarian, Slovak and Polish (at least Silezian)
varieties. (Note that Slovenian, Croatian - all neighbors
of Hungarians for at least 1,000 years - as well as Bulgarian
don't have these "weird" plosives of /d/ or /g/ and /t/ or
/k/ respectively; but they are also used by Western and
Eastern Slavs. (Once, Hungarian was influenced by both Russian
and Western Slavic idioms.)

>I find such things funny, amusing and fascinating.

Being attracted, obsessed by such things, developing even a passion for
them, is something very typical of people interested in linguistics, in
poetry, and in cryptology. Even more interesting that surprisingly many
of them aren't interested in ... math (mostly getting poor results in
school in math/calculus) & physics). I don't know how much all this is
also related to certain levels of "Asperger" (autism).

>If you don’t, I can quite understand that. I am silly and weird. But I
>do no harm to anyone by being like that, so I won’t try to be somebody
>I’m not. More silliness

Other people have become "VIP"s, say, in poetry or in cryptology or
as linguists, because of this "silliness", where they simply have ...
fun.

>Another such silly game is to interpret the Dutch name for Egypt,
>‘Egypte’, as if it were a Hungarian name,

BTW, in the Hung. spelling, there is the need to add an <i>, Egyiptom,
otherwise no Hungarian could utter Egypt(om). Or out of 10 million
only 1-2 thousand might be able. :-) E.g. in your example, if you
split Egypte in Egy (as the Hungarian word for "one") and pte as
if it were something ... Greek (as in Ptolemaios/Ptulmis; or as
in Engl. apt(itude)), then it would be utterable.

>But fun. For me.

[Wenn das ausgerechnet auf SCI.LANG nicht nachvollziehbar wäre, hehe,
dann wäre das Ganze 'ne verkehrte Welt. ;-)]

>Yiddish is an example. Doz hot doz oykh.

This is the American spelling for Yiddish. With <kh> for German <ch>.
OTOH, the <z> spelling is better that what German spelling can offer,
since there is no letter in it for /z/: in German spelling, there
is only <s> in use and has to cover both cases, voiced /z/ and
devoiced /s/ as well as the spelling <ß>.

doz hot, and especially hot illustrates that Yiddish is a South-
German, ie, an "Oberdeutsch", dialect (its closest kinship being the
Bavarian dialect. Despite the Yiddish alternative development
of /o:/ > /oy/ for /au/ (Germ. <auch>), which in the genuine
Bavarian dialect has chosen another evolution path: <aa> (this
is the dialectal contracted form of High German <auch> "too,
as well").

>There is a long version of /a/ = [ɔ]: when an Hungarian spells out
>an internet address that contains a name with á, but all the accents
>were stripped for the URL. Heard on Dankó Rádió. Unfortunately
>I can't remember the example. I think the accouncer's name is
>Popovics László.

(Earlier in the 20th cent. in telegraphy á, é, í, ó, ú were rendered
this way: aa, ee, ii, oo, uu, ie, Dankoo Raadioo. In ASCII emails
I know that only the long ö and ü, that is: ő, ű, are replaced by
circumflexed ones: ô, û.)

I don't understand this comparison. Dankó Rádió is Dɔkooo Raaadiooo.
(With the etymological detail that Dankó is the (South) Slavic Danko,
as a diminutival of Dan(iel).)

Where "aaa" means a real clear a as, say, in Italian, French (when
not nazalized), Bühnendeutsch-High German or in standard Spanish
(with the mouth widely open), and "ooo" for a real perfect /o/
with the smallest mouth/lips aperture you can imagine. If you
choose apertures as in Dutch, German and other idioms, than
you'll invariably get the vowel which is written in Hungarian
a and is incorrectly deemed as a ... "short a", but which is an
/ɔ/ of the purest kind. (Persian also has it: Erɔɔɔɔn and
Afhɔnistɔɔɔɔn. In Bavarian, wɔs hɔb i dia xɔkt? is not always
the same, since in the Vienna area and in several other Austrian
and Bavarian areas this proniunciation is rather wos hob i d(i)a
xokt? Or, as a tourist one often can hear the sentence: "Das
ist mein Platz!" pronounced with this "closed" /o/ where in
standard German should be a real /a/: "Des is mei Bloootz".)

Tim


Ruud Harmsen

unread,
Sep 14, 2022, 7:07:40 AM9/14/22
to
Wed, 14 Sep 2022 10:56:23 +0200: wugi <wu...@scrlt.com> scribeva:
They are automatic, but social perceptions about certain accents
(z-less ones, in this case) can interfere, and create some strange
sort of half-awareness, which is now indeed leading to a serious
language change in the Dutch of the Netherlands, but not that of
Flanders: progressive (forward) assimilation of plosives to sibilants
is replaced by regressive assimilation.

"Het zuiden" (the south) used to have an [s], due to the t, although
"zuidelijk" (southern) has a [z]. The new tendency is to maintain the
z throughout, and avoid the assimilation.

I've already heard a 3 year old do this too, who cannot read yet, and
who hasn't been in contact with z-less dialects of Dutch (like in
Amsterdam). So for her, it is no longer a hypercorrection, but just
the new norm. That means the change is probably irreversible. Sadly,
because I don't like it at all.

I may have mentioned this before:
https://rudhar.com/fonetics/progassi.htm
https://rudhar.com/fonetics/progassj.htm
https://rudhar.com/fonetics/progassk.htm
https://rudhar.com/fonetics/progassl.htm
(all in Dutch, and automatic translation may do a poor job in such
cases).

Tim Lang

unread,
Sep 14, 2022, 7:15:42 AM9/14/22
to
On 14.09.2022 06:50, Ruud Harmsen via Google Groups <goo...@rudhar.com>
wrote:

>>There is a long version of /a/ = [ɔ]: when an Hungarian spells out
>>an internet address that contains a name with á, but all the accents
>>were stripped for the URL. Heard on Dankó Rádió. Unfortunately
>>I can't remember the example. I think the accouncer's name is
>>Popovics László.

BTW, this "means" Vladislav (or Ladislav or Vaclav) Popović/Popovič.

>Or maybe that was just dankoradio.hu, pronounced as written,
>according to Hungarian rules.

Yes, it must be spelled dankoradio.hu for the http-compatible rendering.
There's nothing Hungarians could do about it as long as the http-lines
cannot tolerate signs such as á and ó.

But, otherwise, the correct spelling is that one you already cited:

Dankó Rádió. (In German spelling this could be rendered as
Dannkoh Rahdioh, which would show the same vowel quality as
the Hung. spelling does.) Hungarian spelling always shows the real
pronunciation of the vowels by always adding (ie, with no exception)
the appropriate diacritical sign, here the ´ accent: ó, á.

With the Hungarian pe- culiarity: the accents aren't used to show any
... stress ("accentuation"), but so signalize "watch out: the vowel I'm
riding is a long vowel!"

Tim

Ruud Harmsen

unread,
Sep 14, 2022, 12:47:44 PM9/14/22
to
Mon, 12 Sep 2022 22:29:29 +0200: Tim Lang <m...@privacy.net> scribeva:
>In theory as well as in emphatic, correct pronunciation. But in
>normal colloquial (rapid!) speech the difference between <ssz> and
>sz as well as <n> and <nn> is difficult to hear. (E.g. very frequent
>in building the subjunctive and imperative.)

I do hear a lot of "double" consonants also in the interviews here:
https://www.youtube.com/watch?v=KstWaBoQdTU

OK, but those people are professional actors and singers, so they
don't talk naturally?

Ruud Harmsen

unread,
Sep 14, 2022, 1:32:59 PM9/14/22
to
Wed, 14 Sep 2022 13:15:37 +0200: Tim Lang <m...@privacy.net> scribeva:

>On 14.09.2022 06:50, Ruud Harmsen via Google Groups <goo...@rudhar.com>
>wrote:
>
>>>There is a long version of /a/ = [?]: when an Hungarian spells out
>>>an internet address that contains a name with á, but all the accents
>>>were stripped for the URL. Heard on Dankó Rádió. Unfortunately
>>>I can't remember the example. I think the accouncer's name is
>>>Popovics László.
>
>BTW, this "means" Vladislav (or Ladislav or Vaclav) Popovi?/Popovi?.

Yes, but he sounds native Hungarian.

>>Or maybe that was just dankoradio.hu, pronounced as written,
>>according to Hungarian rules.
>
>Yes, it must be spelled dankoradio.hu for the http-compatible rendering.
>There's nothing Hungarians could do about it as long as the http-lines
>cannot tolerate signs such as á and ó.

I think they can, now, for quite a few years already. Or would that
only be for https?

https://dankórádió.hu/ tries to connect to xn--dankrdi-lwa4ne.hu, but
such domain happens to exist. More at:
https://en.wikipedia.org/wiki/Internationalized_domain_name
https://en.wikipedia.org/wiki/Punycode

>But, otherwise, the correct spelling is that one you already cited:
>
>Dankó Rádió. (In German spelling this could be rendered as
>Dannkoh Rahdioh, which would show the same vowel quality as
>the Hung. spelling does.) Hungarian spelling always shows the real
>pronunciation of the vowels by always adding (ie, with no exception)
>the appropriate diacritical sign, here the ´ accent: ó, á.
>
>With the Hungarian pe- culiarity: the accents aren't used to show any
>... stress ("accentuation"), but so signalize "watch out: the vowel I'm
>riding is a long vowel!"

Yes. But no way to teach non-Hungarians that Bartók has inititial
stress.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 14, 2022, 1:49:18 PM9/14/22
to
On Wednesday, September 14, 2022 at 1:04:17 PM UTC+2, Tim Lang wrote:
> On 14.09.2022 06:46, Ruud Harmsen via Google Groups <goo...@rudhar.com>
> wrote:
> >One way to put it (just a model)
> Exactly showing what's going on. "In a nutshell"
> >is that
>
> ><a> = /a/ = [ɔ; ɒ]
> Without any long variety of it whatsoever (non extant in Hungarian.
> What I'm underscoring here is that this ist not a "short a" in the
> sense of "short /a/", but it is a different vowel, namely /ɔ/, also
> IPa-written /ɒ/.

Yes, it sounds different, but I'm not referring to the sound, but to the
historical, morphologic and grammatical function of it, and of its "long"
pendant. In other words, I'm talking about phonemes, my version of it.
You seem to equate phones and phonemes, so you call [ɔ] /ɔ/ because that's
how it sounds. I don't. I call [ɔ] /a/, because when in grammatical context
it lengthens, it becomes /a:/, which sounds as [a:].

>In Hungarian speech replacing this one with a
> ... short /a/

You mean [a] here, I think. Phonemes, between / /, do not directly
related to how it's pronounced. Indirectly, yes.

> means only this: the speaker is a foreigner; or
> the speaker speaks some Hungarian small dialect from the plains
> between the rivers of Tisza in the East and the Danube in the
> center of "little" Hungary. Everybody would understand what such
> people'd talk about, though.

By using /a/ I do not mean it should sound like [a]. It sounds like [ɔ],
of course, always.

> ><á> = /a:/ = [a:]
> Here too: there is no short variant of this /a/ (which is the genuine
> /a/, in contrast with the vowel written <a> in Hungarian, which is
> no /a/ at all.

There is a double constrast, of length AND timbre, for maximal difference,
maximal clarity, minimal ambiguity. Useful in a language with so many
vowels, and with a high functional load of these 4 phonemes /a/, /a:/,
/e/, and /e:/ (sounding as [ɔ], [a:], [ɛ] and [e:]).

> What's important too is the fact that these Hungarian vowels keep
> staying constant, they do not vary according to individuals - i.e.,
> unlike the case in all German dialects and High German; in French;
> in English, both BE and AE etc, and I dare say in Flemish and
> Dutch.

I know.

Ruud Harmsen

unread,
Sep 14, 2022, 2:07:41 PM9/14/22
to
RH:
>>https://rudhar.com/toerisme/alcercth/en.htm
> [...]
>> I find such things funny, amusing and fascinating.

Wed, 14 Sep 2022 13:04:06 +0200: Tim Lang <m...@privacy.net> scribeva:
>Being attracted, obsessed by such things, developing even a passion for
>them, is something very typical of people interested in linguistics, in
>poetry, and in cryptology. Even more interesting that surprisingly many
>of them aren't interested in ... math (mostly getting poor results in
>school in math/calculus) & physics).

At school I was good at math and physics too. Better than at
languages. Not chemistry, because you had to learn things by heart for
that, couldn't just understand it. I thought only the latter made
sense, found the former a waste of time.

I was very wrong of course.

>I don't know how much all this is
>also related to certain levels of "Asperger" (autism).

Perhaps it is. I don't if I have Asperger. I do think I am different
from the other, and always was. But in what way exactly, how it might
fit into a "spectrum" idea, I don't know.

> >Another such silly game is to interpret the Dutch name for Egypt,
> >‘Egypte’, as if it were a Hungarian name,
>
>BTW, in the Hung. spelling, there is the need to add an <i>, Egyiptom,
>otherwise no Hungarian could utter Egypt(om). Or out of 10 million
>only 1-2 thousand might be able.

Yes, it is impossible for Hungarians. Because it's Dutch and I
deliberately misinterpreted it. The closest Hungarian rendering of the
Dutch pronunciation would be Éhiptö, but then the stress is still
wrong.

> :-) E.g. in your example, if you
>split Egypte in Egy (as the Hungarian word for "one") and pte as
>if it were something ... Greek (as in Ptolemaios/Ptulmis; or as
>in Engl. apt(itude)),

Yes.

>then it would be utterable.
>
> >But fun. For me.
>
>[Wenn das ausgerechnet auf SCI.LANG nicht nachvollziehbar wäre, hehe,
>dann wäre das Ganze 'ne verkehrte Welt. ;-)]
>
> >Yiddish is an example. Doz hot doz oykh.
>
>This is the American spelling for Yiddish. With <kh> for German <ch>.

Yivo translitteration, not specifically American, although Yivo is now
there. (And I may have some details wrong.)

>OTOH, the <z> spelling is better that what German spelling can offer,
>since there is no letter in it for /z/: in German spelling, there
>is only <s> in use and has to cover both cases, voiced /z/ and
>devoiced /s/ as well as the spelling <ß>.
>
>doz hot, and especially hot illustrates that Yiddish is a South-
>German, ie, an "Oberdeutsch", dialect (its closest kinship being the
>Bavarian dialect.

Mitteldeutsch, Central German, it seems. Speyer, Worms, etc. Googling,
I now find
https://nl.wikipedia.org/wiki/ShUM-sites_van_Speyer,_Worms_en_Mainz .

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 14, 2022, 2:12:27 PM9/14/22
to
On Wednesday, September 14, 2022 at 1:04:17 PM UTC+2, Tim Lang wrote:
> (Persian also has it: Erɔɔɔɔn and Afhɔnistɔɔɔɔn.

Yes. But the (historical) length-timbre relation is reversed
in comparison with Hungarian:
Persian: short and clear / long and dark
Hungarian: short and dark / long and clear (or neutral)

Tim Lang

unread,
Sep 14, 2022, 3:45:22 PM9/14/22
to
On 14.09.2022 18:47, Ruud Harmsen wrote:

>I do hear a lot of "double" consonants also in the interviews here:
>https://www.youtube.com/watch?v=KstWaBoQdTU
>
>OK, but those people are professional actors and singers, so they
>don't talk naturally?

Those interviews are very good examples of natural, common colloquial
Hungarian, natural ways of everyday's speech, despite the fact that
most of them are professionals from the business of theater, music
& the like, ie, having a bit more knowledge & awareness concerning
good Hungarian (grammar & style) than the average natives.

Also listen to the following examples (whenever having leisure time
and/or the mood might strike you :-)). All in all, they are very
good/illustrative examples too of genuine Hungarian pronunciations in
various real & natural contexts:

Good examples of common everyday's colloquial Hungarian
(with a few regional and slangy words, but very rich
in showing all kind of colloquial idiosyncrasies as
well as regional or sociolect ones (esp. in the case
of the drill sergeant in the army); as well as for
the benefit of foreigners: richness of all kind of
aspects of speech "melody", accents, and social idio-
syncrasies, which can't be conveyed by the best
teachers). Among typical colloquial words: "aszongya"
the contracted form of "azt mondja" ("s/he says this/that");
"akkó" for "akkor" ("then"), "kü:dte" for "küldte" ("s/he
send him/her/it/them"); position 10:49 min an example
for colloquial/regional transformation of "törölni" (to
wipe) > "türülni" (here "fel kell türülni"); etc:

___the comedian Hofi (Géza Hoffmann), actor
(subject here: soccer)
<https://www.youtube.com/watch?v=mHDk7aN-sag>
(plays the bedlamite - satire)
<https://www.youtube.com/watch?v=KZzwLVQxqRo>
(how was his life and that of his friends,
as a mil. newbie privates; and how dealt with
them the 'gunnery sergeant')
<https://www.youtube.com/watch?v=4eXvL2vVrdQ>

___old movies, comedy:
<https://www.youtube.com/watch?v=l2jbhpEsyd4>

(YT has a vas collection of Hung. older movies, in
full; e.g. a good search words combination would
be "régi magyar filmek teljes". "Kabaré" for
a rich collection of cabaret/satire/humor clips.)

___theater (60s) starring Zoltán Latinovits,
one of the best Hungarian actors in those years:
<https://www.youtube.com/watch?v=l2jbhpEsyd4>

___pop-rock groups, very popular in 70s (not only
in Hungary, but also in other countries of the
East, incl. the USSR). (Influences by Deep
Purple, Led Zeppelin, Uriah Heep & the like are
obvious.)

Omega, "Petróleum lámpa"
<https://www.youtube.com/watch?v=4szfWQ7Sspg>

Omega, "Trombitás Frédi"
<https://www.youtube.com/watch?v=TFt1YgWb7qg>

Skorpió, "Így szólt hozzám a dédapám"
<https://www.youtube.com/watch?v=Cf5A5nFWN9Q>

(You'll hear that -zz- in "hozzám" ("to me")
is not at all exaggerated. And in "jöjj" even
less a doubling of the semivowel -jj- (as y in
English). It is actually doubled in written
since this is the ... imperative form (for "te"
= "you; thou", singular) of the verb "jön".
I reiterate my own opinion: Italian geminates are
a bit stronger in pronunciation than the
Hungarian ones.)

Skorpió, "Jó lenne, ha szeretnél"
<https://www.youtube.com/watch?v=lX0as9tuAoY>

LGT (Lokomotiv GT), "Ringasd el magad"
<https://www.youtube.com/watch?v=sWmm-eId8xg>

Tim

Tim Lang

unread,
Sep 14, 2022, 4:08:05 PM9/14/22
to
On 14.09.2022 19:32, Ruud Harmsen wrote:

>>BTW, this "means" Vladislav (or Ladislav or Vaclav) Popovi?/Popovi?.
>
>Yes, but he sounds native Hungarian.

Of course. I only wish to underscore: don't be surprized by Hungarians
having genuine German, Slovakian, Croat, Serbian names. Millions of
Hungarians have these ethnic origins too, even whole lot of them who
no longer speak the idiom of their ancestors. (Popovics might also
have ... Romanian ancestors, since this name, a derivation of "pop/a",
i.e., the Orthodox priest/parishioner, is as popular in Romania as
it is in Serbia and Russia.)

And also to point out that the name László which is perhaps the
most popular Hungarian male name is a derivation of the Slavic
Vladislav (alias Ladislav alias Vaclav /vatslav/). Its main diminutival
form is Laci /'lɔtsɪ/. Some people perpetuate a mistake that says:
László would be the equivalent of Greek Basilios > Basil, Basilio etc.
But this Greek name (derived from basileos "king") has its equivalent
(derivation) in Hungarian as ... Vazul /'vɔzul/ (ie, not in the least
László).

>I think they can, now, for quite a few years already. Or would that
>only be for https?

Methinks, it is possible nowadays, since I myself have seen such
examples of tolerated diacriticals in http(s) lines.

>Yes. But no way to teach non-Hungarians that Bartók has inititial
>stress.

That's true. Unless one learns that ó in Bartók doesn't refer to
a stressed last syllable, but to know that this /o/ is a long one
(as if written in German would result in Bartohk or Bartook; or
in English ... Barrtawk). BTW: Bartók's paternal old roots were in
today's Slovakia, and his mother was an ethnic German, Voit (which
means Vogt/Voigt/Voight - if it didn't happen to have been a
Slavik Vojta/šek, since there, from where his mother originated,
not only were Germans a compact minority, but also quite many
Slovaks as well as Czechs, and Austrian colonists of Czech or
Moravian or Slovak and Silezian descent!). OTOH, Bartók - a highly
strange name in Hungarian, might very well have been an adaption
of the German name Barth ("beard" or "Bartholomew"). Noteworthy
in this context: almost all Hungarians bearing surnames containing
bart- are called and registered as ... Barta /'bɔrtɔ/. Bartók sounds
as if it were the plural for Barta, but the lengthening of the o-vowel
thwarts this hypothesis.

Tim

Tim Lang

unread,
Sep 14, 2022, 4:37:58 PM9/14/22
to
On 14.09.2022 19:49, Ruud Harmsen via Google Groups <goo...@rudhar.com>
wrote:

>Yes, it sounds different, but I'm not referring to the sound, but to the
>historical, morphologic and grammatical function of it, and of its "long"
>pendant.

Yes, sure - this is another, a different story. But the wording "short
a" might prompt people who study Hungarian think Hungarian <a> in
spelling would be a short <á>, which is a wrong assumption

In other words, I'm talking about phonemes, my version of it.
> You seem to equate phones and phonemes, so you call [ɔ] /ɔ/ because that's
> how it sounds. I don't.

Of course you don't, since you're a Dutch man. If you really had
been exposed to Hungarian (living many years among them), then
you would have gradually seen another "PHON" world whatsoever,
an extreme different one from Dutch, let alone English and (worse)
Danish.

And you still do not get the gist of my "message": Hungarian <a> is
always /ɔ/ and nothing else. It is no A vowel. /ɔ/ is a variety
for /a/ in other languages, starting with all German dialects and
Hochdeutsch, with Dutch, Flemish, French (an, am, en, em also con-
taining variants of /ɔ/ are deemed as variants of /a/), English etc.
I'm merely underscoring the fact that Hungarian <a> /ɔ/ is no
"child" or "parent" of /a/. It might have been 500-1,000 years
ago, but not in the latest 300 years or so.

>I call [ɔ] /a/, because when in grammatical context
>it lengthens, it becomes /a:/, which sounds as [a:].

Because you, as a Dutchman, and esp. in the vicinity of the ... Reich
you are as influenced as you can be. (To native Germans everything
from the darkest /o/ to the most strident /a/ are the same vowel,
generally deemed as an <a> /a(:)/. Even if considerable percentages
of the native speakers pronounce "ea hot xokt, des is mei Plotz",
everybody would automatically decode this as /ɛə hat gəza:kt, das
ist maɪn pla:ts/. If one does that in Hungarian, one get's into
"Teufels Küche".)

>By using /a/ I do not mean it should sound like [a]. It sounds like [ɔ],
>of course, always.

In Hungarian there is no /a/ or [a]. All Hungarian equivalents of this
/a/ are long: <á> /ɑː/. Otherwise Hungarians read it only as /ɔ/.

>There is a double constrast, of length AND timbre, for maximal difference,
>maximal clarity, minimal ambiguity.

Which is the case in Hungarian: there is no mixing up. In gross contrast
with German, Dutch, English and several other languages, where the
confusions/mixing-ups happen almost always even in the standard pronun-
ciation. In Hungarian, they don't have this "freedom" of "improvisation"
and interference with all kind of regiolects as well as personal
(dyslexic) peculiarities. In Hungarian vowels are of one kind, with
no fluctuations. And the few alternative ones are limited to certain
regional dialects.

Useful in a language with so many

"Many"? E.g. French and German has more variants for <ö> than Hungarian
(where short ö has a single alternate: the long one, written ő). Let
alone the vowel richness when followed by nasals in French and Polish.

>vowels, and with a high functional load of these 4 phonemes /a/, /a:/,
>/e/, and /e:/ (sounding as [ɔ], [a:], [ɛ] and [e:]).

Yes, but many other kinds of <e> aren't extant in Hungarian.

>I know.

Na, hát akkor! :-)

Tim

Tim Lang

unread,
Sep 14, 2022, 4:58:36 PM9/14/22
to
On 14.09.2022 20:07, Ruud Harmsen wrote:

>Mitteldeutsch, Central German, it seems. Speyer, Worms, etc. Googling,

It is a wrong information, although real scholars have proved
100-120 years ago that Yiddish is a South German dialect, but
not a Suebian-Alemanian one (Speyer, Worms) in the first place
but, a Bavarian, Franconian, Boemian one. And its alleged
mitteldeutschness is far-fetched. One should listen to the
main vocabulary: it is a typical one of the South-Eastern
German (Bavaria & Austria), incl. the vocabulary peculiarities
that in these provinces almost vanished: <ets> ("ihr" = plural
"you"), <gwen> /gve:n/ = <gewesen; Engl. been>, <enk> = <euch>
(dative and accusative of you), <hejnt> = <heut(e)>, <today>.
In the Worms and Speyer dialect the differences to these and
many other aspects are much bigger. This can be heard by any
of us, just listening. The only thing that makes the comparison
difficult is the Yiddish pronunciation, which is extremely
different from South German dialects, as well as the abundance
of Hebrew words, that, at that, are distorted. E.g. bet "house"
is Bejs(l) in Yiddish (and as such also extant in Bavarian,
Bäjsl and Bäsl); shabbat > Yid. schabbes and schabbəs.
Note that after being driven out of Suebian Speyer, German Jews
lived for centuries in Bohemia and the adiacent Bavarian-Speaking
countries. Even the German dialects once spoken East of Bohemia,
in any of the relevant Polish and Slovak provinces were South-
German, having very much in common with the big group of Bavarian
(all Austria as well as the German Pannonia incl. the Buda
half of Budapest (a.k.a. Ofen) was/is "Bavaria", with the exception
of the province Voralberg neighboring Switzerland: that one is
Alemanian). (By and large, even Mainz is rather a southern German,
e.g. mixture of Mitteldeutsch and Oberdeutsch, in contrast with
the Low German dialects, which in turn represent kinda "continuum"
from the shores of Netherland to former Königsberg = today's
Kaliningrad, where the few Germans living there are linguistically
also southern ones, the descendants of the "Volga-Deutsche" colo-
nists, whom Stalin deported to Central Asia and Siberia in 1941ff.)

Tim

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 15, 2022, 2:26:58 AM9/15/22
to
>> RH:
>>> Yiddish is an example. Doz hot doz oykh.

> Wed, 14 Sep 2022 13:04:06 +0200: Tim Lang <m...@privacy.net> scribeva:
>>This is the American spelling for Yiddish. With <kh> for German <ch>.

> Yivo translitteration, not specifically American, although Yivo is now
> there. (And I may have some details wrong.)

https://rudhar.com/lingtics/intrlnga/poesia/4viages/4viages.htm#Poema

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 15, 2022, 2:36:54 AM9/15/22
to
On Wednesday, September 14, 2022 at 9:45:22 PM UTC+2, Tim Lang wrote:
> ___the comedian Hofi (Géza Hoffmann), actor
> (subject here: soccer)
> <https://www.youtube.com/watch?v=mHDk7aN-sag>
> (plays the bedlamite - satire)

I learnt a new English word!:
https://www.collinsdictionary.com/dictionary/english/bedlam
(although it sounds strangely familiar, is there an expression
that contains the word?)

Interesting etymology! Reminds me of Maasoord, in Poortugaal
(not Portugal), near Rotterdam, a mental health institute. It has
had a different name for a decades now, but colloquially and
regionally the old name is still used to indicate people with
mental issues, as perhaps we might say these days.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 15, 2022, 2:48:27 AM9/15/22
to
On Wednesday, September 14, 2022 at 9:45:22 PM UTC+2, Tim Lang wrote:
> Skorpió, "Így szólt hozzám a dédapám"
> <https://www.youtube.com/watch?v=Cf5A5nFWN9Q>
>
> (You'll hear that -zz- in "hozzám" ("to me")
> is not at all exaggerated. And in "jöjj" even
> less a doubling of the semivowel -jj- (as y in
> English). It is actually doubled in written
> since this is the ... imperative form (for "te"
> = "you; thou", singular) of the verb "jön".
> I reiterate my own opinion: Italian geminates are
> a bit stronger in pronunciation than the
> Hungarian ones.)

Yes, I did notice that jöjj with a double jj, in Hungarian
songs, probably one of these:
https://www.youtube.com/watch?v=EzqbjybiUX4&list=PUQ-p0dzemvz4VWbWOvEsHtg
or:
https://www.youtube.com/watch?v=aAK0nMxH7Zo&list=PLrMT5jYRci1h8TEYHa_jC7EcwasqZDRe2
knowing it belonged to a verb, but not what it meant.

wugi

unread,
Sep 15, 2022, 5:46:00 AM9/15/22
to
Op 14/09/2022 om 22:08 schreef Tim Lang:
> On 14.09.2022 19:32, Ruud Harmsen wrote:

>> Yes. But no way to teach non-Hungarians that Bartók has inititial
>> stress.
>
> That's true. Unless one learns that ó in Bartók doesn't refer to
> a stressed last syllable, but to know that this /o/ is a long one
> (as if written in German would result in Bartohk or Bartook; or
> in English ... Barrtawk).

Er, why not Bortoak instead of Barrtoak, then? (In our Flemish classical
programs I hear him named as BAArtook, strange short a, that).

--
guido wugi

Tim Lang

unread,
Sep 15, 2022, 7:18:05 AM9/15/22
to
On 15.09.2022 11:45, wugi wrote:

>>That's true. Unless one learns that ó in Bartók doesn't refer to
>>a stressed last syllable, but to know that this /o/ is a long one
>>(as if written in German would result in Bartohk or Bartook; or
>>in English ... Barrtawk).
>
>Er, why not Bortoak instead of Barrtoak, then?

Because:

(1) bor- in English would be too /ɔː/ (as in <bore> and <boring>), in
which case the Hungarian spelling would have been different,
namely Bortók, and not Bartók. (@ wiki, there is a picture of him
very young, signed by him with the Hungarian spelling, Bartók).

(2) -oak would be almost automatically interpreted as /oʊk/ and /əʊk/,
which is not the case in the Hungarian pronunciation for Bartók.
The <o> here is a mere /ɔː/. Which according to Hung. spelling
rules could be written like this as well: Bartoók. There are
precedences in the spelling of surnames, esp. in Gaál (for Gál)
and Soós /ʃɔːʃ/, which in German common spelling would look
like this: Gahl/Gaal, Schohsch/Schoosch.

>(In our Flemish classical programs I hear him named as BAArtook,
>strange short a, that).

What do you mean by "AA"? /ʌ/ or /ɔ/? If the latter (as in Danish
Aarhus = Århus), then it would fit.

An interesting and misleading thing is the spelling Bártók, in Slovakian
as well as in Portuguese and Spanish. In which cases the <á> might
show the stress on the 1st syllable and not the length/quality of
the vowel. Yet, for Slovakian I don't know: it might refer to the
quality of the vowel, perhaps to /ʌ/, in which case the vowel would
be completely different from the Hungarian version. But OTOH what
if the real origin of this surname is ... Slovak?!? (The ending
-ok is popular with Slovak surnames. Besides, Bartóks paternal
ancestors were in Slovakia at home until way back in the days of
the 17th century.)

I checked it out, and, bingo (!), the sounds list in the wiki article
"Slovak language" the <á> spelling (like in Hungarian) has the same
value ... [ɑː] (@ wiki written [a:]). This means that this name is
pronounced in Slovakian /'bɑːrtɔːk/ as well, contrasting the Hungarian
/'bɔrtɔːk/ pronunciation. (Slovak <ó> is treated as it is in Hungarian:
a long /o/, ie, /ɔː/.)

However, the wiki article uses the Hungarian, not the Slovak,
spelling of the name: <https://sk.wikipedia.org/wiki/Béla_Bartók>
(perhaps based on the "P.C." idea).

Tim

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 15, 2022, 10:29:47 AM9/15/22
to
On Wednesday, September 14, 2022 at 9:45:22 PM UTC+2, Tim Lang wrote:
> Skorpió, "Jó lenne, ha szeretnél"
> <https://www.youtube.com/watch?v=lX0as9tuAoY>

An example of musical metre not in accordance with
the stress pattern of the languages: they sing:
JÓ lenNE, ha SZERetNÉL
Perhaps that wrecks the proper -nn- in LENne?

Cf. https://rudhar.com/musica/ruivelte/en.htm

I recently noticed the same in "Hej cigány", from the
operetta Marica grófnő (Gräfin Mariza), music by
Emmerich Kalman (Kálmán Imre), lyrics/libretto by
Julius Brammer & Alfred Grünwald:
https://lyricstranslate.com/en/dolhai-attila-hej-cigany-lyrics.html
https://www.youtube.com/watch?v=eq4ypR94Bqg
==
Mit tudjáTOK ti benn, hogy idekinn mi fáj?
Hej, voltam én is egyszer büszke bálkiRÁLY!
NóTÁmat én is százszor elhúzattam!

Ne olyan nagyon fennen, hej ti, odaBENN!

Hej, cigány, hej, cigány!
Húzd a nóTÁM!
Száz feLÉ szakadJON a húr!

Jer ide!
Hej cigány, hej cigány!
Húzd, ne sajNÁLD!
Így mulat minálunk egy úr!
==

Cause: it is a translation from the original German text,
https://aimsgraz.com/wp-content/uploads/2014/09/K-Komm-Zigany.pdf
That German text does have a perfect metrum, with music
and language accents coinciding.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 15, 2022, 10:53:18 AM9/15/22
to
> On 14.09.2022 19:49, Ruud Harmsen via Google Groups <goo...@rudhar.com>
>> In other words, I'm talking about phonemes, my version of it.
>> You seem to equate phones and phonemes, so you call [ɔ] /ɔ/ because that's
>> how it sounds. I don't.

On Wednesday, September 14, 2022 at 10:37:58 PM UTC+2, Tim Lang wrote:
> Of course you don't, since you're a Dutch man.

That has nothing to do with it.

> If you really had
> been exposed to Hungarian (living many years among them), then
> you would have gradually seen another "PHON" world whatsoever,
> an extreme different one from Dutch, let alone English and (worse)
> Danish.

I haven't lived among Hungarians, have never even been to the country.
But I am familiar with the sound of the language, and I know the
spelling-sound correspondences, since 1973 or so. In recent years,
I hear it even more often, thanks to internet radio and Youtube.

> And you still do not get the gist of my "message": Hungarian <a> is
> always /ɔ/ and nothing else.

As I said, I know that, have known it for decades. I also never
denied it. It seems you keep misunderstanding what I intend
to write. You do know the convention <spelling>, [phones],
/phonemes/, don't you?

> It is no A vowel. /ɔ/ is a variety for /a/ in other languages,

Varieties or allophones would be notated in [ ], not in / /.

> starting with all German dialects and
> Hochdeutsch, with Dutch, Flemish, French (an, am, en, em also con-
> taining variants of /ɔ/ are deemed as variants of /a/), English etc.

Irrelevant.

> I'm merely underscoring the fact that Hungarian <a> /ɔ/ is no
> "child" or "parent" of /a/.

Morphologically, it is. Or better: a lengtened <a> is <á>, not
something like [ɔ:], which doesn't exist.

> It might have been 500-1,000 years
> ago, but not in the latest 300 years or so.
> >I call [ɔ] /a/, because when in grammatical context
> >it lengthens, it becomes /a:/, which sounds as [a:].
> Because you, as a Dutchman, and esp. in the vicinity of the ... Reich
> you are as influenced as you can be.

No, really, you don't understand. Dutch and German have
nothing to do with it. I'm looking at Hungarian, nothing
else. I can temporarily switch off my own language when
I look at a different one.

>> By using /a/ I do not mean it should sound like [a]. It sounds like [ɔ],
>> of course, always.
> In Hungarian there is no /a/ or [a].

So you equate phones and phonemes.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 15, 2022, 10:58:33 AM9/15/22
to
On Thursday, September 15, 2022 at 1:18:05 PM UTC+2, Tim Lang wrote:
> (1) bor- in English would be too /ɔː/ (as in <bore> and <boring>), in
> which case the Hungarian spelling would have been different,
> namely Bortók, and not Bartók. (@ wiki, there is a picture of him
> very young, signed by him with the Hungarian spelling, Bartók).
>
> (2) -oak would be almost automatically interpreted as /oʊk/ and /əʊk/,
> which is not the case in the Hungarian pronunciation for Bartók.
> The <o> here is a mere /ɔː/.

So now you are saying the long pendant of your
<a> /ɔ/ [ɔ]
is
<ó> /ɔ:/ [ɔ:]
I can't believe that.

wugi

unread,
Sep 15, 2022, 11:44:12 AM9/15/22
to
Op 15/09/2022 om 16:58 schreef Ruud Harmsen via Google Groups
<goo...@rudhar.com>:
> On Thursday, September 15, 2022 at 1:18:05 PM UTC+2, Tim Lang wrote:

>> (1) bor- in English would be too /ɔː/ (as in <bore> and <boring>), in
>> which case the Hungarian spelling would have been different,

Not what I meant, not my problem that English has no spelling for [bor]
or [bɔr].

>> namely Bortók, and not Bartók. (@ wiki, there is a picture of him
>> very young, signed by him with the Hungarian spelling, Bartók).
>>
>> (2) -oak would be almost automatically interpreted as /oʊk/ and /əʊk/,

I meant [o:], not my problem that English has no spelling for that. I
was merely using approximate spelling.

>> which is not the case in the Hungarian pronunciation for Bartók.
>> The <o> here is a mere /ɔː/.

You want [o:]

> So now you are saying the long pendant of your
> <a> /ɔ/ [ɔ]
> is
> <ó> /ɔ:/ [ɔ:]
> I can't believe that.

I hear a neat [o:] in googletranslate.
https://translate.google.be/?hl=nl&tab=rT&sl=hu&tl=nl&text=B%C3%A9la%20bart%C3%B3k&op=translate

And the short a is a kind of [a], no [ɔ] here, so then short a can sound
like a. That was my first point.

--
guido wugi



Tim Lang

unread,
Sep 15, 2022, 1:07:00 PM9/15/22
to
On 15.09.2022 16:29, Ruud Harmsen wrote:

>>Skorpió, "Jó lenne, ha szeretnél"
>><https://www.youtube.com/watch?v=lX0as9tuAoY>
>
>An example of musical metre not in accordance with
>the stress pattern of the languages: they sing:
>JÓ lenNE, ha SZERetNÉL
>
>Perhaps that wrecks the proper -nn- in LENne?

It doesn't. Besides, -ne and -nél aren't as stressed as is the
natural stress falling on jó and szer- the same way in this
song as in normal colloquial prose. (And -nél might sound
to you as stressed, but there is only that <é> /e:/ which in
Hungarian in general has to be "full of vigor," and not as
"weak"/"flabby" as it happens to sound in the ending syllable
in other languages.)

Jó lenne here is by no means altered whatsoever by being sung.
In normal speech, it sounds the same way. And without the
exaggeration ... "lennnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn-nnnnnnnne"
which you seemingly miss. :-) Most Hungarians in most imaginable
cases rather pronounce /'læ-næ/ almost with no n-elongation.
(Exaggerate the pronunciation esp. in lenne and many other
words, and the mental or spoken reaction might be: "What's
up, man? Being high? What dope, bruh?" :-))

In "German" it would look like this "Joh län(n)ä, hɔ ßärrättnehl".
(I put the doubling here only to underscore that the vowel not
followed by an <h> must be very short.)

The only peculiarity in the context of "lenne" ("would be") is that
many native speakers prefer "volna" as synonym, alternance. So
that, one often can hear "jó volna" instead of "jó lenne". (Having
the same meaning whatsoever.)

>Cf. https://rudhar.com/musica/ruivelte/en.htm
>
>I recently noticed the same in "Hej cigány", from the
>operetta Marica grófnő (Gräfin Mariza), music by
>Emmerich Kalman (Kálmán Imre), lyrics/libretto by
>Julius Brammer & Alfred Grünwald:
> https://lyricstranslate.com/en/dolhai-attila-hej-cigany-lyrics.html
> https://www.youtube.com/watch?v=eq4ypR94Bqg
> ==
> Mit tudjáTOK ti benn, hogy idekinn mi fáj?

Here again: why underlining "-TOK"? What's peculiar about its
pronunciation?

> Hej, voltam én is egyszer büszke bálkiRÁLY!

And why ... "-RÁLY!"? OTOH, note the colloquial words "benn" and
"(ide)kinn", which mean "bent" (indoors; within; inside of) and
"kint" (outside). This is a good example, since this singing here
is quite slow, the pronunciation clear: you can remark here that,
although written "benn" and "kinn", he pronounces the -nn- quasi
as if there wer only one -n-. And in spite of the fact that the
singer would have here the possibility to enhance the stressing
"mode" signaling the gemination: because he has plenty of time
for uttering properly, in contrast with those quick pieces of
pop-rock (by the Skorpió band).

And, if the Hung. libretto author introduced kinn, benn, fenn,
then he could have chosen instead of standard egyszer (once),
e.g. eccer, which is colloquial vernacular. But in the 20's
and 30's of the 20th century, the authors of such dialogs
were way more conservative than other generations have been
60-100 years later on (up to now). :-)

>Cause: it is a translation from the original German text,

"Cause" for what? To me, both versions (in German and in Hung.)
are compatible. (OTOH, the one libretto writer, Grünwald, might have
had knowledge of Hungarian, since both his parents were
immigrants in Vienna, coming from Budapest. It is plausible
that they spoke Hungarian at home.)

>https://aimsgraz.com/wp-content/uploads/2014/09/K-Komm-Zigany.pdf
>That German text does have a perfect metrum, with music
>and language accents coinciding.

I don't know. Cf. in German
<https://www.youtube.com/watch?v=BoTGZpy1CPI>
<https://www.youtube.com/watch?v=lmh6-S2f1T0>
<https://www.youtube.com/watch?v=01ZnCba2FGA>

I only can say that Dolhai's pronunciation is slightly ... exagerated
for an operette singer. So that some sound groupings in the "output"
sound slightly strange. But only in comparison with other professionals
singing those songs contained in the contessa Maritsa operette (within
the frame of this music genre, which is neither opera, nor musichall,
but which in former Austro-Hungary was extremely popular in the
time period approx. 1860-1970/80; its popularity dwindling afterwards
quite considerably).

But compare yourself - e.g. here an "old school" singer singing the
same "nóta". Some tiny differences can be perceived, despite the
poor sound quality: the old guy has a somewhat more standard,
natural, pronunciation, than the younger singer (Dolhai).

<https://www.youtube.com/watch?v=84UAyjczFfc>

By and large, in none individual rendering am I able to discern
any weird or striking "deviation": each of them are typical Hungarian,
even the exaggerations caused by some kinds of "manerisms" typical
of certain styles.

Tim

Tim Lang

unread,
Sep 15, 2022, 1:44:46 PM9/15/22
to
On 15.09.2022 16:53, Ruud Harmsen wrote:

>I haven't lived among Hungarians, have never even been to the country.
>But I am familiar with the sound of the language, and I know the
>spelling-sound correspondences, since 1973 or so. In recent years,
>I hear it even more often, thanks to internet radio and Youtube.

Yea, fine, but, unfortunately, this ain't enough.

>>And you still do not get the gist of my "message": Hungarian <a> is
>>always /ɔ/ and nothing else.
>
>As I said, I know that, have known it for decades. I also never
>denied it. It seems you keep misunderstanding what I intend
>to write.

What you tried to underline here (by misusing "short a") I understood
in the very 1st posting dealing with it: after all, I myself am
a foreigner when I communicate with them. :-) I only wished to
underline the pronunciation aspect of it, so that neither you or
anyone else would conclude "short" written <a> in Hungarian is
the counterpart of the lone one, written <á>.

>You do know the convention <spelling>, [phones],
>/phonemes/, don't you?

Yes, but here this "tetracapillotomy" based on Prussian rigor of
Korinthenkackerei and stubborn exaggerations of the proper
usage of linguistic terminology, is of no help to 99,9% of
the foreigners who try becoming advanced in learning Hungarian.

>>It is no A vowel. /ɔ/ is a variety for /a/ in other languages,
>
>Varieties or allophones would be notated in [ ], not in / /.

But these two vowels above are not in an allophonic relationship -
in Hungarian. They might be in whole lotta other languages, but
those are different "galaxies". :-)

>Morphologically, it is. Or better: a lengtened <a> is <á>, not
>something like [ɔ:], which doesn't exist.

I know that very well. My opposition is only to the unlucky
wording "short a", since this might prompt many to think (at
least for a while), what's written as <a> in Hungarian were
the same vowel as <a>, say, in German, French, Italian, in
the Iberian languages and most of the Slavic ones as well
as Turkish. But it isn't. (That's all.)

>No, really, you don't understand. Dutch and German have
>nothing to do with it. I'm looking at Hungarian, nothing
>else. I can temporarily switch off my own language when
>I look at a different one.

By these examples I underscore that in these languages we've
got the consummate vowel "chaos", whereas the Hungarian
collection of vowels is one of the best examples for
"rigor" and "discipline". In spite the fact that Hungarian,
as a Uralic idiom has that difficult sh*t, "harmony of the
vowels", which they, as natives, don't apply 100% (IMHO, to
a somewhat lesser extent than I assume that Turkish and
other Finnic and those two Ugric languages do.

In German dialects esp. such as Bavarian, Suebian, Saxon,
the usage and differentiation betw. various types or
allophones of A, ranging to the most genuine kinds of O,
is the consummate chaos. Only native speakers of those
dialects are in command of virtually all utterance possi-
bilities. (A Prof. Dr. Dr. as a linguist, coming as a
native speaker of another, even if neighboring, dialect,
won't be able to learn all possibilities. Exception:
those with "photographic"/"savants'" memory.)

Hence my comparisons, since Hungarian is, in contrast,
a shockingly "unitary" language. Nothing compare in
Europe with that. Perhaps only Serbian/Croatian and
Turkish. (I don't know whether or not Finnish and
Eesti as well.)

Tim

Tim Lang

unread,
Sep 15, 2022, 1:58:17 PM9/15/22
to
On 15.09.2022 16:58, Ruud Harmsen wrote:

>>(2) -oak would be almost automatically interpreted as /oʊk/ and /əʊk/,
>>which is not the case in the Hungarian pronunciation for Bartók.
>>The <o> here is a mere /ɔː/.
>
>So now you are saying the long pendant of your
><a> /ɔ/ [ɔ]
>is

Nope!

><ó> /ɔ:/ [ɔ:]
>I can't believe that.

(In the sentence "The <o> here" (I should have written <ó> of course!
Tja, "Eile mit Weile" :-) "is a mere ɔː". It's an additional
explanation, in order to understand that <ó> in Bartók cannot
become oʊ or əʊ, so that a spelling -oak would result in flawed
reading.)

So I didn't made that statement. This relationship is ubiquitous
in standard German and German dialects, but not in Hungarian.
ɔ and ɔ: in Hungarian are, again, two different vowels. The
short variant of the long <ó> is written ... <o> (with no
diacritical signs). And again: you mixed them up and thought
I stated that, because your basics is your west-Germanic idiom,
as well as other languages in which all kinds of A and of O are
on the same "string". A Hungarian cannot assume something like
that.

Tim

Tim Lang

unread,
Sep 15, 2022, 2:08:18 PM9/15/22
to
On 15.09.2022 17:44, wugi wrote:

>>>(2) -oak would be almost automatically interpreted as /oʊk/ and /əʊk/,
>
>I meant [o:], not my problem that English has no spelling for that. I
>was merely using approximate spelling.

This is why I said a -oak spelling shouldn't be chosen; -awk would be
better (but would also imply problems, since in many "Anglo" regions
the pronunciation is not the desired one.

>I hear a neat [o:] in googletranslate.

That's right. I only use [ɔ:] because of the British usage of IPA
transcription, ie, the o's noted that way are real long o's, unlike
the American ones, that tend to make of it [ɔ] or even [ʌ]. Because
all the Anglos use the letter o, but almost never does it get the
value of a genuine, proper, clear-cut long o, as it is even in this
Bartók. (In Hungarian there is no "libertarian"-"anarchic" freedom
to attach to <ó> or to <o> any kind of other vowel values, as the
mood and the regiolect strikes the mumbler. Hence their spelling
is virtually a "phonetic" or "phonemic" if you prefer, whereas
English spelling might have that too if entire "Anglia" would
adopt - by decree :-) - for instance this system ĐA ALFΛBET:
<https://bilderupload.org/bild/e90465791-da-alfabet>)

Tim

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 15, 2022, 2:48:26 PM9/15/22
to
If Wikipedia is anything to go by,
https://en.wikipedia.org/wiki/Hungarian_phonology#Vowels,
<ó> is slightly higher than <o>, and <a> and <á> are surprisingly
close to other, <a> being only slightly higher and slighty backer.
The main differences between <a> and </a> would then be
rounding and length.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 15, 2022, 3:23:25 PM9/15/22
to
On Thursday, September 15, 2022 at 7:07:00 PM UTC+2, Tim Lang wrote:
> On 15.09.2022 16:29, Ruud Harmsen wrote:
>
> >>Skorpió, "Jó lenne, ha szeretnél"
> >><https://www.youtube.com/watch?v=lX0as9tuAoY>
> >
> >An example of musical metre not in accordance with
> >the stress pattern of the languages: they sing:
> >JÓ lenNE, ha SZERetNÉL
> >
> >Perhaps that wrecks the proper -nn- in LENne?
> It doesn't. Besides, -ne and -nél aren't as stressed as is the
> natural stress falling on jó and szer- the same way in this
> song as in normal colloquial prose. (And -nél might sound
> to you as stressed, but there is only that <é> /e:/ which in
> Hungarian in general has to be "full of vigor," and not as
> "weak"/"flabby" as it happens to sound in the ending syllable
> in other languages.)

No, that is not the point. In the spoken language, such a
long vowel doesn't make the syllable seem stressed. My
point that the meter of the sung has accents in the wrong
places.
Jó lenne, ha szeretnél
In the language, per stress (not length):
-...-..
In the music however
-.-.-.-

> Jó lenne here is by no means altered whatsoever by being sung.

That's right, it's still pronounced the same. But the meter
of the music doesn't properly follow the language.

> In normal speech, it sounds the same way. And without the
> exaggeration ... "lennnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn-nnnnnnnne"
> which you seemingly miss. :-)

I can still hear (now that I already know it) that the
n is 'double', i.e. lasts slightly longer than it otherwise
would.

> >Cf. https://rudhar.com/musica/ruivelte/en.htm
> >
> >I recently noticed the same in "Hej cigány", from the
> >operetta Marica grófnő (Gräfin Mariza), music by
> >Emmerich Kalman (Kálmán Imre), lyrics/libretto by
> >Julius Brammer & Alfred Grünwald:
> > https://lyricstranslate.com/en/dolhai-attila-hej-cigany-lyrics.html
> > https://www.youtube.com/watch?v=eq4ypR94Bqg
> > ==
> > Mit tudjáTOK ti benn, hogy idekinn mi fáj?
> Here again: why underlining "-TOK"? What's peculiar about its
> pronunciation?

Did you listen to the song?

I use uppercase to indicate the accents in the music's meter
(or metre; in the sense of, per Collins, "rhythm in verse; measured,
patterned arrangement of syllables, primarily according to stress
or length). It doesn't coincide with the initial stress of the language,
nor with its short and vowels.

In the ideal lyrics/music combination, they should. In the
German original, this correspondence is perfect. The song
was originally written in German, and later translated to
Hungarian. I cannot by whom.

> > Hej, voltam én is egyszer büszke bálkiRÁLY!
> And why ... "-RÁLY!"?

Again, the accents in the music. Did you listen to the
song at all?

>> Cause: it is a translation from the original German text,
> "Cause" for what?

For the mismatches in musical stress and language stress
in the Hungarian version. In German the coincide.

I don't say the translator did a bad job. It must be quite
difficult to make a singable translation between languages
so radically different as German and Hungarian.

> To me, both versions (in German and in Hung.)
> are compatible. (OTOH, the one libretto writer, Grünwald, might have
> had knowledge of Hungarian, since both his parents were
> immigrants in Vienna, coming from Budapest. It is plausible
> that they spoke Hungarian at home.)

Yes, that occurred to me too.

> >https://aimsgraz.com/wp-content/uploads/2014/09/K-Komm-Zigany.pdf
> >That German text does have a perfect metrum, with music
> >and language accents coinciding.

> I don't know. Cf. in German
> <https://www.youtube.com/watch?v=BoTGZpy1CPI>

There is never a musical accent of an unstressed syllable.

One exception, 1:14, "Nimm deine Geige." The syllable "-ne"
is too strongly stressed.

> <https://www.youtube.com/watch?v=lmh6-S2f1T0>
> <https://www.youtube.com/watch?v=01ZnCba2FGA>

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 15, 2022, 3:31:10 PM9/15/22
to
On Thursday, September 15, 2022 at 7:07:00 PM UTC+2, Tim Lang wrote:
> I only can say that Dolhai's pronunciation is slightly ... exagerated
> for an operette singer. So that some sound groupings in the "output"
> sound slightly strange. But only in comparison with other professionals
> singing those songs contained in the contessa Maritsa operette (within
> the frame of this music genre, which is neither opera, nor musichall,
> but which in former Austro-Hungary was extremely popular in the
> time period approx. 1860-1970/80; its popularity dwindling afterwards
> quite considerably).

My point is not singing. I like it.

The point is that the stresses in the music do not coincide
with those in the music, that some unstressed Hungarian
syllables get stressed by the music. In perfect lyrics, that
should never happen. Or only very rarely.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 15, 2022, 3:38:13 PM9/15/22
to
On Thursday, September 15, 2022 at 7:44:46 PM UTC+2, Tim Lang wrote:
> On 15.09.2022 16:53, Ruud Harmsen wrote:
> >You do know the convention <spelling>, [phones],
> >/phonemes/, don't you?

> Yes, but here this "tetracapillotomy" based on Prussian rigor of
> Korinthenkackerei and stubborn exaggerations of the proper
> usage of linguistic terminology, is of no help to 99,9% of
> the foreigners who try becoming advanced in learning Hungarian.

Teaching is not my aim.

> >>It is no A vowel. /ɔ/ is a variety for /a/ in other languages,
> >
> >Varieties or allophones would be notated in [ ], not in / /.
> But these two vowels above are not in an allophonic relationship -
> in Hungarian. They might be in whole lotta other languages, but
> those are different "galaxies". :-)
> >Morphologically, it is. Or better: a lengtened <a> is <á>, not
> >something like [ɔ:], which doesn't exist.
> I know that very well. My opposition is only to the unlucky
> wording "short a", since this might prompt many to think (at
> least for a while),

OK, let's make that "short and rounded a", cause judging from
the Wikipedia vowel triangle, that is closer to the truth.

> what's written as <a> in Hungarian were
> the same vowel as <a>, say, in German, French, Italian, in
> the Iberian languages and most of the Slavic ones as well
> as Turkish. But it isn't. (That's all.)

We agree on that one.
(The Turkish one may seem similar, but it is backer, and
unrounded. Has to be, or it would be /o/.)

Tim Lang

unread,
Sep 15, 2022, 5:01:44 PM9/15/22
to
On 15.09.2022 21:23, Ruud Harmsen wrote:

>> On 15.09.2022 16:29, Ruud Harmsen wrote:
>>
>>>>Skorpió, "Jó lenne, ha szeretnél"
>>>><https://www.youtube.com/watch?v=lX0as9tuAoY>
>>>
>>>An example of musical metre not in accordance with
>>>the stress pattern of the languages: they sing:
>>>JÓ lenNE, ha SZERetNÉL

Now I get it! You don't mean the language, but the ... music,
the one Emmerich Kálmán invented/composed. You mean the
musical cadence & the like; and whether that the libretto
text fits that cadence. In your opinion the German libretto
fits, whereas the Hungarian one doesn't or is "faulty".

I don't know, and I'm not interested in that, although
I am more accustomed to these melodies as well as to
the musical folklore of that south-east European area,
including the whole connotations of "Komm, Zigan!" and
that song -- which I happened to hear zillions of times
in the last 50-60 years, since this kind of Kálmán
& al. melodies, "hits", is the music of countries
from Germany across Austria, Hungary and Romania up to
Ukraine, where the interest for this musical universe
perhaps cease. But it is not my music, I ain't interested
in it. To me, it is far too kitschy. And it is rather
the music of generations born approx. between 1870-1920.
Which is why after 1970-80-90 this operette world vanishes
even in Austria and Hungary (the epicenter of this musical
specialty). It is a real wonder that Dolhai's generation
still produces artists who love this kind of music.

Yet, I wouldn't dare to assume anything, unless knowing
Kálmáns papers (notes and everything). Only what he
composed is valid, irrespective of what music scientists
will say. And how the libretto's own prosody fits the
music, melodies with all their individual peculiarities.
Here, what's naturally "given" in the individual language
might be compatible or not. Nobody's fault. And any
translation must be then, in fact, a poetical re-creation,
re-invention of a big chunk of the text (metaphors, idiomatic
phrases, avoidance of phonetic occurrences that could
be perceived as "ugly," "incompatible" or so). Which is
already seen in the simplest thing which is "Komm,
Zigan" (in fact in German it is not Tsigan, but Zigeuner
/tsigoina/), which means, "come on" or "come here, Gypsy
dude" (ie, the violin soloist, usually the boss of the
band, called in Hungarian "prímás"), whereas the Hungarian
version says "Hej, cigány ..." ie, "hey, Gypsy," and not
for instance "Gyere ide, cigány" (come here) or something
like that. When even such petty things, that are perfectly
translateable, then one can imagine ... BTW, there is
another work by K., where in the German title the Slovenian
Varasdin (Varaždin) is the main venue, whereas in its
Hungarian version they talk of ... Kolozsvár (in German
Klausenburg, in Romanian Cluj; a former capital of the
big province of Transylvania).

Besides, it's a known fact that many directors, esp. when
they themselves are stars, not always respect the composer's
notes/papers, and add their own stuff = "interpretations"
for the show they're responsible of. This also tremendously
influence the rendition of the melodies, songs etc. But all
this is quite irrelevant in analysing linguistic peculiarities
of languages (here Hungarian and German).

But in the case of the pop-rock band Skorpió, you might
interview them, and ask them what they intended in cadencing
which words in which way, that you don't agree with, e.g.
in the refrain line "Jó lenne, ha szeret-szeretnél". Pointing
out, you're interested in some aspects from the viewpoints
of musique theories, harmonies & the like. The composers and
singers in the clip are now approx 68-70; and they sang these
songs of the seventies in recent versions, as old guys, after
2000-2010. Several clips can also be accessed @ YouTube. As
for Kálmán's operettes, I would obtain copies of the original
notes for those songs. Only they are relevant; the arrangements
by one director or another, the version sung by a star or
another (with their "spleens") are rather "footnotes", IMHO. :-)

Tim

Tim Lang

unread,
Sep 15, 2022, 5:07:47 PM9/15/22
to
On 15.09.2022 21:29, Ruud Harmsen wrote:

>My point is not singing. I like it.
>
>The point is that the stresses in the music do not coincide
>with those in the music, that some unstressed Hungarian
>syllables get stressed by the music. In perfect lyrics, that
>should never happen. Or only very rarely.

Why does that bother you after all? Such "distorsions" in
poetry but especially in texts that are sung are ... common
occurrences, "poetic licenses" & the like. Compromise solutions.
Since many languages have whole lotta peculiarities that make
the text-melody compatibility difficult or very difficult.
Does Dutch better fit various kind of music? Or at least the
genres opera and operette.

Tim

Tim Lang

unread,
Sep 15, 2022, 5:57:00 PM9/15/22
to
On 15.09.2022 20:48, Ruud Harmsen wrote:

>If Wikipedia is anything to go by,
>https://en.wikipedia.org/wiki/Hungarian_phonology#Vowels,
><ó> is slightly higher than <o>,

Your point being? The difference between them is so tiny
that only native linguist's ears can discern them; and
those of people exposed to them for 30-40-50-60 years.

I myself, although being capable of pronouncing in Hungarian
as any native, although I am a foreigner, cannot distinguish
the difference shown in that darned graph. Only this one thing,
which is also relevant: the long <ó> easily tend to
turn to a diphtongue [ou]. Especially in several regional
dialects and in sociolects. Esp. in jó "good/well" and
Jóska, József ("Joe, Joseph"): it happens zillion times
that they rather sound like jou, Jouska, Jouzsef.

>and <a> and <á> are surprisingly close to other,

This assertion in turn is an exaggeration: they cannot be as
close as shown in the diagram. In order to utter a perfect
<a> a bare opening of the mouth will do so that a perfect
<a> vowel (that is: ɔ) is generated. But in order to issue
the other one, namely <á>, everybody has to open his/her
mouth in a quite radical way, ie, to generate a considerable
distance between the jaw and the palate, and to force
the throat apparatus to ... co-work. The sound which is
then invariably heard is a "stark" one, then is like a
genuine ʌ, as as e.g. in the Italian "basta, calzature,
Caltanisetta" or in French Camargue, but a long one, as
in Hochdeutsch Maat, Frage, Base, Malerei, Mahlzeit etc.

But with the only difference that this sound must be ... long.
(As in the examples above, in German: Maat ...)

><a> being only slightly higher and slighty backer.
>The main differences between <a> and </a> would then be
>rounding and length.

Such fine details cannot be thoroughly perceived/understood based only
on reading printed descriptions. Real conversations with natives
from all walks of life is necessary - zillions of hours, that
give the brain the necessary additional new neuronal links, incl.
those who prompt the mouth to issue the right or almost correct
sounds. :) BTW: the exclamation ɔɔɔɔɔɔɔɔɔɔɔɔɔɔɔɔɔɔ! usually
means disappointment, where as ʌʌʌʌʌʌʌʌʌʌʌʌʌʌʌʌʌʌʌʌʌ! either
a good surprise or the "heureka" moment: "I gotcha!" (This
is valid in several Eur. languages.)

jó écakát, aludj jól
Tim

PS: écakát: colloqu. for the correct spelling éjszakát.


Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 16, 2022, 5:09:41 AM9/16/22
to
I didn't mix up anything. And West-Germanic idiom, or any other
idiom, has nothing to do with it. I look at Hungarian from an IPA
perspective, disregarding other languages.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 16, 2022, 5:17:17 AM9/16/22
to
I agree that it is kitsch. My parents liked it. (1925-1989, 1922-2014).
Perhaps for that reason, as an adolescent, I did not like it.
But in recent years (I am now 67, born 1955), an appreciation
in hindsight is growing.
https://rudhar.com/musica/oratamen.htm

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 16, 2022, 5:38:18 AM9/16/22
to
On Thursday, September 15, 2022 at 11:57:00 PM UTC+2, Tim Lang wrote:
> On 15.09.2022 20:48, Ruud Harmsen wrote:
> >If Wikipedia is anything to go by,
> >https://en.wikipedia.org/wiki/Hungarian_phonology#Vowels,
> ><ó> is slightly higher than <o>,
> Your point being?

That <ó> = [o:] = /o:/ (in IPA, disregarding English, German,
or whatever other language), and not [ɔ:] or /ɔ:/, as you stated
for a moment, but you later explained that you didn't
really mean that. Likewise <o> = [o] = /o/.

> The difference between them is so tiny
> that only native linguist's ears can discern them; and
> those of people exposed to them for 30-40-50-60 years.

True. So in Hungarian, the difference between <o> and <ó>
is merely and essentially length. Like between <e> and <é>
it is length and height, and between <a> and <á> length
and lip rounding.

> I myself, although being capable of pronouncing in Hungarian
> as any native, although I am a foreigner, cannot distinguish
> the difference shown in that darned graph. Only this one thing,
> which is also relevant: the long <ó> easily tend to
> turn to a diphtongue [ou]. Especially in several regional
> dialects and in sociolects. Esp. in jó "good/well" and
> Jóska, József ("Joe, Joseph"): it happens zillion times
> that they rather sound like jou, Jouska, Jouzsef.
> >and <a> and <á> are surprisingly close to other,
> This assertion in turn is an exaggeration: they cannot be as
> close as shown in the diagram.

Close to each other in terms of height (open/close vowels),
and of backness (front/central/back), as that is the only
thing such diagrams show. But length and lip rounding,
essential here for Hungarian, are not shown in the diagram.
But they follow from the symbols used.

> In order to utter a perfect
> <a> a bare opening of the mouth will do so that a perfect
> <a> vowel (that is: ɔ) is generated.

Opening it, and rounding the lips. And it can be [ɔ] (half open)
or almost [ɒ] open, or something somewhere in between.
Confer https://en.wikipedia.org/wiki/International_Phonetic_Alphabet#Vowels

>But in order to issue
> the other one, namely <á>, everybody has to open his/her
> mouth in a quite radical way, ie, to generate a considerable
> distance between the jaw and the palate, and to force
> the throat apparatus to ... co-work. The sound which is
> then invariably heard is a "stark" one,

Yes, <á> is an open vowel, which leaves room for lots
of air passing.

>then is like a
> genuine ʌ, as as e.g. in the Italian "basta, calzature,
> Caltanisetta" or in French Camargue,

Yes, maybe, not sure. [ʌ] is officially an unrounded [ɔ],
but the symbol is also often used, especially in the
context of English, for somewhat more central vowels.
Your mileage may vary.

but a long one, as
> in Hochdeutsch Maat, Frage, Base, Malerei, Mahlzeit etc.
>
> But with the only difference that this sound must be ... long.

Yes. Long, and unrounded.

> (As in the examples above, in German: Maat ...)
> ><a> being only slightly higher and slighty backer.
> >The main differences between <a> and </a> would then be
> >rounding and length.
> Such fine details cannot be thoroughly perceived/understood based only
> on reading printed descriptions.

The rounding is clearly visible in numerous Youtube videos.
And I don't think they are fine details. I now think rounding
and length are the essential factors for the difference between
Hungarian <a> and <á>, even though the degree of openness
and backness also often varies between them.

> Real conversations with natives
> from all walks of life is necessary - zillions of hours, that
> give the brain the necessary additional new neuronal links, incl.
> those who prompt the mouth to issue the right or almost correct
> sounds.

That's right. I can discern Hungarian vowel, but attempts
to reproduce them, at normal speed and in context,
often fail. Knowing and doing are two different things.
That is why children endlessly practice the doing, without
caring about knowing. Adults tend to do that the wrong
way round.

Tim Lang

unread,
Sep 16, 2022, 6:02:10 AM9/16/22
to
On 16.09.2022 11:09, Ruud Harmsen wrote:

>I didn't mix up anything. And West-Germanic idiom, or any other
>idiom, has nothing to do with it. I look at Hungarian from an IPA
>perspective, disregarding other languages.

It has: because any of us has a peculiar perception and
individual possibilities in pronouncing the -theoretically-
same sound. Yet the real output will always be (strikingly!)
different in the ears of natives (as well as in those of
numerous ... "onlookers"). All depending of our environment
(socialization & esp. the 1st language).

The same sounds noted with the same sign in the average
IPA lists are - in reality - linked completely different pro-
nunciations by people from different language environments.

Even, e.g., whenever Low German speaking native Germans
from the neighboring areas, say, between Aachen and the
Eifel region in the South and Münsterland and Friesland
in the North of that area, whenever going to NL or BE,
as tourists/campers or shopping, wouldn't be able to correctly
utter the same IPA noted sounds as any Dutchman or Fleming
pronounce them; and in accordance with the specific regional
dialect in NL and BE. (I myself was "witness" to this IMHO
linguistically highly interesting phenomenon, when I lived
in the vicinity of Venlo and Heerlen for a short while in
the seventies; and, BTW, when I used to listen to Radio
Hilversum /hilversüm/.) Then one can imagine how this can
be, whenever a Hungarian or a Russian or a Greek, tries
imitating the real Dutch sounds.

Average IPA descriptions are too poor to show us those
"collateral" aspects that in the end make the same sound
to sound different whenever uttered by people coming from
different European or world regions (i.e., linguistic
environments, which means => years of training/conditioning
of the individual brain and the articulatory tools).

Thus, your own æ a ʌ ɒ ɔ ɔː o o: cannot sound exactly as
the Hungarian ones, only based on reading IPA lists and
listening to some old-fashioned Kálmán and Lehár (from a
world that disappeared 80-100 years ago). Unless you, as
an individual, are a highly endowed with the talent of
a "master imitator/impersonator".

Tim

Tim Lang

unread,
Sep 16, 2022, 6:28:41 AM9/16/22
to
On 16.09.2022 11:17, Ruud Harmsen wrote:

>I agree that it is kitsch. My parents liked it. (1925-1989, 1922-2014).

A natural phenomenon: the ... generations. Different generations
have different preferences.

In Hungary and Austria (to a much lesser extent in Slovakia, Romania,
Serbia, Croatia, perhaps Slovenia too) this genre ("Operette") is
still popular, to some extent, even today, because in the century
roughly between 1850-1950, this kind of music (and especially of
such composers as Lehár and Kálmán, kinda "champions" of this
genre) was and almost is a ... national popular music in those
countries. Some of the libretto contents ie themes even refer to
national patriotic issues in those countries. So, other types
of "collateral" emotions are involved. As in various other cultural
and musical worlds of Europe, e.g. in canzonetta of Italy, the
French typical chansons or the Fados in Portugal etc.

>Perhaps for that reason, as an adolescent, I did not like it.

However, it is amazing that you, as a Dutch man in NL, know of
this musical cultural world (typical of Austria and East-South-
East Europe)! Even in Germany, by and large (although most of
these operettes are in German and have been interpreted by
several generations of "superstars" in the German-speaking world)
this genre cease to be as popular as it is in Vienna and Austria
plus Hungary.

OTOH, upon neglecting any other aspect (cultural, genre, kitsch etc),
anyone should acknowledge that many songs of the operette are
really nice or even beautiful => "evergreens" (and "Ohrwürmer", as
the saying goes in German). And despite the attitude of the
"superior" genre, of opera, that always has derided and mocked
the ... "li'l sister". :-)

To conclude the thread jokingly: this operette behaves like the
Hungarian ɔ (or if you prefer) ɒ. Ie, neither opera, nor pop music
(of the "hit parade" kind), and neither folk music or folklore
(although exactly that part, "Hej, cigány!", strongly evokes
the folklore link, ie, the Hung. folklore music played by
professional or amateur rural bands, made chiefly of Gypsies.
Who have been as popular in those E-Eur countries as the Yiddish
klezmer bands in the same areas. (In all areas with strong
Yiddish communities, the "hej, cigány" bands also could play
most of the Yiddish tunes, esp. in Slovakia, Transylvania,
Galicia, Moldavia, Ukraine and I assume in former Southern
Poland as well. Note that certain tunes for folkl. dances are
called in Ukraine, Belarus, Russia ... "vengerka" => "Hungarian
dance"). So, these asp. are reflected by Kálmán's "Gräfin Maritsa,"
and it is this musical context of folklore bands who've enter-
tained on many occasions, esp. in parties which determines
the song "Hej, cigány", showing the coming along of the "primás",
the Gypsy boss and 1st violinist of the band playing music for
the people of the "upper crust". (The main characters of the
play have nobility ranks at least similar to the Brit. earls
or even dukes.) The venue of the story is in the rural area
(hence the the folklore music band) near Cluj a.k.a. Kolozsvár
a.k.a. Klausenburg, the former capital of a former Hungary's
Eastern province, Transylvania. Ie, the real one, and not the
fake province imagined for his horror story, Dracula, by the
Irishman Bram Stoker.)

Tim

Ruud Harmsen

unread,
Sep 16, 2022, 7:45:28 AM9/16/22
to
Fri, 16 Sep 2022 12:02:05 +0200: Tim Lang <m...@privacy.net> scribeva:

>On 16.09.2022 11:09, Ruud Harmsen wrote:
>
> >I didn't mix up anything. And West-Germanic idiom, or any other
> >idiom, has nothing to do with it. I look at Hungarian from an IPA
> >perspective, disregarding other languages.
>
>It has: because any of us has a peculiar perception and
>individual possibilities in pronouncing the -theoretically-
>same sound. Yet the real output will always be (strikingly!)
>different in the ears of natives (as well as in those of
>numerous ... "onlookers"). All depending of our environment
>(socialization & esp. the 1st language).

Yes. But I successfully taught myself to switch that off. If hearing,
anyway, in a less failsafe manner also when pronouncing. Slowly, that
is.

>The same sounds noted with the same sign in the average
>IPA lists are - in reality - linked completely different pro-
>nunciations by people from different language environments.

There is that tendency, as a result of the good principle to use the
least marked symbol as necessary, for phonemes, that is, in a
non-narrow description, ignoring everything that is immaterial in the
language under consideration.

But the differences aren't very great in practice, and aren't so many.

>Even, e.g., whenever Low German speaking native Germans
>from the neighboring areas, say, between Aachen and the
>Eifel region in the South and Münsterland and Friesland
>in the North of that area, whenever going to NL or BE,
>as tourists/campers or shopping, wouldn't be able to correctly
>utter the same IPA noted sounds as any Dutchman or Fleming
>pronounce them; and in accordance with the specific regional
>dialect in NL and BE.

The finer details can be difficult, yes.

> (I myself was "witness" to this IMHO
>linguistically highly interesting phenomenon, when I lived
>in the vicinity of Venlo and Heerlen for a short while in
>the seventies; and, BTW, when I used to listen to Radio
>Hilversum /hilversüm/.)

-söm, acutally. Or usually reduced to a shwa.

>Then one can imagine how this can
>be, whenever a Hungarian or a Russian or a Greek, tries
>imitating the real Dutch sounds.

Yes, that is very difficult, in any direction. Some people do a
reasonably good job, some less so.

>Average IPA descriptions are too poor to show us those
>"collateral" aspects that in the end make the same sound
>to sound different whenever uttered by people coming from
>different European or world regions (i.e., linguistic
>environments, which means => years of training/conditioning
>of the individual brain and the articulatory tools).

With all the IPA symbols available, plus all the diacriticals, we can
come a long way. But sometimes it is so hard that specialists don't
agree, like in the case of the Danish <-d>.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 16, 2022, 7:58:35 AM9/16/22
to
On Friday, September 16, 2022 at 12:02:10 PM UTC+2, Tim Lang wrote:
> Thus, your own æ a ʌ ɒ ɔ ɔː o o: cannot sound exactly as
> the Hungarian ones, only based on reading IPA lists and
> listening to some old-fashioned Kálmán and Lehár (from a
> world that disappeared 80-100 years ago).

I also hear radio presenters talking now.

Of course youŕe right, doing it exactly like a native remains
difficult. But then the internal variation comes into play:
no two native speakers speak exactly alike, and at no
two different events will the same native speaker say it
exactly the same.

>Unless you, as
> an individual, are a highly endowed with the talent of
> a "master imitator/impersonator".

I hear an awful lot more than the average person, and
to some extent, I can also do abnormal things. But only
slowly and highly concentrated, not in running speech.

For example, if I really try, I can say a long [ɑ], then turn
rounding on and off to see (no, hear) what it does. In
other words, alternate between [ɑ] and [ɒ].

Likewise, I can say [u::] and turn rounding off, to see if
that results in that Turkish vowel, 'dotless i'.

Or say [i::] and add rounding, but NOT also move the
vowel back a little (as everybody does do in Dutch,
German, Turkish, Hungarian), in order to get a Swedish
or Norwegian <y>.

Etc. etc.

Tim Lang

unread,
Sep 16, 2022, 8:54:31 AM9/16/22
to
On 16.09.2022 11:38, Ruud Harmsen wrote:

>but you later explained that you didn't really mean that.

Having been in a hurry, I didn't see I had forgotten to put the
´ accent on the <o>.

>Close to each other in terms of height (open/close vowels),
>and of backness (front/central/back), as that is the only
>thing such diagrams show.

Yes of course. And these positions are precious information though.

>But length and lip rounding, essential here for Hungarian, are not shown in the diagram.
> But they follow from the symbols used.

Also: the really real positions of the vowels, front, central, back,
open, close are not always shown correctly in the diagram.

THIS is why I made the comparisons with the quality of /a/ and /o/
in German and in neighboring languages, knowing that you'd instantly
know what I was referring to --- in order to point out the essential
differences in Hungarian. So, the Hungarian o, both short and long,
are way much in a distance to ɔ, and I pointed out in this respect
that standard (radio, TV) German makes of such a written <o> always
an audible ... ɔ. Even in the French name Macron. Hungarians will
never do that, their /o/ is more a real /o/; even if some persons
would be physically (in the mouth area handicapped).

In this respect, several idioms, esp. in the North-West Europe, have
pronunciations of the same sounds, as they were all ... patiens,
in contrast with nations from other, southern or eastern areas,
whose pronunciations are as though after 10 years of schooling.

To a Hungarian, this aspect combined with some of the "worst" spellings
in the world (e.g. English, Irish, Portuguese, but also Danish and
even Dutch) is a ... nightmare. Or the ö's and ü's, incl. the long
variants of them, in Hungarian: they have the same, constant, value.
In contrast with western languages that also have abundant occurrence
of these vowels: they even have 1-2-3 kinds of these vowels, whereas
Hungarians, irrespective of regional dialects, only one kind. (Here
only the length matters.)

>Opening it, and rounding the lips. And it can be [ɔ] (half open)
>or almost [ɒ] open, or something somewhere in between.

In Hungarian there is no rounding of the lips in order to be able
to pronounce correct ɔ, ɒ - or even an exaggeration (by few dialect
speakers and ... foreigners), i.e. when the ɔ almost sounds like
an [a]. The opposite happens when uttering long or short [o]: here,
the rounding of the lips reaches maximal levels (second to none).
Hence my comparisons that obviously have annoyed you: because in
other, esp. western languages, many natives mean to say [o] and
even long [o]'s but their output sounds like an ... [ɔ].

> Yes, <á> is an open vowel, which leaves room for lots
> of air passing.
>
>> then is like a
>> genuine ʌ, as as e.g. in the Italian "basta, calzature,
>> Caltanisetta" or in French Camargue,
>
>Yes, maybe, not sure.

But _I_ am sure: I am in command of Hungarian pronunciation
as any Hungarian native, Viktor Orbán included.


[ʌ] is officially an unrounded [ɔ],
> but the symbol is also often used, especially in the
> context of English, for somewhat more central vowels.
> Your mileage may vary.

Of course - since the Hungarian ɔ is clearer even if compared
with the extreme English pronunciation of the exaggerating
speakers/anchorpeople of the BBC world service or of Queen
Elizabeth's family (incl. Charles III and his sons). Otherwise,
almost the entire Engl.-speaking world's ɔ either sounds closer
to [o] or, americanized, extremely like AAAAAAAAAAAA
("wʌʌʌʌʌʌʌt, Jʌʌʌʌʌʌʌhn, Tʌʌʌm" esp. in the Midwest; and
in the South the diphtong ɑɪ tends to be uttered ʌʌʌ,
which is why in joking spelling "I" and "my" are written
"Ah", "mah".)

Pay attention to the tradition/rule that all diagrams showing
vowel positions for various languages always refer to a
_standard_ language, spoken as such only by minorities
of the "stiff upper lip" society as well as by some radio
and theater specially schooled professionals, in order to
put on display the best "gift of the gab". But over 90%-98%
of the natives don't speak respecting that "ad litteram" what
you read in those diagrams. Among those nations even the
Germans (both of Germany, and of Austria plus Switzerland),
have peculiar pronunciations, caused by the regional dialect,
in many cases even after years and years of special
"logopaedic" training. And German speakers aren't the worst
in this respect within the frame of North-Western Europe.
But in comparison, other nations (even some that otherwise
are poor and "laggard") have natural pronunciations so that
even the "working class" and hobos could read texts on the
mike @ national radio or TV stations. Which in many of
the economically and socially developed nations would be
a sensation.

>Yes. Long, and unrounded.

Of course unrounded: who the heck is able to pronounce a
genuine /a/ in a rounded way? (Perhaps God himself and his
emissaries Mike and Gabriel. :-))

BTW: I remember now a phenomenon typical of a dialect region
of Hungarian, namely in Eastern Transylvania, in the region
of the so-called Szeklers. In their Hungarian numerous words
with <á> are pronounced with ɒ ɔ or ɒ if you prefer this IPA
sign. Due to this, that kind of speech sounds weird, if those
natives use their own dialect and not the standard Hungarian
language. This is the opposite extreme to another extreme,
I already mentioned, somewhere in central Hungary, where
natives make of ɒ/ɔ that resembles an <á> (with strong
opening and without rounding).

But these you won't find described in the diagram, since
it only shows general rules of the standard language. You'll
find some of the "non-conformist" aspects only in pages
(such as at Wiki) dealing with certain dialects/regiolects.

>The rounding is clearly visible in numerous Youtube videos.
>And I don't think they are fine details.

However, it's very positive and useful that you've paid
attention to the rounding. This optical detail gives
an additional confirmation to the rounding degree of
the certain sounds on a "scale" between the vowels <a> /ʌ/
and <o>. Where Hungarian has only one kind of each extant
vowel, whereas whole lotta other languages, esp. in the
North-West (comprising incl. all English dialects) have
some kind of pronunciation "anarchy". The same is valid for
German dialects and standard Hochdeutsch, despite the fact
that "Prussian discipline" is "owned" by the German nation. :-)

>I now think rounding
>and length are the essential factors for the difference between
>Hungarian <a> and <á>, even though the degree of openness
>and backness also often varies between them.

Yay! (Na endlich! Holt das "gold'ne Buch" zur feierlichen
Eintragung! :-))

>That is why children endlessly practice the doing, without
>caring about knowing. Adults tend to do that the wrong
>way round.

That's right! (I myself learnt Hung. pronunciation by
interaction with native children. Even some grammar aspects
concerning the "vowel harmony" or the "agglutination" of
certain components bearing grammar and having vowels
that must be in harmony with the lemma/root vowel. It's
like a miracle: the child's brain can do that, without
knowing what the heck might be a noun, an adjective, an
adverb, the declension, the tenses, prepositions etc
As grown-ups, most people lose this amazing cappacity.
But there are exceptions too, esp. when there is a
flexibility and talent second to none in one person
so that the brain activity, incl. steering the talking
"contraption", and the entire process of mimesis, result
in an output that more and more sounds like the one
produced by natives. But such talented and skilled
contemporaries are "rara avis".)

Tim

Tim Lang

unread,
Sep 16, 2022, 9:06:39 AM9/16/22
to
On 16.09.2022 13:45, Ruud Harmsen wrote:

>>Hilversum /hilversüm/.)
>
>-söm, acutally. Or usually reduced to a shwa.

Now then (-: this is a "Probe aufs Exempel" which I deliver
myself: I always had thaught I was hearing hilversÜm, instead of
hilversÖm. (Perhaps also due to the circumstance that
I thought the <u> in Dutch always would have to be
read [ü], as in Utrecht.)

>With all the IPA symbols available, plus all the diacriticals, we can
>come a long way.

Yes, of course. I agree with that, by and large, although
my insistences might prompt the reader think I'd despise
or deride API-IPA.

>But sometimes it is so hard that specialists don't
>agree, like in the case of the Danish <-d>.

Methinks, descriptions of this kind/interest would be
a tad better if the authors wouldn't limit their
remarks/conclusions to the standard language (which
in numerous cases is deemed by natives as being an
... "artificial" language or version of it. This is
the case in the German-language speaking country too.
"Hochdeutsch" is deemed by all as an "artificial"
or a "paper" German ("papierenes Deutsch").

Tim

Tim Lang

unread,
Sep 16, 2022, 9:28:45 AM9/16/22
to
On 16.09.2022 13:58, Ruud Harmsen wrote:

>Of course youŕe right, doing it exactly like a native remains
>difficult. But then the internal variation comes into play:
>no two native speakers speak exactly alike, and at no
>two different events will the same native speaker say it
>exactly the same.

Yes, this is true. But the cases where individual kinds of
pronunciation are OK are irrelevant in our thread. I mean
only the cases where, in spite of individuality, one certain
kind of sound must be produced by all/everybody among the
natives. And, of course, to pay attention to the dialectal
peculiarities, that might heavily influence the pronunciation
of a native speaker.

>I hear an awful lot more than the average person, and
>to some extent, I can also do abnormal things. But only
>slowly and highly concentrated, not in running speech.

I already assume that as soon as I read a few of your posts.

>Likewise, I can say [u::] and turn rounding off, to see if
>that results in that Turkish vowel, 'dotless i'.

Oh yeah, the vowel/s [ɨ ʉ ɯ]. As for Turkey's Turkish
and several other Turkic languages, this sound has a ...
complication: often (when pronounced by radio speakers
or anchormen) it rather sounds rather like this. But
even the genuine one doesn't sound exactly as its
Russian and Romanian counterparts. (IMHO, the Romanian
one, IPA-noted as [ɨ] seems to be the "purest" one.
In order to compare it with the conterparts in Mongolian,
Korean, Mandarin, Japanese, I'd have to hear how it's
pronounced in those languages. I know only that the
Japanese one is noted with this inverted m: [ɯ].
Anyway, this type of vowel is extremely difficult to
pronounce by most Europeans (who mostly misinterprete
it as being sort of an [ɪ].)

> Or say [i::] and add rounding, but NOT also move the

To utter a proper variant of this, one should avoid
any rounding of the lips. As soon as the rounding is
involved, the vowel gets .... {u]-ish, ie, [ʊ]-ish.
So that the sound is no longer [ɨ ʉ ɯ]-like. You
can vary it only by "moving" it back and forth, by
working with the muscle tissues of the tongue root
and the throat, the mouth apperture being kept with
very small opening (without rounding), almost as
when uttering the vowels [ɪ, ə, e], yet not as in
[u and o], since these two (with any variation of
them) already must have a certain rounding of the
lips.

>in order to get a Swedish or Norwegian <y>.

How does this one sound? Very or slightly different
from the average ... "tyske" (German) <ü>? (BTW:
ü and ö are as well quite nonutterable without any
rounding.)

Tim

Christian Weisgerber

unread,
Sep 16, 2022, 9:30:07 AM9/16/22
to
On 2022-09-14, Tim Lang <m...@privacy.net> wrote:

>>Mitteldeutsch, Central German, it seems. Speyer, Worms, etc. Googling,
>
> It is a wrong information, although real scholars have proved
> 100-120 years ago that Yiddish is a South German dialect, but
> not a Suebian-Alemanian one (Speyer, Worms) in the first place

Whut?! I'm here right in the middle between Speyer and Worms, and
the local speech is not Alemannic in any shape or form. It's Rhenish
Franconian, specifically Palatinate German.

--
Christian "naddy" Weisgerber na...@mips.inka.de

Christian Weisgerber

unread,
Sep 16, 2022, 9:30:07 AM9/16/22
to
On 2022-09-13, Tim Lang <m...@privacy.net> wrote:

> Yes, but in the native Hungarian perception and pronunciation the
> <a> is never an /a/ or /ʌ/, it is always, i.e., without any
> exception whatsoever, an /ɔ; ɒ/. So, in the strict sense of
> the wording "short a" the usage of it is ... wrong: because the
> Hungarian vowel simply written <a> is no /a/ at all. It might
> be perceived as an /a/ in other languages, incl. in German and
> Dutch.

Nobody questions that Hungarian <a> is realized as a rounded back
vowel, [ɒ] or [ɔ].

Yet, denying any connection between <á> and <a> seems unrealistic.

You seem to claim that Hungarians themselves do not associate <a>
with "short a", despite the orthographic and morphological connection.
I'd be more convinced if you could point us to loanwords where
foreign [a] was replaced by Hungarian /aː/ because /ɒ/ is perceived
as the wrong vowel.

This one is not in your favor:
https://hu.wikipedia.org/wiki/Glasznoszty

Tim Lang

unread,
Sep 16, 2022, 10:17:17 AM9/16/22
to
On 16.09.2022 14:54, Christian Weisgerber wrote:

>Whut?! I'm here right in the middle between Speyer and Worms, and
>the local speech is not Alemannic in any shape or form.

Sorry, mea culpa; I had forgotten the "Speyerer Linie",
in the area where the -pp- pronunciation area is
separated from the -pf- pronunciation area (in the
South of it), e.g. Appel versus Apfel. But Speyer
is extremely close to the Suebian-Alemanian area.
(I tend to always take into consideration the
"Benrather Linie" only. :-))

>It's Rhenish Franconian, specifically Palatinate German.

By and large these kinds of German dialects are also
"Southern", although Rhine-Frankish, Mosela-Frankisch
and Cologne Ripuarian Frankish already get various
elements that are typical for the Northern area of
Low German dialects ("Niedersäksesch"), with "wat,
dat, et, Appel" and the like.

Which means that Yiddish dialects have even less in
common with those areas linguistically than some
go on assuming.

Tim
--
BTW: Sagt man bei Euch noch "önch" für "euch"? Oder
ist das ein Mundartmerkmal eher typisch für Eifel,
Kölle, Oche und Düren? Wie dem auch sei, Jiddisch
sagt "enk", genauso wie im echten Altbairischen.

Tim Lang

unread,
Sep 16, 2022, 10:39:38 AM9/16/22
to
On 16.09.2022 15:05, Christian Weisgerber wrote:

>Yet, denying any connection between <á> and <a> seems unrealistic.

In their pronunciation of the vowel - not in the alternation
with <á> in plurals and in other aspects of grammar (declensions,
conjugations etc).

>You seem to claim that Hungarians themselves do not associate <a>
>with "short a",

It is not a short a. We have short a in German, both in
Hochdeutsch and our dialects, but today's Hungarian doesn't
have that. The written <a> is never an <a>. It is not the
same as the <a> in German dialects, where, especially in
the Vienna - Munich area you'll hear the extremest <o>'s
frequently, the natives thinking they'd be saying genuine
<a>'s.

>I'd be more convinced if you could point us to loanwords where
>foreign [a] was replaced by Hungarian /aː/ because /ɒ/ is perceived

There are quite numerous examples, but I'd have to "nachdenken"
a lot, because they aren't so stored in my memory as to be
retrieved spontaneously. :-)

>as the wrong vowel.

Though one spontaneous example:

coffee: in Hungarian it is kávé ['kaaaaa-veeeee; kahweh].
And accordingly cafe (café) is kávéház, and colloquially-
slangy kávézó. (A *kové or *kavé haven't been invented
yet.)

This in spite of the fact that the produce is called
coffea (arabica).

>This one is not in your favor:
>https://hu.wikipedia.org/wiki/Glasznoszty

This is an unappropriate example, since the guy doesn't
pronounce Glasnost' according to the Hungarian spelling,
but he pronunces it "pa russki" = in an evident ... Russian
way.

Even the ending -ty is not exactly as the typical Hungarian
one, but typical of Russian pronunciation. Hence his
rendering 'Glaaaaaß -noßthhhj.

Tim


Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 16, 2022, 12:06:02 PM9/16/22
to
On Friday, September 16, 2022 at 3:06:39 PM UTC+2, Tim Lang wrote:
> On 16.09.2022 13:45, Ruud Harmsen wrote:
>
> >>Hilversum /hilversüm/.)
> >
> >-söm, acutally. Or usually reduced to a shwa.
> Now then (-: this is a "Probe aufs Exempel" which I deliver
> myself: I always had thaught I was hearing hilversÜm, instead of
> hilversÖm. (Perhaps also due to the circumstance that
> I thought the <u> in Dutch always would have to be
> read [ü], as in Utrecht.)

<u> in an open syllable sounds as German short <ü>, in a closed
syllable it is like a short (but high) <ö>. Except when r follows,
then it's long, as in 'duur' (mean Dauer), where <uu> is doubled
because the syllable is closed. Utrecht is supposed to have the
syllable structure u-trecht, not ut-recht. In Hil-ver-sum, the u is
in a closed syllable, because the m follows.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 16, 2022, 12:12:18 PM9/16/22
to
Wikipedia has some nice samples:
https://en.wikipedia.org/wiki/Swedish_phonology#Vowels

Dingbat

unread,
Sep 16, 2022, 4:31:29 PM9/16/22
to
On Sunday, September 11, 2022 at 8:30:06 AM UTC-7, Christian Weisgerber wrote:
> On 2022-09-10, S K <skpf...@gmail.com> wrote:
>
> > Subject: Do the geminate spellings in present day English indicate
> > English had the corresponding geminate sounds in the past?
>
> Only some instances of double consonant spellings in Modern English
> can be traced back to geminates in Old English, e.g. in "will".
>
> Old English had phonemic geminate consonants. IIRC, the vowel
> before those was always short. When the geminates were lost and
> short vowels in open syllables lengthened, the double consonant
> spelling became associated with short vowels and has been inconsistently
> used to mark them, even when there was no etymological geminate,
> as for instance in the morphological spelling rule "trap - trapped".
>
> (More or less the same happened in German, where the principle was
> applied more consistently across the orthography.)
>
> There were also Old English geminates whose modern reflexes are not
> spelled with a doubled consonant, at least -dg- as in "bridge",
> "hedge".
>
> Latin had phonemic geminates, but I think double consonant spellings
> in English loans from Latin have always been purely orthographic.
>
> Western Romance lost the Latin geminates early on, so double
> consonants in French words--including those borrowed into English--are
> purely in imitation of Latin spelling.
>
In Italian Capella, the gemination in the suffix comes from Latin.
How much of the gemination in Italian is NOT of Latin origin and
how did it get those geminates? From a substrate? By sound change?

Dingbat

unread,
Sep 16, 2022, 4:47:53 PM9/16/22
to
On Monday, September 12, 2022 at 3:01:01 AM UTC-7, wugi wrote:
> Op 12/09/2022 om 9:05 schreef Ruud Harmsen:
> > Sun, 11 Sep 2022 14:45:45 -0700 (PDT): S K <skpf...@gmail.com>
> > scribeva:
> >> Way before I knew the remotest thing discussed here I heard an Indian say "Cutty Sark" and though "something is off with that"
> >>
> >> When Raj Brits heard "pukka Sahib" guess they didn't hear the gemination in Hindi.
> >>
> >> How times have changed - Goras connected with Chess these days pronounce Rameshbabu Praggnanandhaa more or less close to the original.
> >>
An Indian shortening of his name is Pragg. It's most unlikely
that they'd know how to pronounce Pragg like an Indian; they're
likely to rhyme it with Tragg (the police lieutenant in Perry Mason
stories).
> >>
> >> Wonder if Standard Indian English has rules for gemination.
> >
> > If you want to hear real and consistent gemination, listen to
> > Hungarian.
> Italian will do.
>
I once read that a waiter cited a diner as ordering Penis Pasta,
for failing to geminate the nasal in Penne.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 17, 2022, 2:25:29 AM9/17/22
to
> On 16.09.2022 11:38, Ruud Harmsen wrote:
>> Opening it, and rounding the lips. And it can be [ɔ] (half open)
>> or almost [ɒ] open, or something somewhere in between.

On Friday, September 16, 2022 at 2:54:31 PM UTC+2, Tim Lang wrote:
> In Hungarian there is no rounding of the lips in order to be able
> to pronounce correct ɔ, ɒ -

That is a contradiction. If Hungarian <a> were unrounded, it would
be or [ʌ] or [ɑ]. But it's not, it's [ɔ] or [ɒ] (or somewhere in between),
I think on that much we agree.

> or even an exaggeration (by few dialect
> speakers and ... foreigners), i.e. when the ɔ almost sounds like
> an [a]. [...]

[...]

>> Yes. Long, and unrounded.
>
> Of course unrounded: who the heck is able to pronounce a
> genuine /a/ in a rounded way? (Perhaps God himself and his
> emissaries Mike and Gabriel. :-))

Ask any speaker of Farsi, Dari or Tadjik, they do it all the time.
Their historically long <a:>. Second and third vowel in 'Afganistan'.
https://nl.wikipedia.org/wiki/Afghanistan
Afghanistan (Pasjtoe / Dari: افغانستان; Pasjtoe: Afġānistān,
IPA: [avɣɒnisˈtɒn]; Dari: Afġānestān, IPA: [avɣɒnesˈtɒn]),

So do many Brits when saying <not> or <dot>:
https://www.collinsdictionary.com/dictionary/english/not
https://www.collinsdictionary.com/dictionary/english/dot
Pronunciation: [nɒt] and [dɒt]. So they should sound very
similar to Hungarian <nat> and <dat> (if such words existed,
I think not).

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 17, 2022, 3:08:43 AM9/17/22
to
>> On 16.09.2022 11:38, Ruud Harmsen wrote:
>>> Opening it, and rounding the lips. And it can be [ɔ] (half open)
>>> or almost [ɒ] open, or something somewhere in between.

> On Friday, September 16, 2022 at 2:54:31 PM UTC+2, Tim Lang wrote:
>> In Hungarian there is no rounding of the lips in order to be able
>> to pronounce correct ɔ, ɒ -
>> [...]

On Saturday, September 17, 2022 at 8:25:29 AM UTC+2, Ruud Harmsen via Google Groups <goo...@rudhar.com> wrote:
> That is a contradiction. If Hungarian <a> were unrounded, it would
> be or [ʌ] or [ɑ]. But it's not, it's [ɔ] or [ɒ] (or somewhere in between),
> I think on that much we agree.

The lip rounding is clearly visible in several Youtube videos. Well,
rounding, it's more like, or also, a protruding, moving the lips
slightly forward. That make a resonance "tube" in the mouth
somewhat longer (like 3 to 5 mm), which then lowers one of the
formant frequencies of the vowel, giving [ɒ] its extra 'dark' quality,
in comparison with an unrounded [ɑ].

Examples:
Korpás Éva in this interview
https://www.youtube.com/watch?v=d1cqu2B1KCc
The question (of which of course I hardly understand a word)
lasts very long, Éva starts talking only after over 4 minutes:
https://www.youtube.com/watch?v=d1cqu2B1KCc&t=4m18s
Note: Youtube offers the possibility to slow down to 75, 50 or
25% of the original speed. Doesn't sound pretty, but the vowels
are still recognizable, and the lip positions can be very well
observed.

Also in the long monologues in her recent online singing lessons:
https://www.youtube.com/watch?v=VWQOsEwqmhQ
https://www.youtube.com/watch?v=wceP0CcVQOY
https://www.youtube.com/watch?v=kSW94BK2Zbw

Pál Eszter in https://www.youtube.com/watch?v=kSW94BK2Zbw
And several other videos in which she sings with the band of
(I think he is) her husband Pál István (search: Szalonna és bandája).

And Magos Zenekar, singer Enyedi Ágnes:
https://www.youtube.com/watch?v=kSW94BK2Zbw
From about 2m10s there are nice close-ups. Sometimes the microphone
is in the way, but often it is not.

And Bognár Szilvia in the same interview series as mentioned
above: https://www.youtube.com/watch?v=oYPNJtIFmBk .

Ruud Harmsen

unread,
Sep 17, 2022, 3:17:02 AM9/17/22
to
Sat, 17 Sep 2022 00:08:42 -0700 (PDT): "Ruud Harmsen via Google Groups
<goo...@rudhar.com>" <goo...@rudhar.com> scribeva:

>And Magos Zenekar, singer Enyedi Ágnes:
>https://www.youtube.com/watch?v=3DkSW94BK2Zbw

Clear example: at about 3:51, what I hear as "csuhaja". I don't know
the language, so it might also be chúhalya or whatever. What matters
now is the too 'short' <a>'s, which in the song become very long,
without changing vowel quality, of course.

You see the tongue moving inside for making the [j] sound, during
which the rounding or protruding of the lips is also less than for the
two adjacent vowels. Direct link:
https://www.youtube.com/watch?v=3DkSW94BK2Zbw&t=3m50s

Ruud Harmsen

unread,
Sep 17, 2022, 3:22:43 AM9/17/22
to
Sat, 17 Sep 2022 09:16:59 +0200: Ruud Harmsen <r...@rudhar.com>
scribeva:

>>And Magos Zenekar, singer Enyedi Ágnes:
>>https://www.youtube.com/watch?v=3DkSW94BK2Zbw

Sorry, wrong link, should be:
https://www.youtube.com/watch?v=nWzi9tGSkLA

>Clear example: at about 3:51, what I hear as "csuhaja". I don't know
>the language, so it might also be chúhalya or whatever. What matters
>now is the too 'short' <a>'s, which in the song become very long,
>without changing vowel quality, of course.
>
>You see the tongue moving inside for making the [j] sound, during
>which the rounding or protruding of the lips is also less than for the
>two adjacent vowels. Direct link:
>https://www.youtube.com/watch?v=3DkSW94BK2Zbw&t=3m50s

Sorry, wrong link, should be:
https://www.youtube.com/watch?v=nWzi9tGSkLA&t=3m50s

And it's csuhajja, his/her mouth, Google Translate tells me.

Ruud Harmsen

unread,
Sep 17, 2022, 3:58:11 AM9/17/22
to
Sat, 17 Sep 2022 00:08:42 -0700 (PDT): "Ruud Harmsen via Google Groups
<goo...@rudhar.com>" <goo...@rudhar.com> scribeva:

>>> On 16.09.2022 11:38, Ruud Harmsen wrote:
>>>> Opening it, and rounding the lips. And it can be [?] (half open)
>>>> or almost [?] open, or something somewhere in between.
>
>> On Friday, September 16, 2022 at 2:54:31 PM UTC+2, Tim Lang wrote:
>>> In Hungarian there is no rounding of the lips in order to be able
>>> to pronounce correct ?, ? -
>>> [...]
>
>On Saturday, September 17, 2022 at 8:25:29 AM UTC+2, Ruud Harmsen via Google Groups <goo...@rudhar.com> wrote:
>> That is a contradiction. If Hungarian <a> were unrounded, it would
>> be or [?] or [?]. But it's not, it's [?] or [?] (or somewhere in between),
>> I think on that much we agree.
>
>The lip rounding is clearly visible in several Youtube videos. Well,
>rounding, it's more like, or also, a protruding, moving the lips
>slightly forward. That make a resonance "tube" in the mouth
>somewhat longer (like 3 to 5 mm), which then lowers one of the
>formant frequencies of the vowel, giving [?] its extra 'dark' quality,
>in comparison with an unrounded [?].
>
>Examples:
>Korpás Éva in this interview
>https://www.youtube.com/watch?v=d1cqu2B1KCc
>The question (of which of course I hardly understand a word)
>lasts very long, Éva starts talking only after over 4 minutes:
>https://www.youtube.com/watch?v=d1cqu2B1KCc&t=4m18s
>Note: Youtube offers the possibility to slow down to 75, 50 or
>25% of the original speed. Doesn't sound pretty, but the vowels
>are still recognizable, and the lip positions can be very well
>observed.
>
>Also in the long monologues in her recent online singing lessons:
>https://www.youtube.com/watch?v=VWQOsEwqmhQ
>https://www.youtube.com/watch?v=wceP0CcVQOY
>https://www.youtube.com/watch?v=kSW94BK2Zbw
>
>Pál Eszter in https://www.youtube.com/watch?v=kSW94BK2Zbw
>And several other videos in which she sings with the band of
>(I think he is) her husband Pál István (search: Szalonna és bandája).
>
>And Magos Zenekar, singer Enyedi Ágnes:
>https://www.youtube.com/watch?v=kSW94BK2Zbw
>From about 2m10s there are nice close-ups. Sometimes the microphone
>is in the way, but often it is not.
>
>And Bognár Szilvia in the same interview series as mentioned
>above: https://www.youtube.com/watch?v=oYPNJtIFmBk .

Tim Lang

unread,
Sep 17, 2022, 4:01:59 AM9/17/22
to
On 17.09.2022 08:25, Ruud Harmsen wrote:

>That is a contradiction. If Hungarian <a> were unrounded, it would
>be or [ʌ] or [ɑ].

I knew that you'd react. :-) But the Hungarian variant of
this vowel can and always is pronounced only keeping
the mouth aperture opend, without giving any impulse to
the lips. This is why you can't grasp the difference of
heir [ɔ] and e.g. yours or the English one. (Moreover,
the Hungarian way of pronouncing it has also influenced
neighboring languages, e.g. Romanian dialects spoken in
Transylvania and Banate.)

But it would suffice that one native show you - for seconds -
how it's done, and you'd be able yourself to pronounce
this "alternative" kinda [ɔ] (or the inverted ɑ would
fit better, since ɔ in too many idioms is actually to
close to ... [o].

>But it's not, it's [ɔ] or [ɒ] (or somewhere in between),
>I think on that much we agree.

Of course, somewhre, "in the middle", between a genuine [o]
and a genuine [ʌ] or ... "neither nor".

> Ask any speaker of Farsi, Dari or Tadjik, they do it all the time.
> Their historically long <a:>. Second and third vowel in 'Afganistan'.

Yes, but to my own ears, the ɔ in their ɛ-rɔɔɔn and afhɔnistɔɔɔn
sounds like the Hungarian one, indeed, but is closer to o, ie,
the Hungarian ɔ is remoter. (Seemingly, the same happens
comparing Swedish-Danish å with Hungarian <a> [ɒ].)

>https://nl.wikipedia.org/wiki/Afghanistan
>Afghanistan (Pasjtoe / Dari: افغانستان; Pasjtoe: Afġānistān,
>IPA: [avɣɒnisˈtɒn]; Dari:

Oh, I like this sign, ɣ; I should have used it myself.

Afġānestān, IPA: [avɣɒnesˈtɒn]

This is a good hint. Farsi's and Dari's ɒ really sounds
much like the Hungarian ɒ, even if there is a slight
difference though (the Persian/Iranian variant is a bit
o-isher than the "Puszta"-ɒ).

You gotta listen to the native speakers (to more than one
native speaker; and from various provinces). In any
language. In order to get the real idea of the real
pronunciations, of which IPA-API lists select only
the "most important" variants.

> So do many Brits when saying <not> or <dot>:
> https://www.collinsdictionary.com/dictionary/english/not
> https://www.collinsdictionary.com/dictionary/english/dot
> Pronunciation: [nɒt] and [dɒt]. So they should sound very
> similar to Hungarian <nat> and <dat> (if such words existed,
> I think not).

Yes! I myself pointed this out (at least 2-3 times). I
even mentioned "BBC-WS English" and "Queen's English".
Within the frame of this "received" high-brow stiff
upper lip English of the upper crust of England, these
words are pronounced with the vowel that sounds (99,x%)
as the Hungarian (standard) counterpart. E.g. the word
hot in this kind of BritEnglish sounds as the Hungarian
word hat meaning "six". But in Hungarian there is no
"liberalism" in allowing the native to give their ɒ
"oscillations", ie, unlike the esp. American one, hʌʌʌt
or hɑːɑːɑːt, or the opposite exaggeration, the "dark"
o or ɔː of some Bavarian pronunciations, including the
Viennese one: hot (hooot). (Dea hooooot xoookt: Der hat
gesagt.)

Tim

Ruud Harmsen

unread,
Sep 17, 2022, 4:06:09 AM9/17/22
to
Sat, 17 Sep 2022 00:08:42 -0700 (PDT): "Ruud Harmsen via Google Groups
<goo...@rudhar.com>" <goo...@rudhar.com> scribeva:
No, that's:
https://www.youtube.com/watch?v=YAWy8F6Q3Yg

>And several other videos in which she sings with the band of
>(I think he is) her husband Pál István (search: Szalonna és bandája).

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 17, 2022, 4:30:23 AM9/17/22
to
On Saturday, September 17, 2022 at 10:01:59 AM UTC+2, Tim Lang wrote:
> On 17.09.2022 08:25, Ruud Harmsen wrote:
>> That is a contradiction. If Hungarian <a> were unrounded, it would
>> be or [ʌ] or [ɑ].
> I knew that you'd react. :-) But the Hungarian variant of
> this vowel can and always is pronounced only keeping
> the mouth aperture opened,

I never said it wasn't. Hungarian <a> is open or half-open, so
the lips are also relatively open, even if rounded and protruded.
They must be, because they are attached to the jaw, and so is
the tongue mass.

High rounded vowels like [o] [u] have a smaller lip opening than
opener rounded vowels. That's only natural.

> without giving any impulse to the lips.

I don't know what you mean by "impulse", it's not a phonetic
term. If you mean protruding, moving forward: the many
videos show otherwise.

> This is why you can't grasp the difference of
> [t]heir [ɔ] and e.g. yours or the English one.

A few lines later in this same message, you are contradicting
yourself again:

> these
> words are pronounced with the vowel that sounds (99,x%)
> as the Hungarian (standard) counterpart. E.g. the word
> hot in this kind of BritEnglish sounds as the Hungarian
> word hat meaning "six".

> (Moreover,
> the Hungarian way of pronouncing it has also influenced
> neighboring languages, e.g. Romanian dialects spoken in
> Transylvania and Banate.)
>
> But it would suffice that one native show you - for seconds -
> how it's done,

You apparently missed all the links to the videos. (OK, some
were wrong, somehow Youtube, or Firefox, or Agent sometimes
mess them up. Or maybe I did.)

> and you'd be able yourself to pronounce
> this "alternative" kinda [ɔ] (or the inverted ɑ would
> fit better, since ɔ in too many idioms is actually to
> close to ... [o].

Yeah right.

How do you know I can't make a proper [ɒ]? You never
heard me speak.

> >But it's not, it's [ɔ] or [ɒ] (or somewhere in between),
> >I think on that much we agree.
> Of course, somewhre, "in the middle", between a genuine [o]
> and a genuine [ʌ] or ... "neither nor".

[ʌ] is unrounded, back, half open.
[o] is rounded, back, half close.

So in between that would be half rounded.

> > Ask any speaker of Farsi, Dari or Tadjik, they do it all the time.
> > Their historically long <a:>. Second and third vowel in 'Afganistan'.
> Yes, but to my own ears, the ɔ in their ɛ-rɔɔɔn and afhɔnistɔɔɔn
> sounds like the Hungarian one, indeed, but is closer to o, ie,
> the Hungarian ɔ is remoter.

Yes, lower, opener, in other words, closer to cardinal [ɒ] than
[ɔ]. Which is also what the Wikipedia diagram shows.

> You gotta listen to the native speakers (to more than one
> native speaker; and from various provinces). In any
> language. In order to get the real idea of the real
> pronunciations, of which IPA-API lists select only
> the "most important" variants.

You really think I didn't?

Except for the provinces, I admit I know nothing about
regional Hungarian pronunciation, to my ears the language
is 100% uniform.

> > So do many Brits when saying <not> or <dot>:
> > https://www.collinsdictionary.com/dictionary/english/not
> > https://www.collinsdictionary.com/dictionary/english/dot
> > Pronunciation: [nɒt] and [dɒt]. So they should sound very
> > similar to Hungarian <nat> and <dat> (if such words existed,
> > I think not).
> Yes! I myself pointed this out (at least 2-3 times). I
> even mentioned "BBC-WS English" and "Queen's English".
> Within the frame of this "received" high-brow stiff
> upper lip English of the upper crust of England, these
> words are pronounced with the vowel that sounds (99,x%)
> as the Hungarian (standard) counterpart. E.g. the word
> hot in this kind of BritEnglish sounds as the Hungarian
> word hat meaning "six".

See above. You denied it there.

Tim Lang

unread,
Sep 17, 2022, 4:47:54 AM9/17/22
to
On 17.09.2022 09:08, Ruud Harmsen wrote:

>The lip rounding is clearly visible in several Youtube videos. Well,

This can't be. Look yourself at the mouth of Hungarians and tell
them to slowly say "a, az, jaj, hat, Magyar, anya, apja, akar,
kanyar, hamar, kaja". And stay face to face and watch closely.

BTW: <a> versus <á>: kanyar (curve) vs. surname Kányádi.
And hamar (soon; immediate) vs. ha már (if after all; if anyway).

Then prompt them say words with real <o> in them, e.g. "komondor,
Moson, komoly, golyó": here you'll get the max. rounding (describing
a small ... o-circle). This kind of [o], long or short, is
the the other opposite extreme to Hung - the other one
(in the frame of ... Scylla and Carybdis :-)) being the genuine
clear [a], as in German "ach, Krach, Arschbacke")

The Hungarian difference _ɒ versus O and A_ respectively is
perhaps fit for the 1st place in a top of EuroAsian idioms.
To grasp that, many a European (incl. Dutch, British, German)
need intense "drill" of the ear, brain and the masticatory
"engine" to get it. IPA descriptions won't be enough.
(And watching Hungarians, as shown in those clips, but
when talking normally and quickly, would be of too little
help. A native should show you slowly and in an axaggerated
manner. Until you get the difference, then, if you wanna
imitate, drill-exercise-Übung.)

(Or else: by "rounding" you might mean something totally different.)

>rounding, it's more like, or also, a protruding,

"Protruding" at that? :-O Oh, no - that'll be used only
for .... [o], [u], [ö], [ü] purposes, both short and long.

Ask any Hungarian to pronounce "jaj" and "juj". And watch
his/her mouth/lips when repeating these ubiquitous Hung.
interjections. Then you'd have your "revelation".

And for the protrusion: "gyönyörü" (wonderful). Here, when
there were no protrusion, then you'll get *gyanyara,
*gyenyere, *gyinyiri and even *gyányárá. :-)

>moving the lips slightly forward.

Nope! The Hungarian "short a" is pronounced as though the lips
and the entire jawl muscular parts were ... paralysed. Even
people with the typical palsy damages e.g. after apoplexia
can pronounce a perfect Hungarian ɒ.

I understand why you can't ... understand: because your mind
is "programmed" to associate all kind of ɒ with the big family
of O. In this respect, the Hungarian variant is indeed out-
standing => radical. And acc. to my own opinion is the most
genuine ɒ in Eurasia, since it is placed in a position which
is the remotest when compared with all possibilities that
are in reality closer to ... O. Something that can't be
reflected by the IPA collection of letters/symbols.

>That make a resonance "tube" in the mouth
>somewhat longer (like 3 to 5 mm), which then lowers one of the
>formant frequencies of the vowel, giving [ɒ] its extra 'dark' quality,
>in comparison with an unrounded [ɑ].

From another "angle": the Hungarian ɒ is a bit closer to ... a,
than various ɒ's in other European idioms (in which ɒ has
more "o-contamination" and this is why the lips also are
involved to help getting the correct pronunciation).

>Examples:
>Korpás Éva in this interview

But you'll never observe lips forming more and more kinda "O"-circle
whenever pronouncing the Hungarian ɒ, called "short a". Beyond
any explanation: natives and "advanced" foreigners don't need
any rounding for pronouncing this vowel.

Think of keeping the mouth opened for uttering an unemphatic [a].
In this position, without any movement of the lips whatsoever
try pronouncing ɒ, only by means of the inner configuration,
lingual, laryngeal and vocal chords. It's only a matter of
"teacher" showing you once, than for you some "drill"/"training"
time, than you'll succeed.

>Szalonna

This is a good word for practicing ɒ in alternation with the
genuine [o], in order to audio-mental differentiation of them;
which difference in Hungarian is at a maximum level as compared
with many other Eur. languages that always make approximations
(and where nobody is bothered by occasional "mixing up".

(The word means "lard, bacon", is a Slavic loanie; kinship:
Slovak slanina and Polish słonina.)

Or ask your Hung. Interlingua colleague to show you how to
pronounce "hajrá, Magyarok!" (sort of a "battle" shouting).

Tim

Tim Lang

unread,
Sep 17, 2022, 4:56:17 AM9/17/22
to
On 17.09.2022 09:16, Ruud Harmsen wrote:

>> And Magos Zenekar

<a> in Magos ("seedy") and zenekar (band; orchestra) is the same ɒ.

> Clear example: at about 3:51, what I hear as "csuhaja". I don't know
> the language, so it might also be chúhalya or whatever. What matters
> now is the too 'short' <a>'s, which in the song become very long,
> without changing vowel quality, of course.

Of course it is a short wovel! That I myself reiterated
zillion times. I only pointed out that the wovel in
reality is no real A at all. I.e., in a "microscopic"
analysis phonetically, allophonically etc. I only
said, "short a" is a misleading wording, since students
of this idiom would be prompted to deem this wovel as
an A, and by using the common IPA sign ɔ, the readers
might think of a closeness to ... O; hence I admitted
that we'd better use ɒ (which seems to have been
popular in Europe, for the Engl. ph. transcriptions,
already since the seventies).

Tim



Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 17, 2022, 5:30:16 AM9/17/22
to
On Saturday, September 17, 2022 at 9:22:43 AM UTC+2, Ruud Harmsen wrote:
> >https://www.youtube.com/watch?v=3DkSW94BK2Zbw&t=3m50s
> Sorry, wrong link, should be:
> https://www.youtube.com/watch?v=nWzi9tGSkLA&t=3m50s
>
> And it's csuhajja, his/her mouth, Google Translate tells me.

Here are the lyrics:
https://folkradio.hu/mediatar/felvetelek/1822/banat-banat-keseruseg-meg-a-tesver-is-ellenseg-csuhajja

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 17, 2022, 5:58:09 AM9/17/22
to
On Saturday, September 17, 2022 at 10:47:54 AM UTC+2, Tim Lang wrote:
> (Or else: by "rounding" you might mean something totally different.)
> >rounding, it's more like, or also, a protruding,
> "Protruding" at that? :-O Oh, no - that'll be used only
> for .... [o], [u], [ö], [ü] purposes, both short and long.

The mouth and lips are less opened for close rounded vowels,
than for open vowels rounded vowels, yes.

> Ask any Hungarian to pronounce "jaj" and "juj". And watch
> his/her mouth/lips when repeating these ubiquitous Hung.
> interjections. Then you'd have your "revelation".

All of that happens numerous times in the videos I watched.

> And for the protrusion: "gyönyörü" (wonderful). Here, when
> there were no protrusion, then you'll get *gyanyara,
> *gyenyere, *gyinyiri and even *gyányárá. :-)
> >moving the lips slightly forward.
> Nope!

I SEE it in numerous videos. And I hear the effect.
But you deny it.
I give up.
It makes no sense to go like this.

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 17, 2022, 6:11:13 AM9/17/22
to
On Saturday, September 17, 2022 at 10:47:54 AM UTC+2, Tim Lang wrote:
> I understand why you can't ... understand: because your mind
> is "programmed" to associate all kind of ɒ with the big family
> of O. In this respect, the Hungarian variant is indeed out-
> standing => radical. And acc. to my own opinion is the most
> genuine ɒ in Eurasia, since it is placed in a position which
> is the remotest when compared with all possibilities that
> are in reality closer to ... O. Something that can't be
> reflected by the IPA collection of letters/symbols.

Sorry, this is nonsense.

I am not programmed, because I unprogrammed myself, already
since over 50 years ago.

And IPA describes it perfectly (albeit it with symbols I didn't
know back then, or maybe they even didn't exist; I learnt them
maybe 10 or 15 years ago.) The IPA symbol [ɒ] perfectlt describes
the Hungarian vowel. It is in line with I see AND hear.

Cf. https://jbdowse.com/ipa/, a complete IPA chart, with sound samples.

> But you'll never observe lips forming more and more kinda "O"-circle
> whenever pronouncing the Hungarian ɒ, called "short a". Beyond
> any explanation: natives and "advanced" foreigners don't need
> any rounding for pronouncing this vowel.

High rounded vowels have a smaller mouth opening than lower
rounded vowels, yes. But that doesn't mean they're not rounded.

> Think of keeping the mouth opened for uttering an unemphatic [a].

Central or front? This symbol is ambiguous in IPA.

> In this position, without any movement of the lips whatsoever
> try pronouncing ɒ, only by means of the inner configuration,
> lingual, laryngeal and vocal chords. It's only a matter of
> "teacher" showing you once, than for you some "drill"/"training"
> time, than you'll succeed.

I can already do that -- but I start with a back unrounded ɑ, of course,
not a central or front vowel.

> Or ask your Hung. Interlingua colleague to show you how to
> pronounce "hajrá, Magyarok!" (sort of a "battle" shouting).

There's no need. I already know. I have heard a lot of Hungarian
in my life. I like the sounds of the language, even if I hardly understand
anything.

Tim Lang

unread,
Sep 17, 2022, 8:24:20 AM9/17/22
to
On 17.09.2022 10:30, Ruud Harmsen wrote:

>I never said it wasn't. Hungarian <a> is open or half-open, so
>the lips are also relatively open,

This is it = the correct perception! ("Ez az!") That is, less open
than if you wish to say <á> (as if sitting in the ... dentist's
"shop" :-)).

>even if rounded and protruded.

But there ain't any need whatsoever to round and protruded
the lips for the Hungarian <a>. (Theoretically, only when
you choose to bite into a sour lemon before performing the
<a> utterance. :-))

>I don't know what you mean by "impulse",

The brain => via (facial) nerves => to those peripheral
areas that ... execute the motions that cause the correct
or incorrect sounds.

>it's not a phonetic term.

I ain't here in a séance of a central or steering body
of a totalitarian party, so I dare ... disobey.

In what I'm talking about (this friggin' Hung. <a>) even
the annoying insistence on using "phonemic" and/or
"phonetic" and the darned brackets are of less (read: no)
importance. What is important for "outsiders" is to
hear the real sound and (if they're interested) to
see the motion of the ... anatomy that generates the
appropriate sound (which is not the sound you already
know from Dutch, English, Portuguese and German).
(My mistake: I thought you are interested exactly
in these aspects; and it was because that you listened
to various fragments of tv shows in Hungarian, even as
a newbie.)

>If you mean protruding, moving forward: the many
>videos show otherwise.

Then you perceive them with a ... delay, thus seeing
other vowels and not the <a> one.
>
>> This is why you can't grasp the difference of
>> [t]heir [ɔ] and e.g. yours or the English one.
>
>A few lines later in this same message, you are contradicting
>yourself again:

How on earth do I contradict myself, when I explain that
the Hungarian vowel for <a> is exactly like the English
version of that sound pronounced by a minority of the
English people from England (incl. the Windsor and
Mountbatten family; and some of the BBC-World Service
speakers (especially a few decades earlier, since in
recent times, more and more people there talking live
have regional, and even US Midwestern pronunciations).

This means, most of the English dialects/pronunciations
have different kinds of this vowel, as compared with
Hungarian, but a few English people utter the same vowel.
And attention: in most European countries in English
classes it is this vowel that is taught: the other
variants that are ubiquitous in the entire English
language world, such European students hear and learn
only in contact with native speakers from Britain,
Ireland, USA, Canada, Australia, New Zealand.

It must be the [ɔ] sign that constantly misleads you.
I use it because this is the one I was taught. The
"better" one (that I assume would better fit the
Hungarian <a>, namely, [ɑ] have been more and more
used than the other in recent times.

I could've used instead the American sign for this, [ȯ],
e.g. as in the word "loss" (pronounced like @ Merriam-
Webster's, online). But I'm not at all accustomed to
using [ȯ].

>You apparently missed all the links to the videos.

I don't need to do that: I can myself proncounce any
Hungarian word from "shit" to "God" like any native.
I've communicated/interacted with hundreds and hundreds
of natives from various Hungarian-speaking regions. I
know what I am talking about. Moreover, I can offer
comparisons, with the help of other languages, which
most of the persons you see in thos clips aren't capable
of. Especially since most Hungarians, when speaking
other languages have terrible accents, exactly because
they don't perceive sounds correctly, and especially
since almost all Hungarians do not HEAR that other
languages have real "short a"'s, ie genuine [a], so
that it is a mistake to pronounce them as they were
Hungarian <a>'s. (This is how you can detect out of
100 Hungarians talking in other idioms approx 98
as Hungarians. Only the rest is so "talented", that
you cannot tell their origin, ie, you'll be misled.)

>were wrong, somehow Youtube, or Firefox, or Agent sometimes
>mess them up. Or maybe I did.)

Nothing went wrong: I just don't need watching that.

>How do you know I can't make a proper [ɒ]? You never
>heard me speak.

Because, if you can, it'd mean that you 100% understand
what I am talking about, so all my comments were
superfluous, and you'd never put "rounding" and "protruding"
in the context of this Hungarian type of "neither-A-or-O".

> [ʌ] is unrounded, back, half open.
> [o] is rounded, back, half close.
>
>So in between that would be half rounded.

With the difference that in Hungarian you can issue a perfect
variant of this vowel with absolutely no need to round anything.
Other languages having this weird vowel do feel like needing
a bilabial rounding and protruding, and of course think the
vowel is kind of an [o] (hence the old sign that looks like
an inverted C, in order to represent one half of a circle).

>Yes, lower, opener, in other words, closer to cardinal [ɒ] than
>[ɔ].

Yes, I'll use [ɒ] from now on and cease writing [ɔ], if this
is the problem bothering you.

>Which is also what the Wikipedia diagram shows.

The Wiki diagram doesn't use the inverted C at all; it uses
the inverted cursive a only. (Well, this is new to me, as
I said, because the inverted cursive a is a newer option
than the way older inverted C.)

<https://upload.wikimedia.org/wikipedia/commons/6/62/Hungarian_vowel_chart_with_rounded_short_a.svg>

Okay. But, again, I'd criticise this diagram (which is also
cited in the article in Hungarian about the same subject,
"vowels"), because the distance between [ɒ] and [a:]
is the same as the distance between the to o's,
[o] and [o:]. But there is a major difference caused
by the aperture: the mouth must be widely opened for
a real [a:], whereas both kinds of [o] (long and short)
the lips and mouth aperture don't change positions.

So, only the simple three-tired diagram showing only
the position of the vowels is a ... shortcoming, because
you, as an "onlooker" from another phonemic and phonetic
galaxy will forever miss other important features that
are essential for rendering the "stuff" correctly. You'll
forever end hearing "Hilversüm" instead of "Hilversöm". :-)

OTOH, of course, these diagrams will do for any foreign student,
since nobody will wish to adapt in such a masterful way,
as to get, one day, such an exactness, as that of trained
natives, in order to act as an actor or to declamatorily
speak (or solemnly read gov't communiques), as s/he were
no foreigner at all.

So, if the outsider pronounces as he can, the natives
will understand. So you can pronounce for "az a szép
akinek a szeme kék" this way too: "oz o sijp o kijnaak
o saamaa kijk". Hungarians might think they hear a
South African or a Chinese, but they'd understand. :-)
Even more, if as a foreigner you completely replace
the "short a" by systematically using <á> instead.
For mutual intelligibility, it wouldn't be OK.

Tim



Tim Lang

unread,
Sep 17, 2022, 8:49:55 AM9/17/22
to
On 17.09.2022 12:11, Ruud Harmsen wrote:

>And IPA describes it perfectly

Yes, of course. Yet the IPA will never show you the "dark side
of the moon". I could show it to you, but you don't need this
and now I understand you don't wanna see it. My mistake.

>The IPA symbol [ɒ] perfectlt describes the Hungarian vowel.

Of course. And now it's clear to me that I should have used
it exclusivele, and the inverted C not at all. (Hungarian
or foreign scholars should have invented a third sign for
this, since Hungarian [ɒ] is extremely different from other
variants - provided that one has "real ears".

>High rounded vowels have a smaller mouth opening than lower
>rounded vowels, yes. But that doesn't mean they're not rounded.

OK, your own obervations and opinion.

>>Think of keeping the mouth opened for uttering an unemphatic [a].
>
>Central or front? This symbol is ambiguous in IPA.

The position might be that in the diagrams 100%. What matters
is the mouth opening: depending on it (and slightly on some
muscular tissue: tongue and throat) this a would differ. I myself
can produce at least 3-4 types of [a] and at least 2 [a] keeping
the lips (with no rounding and pouting whatsoever) and the
mouth opened in the same positions. Many Hungarians unless
they don't learn a second and a third language won't be aware
themselves of how many variants these vowels A and O ("alpha
and omega") can have (in Europe; let alone in "exotic" languages
of Asia, Africa and "native" Americas).

>> Or ask your Hung. Interlingua colleague to show you how to
>> pronounce "hajrá, Magyarok!" (sort of a "battle" shouting).
>
>There's no need. I already know. I have heard a lot of Hungarian
>in my life. I like the sounds of the language, even if I hardly understand
>anything.

OK, it might be so. But I'm highly skeptical, because of this
additional information: "I hardly understand anything". It
would rather make sense if you wrote "I understand 95% to 98%",
which would imply getting the "melody" of the sentences depending
on ... moods (sorrow, joy, wrath, annoyance etc, including the
implications of such "primeval" interjections as "jaj" and "juj",
which "comprise" like in a "nutshell" whole complexities of
feelings and psyche).

Tim


Christian Weisgerber

unread,
Sep 17, 2022, 10:30:07 AM9/17/22
to
On 2022-09-16, Dingbat <ranjit_...@yahoo.com> wrote:

> In Italian Capella, the gemination in the suffix comes from Latin.
> How much of the gemination in Italian is NOT of Latin origin and
> how did it get those geminates? From a substrate? By sound change?

Funny that you should ask, because I've recently started wondering
about this myself. I haven't even googled for it yet, though, so
here are just some initial thoughts:

A lot of Italian gemination must be due to sound changes, but I
don't know the details. The gemination in the present subjunctive
stem of some common irregular verbs such as sapere/sapp-, fa[ce]re/facc-
makes me suspect a development Cj > CCj. (This would be parallel
to West Germanic gemination.)

There is also "syntactic gemination", where some common monosyllabic
words trigger gemination in the initial consonant of the next word,
e.g. "va bene" is /ˌvabˈbɛne/. I don't know what to make of this.

Ruud Harmsen

unread,
Sep 17, 2022, 11:05:55 AM9/17/22
to
Sat, 17 Sep 2022 14:49:48 +0200: Tim Lang <m...@privacy.net> scribeva:

>>There's no need. I already know. I have heard a lot of Hungarian
>>in my life. I like the sounds of the language, even if I hardly understand
>>anything.
>
>OK, it might be so. But I'm highly skeptical, because of this
>additional information: "I hardly understand anything".

In Hungarian, I discern nearly all of the speech sounds, but I don't
have enough grammar and vocabulary to understand the meaning.

In European Portuguese, in 2000 of or, I knew thousands of words and
lots of grammar, but understood to zero of the spoken language,
because it's hard to find the sounds in the endless stream of fast
sounds.

Ruud Harmsen

unread,
Sep 17, 2022, 11:07:37 AM9/17/22
to
Sat, 17 Sep 2022 14:24:13 +0200: Tim Lang <m...@privacy.net> scribeva:

>On 17.09.2022 10:30, Ruud Harmsen wrote:
>
>>I never said it wasn't. Hungarian <a> is open or half-open, so
>>the lips are also relatively open,
>
>This is it = the correct perception! ("Ez az!") That is, less open
>than if you wish to say <á> (as if sitting in the ... dentist's
>"shop" :-)).
>
>>even if rounded and protruded.
>
>But there ain't any need whatsoever to round and protruded
>the lips for the Hungarian <a>.

Yes, there is, otherwise it is Hungarian with a Dutch accent.

>(Theoretically, only when
>you choose to bite into a sour lemon before performing the
><a> utterance. :-))

Ruud Harmsen

unread,
Sep 17, 2022, 11:09:32 AM9/17/22
to
Sat, 17 Sep 2022 14:24:13 +0200: Tim Lang <m...@privacy.net> scribeva:

>>If you mean protruding, moving forward: the many
>>videos show otherwise.
>
>Then you perceive them with a ... delay, thus seeing
>other vowels and not the <a> one.

Utter nonsense. In the videos I mentioned sound and vision are in
perfect sync.

Ruud Harmsen

unread,
Sep 17, 2022, 11:13:05 AM9/17/22
to
Sat, 17 Sep 2022 14:24:13 +0200: Tim Lang <m...@privacy.net> scribeva:

>>A few lines later in this same message, you are contradicting
>>yourself again:
>
>How on earth do I contradict myself, when I explain that
>the Hungarian vowel for <a> is exactly like the English
>version of that sound pronounced by a minority of the
>English people from England (incl. the Windsor and
>Mountbatten family; and some of the BBC-World Service
>speakers (especially a few decades earlier, since in
>recent times, more and more people there talking live
>have regional, and even US Midwestern pronunciations).

In one place you say Hungarian <a> isn't rounded, then you say it is
exactly like the LOT-vowel of British. The difference between a
British LOT and American LOT is that the latter is not rounded and the
former is.
https://www.merriam-webster.com/dictionary/lot
https://www.collinsdictionary.com/dictionary/english/lot

wugi

unread,
Sep 17, 2022, 11:22:22 AM9/17/22
to
Op 17/09/2022 om 15:42 schreef Christian Weisgerber:
> On 2022-09-16, Dingbat <ranjit_...@yahoo.com> wrote:
>
>> In Italian Capella, the gemination in the suffix comes from Latin.
>> How much of the gemination in Italian is NOT of Latin origin and
>> how did it get those geminates? From a substrate? By sound change?

Normal genuine sound change.

> Funny that you should ask, because I've recently started wondering
> about this myself. I haven't even googled for it yet, though, so
> here are just some initial thoughts:
>
> A lot of Italian gemination must be due to sound changes, but I
> don't know the details. The gemination in the present subjunctive
> stem of some common irregular verbs such as sapere/sapp-, fa[ce]re/facc-
> makes me suspect a development Cj > CCj. (This would be parallel
> to West Germanic gemination.)
>
> There is also "syntactic gemination", where some common monosyllabic
> words trigger gemination in the initial consonant of the next word,
> e.g. "va bene" is /ˌvabˈbɛne/. I don't know what to make of this.

And acqua < aqua, repubblica < republica.

Whatever its reason, I'm sure it has nothing to do with substrates, it
is a proper development of Italian (which can be seen as an extra
geminated Latin:).

Spanish has largely got rid of Latin gemination: ilegal, inocente; or
worked around it: inmenso, inmóvil. Only needing it by exception:
innovar. Apart from some -rr- geminates: irresponsable... There are
cases of -rr- representing geminated -r- in joined words, but I can't
think of one just now.

Other languages have kept Latin geminates but either ignoring them in
pronunciation or merely for keeping vowels short: command, common, oppose.

Italian has not only maintained Latin gemination but extended it into
new terrain: osservatore, lassativo, lassú, tassi, co[]struzione,
i[]spirare, della/dalla, petto, ettogrammo, otto, pizza, mezzo... and
acqua, pubblico.

--
guido wugi

Ruud Harmsen via Google Groups <google@rudhar.com>

unread,
Sep 17, 2022, 11:34:16 AM9/17/22
to
On Saturday, September 17, 2022 at 2:24:20 PM UTC+2, Tim Lang wrote:
> It must be the [ɔ] sign that constantly misleads you.

It's a perfectly fine IPA symbol. And it denotes a rounded vowel.
The unrounded counterpart is [ʌ].

> I use it because this is the one I was taught. The
> "better" one (that I assume would better fit the
> Hungarian <a>, namely, [ɑ] have been more and more
> used than the other in recent times.

[ɑ] is unrounded, [ɒ] is rounded. If the Hungarian vowel
were [ɑ], it would be the same as in Dutch. But I think it
sounds slightly different.

> I could've used instead the American sign for this, [ȯ],
> e.g. as in the word "loss" (pronounced like @ Merriam-
> Webster's, online). But I'm not at all accustomed to
> using [ȯ].

Those symbols are only for American English, and unusable
and confusing for anything else. Also because American
English isn't the same everywhere.

IPA is universal.

> Nothing went wrong: I just don't need watching that.

Cognitive dissonance.

> With the difference that in Hungarian you can issue a perfect
> variant of this vowel with absolutely no need to round anything.

Numerous videos prove you wrong. Hungarian <a> is clearly a
rounded vowel.

> Yes, I'll use [ɒ] from now on and cease writing [ɔ], if this
> is the problem bothering you.

Thereby you'd admit that the vowel is rounded, because
both symbols mean rounded vowels in IPA.

But you deny rounding (litterally closing your eyes to it),
so you contradict yourself.

> >Which is also what the Wikipedia diagram shows.
> The Wiki diagram doesn't use the inverted C at all; it uses
> the inverted cursive a only. (Well, this is new to me, as
> I said, because the inverted cursive a is a newer option
> than the way older inverted C.)

Inverted c is half open, turned scrip a is open (or almost so).
I also know the inverted c much longer that the other one,
which may be a later addition to IPA.

> Okay. But, again, I'd criticise this diagram (which is also
> cited in the article in Hungarian about the same subject,
> "vowels"), because the distance between [ɒ] and [a:]
> is the same as the distance between the to o's,
> [o] and [o:]. But there is a major difference caused
> by the aperture: the mouth must be widely opened for
> a real [a:],

Yes, <á> is fully opened and unrounded, <a> is almost open,
or even half open, and rounded. Clearly visible in numerous
videos, where <á> and <a> are close to each other.

> whereas both kinds of [o] (long and short)
> the lips and mouth aperture don't change positions.

That's right. The difference there is largely only length.

> So, only the simple three-tired diagram showing only
> the position of the vowels is a ... shortcoming, because
> you, as an "onlooker" from another phonemic and phonetic
> galaxy will forever miss other important features that
> are essential for rendering the "stuff" correctly.

I miss nothing, and IPA misses nothing. You missed
some if IPA, and you are unwilling to learn.

Tim Lang

unread,
Sep 17, 2022, 12:42:06 PM9/17/22
to
On 17.09.2022 17:05, Ruud Harmsen wrote:

>In Hungarian, I discern nearly all of the speech sounds, but I don't
>have enough grammar and vocabulary to understand the meaning.

That's fine, congrats.

>In European Portuguese, in 2000 of or, I knew thousands of words and
>lots of grammar, but understood to zero of the spoken language,
>because it's hard to find the sounds in the endless stream of fast
>sounds.

Oh, yeah, this is understandable: pronouncing words in Portuguese
(as well as Brasilian Portuguese) cannot compare with the pronun-
ciation in Hungarian: the Hungarian one sounds very "clear", something
ideal, I'd say, for anyone wishing to learn Hungarian. Portuguese
must be difficult to understand (because of the pronunciation) even
for most Italians, French, Castilians, Rheto-Romans, Romanians.

Tim

Tim Lang

unread,
Sep 17, 2022, 1:21:46 PM9/17/22
to
On 17.09.2022 17:13, Ruud Harmsen wrote:

>In one place you say Hungarian <a> isn't rounded, then you say it is
>exactly like the LOT-vowel of British.

Yes, but what you can't see and hear: _me_ doing both the
Hungarian with unmoved lips and English with a slight
rounding adaption. To me, all these variants are very
easy to distinguish and pronounce. And I hear daily
the Bavarian variants of both English and Hungarian
vowels, which would be (to me) the 3rd, 4th ɔ or ɒ
variant. 3-4 times a month I also hear the German
variant (number 5 :-)) from Saxony (from Dresden,
Leipzig, Halle), uttered by people who are immigrants
in my province. So ... And I can imitate most of these
ɒ-sounds. Even know whenever I have to make them sound
rather o-like, for those German regional ears where
the attached mouth rather says: "wos hob i da xogt?"
(High German: "was hab' ich dir gesagt?) To them, this
vowel still is an "a" or at least an "ɒ", but to me,
such exaggerations are as o-like as they come.

Example where I'd encounter difficulties, for instance imitating
the diferent [ʒ, tʃ] in Polish, as well as differentiate <sz>,
<ś>; <rz>, <ż>. Some natives would have to show me; these too,
the nasalized a's and e's, written with a cedilla, since I've
noticed the Polish similar sounds are different from French
<an, am, en, em> and <(e)in>. And perhaps even more difficult
to imitate some of the Danish sounds. And some of the genuine
Dutch pronunciations I've heard (radio, TV and in the street).
I assume, I'd need whole lotta drills, training.

>The difference between a
>British LOT and American LOT is that the latter is not rounded and the
>former is.

I've known that even empirically for decades, not only thanks
to movies, radio, TV, but also thanks to various persons of
British descent (England and Scotland) and of Amer. descent
(both East Coast & New York and Midwest and one L.A.), among
which a few were teachers of mine. And some of them I often
bothered insisting to ask "why do you pronunce this or that
that way?" "How do you pronounce this and that?" I remember
how some of them, Americans, hadn't known that "wrath" is
pronounced in BE [wrɔθ]

>https://www.merriam-webster.com/dictionary/lot
>https://www.collinsdictionary.com/dictionary/english/lot

For the British one, this is exactly how I was teached to
pronounce. And this is the pronunciation (the classic one)
I heard by BBC-World Service people; and by the family
of Queen Elizabeth II and her son Charles III (and by
them themselves).

But the American pronunciation by that women is a surprize:
because that "lot" rather sounds as another ... British
one. Despite the fact that her vowel in "lot" is not the
same as the one in the Brit. guy's pronunciation.

The overwhelming majority of Americans I listened
to say a much intenser, opener, "broader", ä in [lät].
The Webster woman's [ä] doesn't sound American (I'd dare
say, but only in the absence of American witnesses :-)).

(Is she an actress? Or from New England? Anyway: a bit
(nanomillimetrically) closer to A than the average
Hungarian <a>. The vowel pronounced by the Brit is
almost ... "Hungarian". "Almost" because not 100%, but
up to 85%-95%.)

Tim

Tim Lang

unread,
Sep 17, 2022, 2:15:39 PM9/17/22
to
On 17.09.2022 17:34, Ruud Harmsen wrote:

>I miss nothing, and IPA misses nothing. You missed
>some if IPA, and you are unwilling to learn.

Ezzel befejezem veled, mert szemtelenkedsz.
Tôlem, el mehetsz Kukutyinba zabot hegyezni.
Azt hittem, hogy egy komoly ember vagy, de
nagyon tévedtem.

Tim

Daud Deden

unread,
Sep 17, 2022, 6:16:16 PM9/17/22
to
Tim, I have only ever heard 'wrath' pronounced as rat (subtract t and add the th of thistle). What other pronunciation could it have? (I don't know the symbol (backwards c)).
Message has been deleted
It is loading more messages.
0 new messages