Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, LINGUISTICS ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 22 September 2017

Middle English

Summary and Keywords

Middle English is the name given to the English of the period from approximately 1100 to approximately 1450. This period is marked by substantial developments in all areas of English grammar. It is also the period of English when different dialects are the most fully attested in the texts. At the beginning of the Middle English period, the sociolinguistic status of English was low due to the Norman Invasion, and although religious texts of Old English composition continued to be copied and updated, few original compositions are extant. By the end of the period, English had regained its status as the language of government, law, and literature generally.

Although some notable changes to the phonemic inventory of consonants date from the Middle English period, the most dramatic phonological developments of the period involve vowels. The reduction of the vowels of unstressed syllables, one of the changes that marks the beginning of the Middle English period, is a phonological change with substantial morphological effects, as it substantially reduced the number of distinctive inflectional forms. Constituent order replaced case marking as the primary means of signaling grammatical relations. By the end of the Middle English period, subject-verb-object order had become established as the norm.

The lexicon of English was transformed in this period by an enormous influx of French words. The role of derivational morphology declined as its functions were to some extent replaced by the adoption of French words. Most Scandinavian loans in English first appear in the texts of this period. The Scandinavian loans are typically everyday words, while the words adopted from French are more often in areas of government, law, and higher culture, reflecting the nature of the contact between English speakers and the speakers of these languages.

The density of the Scandinavian population in the northern part of England is generally held to be responsible for the earlier appearance of changes in the north than in the south. The replacement of the third person plural personal pronoun hie by the Scandinavian they is an example of a development which is apparent only in the north early in Middle English but became general in English by the end of this period.

An important phonological development of later Middle English is the beginning of the Great Vowel Shift, which affected long vowels and involved successive changes and was implemented differently in different dialects, the north-south divide being the most evident.

Early Middle English is a language that cannot be understood by Modern English readers without special study, while the language of the late Middle English period, especially that coming from the London area, can be understood with the heavy use of explanatory notes.

Keywords: history of English, diachronic linguistics, language change, English dialects, language contact, borrowing

1. Overview

Middle English is the name given to the English of the period from approximately 1100 to 1450 or 1500. This period is marked by substantial developments in all areas of the language. Middle English was the period when English developed from a language with moderate inflection, in which grammatical roles were primarily encoded by case marking, to one in which inflection was greatly reduced.

Materials from early Middle English are the sparsest for any recorded period of English, due to the dominance of French as the language of government, law, and literature, and Latin as the language of scholarship. English writings from this period are mainly copies of Old English materials, although there are valuable exceptions. A Linguistic Atlas of Early Middle English (LAEME, Laing, 2013–) is an invaluable resource containing a large corpus of tagged English texts from 1150 to 1325 as well as up-to-date scholarship on the dating and dialect of these texts, especially valuable to linguists looking at variation because each scribal hand is given a separate entry. LAEME is a work in progress and does not yet cover all texts from the period, but it contains many texts which have not yet been published due to the usual focus on ‘literary’ texts.

English writings become more substantial from the beginning of the 13th century, but surviving texts from the far north are scanty until about 1350; texts from the north midlands are more numerous. By the end of the Middle English period, English had reasserted its position as the official language of parliament and the law courts, as well a language suitable for courtly literature. A valuable resource for information on the writings of this period is A Linguistic Atlas of Late Mediaeval English (LALME; McIntosh, Samuels, Benskin, Laing, & Williamson, 1986), an electronic version of which is available at A Linguistic Atlas of Late Mediaeval English. As well as information about the manuscripts and dialects, LALME contains maps documenting the geographical occurrence of a large number of linguistic features in England in the period 1350 to 1450. The Middle English Grammar Project currently being carried out by researchers at the University of Glasgow, Scotland, and Stavanger University College, Norway, is investigating all aspects of late Middle English grammar in the broadest sense, including lexicon and phonology. The availability of electronic materials has generally made Middle English texts and related materials much more accessible than they were to previous generations of scholars. The Innsbruck Corpus of Middle English Prose makes 129 complete Middle English texts available. Many Middle English texts are available from the University of Oxford Text Archive.

More than any other period of English, Middle English was a time when regional dialects were well represented in writing. To oversimplify the complex network of dialects, we can note that a major division is made between northern and southern dialects, on the one hand, and eastern and western ones, on the other. A survey like this cannot cover more than a few differences in developments in different dialects and must focus primarily on changes affecting the southeastern varieties that form the basis for the modern national standards of England and the former colonies.

Middle EnglishClick to view larger

Figure 1. The introduction to “The Knight’s Tale”, Ellesmere Chaucer folio 10r, ca. 1400–1405. Huntington Library, San Marino, California.

Middle English was substantially affected by contact with other languages, although there are sharply differing views on to what extent contact shaped English and the details of the effects. It has always been recognized that Norse (the cover term adopted here for all varieties of early Scandinavian languages) and French influenced the development of English in important ways, but the possible influence of the Celtic substratum on English has been given serious consideration only fairly recently. While Celtic influence in the spoken language must primarily have been in the Old English period, it has been argued that some features that show up only in Middle English are of Celtic origin, and the reason they did not appear in writing was that the written medium was dominated by an Anglo-Saxon elite. For more discussion of the ‘Celtic Hypothesis,’ see especially Filppula, Klemola, and Paulasto (2008) and Hickey (2012).

2. Lexicon and Semantics

The lexicon is the area in which there is most agreement concerning the effects of contact. Many words of Norse origin which are first found in texts of the early Middle English period came into English as a result of the Scandinavian settlement and Danish rule which predates this period. It is impossible to know whether these words were used in speech for a substantial period before they were used in writing. It is also difficult to put an exact number on Norse loans into English because in some cases the shared Germanic origins of the languages make it unclear whether a word that is first attested in early Middle English is a Norse loan or a word that was used, but not recorded, in Old English. For example, die, not attested until the 12th century, may have been a loan from Old Norse deyja, but may have been the continuation of a verb used in speech in Old English. In other cases, Norse loans have resulted in doublets such as shirt, showing the Old English change [sk] > [ʃ], and skirt, which reveals itself to be a loan from a period after this change was accomplished.

The impact of French loan words on English was enormous; Kastovsky (2006, p. 249) gives a figure of about 10,000 words borrowed from French in the Middle English period, more than 90% of them first attested after 1250. Borrowing from different French dialects at different times resulted in many couplets, e.g., catch, from the Anglo-Norman cachier, versus chase, from chacer, the Central French version of the same word. For a discussion of the different periods of borrowing and how loans into Middle English reflected the nature of language contact in those periods, see Schendl (2012).

The French and Norse loans are strikingly different in their nature, the former typically dealing with higher spheres of life such as government, religion, and law, such as baron, court, crime, religion, and prayer, and the latter typically more everyday words such as leg and sky, although there are exceptions such as poor, an everyday word from French, and the word law itself, a Norse loan. This traditional view, based mainly on literary texts, is likely to need modification in view of the findings concerning the penetration of French and Scandinavian into occupational vocabularies, being investigated currently by A Bilingual Thesaurus of Everyday Life in Medieval England. While the French loans are indicative of a situation in which words from the language of a superordinate people are adopted, the Norse loans are more indicative of intermarriage. The borrowing of function words such as the pronouns they, their, and them, as well as prepositions like til, also points to more intimate contact. Lexical borrowing had effects beyond the lexicon in all linguistic levels.

Some of the loan words, such as baron, refer to new concepts or organization of society. Many of the borrowed words, such as poor, replaced essentially synonymous native words. In some cases, a native word did not disappear but underwent semantic change that differentiated it from the loan. Such is the case with ghost, which has been replaced by the borrowed spirit in its old meaning except in phrases such as give up the ghost and Holy Ghost. Its usual current meaning of ‘spirit of a dead person’ appeared in late Middle English. Spirit is a good illustration of the fact that it is sometimes difficult to determine whether a given word is derived from late Latin or from French. What can be said with confidence is that the influx of French words as well as the Norman domination in the higher echelons of the Church made borrowing from Latin easier in the Middle English period than previously. Essential sources for studying the lexicon and semantics of Middle English include The Middle English Dictionary (Kurath & Kuhn, 1956), an electronic version of which is freely available at The Middle English Dictionary, and the Oxford English Dictionary.

3. Phonology

Too many phonological changes took place over this period of about 400 years to cover more than an outline of a few of the most notable differences from other periods of English. For a more detailed discussion, see especially Lass (1992) and Ritt (2012).

It is important to distinguish sound change from orthographical change. The spelling <hous> for what was spelled <hus> earlier does not represent a new pronunciation, but rather the application of French scribal practice for [u:]. It is also probable that certain sound changes that became apparent only in early Middle English had started in the Old English period but give the appearance of happening later because of the conservative influence of what is often known as the West Saxon Schriftsprache (writing language). The interpretation of the relationship between spellings and sound systems is always difficult. There is therefore uncertainty and scholarly disagreement about the timing and details of most of the changes discussed below.

3.1 Word Stress and Syllable Structure

The area of stress patterns is one where it can uncontroversially be said that borrowing played an important role in Middle English phonology. The influx of French words with mainly penultimate stress complicated the old Germanic rule of stress on the first syllable (with the exception of certain prefixes) and produced a system in which the stress of polysyllabic words is to a large extent unpredictable.

Syllable structure was fairly free in Old English in terms of what combinations of ‘weight’ of weak and light feet could occur in stressed syllables. A number of the early changes to segmental phonology in Middle English, some of which are discussed briefly in the next section, seem to work toward the same end of producing a ‘referred’ syllable structure of simple heavy rhymes in the strong position of a metrical foot (Lass, 1992, p. 75).

3.2 Segmental Phonology

3.2.1 Phonemic Inventory

The phonemic inventory underwent some notable changes in Middle English. A new voicing opposition arose in the fricatives, with [v], [ð], and [z] now appearing word initially and finally for the first time and also contrasting with voiceless correspondents medially. This was a complicated matter which proceeded differently in the different positions. It is probable that French loan words played some role in this development, with loans like veal introducing a voicing contrast in word-initial fricatives not found in Old English. However, the picture is complicated by the fact that voicing in initial position was a feature of southern dialects generally; compare fox with its feminine form vixen, where the pronunciation is taken from what is now a nonstandard dialect.

While the new phonemes /f/, /ð/, and /z/ are an instance of phonemic split, /ʃ/ had never had a voiced allophone, being the result of the Old English change of the voiceless sequence /sk/ > /ʃ/. The phoneme [ʒ] was an entirely new sound brought in by borrowing from French and is still limited in Modern English to words borrowed from that language, e.g., rouge, starting in the Middle English period.

As well as the phonemic split of consonants, we find mergers. Most notably, the shortening of geminate consonants removed the contrast between /p:/ and /p/, for example. There is disagreement on exactly how this change proceeded, due to difficulties in interpreting the orthography. Lass (1992, p. 59) states that degemination began in the north ca. 1200 and spread south.

3.2.2 Vowels: Quality and Quantity

Changes to the vowel system were more sweeping than changes to the consonant system in this period. There were shifts in both quantity and quality, not to mention the repeated creation of front rounded vowels and their subsequent unrounding. One change traditionally referred to as Open Syllable Lengthening has been the subject of an especially large amount of scholarly debate. This name is based on the traditional description of the prototypical change as a lengthening of short vowels in certain ‘open’ syllables, so that [nama] > [na:mə] (‘name’). Things are much more complicated than this, as Minkova (1982) demonstrated, with some vowels failing to lengthen where expected and others unexpectedly lengthening in monosyllabic words. The adjustments in quantity of vowels is currently still a matter of active research, but as noted by both Lass (1992) and Ritt (2012), the details of the changes make most sense when seen in the context of their effects on weight of the syllable.

One notable feature of early Middle English phonology involving word and syllable structure is that unstressed syllables were prone to weakening. This is a continuation of a trend already found in Old English, where fewer vowel distinctions were found in unstressed syllables than in stressed syllables. The ‘confusion’ of unstressed vowels continued in early Middle English, resulting in a near total collapse of distinctions in final unstressed syllables into [ə], usually spelled <e>. The merger of /m/ and /n/ into /n/ in these syllables represents another weakening. By the end of the Middle English period, both this word-final nasal and the vowel were subject to deletion, developments which, as discussed below, have traditionally been seen as deeply affecting the morpho-syntactic system.

One change in vowel quality that deserves special attention here concerns the fate of Old English [ɑ:], both because it represents a major north-south isogloss and because it represents the beginning of the Great Vowel Shift, a change which belongs mainly to the Early Modern English period. In a nutshell, this vowel moved up in the southern dialects, beginning as early as the 12th century, and eventually rounded, so that Old English /hɑ:m/ > /hɔ:m/. In the north, this vowel moved forward before joining the upward movement of the Great Vowel Shift, resulting in forms like /hɩəm/ for ‘home.’

4. Morphology

The Middle English period is marked by substantial changes to morphology, both derivational and inflectional. Massive borrowing from French is responsible for many (but not all) important changes to derivational morphology. Contact with Norse is widely held to be responsible for the drastic changes in inflectional morphology of this period, but the phonological changes that had already started reducing the distinctiveness of inflections before the Scandinavian invasions are also usually seen as playing an important role.

Linguists studying Middle English morphology usually depend on edited texts. Much can be learned in this way, but researchers should be aware that common editorial practices such as the silent expansion of abbreviations may lead to unreliable conclusions. For early Middle English, the LAEME corpus should be used where possible, as it transcribes manuscripts without editing.

4.1 Word Formation

Old English made extensive use of compounding and derivational morphology to create new words. The huge number of words borrowed from French into Middle English had a significant effect on this system. Sylvester (2012, p. 462) states that of the approximately 30 derivational suffixes of Old English, about three-quarters persisted into Middle English. In addition to these old suffixes, new derivational morphemes entered the language from French. Most of these are found only attached to words of Romance origin, but a few, such as -able, became nativized within the Middle English period, such as unspeakable, first attested around 1400. Dalton-Puffer (1996) remains an important in-depth study of the effects of French on Middle English morphology.

Compounding remained a productive resource of the language, but was less frequently employed than in Old English. This can largely be attributed to the enthusiasm of English speakers in this period for borrowing a word for a new concept, rather than making one up from native resources. However, Sylvester (2012) mentions new types of compounds that first appear in Middle English. Contact with Norse had much less effect on Middle English derivational morphology, but Dance (2012) mentions the increased productivity some suffixes common to Norse and Old English, e.g., the -n of harden, as a likely effect of this contact.

An area that spans syntax and derivational morphology is the loss of the verbal prefixes found in Old English, such as the a- of arisan, ‘arise.’ These pretty much disappeared quickly in the early Middle English period, with a concomitant increase in phrasal verbs such as get up. The classic study is Hiltunen (1983). The similarity between English and the modern Scandinavian languages has led to the assumption that this development was caused by contact with Norse, e.g., in McWhorter (2002). However, Hiltunen’s conclusion was that the influence of Norse was at most to speed up a development that was already taking place in Old English, where phrasal verbs such as teon up ‘give up,’ and faran ut, ‘go out,’ were already found. The old verbal prefixes were originally transparent in their semantic (originally locative) effects, but had become opaque and sometimes quite idiosyncratic, in some instances before the first Old English texts, and so were ripe for replacement by phrasal verbs.

4.2 Inflectional Morphology

Early Middle English is a period of striking dialect differences in inflectional morphology. Broadly speaking, deflexion (the loss of inflection) was greatly advanced in northern areas early in the period, while in the southeast, we still find nominal case marking, albeit of a reduced sort, in the Aʒenbite of Inwyt, written (probably by an elderly man) in 1340. This dialect situation suggests that contact with Norse speakers in the north is responsible for the greater degree of deflexion found in that area. However, it is overly simplistic to speak to ‘more conservative’ dialects. In her treatment of the grammar of the early 13th century southwest midlands dialect known as dialect AB, d’Ardenne (1961, §58) comments, ‘The accidence of AB is marked by conservatism in the verbal system contrasted with advanced simplification of the declension of nouns, and still more of adjectives.’ If we assume that the simplification of nominal inflection is due primarily to language contact, it is difficult to explain why the same contact did not result in the more advanced simplification of the verbal system found in other dialects that had similar simplification to AB in the nominal system.

4.2.1 Morphology Within the Noun Phrase

Old English had a system of four central cases, two numbers, and three genders. Determiners and adjectives agreed with the nouns they modified in these features. The form of the adjective also varied, the ‘weak’ form being used essentially after a definite determiner or possessive/genitive modifier, and the ‘strong’ form being used otherwise. Syncretism in the inflections of nouns was already well advanced in Old English, and it was often only the determiner or adjective that signaled the grammatical features of a noun. By the end of the Middle English period, the system of grammatical gender had been replaced by natural gender, and the case marking system was as severely reduced as it is in Modern English. In adjectives, both the singular/plural and the weak/strong distinction were maintained with some systematicity even in the works of Chaucer toward the end of the 14th century, but the marking, -e for both weak and plural adjectives, had become highly ambiguous.

The transformation of the old system of grammatical features took place at different times and in different ways in the various dialects, but some generalizations can be made. The dative/accusative distinction disappeared early, with the distinction preserved longest in the masculine singular accusative pronoun hine versus dative him. A number of texts use him in old accusative contexts (e.g., as the object of a verb) but also use hine in the same contexts without extending it to traditionally dative contexts. This situation indicates that the writers commanded the dative/accusative category distinction but regarded him as a form that could be used for either case. Old forms deriving historically from the dative are best regarded as markers of a prepositional case in some texts.

Texts composed in the first half of the 12th century in a southeastern dialect show essentially the Old English system of grammatical features and agreement within the noun phrase, although with considerable syncretism of forms. By the end of the 12th century, the evidence suggests that even in the most ‘case rich’ dialects, the inflection of elements within the noun phrase for the old grammatical features had essentially become optional.

Features continued to be most clearly marked in the determiner system, and with the replacement of the inflected forms of the determiner came the loss of grammatical gender. The genitive was the most robust of the cases; in the texts in which indeclinable þe vied with inflected forms of the definite determiner, an inflected form of the determiner is more often used in the genitive than with a subject or object. Most nouns inflected with -(e)s for the genitive case in the singular, a suffix which derived from the inflection for the major masculine and neuter paradigms.

The interpretation of the variation found between inflected and uninflected forms in some texts is far from straightforward. A given text might be written by more than one scribe, and most texts are copies of earlier ones. However, close inspection of the most reliable texts shows that the traditional description of the inflection as ‘confused’ must be rejected. The use of the forms is too systematic to be likely to have been consciously learned and it seems most likely that the writers of the texts had internalized a productive system of features even though the marking of those features was not always obligatory.

The loss of inflection is intimately tied up with the reduction of unstressed syllables, and this language-internal explanation is attractive for the loss of inflection in dialects in which contact with Norse was not intense. However, the reduction of phonological distinctions does not account completely for the collapse of case marking. Notably, it does not explain the complete loss the inflection on the determiner the. While phonological reduction caused the syncretism of some forms of this determiner, it does not explain the loss of the distinctive genitive form, which in the masculine and neuter singular ended in an -s. We can compare the situation here with modern German, in which nouns have very little morphology signaling grammatical features, but these are still signaled on determiners and adjectives. It is likely that the low sociolinguistic status of English contributed to an acceleration of changes already in evidence in Old English and combined with contact and phonological reduction of inflections to do away with most of the old system.

Number, still a feature of Modern English, was the hardiest of the grammatical features. The -as nominative and accusative plural inflection of the largest class of nouns in Old English, written <es> in Middle English, quickly spread to other noun classes as the general plural. However, -en, the reflex of the -an that marked most combinations of case, gender, and number in the ‘weak’ nouns of Old English, was for a while a contender to become the regular plural marker, as innovative forms like honden, ‘hands,’ illustrate.

A final morphological change that deserves mention is the extension of the originally plural personal pronoun ye to singular addressees, depending on the social relationships involved.

4.2.2 Verbal Morphology

As with nominal morphology, Middle English was a period of drastic reduction as well as diatopic difference in verbal morphology. One major dialect distinction concerns plural and third person singular suffixes in the present tense. In Old English, the plural suffix (which was the same for all persons) was always distinct from the third person singular suffix. For example, weak verbs of Class I had -eþ in the third person singular and ‑aþ in the plural. With the reduction of vowels in unstressed syllables, these inflections became ambiguous. Speakers in the most southern dialects simply lived with this ambiguity, but in the midlands, speakers extended the past tense plural suffix ‑en to the plural of the present tense, e.g., comen, ‘come (plural).’ By the end of the Middle English period, this suffix was on its way to becoming highly syncretic, as the loss of the final nasal and then the vowel made the first person singular form identical with the plural form, e.g., come. In the North Midlands, the third person singular and general plural forms were kept distinct in another way, as -es replaced ‑ as the third person singular suffix. However, in this region the second person marker was reduced from -est to -es, making the second and third person singular forms identical. In the far north, -es was generalized to all persons and numbers and became a marker of present tense only.

Other notable simplifications include the reduction of the weak verbs to one class and the leveling of allomorphy in strong verbs. In Old English, the plural of Class II weak verbs in the present indicative ended in -iaþ, as opposed to - of Class I verbs. In most dialects this and other differences between the two classes of verbs disappeared early. However, in the 13th-century West Midlands dialect AB, we have retention of -ieþ vs. -. This dialect, advanced in its reduction of nominal morphology, was similarly conservative in retaining some distinct allomorphs of strong verb stems. For example, the paradigm for the IV strong verb beoren ‘bear’ in this dialect has an alternation between eo and e (see Table 1):

Table 1: Paradigm of beoren, ‘to bear,’ in dialect AB

Middle English

The variation in the vowels is the reflex of sound changes that took place in Old English, the details of which are not important here. The important fact is that the retention of two present-tense stems is a conservative feature in a dialect that shows advanced simplification in other areas of morphology. For a thorough discussion of the morphology of dialect AB, see d’Ardenne (1961).

Another simplification in the strong verbs was the leveling of past stem allomorphs. In Old English, most classes of strong verbs had one vowel in the past tense stem in the first and third person singular opposed to another stem vowel shared by the second person singular and the plural. Thus in dialect AB we have ich draf, þu driue, we driuen ‘I/you sg./we drove.’ In the course of Middle English, the stem vowel of the past tense became invariable.

As with nominal inflection, reduction in verbal inflection is usually evident earliest in the northern dialects. The replacement of -eþ by -es in the northern dialects is often attributed to Norse influence. Dance (2012, p. 1735) notes that it is difficult to derive this change directly from Norse morphology, and it is more likely to be an analogical extension of the -s found in the second person singular already in Old English. Contact may have played a role in this analogical extension, because this sort of simplification happens in contact situations even when it does not involve direct borrowing. A more convincing case of direct borrowing of morphology can be made for the -ande form of the present participle found in northern and eastern dialects in contrast with the -ende form inherited from Old English.

5. Syntax

Middle English was a period of extensive syntactic change, and only a selection of the changes that have attracted a good deal of scholarly attention can be mentioned here. The role of language contact in shaping Middle English syntax is currently an issue of considerable debate. An influential paper here is McWhorter (2002), who argues for a large number of syntactic changes that he attributes to contact with Norse. Each of the changes needs detailed scrutiny before it can be accepted as contact-induced. A difficulty here is the fact that contact effects may be direct, involving the importation of a construction directly, or indirect, as in causing simplification in the grammar that is hard to distinguish from internal, structurally motivated changes. Furthermore, grammatical simplification is hard to define and depends on theoretical assumptions. For a discussion of these matters and an overview of the role of contact in syntactic change in Old and Middle English, see Fischer (2013), who concludes that there was substantial, but indirect, Norse influence on English syntax. In contrast to its impact in other areas of the grammar, French is generally assumed to have had only fairly superficial effects in syntax.

5.1 Deflexion and Syntactic Change

A good deal of effort has been put into investigating the decline of inflection as a ‘trigger’ for many of these changes. A useful reference looking at the relationship between morphological and syntactic change in Middle English is Haeberli (2004).

One problem facing attempts to link morphological and syntactic changes is the fact that not many syntactic differences can be found in the different dialects. This is not what we would expect if the loss of inflections caused the syntactic developments for which it has often been held responsible. So, for example, if loss of inflection triggered the fixing of word order, we would expect to find more fixed word order earlier in the north, with its ‘case impoverished’ morphology, than in the south, with its generally more ‘case rich’ dialects. However, as far as we can tell from our texts, there was a move toward more fixed word order in both types of dialect around the same time. Although the order of major constituents was freer in Old English than in Middle English, there were already unmarked orders in Old English, which became grammaticalized in Middle English in a way that does not correlate with the amount of inflectional morphology in the different dialects. However, it can be said that at the time when word order was becoming more fixed, nominal inflection had become at least optional even in the southern dialects. There was thus a general shift toward depending on constituent order to mark grammatical relations even in areas where inflection could in principle still do this job.

One change that has traditionally been attributed to the loss of case-marking distinctions is the loss of ‘impersonal’ verbs such as methinks. The general idea is that the loss of clear dative marking on the fronted noun phrases of the impersonal constructions resulted in their reanalysis as subjects instead of objects. There are numerous problems with this idea, including the fact that nouns were not usually fronted with these verbs; only pronouns were normally put in the preverbal position with many of these verbs, and with pronouns, the non-nominative case marking was clear. Also problematic is the fact that in early Middle English, the very period when the case marking system was collapsing, new impersonal verbs actually entered the language and some native verbs began to be used impersonally for the first time (Miura, 2014). There is also no need to appeal to the loss of case marking to explain changes in the assignment of arguments to grammatical relations with individual verbs such as like, as other explanations in terms of pragmatics and semantics can be put forward. With like, the shift of the grammatical relation of the experiencer from (dative) object to subject can be accounted for by the fact that the experiencer became increasingly more similar in its semantics to a typical subject; it had never been a very typical object (Allen, 1995). Other verbs such as behoove went in the ‘wrong direction’ for an explanation in terms of case marking confusion. That is, the ‘case confusion’ explanation could work only for verbs whose experiencer was in the dative case in Old English but sometimes occurred in preverbal position, making the experiencer liable to reanalysis as the subject. With behoove, however, the shift was not from object to subject but from subject to object. In Old English, the experience of behofian was overwhelmingly in the nominative case, and clearly the subject. In early Middle English, however, we start to find examples in which the experiencer is in the dative case, and in the postverbal position typical of an object, resulting eventually in sentences like (the now rather archaic) ‘it behooves us to proceed cautiously.’ The loss of case distinctions could not have caused speakers to reanalyze an unambiguous subject as an object, but the change can be linked to a shifts in the meaning of behoove. All this is not to deny that the decreased reliance on case marking played no role in the loss of ‘impersonal’ verbs. While deflexion cannot be seen as making the old analysis of these verbs impossible, it can be seen as favoring a system in which such verbs had no place because speakers now expected to be able to read grammatical relations off directly from constituent structure. See Allen (1995) for further discussion.

5.2 Constituent Order

The fixing of constituent order in Middle English to essentially its modern state has received a great deal of attention. Both the position of the verb and the position of noun phrases were affected.

5.2.1 Decline of Verb-Second

Although the positioning of the verb of a main clause was never as fixed in English as it is in the other Germanic languages today, it is generally accepted that Old English had some sort of verb-second rule which was lost in Middle English. This loss is one syntactic change which has been claimed to have proceeded differently in the dialects more and less influenced by Norse contact. Kroch, Taylor, and Ringe (2000) argue that northern dialects had a different type of verb-second phenomenon from southern dialects in Middle English, caused by contact with Norse, although probably not simply a shift to the Norse type, and that confusion caused by contact between the southern and northern dialects probably caused verb-seconding to disappear entirely. By the end of the Middle English period, the verb’s normal position was after the subject, rather than after the initial constituent. Fischer (2013) notes that while this explanation depends on theory-internal assumptions, the evidence of a dialect difference in surface patterns is solid. Other useful references including data and discussions of possible explanations include Fischer, Kemenade, Koopman, and Wurff (2000) and Westergaard (2009).

5.2.2 The Position of Verbs and Objects

The Middle English period saw the near total loss of verb-final order, an order mostly, but not completely, limited to subordinate clauses in Old English. Object-verb order did not disappear suddenly in Middle English but became progressively restricted to certain stylistic and syntactic contexts; for an overview of the Middle English situation and references, see Moerenhout and van der Wurff (2005). Object-verb order was more common in poetry than in prose in Middle English, and even in poetry the frequency of this order had decreased by about 50% in the period 1340–1470. The order did not completely disappear until into the Early Modern English period. Trips (2002) provides useful statistics, although her argument for the role of Scandinavian contact in the loss of object-verb order is controversial.

5.2.3 Fixing of the Position of Objects

Grammatical relations are generally predicable from the position of a nominal phrase within the sentence by the end of the period. The positioning of the subject before the object, already the more frequent order for pragmatic reasons in Old English, became a fixed rule, although as late as the mid-15th century we find Capgrave writing Heretikes, hated þis man with an holy anger (Munro, 1910, p. 52, line 9). The rule at this time seems to have been that a noun phrase (NP) preceding the verb was the subject, unless context made it clear that it was something else, as is the case in this sentence, since heretics could not be the subject, but only the object, of a holy anger.

The order of two objects also became fixed in this period to the current indirect-direct object order (IO-DO) for two nonpronominal objects. As discussed in Allen (1995), attempts to link this development directly to the loss of morphology distinguishing the two types of objects must make appeals to highly abstract morphological categories, since the order did not become fixed until long after the overt morphology had disappeared. An indirect link can be made; although neither morphology nor a fixed order was really necessary to disambiguate the grammatical relations in most cases because the nominal phrases would differ in animacy, the gradual favoring and then fixing of one order was in line with the general trend toward signaling grammatical relations configurationally.

The fixing of the order of two objects is linked with another change, the introduction of the ‘indirect’ or ‘recipient’ passive. The first convincing example is Item as for the Parke she is a lowyd Every yere a dere ‘Item: as for the park, she is allowed a deer every year,’ from Award Blount of 1375 (edited by Cooke, 1925). Since this is around the time when the order of two objects had become fixed, it is reasonable to assume that speakers reanalyzed the active-passive relationship as one creating a derived subject from the first object of a verb, rather than from the direct object.

5.3 The Noun Phrase

The determiner system took its present shape in Middle English. Despite the widely held view that Old English did not have a definite determiner, the transition of what was a demonstrative to the definite determiner arguably took place in the Old English period (see Crisma, 2011, on Old English and Allen, 2016, for the development from Old to early Middle English). There is universal agreement that the indefinite article only developed out of a reduced form of one in Middle English.

Not every change in the surface patterns within the nominal phrase can be mentioned here, but a change that spans morphology and syntax and has attracted a good deal of discussion is the development of the ‘group genitive,’ the name given by Otto Jespersen (1894) to possessives in which the marking is at the end of a syntactic group, e.g., God of Loves servantz, ‘god of love’s servants,’ found at I.15 in Chaucer’s Troilus and Criseyde.1 Here an inflection has become what is variably described as a clitic or a phrasal affix, attaching to the end of the nominal phrase. Chaucer used the older ‘split genitive’ construction, with attachment of this marker to the possessor, much more frequently, e.g., The Prologe of the Wyves Tale of Bathe ‘the prologue of the wife of Bath’s tale.’ The earliest examples of the new group genitive appear in the late 14th century, in Chaucer’s works and those of some other authors. Again, loss of inflection played a role—when agreement was lost within the nominal phrase and the possessive marker became generalized to -(e)s, the marking was ambiguously on the head or the right edge of the phrase and available for reanalysis (Allen, 2008).

5.4 Other Changes

Numerous other syntactic changes took place in Middle English which can only briefly be mentioned here. Changes to one system were frequently interconnected with changes to other systems; for example, changes to negation in Middle English interacted with most of the changes discussed below. Useful references include van Kemenade (2000) and Iyeiri (2001). The study of the interaction between syntax, discourse, and information structure has recently become an area attracting both empirical and theoretical research by linguists; see the papers in Meurman-Solin, López-Couso, and Los (2012) as well as Los and van Kemenade (2012).

5.4.1 Auxiliary Verbs

One change that has been much discussed in the literature is the rise of do as a ‘dummy auxiliary’; Kroch (1989) gives a summary of earlier scholarship on this question. This change deserves particular notice, not only because it has been the subject of much debate concerning how the various stages of the language should be analyzed and of theoretical discussions about the mechanisms of syntactic change but also because there is a long history of attributing this change to the effects of Celtic speakers learning English. The reason why this idea is attractive is because this typologically rather unusual use of an auxiliary is also found in relevant Celtic languages. For discussions of this and other proposed effects of contact with Celtic languages, see the papers of the special issue of English Language and Linguistics edited by Filppula and Klemola (2009). A less dramatic development in the syntax of auxiliaries is the expansion of combinations of auxiliaries, as discussed in Denison (2000). Denison (1993) gives an excellent overview of developments in auxiliaries and verbal syntax generally.

5.4.2 Subordinate Clauses

Notable changes took place in subordinate clauses in addition to the loss of verb-final order. One such development was the introduction of wh- words as relative pronouns.2 In the early Middle English period, demonstrative pronouns ceased to be used as relative pronouns and an uninflected relative marker þe or þat was the norm; þe was the Old English form, at first in variation with and then replaced by the innovative þat ‘that’ (originally a neuter demonstrative pronoun) in early Middle English. However, when the relativized phrase was oblique, the indeclinable relative marker could not be used, and in such situations, the interrogative pronouns served as a replacement for the earlier demonstrative pronouns. The first examples of wh- words used as relative pronouns date to the early 12th century, when we find occasional examples of whose as well as whom in the role of an object of a preposition, e.g., for hwan manega mynstras & turas & huses gefeollon ‘for which [i.e., on account of which] many minters and towers and houses fell’ (Peterborough Chronicle annal for 1117, Irvine, 2004 edition). While whom is not found in the direct object role in the very earliest examples, we do find examples already by the end of the 12th century. However, who is not found in subject function until late Middle English; Rydén (1983) puts the earliest attested example at 1426, in contrast to which, a relative pronoun that was available at an earlier date and was used with animate as well as inanimate antecedents.

Another change was the extension of preposition stranding, already frequent in Old English in limited situations, to new syntactic contexts. This change affected both subordinate and main clauses in Middle English, but only the effects in subordinate clauses will be discussed here. In Old English, preposition stranding was the only option when a relative clause was introduced by the indeclinable relative particle. Early Middle English examples like ʒef þu wult leauen þe lahen þet tu liuest in ‘if you will leave the laws [religion] that you live in’3 therefore continue the old pattern, except that in Old English the stranded preposition would usually have been before the verb of the subordinate clause, rather than after it. Such examples are naturally very common in early Middle English, since the use of relative pronouns was unusual. In Old English, when a relative pronoun was used, preposition stranding was never found. The same was true in interrogatives, both direct and indirect. However, fairly early in the Middle English period, we find a few examples of stranding in both indirect questions and relative clauses. We find both types in Genesis and Exodus, e.g., Wiste noman of werlðe ðo Quat kinde he was kumen fro ‘No one knew what kind [descent] he had come from,’ and And getenisse men ben in ebon Quilc men mai get wundren on ‘And there are gigantic men in Hebron, which men may still wonder at.’4 Preposition stranding remained unusual in the new contexts in early Middle English but became more common in later Middle English. For a comprehensive refutation of a recent proposal that the increase in preposition stranding represented the importation of Norse syntax, see Bech and Walkden (2016). Given that preposition stranding was established in commonly occurring contexts in Old English, its spread into new territory as a grammatical option needs no explanation in terms of language contact, although the possibility that contact affected the frequency in different dialects deserves serious scrutiny.

The first few examples of ‘split’ infinitives come from the 13th century. An early example and he cleopede him to; alle his wise cnihtes. for to him reade ‘and he called all his knights to him to advise him’ is found in Laʒamon’s Brut.5 Examples remain unusual all through Middle English; for details see Calle-Martín (2015). Similarities between modern Scandinavian languages and Middle English make it tempting to see the influence of Old Norse here, but the fact that this and other innovations in subordinate clauses do not appear first in the most Scandinavianized dialects is problematic for such an explanation, as is the lack of evidence that this syntax was present earlier in Norse than in English. Parallel development is more likely, although contact with Norse might have reinforced tendencies toward these developments in English. For a detailed discussion of the problems facing a Norse origin of split infinitives in English, see Bech and Walkden (2016).

Infinitival clauses gained ground over tensed subordinate clauses and for became a complementizer used to introduce infinitival clauses, while the to-infinitive ousted the bare infinitive except as the complement to a limited number of verbs. Developments in infinitival clauses are extensively discussed in Los (2005).

For more extensive discussion of Middle English syntax, see Fischer (1992) and the shorter but updated chapter by Fischer and Wurff (2006). The study of early English prose syntax has been revolutionized by the advent of syntactically parsed corpora. Of particular relevance to Middle English is the Penn-Helsinki Parsed Corpus of Middle English (PPCME2) (Kroch & Taylor, 2000). A description and information about purchasing this corpus are available at Penn-Helsinki Parsed Corpus of Middle English (PPCME2).

Further Reading

Bergs, A., & Brinton, L. J.. (2012). Middle English. In A. Bergs & L. J. Brinton (Eds.), English historical linguistics: An international handbook: Vol. 1, Part IV. Middle English (pp. 399–588). Berlin: Mouton de Gruyter.Find this resource:

Blake, N. (Ed.). (1992). The Cambridge history of the English language: Vol. 2. 1066–1476. Cambridge: Cambridge University Press.Find this resource:

Horobin, S. (2013). Chaucer’s language (2d ed.). Houndsmills, Hampshire, and New York: Palgrave Macmillan.Find this resource:

Horobin, S., & Smith, J. J. (2002). An introduction to Middle English. New York: Oxford University Press.Find this resource:

Kastovsky, D. (2006). Typological changes in derivational morphology. In A. van Kemenade & B. Los (Eds.), The handbook of the history of English (pp. 151–176). Oxford: Blackwell.Find this resource:

Los, B. (2015). A historical syntax of English. Edinburgh: Edinburgh University Press.Find this resource:

Mustanoja, T. F. (1960). A Middle English syntax. Helsinki: Société Néophilologique.Find this resource:

Nevalainen, T., & Traugott, E. C. (Eds.). (2012). The Oxford handbook of the history of English. Oxford: Oxford University Press.Find this resource:

Thomason, S., & Kaufman, T. (1988). Language contact, creolization, and genetic linguistics. University of California Press.Find this resource:

Turville-Petre, T., & Burrow, J. A. (1996). A book of Middle English. Oxford: Blackwell.Find this resource:

Warner, A. (1993). English auxiliaries. Cambridge: Cambridge University Press.Find this resource:

Wright, J., & Wright, E. M. (1934). An elementary Middle English grammar. London: Oxford University Press.Find this resource:


Allen, C. L. (1995). Case marking and reanalysis: Grammatical relations from Old to Early Modern English. Oxford and New York: Clarendon Press.Find this resource:

Allen, C. L. (2008). Genitives in early English: Typology and evidence. Oxford: Oxford University Press.Find this resource:

Allen, C. L. (2016). The definite article: What happened with þe? In S. Vikner, H. Jorgensen, & E. v. Gelderen (Eds.), Let us have articles betwixt us: Papers in historical and comparative linguistics in honour of Johanna L. Wood (pp. 43–82). Aarhus: Department of English, School of Communication & Culture, Aarhus University.Find this resource:

Arngart, O. (Ed.). (1968). The Middle English Genesis and Exodus. Lund: C. W. K. Gleerup.Find this resource:

Bech, K., & Walkden, G. (2016). English is (still) a West Germanic language. Nordic Journal of Linguistics, 39(1), 65–100.Find this resource:

Benson, L. D. (Ed.). (1987). The Riverside Chaucer. Boston: Houghton Mifflin.Find this resource:

Brook, G. L., & Leslie, R. F. (Eds.). ([1963] 1978). Laʒamon: Brut. London: Oxford University Press.Find this resource:

Calle-Martín, J. (2015). The split infinitive in Middle English. North-western European Language Evolution, 68(2), 227–250.Find this resource:

Cooke, A. H. (1925). The early history of Mapledurham. Oxfordshire Record Society, 7, 204–206.Find this resource:

Crisma, P. (2011). The emergence of the definite article in English: A contact-induced change? In A. P. Sleeman & H. Perridon (Eds.), The noun phrase in Romance and Germanic: Structure, variation, and change (pp. 175–192). Amsterdam and Philadelphia: John Benjamins.Find this resource:

Dalton-Puffer, C. (1996). The French influence on Middle English morphology: A corpus-based study of derivation. Berlin and New York: Mouton de Gruyter.Find this resource:

Dance, R. (2012). English in Contact: Norse. In A. Bergs & L. J. Brinton (Eds.), Historical linguistics of English: An international handbook (Vol. 2, pp. 1724–1737). Berlin: Mouton de Gruyter.Find this resource:

d’Ardenne, S. R. T. O. (Ed.). (1961). Þe liflade and te passiun of Seinte Iuliene. London, Toronto, and New York: Oxford University Press.Find this resource:

Denison, D. (1993). English historical syntax. London and New York: Longman.Find this resource:

Denison, D. (2000). Combining English auxiliaries. In O. Fischer, A. Rosenbach, & D. Stein (Eds.), Pathways of change: Grammaticalization in English (pp. 111–147). Amsterdam and Philadelphia: John Benjamins.Find this resource:

Filppula, M., & Klemola, J. (2009). Re-evaluating the Celtic Hypothesis [Special Issue]. English Language and Linguistics, 13(2).Find this resource:

Filppula, M., Klemola, J., & Paulasto, H. (2008). English and Celtic in contact. New York: Routledge.Find this resource:

Fischer, O. (1992). Syntax. In N. Blake (Ed.), The Cambridge history of the English language. Vol. 2: 1066–1476 (pp. 207–408). Cambridge: Cambridge University Press.Find this resource:

Fischer, O. (2013). The role of contact in English syntactic change in the Old and Middle English periods. In D. Schreier & M. Hundt (Eds.), English as a contact language (pp. 18–44). Cambridge: Cambridge University Press.Find this resource:

Fischer, O., Kemenade, A. van, Koopman, W., & Wurff, W. van der. (2000). The syntax of early English. Cambridge and New York: Cambridge University Press.Find this resource:

Fischer, O., & van der Wurff, W. (2006). Syntax. In R. M. Hogg & D. Denison (Eds.), A history of the English language (pp. 109–198). Cambridge and New York: Cambridge University Press.Find this resource:

Haeberli, E. (2004). Syntactic effects of inflectional morphology and competing grammars. In E. Fuss & C. Trips (Eds.), Diachronic clues to synchronic grammar (pp. 101–129). Amsterdam and Philadelphia: John Benjamins.Find this resource:

Hickey, R. (2012). English and the Celtic hypothesis. In T. Nevalainen & E. C. Traugott (Eds.), The Oxford handbook of the history of English (pp. 497–507). Oxford: Oxford University Press.Find this resource:

Hiltunen, R. (1983). The decline of the prefixes and the beginnings of the English phrasal verb: The evidence from some Old and Early Middle English texts. Turku: Turun Yliopisto.Find this resource:

Irvine, S. (Ed.). (2004). The Anglo-Saxon chronicle: A collaborative edition: Vol. 7. MS E. Cambridge: D. S. Brewer.Find this resource:

Iyeiri, Y. (2001). Negative constructions in Middle English. Fukuoka: Kyushu University Press.Find this resource:

Jespersen, O. (1894). Progress in language: With special reference to English. London: Swan Sonnenschein.Find this resource:

Kastovsky, D. (2006). Vocabulary. In R. M. Hogg & D. Denison (Eds.), A history of the English language (pp. 199–270). Cambridge and New York: Cambridge University Press.Find this resource:

Kroch, A. S. (1989). Function and grammar in the history of English: Periphrastic do. In R. W. Fasold & D. Schiffrin (Eds.), Language change and variation (pp. 133–172). Amsterdam and Philadelphia: John Benjamins.Find this resource:

Kroch, A. S., & Taylor, A. (2000). The Penn-Helsinki parsed corpus of Middle English 2 (PPCME2). Department of Linguistics, University of Pennsylvania. CD-ROM, second edition.Find this resource:

Kroch, A. S., Taylor, A., & Ringe, D. A. (2000). The Middle English verb-second constraint: A case study in language contact and language change. In S. Herring, P. van Rienen, & L. Schøsler (Eds.), Textual parameters in older language (pp. 353–392). Amsterdam and Philadelphia: John Benjamins.Find this resource:

Kurath, H., & Kuhn, S. (Eds.), (1956). Middle English dictionary. Ann Arbor: University of Michigan Press.Find this resource:

Laing, M. (2013–). A linguistic atlas of Early Middle English, 1150 to 1325. Version 3.2. Edinburgh: Angus McIntosh Centre for Historical Linguistics.Find this resource:

Lass, R. (1992). Phonology and morphology. In N. Blake (Ed.), The Cambridge history of the English language: Vol. 2: 1066–1476 (pp. 23–155). Cambridge: Cambridge University Press.Find this resource:

Los, B. (2005). The rise of the to-infinitive. Oxford: Oxford University Press.Find this resource:

Los, B., & van Kemenade, A. (2012). Information structure and syntax in the history of English. In A. Bergs & L. J. Brinton (Eds.), English historical linguistics: An international handbook (Vol. 2, pp. 1475–1490). Berlin: Mouton de Gruyter.Find this resource:

McIntosh, A., Samuels, M., Benskin, M., Laing, M., & Williamson, K. (1986). A linguistic atlas of late mediaeval English. Aberdeen: Aberdeen University Press.Find this resource:

McWhorter, J. (2002). What happened to English? Diachronica, 19, 217–272.Find this resource:

Meurman-Solin, A., López-Couso, M. J., & Los, B. (2012). Information structure and syntactic change in the history of English. Oxford and New York: Oxford University Press.Find this resource:

Minkova, D. (1982). The environment for Middle English Open Syllable Lengthening. Folia Linguistica Historica, 3, 29–58.Find this resource:

Miura, A. (2014). Middle English verbs of emotion and impersonal constructions: Verb meaning and syntax in diachrony. Oxford: Oxford University Press.Find this resource:

Moerenhout, M., & van der Wurff, W. (2005). Object-verb order in early sixteenth-century English prose: An exploratory study. English Language and Linguistics, 9(1), 83–114.Find this resource:

Munro, J. J. (Ed.). (1910). Lives of St. Augustine and St. Gilbert of Sempringham and a sermon. London: Oxford University Press.Find this resource:

Ritt, N. (2012). Middle English: Phonology. In A. Bergs & L. J. Brinton (Eds.), Historical linguistics of English: An international handbook (Vol. 1, pp. 399–414). Berlin: Mouton de Gruyter.Find this resource:

Rydén, M. (1983). The emergence of who as a relativizer. Studia Linguistica, 37, 126–134.Find this resource:

Schendl, H. (2012). Middle English: Language contact. In A. Bergs & L. J. Brinton (Eds.), Historical linguistics of English: An international handbook (Vol. 1, pp. 505–518). Berlin: Mouton de Gruyter.Find this resource:

Sylvester, L. (2012). Middle English: Semantics and lexicon In A. Bergs & L. J. Brinton (Eds.), Historical linguistics of English: An international handbook (Vol. 1, pp. 450–466). Berlin: Mouton de Gruyter.Find this resource:

Trips, C. (2002). From OV to VO in Early Middle English. Amsterdam: John Benjamins.Find this resource:

van Coetsem, F. (2000). A general and unified theory of the transmission process in language contact. Heidelberg: Winter.Find this resource:

van Kemenade, A. (2000). Jespersen’s cycle revisited: Formal properties of grammaticalization. In S. Pintzuk, G. Tsoulas, & A. Warner (Eds.), Diachronic syntax: Models and mechanisms (pp. 51–74). Oxford: Oxford University Press.Find this resource:

Westergaard, M. (2009). Word Order in Old and Middle English: The role of information structure and first language acquisition. Diachronica, 26(1), 1–65.Find this resource:


(1.) All Chaucerian references are from The Riverside Chaucer (Benson, 1987).

(2.) The conventional wh- is used as a cover term here for variant spellings, usually hw-.

(3.) The example is from the 13th-century life of St. Juliana, p. 9, lines 78–79 of d’Ardenne’s edition (1961).

(4.) This poem is thought to have been composed in the mid-13th century but is found in a manuscript assigned to the first half of the 14th century by LAEME. The first example is at line 901 and the second is at line 3716.

(5.) The example is at lines 5495–5496 (Otho version) in Brook and Leslie’s (1978) edition. This manuscript is from the second half of the 13th century, although the poem is thought to have been composed considerably earlier.