Germanic language with the most consistent spelling

[Nino83]

Hello everybody.

How would you rank Germanic languages (English, German, Dutch, Swedish, Danish, Norwegian, Icelandic) from that with the most consistent spelling to that with the most inaccurate one?

For example, I know that Danish has something like 35 vowels (this is why "A 15-month-old Croatian child understands approximately 150 words, while a Danish child of the same age understands just 84 on average." http://www.krusekronicle.com/kruse_...es-irritable-vowel-syndrome.html#.U3Hn4vl_uBQ) and the spelling is not so consistent ([æ] [ɑ] and [a] can be written with /a/ or /e/, [ɛ] with /e/, /æ/ or /i/, [ɔ] with /o/ or /å/, [o] with /o/ or /u/, [ø] [œ] and [ɶ] with /y/ or /ø/, the soft d (transcribed as [ð̞ˠ̠]) that sounds like an [ɫ], silent h (before v), d (after l, n, r) and e (after r), non-initial /p/ /t/ /k/ are pronounced [d] [g] and /v/ /r/ and /g/ as [ʊ̯] [ɐ̯] and [ɪ̯]/[ʊ̯] in syllable coda).

So probably Danish language has the worst spelling system.

Is there any Germanic language which is read (almost) as it is written (like Italian or Spanish)?
The main reason why English spelling reforms were rejected is that there are too many English accents so it's unlikely to reach an agreement.
Why does nobody propose to reform Danish orthographic system (seing that it is spoken only in Denmark)?

 

[Frank78]

H
Is there any Germanic language which is read (almost) as it is written (like Italian or Spanish)?

German is "spoken as its written" to quite a high degree but I'm unable to compare to any other language than English.

Here you have all the sounds: http://www.kuwi.europa-uni.de/de/le..._Ursula_Bock/phonetik/beispielw__rter_ipa.pdf

 

[ahvalj]

To perform a major reform of an orthographic system is almost as difficult as to change any other primary habit of adult people: more so, since orthography is one of the bases of the visual orientation of a literate person, its significant change is akin to, say, a change in the way the person moves. One is suggested to forget the old locomotory skills and to learn new ones. This is the reason why major reforms were extremely rare in history and were mostly tied with the change of an alphabet. I cannot imagine any major reform in e. g. English not involving a switch to a completely different (non-Roman) writing system. Yu jast cænt meik piipl to rait in æ touteli nyu wei widhaut æbændening ol thæ vizhuæl links tu thæ priivies sistem.

I think, indeed, German and Dutch orthographies are the least troublesome. Swedish and Norwegian have major problems with the letter o and with omitting the pitch tone.

 

[Nino83]

German is "spoken as its written" to quite a high degree but I'm unable to compare to any other language than English.

Here you have all the sounds: http://www.kuwi.europa-uni.de/de/le..._Ursula_Bock/phonetik/beispielw__rter_ipa.pdf

Thank you.
So hasn't German "silent letters"?
Are vowels and diphtongs written always with the same letters/diagraphs (for example [iː] is always written "ie" in open syllables)?
Wouldn't be better to write final /b/ /d/ /g/ with /p/ /t/ /k/?

This is the reason why major reforms were extremely rare in history and were mostly tied with the change of an alphabet. I cannot imagine any major reform in e. g. English not involving a switch to a completely different (non-Roman) writing system. Yu jast cænt meik piipl to rait in æ touteli nyu wei widhaut æbændening ol thæ vizhuæl links tu thæ priivies sistem.

Why non-Roman?
English doesn't have problems with consonants. Removing silent letters would be sufficient (knife, write, walk, should, island, debt, bomb, listen).
For example, it's not necessary to write "vizhuæl". The rule is simple, unstressed "s" before "ju" is a [ʒ], as in Italian a "c" before "e, i" is a [ʧ]. It's sufficient to write "visiual (with an "i", or "j" or "y" before the "u") or as it is "visual".
If we're speaking about vowels, they could make the "silent e" rule more consistent (ar, hav, liv, giv) (or write diphthong as they are pronounced, as you did in you comment, it was very clear to read, but this second change would be too drastic), eliminate the few irregular "ea" (breik, greit, steik, si), "ei" (why not "seeze" instead of "seize", "receeve" instead of "receive"?), the "ough" words (thaut, thru, tho, drout, cof, tuf).

It's not important to write every single vowel exactly as it sounds (there is the IPA for this purpose). It's sufficient to have few diagraphs with regular rules.

 

[ahvalj]

Why non-Roman?
English doesn't have problems with consonants. Removing silent letters would be sufficient (knife, write, walk, should, island, debt, bomb, listen).
For example, it's not necessary to write "vizhuæl". The rule is simple, unstressed "s" before "ju" is a [ʒ], as in Italian a "c" before "e, i" is a [ʧ]. It's sufficient to write "visiual (with an "i", or "j" or "y" before the "u") or as it is "visual".
If we're speaking about vowels, they could make the "silent e" rule more consistent (ar, hav, liv, giv) (or write diphthong as they are pronounced, as you did in you comment, it was very clear to read, but this second change would be too drastic), eliminate the few irregular "ea" (breik, greit, steik, si), "ei" (why not "seeze" instead of "seize", "receeve" instead of "receive"?), the "ough" words (thaut, thru, tho, drout, cof, tuf).

It's not important to write every single vowel exactly as it sounds (there is the IPA for this purpose). It's sufficient to have few diagraphs with regular rules.

Well, then why feel and feeling and not fele and feling? Read and not rede? Breath and not breth? Wait and not wate? With and not widh (cp. myth)? What happens in the Latin borrowings with the open and closed syllables (confuse/confusion but provide/provision). What about the permanently unstressed vowels (bottom)? And what about latest borrowings that do not follow these rules (machine instead of mashene)? If they start correcting all these inconsistencies, the overall aspect of the text will change drastically. Hence my comment about the non-Roman script.

 

[bearded]

@ Ahvalj
Just a marginal remark or Randbemerkung: your ''caent meik'' in #3 reveals that you adopt the US pronunciation, since a Briton would probably have written ''caant meik''. Now suppose a different alphabet was accepted to the purpose of making orthography consistent with pronunciation: what would happen with those words that are being written in the same way but pronounced differently here and there? Would there be two or more orthographies of English? Thank you for letting me know your idea.

 

[ahvalj]

@ Ahvalj
Just a marginal remark or Randbemerkung: your ''caent meik'' in #3 reveals that you adopt the US pronunciation, since a Briton would probably have written ''caant meik''. Now suppose a different alphabet was accepted to the purpose of making orthography consistent with pronunciation: what would happen with those words that are being written in the same way but pronounced differently here and there? Would there be two or more orthographies of English? Thank you for letting me know your idea.

I am actually not suggesting any change: in contrast, I am trying to explain why it most probably won't happen (at least, without major geopolitical and cultural shifts). The new orthography, if eventually introduced, should of course be a result of a broad consensus, to be applicable to all the regional literary variants. In cases like can't there probably will be separate spellings for each pronunciation (e. g. cænt and cant).

 

[Nino83]

The new orthography, if eventually introduced, should of course be a result of a broad consensus, to be applicable to all the regional literary variants. In cases like can't there probably will be separate spellings for each pronunciation (e. g. cænt and cant).

The important thing is the correspondence between grapheme and phoneme (short /a/ can be pronounced [a] or [æ], this wouldn't be a real problem).

 

[Frank78]

Thank you.
So hasn't German "silent letters"?
Are vowels and diphtongs written always with the same letters/diagraphs (for example [iː] is always written "ie" in open syllables)?
Wouldn't be better to write final /b/ /d/ /g/ with /p/ /t/ /k/?

That's why I wrote "to a high degree". Final obstruent devoicing speaks against it.

Real silent letter are rare in German, no example comes to my mind right now. Even If they are not spoken they have a function, e.g. the h in "stehen" indicates that the preceding vowel must be spoken long.

 

[ahvalj]

German has a number of cases when the distinction between long and short vowels remains not reflected in the orthography. Thus, probably Dutch (+Afrikaans) should be considered the Germanic language with the most consistent spelling.

 

[Nino83]

Real silent letter are rare in German, no example comes to my mind right now. Even If they are not spoken they have a function, e.g. the h in "stehen" indicates that the preceding vowel must be spoken long.

Thank you. So would it be correct to say that German and Dutch have the most accurate orthography, English and Danish the least one and Swedish and Norwegian are in the middle?
And Icelandic? Does anyone know something about it?

 

[ahvalj]

Thank you. So would it be correct to say that German and Dutch have the most accurate orthography, English and Danish the least one and Swedish and Norwegian are in the middle?
And Icelandic? Does anyone know something about it?

Icelandic is pretty regular: the problem is that the rules are too numerous (a bit like French: every letter in certain circumstances has only one reading, you just have to memorize each of these zillion circumstances). Faroese has a very historical orthography, like English, if not more. But the most complicated is probably Frisian.

 

[itreius]

Dutch has some inconsistencies in its orthography, -lijk and -en immediately spring to mind, although the orthography as a whole is fairly straight-forward. Certainly much more intuitive for me than the Danish one.

 

[Nino83]

Icelandic is pretty regular: the problem is that the rules are too numerous (a bit like French: every letter in certain circumstances has only one reading, you just have to memorize each of these zillion circumstances). Faroese has a very historical orthography, like English, if not more. But the most complicated is probably Frisian.

Thanks.
Where would you put Icelandic? Before or after Swedish and Norwegian?

 

[berndf]

The important thing is the correspondence between grapheme and phoneme (short /a/ can be pronounced [a] or [æ], this wouldn't be a real problem).

That is exactly the problem. In British English* can't does not contain a short /a/ but a long /ɑ:/ and in American English* this would collide phonemically with a short <o>.

________________
*Of course simplified. There are dialects in both countries where this is not true.

 

[Nino83]

That is exactly the problem. In British English* can't does not contain a short /a/ but a long /ɑ:/ and in American English* this would collide phonemically be a short <o>.

Ah, ok. Oxford Learner's Dictionary is clear on the matter (BrE has an [æ] in can and an [ɑ:] in can't, quite irregular).
So things would be complicated with open vowels.

 

[ahvalj]

Thanks.
Where would you put Icelandic? Before or after Swedish and Norwegian?

Icelandic pronunciation is 100% deducible from the orthography, if I am not mistaken (well, except the shortenings and assimilations in the less careful speech), hence before Swedish and Norwegian, but the rules are much, much more numerous and the overall phonetic aspect of the language is much more different from the written image, not less than in Danish. In principle, you can try to pronounce the Swedish text in an Italian way (not forgetting the initial stress and a few basic rules) and you most likely will be understood (I heard an Italian colleague doing this), which is definitely not so in Icelandic and Danish.

 

[berndf]

Real silent letter are rare in German, no example comes to my mind right now. Even If they are not spoken they have a function, e.g. the h in "stehen" indicates that the preceding vowel must be spoken long.

But you have no way to predict, e.g. if /ʃlaːfən/ produces schlafen, *schlaafen or *schlahfen. There are ambiguities spelling->pronunciation (mainly the non-marking of vowel length, stress and morpheme-boundary (Wachstube could be Wach-stube or Wachs-tube)) and there are ambiguities pronunciation->spelling.

 

[Stoggler]

Dutch has some inconsistencies in its orthography, -lijk and -en immediately spring to mind, although the orthography as a whole is fairly straight-forward. Certainly much more intuitive for me than the Danish one.

True but such inconsistencies (or exceptions if you will) are rare compared to most other languages and it takes little effort to learn them.

 

[Nino83]

So it would be: Dutch, German (vowel lenght ambiguities), Swedish, then Icelandic (predictability but too many rules), Norwegian, Danish.

 

[Stoggler]

So it would be: Dutch, German (vowel lenght ambiguities), Swedish, then Icelandic (predictability but too many rules), Norwegian, Danish.

Why are Norwegian and Swedish split by Icelandic?

And what about the two written standards for Norwegian (Bokmål and Nynorsk)? How do they fit into the equation?

I can't speak for Nynorsk but the standard Norwegian courses you can buy in bookshops here cover Bokmål and that seems more consistent than Swedish to me (I am on thin ice with Norwegian though - I can talk about Swedish and that has issues with the sk- and sj spellings, nothing too difficult but there are inconsistencies in their spellings)

 

[Nino83]

Why are Norwegian and Swedish split by Icelandic?

I don't know (I made a supposition)
Is there a great difference between Bokmål and Danish in spelling?

 

[itreius]

Written Bokmaal and Danish are close, but the first is a better representation of speech than the latter.

 

[Stoggler]

Is there a great difference between Bokmål and Danish in spelling?

I don't think there are major differences between the two written languages - they are very similar in writing:

http://en.wikipedia.org/wiki/Comparison_of_Norwegian_Bokmål_and_Standard_Danish

The spoken languages though are a different matter

 

[Nino83]

So would you put Norwegian (Bokmål) between Swedish and Icelandic?

 

[Dib]

As far as I understand, as a matter of official stand, Norwegian has no standard pronunciation nor any standard spoken language. Both Bokmål and Nynorsk are supposed to be purely written languages. Things may be a bit less diverse in practice, but that certainly adds another level of "problem"?

 

[ahvalj]

So, speaking purely of consistency, for those who like charts (not a Russian habit, I should confess), I would suggest something like this:
Dutch/Afrikaans (well thought and consistent, looking much better than sounding ,-)
Icelandic (highly etymological but very consistent spelling)
Soviet Yiddish (with all vowels marked and basically no redundant letters for x etc., though with some troubles as to the Slavic borrowings and influences)
German (v/f, vowel length, -h-)
Swedish/Bokmål/Nynorsk (ambiguous o, no pitch tone marked, sk-/sj-/skj-/stj-, j-/dj-/gj-)
Faroese (extremely etymological but rather consistent spelling)
English (etymological and rather inconsistent spelling)
Frisian (not well established spelling)
Danish (etymological and rather inconsistent spelling, no stød marked)
traditional Yiddish (part of vowels omitted, historical spelling of Hebrew words).

 

[berndf]

Soviet Yiddish (with all vowels marked and basically no redundant letters for x etc., though with some troubles as to the Slavic borrowings and influences)

What do you mean by "Soviet Yiddish"? The YIVO standard (which is based on the Lithuanian dialect)?

 

[ahvalj]

What do you mean by "Soviet Yiddish"? The YIVO standard?

I am not an expert and I am based on the comments in the manuals and overviews: it was the orthography used e. g. in the newspaper Sovyetish Heymland and in the books edited in the USSR. Wikipedia states (http://en.wikipedia.org/wiki/Yiddish_orthography):

The first action formally undertaken by a government was in the Soviet Union in 1920, which prescribed the abolition of the separate etymological orthography for words of Semitic (i.e., Hebrew and Aramaic) origin. This was extended twelve years later with the elimination of the five separate final-form consonants (as indicated in the table below) which were, however, widely reintroduced in 1961. The changes are both illustrated in the way the name of the author Sholem Aleichem is written. His own work uses the form שלום־עליכם but in Soviet publication this is respelled phonetically to שאָלעמ־אלײכעמ also dispensing with the separate final-form mem and using the initial/medial form instead. This can be seen, together with a respelling of the name of the protagonist of his Tevye der milkhiker (originally טביה, changed to טעוויע), by comparing the title pages of that work in the U.S. and Soviet editions illustrated next to this paragraph. Note also the Germanized מילכיגער (milkhiger) in the former exemplifying another widespread trend, daytshmerish, discussed further below.

 

[berndf]

Interesting, thank you. I didn't know such a convention existed. The YIVO standard also unifies some variant spellings but it never went so far as to respell Hebrew and Aramaic words or abolish ף ,ן ,ם ,ך and ץ.

 

[Nino83]

So, speaking purely of consistency, for those who like charts (not a Russian habit, I should confess), I would suggest something like this:

Thank you for the chart!

 

[Dan2]

A couple of tangential comments if I may...

I know that Danish has something like 35 vowels

Danish certainly has a lot of vowels, but I think it's unwise to make a statement like this without more specificity. Danish certainly doesn't have anything like 35 vowel phonemes, so the number 35 clearly includes allophones. Then the question arises of how different two sounds have to be to call them different allophones. American English has about 15 phonemically distinct vowels and diphthongs, but the number of allophones is not clearly defined. For ex., the words "bet", "bed", and "Ben" would all be said to contain the same vowel phoneme, but the vowel if "bet" is distinctly shorter than that of the other two. Of the other two, the vowel of "Ben" is partially nasalized. With respect to these three words then, do we have just one allophone (since the vowel quality in all of these words is essentially [ɛ]) or two (long vs short) or three? With effort I might be able to identify 35 or even more "vowels" for American English.

The main reason why English spelling reforms were rejected is that there are too many English accents so it's unlikely to reach an agreement.

This is often said, but I think it represents a misunderstanding. No one is suggesting replacing English orthography with something like IPA, which would require choosing an absolute pronunciation for each word. Rather, spellings would simply be made more consistent. Here's an example of how dialectal variation is much less important than one might think: One can distinguish about a half-dozen dialectically different pronunciations of the word "heart" ([hɑrt], [hart], [hɑ:t], [ha:t], etc.) and yet speakers with every one of these pronunciations would agree on the following: a) "heart" has an exceptional spelling, b) "heart" rhymes with "part" and "cart", c) respelling it as "hart" (even if unwise for other reasons) would make its spelling regular. So the dialectal variation here is irrelevant to spelling reform. Similarly, a British friend and I pronounce "go" and "got" quite differently from one another, but we would both insist that "go" and "got" are absolutely fine spellings for these words.

Sure, there are some cases where major dialects have irreconcilable differences (Amer vs Brit "can't", "bath", etc.), but claiming that it's impossible to "fix" thousands of irregular spellings because a small residue of inconsistencies would remain is illogical.

But there are good reasons to reject a major spelling reform for English: the loss of continuity with earlier texts; the obscuring of relationships between words ("nation" and "democrat" have different vowel phonemes compared to "national" and "democracy"); the obscuring of relationships across languages (words like "nation" and "democracy" exist in other important languages); the difficulties that would be imposed on generations alive during the change (as pointed out by ahvalj); and an aesthetic sense that many of us have that the appearance of "phonetic" English is quite unappealing. (I personally would favor the respelling of only a small handful of words that are not part of paradigms and that give natives and foreigners a lot of problems, like "through" and "though".)

 

[Nino83]

Danish certainly doesn't have anything like 35 vowel phonemes, so the number 35 clearly includes allophones.

Yes but the "problem" is that in Danish, often, these vowels overlap so the same sound in the same circumstances is written with different letters. For example, in Brazilian Portuguese you have represented by /i/ or final unstressed /e/ but no word ends with an /i/ so you can predict whether the sound is a written /i/ or /e/.
If you can't predict how the same sound is written, the spelling is, simply, inconsistent.

This is often said, but I think it represents a misunderstanding. No one is suggesting replacing English orthography with something like IPA, which would require choosing an absolute pronunciation for each word. (I personally would favor the respelling of only a small handful of words that are not part of paradigms and that give natives and foreigners a lot of problems, like "through" and "though".)

In fact I supported spelling reform of ough words, the removing of the silent letters (that are never pronounced), and to make the "silent e" rule more consistent (in verbs like have, give etc.). The fact that "nation" and "nationality" have a different vowel is due to the fact that the first is stressed and the second has a secondary stress, so it's not a problem if the change is always the same (from [eɪ] to [æ], as in nation/nationality and in nature/naturalistic). The rule is found and the spelling is consistent.

 

[berndf]

Yes but the "problem" is that in Danish, often, these vowels overlap so the same sound in the same circumstances is written with different letters. For example, in Brazilian Portuguese you have represented by /i/ or final unstressed /e/ but no word ends with an /i/ so you can predict whether the sound is a written /i/ or /e/.
If you can't predict how the same sound is written, the spelling is, simply, inconsistent.

I'd say the opposite is true. In "consistent" spelling, each grapheme should always represent the same phoneme, independently of how it is realized.

The fact that "nation" and "nationality" have a different vowel is due to the fact that the first is stressed and the second has a secondary stress, so it's not a problem if the change is always the same (from [eɪ] to [æ], as in nation/nationality and in nature/naturalistic). The rule is found and the spelling is consistent.

That is the situation we have now (lots of at hoc rules to explain certain shifts). If this is not a problem for you you don't need a reform at all.

 

[Nino83]

In fact (for me) it would be sufficient to remove exceptions (why to write break or receive?).
About regular shifts, I don't understand why nature and natural don't have the same vowel (another exception?).
Finally, I think that if a language have all these exceptions to the rule, there won't be any system which can make spelling consistent without obscuring the relationship between words.

 

[berndf]

About regular shifts, I don't understand why nature and natural don't have the same vowel (another exception?).

I don't really know. Interesting question. The MED gives states the pronunciation with three long vowels ([na:tu:ra:l]). I only guess that a three syllable word with long vowels in all three syllables was too cumbersome and two vowels were shortened ([natu:ral]).

 

[Nino83]

According to Oxford Dictionary: [ˈneɪtʃə(r)] and [ˈnætʃrəl] (so the /u/ is elided and the /a/ shortened).
With these changes it's impossible to have your cake and eat it.

 

[berndf]

Yea, that's what I said, both /a:/s were shortened. The /u:/ is not elided. Its post-great-vowel-shift-reflex in an unstressed syllable is [ʃ(ə)].

 

[Nino83]

So is it [ˈnætʃərəl]?

 

[berndf]

So is it [ˈnætʃərəl]?

In my experience there is free variation between [ˈnætʃərəl] and [ˈnætʃrəl] (see phonetic transcription in Webster's).

You might even call it an arbitrariness of transcription: The realization of some Schwas can be very reduced so it almost doesn't matter anymore, if it is audible or not. You would still "feel" it. It is similar to the "flüchtiges e" phenomenon in German (unsre vs. unsere).

 

[Ben Jamin]

Thank you.
So hasn't German "silent letters"?
Are vowels and diphtongs written always with the same letters/diagraphs (for example [iː] is always written "ie" in open syllables)?
Wouldn't be better to write final /b/ /d/ /g/ with /p/ /t/ /k/?



Why non-Roman?
English doesn't have problems with consonants. Removing silent letters would be sufficient (knife, write, walk, should, island, debt, bomb, listen).
For example, it's not necessary to write "vizhuæl". The rule is simple, unstressed "s" before "ju" is a [ʒ], as in Italian a "c" before "e, i" is a [ʧ]. It's sufficient to write "visiual (with an "i", or "j" or "y" before the "u") or as it is "visual".
If we're speaking about vowels, they could make the "silent e" rule more consistent (ar, hav, liv, giv) (or write diphthong as they are pronounced, as you did in you comment, it was very clear to read, but this second change would be too drastic), eliminate the few irregular "ea" (breik, greit, steik, si), "ei" (why not "seeze" instead of "seize", "receeve" instead of "receive"?), the "ough" words (thaut, thru, tho, drout, cof, tuf).

It's not important to write every single vowel exactly as it sounds (there is the IPA for this purpose). It's sufficient to have few diagraphs with regular rules.

You just quote the simple words in English, but what about Cholmondeley, Worcestershire, cough, dough, voicing or not voicing of intervocalic "s", and many others?

 

[Nino83]

You just quote the simple words in English, but what about Cholmondeley, Worcestershire, cough, dough, voicing or not voicing of intervocalic "s", and many others?

Cough is cof, dough has the same vowel of though (tho).
I agree with you but it's better (I think) to have less exceptions than more ones.
Anyway, it's clear that it's no use reforming English spelling, seeing that the word nation (stressed /a/ in open syllable) has an [eɪ]), national (stressed /a/ in open syllable followed by two unstressed syllables) has an [æ] while nationhood (equal to national) has an [eɪ]. It impossible to put toghether consistent spelling preserving, at the same time, ethymological relationships.

 

[berndf]

Anyway, it's clear that it's no use reforming English spelling, seeing that the word nation (stressed /a/ in open syllable) has an [eɪ]), national (stressed /a/ in open syllable followed by two unstressed syllables) has an [æ] while nationhood (equal to national) has an [eɪ]. It impossible to put toghether consistent spelling preserving, at the same time, ethymological relationships.

You could add markers for long short and reduced varieties of a vowel, e.g. macron for long, nothing for short and breve for reduced, nation > nāshŏn, national > nashŏnăl, nationhood > nāshŏnhud. That would be a compromise between etymological and phonemic spelling. Not that I advocate it but at least it doesn't look too awful.

 

[Peterdg]

I don't know why you consider Dutch spelling consistent. (my mother tongue is Dutch).

First of all, you have no way to know where the stress of a word falls. However, you need to know where the stress of the word falls to be able to deduct the pronunciation of a vowel (you need to be able to determine if a syllable is open or closed and stressed to know if a single written vowel has to be pronounced long or short).

E.g.: "bedelen". If the stress falls on the the first syllable, you pronounce the first "e" long and it means "to beg". If the stress falls on the second "e", then that one is pronounced long and it means "to endow".

Second, you have "ij" and "ei". In some cases, they are pronounced exactly the same, in other cases "ij" is pronounced as a schwa.

Third, you have "au" and "ou" which are pronounced exactly the same.

Then there are the eternal "s", "c", "k" ambiguities. "Sex" in English is written "seks" in Dutch. Don't ask me why.

We also have the problem of the intermediate "n" in compound words: that is a real disaster. In a compound word consisting of two independent existing nouns, you have to write an intermediate "n" if the first noun only has a plural in "n". If it also has a plural in "s" (or only has a plural in "s"), the you don't write the intermediate "n". So far so good: then you have what they call the "petrified" cases: some words don't follow the above rule because "they are petrified". Only God knows why, and even then...

I can go on and on about our spelling but the base line is that nobody can write correctly in Dutch.

I have written this post without any spelling checker. When I write in the Spanish forum, I always do that without spelling checker. When I want to write something in the Dutch forum (my own mother tongue for Pete's sake), I do not dare to post without passing what I have written through a spelling checker first. Sad, very sad.

BTW: Since I was born, we have had at least 4 spelling reforms and none of them has made it any better.

 

[Nino83]

I'd distinguish between accent and other orthographic matters.
For example, also in Italian there are minimal pairs differing in accent (lèggere, to read, leggère, lighweight f. pl.). I'd be in favor of marking the accent when it doesn't fall on the penultimate syllable (as in Portuguese or Spanish).

The problem of English language is that the same written letter with the same accent is often pronounced in a different manner (unpredictability).
In Dutch, have vowels the same pronunciation when they are stressed? Is it easy to determine if a syllable is open or closed?
If the answer is yes, I don't see many problems.

You could add markers for long short and reduced varieties of a vowel, e.g. macron for long, nothing for short and breve for reduced.

If I'm not wrong, long vowels are present only in stressed syllables (is it right?), so for me it would be sufficient to introduce two accents: grave accent (`) for long vowels and (´) for short vowel (to use only when the stress doesn't fall on the first syllable).
So: nàtion, national, nationálity, nàtionhood.

Anyway, the introduction of accents would be healty in (some) Germanic languages (and in Italian, as in the example above).

 

[ahvalj]

Yes, mea culpa, I haven't thought about ei/ij and ou/au and minor problems in Dutch. It turns out that the most consistent spelling is found in Icelandic, which seems really odd considering the complexity of the etymological Icelandic orthography. Anyway, these four languages — Dutch, Afrikaans, Icelandic and German — have the least problematic orthographies of all the Germanic languages. The Soviet Yiddish, which had the most consistent orthography in the 30—50's, was spoiled later, and in any case seems to be dead now.

 

[berndf]

If I'm not wrong, long vowels are present only in stressed syllables (is it right?), so for me it would be sufficient to introduce two accents: grave accent (`) for long vowels and (´) for short vowel (to use only when the stress doesn't fall on the first syllable).
So: nàtion, national, nationálity, nàtionhood.

Anyway, the introduction of accents would be healty in (some) Germanic languages (and in Italian, as in the example above).

Remember, Germanic distinguishes three and not just two levels of stress and also three quantities and not just two, so marking primary stress alone wouldn't suffice. In your example you would lose the information that the unaccented <a> in nationálity is short and not reduced while the <o> is reduced. In my notations that would be marked: nashŏnalĭty. The shortcoming of my notation would be that primary and secondary stress aren't distinguished.

 

[Nino83]

Yes but often secondary stress is predictable (e.g. stress-reduced-secondary or secondary-unstressed-stressed, unstressed-stressed), i.e the secondary stress doesn't fall immediately before or after the stressed syllable.
My notation shows where the primary stress is.

EDIT: It's not always so, e.g. assòciàtion. So also my notation is not able to show where the primary stress is.

 

[Testing1234567]

Why don't we just use IPA to write English, for that matter.

 

[Peterdg]

Why don't we just use IPA to write English, for that matter.

Because of this:

No one is suggesting replacing English orthography with something like IPA, which would require choosing an absolute pronunciation for each word.