This is an old revision of the document!
Content
Lateral approximants
Nasals
M/m (runic ᛗ) for [m]
N/n (runic ᚾ) for [n]
[ŋ]?
some proposals also include [ŋ], what makes sense since this sound is quite common in Germanic languages
mostly the digraph ng is proposed as its written representation
another possibility would be to take an unused symbol, e.g. q (or ᛝ in runes)
maybe only have it as an allophone of /n/ before /k, g/ thus only having [ŋk, ŋg]? Some varieties of Germanic have it this way; others have multiple sources for [ŋ], such as Thuringian which has it for *nd (“finge(n)” 'find')
Plosives
P/p (runic ᛈ) for [p]
B/b (runic ᛒ) for [b]
T/t (runic ᛏ) for [t]
D/d (runic ᛞ) for [d]
K/k (runic ᚲ) for [k]
G/g (runic
ᚷ) for [g] (not totally clear, e.g.
sko03 proposes to speak this letter as [j] in words ending with
-ig
[θ] and [ð]?
most proposals don't include these two as most Germanic languages eliminated them except of English and Icelandic
by the bigramms Th/th and Dh/dh (almost like in English)?
by the symbols Þ/þ (runic ᚦ) and Đ/ð (like in Icelandic)?
this poll resulted in replacing these sounds with [d] (written
d) and/or [t] (written
t)
Fricatives
H/h (runic ᚺ) for [h]
F/f (runic ᚠ) for [f]
S/s (runic ᛋ) for [s]?
Z/z (runic ᛉ) for [z]?
R/r (runic ᚱ) for [r], [ʁ], …?
V/v for [v]?
[ʃ]? (and [ɕ]?)
most Germanic languages have that sound
but none has a certain symbol for it (except for Yiddish with Hebrew letters)
digraph sh like in English?
unused letter, e.g. c like in Lojban or x like in Catalan and Portuguese?
diacritic symbol on another letter, e.g. ŝ like in Esperanto?
sc, with c not otherwise used?
sj? (would have the mildly adverse effect of blurring the distinction between [ʃ] and [sj]).
[x ~ χ] ([ç] as well???)?
digraph
ch like in German and Scottish? (according to
poll 1 and
poll 2 most people prefer this)
h as in Old English and Old Saxon
unused letter, e.g. x like in some Slavic languages?
diacritic symbol on another letter, e.g. ĥ like in Esperanto?
Semivowels
Vowels
I/i (runic ᛁ or ᛇ) for [i] or [ɪ]
E/e (runic ᛖ) for [e], [ɛ] or [ə]
A/a (runic ᚨ) for [a], [ɑ], [æ] or [ʌ]
O/o (runic ᛟ) for [o] or [ʊ]
U/u (runic ᚢ) for [u] or [ʉ]
Ö/ö or Oy/oy (runic ᛟᛇ) for [ø]
Ü/ü or Uy/uy (runic ᚢᛇ) for [y] or [ʏ]
Length
A controversial subject is the question about long and short vowels:
Variant 0a: No distinction
Variant 0b: No indication
Variant 1a: Vowel doubling
Variant 1b: Lengthening-symbol
Variant 1c: Distinct letters
Variant 2: Implicitly on consonant doubling/clustering
Variant 3 (or "1.5"): Vowel AND consonant doubling
description
a hybrid of variants 1 and 2
every vowel is short by default, but long if it is in the first syllable of a multisyllabic word and there is only one consonant separating it from the next vowel.
final A and E are short by default; I, O, and U are long by default (?)
if an otherwise-short vowel needs to be made long, it is doubled.
if an otherwise-long vowel needs to be be made short, the consonant after it is doubled.
example: ?
advantages
disadvantages
Of course these variants can be mixed or bound to conditions (like the stress or the position of the vowel in the word).
Non-standard vowels
Another question is the inclusion of non-standard vowels, which are present in most Germanic languages:
[ø]/[ö]
present in all Germanic languages
several cases in which having such a sound is practiclly inevitable (e.g. the word for to hear, høra seems to be the only viable solution)
possibilities
representation by the symbol ö
representation by the symbol ø
representation by a yet unused symbol, e.g. c or q (e.g. hcra/hqra)
representation by the symbol group oe
representation by the symbol group oy
[y]/[ü]
present in many Germanic languages, but not in all
several cases in which having such a sound is practiclly inevitable (e.g. the word for south, syd seems to be the only viable solution)
possibilities
representation by the symbol ü
representation by a yet unused symbol, e.g. y (e.g. syd)
representation by the symbol group ui
representation by the symbol group ue
representation by the symbol group uy
[æ]
Z and X
Are we going to use the letters Z and X to represent consonant pairs? If so, the first question is which values precisely they will take; Z will be /ts/ and/or /dz/, and X will be /ks/ and/or /gz/. A second and more esoteric question is how to double them, assuming a consonant-doubling system is chosen. The following options exist:
no doubling (Z and X) which is simple but either leaves the length of the preceding vowel unclear, or means that it must always be short;
simple doubling (ZZ and XX) which though simple, is misleading, because you'd think the pairs should be pronounced “tsts” and “ksks”;
explicitness (TS and KS) which is simple but which is incongruous alongside the “implicit” originals;
first-half doubling (TZ and KX) which chimes nicely with German for TZ (e.g. Schnitzel, rather than Schnizel) but which is unfamiliar with KX; and 5, second-half doubling (ZS and XS) which is functionally just like the previous option but visually more unfamiliar.
Consonant-cluster-shortcuts like these are more common in Romanic and Slavic languages. The orthography becomes less phonetic when using them and as shown above it starts to get weird when having short and long vowels in the language. So i guess they are only relevant if using vowel-variant 0a or 0b, since except of #1 none of the options above is common in any Germanic language or any other language i know. I tend to use the explicit forms ks, ts, …
— Fenris Wolf 2014/10/31 11:35
Open questions
What are the names of the letters of the alphabet? A simple approach would be to let the vowels be themselves, and the consonants be themselves plus A; thus a, ba, ca, etc. This way, the names are obvious, and conflict with other words is minimised (if the letter D was di instead of da, for example, there would be ambiguity between the pronoun and the letter).
Something else that people might like to think about is the possibility of enabling “invisible little schwas”, such as exist in words like apl and spasm. NB, the only other popular auxlang that does this is Idiom Neutral.
See also
Discussion
Just a note: there could theoretically be another variant, just using the most “common” spelling of each word, which would give a mix of variants. If the languages do not differ to much, or if one minimises the differences, this would give a not to complex orthography - interlingua is done in this way, keeping differences but minimizing them. If not, this would just be inconsequent and cluttery.
A way of deciding, on the other hand, would be to look in more detail on how common and widespread each system are, and chose the most common (or, as in the above, two very common).
The possible disadvantage with consequence is that it makes some words less recognizible. Depending on the individual, this may make it harder to understand. Likewise, some will probably feel like inconsequent spelling makes the language hard, and inconsequence may for others not be any bigger obstacle. I'm getting lost here, but I don't think that inconsistent spelling, nor a bit lessened recognizability is any larger problems. (And personally, inconsistent spelling hasn't been any big obstacle when learning a language.. As you probably even can see here though, I'm not sure about the english spelling in all cases. But practically, spelling errors is not that much of a flaw in the understanding and usage of a language, it shouldn't be over-estimated. One still understands the language, and can make yourself understood - the disadvantage is that it has a low status and risc making you look incompetent..).
The doubling idea (from user Ob)
In a nutshell, vowel doubling is used, and accents are only necessary in special cases; for example, ruum [ru:m], bu [bu:], but bùung/buûng [bu:UN]. Here's a full listing (NB, macrons could be used instead of graves), where the options in each slot are ranked from left to right in order of preference, and where the main vowels are in bold.
Vowel | Pre cons. | Pre vowel | Final | Vowel | Pre cons. | Pre vowel | |
/a/ | a,â | a,â | a,â | /A:/ | aa,à | à | aa,à |
/E/,/@/ | e,ê | ê | e,ê | /e:/ | ee,è | e,è | ee,è |
/I/ | i,î | î | î | /i:/ | ii,ij,ì | i,ij,ì | i,ii,ij,ì |
/O/ | o,ô | ô | ô | /o:/ | oo,ò | o,ò | o,oo,ò |
/U/ | u,û | û | û | /u:/ | uu,ù | u,ù | u,uu,ù |
/9/ | ø | ø | ø | /2:/ | ø | ø | ø |
/Y/ | y | y | y | /y:/ | y | y | y |
As you can see, I don't know what to do about the front rounded vowels yet…
J-diphthongs are easy to form (aj, oj) but with U-diphthongs there's a choice between U and W. I suggest writing /aU/ as au, and using W for the other, rarer ones.
Confession: I'm not sure what the situation is in the Continental Germanic languages regarding final vowels and vowels before other vowels (my system is English-inspired in this regard) in terms of the long/short distinction, so it may be that the system can be improved there. I also have a distorted perception of certain instances of vowel + R.
Decisions made so far
L/l for [l]
M/m for [m]
N/n for [n]
bigramm ng for [ŋ] since this representation is used in all Germanic languages and therefore it is the one that is most easily recognizable (according to goal #1)
P/p for [p]
B/b for [b]
K/k for [k]
G/g for [g]
T/t for [t]
D/d for [d]
[θ] and [ð] are not included and get replaced by [d] or sometimes [t] in the word-derivation from English and Icelandic (the only two Germanic languages that still have these sounds)
H/h for [h]
S/s for [s]
V/v for [v]
F/f for [f]
R/r for [ɾ]
J/j for [j]
[w] is not included
[x]/[χ] is not included
C/c for [ʃ] — whether this one is used in normal words will be decided later
it is allowed to pronounce the letters a little different, for example v as [ʋ] or s as [z], where the sounds from above are the default
If you've got strong arguments against any of these decissions, feel free to add them here or visit the IRC-channel!