Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, LINGUISTICS (linguistics.oxfordre.com). (c) Oxford University Press USA, 2018. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 15 November 2018

History of the English Language

Summary and Keywords

The status of English in the early 21st century makes it hard to imagine that the language started out as an assortment of North Sea Germanic dialects spoken in parts of England only by immigrants from the continent. Itself soon under threat, first from the language(s) spoken by Viking invaders, then from French as spoken by the Norman conquerors, English continued to thrive as an essentially West-Germanic language that did, however, undergo some profound changes resulting from contact with Scandinavian and French. A further decisive period of change is the late Middle Ages, which started a tremendous societal scale-up that triggered pervasive multilingualism. These repeated layers of contact between different populations, first locally, then nationally, followed by standardization and 18th-century codification, metamorphosed English into a language closely related to, yet quite distinct from, its closest relatives Dutch and German in nearly all language domains, not least in word order, grammar, and pronunciation.

Keywords: history of English, Old English, Middle English, Early Modern English, phonology, morphology, word order change, grammaticalization

1. Historical Background

The earliest English grew out of the West-Germanic dialects (including English, Dutch, Frisian, and German) spoken by continental immigrants. Britain was a divided land after the break-up of Roman control in ad 410, and the coming of westward-moving tribes from the continent, whether by invitation or not, must have been waiting to happen. The Venerable Bede, a monk and historian in Northumbria in the 8th century, gives us a (perhaps not quite accurate) account of how two Frisian thanes—Hengist and Horsa—were invited by the Celtic king Vortigern to strengthen his troops against the Picts. The Frisian tribes managed to establish a powerbase in Kent, with other groups settling on the east coast north of the Humber and in East Anglia. Bede names three groups: Angles, Saxons, and Jutes. Over the next two centuries, they gained control over England, and by that time called themselves Angelcynn ‘English people.’ This term seems to have referred to “English” (rather than just the Anglo- part of the Anglo-Saxons), in contrast to people of Celtic origin.

The dialects spoken by these tribes were closely related, much more so than the present-day languages, which have undergone over a millennium of further divergence from their common (West-)Germanic origin. In spite of this dialectal variation known to have existed, Old English is often presented as a relatively unified language. This is no doubt because the vast majority of the available texts hail from the West-Saxon area, where King Alfred the Great (reign 871–899) initiated a national literary tradition. This is also why there is no hardly any direct linguistic evidence for the impact of the Scandinavian conquest in the northeast of England during and following the reign of King Alfred.

In the transition from Old English to Middle English, English underwent a number of pervasive changes in phonology, morphology, vocabulary, and syntax. These changes have often been ascribed to French influence following the Norman Conquest of England in 1066. It is doubtful whether this is correct, though. The Norman Conquest started in the south of the country and left its imprint mainly in the top layers of society and in the written language. It is clear that this caused a tremendous influx of French loanwords. There is little evidence, however, that French influence penetrated the language much deeper than that. The changes in the phonology of unaccented syllables (reduction of unstressed vowels to schwa) that had a domino effect in the morphology (reducing case endings) had already been underway in the north of England in the Old English period, before French influence could take effect. Recent work on the matter hypothesizes that it is the Viking conquest of the north and east of England in the 9th and 10th centuries that had the more profound influence, as it resulted in long-term settlement and assimilation of invading and native populations. However plausible this is, we cannot directly verify it because the text material from the relevant period is written in the West-Saxon dialect.

The West-Saxon literary tradition of the Old English period was broken by the Norman Conquest, and it was not before the second part of the 14th century that another one started developing. Pervasive social changes towards the end of the Middle Ages, and the growth of cities resulting from increased trade and commerce, led to unprecedented interregional migration, and London became a melting-pot city, with conditions of dialect contact and multilingualism. This third layer of language contact was the final force that shaped the development of the present-day language: with loss of (verb) inflections, fixing of SVO word order, the rise of a system of auxiliaries, and the incipient creation of a standard language. The first new literary standard to arise was that apparent in the work of Wycliffe and his followers (late 14th, early 15th century). The dialect of these works has proved difficult to localize precisely and appears to be based on the dialects of the central Midland counties. The clearest indication that it was a literary standard is the fact that it survived in written form unchanged until the late 15th century. There is good reason to assume that it was one of the sources for the emerging modern standard norm. Standardization definitely set in once King Henry V started using English in his official letters (ca. 1420), and English became the official language of the court and the king’s chancery.

2. The Phonological Makeup of Older English

This section will first survey the sound system of Old English, followed by a brief treatment of the developments that distinguish Old English from the other West-Germanic languages phonologically. We will then highlight some of the developments that shaped the phonology of English in its transition to the present-day language.

2.1 Old English Vowel System

2.1.1 Vowels

The system of Old English vowel phonemes in stressed syllables is presented in Table 1. The spelling of the vowels corresponds fairly consistently and transparently with these. Vowels can be long or short; the distinction appears to be simply quantitative. Vowel length is not often marked, but when it is, it is done by a length-mark over the vowel, such as in snīðan ‘cut’ vs. its participle gesniden. There are seven pairs of long/short vowels.

Table 1. Vowels of Old English

History of the English Language

Vowels in unstressed syllables were spelled as distinct in the 9th and perhaps the 10th centuries, but it is generally accepted that they had been reduced to one centralized vowel schwa by the end of the Old English period. This reduction played a major role in the development of inflectional endings, to which we will return. Old English had two pairs of diphthongs which could be either long or short: /e(ː)a/, /e(ː)o/. This leaves aside <io>, which is considered a spelling variant of /eo/, and <ie>, which is used for diphthongs that have undergone umlaut, to be discussed below.

The Anglo-Saxon migrations to England had heralded a period of rapid language change. A series of sound changes distinguish Old English from its West-Germanic antecedents. The most prominent among these are three vowel assimilation processes: breaking (shared with Frisian), back mutation (aka back umlaut), and i-mutation (aka i-umlaut). Breaking and back mutation have very similar effects of creating diphthongs. In the case of breaking, this is triggered by the following consonant: /æ/ becomes [æa], spelt <ea>, and /e/ becomes [eo], spelt <eo>, when followed by a velar fricative or by a liquid and a further consonant. Examples are neah ‘near’ (cf. German nahe), feoh ‘property’; earm ‘arm’ (German Arm), heorte ‘heart’ (cf. German Herz); eall ‘all’ (German alle), seolfor ‘silver.’ Back mutation (which is less regular across dialects) achieves the same result, but is triggered by a back vowel in the following syllable. Examples are ealond ‘island’ (cf. Dutch eiland), and weoruld ‘world.’ I-umlaut, on the other hand, had a fronting effect, triggered by a /i/ or /j/ in the following syllable. Examples are hǣlan (from *haljan) ‘heal’, ele (from *oli) ‘oil’, and trymman (from *trumjan) ‘strengthen.’ Back mutation, and i-mutation even more so, had a tremendous impact across complete inflectional classes, such as the class I weak verbs, which by virtue of the ‑i- in their stem extension had an umlauted stem vowel (hǣlan and trymman are examples of this), and nouns of the athematic noun declension, whose plural suffix had been ‑i. This still underlies some irregular plurals in the present-day language, such as goose/geese (from *gosi), mouse/mice (from *musi).

2.1.2 Consonants

The consonant phonemes of Old English are represented in Table 2.

Table 2. Consonants of Old English

History of the English Language

The phoneme symbols /f, θ‎, s, x/ do not reflect a voice opposition. The voice opposition for fricatives was presumably not phonemic in Old English, and the voiced/voiceless values were in complementary distribution as a result of a prehistoric sound change voicing fricatives in voiced environments. Thus, in all but these environments [f, θ‎, s] are voiceless. To illustrate, in fōt ‘foot’, geaf ‘gave’, and sceaft ‘shaft’, the fricative would have been voiceless as it is word-initial (fōt), word final (geaf) or precedes a voiceless stop (sceaft); in drīfan ‘drive,’ wulfas ‘wolves,’ and hræfn ‘raven,’ it is voiced, in intervocalic position, or preceding a nasal or liquid. Likewise, the distinction between palatal and velar consonants (and affricates) was perhaps not phonemic: The palatal variants were the result of palatalization, a sound change affecting velar consonants when they are followed by an original front vowel (i.e., a front vowel not the result of i-mutation). Examples of the palatal affricate /ʧ, ʤ/, spelled <c, cg>, are læccan (*lakjan) ‘seize’, ceace (*kaak) ‘cheek’, bycgan (*bugjan) ‘buy’, ceorl (*kerl) ‘churl’, cirice (*kirike) ‘church’, geard (*gard) ‘yard’, pic (*pik) ‘pitch’, dæg (*dag) ‘day’, all with the palatal value, vs. climban ‘climb’, cræft ‘craft’, cu ‘cow’, gold ‘gold’, boc ‘book’, plog ‘plough’, all with the velar quality. The sibilant /ʃ/, spelled <sc>, is an exception here, since it arose out of palatalization of *sx, and its palatalization became general, not limited to front environments: sceolde ‘should,’ scinan ‘shine,’ but also scrud ‘shroud,’ all with palatal sc. Original Anglo-Saxon words with palatalized consonants may be contrasted with later Scandinavian loans that have the nonpalatalized variant: native shirt vs. Scandinavian skirt; native yard vs. Scandinavian garden; native eie (dialectal) vs. Scandinavian egg.

2.2 From Old English to Middle English

2.2.1 Vowels

In the transition to Middle English, vowels in stressed syllables underwent a number of quantitative changes, variously involving lengthening and shortening of vowels. These are interpreted by some scholars as a response to changes in word prosody resulting from loss of inflectional endings. Vowel lengthening took place before a nasal or liquid followed by a homorganic voiced stop: <-nd, -mb, -ld, -rd, -ng>, as in hand > hānd ‘hand’, climban > clīmben ‘climb’, cald > cāld ‘cold’, word > wōrd ‘word’, strang > strāng ‘strong.’ This change did not apply to clusters with three consonants: While the singular of ‘child’ was lengthened, cild > cīld, the plural remained cildru with a short vowel. Open syllable lengthening is another case, curiously shared in varying degrees by the other West-Germanic languages. Open syllables consist of a stem vowel, followed by a single consonant; some examples are: beran > bēren ‘bear’, mete > mēte ‘meat’, macian > māken ‘make’, hopian > hōpen ‘hope.’ In all other environments, i.e., those involving consonant clusters other than the above, and double consonants, vowels were shortened: sōfte > soft ‘soft’, fīftig > fiftig ‘fifty’, also cēpen ‘keep’ vs. past tense cepte ‘kept’, etc. Moreover, vowels were shortened in the first syllable of trisyllabic words: crīstendōm > cristendōm ‘Christendom’, sūþerne > suþerne ‘southern’, etc.

A major development in the transition from Old to Middle English affects unstressed syllables, and this primarily had an impact on inflectional endings on nouns, verbs, and adjectives: To the extent that their quality was distinct towards the end of the Old English period, the vowels of unaccented syllables in Middle English are spelled <e>, presumably reflecting a schwa-like quality. This process seems to have started earlier in the North than in the South, and may be attributed to language contact with Scandinavian. In addition, word-final ‑n was lost in the North, which had a further impact on inflectional endings of weak nouns, as we will see below.

2.2.2 Consonants

In the course of the Middle English period and the 16th century, specific consonants were lost or vocalized, yielding new long vowels or diphthongs.

The first of these consonants was the voiced velar fricative, spelled <g>. It first developed to the palatal approximant [j] following a front vowel, and it subsequently became the second part of a long front diphthong, thus: OE dæg > ME dai ‘day’; OE weg > ME wei ‘way’; OE segl > ME sail ‘sail’; OE ege > ME eie ‘eye.’ After back vowels, the voiced velar fricative changed first to [w], then to the second part of a back diphthong: OE lagu > ME lawe ‘law’; OE agan > ME owen ‘own’; OE boga > ME bowe ‘bow’; OE plogas > ME plowes ‘ploughs.’

Next in line was the voiceless velar fricative, spelled <h> or <gh>. The first stage was the formation of a glide between a vowel and the voiceless velar fricative /x/ (spelled <h> or <gh>). After a front vowel, a front glide developed: ME eghte > eighte ‘eight’, ME heh > heih ‘high.’ After a back vowel, a back glide developed: ME tahte > taughte ‘taught’, ME dohter > doughter ‘daughter.’ The next, or perhaps concurrent, step was the loss of the velar, leaving the stem vowel long and feeding into the Great Vowel Shift, which we will discuss below. This is illustrated by the present-day pronunciation of these same examples. There was an alternative development when the stem vowel was a back vowel: The velar could develop into /f/, often but not always in combination with a short stem vowel. Examples of this are rough, tough, enough, laughter.

2.3 From Middle English to Early Modern English

The most famous sound change, indeed the mother of all sound changes in the history of English, is the Great Vowel Shift (GVS), which took place over an extended period of time spanning late Middle English and Early Modern English. The Great Vowel Shift affected long vowels in stressed syllables. It was largely unconditioned. It has long been thought of as a chain shift change, where the various steps in the chain are causally related. A typical diagrammatic representation is provided in Figure 1.

History of the English LanguageClick to view larger

Figure 1. A typical diagrammatic representation of the Great Vowel Shift.

The chain shift view implies that the shift started at the top end or bottom end of the vowel diagram, at the top end with diphthongization of the high vowels, necessitating the raising of lower vowels by one position (the drag chain view). Alternatively, it started with the raising of low vowels, necessitating the raising of the higher ones (the push chain view). Recent research shows that the neat diagram may well apply to the end result of the chain, but that the process that led to it was anything but unitary. Rather, it was the result of long-term variation and diffusion, in which various stages can be identified. There was, moreover, differentiation between dialects, which came into contact in a complex sociolinguistic setting. Broader research into vowel shift changes shows that they usually involve long vowels and suggests that these are phonetically unstable, hence prone to change in situations of dialect contact, a typical result of the interregional migration and societal upscaling of the late Middle Ages. Incipient standardization of the language also created prestige varieties that had a further impact as social models for upwardly mobile folk such as the relatively new middle classes.

3. Inflectional Morphology

In the stages of Germanic prior to Old English, nouns and verbs typically consisted of a stem, a stem suffix determining the inflectional class, and an inflectional ending. Old English inflectional classes are still categorized on this basis, though it is not clear to what extent the addition of the stem suffix was still productive in Old English; it had been lost in most cases through phonological developments, which left their traces on the phonological makeup of the stem. One example of this, persisting even to the present day, is the plural of nouns of the athematic declension (e.g., mouse/mice, plural *musi < i-umlaut mys < unrounding mīs < GVS [mɑis]; goose/geese, plural *gosi < i-umlaut gēs < GVS [gi:s]).

3.1 Nouns

Old English nouns are divided into stem classes distinguished on the basis of their original proto-Germanic stem suffixes. If the original stem suffix ended in a vowel (a, ō, i, u), the declension was strong; if the original stem suffix ended in n preceded by a, ō, or i, the declension was weak. There are a few additional minor declensions with stems ending in -nd or -r, as well as athematic stems.

Gender in Old English is grammatical, not natural as in Modern English. Gender depends partly on the stem class: a-stems are masculine or neuter, ō-stems are feminine, i-stems are masculine, feminine, or neuter, u-nouns are masculine or feminine. The gender of the weak stems is determined by the vowel preceding the -n as in the strong stems. The gender of the athematic stems is masculine or feminine; that of the -nd stems masculine or feminine; that of the -r stems masculine or feminine.

Nouns have a two-term number contrast: singular and plural, and there are five cases: Nominative, Accusative, Dative, Genitive, Instrumental. The last of these is being lost as early as the Old English period. Number and case together are expressed by a single inflectional ending.

Table 3 gives the inflectional paradigm of the most prominent strong nouns, the a-nouns including masculine and neuter nouns. The weak declensions are exemplified in Table 4.

Table 3. Declension of a-stems

Singular

Plural

masc.

neut.

neut.

masc.

neut.

neut.

Nom.

stān

scip

word

stānas

scipu

word

Acc.

stān

scip

word

stānas

scipu

word

Gen.

stānes

scipes

wordes

stāna

scipa

worda

Dat.

stāne

scipe

worde

stānum

scipum

wordum

‘stone’

‘ship’

‘word’

Table 4. Weak Declensions of Nouns

Singular

Plural

masc.

fem.

neut.

masc.

fem.

neut.

Nom.

guma

tunge

ēage

guman

tungan

ēagan

Acc.

guman

tungan

ēage

guman

tungan

ēagan

Gen.

guman

tungan

ēagan

gumena

tungena

ēagena

Dat.

guman

tungan

ēagan

gumum

tungum

ēagum

‘man’

‘tongue’

‘eye’

Case morphology was reduced drastically in the transition to Middle English. This started with the reduction of distinct vowels in unstressed syllables to schwa (spelled -e). The fact that this leveling of inflectional endings was earlier in the north and east of England supports the idea that it was the result of language contact with Scandinavian: The first traces of it can be found in the 10th-century Northern Lindisfarne gospels. As noted, there is, however, little direct linguistic evidence to support this conclusion. In a text as early as the Peterborough Chronicle (early 12th-century East Midlands), two things about the reduction of noun endings stand out: first, the nominative and accusative plural of the older masculine and neuter a-stems is reduced to -es and extended to all the plurals (including dative) of all declensions. The dative ending on nouns was lost completely. Southern and Western texts at this stage or even later have not advanced quite as far as this, but the general trend is the same.

3.2 Pronouns

The set of personal pronouns in OE is given in Table 5, in the standard spelling. The personal pronoun forms in Old English are used for pronominal reference to persons and entities, and for reflexivization with or without emphatic self, thus: ic bletsige me ‘I bless myself.’

Table 5. Personal Pronouns

Sg

1

2

3

Pl

1

2

3

.

masc

Fem.

neut

Nom.

ic

þu

hēo

hit

Acc.

þē

hine

hit

ūs

ēow

Gen.

mīn

þīn

his

hire

his

ūre

ēower

hira

Dat.

þē

him

hire

him

ūs

ēow

him

Number is distinguished as singular and plural, with the relatively infrequent dual form omitted. There are four cases. Gender is distinguished in the third person singular. The nominative–accusative distinction is made in all persons and numbers except the third person singular neuter and the third person plural. The dative–accusative distinction is made in the third person paradigms, both singular and plural.

Old English also had two series of demonstrative pronouns, which are shown in Table 6.

Table 6. Demonstrative Pronouns

Demonstrative

Deictic

Sg.

Pl.

Sg.

Pl.

masc.

fem.

neut.

common

masc.

fem.

neut.

common

Nom.

se

sēo

þæt

þā

þes

þēos

þis

þās

Acc.

þone

þā

þæt

þā

þisne

þās

þis

þās

Gen.

þæs

þǣre

þæs

þǣra

þisses

þisse

þisses

þissa

Dat.

þǣm

þǣre

þǣm

þǣm

þissum

þisse

þissum

þissum

The deictic variant translates as ‘this, that, these, those’, which could be used as a deictic pro-form or to mark an NP. The demonstrative paradigm, aka the se paradigm, is rather interesting as it marked several core functions in Old English: The first is that of demonstrative determiner and definiteness marker. It could as such simply mean ‘the.’ It could also be used as an independent pronoun, often in clause-initial position, with a deictic reading. (1a) illustrates this. (1b) illustrates its use as a relative pronoun:

(1)

History of the English Language

The demonstrative pronoun appears to be used in clause-initial position as a topic-switcher, referring to the nonsubject referent in the previous clause as in (1a). It clearly served an important function as an interclausal reference tracker.

The transition to Middle English affected the pronominal paradigms in crucial ways. The dative–accusative distinction for personal pronouns was neutralized, to first singular , second singular þē, and third singular him, her, and hit. Feminine nominative hēo developed to she, probably to disambiguate it from other third person subject forms. It has long been maintained that the present-day third person plural pronouns they, their, and them were borrowed from Scandinavian in the Northeast, illustrating the intensity of contact with Old Norse, but Cole (forthcoming) makes an interesting case that they could be a Northern remodeling of the plural demonstratives þā, þǣra, þǣm, perhaps promoted by Scandinavian contact. The demonstrative pronoun paradigm was lost in its entirety, with relic forms being in demise over the 12th century, and the indefinite article the replaced their determiner function. Their topic switch function was lost and to some extent replaced by relative clauses, while their relative pronoun function was taken over by the wh-pronouns, which up to that point had served as interrogatives, indefinites, and quantifiers. Finally, over the 16th and 17th centuries the second plural form ēow, ēower, which developed into you, your, took over from the original second singular þ-series, a change attested in many languages. This development is thought by some to be related to the original second singular -st verb ending, but this is by no means firmly established.

3.3 Adjectives

In prehistoric times, adjectives had been inflected as nouns, in agreement with the noun modified, and we can still trace the two main Old English adjectival declensions back to the main nominal declensions. The so-called strong declension or indefinite declension corresponds for the most part to strong nominal declensions. It is employed when an adjective is used predicatively, or when it modifies an indefinite NP, such as in we habbað godne gast ‘we have a good spirit.’ The weak or definite declension largely corresponds with the weak nominal declension and is used when the attributive adjective follows a demonstrative determiner, as in þæt mon þa godan Mammeam gemartrade ‘that they should martyr the good Mammea.’ Vestiges of this subtle system of definiteness marking are last found in Chaucerian English.

3.4 Verbs

As in the other Germanic languages, there is a distinction between weak and strong verbs. The weak verbs originally consisted of a stem plus a stem suffix, and the past tense is expressed by adding a dental suffix to the stem suffix. Strong verbs have no stem suffix, and the past tense is expressed through ablaut alternations in the root. The tense system has only one opposition: preterite vs. present.

3.4.1 Verb Classes

Weak verbs. Two original stem suffixes distinguish the various classes of weak verbs in Old English, are important: Class I of the weak verbs had a stem suffix i/j; class II had a stem suffix oi/oj. and left its mark in the form of i-umlaut of the root vowel, thus Gmc. *framjan > OEng. fremman ‘do, perform’; Gmc. *narjan > OEng. nerian ‘save.’ In class II, the i/j suffix similarly disappeared. No i-umlaut took effect since the o suffix intervened. The latter is still found in the Old English preterites of class II, cf. the contrast between class I infinitive fremman, pret. 1 sg. fremede and class II infinitive lufian ‘love’, pret. 1 sg. lufode. It is customary to distinguish a weak class III, which consists of four rather irregular verbs with dental past endings: habban ‘have’, libban ‘live’, secgan ‘say’, hycgan ‘think.’

Strong verbs. The strong verbs form their past tenses by means of vowel ablaut. There are seven ablaut classes for the strong verbs. The present-day irregular verbs of English are relics of these ablaut classes.

In addition to the verb classes mentioned so far, there is a class of so-called preterite-present verbs: a class of some 14 originally perfective verb stems which developed a new weak past tense in Germanic. It includes the modals scæl ‘shall’, mæg ‘may’, cunn ‘can’, mōt ‘must’, and several more, such as þearf ‘need’ and dearr ‘dare.’ The preterite presents what we may call a defective paradigm, lacking nonfinite forms, and showing only a two-term contrast in the present and past indicative. There is furthermore a small group of highly frequent, highly irregular verbs: beon wesan ‘be’, willan ‘wish, will’, don ‘do’, gan ‘go.’

The inflectional categories for verb forms are illustrated here for the weak verb fremman ‘perform, do.’ The following grammatical distinctions may be noted: finite/infinitival/participial (present/past); present/past tense; indicative/subjunctive/imperative mood; singular/plural; 1st, 2nd, and 3rd person in the singular.

Fremman

present

past

Indicative

1sg ic

fremme

fremede

2sg þū

fremest

fremedest

3sg hē, hēo, hit

fremest

fremede

1-3pl wē, gē, hī

fremmað

fremedon

Subjunctive

1-3sg ic, þū, hē, hēo, hit

fremme

fremede

1-3pl wē, gē, hī

fremmen

fremeden

Imperative

2sg

freme

2pl

fremmað

Participle

fremmende

gefremed

It was observed above that the inflections on nouns were reduced rather drastically in the transition to Middle English, as a result of vowel weakening in unstressed syllables. Distinct vowels were reduced in verb endings as well, but their consonantal properties were maintained and the system of verb endings in early Middle English (subject to some dialectal variation) still made a distinction for three persons in the singular present indicative, two in the singular preterite indicative, and had a clearly marked number contrast in the present as well as the preterite. Further reductions that led to the Modern English system took place largely in the 14th and 15th centuries, leading to the elimination of distinct person endings in the singular (apart from third singular -s) and the plural, and the loss of the subjunctive mood.

4. Syntax

The syntax of any language is closely interwoven with its inflectional morphology, since inflections manifest the expression of grammatical categories, such as tense, aspect, mood, negation, clause type, and so on. Inflections also contribute importantly towards determining word order. The classical observation in this respect is that rich inflection tends to correlate with free word order. We will nuance this view below. This section focuses on two domains of syntactic change: on the changing expression and structure of grammatical categories (grammaticalization) and on word order.

4.1 Grammaticalization

We use the term “grammaticalization” for the process by which grammatical categories arise. The term is also applied to a specific theoretical model called grammaticalization theory, but now that the fierce debate of the 1990s and 2000s on a “theory” of grammaticalization has abated, we forego discussion of this. Grammaticalization is generally seen as a change by which a lexical item develops into a grammatical marker. This is accompanied by a reduction in phonetic substance, loss of syntactic independence, and of lexical meaning. The semantic pathway is often held to be one of subjectification of meaning. The reduction can be described as follows:

(2)

History of the English Language

This developmental path immediately clarifies why inflectional endings on, for example, nouns and verbs find themselves towards the end of it. We can think of this as a cyclical development: Once inflections have been reduced to zero, as happened to many of the Old English inflections, the cycle starts again and some lexical item is recruited as a grammatical marker. The notion of “cycle” was famously coined by Otto Jespersen in his 1917 book on negation, in which he establishes what is often called the negative cycle, found in many languages:

  • Stage 1: Negation is expressed by one negative marker.

  • Stage 2: Negation is expressed by a negative marker in combination with a negative adverb or noun phrase.

  • Stage 3: The second element in stage 2 takes on the function of expressing negation by itself; the original negative marker becomes optional.

  • Stage 4: The original negative marker becomes extinct.

This paves the way for our first example of grammaticalization: Old English was at stage 1 with a hint of stage 2—negation was typically expressed by the preverbal clitic ne, often contracted with the finite verb as in (3); a secondary sentence negator occurs at low frequencies, but the precursor of later not is the negated NP no wiht, lit. ‘no creature’, which in Old English can be used as an emphatic negative adverb, as in (4):

(3)

History of the English Language

(4)

History of the English Language

Early Middle English is firmly at stage 2: Most negated contexts show a combination of ne–not, where not is now grammaticalized as an (unemphatic) negative marker:

(5)

History of the English Language

Moving on to later Middle English, ne is going extinct, and Chaucer writes about the monk in the Canterbury tales:

(6)

History of the English Language

Once ne is lost, we see the first signs of negators contracted with the verb, as we see it in present-day English He didn’t care a fig. In overlapping stages, the picture looks like Figure 2.

History of the English LanguageClick to view larger

Figure 2. Overlapping stages of negators contracted with the verb.

The negative cycle is a schoolbook illustration of a grammaticalization path as in (2). The next question is how these developments work in terms of language change. This particular cyclical development spans a millennium as if it is part of a grand plan, but this is not something that speakers and language learners over the centuries could have been aware of. There may be what seems like a continuity diachronically, but we cannot but assume that each speaker’s grammar represents one synchronic stage. Note in this context that ne at stage 1 is a preverbal clitic, whereas the origin of not is a negated noun na wiht. Not is then grammaticalized to the status of negative marker at stage 2, i.e., it was a full (negated) constituent earlier, and it was reanalyzed by speakers as a grammatical word marking negation. We will consider this further when we have outlined another, closely related, case of grammaticalization: that of the rise of auxiliaries in English, involving the auxiliation of modal verbs, and the rise of what is often called the dummy auxiliary do as in Why did you go there?

The auxiliation of modals is a complex of changes in the verb system between 1400 and 1800; it shows how the loss of inflections, its syntactic effects, and grammaticalization can be seen to interact and complement each other. Most modal verbs in Old English belonged to the class of preterite present verbs, a verbal inflectional class that had a defective morphological paradigm, significantly lacking nonfinite forms and showing only a two-term person contrast in the present and past indicative singular. They had incipient purely grammatical uses (e.g., sceal ‘shall’ marking future tense), but they were themselves marked for an indicative/subjunctive mood distinction. They also had a number of main verb characteristics, and could, for instance, be used without an infinitive or other complement (Wel þæt swa mæg ‘that may well be so’, lit. ‘Well that so may’), or have that-clause complements or objects. While they were very predominantly used with bare infinitives, these main verb uses and their mood marking show that modals were not simply auxiliaries in Old English.

The complex of change affecting the modals in its detail probably almost spans the full history of English, but there are a number of clear shifts in the period 1400–1700:

  1. (a) Morphology marking the subjunctive mood was lost, and singular/plural contrasts in the indicative mood were lost.

  2. (b) Modal verbs were losing their main verb properties, in particular complementation by that-clauses and objects, and their use as independent verbs without any complement. They lost their status as full verbs and became finite auxiliaries occurring strictly to the left of the negator not.

  3. (c) Finite main verbs ceased to occur on the left of adverbs, negative or otherwise, in sentences like He saw not John, which were still attested in the 14th century.

  4. (d) The rise of periphrastic do.

A well-accepted account of these changes is that they reflect a reorganization of the verbal system such that modals became auxiliaries, that is, grammaticalized modality markers which were obligatorily placed to the left of not, whereas finite main verbs could no longer be placed left of not. Once this new category auxiliary was in place, periphrastic do, originally a causative verb, was grammaticalized as a default auxiliary in the NICE contexts (Negation, Inversion, Code (ellipsis), Emphasis). A plausible way of framing this insight is that modals were lexical verbs in Old English, and like finite lexical verbs could be placed in second position in questions, and to the left of not in negation contexts. (7) and (8) exemplify questions, with a lexical verb and with a modal verb respectively:

(7)

History of the English Language

(8)

History of the English Language

In other words, finite lexical verbs and modal verbs were in the same positions in the clause. The essence of the reorganization of the verbal system between 1400–1700 was that it gave rise to a division of labor between auxiliaries, used in the NICE contexts, and lexical verbs, which could no longer be placed in the position now taken by auxiliaries. Negative not could, once ne was lost, be contracted with the auxiliary, as in He didn’t care a fig.

Auxiliation was not restricted to modals and do: A wide range of periphrastic verbal expressions arose during this same period, for instance have to, used to, to be to. These are usually called semi-auxiliaries, indicating that they mix main verb and auxiliary properties. They have in common that they express grammatical meaning (have to expresses obligation, used to habitual aspect in the past, to be to future activity), which is part and parcel of grammaticalization processes.

The negative cycle and the auxiliation of modals are two particularly well-known examples of grammaticalization, and both had a profound influence on the historical development of English syntax. There are many more examples though, including, for example, the grammaticalization of the demonstrative pronoun that to a conjunction marking a subclause (the man that I love), of the locative adverb there to an expletive subject (there is a man in the garden), and of the cardinal number one to a pronoun (a red one), to name but a few.

4.2 Word Order

Word order is another domain of major change in English. Example (7) above shows that subject–verb inversion was not restricted to questions in Old English, and (8) shows that objects could precede the verb in Old English. Both these facts, and the changes that resulted in the present-day situation, have been subject to considerable debate. Present-day English Subject Aux Inversion, as in Why did you go there? is a residue of a broader phenomenon in Old English that should be viewed in a larger Germanic perspective of the “Verb Second” constraint (V2). Old English OV order likewise invites comparison with the continental West Germanic languages; we will consider them in turn.

4.2.1 Subject Aux Inversion and Verb Second

Subject-initial clauses in Old English are often SVO as in (9)

(9)

History of the English Language

The word order typical of questions as in (7–8) is not restricted to auxiliaries as noted above, and it is found in several other contexts as well: negative-initial clauses and clauses introduced by discourse-sequencing and correlative adverbs such as þa and þonne ‘then’, as the examples in (10) show:

(10)

History of the English Language

Inversion in contexts with questions is restricted to main clauses except with specific types of intransitive verbs. This is the primary consideration for analyzing Subject Aux Inversion structurally along the lines of the Verb Second (V2) constraint that is typical of the other Germanic languages: finite verb (Vf) fronting, following a first constituent other than the subject, is taken to be to the same position as the conjunction in subclauses, in something like the following template, in which the subclause is constructed:

Main clause

Clausal prefield Vf

Clause proper

Hwi

wolde

God

swa lytles þinges

him

forwyrnan

Subclause

Clausal prefield

Clause proper

Saga me

hwæðere

God

swa lytles þinges

him

forwyrnan

wolde

Tell me

whether

God

such a small thing

him

deny would

In a way then, the position of the finite verb signals a main clause, whereas a conjunction like whether signals a subclause. This is still the case in present-day Dutch and German. This has been taken to suggest (Kiparsky, 1995) a possible motivation for the rise of V2 as an early Germanic innovation: it served to distinguish main clauses from subclauses, as clausal relations became more hierarchical, or hypotactic, as it is usually called.

There is a second type of word order that resembles V2, although it shows an interesting quirk: a clause-initial prepositional phrase or an object typically yields inversion when the subject is nominal, but not when it is pronominal: This is illustrated in (11):

(11)

History of the English Language

This suggests that the organization of the clausal prefield in Old English is more complex than that of the present-day (West)-Germanic languages—it suggests in particular that pronominal subjects are in an earlier position in the clause than nominal subjects, with another position for the finite verb in between:

Main clause

clausal prefield (Vf)

Clause proper Vf

Be ðæm

we

magon

swiðe swutule

oncnawan

that…

By that

we

may

quite clearly

perceive

that…

This later fronted position for the finite verb might occur in subclauses as well as it part of the clause proper, and it does, although not with inversion. (11) shows that there is a form of finite verb fronting in embedded clauses as well:

(12)

History of the English Language

Verb Second continued to thrive in Middle English, but substantial parts of it were lost between roughly 1400–1650, and SVO word order became fixed. The type exemplified in (11) was lost first, as nominal subjects increasingly occurred in preverbal position over the 15th century. The more systematic type of V2 in questions and negative-initial clauses and then-initial clauses remained more robust, until the then-initial context was lost over the 15th–16th centuries. This left questions and negative-initial contexts as inversion contexts, where inversion became restricted to auxiliaries, as the auxiliation of the modals and the rise of do-support kicked in.

It is worth emphasizing that the phenomenon of Verb Second is in principle independent of the order of object and verb. This can be seen very clearly by comparing Modern Dutch with Modern Swedish. Both languages regularly have verb-second in main clauses, but Modern Dutch has object–verb order, while Modern Swedish has verb–object order. Therefore, a Dutch object will precede any verb that is not in second position, while a Swedish object follows it; examples are given in (13) and (14).

(13)

History of the English Language

(14)

History of the English Language

This is also clear from the fact that object–verb (OV) order was almost completely lost before Verb Second even started to be lost. We will turn to object–verb order now.

4.2.2 Object–Verb Order and Its Loss

The position of the nonfinite verb is where we get a precise picture of variation between OV and VO word orders. Objects and prepositional groups are the constituents whose position is variable with respect to the nonfinite verb. The examples below give an illustration of the five main word order patterns with objects, distinguished according to whether the finite verb precedes or follows the nonfinite verb (AuxV vs. VAux):

(15)

History of the English Language

(16)

History of the English Language

Note that these are nominal objects; objects that are pronouns occur in OV orders a good deal more frequently. To give an idea of the proportions of the word order patterns: Of all the subclauses with a finite as well as a nonfinite verb, some 75% of nominal objects occur in one of the OV word orders (15a–b) and (16a). A very substantial part of these were lost in the transition to Middle English: OVAux was gone by 1250, and OAuxV and AuxOV became restricted to pronoun objects by 1350. OV order had virtually disappeared by the end of the Middle English period.

It is a much debated question what factors determined the pervasiveness of OV word order in Old English, and what, in turn, its demise. This seems to be due in part to what is usually called “basic word order,” in part to case marking on objects, allowing them more positional freedom, but there is also good evidence that OV order was more frequent when the object refers back to a person or entity in the context. For instance, þæt bysmor ‘the disgrace’ in (16a) refers back to the story betrayed by Melantia that Eugenia wished to sleep with her. This idea is confirmed by the fact that the remnants of OV word order in the earliest Middle English are exclusively of that type, exemplified in (17):

(17)

History of the English Language

The object ðese eadi hope refers back in the immediate context to God’s promise to turn humility into great brightness. One hypothesis on the causes of the loss of OV is that it was caused by contact with Scandinavian in the Danelaw area. Scandinavian is thought to have been VO at that point, and language contact supposedly accelerated the loss of case. It is also clear that the loss of OV was more advanced in Northeastern parts in Middle English than it was in the South.

5. Words and Word Formation

The major part of the Old English vocabulary is from common Germanic stock. Old English had a number of typically Germanic word-formation processes. Some of these have survived to the present day, such as prefixation by un- as in to undo, and suffixation by -ness as in boldness, or by -hood, as in sisterhood. The suffix -ly as in beautifully derives from the suffix -lic, although this was not an adverb-forming one in Old English, but became so later. Most of these processes lost their productivity and have left us some relics, such as withstand from the productive verbal prefixing process expressing ‘against-V’, cf. OEng wiþcweþan ‘contradict (lit. against-speak).’ Another example is remnants with be-, originally a transitivizing prefix, for instance, bespeak, bemourn. This prefix is still (or again?) productive, it seems, in informal registers. Many of the OE word-formation processes have left no trace in the standard language; French and Latin processes have taken their place, for instance suffixation by -ment, -ous, -able, -al, -ance, -ate, -ize, -ant; prefixation by circum- (OE ymb-), counter- (OEng. with-), dis-, mis-, inter-, mal-, pre-, post-, re-, semi-, sub-, vice-. The list is far from complete. French loan-formations are due to the influence of the French invaders after 1066; Latin loans during and after the Middle English are usually the result of prestige borrowing.

The earliest loanwords in Old English are loans from Latin, and some from languages that the Anglo-Saxons had contacts with, such as Celtic. The Scandinavian invasions of the 9th and 10th centuries have left their mark on the vocabulary, invariably words for everyday things. Some of them are recognizable on the basis of their velar consonants where native Anglo-Saxon words would have palatalized consonants: bank, rig, skirt, scar, car, gate, crook, garden, but also fell, holme, force, tarn, wath, etc. The Norman Conquest led to a tremendous influx of loanwords from French, largely naming notions relating to royalty, court matters, law terms and military terms, and food terms for the rich. The Renaissance brought further layers of prestige borrowings from Latin and Greek.

6. Critical Analysis of Scholarship

The references briefly discussed here can mostly be found in the article “History of the English Language” in Oxford Bibliographies Online. If not, they are listed in the references section.

The study of the history of the English language has a long and rich tradition, starting with a range of editions of important Old and Middle English texts in the middle of the 19th century, many of which are still available as reprints from the early English Text Society. The linguistic study of the history of English took off in the 20th century with a range of traditional grammars usually concerned with the phonology and morphology of Old and Middle English and a further range of detailed studies of the language of particular texts and of particular dialects areas. Since the 1970s and in the wake of the development of functionalist and formalist models of language structure, language use, and diachronic change, the various historical stages of English and the diachronic changes in the domains of phonology, morphology, syntax, and pragmatics have also become objects of intense study and debate for historical linguists. The study of these aspects has been greatly enhanced by the recent boost of computerized corpora, including text corpora as well as corpora enriched with various types of linguistic information. The history of English in all its breadth has thus become a field of study that draws both on rich documentation and on linguistic and methodological sophistication.

We will now briefly highlight some of the current debates in the field. The most recent one is on the extent of Scandinavian influence on the development of English. Most of the evidence we have on this is indirect, as noted in the main article, and so there is considerable room for interpretation and even speculation. Clearly, the Scandinavians settled in the northeast of England following initial hostilities, and language contact may well have been close and daily. Townend (2002) argues that Old Norse and Old English (OE) were mutually intelligible, and that there was societal though not individual bilingualism in Anglo-Saxon times in the Danelaw area. Emonds and Faarlund (2014) present the radical (and much-criticized) scenario that Old Norse dominated English in the Danelaw and claim that the grammatical skeleton of present-day English descends directly from Old Norse. For critical discussion of this, see Walkden and Bech (2016); and a special issue of Language Dynamics and Change, 6.1, which is devoted in its entirety to this topic.

Work on sound change in Middle English has received a new boost from the appearance and online availability of A Linguistic Atlas of Early Middle English, 1150–1325, and the Corpus of Narrative Etymologies. These two resources allow meticulously detailed research into the dialectology of early Middle English, and promote work on the details of sound change in that period. Some recent literature emerging from this is: Laing and Lass (2014) on the development of Middle English she from Old English heo; Minkova’s (2014) account of the Great Vowel Shift; Alcorn (2015) on the innovation of an alternative plural pronoun; Stenbrenden (2016) on vowel shifts in Middle English.

A major issue in the current study of word order change is the relation between syntactic structure and information structure in word order change: The latter concerns the ordering of information within the clause according the “given before new” principle, which brings issues of language use to bear on word order, beside structural analysis. Recent work that contributes to our insight into the loss of OV word order and the loss of Verb Second from this perspective can be found in the section on Interfaces of Traugott and Nevalainen (2012), Meurman-Solin, Lopez Couso, and Los (2012), and Bech and Eide (2014).

Further Reading

The references mentioned here are briefly discussed in the article “History of the English Language” in Oxford Bibliographies Online.

Crystal, D. (1995). The Cambridge encyclopedia of the English language. Cambridge, U.K.: Cambridge University Press.Find this resource:

    Hogg, R. M. (Ed.). (1992–2001). The Cambridge history of the English language. 6 vols. Cambridge, U.K.: Cambridge University Press.Find this resource:

      Hogg, R. M., & Denison, D. (Eds.). (2006). A history of the English language. Cambridge, U.K.: Cambridge University Press.Find this resource:

        Jucker, A. H., & Taavitsainen, I. (2013). English historical pragmatics. Edinburgh Textbooks on the English Language—Advanced. Edinburgh: Edinburgh University Press.Find this resource:

          Kay, C., & Allan, K. (2015). English historical semantics. Edinburgh Textbooks on the English Language—Advanced. Edinburgh: Edinburgh University Press.Find this resource:

            Kytö, M., & Pahta, P. (2016). The Cambridge handbook of English historical linguistics. Cambridge, U.K.: Cambridge University Press.Find this resource:

              Los, B. (2015). A historical syntax of English. Edinburgh Textbooks on the English Language—Advanced. Edinburgh: Edinburgh University Press.Find this resource:

                Mair, C. (2006). Twentieth-century English: History, variation and standardization. Cambridge, U.K.: Cambridge University Press.Find this resource:

                  McColl Millar, R. (2012). English historical sociolinguistics. Edinburgh Textbooks on the English Language—Advanced. Edinburgh: Edinburgh University Press.Find this resource:

                    Minkova, D. (2014). A historical phonology of English. Edinburgh Textbooks on the English Language—Advanced. Edinburgh: Edinburgh University Press.Find this resource:

                      Nevalainen, T. (2006). An introduction to Early Modern English. Edinburgh: Edinburgh University Press.Find this resource:

                        Ringe, D., & Taylor, A. (2014). The development of Old English. A linguistic history of English: Volume II]. Oxford: Oxford University Press.Find this resource:

                          Tieken-Boon van Ostade, I. (2009). An introduction to late modern English. Edinburgh: Edinburgh University Press.Find this resource:

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

                              References

                              Alcorn, R. (2015). Pronoun innovation in Middle English. Folia Linguistica Historica, 36(1), 1–17.Find this resource:

                                Bech, K., & Eide, K.-G. (Eds.). (2014). Information structure and syntactic change in Germanic and Romance languages. [Linguistik Aktuell 213]. Amsterdam: John Benjamins.Find this resource:

                                  Cole, M. (forthcoming). A native origin for they, their, them: Tracing the linguistic path of Old English þa, þara, þam from Old Northumbrian to northern Middle English.Find this resource:

                                    Emonds, J. E., & Faarlund, J. T. (2014). English: the Language of the Vikings. Olomouc Modern Language Monographs, Palacký University.Find this resource:

                                      Kiparsky, P. (1995). Indo-European origins of Germanic syntax. In A. Battye & I. Roberts (Eds.), Clause structure and language change (pp. 140–169). Oxford: Oxford University Press.Find this resource:

                                        Laing, M., & Lass, R. (2014). On Middle English she, sho: A refurbished narrative. Folia Linguistica Historica, 35(1), 201–240.Find this resource:

                                          Meurmann-Solin, A., Lopez-Couso, M. J., & Los, B. (eds.). (2012). Information Structure and Syntactic Change. Oxford Studies in the History of English 1. New York: Oxford University Press.Find this resource:

                                            Stenbrenden, G. (2016). Long-Vowel Shifts in English, c. 1050–1700. Studies in English Language. Cambridge, UK: Cambridge University Press.Find this resource:

                                              Townend, M. (2002). Language and History in Viking Age England: Linguistic Relations between Speakers of Old Norse and Old English. Turnhout: Brepols Publishers.Find this resource:

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