anglo-saxon warriors
anglo-saxon warriors
"The most obvious kind of evidence for looking at the [linguistic] history of Old English is the evidence provided by written texts." (Leith, 1996: 110)


Nobody actually knows what Old English (OE) sounded like. The only evidence we have of the language as it existed then are a number of Old English texts that have survived into the present day. The problem is, as noted by Leith (1996), that no such material was produced prior to the 8th century.

Regarding pronunciation, Leith (1996) claims that it is commonly assumed by many academics that OE spelling was more phonetic than the present-day system, i.e. that the orthography reflected the actual speech sounds of the language more precisely than is the case today. Moreover, the assumption is that there were no silent letters, e.g. the <h> would be pronounced in hring (ring), the <g> in gnornian (mourn) etc. It is further surmised that the letters employed represented the same sound values as those in spoken Latin (see below). Accepting all this to be true than it is possible to make specific deductions concerning the probable pronunciation of OE

Crystal (1995) claims that current views regarding OE speech sounds are based on four principal forms of evidence: comparative reconstruction, sound change, poetic evidence and alphabetical logic. These principals are to be examined first after which examples of OE sounds in terms of both vowels and consonants are given in table form (the information contained therein being drawn from Mitchell and Robinson (1968)). Finally, there is the opportunity to hear the Lord's Prayer spoken in West Saxon dialect.


Investigating OE pronunciation

Comparative reconstruction
This is, in effect, a retrogressive investigation based on the phonological examination of modern English, particularly dialect. The assumption is that these modern sounds are likely to resemble something near those of OE. However, although it might seem improbable that consonants are pronounced much differently, the sound values of vowels are much more problematic. For example, as Crystal (1995) observes, although we currently think in terms of five vowels there are in fact around twenty vowel sounds in modern English. Furthermore, the quality of these sounds can vary greatly according to accent. Accordingly the modern perception of OE vowel sounds could be viewed as an approximation.

Sound change
There is an extensive body of knowledge relating to the sound changes which occur as a language develops and alters. It is thus possible to theorize with regard to the manner in which OE letters were pronounced. By way of example Crystal (1995) cites OE hit ("it"). It might be hypothesized the originally the <h> was pronounced but later dropped as the language evolved. This does seem reasonable given that h-dropping in unstressed syllables is very much a modern phenomenon.

Poetic evidence
Alliteration and rhymes present in OE texts provide clues in relation to how the sound system worked. The poetic metre of a poem is also indicative of the way words were stressed. Knowing the stress patterns in a line of poetry provides clues as to how a vowel in an unstressed syllable might have been realized.

Alphabetical logic
As with sound change, much is known about the pronunciation of the letters of the Latin alphabet. Moreover, the view that their adoption into OE was conducted in a rational and uniform manner seems to be a reasonable one. Thus, the notion that the Latin letter p was employed to signify the identical sound in OE seems to be logical.


However, with regard to textual evidence on which much of the above is based, Smith (1996) does identify certain problems. Among other things, he points out that the amount of surviving material is fairly meagre and unevenly distributed with many areas of the country "unrepresented". Furthermore, OE texts may not always provide reliable clues to pronunciation. As an example, Smith cites the Ruswell Gospels, an 8th / 9th century copy of the Gospels written in Latin. This MS was given a an interlinear gloss during the 10th century by two different scribes identified as Farman and Owun. The latter wrote in Old Northumbrian whilst Farman's language is a mixture of West Saxon and Mercian. This, however, is not an indication of his manner of speech but rather the style of his writing. By way of explanation Smith cites Kuhn (1945: 641-2) who states that while this individual's native speech was Mercian, "[He] was trying to imitate the language of his temporal and ecclesiastical superiors. As a consequence of this imitation he introduced numerous Saxonisms into his glosses".




The convention of distinguishing the long vowels from the short ones by placing a macron over the former is a modern one. It is not found in Old English texts.

Short vowels
Example words Approximate realization
cald (cold)

[a] as in the initial vowel of aha

denu (valley)

[e] as in pet

rinc (man) [i] as in pin
folde (earth) [o] as in pot
sumor (summer) [u] as in wool
synn (sin) [y] as in French tu
wter (water) [a] as in hat


Long vowels
Example words Approximate realization

[a:] as in father


[e:] as in German zehn

(ride) [i:] as in tree
(food) [o:] as in German wo
(house) [u:] as in chew
(hide) [y:] as in German Bhne
(sleep) [a:] as in a lengthened form of hat




On the premise that OE orthography was an accurate representation of the actual speech sounds (see above) it is assumed that, in contrast to modern English, all consonants were pronounced. Letters in common use today were either nonexistent (<j> and <v>) or extremely rare (<q.>, <x> and <z>). Moreover, some characters had more than one sound value whilst others are no longer employed. For example:


Example words Realization
cald (cold) preceding <a>,< o>, <u> or <y> - [ k ] as in kick
(cheese) preceding <e> an <i> - [ t] as in child
gafol (tribute) between vowels or other voiced sounds - [ v ] as in very
folc (people) elsewhere such as the beginning and end of words - [ f ] as in fill
hund (dog) word initially - [ h ] as in hat
niht (night) elsewhere - [x] as in Scottish loch
( rise)

between vowels or other voiced sounds - [ z ] as in zebra

steorra (star)

elsewhere such as the beginning and end of words - [ s ] as in set

(further) between vowels or other voiced sounds - [ th ] as in though or clothe
(thing) elsewhere such as the beginning and end of words - [ th ] as in thin or moth
(worthy) between vowels or other voiced sounds -[ th ] as in though or clothe
(oath) elsewhere such as the end of words - [ th ] as in thin or moth
(OE equivalent to <g>)
(gold) preceding <a>,< o>, <u> or <y> - [ g ] as in modern good
(jewel) preceding <e> and <i> - [ j ] as in yet
(bow) following or between back vowels - [x] as in Scottish loch
(Vikings) [ w ] as in wall


OE consonant groups also appear strange to the modern eye:

Consonant group
Example words Approximate realization

[] as in ship

ecg (edge)

[] as in judge




Spoken Old English

The general consensus is that the sounds of OE no doubt varied from region to region. Pyles and Algeo (1993) even go so far as to suggest that variation may also have existed between social groups. They claim that "a period in which all members of a given linguistic community speak exactly alike, let alone an entire nation, is inconceivable." (Pyles and Algeo, 1993: 102-3).

With regard to regional variation, set out below is the Lord's Prayer written in both West Saxon and Northumbrian dialects. To hear the spoken West Saxon version click here. (It should be noted that the final West Saxon word shown in parentheses is not featured in Crystal's version.)

West Saxon Northumbrian
Fder ure, Fder ure,
u e eart on heofunum, u ar in heofnum.
si in nama gehaldgod. Sie gehalgad noma in.
To becume in rice.

Tocyme ric in.

Gewure in willa on eoran swa swa on heofonum. Sie willo in su is in heofne and in eoro.
Urne gedghwamlican hlaf syle us to dg. Hlaf usenne oferwistlic sel us to dg.
And forgyf us ure gyltas, swa swa we forgyfa urum gyltendum. And forgef us scylda usra, su uoe forgefon scyldgum usum
And ne geld u us on costnunge, And NE inld usih in costunge
ac alys us of yfele. (solice). ah gefrig usich from yfle.
(Crystal 1995: 27)
(Brook 1963: 52)


The recording features the voice of Professor Cathy Ball of the Department of Linguistics, Georgetown University, Washington DC. Anyone interested in learning more about Old English may wish to access Professor Ball's website. Further details regarding the above recording can be obtained from the "paternoster" web page. Additional examples of Old English speech are available on the "audio" web page.



Ball, C. (2000) Old English Pages,

Brook, G. L. (1963) English Dialects, London: Andre Deutsch

Crystal, D. (1995) The Cambridge Encyclopedia of the English Language, Cambridge: Cambridge University Press.

Kuhn, S. (1945) e and in Farman's Mercian Glosses, Publications of the Modern Language Assocition of America, 60, 631 -669.

Leith, D. (1996) The Origins of English. In Graddol, D, Leith, D. and Swann, J. (1996) English history, diversity and change, London: Routledge.

Mitchell, B. and Robinson, F.C. (1968) A Guide to Old English, Second Edition, Oxford: Blackwell.

Pyles, T. and Algeo, J. (1993) The Origins and Development of the English Language, Fourth Edition, London: Harcourt Brace.

Smith, J. (1996) An historical study of English, London: Routledge.