Old English—an overview
Philip Durkin, OED
Old English is the name given to the earliest recorded stage of the English language, up to approximately 1150AD (when the Middle English period is generally taken to have begun). It refers to the language as it was used in the long period of time from the coming of Germanic invaders and settlers to Britain—in the period following the collapse of Roman Britain in the early fifth century—up to the Norman Conquest of 1066, and beyond into the first century of Norman rule in England. It is thus first and foremost the language of the people normally referred to by historians as the Anglo-Saxons.
Before the coming of the Anglo-Saxons, the majority of the population of Britain spoke Celtic languages. In Roman Britain, Latin had been in extensive use as the language of government and the military and probably also in other functions, especially in urban areas and among the upper echelons of society. However, it is uncertain how much Latin remained in use in the post-Roman period.
During the course of the next several hundred years, gradually more and more of the territory in the area, later to be known as England, came under Anglo-Saxon control. (On the history of the name, see England n.)
Precisely what fate befell the majority of the (Romano-)British population in these areas is a matter of much debate. Certainly very few words were borrowed into English from Celtic (it is uncertain whether there may have been more influence in some areas of grammar and pronunciation), and practically all of the Latin borrowings found in Old English could be explained as having been borrowed either on the continent (i.e. beforehand) or during or after the conversion to Christianity (i.e. later).
The conversion of the Anglo-Saxons to Christianity, which began in the late sixth century and was largely complete by the late seventh century, was an event of huge cultural importance. One of its many areas of impact was the introduction of writing extensive texts in the Roman alphabet on parchment (as opposed to inscribing very short inscriptions on wood, bone, or stone in runic characters). Nearly all of our surviving documentary evidence for Old English is mediated through the Church, and the impress of the literary culture of Latin Christianity is deep on nearly everything that survives written in Old English.
Conflict and interaction with raiders and settlers of Scandinavian origin is a central theme in Anglo-Saxon history essentially from the time of the first recorded raids in the late eighth century onwards. However, the linguistic impact of this contact is mainly evident only in the Middle English period. Likewise, the cataclysmic political events of the Norman Conquest took some time to show their full impact on the English language.
Some distinguishing features of Old English
In grammar, Old English is chiefly distinguished from later stages in the history of English by greater use of a larger set of inflections in verbs, nouns, adjectives, and pronouns, and also (connected with this) by a rather less fixed word order; it also preserves grammatical gender in nouns and adjectives.
An example: The following couple of lines from Ælfric's De temporibus anni:
‘Ðunor cymð of hætan & of wætan. Seo lyft tyhð þone wætan to hire neoðan & ða hætan ufan.’
may be translated word-for-word as:
Thunder comes from heat and from moisture. The air draws the moisture to it from below and the heat from above.
To pick out a very few grammatical features:
The nouns hæte, ‘heat’, and wæta, ‘moisture’, both have the inflection -an in the first sentence, because both are in the dative case, governed by the preposition of ‘from’.
In the second sentence they both again have the inflection -an, but this time they are in the accusative case, as the direct objects of tyhð ‘draws’.
The forms of the definite article agree with these nouns, but you will note that they are different in each instance, þone wætan ‘the moisture’ (direct object), but ða hætan ‘the heat’ (also direct object). The difference arises because wæta ‘moisture’ is masculine but hæte ‘heat’ is feminine, and the article (like other adjectives) agrees in gender as well as case.
For another example of gender agreement, look at the pronoun hire (i.e. the antecedent of modern English her) referring to seo lyft (feminine) ‘the air’.
In vocabulary, Old English is much more homogeneous than later stages in the history of English. Some borrowings from Latin date back to before the coming of the Anglo-Saxons to Britain (i.e. they were borrowed on the continent), while many others date from the period of the conversion to Christianity and later. However, words borrowed from Latin or from other languages make up only a tiny percentage of the vocabulary of Old English, and the major influx of words from French and from Latin belongs to the Middle English period and later. (There are also numerous loan translations and semantic loans from Latin in Old English, reflecting the influence of Latin on the language of religion and learning.)
Some borrowing from early Scandinavian is attested in later Old English, but again the major impact of contact with Scandinavian settlers becomes evident only in Middle English.
There is also a great deal of continuity between Old English and later stages in the history of the language. A great deal of the core vocabulary of modern English goes back to Old English, including most of the words most frequently used today.
For a very few examples see I pron. and n.², one adj., n., and pron., and conj.¹, adv., and n., man n.¹ (and int.), woman n.
For further information on which Old English words are included in the OED, and on how Old English material is dated in the dictionary, see Old English in the OED by Anthony Esposito.
Some letters from the Old English alphabet which modern English has lost:
- þ, ð both represent the same sounds as modern th, as e.g. in thin or then;
- æ and a represent distinct sounds in Old English, formed with the tongue respectively at the front and back of the mouth.
The pronunciation of e.g. trap or man in many modern varieties of English comes close to Old English æ, whereas Old English a was more like the sound in modern German Mann ‘man’ or Spanish mano ‘hand’ (like the sound in modern English father, but shorter).
The beginning of Old English …
It is very difficult to say when Old English began, because this pushes us back beyond the date of our earliest records for either Old English or any of its closest relatives (with the exception of very occasional inscriptions and the evidence of words and names occurring in Latin or in other languages). Everyone agrees in calling the language of our earliest extensive sources found in contemporary copies ‘Old English’: these are Latin-English glossaries from around the year 700. (Some other material was certainly composed before 700, but survives only in later copies.) By this time Old English was already very distinct from its Germanic sister languages (see below) as a result of many sound changes (i.e. changes in how certain sounds were pronounced, chiefly when they occurred near to certain other sounds) and other linguistic developments. In fact, most of the most important changes which we can trace through our surviving Old English documents had already happened before this time. Some of them were very probably well in progress or even complete before the time of the settlement in England.
Some Latin-English glosses from one of our earliest sources (the Épinal Glossary):
- anser goos (i.e. ‘goose’)
- lepus, leporis hara (i.e. ‘hare’)
- nimbus storm (i.e. ‘storm’)
- olor suan (i.e. ‘swan’)
Some scholars distinguish the undocumented period before our earliest texts as ‘pre-Old English’, while others are happy just to use the name ‘Old English’ for this period as well as for the documented period. In practice, the dividing line is hazy. Most of our documentary evidence for Old English comes from much later (late ninth century and onwards), and even in the later period there is much that we do not know. In the earlier part of the documented period, the gaps and uncertainties mean that we often know just as little about a certain topic as we do for the preceding undocumented period.
If we trace its history back further, Old English belongs to the West Germanic branch of the Germanic languages, along with Old Frisian, Old Saxon, Old High German, and the various dialects which later gave rise to Old Dutch. The major early representatives of the North Germanic branch are Old Icelandic, Old Norwegian, Old Swedish, and Old Danish (although the earliest extensive remains for all of these are much later than the earliest Old English documents), while the only representative of the East Germanic branch for which extensive remains survive is Gothic. Ultimately, all of these branches diverged from a single hypothetical ancestor, (proto-)Germanic, which itself constitutes a branch of the larger Indo-European language family. Other branches of Indo-European include Celtic, Italic (including Latin and hence the Romance languages), Greek, Indo-Iranian (including Sanskrit and Persian), Baltic, and Slavonic (these last two being regarded by many as a single branch, Balto-Slavonic).
In fact, very many details of the pre-historic relationships between Old English and the other Germanic languages are much debated and very controversial, which greatly complicates any attempt to say when ‘Old English’ began.
The end of Old English
The conventional dividing date of approximately 1150 between Old English and Middle English reflects (very roughly) the period when these changes in grammar and vocabulary begin to become noticeable in most of the surviving texts (which are not very numerous from this transitional period). In what is often called ‘transitional English’ the number of distinct inflections becomes fewer, and word order takes on an increasing functional load. At the same time borrowings from French and (especially in northern and eastern texts) from early Scandinavian become more frequent. All of these processes were extremely gradual, and did not happen at the same rate in all places. Therefore any dividing date is very arbitrary, and can only reflect these developments very approximately.
Old English dialects
The surviving Old English documents are traditionally attributed to four different major dialects: Kentish (in the south-east), West Saxon (in the south-west), Mercian (in the midland territories of Mercia), and Northumbrian (in the north); because of various similarities they show, Mercian and Northumbrian are often grouped together as Anglian. This division is largely based on linguistic differences shown by various of the major early sources, although many of the details are highly controversial, and some scholars are very critical of the traditional association of these linguistic differences (however approximately) with the boundaries of various politically defined areas (which are themselves only poorly understood), and today many of the details of where each variety was centred geographically are subject to debate. For political and cultural reasons, manuscripts written in the West Saxon dialect hugely predominate among our later records (although much of the verse is something of a special case), reflecting the widespread adoption of a form of West Saxon as a written language in the later Old English period.
Old English verbs
Verbs in Old English show an extensive range of inflections, reflecting distinctions of person and number (e.g. first person singular, first person plural, etc.), tense (present or past), and mood (indicative, subjunctive, or imperative); many other distinctions are realized by periphrastic constructions with be v., worth v., will v., or shall v. as auxiliary in combination with non-finite forms of the verb.
With the exception of some (mostly high frequency) irregular or anomalous verbs, Old English verbs belong to one of two main groupings: strong verbs and weak verbs.
The strong verbs realize differences of tense by variation in the stem vowel. They are assigned to seven main classes, according to the vowel variation shown. Thus RIDE v., a Class I strong verb, shows the following vowel gradation in its "principal parts", from which all of its other inflections can be inferred:
- infinitive: rīdan
- past tense singular: rād
- past tense plural: ridon
- past participle: (ge)riden
Similarly, the Class III strong verb BIND v. shows the following principal parts:
- infinitive: bindan
- past tense singular: band (or bond)
- past tense plural: bundon
- past participle: (ge)bunden
The principal parts of the various classes can simply be memorized as fairly arbitrary sets (with various subclasses and exceptions). To understand the causes of this variation we need to go back to a much earlier system of vowel gradation called ablaut, which Germanic inherited from Indo-European, and which Germanic made extensive use of in the strong verb system.
Since ablaut also ultimately explains the relationships between many other Old English words, it can be very useful to have some understanding of how it works, although it is far from simple. See the text box for a very short sketch.
A very short introduction to ablaut The stem vowels ī, ā, i, i shown by rīdan ultimately reflect Indo-European *ei, *oi, *i, *i (giving by regular development Germanic *ī, *ai, *i, *i, giving ultimately Old English ī, ā, i, i). Thus the principal parts in Old English can be explained as reflecting Indo-European *i in combination with either *e (hence *ei), *o (hence *oi), or nothing (hence *i). For these reasons, the infinitive rīdan is said to show the Indo-European e-grade, the past tense singular rād is said to show the Indo-European o-grade, and the past tense plural ridon and past participle (ge)riden are said to show the Indo-European zero-grade, even though, confusingly, the Old English forms themselves do not show e, o, or zero. Similarly bindan ultimately reflects a sequence *en, *on, *n, *n, in which *e, *o, or nothing appear in combination with *n. Similar variation figures largely in a great many etymologies: for some examples see e.g. LOVE n.¹, OWE v., RAW adj. and n.¹, COOL adj., adv., and int., RED adj., n., (and adv.), RIFT n.,
The weak verbs form the past tense and past participle in a quite different way, using a suffix with a vowel followed by -d-, which is the ancestor of the modern inflection in -ed (see ‘-ED’ suffix¹). Thus lufian LOVE v.¹ (a weak Class II verb) shows 1st and 3rd person past singular lufode.
Weak verbs often originated as derivative formations, and often preserve some aspect of this in their meaning, as for example showing causative or inchoative meaning: see below on cēlan ‘to (cause to) cool’ and cōlian ‘to become cool’.
Derivational relationships and sound changes
Many Old English words belong to large groups of words all derived ultimately from the same base, and are related to one another in ways that would have been fairly transparent to speakers of the language. However, in the period of our literary documents the relationships between words were often much less clear than they are likely to have been earlier, because sound changes and other developments had obscured the derivational relationships.
For example, cōl ‘cool’ (see COOL adj., adv., and int.) has a small family of related words in Old English, including cōlnes COOLNESS n., which clearly shows the same base plus ‘-NESS’ suffix. The relationship is similarly clear in the case of the derivative Class II weak verb cōlian ‘to become cool’ (see COOL v.¹).
However, the relationship is less immediately clear in the case of the derivative Class I weak verb cēlan ‘to (cause to) cool’ (see KEEL v.¹). In this case the difference in the stem vowel was caused by an important process called i-mutation which occurred before the date of our earliest records. The earlier form was probably *kōljan. In the process called i-mutation an i or j caused a change in the vowel in the preceding syllable, in this case *ō > *ē. In this word (as in many others) the j was then itself lost, so that by the time of our surviving texts we find cēlan in the same word family as cōl, cōlnes, and cōlian.
The same process explains the variation that we find in the stem vowel in the plural of some words. The word mouse of course shows in modern English the plural form mice; similarly in Old English we find singular mūs but plural mȳs. The earlier forms would have been singular *mūs, plural *mūsi (earlier *mūsiz); i-mutation caused the change *ū > *ȳ in the plural, and then the i was in turn lost, so that in our surviving texts we find singular mūs but plural mȳs.
This and similar processes explain many of the rather complex relationships between related word forms in Old English.
Further reading on Old English
- Richard Hogg, An Introduction to Old English (2002)
- Bruce Mitchell and Fred C. Robinson, A Guide to Old English (7th edn., 2006)
- Roger Lass, Old English: A Historical Linguistic Companion (1994)
- Richard Hogg ed., The Cambridge History of the English Language vol. i: The Beginnings to 1066 (1992)
- Philip Durkin, The Oxford Guide to Etymology (2009)
Where now with the OED Online?
- there's a growing list of commentaries on English in time, charting historical lexicography from Old English to the modern day. As well as this introduction to Old English, you can also read a similar overview of early modern English by Edmund Weiner, deputy editor of the OED, as well as a guide to Old English in OED.
- the OED Online includes more than 7500 entries for which the first evidence of use is dated 1150 or earlier.
How do I search for these? With subscriber access to the OED Online you can search for entries by date, usage, origin, region, and subject using the Advanced Search option. To group entries by time period, use Advanced Search/date of entry or entry range. All results can be displayed as timelines (simply click on the link at the top of the results list), or you can browse the OED via the Timelines option.