Old Norse -> Articles -> Pronunciation of Old Norse (standard)

Pronunciation of Old Norse (standard)

by skar Gulaugsson 2000
The most effective way of identifying sound values in brevity is the use of phonetic alphabets such as the International Phonetic Alphabet (IPA), which uses various characters from the Latin and Greek alphabets, capital and small, to represent all the different sounds enunciable by human beings. There are sounds in the IPA chart that most people will never have heard of, as the range of sounds used in human speech is enormous.
The main problem with the IPA system is that, with its heavy use of unconventional characters, phoneticizing with it is practically impossible in computer documents; no normal keyboards support the less conventional IPA characters, and even if one downloads the fonts available for it, the sheer amount of different IPA symbols makes it impossible to type it fast enough. Not to mention problems with converting the results onto other people's computer screens.
The solution to this has been the adoption of a new version of IPA, one whose characters are all available on "normal" keyboards (such as American ones). Instead of interchanging Greek, Latin, and various arcane letters, this new computer standard, called SAMPA, fully exploits differentiation by capitals and small caps, and also uses special characters such as { } @ & etc, as well as numbers.
For the purposes of this course, we will only present here relevant characters, i.e. those representing sounds in Modern English and Old Norse. To begin with, here are "standard" English sounds (both General American (GA) and Received Pronunciation (RP)):

(in phonetics, [] is always used to contain phonetic transcriptions)

[:] following a vowel represents that vowel being long, i.e. of roughly double duration to short of vowels (do not confuse with common English definitions of 'long a as in date, short as in cat'). Diphthongs (two different vowels combined) are long by default.
Orthographic	as in...	SAMPA	Notes

'a', 'ai',	'date'
 or 'ay'	'wait'		[eI]
'a', 'ai'	'fair' 				
		'hair'		[E:]
'e', 'ee',	'Pete'
 'ea', etc	'feet'		[i:]
'i' or 'igh'    'bite'
		'high'		[aI]
'o' or 'oa'	'note'
		'boat'		[@U]
'u' or 'ew'	'cute'
		'few'		[ju:]
'u' or 'ew'	'Sue'			
		'sew'		[u:]	after [s, n, l, r]
'oo'		'tool'		[U:]
'oo'		'foot		[U]
'ou' or 'ow'	'louse'
		'brown'		[AU]
'oi' or 'oy'	'void'
		'boy'		[OI]
'au'		'cause'		[O:]
'a'		'pat'		[{]
'e'		'pet'		[e]
'i'		'pit'		[I]
'o'		'pot'		[Q]
'u'		'cut'		[V]
any vowel	'a' in			"schwa"; the weak 
		'another'	[@]	unstressed vowel that
					people have trouble
'ur'		'fur'		[3:]	in non-rhotic
					dialect (without final
					'r' sound, as in RP)
'ar'		'star'		[A:]	non-rhotic dialect
'ere' or 'ear'	'ear'		[I@]  	non-rhotic
'air'		'air'		[e@]	non-rhotic
'ure'		'cure'		[jU@]	non-rhotic
I will not go into the proper transcription of the last few cases in rhotic dialects; it is not relevant to the study of Old Norse.
It is very important to differentiate capitals vs small-caps in SAMPA (cannot be said too often). [A] for example is a different sound to [a] (the latter being closer to [O]). It is also very important to note that anything within [] represents phonetic transcription; that must therefore not be read as English orthography!
You will probably be finding the SAMPA transcription for English rather "bulky" or inconvenient; that is normal, since the transcription would not do well to define itself by English sounds, which aren't necessarily "standard" in comparison to other languages. Remember, phonetics is the science of all languages, not only English.
Regarding some English consonants:
Orthographic	as in...	SAMPA		description

'p'		'pick'		[p^h]		[^h] represents aspiration
'p'		'spoil'		[p]		not aspirated after [s]
't'		'tick'		[t^h]		
't'		'store'		[t]		
'c' or 'k'	'cat'		[k^h]		
'c' or 'k'	'skull'		[k]		
'k' before	'kin'		[c^h]		"palatalized" 'k', i.e. 
  'i' or 'e'					pronounced more forward
'k' before	'skin'		[c]
  'i' or 'e'
'b'		'bay'		[b]		voiced
'd'		'day'		[d]		"
'g'		'good'		[g]		"
'g'            	'gill' 		[C]             voiced ("soft") version of [c]
'g'		'gin'		[dZ]		voiced version of 'ch' [tS]
'ch'		'chin'		[tS]		unvoiced affricate
'th'            'that'          [D]             voiced
'th'            'thick'         [T]             unvoiced
'j'	  	'jest'          [dZ]		voiced version of 'ch' [tS]
'y'             'you'           [j]             *note* that phonetic [j] is not English 'j'!
'w'             'win'           [w]                
'wh'            'white          [w^0]           unvoiced, only in some dialects
'v'             'vine'          [v]
'ng'            'young'         [N]
's'             'see'           [s]             unvoiced
's'             'house'         [z]		voiced
[^0] as in [w^0] means devoicing.
And now to the sounds of Old Norse, starting with vowels:
Once again, remember that phonetic vowel length, represented by [:], has nothing to do with English definitions of long vowels, like 'a as in date'.
Orthographic	SAMPA	Notes

              [A:]     note the length mark; not a 			common sound in English
a              [A]      dialects; see 'ar' above, also like
			Spanish 'a', for example
e              [e]
i              [i]
              [o:]     does not exist on its own in
			English; like 'o' in Spanish, 
o              [o]      French, German, and many others
u              [u]
              [y:]     this is the dreaded 'rounded
			[i]'; if you lack Old Norse
y              [y]      samples for this, look for French
			pronunciation of 'u',
                        or German '', which are also
			pronounced [y]; it is pronounced by rounding
			the lips (as when one pronounces [u] or [o])
			and saying [i]
 (i.e. tailed o)
oe	        [2:]    doesn't exist in English; look
			for 'eu' as in 'feu' in French
au              [Au]    trivial difference to English [AU];
			more like Spanish 'au'
ei              [Ei]    similar to Spanish 'ei'
ey              [9y]    once again, look to French: like
			'euille' in 'feuille'
On to the consonants:
Orthographic    ...as in...	SAMPA	Notes

p		prestr		[p^h]	aspirated as in English
p               sp             [p]     same pattern as in English,
					no aspiration after [s]
p               api             [p]     between vowels, also not
					aspirated (unlike English)
t               tl             [t^h]
t               standa		[t]
t               hata            [t]     not aspirated in intervocalic
					position, like [p]
k               kpa            [k^h]
k               sk             [k]
k               aka             [k]
k               kerti           [c^h]
k               skella          [c]
k               haki            [c]
b               bor            [b]
d               dagr            [d]
g               g             [g]           
g               geta            [C]    	like English 'gill'
gj              gjta           [C]
g               saga            [G]     between vowels; like
					'g' in Spanish 'fuego'
               ykk            [T]
               ea             [D]
j               jl             [j]     not [dZ] as English 					'j', but [j] as 'y'
v               vandi           [w]     
hv	        hvt		[w_0]
f               f  	        [f]            
f               hafa            [v]
h               hafa            [h]
r               rs             [r]     trilled 'r', as in
					Spanish; *not* to be pronounced
					like standard English 'r'
hr              hrs            [r^0]   unvoiced [r]; as an
					approximation, try saying [r]
					exhaling forcefully at the
					same time, as if saying [h];
					consider also the relationship
                                        between [w] and [w^0], as
n               n              [n]
ng              ung             [Ng]    unlike English, the 'g'
					is also pronounced
hn              hnefi           [n^0]    same relationship as
					between [r] and [r^0]
l               l              [l]     always as Spanish
					'l', not like English 'l'
					in 'cold'
hl              hl             [l^0]   unvoiced, like [r^0]
s               s              [s]     never [z]
x               sax             [ks]    like English
pt              krapt           [ft]           
kt              rakt            [xt]    [x] is like 'ch' in
					Scottish 'loch', or 'j'
					in Spanish 'ojo'
gt              sagt            [Gt]
gg              hagga           [gg]    see below
Double consonants: pronounced with double length, i.e. 'alla' is [AllA]. This is not normally done in English, though sequences of words may produce double consonants (geminates), e.g. 'black cow' [bl{k kAU]. If you pronounce 'black cow' as one word, you have [kk] in the middle.
Stress in Old Norse always falls on the first syllable, or rather, all stems are stressed. ON words are always just <stem> + <ending>, stems never being more than one syllable; longer words, i.e. with more than one stem syllable, are composed. In a composed word, e.g. 'fingrbjrg', stress falls primarily on the first syllable (the first stem), 'fing', and secondarily on the next stem, 'bjrg'. Endings only contain the short vowels 'a, i, u'; syllables with long (accented) vowels or diphthongs can thus always be recognized as stem syllables, and therefore entitled to stress.

We have not described the pronunciation of modern Icelandic. Guides can be found on the Net. This one is fairly good.

Icelandic Phrase Page[Link rotten.]