The old vowel-system - standardized 19th century spelling (as seen
in conservative editions of old poetry, for example) - (n)=nasalized
(which is seldom, if ever, seen in printed editions):
1. i í í(n)
2. y ý ý(n)
3. e é é(n)
4. ö oe oe(n)
5. ,e æ æ(n)
6. u ú ú(n)
7. o ó ó(n)
8. ,o ,ó ,ó(n)
9. a á á(n)
There are 9 basic simple vowels shown above. Each row ends with a
nasalized version of the long grade. In row 4, 'ö' stands for the
letter 'o' with a line through it - a symbol my keyboard is not set
up to make). What do you notice first about these character rows?
Perhaps you can see that the same letter occurs 3 times in every row
except in rows 4 and 5 ? Please bare in mind that the commas seen
before letters in rows 5 and 8 should appear as hooks below these
letters - a feature my keyboard is not set up to represent. Let us
now answer the question why rows 4 and 5 are unique in not showing
the same letter thrice. The answer is printed convention. How do we
know this? The answer is that we know this from early manuscripts,
where the letter ',e' (1st in row 5) is often written as 'æ' (like
the next letter in the same row) and 'ö' as 'oe' (or visa versa).
The hook under ',e' is almost always omitted in modern editions of
old poetry - a fact which goes for the most part unnoticed, as the
sound became identical with 'e'(1st in row 3) about the 13th century
in Icelandic; also, the pronounciation of 'æ' (2nd in the same row)
changed to a different in Icelandic as well. Regarding row 4, early
manuscripts show no consensus on which form of the vowel to write -
a fact which may have encouraged modern editors to use two different
symbols in modern editions - one for the long and another for the
short. Given the habits of modern editors (and many old spellers),
how could we represent nasalization if we wanted too? Should we set
a dot over the letter (like the grammarian who first described the
above system in writing)? What about the fact that in rows 4 and 5
the long and short of the same sound is represented by 2 different
characters? How could we make this more clear? The answers to these
questions can be found in early manuscripts. To clarify row 4, we
simply choose either 'oe' or 'o' (with a line through it) and place
an accent over the long variety (please note that accents are very
inconsistently used in old manuscripts). To clarify row 5, we simply
write 'æ' for the short and accented 'æ' for the long - that is to
say, 'viltu sægja oss æitthvat nýtt'? ('will you tell us something
new'?) instead of 'viltu segja oss eitthvat nýtt'. This spelling is
common in the earliest manuscripts. Why? Because short 'æ' was still
distinct from short 'e' - only words with short natural 'e' would
correctly be spelled with 'e' (words such as 'ek' I, 'vegr' way, 'at
vera/at vesa' to be). Also, the diphthongs 'ei' and 'ey' are often
spelled 'æi' and 'æy'. Why? For the same reason: the 'e' in these
diphthongs was a the 'hooked' variety - that is to say, the short of
long 'æ'. Where did these vowels come from? The answer is that short
'æ' is from short 'a'(á) and long 'æ' from long 'a'(á) - simply put,
they are mutations. For those interest in understanding how the old
vowels were pronounced, the spellings 'æ' for the short and '´æ' for
the long should be quite appealing. Are there any added benefits to
this way of spelling? Yes, consider the following: ',e' shows no
apparent relationship with 'a', whereas 'æ' shows this relationship -
in fact, the old grammarian mentioned above actually tells us this
much: the sound is somewhere between 'a' and 'e'. He also tells us
that the symbols ',o ,ó ,ó(n)' in row 8 stand for a sound which is
somewhere between 'a' and 'o' - in fact, we see the glyph 'ao' used
rather frequently in early manuscripts. Are there any benefits to
using this glyph? Yes, consider the following: 'ao' clearly shows
its relationship to 'a', whereas as ',o' does not. So based on what
this old grammarian had to say (around 1150), modern linguistics,
and on actual spelling conventions used in early manuscripts, we
arrive at the following 9 rows for use in printing early poetry (or
whatever else we choose to so print for that matter):
1. i í í(n)
2. y ý ý(n)
3. e é é(n)
4. oe oe' oe'(n)
5. æ æ' æ'(n)
6. u ú ú(n)
7. o ó ó(n)
8. ao aó aó(n)
9. a á á(n)
The diphthongs thus:
1. æi
2. æy
3. au
So what about the nasals? Should we use a dot even where there is
already an accent? Well, notice that none of the characters seen in
any of the rows above are 'hooked' - there are no 'tails' below any
of the characters. This means that we can use the 'tail' to signify
nasalization - so simple is that. The same text can be printed and
shown with or without nasalization 'tails' - no need to change the
vowel symbol, just add or subtract the 'tail' - so simple is that.
So maybe you wanted to know what vowels the vikings 'really' used?
Well, if you´re asking about the ones who spoke Old West Norse, take
a long hard look at rows above (imagine the 'tails' under the 3rd
letters in each of the 9 rows). Think about it like this:
Accent on top = lengthened grade (long vowel)
Tail below = nasalized variety of the same
What evidence do we have that this is how the vowel-system 'really'
was in Old West Norse? We have records of Faroese, Icelandic, and
West Norwegian - all of which derive from the same 9th-10th century
Old West Norse. Anything else? Yes, we have a very large number of
runic inscriptions in the same language - in other words, all the
evidence we need to win in court.
I doubt if the East Norse had any problem undestanding their Western
kin - the inscriptions tell us that much. What do you say, Arnljótr?