3.1 Spelling

If by a standard language we understand a variety with minimal variation in form and maximal variation in use, spelling is a much better candidate for Early Modern Standard English than speech.² Spelling regularization can be approached as a process of focusing that had started around the mid-14th century, when English gained new functions as a written language. Both professional copyists and printers were instrumental in spreading southern forms and spelling conventions to the rest of the country. But the process was not straightforward: arising from the concerns of the teaching profession, the question of a spelling reform sparked heated debates in the 16th century. Those in favor of a reform argued that English spelling had become too far removed from the pronunciation of the language and that a closer connection should be (re)established. Those supporting established usage countered them by arguing that there was too much variation in speech to form a basis for a renewed orthography.

New means of regularizing established spelling conventions were introduced to show the “right writing” of words, notably word lists appended to teaching manuals such as Richard Mulcaster’s Elementarie (1582). When analyzing this list of ca. 8,500 common English words, Barber (1997, p. 86) found that over half of them are identical with their modern standard spelling, and considerably more if the regular <i>/<j> and <u>/<v> alternations (see below) are excluded from the calculation.

Teaching manuals and spelling books made an important contribution to fixing English spelling norms in the 17th century. One worth mentioning is Edmund Coote’s The English Schoole-maister (1596), which offers not only syllable and word lists displaying spelling rules but also a 1,400-word dictionary. Scragg (1974, pp. 78–80) attributes part of the success of Coote’s manual—it went through some fifty editions in the 17th century—to his method of selecting for each word the spelling that was most common in print. Coote promoted, for example, spellings such as seem and seen instead of seme and sene, proposed by Mulcaster (1582)

In the course of the 17th century, the <u>/<v> alternation (<v> being used word-initially and <u> word-medially for both the vowel and the consonant) was also regularized to its modern standard form. Figure 1 shows the crossing over from the previously regular word-medial <u> to <v> in have in the EEBO-TCP database between the 1620s and 1640s. A similar process took place around the same time with the former use of <i> for both <i> and <j>.

Open in new tab

Figure 1. Regularization of the <u>/<v> alternation in 17th-century print: <haue>/<have>. (http://earlyprint.wustl.edu/eebotcpngrambrowser.html)

These processes apply to spelling regularization in print, which displays remarkable stability by the mid-17th century. The same cannot be said about private spelling habits, which show much more variation depending on the writer’s level of education. Moreover, punctuation and capitalization are standardized later than spelling in print as well. For example, the apostrophe only begins to be used to signal the possessive -s in the 17th century, earlier in the singular than in the plural, but it can mark the elision of the vowel in regular past participle forms (e.g., us’d for used) well into the Late Modern English period.

Looking at the big picture, Baron, Rayson and Archer (2009) calculated the relative frequencies of spelling variants over time in several corpora and databases, covering the period from 1400 to 1800 and including the Shakespeare Corpus and the EEBO database. On average, their data showed variation in some 50% of word types towards the end of the 16th century, and at the end of the 17th century variation was still found in about 20% of word types but in less than 10% of word tokens.³

3.2 Vocabulary

Automatic studies of spelling variation do not necessarily take into account the growth of English vocabulary over time, which vastly increases the number of rare words in the language in the 16th and 17th centuries. Vocabulary growth represents the widening range of domains of the vernacular, thus showcasing the tendency of a standardizing language to maximal variation in use. Lancashire (2012, p. 637) writes: “[t]he mother tongue remained small, well under 10,000 words, until the 17th century. Printed books, however, by saving and disseminating learned and technical words, expanded available vocabulary by 75% from 1500 to 1600.”

Various aspects of this growth have been explored using the Oxford English Dictionary. Work on the new edition of this lexical resource (OED3) suggests that twice as many words were first recorded between 1550 and 1699 as in the previous 150 years from 1400 to 1549 (Durkin & Allan, 2016, p. 72). Dividing the EModE period into fifty-year subperiods, Durkin (2014, p. 34) shows that the frequency of new words peaks in the first half of the 17th century. Figure 2 presents a very similar overview based on the Chronological English Dictionary (CED; Wermser, 1976, p. 23; see Nevalainen, 1999, pp. 336–339). Lancashire (2012, p. 641) predicts that the process of antedating OED headwords “is spreading vocabulary growth out more evenly through the Tudor period and may diminish the ‘acceleration’ effect” found in the 16th century.⁴

Open in new tab

Figure 2. Lexical growth in English 1500–1700.

Lexical innovations reflect the ongoing processes of modernization and vernacularization in the EModE period. The vernacular gained currency in a wide variety of specialist fields and replaced Latin and French, for example, in many legal, religious, and scientific domains. The growing range of specializations of contemporary knowledge is recorded in monolingual dictionaries and glossaries, which cover fields as diverse as alchemy and architecture, classics and cosmography, fencing and heraldry, hunting and falconry, mathematics and minerals, philosophy and poetics, and weights and measures (Nevalainen, 1999, pp. 342–347).

As English was spreading to a widening range of discourses, including the most prestigious ones, the “insufficiency” of the mother tongue became a common cause of complaint: English was felt to lack the prestige of Latin or French as a language of learning and literature. As a result, several often competing methods of vocabulary expansion were adopted, including both borrowing and coinages based on native word formation processes (Nevalainen, 1999, pp. 336–342). These means were used to introduce new lexis to meet the needs of vernacularization but also to consciously enrich and embellish the language. Deliberate learned loans also met with opposition and were branded as “inkhorn” terms (with reference to the common inkwell, originally made of a horn).

One of the early borrowers of the utilitarian kind was Sir Thomas Elyot, a scholar and diplomat. Most of the loans he used come from the classical languages, and include a number of words that are still current, such as encyclopedia, frugality, metamorphosis, modesty, and persist (Barber, 1976, p. 79). He was also among the first to use education in print (ex. 4):

(4)

Reflecting its cultural dominance at the time, Latin was indeed the primary source of borrowed lexis in EModE (Durkin, 2014, pp. 255–269). Beginning with Robert Cawdey’s A Table Alphabeticall (1604), specialist vocabulary started to be coded in “hard-word” dictionaries in the 17th century for the benefit of the vast majority of the population who would not understand words borrowed from “Hebrewe, Greeke, Latine, or French,” as Cawdrey pointed out in his preface. In The English Dictionarie (1623), Henry Cockeram also introduced a large number of new words into English by anglicizing Latin entries in Thomas Thomas’s Latin–English dictionary.

Borrowing often resulted in native and borrowed lexemes co-occurring in the same lexical sets. Loan words typically appeared in technical uses, as in the following word pairs, where the first member has its roots in Old English and the second in Latin or French (the fields and dates come from the EEBO database): belly/ventral (fiction 1693, medicine 1695), book/bibliography (religious miscellany 1664), egg/ovum (biology 1672), naked/nudity (zoology 1607), saying/dictum (poetry 1586), and sun/heliocentric (travel, 1667). It is noteworthy that loan words did not necessarily have the same meaning as they have today. Bibliography, for example, was first introduced in the sense of ‘the writing of books’ (Nevalainen, 1999, pp. 362–364).

The case of ovum nicely illustrates the prevalence of Latin in the English print culture at the time; one of its first occurrences in the EEBO database goes back to a dietary guide from 1599, where it simply provides the Latin term for egg in a header. Well into the 17th century, the word appears in Latin passages or collocates such as malum ovum (‘bad egg’) in a wide variety of sources ranging from treatises on embryology, natural history, and theology to Latin–English dictionaries and textbooks. It appears as an integrated specialist term in English in the latter half of the century, the OED dating this technical use in Philosophical Transactions to 1672.

The fact that specialist terms are well documented does not mean that they have survived until the present. Analyzing The Shorter Oxford English Dictionary, Neuhaus (1971) found that between 1460 and 1620 many more new words were introduced than old ones lost. By contrast, between 1640 and 1680, words introduced after 1530 showed a higher than average rate of disappearance. This comparison suggests that the intensive period of vocabulary enrichment was followed by a corresponding increase in obsolete words. Most of them disappeared during their first decade, and many were cited only once. It would appear that they had never made it to wider use. This was true of one-third of the Latinate neologisms attributed to Shakespeare and of a large number of coinages introduced by lexicographers like Cockeram (1623), who endeavored to enrich the EModE lexicon with parallel formations (Nevalainen, 1999, pp. 340–341, 349).

Measured in terms of the intake of new words, lexical standardization remains an abstraction. It does not take place in a cultural vacuum but constitutes an ongoing process with both gains and losses, and takes its time to be integrated into actual use. Importantly, the expanding vocabulary of EModE did not become the property of the population at large; as Lancashire (2012, p. 643) notes, it rather shows the growth of an increasing variety of occupational sociolects.

4.1 Pronunciation

If vocabulary growth did not directly affect all English speakers in the EModE era, changes in pronunciation and grammar did. As noted above, pronunciation can only be reconstructed from written evidence before it was integrated into monolingual English dictionaries using a systematic phonetic notation in the 18th and 19th centuries. Although there is a fair amount of commentary provided by early phoneticians and orthoepists, it is far from unproblematic. Lass (1999, p. 59) notes that we have to discriminate between

those who understood the difference between sound and spelling and those who didn’t, those whose normative biases led them to propose purely ‘theoretical’ and non-existent pronunciations and more objective observers, etc.

Despite these problems of primary evidence, there are comprehensive studies, such as Dobson (1968), which derive their evidence from contemporary EModE grammarians, phoneticians, and spelling reformers.

Scholarly opinion agrees that while some major changes happened in the EModE vowel system, the consonant system remained comparatively stable.⁵ In all varieties, /r/ was pronounced after the vowel (car, door). There is, however, occasional evidence for /r/-deletion in private writings (e.g., skasely ‘scarcely’) and some phonetic weakening of /r/ in the 17th century. By contrast, /h/-dropping was common word-initially in weakly stressed words such as pronouns (e.g., hit > it). It was commonly dropped in French loan words (history, honor) but often reintroduced into these words on etymological grounds. Dialectally, /h/ could also be dropped from words of Germanic origin, such as hand and heart. Chronicling public events in 16th-century London, Henry Machyn, a merchant tailor, spells hard without an <h> but inserts one into playing so that it becomes plahyng (ex. 5; Nevalainen, 2006, p. 127). Shakespeare and other dramatists often drew on /h/-dropping and /h/-insertion for puns (air, hair, heir).

(5)

Just like today, unstressed syllables could be dropped in connected speech and grammatical words could assume their contracted forms (they are > they’re). But preceding an auxiliary, the pronoun it was clipped (it is > ’tis), not the verb as today (it’s).

The most dramatic changes took place in stressed syllables, as long vowels were redistributed in a series of processes which resulted in what is known as the Great Vowel Shift (GVS). The process was probably set in motion by the high vowels /iː/ and /uː/ in the late 15th century: the front vowel /iː/ came to acquire the quality of /ei/, and the back vowel /uː/ the quality of /ou/. The outcome was that words like /liːf/ (life) and /tiːm/ (time) were pronounced as /leif/ and /teim/, and /huːs/ (house) and /uːt/ (out) as /hous/ and /out/.

The rest of the long vowels systematically moved up one step in the vowel space: words with the high-mid front vowel /eː/ (meet, see) came to be pronounced with the high vowel /iː/, and those with the high-mid back vowel /oː/ (do, loose) with /uː/. Words pronounced with the low-mid front vowel /ɛ‎ː/ (meat, sea) in turn acquired the closer quality of /eː/, and those that had the back vowel /ɔː/ (boat, rose) were raised to /oː/. Finally, the low /aː/ vowel in words like make and name rose to low-mid /ε‎ː/.

It is noteworthy that the Great Vowel Shift was not completed in the northern dialects, where it affected the front vowels but not all back vowels. Long vowels also emerged from diphthongs (i.e., vowels with a changing quality). In almost all these changes, a Middle English diphthong was reduced to a pure long vowel. These processes resulted in vowel mergers, and many word pairs came to sound alike (see further Lass, 1999).

4.2 Grammar

Unlike pronunciation, grammar change can be observed in most genres and registers, from formal kinds of writing such as statutes to informal kinds such as comedy. Personal communication is available in correspondence and approximate records of speech in trial proceedings. The EModE section of the Helsinki Corpus of English Texts (HC) comprises fifteen different genres and has provided primary data for a large number of studies on EModE grammar. The Corpus of Early English Correspondence (CEEC) in turn enables systematic historical sociolinguistic research into the Early Modern period, and speech-based data is included in the Corpus of English Dialogues (CED).⁶

There are some broad trends that characterize many processes of grammatical change in EModE. One of the most conspicuous is greater transparency. This analytic trend is found in many of the new forms and constructions that emerge in competition with older, more synthetic expressions in different grammatical domains. At the same time, certain grammatical distinctions are lost altogether.⁷

There are both gains and losses in the EModE system of personal pronouns. Perhaps the most noticeable loss is the second-person singular pronoun thou, which was replaced by the plural form you. Originally the polite form of address, you is the common form found, for example, in personal correspondence throughout the period, and thou only occurs in letters written by mothers to their children and by husbands to their wives. Example (6) shows both of them occurring in the same letter. Based on dialogues ranging from court proceedings to comedy, Walker (2007) draws the conclusion that social superiors could address their inferiors with thou but inferiors replied to them with you.

(6)

The generalization of you for singular as well as plural addressees also meant a loss in the number contrast in second-person pronouns. The case contrast between the subject form ye and object form you was also lost, and you began to be used in both functions. This process was completed in data sources such as personal letters in the 16th century, but ye was retained dialectally much longer (for example, in northern regional varieties), as indeed was thou.

Two other changes are connected with the trend of making a distinction between personal and nonpersonal reference. The personal pronoun system was enriched by the introduction of the neuter possessive pronoun its at the end of the 16th century. It replaced the historical form his, which had been used for both masculine and neuter referents, as in (7).

(7)

The same principle is manifest in the restriction of the relative pronoun which to nonhuman referents. The older use is still found in example (4), above: the chylde of a gentylman, whiche is to haue auctoritie. In the course of the 16th and 17th centuries, it came to be replaced by who, which alternated with the older that, as it continues to do today.

Verb forms also underwent changes. Verbal ‑s gradually replaced ‑th in the third-person present singular indicative (e.g., it cometh > it comes) and are replaced be in the present indicative plural of the verb to be. These forms were both originally northern dialect features. The southern forms were retained dialectally and in formal written contexts much longer than in speech (cf. the powers that be in the Tyndale Bible, 1534). Verbal constructions were also established, providing grammatical means of expressing the progressive aspect (be + ‑ing) and indicating future time by means of the be + going to construction. They are illustrated in (8) and (9), respectively. The new expression of future time was added to earlier ones, such as the auxiliaries shall and will.

(8)

(9)

A number of developments were connected with the order of sentence elements. The subject–verb–object order, with the verb following the subject, was established as the basic word-order type in EModE declarative sentences. The inverted order was still found after adverbs like here, now, then, and thus (10), but also after longer sentence elements. Bækken (1998, p. 60) estimated that the rate of inversion fell from the average of one inverted declarative main clause in five in 1500 to one in ten in 1700.

(10)

A converse development took place with initial negative elements such as neither, never, and nor, which started to require inversion in declarative sentences. This process also supported the rise of the auxiliary do in negative contexts. Examples (11) and (12) show how an individual who used the inverted word order after neither changed from main-verb inversion (11) to auxiliary inversion by means of do (12). The presence of an auxiliary also prevents the inversion of the subject and the main verb in interrogative sentences, and maintains the regular verb–object order in negative sentences.

(11)

(12)

The introduction of do into the verb phrase makes it more transparent, as the auxiliary then carries features such as tense (present vs. past), mood (indicative vs. subjunctive), and polarity (affirmative vs. negative). In the absence of another auxiliary, the use of do was first generalized in questions and, in particular, negative questions in the course of the EModE period. The process took longer in negative statements, where do was used in about two-thirds of the cases with most verbs at the end of the 17th century.

Running counter to this trend, do failed to become a regular feature in affirmative statements. After picking up in the 16th century, this use declined sharply in the course of the 17th century. At the height of its popularity, it was common in both emphatic and nonemphatic uses and in formal and informal contexts, as in (13) and (14), while today it is used systematically only for emphasis and contrast in this sentence type. It has been suggested that the rapid decline of do in affirmative statements was due to dialect contact or that it may have been a manifestation of some more abstract grammatical processes of change, but no definitive answer has been proposed so far (Warner, 2012).

(13)

(14)

The grammar of negation also underwent changes. Like today, not was used to negate the sentence in the 16th century, but when associated with indefinite expressions, it was often accompanied by other negative forms (no, never) rather than their nonassertive counterparts (any, ever). Multiple negation occurred in the writings of educated and uneducated people alike. The examples below come from the corpus of personal letters and date to the 1540s. The passage in (15) was written by a merchant’s wife, and those in (16) and (17) by a bishop. Although the writers show different degrees of spelling variation, reflecting their differences in education, both of them use multiple and single negation side by side.

(15)

(16)

(17)

Open in new tab

Figure 3. Decline of multiple negation in simple and co-ordinate constructions (CEEC).

As shown by Figure 3, during the EModE period multiple negation as good as disappeared from simple contexts with indefinite pronouns like those in (15) to (17) in the personal correspondence of literate people. It was retained in coordinate constructions such as not. . . neither into the Late Modern English period. The use of multiple negation was also socially stratified: although it was generally on its way out, it disappeared from the language of the lower social ranks more slowly than from the higher and professional social strata (Nevalainen & Raumolin-Brunberg, 2017, pp. 71–72, 145–146). Like many other recessive features, multiple negation persisted regionally and was stigmatized by prescriptive grammarians in the 18th century.