Historical period

The chronological boundaries of the Middle English period are not easy to define, and scholarly opinions vary. The dates that OED3 has settled on are 1150-1500. (Before 1150 being the Old English period, and after 1500 being the early modern English period.) In terms of ‘external’ history, Middle English is framed at its beginning by the after-effects of the Norman Conquest of 1066, and at its end by the arrival in Britain of printing (in 1476) and by the important social and cultural impacts of the English Reformation (from the 1530s onwards) and of the ideas of the continental Renaissance.

Back to top

The most important linguistic developments

Two very important linguistic developments characterize Middle English:

• in grammar, English came to rely less on inflectional endings and more on word order to convey grammatical information. (If we put this in more technical terms, it became less ‘synthetic’ and more ‘analytic’.)Change was gradual, and has different outcomes in different regional varieties of Middle English, but the ultimate effects were huge: the grammar of English c.1500 was radically different from that of Old English.Grammatical gender was lost early in Middle English. The range of inflections, particularly in the noun, was reduced drastically (partly as a result of reduction of vowels in unstressed final syllables), as was the number of distinct paradigms: in most early Middle English texts most nouns have distinctive forms only for singular vs. plural, genitive, and occasional traces of the old dative in forms with final –eoccurring after a preposition.In some other parts of the system some distinctions were more persistent, but by late Middle English the range of endings and their use among London writers shows relatively few differences from the sixteenth-century language of, for example, Shakespeare: probably the most prominent morphological difference from Shakespeare’s language is that verb plurals and infinitives still generally ended in –en (at least in writing).
• in vocabulary, English became much more heterogeneous, showing many borrowings from French, Latin, and Scandinavian. Large-scale borrowing of new words often had serious consequences for the meanings and the stylistic register of those words which survived from Old English. Eventually, various new stylistic layers emerged in the lexicon, which could be employed for a variety of different purposes.

One other factor marks out the bulk of our Middle English evidence from the bulk of our Old English or early modern English evidence, although it is less directly a matter of change in the language than in how it is represented in writing:

• the surviving Middle English material is dominated by regional variation, and by (sometimes extreme) variation in how the same underlying linguistic units are represented in writing. This is not because people suddenly started using language in different ways in different places in the Middle English period, but because the fairly standardized late Old English literary variety broke down completely, and writing in English became fragmented, localized, and to a large extent improvised.

Back to top

A multilingual context

Medieval Britain had many languages. English continued to be in contact with Celtic languages on many of the internal frontiers within the British Isles. Until the use of Scandinavian languages in mainland Britain died out (the precise date of which is a matter of uncertainty), it continued to be in contact with these also. And, crucially, it was in contact with Latin and with French.

After the Norman Conquest, the ruling elite in England (in church as well as state) were French speakers. Before the Conquest, England had been relatively ‘advanced’ in the extent to which the vernacular language, rather than Latin, was used in writing. After the Conquest, English became pushed out of these functions almost entirely. Latin predominates in most types of writing in the immediately post-Conquest period. When, quite soon afterwards, we find a flowering of vernacular writing in a number of different text types and genres, this is in French, not English. Likewise it was French, not English, that generally vied with Latin in a wide range of technical and official functions until very near the end of the Middle English period. (What to call the French used in Britain in this period is a difficult scholarly question. Traditionally the term ‘Anglo-Norman’ has been used, notably in the title of The Anglo-Norman Dictionary. In fact, the present-day editors of that dictionary note that in many ways ‘Anglo-French’ is a more appropriate term, since it better reflects the wide variety of inputs shown by the French used in medieval Britain. OED3 retains the term ‘Anglo-Norman’ largely to maintain consistency with the title of The Anglo-Norman Dictionary.)

Up until about the middle of the fourteenth century, our surviving written records for Middle English of any variety are patchy, and can be characterized as a number of more or less isolated ‘islands’ of usage, reflecting the English of particular communities or even individuals who felt motivated, for various different reasons, to write something down in English. We have some substantial literary texts, such as the Ormulum or the Ancrene Wisse (both of which we will look at more closely below); in a very few cases, like the Ancrene Wisse and a small group of texts in a very similar language apparently from a very similar milieu, we can identify mini-traditions of English writing; but what we do not have are clear, well-established, persistent traditions of writing in English (whether for literary or non-literary purposes) from which any sort of standard written variety could grow.

From the later fourteenth century our records become more plentiful, especially for London, as the use of English increased in literary contexts and in a variety of different technical and official functions. English began more and more to be the default choice for major (broadly metropolitan) literary writers such as, in the late fourteenth century, Geoffrey Chaucer, John Gower (who still also wrote major poems in French and Latin), and (although his milieu was rather different) William Langland. We also continue to find substantial literary works from parts of the country far removed from London, and reflecting very distinct local varieties of English, such as Sir Gawain and the Green Knight.

In this same period religious writings in English become more and more common; these include the first complete English translation of the Bible, the Wycliffite Bible, which emerged from the circle of followers of the reformer John Wyclif. We also find increasing numbers of scientific and medical texts written in English.

As it came to share and, eventually, take over various functions from Latin and French, English was hugely influenced by these languages, in its stock of word forms, in the meanings these words showed, and in the phrases and structures in which they were used. Thus the vocabulary of such fields as law, government, business, and religion (among many others) became filled with words of Latin or French origin, as people began using English to express technical matters which had previously been the domain of Latin or French.

Back to top

Borrowing from early Scandinavian

The long succession of Viking Age raids, settlements, conquests, and political take-overs that played such a large part in Anglo-Saxon history from the late-eighth century onwards resulted in many speakers of varieties of early Scandinavian being found in Britain. In particular, there were areas of significant Scandinavian settlement in the east and north east of England (chiefly of speakers of East Norse varieties) and in the north west of England (chiefly of speakers of West Norse varieties), as well as in parts of Scotland. We speak of ‘early Scandinavian’ in this context because we are dealing with the antecedent stage of the later Scandinavian languages, Icelandic, Norwegian, Swedish, Danish, etc. (As regards the divisions among the Scandinavian languages, Icelandic and Norwegian are both West Norse languages, while Swedish and Danish are East Norse languages; however, very few of the Scandinavian loanwords in English can be assigned with any confidence to specifically East Norse or West Norse input.)

Gradually, over the course of generations, the use of early Scandinavian died out in England, but not without leaving a significant impact on the vocabulary of English. When most borrowings occurred is a matter of some uncertainty; Old English texts up to about the year 1100 are estimated to contain only about 100 Scandinavian loanwords, many of them in isolated examples. Most of these words come from semantic areas in which there was significant cultural influence from the Scandinavians, such as seafaring, warfare, social ranks, law, or coins and measures. Many, many more Scandinavian borrowings are first recorded in Middle English texts, but it is very possible (and indeed likely) that most of these first entered some varieties of English in the Old English period. One major indicator of this is that very early Middle English texts from areas of high Scandinavian settlement are full of Scandinavian borrowings.

The long homiletic poem entitled the Ormulum is the work of an Augustinian canon called Orm (a name of Scandinavian origin) who probably lived in south Lincolnshire; the dating is controversial, but Orm may have started work on the text as early as the middle of the twelfth century and continued well into old age. It contains well over a hundred words of either certain or likely Scandinavian origin, including some which are of common occurrence in modern English such as to anger, to bait, bloom, boon, booth, bull, to die, to flit, ill, law, low, meek, to raise, root, to scare, skill, skin, to take, though, to thrive, wand, to want, wing, wrong. Perhaps most interestingly of all, it contains some of the earliest evidence for one of the most important Scandinavian borrowings, the pronoun they and the related object form them and possessive their.

The example of they, them, and their is very instructive about the nature and extent of Scandinavian influence on English. It is very rare for pronouns to be borrowed; the fact that these forms were borrowed probably reflects both the very close contact between Scandinavian and English speakers, and the close structural and lexical similarities between the two languages. Because so many words, forms, and constructions were already either identical or very similar, this made it much easier for even grammatical words to be borrowed.

Something else illustrated by they, them, and their is the long process of internal spread, from variety to variety, shown by many words of Scandinavian origin after they entered English. Orm uses they invariably, but them and their vary in his text with the native forms hem and her. In later northern or eastern texts them and their quite quickly become the normal forms, but this takes much longer in other varieties: the most important early Chaucer manuscripts, from London in the late fourteenth or early fifteenth centuries, have typically they for the subject form but still hem and her for the object and possessive forms.

The inherited similarities between English and early Scandinavian also make it extremely difficult to be certain in very many cases whether a word actually shows a Scandinavian borrowing at all, or an Old English word which is simply poorly attested in our surviving sources. The Scandinavian component in the total vocabulary of Middle English perhaps amounts to somewhere in the region of 2 or 3 per cent, but any figures must be treated with a good deal of caution. In spite of the relatively small total, many of the words occur with quite high frequency, especially in texts from more northerly and easterly areas. Some Scandinavian borrowings which were doubtless borrowed in either Old English or Middle English are first attested much later; this is especially the case with words preserved only in regional use.

Back to top

Borrowing from Latin and/or French

The Latin component in the vocabulary of Old English was small, only amounting to a few per cent of the total of surviving Old English words, and many (but by no means all) of these words were doubtless of very rare occurrence, confined to very occasional use by scholars. The securely identified pre-Conquest borrowings from French amount to barely a handful, and even in very late, post-Conquest Old English not many more are recorded.

In Middle English this picture changes radically. If we look at the vocabulary of Middle English as a whole, the evidence of dictionaries suggests that the number of words borrowed from French and/or Latin outstrips the number of words surviving from Old English by quite a margin. However, words surviving from Old English (as well as a few of the Scandinavian borrowings, especially they) continue to top the high frequency lists (as indeed mostly remains the case even in modern-day English).

The formulation ‘French and/or Latin’ is an important one in this period. Often we can tell that a word has come from French rather than Latin very clearly because of differences of word form: for instance, English peace is clearly a borrowing from Anglo-Norman and Old French pais, not from Latin pac-, pāx. Some other pretty clear examples are marble, mercy, prison, palfrey, to pay, poor, and rule. It is often much more difficult to be certain that a Middle English word has come solely from Latin and not partly also from French; this is because, in addition to the words it inherited from Latin (which typically showed centuries of change in word form), French also borrowed extensively from Latin (often re-borrowing words which already existed in a distinct form). Some typical examples are animal, imagination, to inform, patient, perfection, profession, religion, remedy.

Given these factors, any figures for the relative proportions of French and Latin borrowings in the Middle English period have to be hedged about with many provisos. However, the broad picture is clear. In Middle English, borrowing from French is at least as frequent as borrowing from Latin, and probably rather more frequent.

By 1500, over 40 per cent of all of the words that English has borrowed from French had made a first appearance in the language, including a very high proportion of those French words which have come to play a central part in the vocabulary of modern English. By contrast, the greatest peak of borrowing from Latin was still to come, in the early modern period; by 1500, under 20 per cent of the Latin borrowings found in modern English had yet entered the language.

The greatest peak of first examples of French borrowings in English comes in the late fourteenth and early fifteenth centuries. This probably largely corresponds to the realities of linguistic change, since we know that this is the period in which English was taking on many technical functions from Latin and, especially, French, at least so far as written records were concerned. However, this is precisely when our volume of surviving Middle English material also goes up dramatically, and so we cannot always rule out the possibility that words existed in English rather earlier. Certainly, some much earlier texts, such as the thirteenth-century Ancrene Wisse, show considerable borrowing from French at an early date, and we cannot always be certain that an absence of earlier attestations necessarily means that a word did not exist in at least some varieties of English at an earlier date.

Mixed language texts pose many difficult challenges. One quite common pattern is for accounts, records, and other official documents to have Latin as the ‘matrix’ language, but to switch freely to a vernacular language to name particular things or concepts. Whether the vernacular language in question is French or English can be very difficult to tell, or in many cases plain impossible. In fact, many scholars who have spent time working on such documents take the view that the writers themselves probably did not always distinguish very clearly between one clearly defined vocabulary as ‘English’ and another as ‘French’; the considerable overlap, of words belonging to both languages (as a result of earlier borrowing), in a context in which new words were being borrowed all of the time, would indeed have made it almost impossible to make such a clear distinction, especially in many areas of technical vocabulary. For some examples of some of the implications for OED data see the entries for oillet n., pane n.¹, pastern n., pullen n., rack v.², russet n. and adj.

Back to top

Pronunciation

Since our surviving Middle English evidence is so characterized by regional variation, it is very difficult to summarize ‘typical’ Middle English pronunciation, just as it is difficult to summarize ‘typical’ Middle English morphology, or grammar.

As a general rule of thumb, anyone entirely unfamiliar with Middle English who wants to be able to pronounce Middle English word forms is better off trusting the Middle English spelling, rather than making assumptions on the basis of the modern English pronunciation. In particular, vowel letters normally have values much closer to what is typical in modern continental European languages, than to the values that they have in modern English.

• for example, the i in fīn ‘fine’ represents a long monophthong similar to that in modern English meet, while the e in mēten ‘to meet’ represents a sound more similar to that in modern English make (but a monophthong, not a diphthong).

See Edmund Weiner’s piece on early modern English to see how the Great Vowel Shift changed this situation. See also the OED entries for A n., E n., I n., O n., U n. for much more detail on the development of the various sounds represented by these letters.

Back to top

A period characterized by variation

The majority of later Old English texts are written in a fairly uniform type of literary language, based on the West Saxon dialect. The linguistic forms employed show considerable regularity, as do the spellings used to represent them.

The political and cultural upheavals of the Norman Conquest completely changed this situation: people who chose to write in English in the early Middle English period typically had to improvise, in order to find ways of representing a particular local variety of Middle English in writing. To do this they often had to draw upon spelling traditions that were more typically used in writing Latin or French. Variation reigns supreme. Some groups of manuscripts show very similar language represented in very similar orthography, but in the broader picture these appear isolated pockets.

In later Middle English spelling habits typically become rather more stable, and we generally find more consistency in the strategies used for representing particular sounds in writing. However, a considerable degree of spelling variation remains the rule rather than the exception, and it is quite typical to find the same word spelled in slightly different ways within a single page of a single manuscript. If we look at the full repertory of surviving spelling forms, the situation can still seem quite bewildering; for instance, the Linguistic Atlas of Late Mediaeval English records around 500 different spellings for through.

As well as showing variation in how to represent sounds in spelling, our surviving late Middle English writings also continue to reflect a wide variety of different regional varieties of English. Although London and its dialect became of increasing importance in official functions and in literary production, and many of the major late Middle English writers were based in or near the capital, the real dominance of a metropolitan variety over all others in literary use comes only in the early modern period.

London English of the late-fourteenth and fifteenth centuries showed a wide variety of inputs, among which a number of features from the central and east midlands figured strongly. It is in no way an interrupted continuation of the predominantly south-western Old English literary language, and in many key respects it reflects the language of parts of the country for which we have little or no evidence from the Old English period.

There also continued to be a great deal of variation within London English, in written forms as well as spoken. The focused usage of a number of official documents, often referred to as ‘Chancery English’, had a significant input into the practices of early modern English printers, but this is only one aspect of a very complex story, which is still subject to considerable uncertainty and debate.

This complicated picture is complicated still more by the nature of our surviving documents, which is discussed in the following section.

Back to top

Our surviving documents

We have much more surviving Middle English evidence than we have for Old English, but still far less than we have for the developing, London-based standard language of the sixteenth century and later. The information that we do have is patchy and uneven: we have a pretty good record for London and the surrounding area from about the end of the fourteenth century onwards, but for most parts of Britain throughout the period we have only isolated flashes of illumination.

Our surviving evidence for Middle English also poses a number of interesting challenges for historical lexicography. The overwhelming majority of our information comes from hand-written manuscripts. (From the last quarter of the fifteenth century onwards there are also printed books, and of course there is also some written text on coins, paintings, memorials, etc.) Manuscript evidence can present many difficult challenges for dating and interpretation.

Many (but by no means all) collections of functional records, e.g. recording business transactions, are in hands which are either contemporary or very nearly contemporary with the information being recorded. But this is much more rarely the case with literary works (taking this in a broad sense, to include e.g. technical or religious treatises); these are often recorded only in much later manuscripts, and even when the manuscripts are contemporary or nearly contemporary, they may show extensive departures from the language of the author.

In a very few cases, we have manuscripts surviving in the hand of the author, known technically as ‘holograph’ manuscripts. Pretty certain cases include: the Ormulum (see above); from the fourteenth century, the Ayenbite of Inwyt by Dan Michel of the Northgate; and, from the fifteenth century, various works by Thomas Hoccleve and John Capgrave. Most literary works survive in copies by non-authorial hands. These pose various interconnected problems.

Firstly, we need to assign a date to the manuscript in which our evidence occurs. This is often not a simple matter. Some manuscripts are dated on the basis of pieces of internal evidence, such as a dated inscription in one of the scribal hands, or a reference to a particular historical event. Other manuscripts contain no clear indication of date themselves, but are dated on the basis of careful comparison with the hands of other manuscripts which can be dated more confidently on other grounds. In this way, palaeographers have built up a careful picture of the development of the various different scripts that scribes used in medieval Britain. However, very many hedges, provisos, and qualifications are necessary at every stage in this process: even datable manuscripts can often only be dated very approximately, and dating to a particular year can only rarely be relied on as 100 per cent secure; the palaeographical dating that builds on these foundations is dependent on the skill and judgements of palaeographers, who will rarely claim precision for a particular dating, and who will often differ from one another in their judgements. Normally, palaeographical datings are expressed as an approximate date range. In some cases, palaeographers may only feel confident in assigning a manuscript to somewhere within a period of as much as a hundred years (this is quite often the case with fifteenth-century manuscripts).

Once we have a date for our manuscript, we then have the problem of trying to decide whether it is reflecting the contemporary language of the scribe, or the language of the original author, or of an earlier stage in a chain of copying, or whether it shows some sort of mixed language, with features from various different points in the chain.

Modern work on the habits of medieval English scribes suggests that their behaviour can be divided into three types:

• scribes who ‘translate’ consistently into their own dialect
• scribes who copy more-or-less precisely, letter-for-letter, from their exemplar
• scribes who ‘translate’ only partially, replacing some words or forms with those from their own dialect, but leaving others unchanged

Since our surviving manuscripts sometimes stand at the end of a long chain of copying, in which successive scribes may have adopted different approaches, the possible permutations become very complex indeed.

All of this has some important implications for historical lexicographers, including:

• it is only quite rarely, and in very special circumstances, that we can be absolutely certain that the precise reading we find in a manuscript is authorial.
• but equally, we cannot normally assume that the language of a manuscript precisely reflects the contemporary usage of its scribe, especially as regards vocabulary: even a consistent ‘translator’ may have left in some words or forms which he would not have selected in his own day-to-day linguistic usage.
• comparison between different manuscripts of a work often indicates that a particular word is very likely to have been used by the original author, but various scribes have made their own choices about spelling; the different spellings adopted may well correspond to different pronunciations, and leave us in doubt about the authorial form.
• thus, dating of words and forms from the Middle English period is often hedged around with uncertainty – not only do we have only a very partial reflection of actual linguistic use, but we also cannot be certain that we even have a faithful ‘snapshot’ of a particular moment in time.

Back to top