The Pronunciation of Old English
Introduction
The sounds of Old English should not prove difficult, with a few exceptions, for speakers of Modern English. It can be hard at first to get used to some of the spelling conventions, such as the fact that all letters—including final e—are pronounced; but on the whole Old English does not have many sounds that are not the same as in Modern English, and, in most cases, indicated by the same letters (you can read a brief tutorial on Old English script here).
This is particularly true of the short vowels and the consonants, most of which are thought to have been largely the same as their Modern English equivalents. The lax e /ɛ/ in Modern English edge, for example, is probably not all that different from the short e [ɛ] the word’s Old English ancestor, ecg [1]. Likewise the d [d] in Old English dysiġ (‘foolish’), was pronounced much the same as the same letter in dizzy, the Modern English descendant of dysiġ.
There are two major exceptions to this. The first involves the Old English long vowels, which underwent great changes in the transition from late Middle to early Modern English (roughly the period between Chaucer and Shakespeare). The second is the existence of a few vowel and consonant sounds that are rare or no longer exist in Modern English or that fail to reflect distinctions we now consider important. These include the sounds associated with the letter y, some of the sounds associated with c, g, and h, and the fact that the sounds we now associate with s and z and with f and v do not seem to have been seen as distinct meaningful sounds (phonemes) in Anglo-Saxon England.
The following sections discuss each of the major groups of sounds in turn. In each case you are given some examples of the sound in Old and Modern English and a sound clip illustrating the sound in and Old English context. At the end of each section there is an exercise, suitable for downloading to an Digital Audio Player (such as an iPod or similar), that allows you to practice the sounds out loud while comparing yourself to a reference sample.
Consonants
With a very few exceptions, the Old English consonant system is essentially identical that of Old English. Hence the sound spelled by the Old English letter b was pronounced more or less as is that spelled by our modern b: Old English bār, Modern English boar (i.e. wild pig).
Even when Anglo-Saxons used different characters or spellings, the actual sounds were mostly the same:
- The sounds represented by the Anglo-Saxon letters þ and ð were pronounced as are the sounds represented by their Modern English equivalents: th (as in then [ð] and thigh [θ] ).
- ƿ represented the same sound as Modern English w [w], e.g. Old English weġ, Modern English way.
- sc was used for sounds equivalent to those represented by sh [ʃ] (and sometimes sk [sk]) in Modern English: Old English scip; Modern English ship; Old English āscian, Modern English to ask.
- cg was used to represent the sound we now generally write using dge [ʤ], e.g. Old English bricg, Modern English bridge.
- f was used to represent both the sound we usually write as f [f] and the sound we often write as v [v]: Old English wīf (‘woman’), Modern English wife; Old English wīfum (‘to/by/for/with the women’), Modern English wives:
- f was unvoiced (i.e. spoken like an f) when it appeared on its own at the beginning or end of a word, or in a cluster of unvoiced consonants (i.e. with consonants like t, p, s, k);
- it was voiced (i.e. spoken like a v) when it was found by itself between vowels, or in a cluster of voiced consonants (i.e. alongside consonants like d, b, z, g, m, n).
- s was used to represent the sounds we usually spell using s [s] and z [z]: Old English seġl, Modern English sail; Old English cēosan, Modern English to choose.
- Like f, s was unvoiced (i.e. spoken like an s) when it appeared on its own at the beginning or end of a word, or in a cluster of unvoiced consonants (i.e. with consonants like t, p, s, k).
- it was voiced (i.e. spoken like a z) when it was found by itself between vowels, or in a cluster of voiced consonants (i.e. alongside consonants like d, b, z, g, m, n).
- The letters c, g, and h represented a number of sounds we now spelling using different characters, depending on their context and etymology:
- c could represent the sounds we commonly spell using k [k] (Old English cyning, Modern English king) or ch [ʧ] (Old English ciese, Modern English cheese).
- g]* was used for the sounds we now often spell as g [g] (Old English god, Modern English God); the sound we now often spell using y or j [j] (Old English geard, Modern English yard), and a sound that is no longer found in Modern English, but is similar to Dutch g [ɣ] in words like Old English dragan, Dutch dragen, Modern English to draw (i.e. pull, drag) and Old English būgan, Dutch bugen, Modern English to bow. As these two examples suggest, this sound became Modern English /w/, commonly spelled with a w, in the later history of the language.
- h]* could represent either the sound in Modern English head [h] (Old English hēafod) or a sound similar to the ‘Scottish’ or ‘German’ ch [x] as in Scottish Loch Lomand, German ich: Old English cniht, Modern English knight. As the Modern English spelling suggests this form was usually spelled gh in Middle English and disappeared or changed to a sound like f in Modern English.
Old English also distinguished between long and short consonants. This distinction, which is found in Modern Italian, has no equivalent in Modern English. When pronouncing a long consonant, which were usually written as double consonants, try to hold the consonant longer than you would for a regular (i.e. short) consonant.
| Consonant | IPA Value1 | Old English Example | Comments |
| b | b | bār, ‘boar’ | |
| c | k | cyning, ‘king’ | Often spelled with a k in Middle and Modern English |
| ʧ | ċīese | The dot above the letter for this sound is a modern convention not found in the manuscripts. | |
| cg | ʤ | ecg, ‘edge’ | |
| d | d | dysiġ, ‘foolish’ | |
| f | f | wīf, ‘woman’ | |
| v | wīfum, ‘to/by/for/with the women’ | ||
| g | g | gār, ‘spear’ | |
| ɣ | būgan, ‘to bow’ | These sounds are commonly spelled with a w in modern English. | |
| j | ġeard, ‘enclosure, yard’ | The dot above the letter for this sound is a modern convention not found in the manuscripts. | |
| h | h | hlūde, ‘loud’ | Initially |
| x | cniht | Medially and finally; often spelled gh in Middle and Modern English | |
| l | l | lār, ‘teaching, lore’ | &#:x00A0; |
| m | m | miċel, ‘great, big; Yorkshire: mickle’ | |
| n | n | nȳd, ‘necessity’ | |
| ŋ | sang | When in combination with g, say the sound as in modern English -ing words, but then also add the g. | |
| p | p | pleoh, ‘danger, risk’ | |
| r | r | rōf, ‘strong’ | Rolled. |
| s | s | seġl, ‘sail’ | |
| z | cēosan, ‘to choose’ | ||
| sc | ʃ | æsc, ‘ash’ | |
| sk | ascian, ‘to ask’ | Many (but not all) words with sk in Modern English are of Norse original while those with sh are from Old English; cf. shirt (from Old English) and skirt (same word, but from Norse) | |
| t | t | til, ‘good’ | |
| þ or ð | θ | þær (or ðær), ‘there’ | |
| ð | cweðan (or cweþan), ‘to say’ |
Vowels
Old English made a quantitative and probably qualitative distinction between long and short vowels. This means that vowels were distinguished by how long they were held as much or more than by differences in how they sound. The vowel in Old English god (Modern English God) was certainly shorter and probably had a different sound (though not as dramatically different a sound as its Modern English descendent has) than gōd (Modern English good); the Old English words āc (Modern English oak) and ac (‘but, and’) differed primarily in length.
In Modern English, we no longer recognise length as a meaningful distinction between vowels. Instead we recognise differences in tenseness and quality. What are often called “short” vowels in Old English are really lax; so-called “long vowels” are actually tense. Fortunately, however, Modern English short vowels correspond quite closely to Old English short vowels; and while Modern English tense vowels sound quite different from Old English long vowels, our set of tense vowels matches the Old English set of long vowels.
Old English did have one set of vowels, at least in the earlier periods, which we no longer have: front rounded vowels, long and short, written as y. This is similar to an umlauted u in German, or the u in French tu. To form it, begin by saying (and holding) a long or short i. While you are saying the i, round your lips as if you were saying a u. When your lips are rounded, you’ll have the sound. Listen to the reference examples to practice—but even if you can’t do it in the end, take heart; by the end of the Anglo-Saxon period, the sounds were more or less indistinguishable from the corresponding form of i.
Long vowels
Because length is not a quality we hear naturally any more as native speakers of English, it is the quality of the long vowels that probably is most striking to our ears. In addition to being longer than their Modern English descendants, Old English long vowels also tend to have the values associated with them in continental languages like French, Dutch, German, or Italian, rather than their Modern English sounds:
| Old English Vowel | IPA Symbol1 | Mnemonic | Old English Example |
| ȳ | yː | French ruse | cȳþþ ‘kith, friends’ |
| ī | iː | beat | wīte, ‘punishment’ |
| ē | eː | bait | ēþel, ‘homeland, territory’ |
| ā | ɑː | aunt (non-Canadian pronunciation) [2] | stān, ‘stone’ |
| ǣ | æː | bat | bæde, ‘bade, i.e. asked, commanded’ |
| ō | oː | boat | gōd, ‘good’ |
| ū | uː | boot | brūcan, ‘to enjoy’ |
Listen to the Long vowels and example words
Short Vowels
As noted above the short vowels correspond very closely, with the exception of y, to our Modern English lax vowels:
| Old English Vowel | IPA Symbol1 | Mnemonic | Old English Example |
| y | y | French tu | cynn ‘kin, family’ |
| i | ɪ | bit | scip, ‘ship’ |
| e | ɛ | bet | ecg, ‘ecg, sword’ |
| a | ɑ | father | pað, ‘path’ |
| æ | æ | bad | cræft, ‘skill or trade’ |
| o | ɔ | bought | god, ‘God’ |
| u | ʊ | book | sunu, ‘son’ |
Listen to the short vowels and example words
Diphthongs
Old English had three main sets of long and short diphthongs: ea*/[*ēa], eo*/[*ēo], and ie*/[*īe]:
| Old English Diphthong | IPA Symbol1 | Equivalent to | Old English Example | |
| ea | æa | æ+a | ||
| ēa | æːa | ǣ+a | ||
| eo | ɛɔ | e+o | ||
| ēo | eːɔ | ē+o | ||
| ie | ɪɛ | i+e | ||
| īe | iːɛ | ī+e |
The Old English diphthongs were strongly falling. That is to say they gave a much stronger prominence to the first part of the glide than the second. By the late Old English period, all were probably equivalent to long and short æ, e, and i, respectively.
Exercise
The following podcast contains an exercise that will let you practice your Old English consonants, long and short vowels, and diphthongs.
You complete the exercise as follows:
- Listen to the Old English sound, example word, and Modern English translation
- In the pause that follows this, trying saying the Old English sound and example word yourself
- Listen again as the sound, Old English word, and Modern English translation is repeated
- Then try the Old English sound and word again, correcting any problems.
- If you are still having difficulty, repeat the exercise for the sound and word, before moving on to the next.
Listen to an exercise involving long vowels
Listen to an exercise involving short vowels
Notes
1 Symbols in square or slanted brackets (e.g. /ɛ/ or [ɛ]) are notations for the relevant sound in IPA. Not all students will understand this system, so I have also provided spelling pronunciations based on North American English. If you would like to hear the sounds associated with the IPA symbols, a good site to visit is the following: http://www.phonetics.ucla.edu/course/chapter1/chapter1.html (thanks to Lee Ann Schneider for the link).
2 In Canadian English aunt (i.e. the sister of your father or mother) sounds the same as (is a homophone of) ant (i.e. the creepy crawly thing). Old English ā sounds like the first sound in aunt when spoken by people who can hear the difference between aunts and ants.
Comment [10]
Kyungsik Joe from S.Korea (Thu Oct 16, 2008 (21:28:58)) [PermLink]: You said that ‘H’ of old english is pronounced as a [x]after vowel ‘i’ but it’s actually not [x] but a voiceless velar fricative(or palatal fricative). Check it up.
dan (Sun Oct 19, 2008 (11:15:51)) [PermLink]: Thank you Kyungsik Joe for pointing out an error on the site. I meant to indicate a voiceless velar fricative, but it turns out I've been using the wrong IPA symbol (unfortunately for quite a while), so I'm glad for the correction.
Since this is a teaching site, I hope you'll forgive me for pointing out that there is an error in your comment (the opposite from mine, actually): [x] is the symbol for voiceless velar fricative; the problem was that my site originally had chi [χ], which is the symbol for voiceless uvular fricative. So you are right I was indicating the wrong sound, but the character you list in your comment (which is not the one I originally had) is in fact the correct symbol for the sound I should have (and meant to be) indicating. Whew! Confusing!
So to sum up: the part of the site in question has now been corrected: the sound of the letter h in contexts like cniht is a voiceless velar fricative and this sound is indicated by the IPA symbol [x].
Thanks very much again for pointing this out!
Me (Fri Jun 12, 2009 (23:43:58)) [PermLink]: [x] had an allophone [ç] after front vowels.
University of Bergen, Engl 303 (Thu Aug 27, 2009 (02:22:56)) [PermLink]: Thank you for the helpful guide. We’d like to add that Old English when pronounced [ɣ] is voiced.
dan (Tue Sep 1, 2009 (08:37:05)) [PermLink]: Thanks very much. You are right. The problem is actually that I’m wrong about how Dutch is pronounced: everybody I know who is Dutch claims it is voiceless—except maybe in Belgium. So OE can be /ɣ/; Dutch should be /x/. In addition to being an obscure example, it turns out it’s just wrong, too.
Ali (Sun Dec 13, 2009 (03:31:12)) [PermLink]: How can you know if y (Aepelbryht, norphyte) is pronounced “Ü” or “i” ?
dan (Mon Dec 14, 2009 (08:30:34)) [PermLink]: Ali,
It depends a little on the period: early on in the period, y is used for a rounded front vowel that is distinct from the unrounded version. Later in the period, y and i become identical, if we are to judge from the spelling confusions in manuscripts.
Ehsan ehsan_dik@yahoo.com (Sun Jul 10, 2011 (22:49:49)) [PermLink]: i need help in describing the each english consonanat
Adrienne (Tue Sep 13, 2011 (12:27:57)) [PermLink]: Hi Dan, really interesting and accessible guide. But your comments about short and long vowels don’t ring true. Native American speakers indeed do not seem to hear this distinction – I was amazed when a friend of mine could not hear the distinction between “Mary” and “Merry”! But – I can assure you there are a good few long and short vowels alive and kicking among native English speakers.
dan (Thu Sep 15, 2011 (18:41:21)) [PermLink]: Hi Adrienne,
Thank you very much for your nice comments. I’m glad it was useful.
Since I am teaching mostly North American (more specifically mostly Western Canadian) students, I’ve written the guide to fit their pronunciation habits.
But more importantly, when I’m talking about length (i.e. long and short vowels, above), I’m using the term in the technical sense of "meaningful linguistic contrast involving duration as opposed to quality."
In that specific sense, Modern English (in all dialects) really is usually said to distinguish meaningfully ("phonemically") by quality rather than quantity. So speaking very technically, we don’t have a distinction between long and short vowels, even if they sometimes some _sound_ longer and we common speak informally of "long" and "short" vowels.
In fact the sounds we informally call "long" and "short" actually differ meaningfully (i.e. "phonemically") in terms of how they sound rather than how long they sound. You might say “short e” more quickly than you would say “long e,” but the important difference is actually that they are said in slightly different places in your mouth and with a difference in relative tenseness or laxness. The sound in merry (which we write as /ɛ/ in the phonetic alphabet) is relatively speaking lower and maybe a little farther back than the sound in Mary (which we write as /e/). But the following /r/ can make this hard to hear for most people. You might hear the difference more clearly in examples like bet vs. bait.
This is not to say that you are not hearing a difference in length; but rather that this difference in length is not phonemically significant. Not all phonetic differences are phonemically meaningful--in fact most are not. The /k/ sound in cut is different in most varieties of English from the /k/ sound in kit (they are usually said with the tongue in different positions). But although this distinction can be meaningul in other languages, it is not in English: both are just different phonetic realisations of the same phoneme /k/.
There is a Canadian example that neatly illustrates this distinction between phonemic and phonetic difference. Canadians can hear a distinction between rider and writer that other North Americans can’t. But they do not rely on the thing that most non-North Americans use to tell them apart (i.e. a difference in phonetic realisation of /d/ and /t/): in Canadian English, as in U.S. English, the middle consonant of both words is pronounced as a “flap”--meaning that in both varieties of English, it sounds like a [d] in both words (compare standard American/Canadian and British pronunciations of butter, where the same thing happens).
Instead, in Canadian English, the words sound difference because of the realisation of the initial diphthong. In Canadian English, the diphthong /aI/ is "raised" to /əI/ whenever it is found before a phonemically voiceless consonant, but pronounced without raising whenever it is found before a phonemically voiced one. In the case of writer and rider, Canadians apply this raising to writer but not rider, even though the realisation of the medial consonant sounds the same (and voiced) in each word.
What this suggests is that in Canadian English, the medial consonant is phonetically voiced, but phonemically voiceless--or in other words, they say it with voicing, but they think of it as voiceless. If they didn't, they wouldn't apply Canadian raising. I have often shown this is class by saying rider and rider in a variety of different orders and asking students to indicate by hand movements whether they think I mean a person who works with a pen or rides a horse: Canadian students always give the right signal for the word I am intending to say; most of my American students have a much harder time figuring out which one I mean, if they can get it at all.
Thanks again for some great examples and your kind words!
Commenting is closed for this article.
