Top Contents Bibliography 2.1. Quick start
2.1.1. Vowels and diphthongs 2.1.2. Consonants 2.1.3. Sermonette
2.2. More about vowels
2.2.1. Short a, æ and ea 2.2.2. I-mutation 2.2.3. Silent e and o for u
2.3. More about c and g 2.4. Syllable length 2.5. Accentuation 2.6. On-line pronunciation practice 2.7. Summary

2. Pronunciation

2.1. Quick start

No one knows exactly how Old English sounded, for no native speakers survive to inform us. Rather, linguists have painstakingly reconstructed the pronunciation of the language from various kinds of evidence: what we know of Latin pronunciation (since the Anglo-Saxons adapted the Latin alphabet to write their own language), comparisons with other Germanic languages and with later stages of English, and the accentuation and quantity of syllables in Old English poetry. We believe that our reconstruction of Old English pronunciation is reasonably accurate; but some aspects of the subject remain controversial, and it is likely that we will never attain certainty about them. The greatest Old English scholar in the world today might very well have difficulty being understood on the streets of King Alfred’s Winchester.

Despite the uncertainties, you should learn Old English pronunciation and get into the habit of reading texts aloud to yourself. Doing so will give you a clearer idea of the relationship between Old and Modern English and a more accurate understanding of Old English meter, and will also enhance the pleasure of learning the language.

If you find any of the terminology or the phonetic symbols in this chapter unfamiliar, you should consult Appendix B, “Phonetic Symbols and Terms”.

2.1.1. Vowels and diphthongs

Old English had six simple vowels, spelled a, æ, i, o, u and y, and probably a seventh, spelled ie. It also had two diphthongs (two-part vowels), ea and eo. Each of these sounds came in short and long versions. Long vowels are always marked with macrons (e.g. ā) in modern editions for students, and also in some scholarly editions. However, vowels are never so marked in Old English manuscripts.

When we speak of vowel length in Old English, we are speaking of duration, that is, how long it takes to pronounce a vowel. This fact can trip up the modern student, for when we speak of “length” in Modern English, we are actually speaking of differences in the quality of a vowel. If you listen carefully when you say sit (with “short” i) and site (with “long” ī), you’ll notice that the vowels are quite different: the “short” version has a simple vowel [ɪ], while the “long” version is a diphthong, starting with a sound like the u in but and ending with a sound like the i in sit [ʌɪ].[1] The same is true of other long/short pairs in Modern English: they are always qualitatively different. We do give some vowels a longer duration than others (listen to yourself as you pronounce beat and bead), but this difference in duration is never significant: that is, it does not make a difference in the meaning of a word. Rather, we pronounce some vowels long and others short because of the influence of nearby sounds.

This book frequently uses symbols from the International Phonetic Alphabet (IPA) for convenience of reference, though it also gives examples wherever possible. For a table of the IPA symbols relevant to the study of Old English, see Appendix B.

Vowel length (that is, duration) is significant in Old English because it does make a difference in the meanings of words. For example, Old English is means ‘is’ while īs means ‘ice’, ac means ‘but’ while āc means ‘oak’, and ġe means ‘and’ while ġē means ‘you’ (plural). The significance of length means that the macrons that appear in the texts you will be reading are not there only as guides to pronunciation, but also to help you decide what words mean. If you absent-mindedly read mǣġ ‘kinsman’ as mæġ ‘may’, you will never figure out the meaning of the sentence you are reading.

Simple vowels

The following list of vowels deals with quality only; you may assume that the short and long vowels sound alike except for a difference in duration. The list cites a number of Modern English words for comparison: these are from the Mid-Atlantic dialect of American English and may not be valid for speakers of British English or other American dialects.

a is pronounced [ɑ], as in Modern English father. Examples: macian ‘make’, bāt ‘boat’.
æ is pronounced [æ], as in Modern English cat. Bæc ‘back’, rǣdan ‘read’.
e is pronounced [e], as in Modern English fate; that is, it is like the e of a continental European language, not like the “long” or “short” e of Modern English (actually [i] or [ɛ]). Helpan ‘help’, fēdan ‘feed’.
i is pronounced [i], as in Modern English feet; that is, it is like the i of a continental European language, not like the “long” or “short” i of Modern English (actually [ʌɪ] or [ɪ]). Sittan ‘sit’, līf ‘life’.
o is pronounced [o], as in Modern English boat. God ‘God’, gōd ‘good’.
u is pronounced [u], as in Modern English tool; it is never pronounced [ʌ] as in Modern English but. Full ‘full’, fūl ‘foul’.
y is pronounced [y], like the ü in German über or Füße, or like the u in French tu or dur. Make it by positioning the tongue as you do to say feet while rounding the lips as you do to say tool. Cyning ‘king’, brȳd ‘bride’.
ie which appears mainly in early West Saxon, is difficult to interpret. It was probably approximately [ɪ], like the i of Modern English sit. In late West Saxon, words that contained this vowel are rarely spelled with ie, but rather with i or y. Ieldesta ‘eldest’, hīeran ‘hear’.

In unaccented syllables, where few vowel sounds were distinguished, vowels were probably pronounced less distinctly than in accented syllables. In late Old English (ca. 1000 and later), frequent spelling confusion shows that by then the language was beginning to approach the Middle English situation in which all vowels in unaccented syllables were pronounced [ə] (a neutral schwa, like the a in China). But unaccented vowels were distinguished in Old English, and it is important to pronounce them, for vowel quality often is the only thing that distinguishes one ending from another. For example, dative singular cyninge and genitive plural cyninga, genitive singular cyninges and nominative plural cyningas are distinguished only by vowel quality.

Diphthongs

Old English has two digraphs (pairs of letters) that are commonly interpreted as diphthongs: ea and eo.[2] Both ea and eo can represent short or long sounds, equivalent in length to the short and long vowels. Beyond this generally agreed fact, there is controversy about what sound these digraphs represent. Here we present the most widely accepted view.

A digraph io appears primarily in early texts, and for the student’s purposes is best taken as a variant of eo.
eo represents [eo] or [eʊ], a diphthong that started with [e] and glided to a rounded sound, [o] or [ʊ]. Examples: ċeorl ‘freeman’ (Modern English churl), dēop ‘deep’. 
ea represents [æɑ], a diphthong that started with [æ] and glided to [ɑ] (as in father). Feallan ‘fall’, rēad ‘red’.

Some grammar books say that the spelling ie also represents a diphthong, but this book interprets it as a simple vowel.

Perhaps the most common error students make when trying to pronounce Old English diphthongs is to break them into two syllables—for example, to pronounce Bēowulf as a three-syllable word when in fact it has only two syllables. Remember that there is a smooth transition between the two vowels of a diphthong, and this is as true of the unfamiliar diphthongs of Old English as it is of the familiar ones of Modern English (like those of site and sound).

2.1.2. Consonants

Most Old English consonants are pronounced as in Modern English, and most of the differences from Modern English are straightforward:

  1. Old English scribes wrote the letters þ (“thorn”) and ð (“eth”) interchangeably to represent [θ] and [ð], the sounds spelled th in Modern English. Examples: þing ‘thing’, brōðor ‘brother’. 
  2. There are no silent consonants. Old English cniht (which comes to Modern English as knight) actually begins with [k]. Similarly hlāf (Modern English loaf) and hring (ring) begin with [h], gnæt (gnat) with [ɡ], and wrīðan (writhe) with [w]. Some Old English consonant combinations may be difficult to pronounce because they are not in Modern English. If you find this to be so, just do your best. 
  3. The consonants spelled f, s and þ/ð are pronounced as voiced [v], [z] and [ð] (as in then) when they fall between vowels or other voiced sounds. For example, the f of heofon ‘heaven’, hæfde ‘had’ and wulfas ‘wolves’ is voiced. So are the s of ċēosan ‘choose’ and the ð of feðer ‘feather’. This distinction remains not only in such Modern English singular/plural pairs as wolf/wolves, but also in such pairs as noun bath and verb bathe, noun cloth and derivative clothes.
  4. These same consonants were pronounced as unvoiced [f], [s], and [θ] (as in thin) when they came at the beginning or end of a word or adjacent to at least one unvoiced sound. So f is unvoiced in ful ‘full’, cræft ‘craft’ and wulf ‘wolf’. Similarly s is unvoiced in settan ‘set’, frost ‘frost’, and wulfas ‘wolves’, and þ/ð is unvoiced in þæt ‘that’ and strengð ‘strength’. 
  5. When written double, consonants must be pronounced double, or held longer. We pronounce consonants long in Modern English phrases like “big gun” and “hat trick,” though never within words. In Old English, wile ‘he will’ must be distinguished from wille ‘I will’, and freme ‘do’ (imperative) from fremme ‘I do’. 
  6. This book sometimes prints c with a dot (ċ) and sometimes without. Undotted c is pronounced [k]; dotted ċ is pronounced [ʧ], like the ch in Modern English chin. This letter is never pronounced [s] in Old English. It has a special function in the combination sc (see item 10 below). 
  7. The letter g, like c, is sometimes printed with a dot and sometimes without. Dotless g is pronounced [ɡ], as in good, when it comes at the beginning of a word or syllable. Between voiced sounds dotless g is pronounced [ɣ], a voiced velar spirant.[3] This sound became [w] in Middle English, so English no longer has it. Dotted ġ is usually pronounced [j], as in Modern English yes, but when it follows an n it is pronounced [ʤ], as in Modern English angel.
  8. The combination cg is pronounced [ʤ], like the dge of Modern English sedge. Examples: hrycg ‘ridge, back’, brycg ‘bridge’, ecg ‘edge’. 
  9. Old English h is pronounced [h], as in Modern English, at the beginnings of syllables, but elsewhere it is pronounced approximately like German ch in Nacht or ich—that is, as a velar [x] or palatal [ç] unvoiced spirant (pronounced with the tongue against the velum [soft palate] or, after front vowels, against the hard palate). Examples: nēah ‘near’, niht ‘night’, þēah ‘though’, dweorh ‘dwarf’. 
  10. The combination sc is usually pronounced [ʃ], like Modern English sh: scip ‘ship’, æsc ‘ash (wood)’, wȳscan ‘wish’. But within a word, if sc occurs before a back vowel (a, o, u), or if it occurs after a back vowel at the end of a word, it is pronounced [sk]: ascian ‘ask’ (where sc was formerly followed by a back vowel), tūsc ‘tusk’. When sc was pronounced [sk] it sometimes underwent metathesis (the sounds got reversed to [ks]) and was written x: axian for ascian, tux for tusc. Sometimes sc is pronounced [ʃ] in one form of a word and [sk] or [ks] in another: fisc ‘fish’, fiscas/fixas ‘fishes’.
Practice making this sound: Raise the back of your tongue to the velum (the soft palate) as you do when pronouncing a k. Instead of a stop, though, pronounce a spirant, somewhat like the ch of German Nacht, but voiced. If you are sure you cannot pronounce the [ɣ], pronounce it [w] instead.

2.1.3. Sermonette

When students of Old English go wrong in translating, it is often because they have done a sloppy job of looking up words in a dictionary or glossary. Remember, when you look up words, that vowel length is significant, and so is the doubling of consonants. Biddan ‘ask, pray’ and bīdan ‘await, experience’ are completely different words, but some students mess up their translations because they look at them as equivalent. Don’t fall into this trap!

On a related point, you will notice as you go along that the spelling of Old English is somewhat variable. Scribes at that time lacked our modern obsession with consistency. Rather than insisting that a word always be spelled the same way, they applied a set of rules for rendering the sounds of their language in writing, and these rules sometimes allowed them to get the job done in more than one way. Further, scribes sometimes mixed up the dialects of Old English, writing (for example) Mercian þēostru ‘darkness’ instead of West Saxon þīestru. These minor inconsistencies sometimes lead students to believe that anything goes in Old English spelling, and this belief leads them into error.

It is not true that anything goes in Old English spelling. Though you will have to get used to frequent variations, such as ie/i/y and iung for ġeong ‘young’, you won’t often see confusion of æ and ea, or indeed of most vowels, or of single and double consonants, or of one consonant with another. For a list of spelling variants that you will frequently see, consult Appendix A.

Get into the habit of recognizing the distinctions that are important in Old English and doing an accurate job of looking up words, and you will avoid a lot of frustration.

2.2 More about vowels

2.2.1. Short a, æ and ea

The short sounds spelled a, æ and ea are all derived from the same vowel (spelled a in most other Germanic languages). The split of one vowel into two vowels and a diphthong, which occurred before the period of our written texts, was conditioned by the sounds that surrounded it in the word (the details are complex and controversial: see Lass 1994, pp. 41-53). The effects of this split were not long-lasting; by the Middle English period a, æ and ea had coalesced into one vowel, spelled a.

The reason it is important for you to know about the relationship of a, æ and ea is that these sounds vary within paradigms. If æ or ea occurs in a short syllable and a back vowel (a, o, u) follows, the æ or ea becomes a. Add the plural ending -as to dæġ ‘day’ and you get dagas; add plural -u to ġeat ‘gate’ and you get gatu.

2.2.2. I-mutation

I-mutation is a shift in the quality of a vowel so that it is pronounced with the tongue higher and farther forward than usual—closer to its position when you pronounce the vowel [i] (as in feet).[4] The correspondences between normal and mutated vowels are shown in table 2.1. Notice that the i-mutation of a produces a different result depending on whether a nasal consonant (m or n) follows.

German linguists call it Umlaut. Because of the great influence of German linguistics at the time when the historical evolution of the Germanic languages was being worked out, you will occasionally see this term even in grammars written in English.
Table 2.1. i-mutation
short long
unmutated mutated unmutated mutated
a æ ā ǣ
an/am en/em  
æ e
e i
ea ie (i, y) ēa īe (ī, ȳ)
eo ie (i, y) ēo īe (ī, ȳ)
o e ō ē
u y ū ȳ

The effects of i-mutation are still evident in Modern English. The vowels of such athematic plurals as men (singular man), lice (louse) and teeth (tooth) exhibit i-mutation, as does the comparative adjective elder (old); and i-mutation accounts for most of the verbs that both change their vowels and add a past-tense ending (e.g. sell/sold, buy/bought, in which the present has i-mutation but the past does not).

All of these categories of Modern English words exhibiting i-mutation were already present in Old English. I-mutation also appears in some forms of certain nouns of relationship, some comparative adverbs, and many verb forms. Examples: the nominative plural of mann ‘man’ is menn; the nominative plural of lūs ‘louse’ is lȳs; the comparative of eald ‘old’ is ieldra; the comparative of the adverb feor is fier; the third-person singular of the strong verb ċēosan ‘choose’ is ċīest.

2.2.3. Silent e; o for u

When ċ, ġ or sc (pronounced [ʃ]) occurs before a back vowel, it is sometimes followed by an e, which probably should not be pronounced, but merely indicates that the ċ should be pronounced [ʧ], the ġ [j] or [ʤ], and the sc [ʃ]. For example, you will see sēċean ‘seek’ as well as sēċan, ġeþinġea ‘of agreements’ as well as ġeþinġa, and sceolon ‘must’ (plural) as well as sculon.

Notice that sceolon has o in the first syllable while sculon has u. These two spellings do not indicate different pronunciations; rather, the Old English spelling system appears (for unknown reasons) to have prohibited the letter-sequence eu, and scribes sometimes wrote eo instead to avoid it. Other words that are spelled with o but pronounced [u] are ġeō ‘formerly’, ġeong ‘young’, ġeoguð ‘youth’ and Ġeōl ‘Yule’. For these you may also encounter the spellings , iung, iuguð, Ġiūl and Iūl.

2.3. More about c and g

The dots that we print over c and g are not in the manuscripts that preserve the Old English language for us; rather, modern scholars have supplied them. Further, the relationship between Old English pronunciation and Modern English outcome is not always straightforward, as you can see from Modern English seek, which comes from Old English sēċan. So what are the rules for the pronunciation of Old English c and g? We print dots over c and g when they come in these environments:

Otherwise, we generally print plain c and g.

C was pronounced [k] in camb ‘comb’, cǣġ ‘key’, cēne ‘keen, brave’, bacan ‘bake’, bōc ‘book’. It was pronounced [ʧ] in ċeaf ‘chaff’, ċīdan ‘chide’, ċierran (late West Saxon ċyrran) ‘turn’, ‘I’.

G was pronounced [ɡ] in gōd ‘good’, glæd ‘glad’. It was pronounced [ɣ] (the voiced velar spirant) in dagas ‘days’, sorga ‘sorrows’, sīgan ‘descend’. It was pronounced [j] in ġiestrandæġ ‘yesterday’, sleġen ‘slain’, mæġ ‘may’, seġl ‘sail’ (noun), seġlode ‘sailed’. It was pronounced [ʤ] in enġel ‘angel’, senġe ‘I singe’.

As soon as you start to read Old English texts you will notice that these rules apply well enough at the beginnings of syllables, but don’t always seem to work elsewhere. For example, the c in sēċan ‘seek’ has a dot even though it comes before a back vowel, and the c in macian ‘make’ lacks a dot even though it comes before a front vowel. Such anomalies arise from the fact that the changes that produced the sounds spelled ċ and ġ took place long before the time of our written texts, and the sounds that produced those changes often disappeared later as a result of the simplification of unaccented syllables that is characteristic of Old English.[5] This fact is inconvenient for students of Old English, for it means that you cannot be certain how to pronounce some words unless you know their pre-history.

We can tell what these sounds were because they are often preserved unchanged in related languages. For example, in Old Saxon the word that appears in Old English as sēċan is sōkian, and in Gothic it is sokjan—the sound that produced i-mutation and changed [k] to [ʧ] is still present in those languages.

Often it is enough to know about the grammar of a word to decide how to pronounce it. In class 1 weak verbs, the root syllable had formerly been followed by [i], which either disappeared or came to be spelled e, or [j], which usually disappeared; so c and g should generally be dotted at the ends of those syllables. Examples: senġan ‘singe’, senċan ‘cause to sink’, sēċan ‘seek’, īeċan ‘increase’, bīeġan ‘bend’. In class 2 weak verbs, the root syllable had formerly been followed by a back vowel, even though that vowel often disappeared; so c and g at the ends of those root syllables should not be dotted. Examples: macian ‘make’, bōgian ‘dwell’, swīgian ‘fall silent’.

When the vowel of any syllable has undergone i-mutation, that is a sign that [i] or [j] once followed, and so c or g at the end of such a syllable should be dotted. Athematic nouns like man/men, which change their vowels, do so as a result of i-mutation; so the plural of bōc ‘book’ is bēċ, and the plural of burg ‘stronghold’ is byrġ.

2.4. Syllable length

The length of a syllable (sometimes called its weight) is important in both Old English grammar and meter. A long syllable has a long vowel or long diphthong, or it ends with at least one consonant. These one-syllable words are long: ‘sea’, fæt ‘container’, blind ‘blind’, dǣd ‘deed’, hēng ‘hung’. A short syllable must have a short vowel or diphthong and must not end with a consonant. The demonstrative pronoun se is a short syllable.

When a single consonant falls between two syllables, it belongs to the second. Add an ending to fæt ‘container’, for example fæte, and the -t- no longer belongs to the first syllable, but rather to the second: fæ-te, in which the first syllable is now short rather than long. Add an ending to dǣd ‘deed’ (dǣ-de), and the first syllable is still long because it contains a long vowel.

Two short syllables may count as one long one, so a two-syllable word like reċed ‘hall’ behaves like a word with one long syllable. But when a two-syllable word begins with a long syllable—for example, hēafod ‘head’—the second syllable counts as short, even if a consonant ends it. If you ponder this long enough, it may start to make some sense.

2.5. Accentuation

All Old English words are accented on the first syllable, except that words beginning with the prefix ġe- are accented on the second syllable, and verbs beginning with prefixes are accented on the next syllable after the prefix. It may seem odd, but it is a fact that nouns and adjectives with prefixes (except ġe-) are accented on the prefixes. The verb forwéorðan ‘perish’ is accented on the second syllable; a noun derived from it, fórwyrd ‘destruction’, is accented on the prefix.

Words borrowed from Latin are accented on the first syllable, despite Latin rules of accentuation. So paradīsus ‘paradise’ is accented on the first syllable (páradīsus) instead of on the penultimate (paradísus), as in Latin.

2.6. On-line pronunciation practice

You will find pronunciation exercises at faculty.virginia.edu/OldEnglish/. Audio also accompanies the Old English Aerobics texts “The Fall of Adam and Eve” and “The Story of Cædmon”.

2.7.Summary

Table 2.2 presents the Old English pronunciation rules in summary form. Make a copy of it and keep it by your side as you practice reading aloud.

Table 2.2. Old English pronunciation
Spelling Pronunciation
a [ɑ] as in Modern English father
æ [æ] as in Modern English cat
e [e] as in Modern English fate
ea [æɑ] a diphthong, starting with [æ] and ending with [ɑ]
eo [eo] or [eʊ] a diphthong, starting with [e] and ending with [o] or [ʊ]
i [i] as in Modern English feet
ie [ɪ] as in Modern English sit
o [o] as in Modern English boat
u [u] as in Modern English fool
y [y] as in German über or Füße, French tu or dur
 
c [k] as in Modern English cow
ċ [ʧ] as in Modern English chew
cg [ʤ] like the dge in Modern English edge
f [f] as in Modern English fox; between voiced sounds [v]
g [ɡ] as in Modern English good; between voiced sounds [ɣ], a voiced velar spirant
ġ [j] as in Modern English yes; after n [ʤ] as in angel
h within words or finally, [x] or [ç] like German ch
s [s] as in Modern English sin; between voiced sounds [z]
sc [ʃ] usually as in Modern English show; occasionally [sk]
þ/ð [θ] as in Modern English thin; between voiced sounds, [ð] as in then