Early Modern English religious prose – A conservative register?

Thomas Kohnen, Tanja Rütten & Ingvilt Marcoe
English Department, University of Cologne

Abstract

1. Introduction

Early Modern English religious prose is generally described as a conservative register, that is, a register resistant to language change and linguistic innovation (see, for example, Crystal & Davy 1969; Barber 1997; Nevalainen 2006: 135). A number of morpho-syntactic features which were typical of the evolving modern standard did not spread to Bible translations and liturgical texts. For example, the third person suffix -th, which was replaced by -s in the course of the sixteenth century, was preserved longer in religious texts such as the King James Bible from 1611 and the 1662-version of The Book of Common Prayer (Nevalainen & Tieken-Boon van Ostade 2006: 294). The ye/you distinction, which was also lost in the course of the sixteenth century, was still upheld in the King James Bible (Nevalainen 2006: 80). Moreover, the substitution of the second-person singular pronoun thou by you is not found in liturgical language (Barber 1997: 153-154). The impression one gets of religious prose is that of a generally conservative language variety, robust to change and rather traditional. However, an assessment of religious language other than that found in liturgical texts or the Bible is still lacking. Thus, it is not clear whether all religious genres were affected by this archaic tendency in the same way.

In our paper, we discuss in how far a uniform description as archaic can be applied to all religious genres and to what extent Early Modern religious prose may be considered conservative with regard to language change. Our investigation includes four features that reflect the development towards modern Standard English (thou vs. you, -th vs. -s, be vs. are and the which vs. which). These features are analysed in prayers, catechisms, sermons and religious biographies. The data is part of the Corpus of English Religious Prose (COERP), which is currently being compiled at the University of Cologne. [1]

The paper falls into four parts. After this introduction, section two gives a short overview of the background and data of our study. In the third section the results of the analysis of the four linguistic features are presented. The conclusion section contains a short discussion of the conservative character of religious prose, the usefulness of the four features as diagnostic tools for religious language and some further perspectives of this study.

2. Background and data

The choice of linguistic features for our analysis was based on the following considerations. Our initial intention was to start with a selection of four features from Nevalainen and Raumolin-Brunberg’s list of fourteen changes (2003) which are relevant for the emerging standard language in Early Modern English. This list offers a good point of comparison for the relative “conservatism” of religious language since these features (and their development) have been analysed in the Corpus of Early English Correspondence. A comparison of the results should allow us to determine whether letters, a genre usually considered informal and close to oral modes of communication, were more “advanced” in the development towards the standard language than religious writing. Two features which are often associated with religious language and a conservative style (the late replacement of thou by you and the indicative plural be by are) are discussed by Nevalainen and Raumolin-Brunberg (2003) but are not included in their analysis. [2] Since we thought that these two features were of major importance, we gave priority to them and only selected two more features from Nevalainen and Raumolin-Brunberg’s list (-th vs. -s and the which vs. which). Thus, our selection for this study includes features that are typical of religious discourse, that have, at least in part, been analysed in letters and that include the morphological, the syntactic, and also the pragmatic level.

The corpus of our analysis, which is part of the Corpus of English Religious Prose (COERP), consists of the four genres: prayers, catechisms, sermons, and religious biographies. These genres illustrate well the diversity of religious discourse since they reflect various communicative situations and constellations of discourse participants in the religious domain. [3] In prayers (members of) the Christian community address(es) God (or a saint); in sermons and catechisms, two genres typical of religious instruction, the Christian community is addressed by a (usually superior) fellow Christian; religious biographies, on the other hand, include narrative accounts of the lives of model Christians.

Table 1 gives an overview of the corpus used for our analysis. [4] The corpus consists of about 709,000 words in total, subdivided into four periods of fifty years, covering the sixteenth and seventeenth centuries. The three genres, catechisms, sermons and religious biographies each have ca 40,000 words per period, whereas prayers, as a separate sphere (see Kohnen 2010), have ca 70,000 words per period. The smaller number of words in period 1 in prayers is due to the fact that vernacular prayers are fairly rare during the early decades of the sixteenth century. The lack of data for catechisms in period 1 has similar reasons. Since the earliest catechisms were first published in 1549 and do not make up sufficient data to represent one sub-period by themselves, they were included in period 2. [5]

For the presentation of the results we decided to use proportions, that is, percentages, rather than normalised frequencies. [6] Clearly, proportional distributions are a better tool for showing the spread or decline of two competing forms, independently of their respective frequencies in the corpus (see also Nevalainen & Raumolin-Brunberg 2003: 68, 74 for a similar presentation). Proportions also offer a better measure of comparison across genres, since they are independent of differing high or low normalised frequencies in the four genres.

3. Analysis of linguistic features

3.1 thou vs. you

The persistent use of thou instead of you as the second-person singular pronoun is often mentioned as a typical feature of religious language and its conservative nature (see, for example, Barber 1997: 153-154).

Figure 1 shows the proportions of you-forms with a singular referent (as opposed to thou-forms) across the four religious genres. [7] The proportions suggest that this feature clearly discriminates between the four genres.

On the one hand we find prayers, which do not contain any singular you-form. All you-forms occur in the plural and refer to more than one addressee. Here, they are mostly used to address (several) saints or a congregation of Christians in blessings, or they are found in Bible quotations. The few instances where you is, in fact, used to refer to a singular referent are not found in the actual prayers but in the metatext, which, for example, gives instructions on how to make use of the prayer (see examples (1) and (2)).

(1)

A Prayer to be saide when you go to bed. I Thank thee, O heauenly father, by thy dearely beloued sonne Iesus Christ, our Lord and sauiour, that of thy free mercie thou hast preserued me.

(Anon., A prymmer or boke of priuate prayer, 1580)

(2)

PRAIERS AT NIGHT AND BED-TIME
Retire your self again; AND when you have thought upon all the day past, how you have spent it, that is, what good or evil Actions you have done, and what bad Inclinations you have resisted or amended, asking God pardon for all your offences, use these Praiers.

The Psalm and Praiers.
LEt my Praier be set forth in thy sight, O Lord, as the Incense; And the lifting up of my hands be as an Evening Sacrifice. In thee hath been my hope all the day long, and under the shadow of thy Wings shall be my refuge day and night for ever.

(Anne Douglas Morton, The Countess of Morton’s daily exercise, 1666)

Prayers form one end of a scale in the proportions of thou- and you-forms used with a singular referent. On the other end of this scale we find religious biographies, which show a rather high proportion of you-forms with a singular referent. In periods 1 and 3 they have more than 40 per cent and in periods 2 and 4 more than 50%. In between those two genres (prayers and religious biographies) there are catechisms and sermons, which seem to be rather flexible in their representation of you-forms. Catechisms show a variable, but generally smaller proportion of you-forms than religious biographies (except for period 3, where they exceed religious biographies by roughly 7%). Sermons, generally, have a very low proportion of you-forms used with a singular addressee. Due to the communicative setting of sermons, hardly any occasion is found where one single person is addressed directly by the preacher at all. The usual form of address is you directed at the whole congregation. [9]

In our analysis of the pronominal system of religious language we found it useful to distinguish primary and secondary items in order to reveal the different settings in which the address terms are used (cf. Kohnen 2000: 309). Primary items are those pronouns which directly address the recipient of the text. For example, in (3) the reader of the biography is addressed.

(3)

Thus hast thou heard (gentle Reader) the discourse of the vertuous life and Christian death of this blessed and faithfull servant of God, Mistris Katherine Stubs.

(Phillip Stubbes, A crystall glasse, 1591)

In catechisms, we supposed that the text is used in its performative function, that is, that the dialogue is actually performed between an instructor and a catechumen. Thus, the pronouns in example (4) count as primary items.

(4)

Master. How doest thou call bread thyne, which thou prayest to haue geuen thee of God?

(Alexander Nowell, A catechisme, 1570)

Secondary items are those pronouns which are used in reported or fictional interaction. In our corpus, secondary items are most common in Bible quotations (see example (5)) and prayers or invocations to God inserted in the texts.

(5)

Question. What is the ninth commaundement?
Answer. 9. Thou shalt not beare false witnesse, &c.

(Alexander Nowell, A catechisme, 1570)

As can be seen from the proportions of you-forms as opposed to thou-forms in Figure 1, thou-forms provide the large majority of pronominal singular address terms in all four genres. However, it turns out that many of these forms are secondary items, that is, they are not used to address the recipient of the text. Figure 2 shows the proportions of secondary items among all thou-forms found in sermons, biographies and catechisms. [10]

In all genres the proportions of secondary items are fairly high: in religious biographies they range between 86 and 100%, in catechisms between 40 and 88% and in sermons between 32 and 81%. Especially in catechisms and sermons, the proportions of thou used as a secondary item increase over time. This means that the primary use of thou (to refer directly to the text recipient) decreases, which reflects the general development of the pronoun system in Early Modern English.

In addition, it is noteworthy that an increasing share of all thou-forms occurs in Bible quotations or addresses to God (in invocations or short prayers) which are inserted into the text (see examples (6) and (7)).

(6)

When the children of GOD, that is, the good Angels came before the LORD, Sathan stoode amongst them, and the Lord sayd vnto him, Whence commest thou? and hee answered, From compassing the Earth too and fro: and the LORD sayde. Hast thou not considered my seruant Iob, how there is none like him in the earth, an vpright and iust man? Sathan answered: It is not for nothing that Iob feareth thee, hast thou not made a hedge about him and his house, and about euery thing which hee hath on euery side? but touch all that hee hath, and then see whether hee will not blaspheme thee to thy face.

(Arthur Dent, Christes Miracles Deliuered in a Sermon, 1608)

(7)

Lord, we may diligently follow the workes of our calling, and so continually receiue a blessing from thee thorough Iesus Christ:

(Thomas Playfere, A Sermon Preached at Winsor, 1604)

Figure 3 shows the proportions of such thou-forms found in Bible quotations and prayer sections in catechisms, sermons and religious biographies. It nicely illustrates that a large number of all thou-forms belong to that category. In period 3 the proportion ranges between 50 and 88%, and in period 4 between 64 and 75%. We can thus conclude that the archaic “thou-character“ of these three religious genres, especially in the seventeenth century, is based on the high proportion of Bible quotations and prayer sections contained in them and not on the other parts of these texts.

A closer look at the proportions of you used as a primary item is also interesting. Having excluded one rather idiosyncratic biography in period 3, [11] it is relatively clear that the proportions of you-forms used as primary items increase over time, while the proportions of thou-forms decrease. Figure 4 shows an increase in catechisms from 25 to 67% (88% in period 3), and in sermons from 1 to 21%. Religious biographies already show 100% of primary items in period 1, [12] but after a decrease in period 2 they follow a similar tendency of increase. However, one should bear in mind that the number of tokens in religious biographies is significantly smaller than in the other two genres. For example, in period 1 there are 21 tokens in religious biographies and 117 tokens in sermons; in period 2 we find nine tokens in religious biographies as opposed to 445 tokens in catechisms and 51 tokens in sermons. Thus the “dip” in Figure 4 should not be overinterpreted.

Figure 4 also shows the pattern observed in Figure 1. We find catechisms and sermons in-between prayers and religious biographies, as the two extremes in this development. [13]

In catechisms we noticed a further interesting development in the pronoun system. Figure 5 contains all occurrences of you and thou as primary items arranged by speakers. Here the total of all primary you-forms was split up into you used by the teacher and you used by the pupil. In the same way the thou-forms were subdivided. A clear stratification can be observed. While the teacher uses both you and thou to address his pupil, the pupil always returns you (see examples (8) and (9) below). [14]

(8)

Question. BY what meanes art thou receiued into the Church of God?
Ans. Outwardly by baptisme but inwardly by faith in Christ,
Quest. What benefits receiued you by baptisme?
Ans. Three: first I was made a member of Christ, Secondly, I was adopted the childe of God, Thirdly, I was made the heire of the kingdom of glorie.

(William Hill, The first principles of a Christian, 1616)

(9)

S(on). I beg your excuse that I cannot yet yield a ful assent to what you drive at, until you pleas to explain more particularly to me what is that very degree of perfection which you say our Natures ar capable of, that so I may examin whether in effect it be ever attained or no.
F(ather). Thou dost wel: and I wil endeavor to satisfy thee.

(William Popple, A rational catechism, 1687)

The distribution of you and thou thus follows the general trend observed in the development of the pronoun system in Early Modern English in that you and thou are clearly stratified according to speaker role.

The results of the analysis of the first feature can be summed up in three points. First, among the four genres under analysis only prayers can be called “typically religious” with regard to the use of thousince a basically primary use of thou is retained only in this genre. Religious biographies, in contrast, have the lowest proportion of thou, with catechisms and sermons situated in the middle. Second, the archaic nature of the non-prayer genres is probably in part caused by the large share of thou in Bible quotations, inserted prayers and invocations to God. Biblical language and prayer language thus seem to add an additional layer to the language used in the other religious genres. Third, catechisms, sermons and religious biographies follow the general trend of the development of the English language, though at a slower pace and to a different extent.

3.2 -th vs. -s

The second feature to be discussed is the replacement of the third-person singular present indicative form -th by -s. Forms like showeth or withereth are often seen as typical examples of religious language, and Crystal and Davy (1969) mention -th as one of the inveterate properties of the religious register. In our analysis, have and do have been excluded due to their longer retention of the th-suffix (see also Nevalainen & Raumolin-Brunberg 2003: 67). In addition, the few s-forms occurring in verse were not counted either.

Table 2 gives the proportions of s-forms in the data. The general development in the sixteenth century shows extremely small shares of s-forms in all religious genres, the largest proportion in period 2 being 3.2% in prayers. The proportions are clearly smaller than those found by Nevalainen and Raumolin-Brunberg (2003) for letters, which rise from 6.8% to 23.6%.

In the seventeenth century we witness a sudden rise in the use of the s-suffix in religious biographies (39.3 and 77.3%) and sermons (28 and 88.7%). This rise does not happen until period 4 in prayers (35%) and catechisms (48.2%), with prayers clearly showing the smallest proportion. Apart from sermons in period 4, all religious genres have clearly smaller proportions of s-forms than the letters in the seventeenth century.

Interestingly, it is prayers which first show s-forms. But s-forms do not seem to have spread in the genre evenly. Until the middle of the seventeenth century the large majority of all items occur in single collections. In period 1 the three s-forms stem from two texts, Devout Prayers and York Bidding Prayers. [16] In period 2 all ten s-forms occur in one text, The Fifth Lampe of Virginity, and even in period 3, twelve out of thirteen s-forms stem from one collection, The Crums of Comfort. This distribution of the data suggests that the s-form did not affect the whole genre. Thus, prayers as a whole may still be described as fairly conservative.

Looking at the instances of th-forms in catechisms, sermons and religious biographies in period 4 in more detail, it can be noted that th-forms are frequently found in Bible quotations and prayers that are inserted in the texts (see examples (10) and (11). [17] In religious biographies, for instance, th-forms in Bible and prayer sections make up 50% and in sermons 43.1% of all th-forms.

(10)

Consider of that place, To me says God, belongeth Vengeance and Recompence; their foot shall slide in due time, for the Day of their Calamity is at hand, and the thing that shall come upon thee maketh hast.

(Thomas Doolittle, A call to delaying sinners, 1698)

(11)

How is it, Lord, that thou shouldst manifest thy self unto me, and not unto others, even so Father, because it seemeth good in thy eyes, Thou wilt have mercy because thou wilt have mercy.

(James Janeway, Invisibles, realities, 1673)

A large proportion of th-forms retained in other than Bible or prayer contexts may be explained by phonological constraints. Nevalainen and Raumolin-Brunberg (2000a: 235; 2003: 67) point out that the replacement of -s by -th can be related to a long-term morpho-phonemic change in English, the deletion of the preconsonantal vowel /e/ in inflectional endings except after stem-final sibilants. In period 4 in our data, non-syncopated eth-forms are accordingly retained longer after stem-final sibilants, as can be seen in example (12), where s- and th-endings occur side by side.

(12)

No sooner do Men place Religion where God placeth none, but Sathan obtains a great Power and Influence over them, and at last hurries them headlong into strange and wild Extravagancies in the Defence, and for the Propagation of those Points.

(Samuel Bold, A sermon against persecution, 1682)

In addition, th-forms are preserved longer in our data in the fixed expression as followeth (see example (13)) and in combination with the relatively frequent verb say (see also Kytö 1993: 121).

(13)

… we had (as a reall Testimony of her Fathers thankfulnesse to God, and reall respects to us) the joyfull news of her marvellous Recovery transmitted to us, which take as followeth, …

(James Fisher, The wise virgin, 1653)

Generally, we may conclude that the spread of -s forms in the religious genres does not differ significantly from other genres. Although the spread may happen later and more slowly, especially in prayers and catechisms, [18] none of the four genres under analysis preserves th-forms exclusively. This seems to apply only to Biblical writing. In addition, the order of genres observed with the first feature seems to be confirmed here: prayers are most reluctant to adopt the new feature, whereas religious biographies show large proportions already in the early seventeenth century. Sermons and catechisms are found in-between.

3.3 be vs. are

The northern variant are of the present plural indicative started to replace the southern form be in the course of the sixteenth century (Nevalainen 2000: 342). A prolonged retention of the archaic form be in religious contexts could be taken as an indicator of a conservative nature of religious genres.

In our analysis, all instances of be which could be subjunctive forms have been excluded. These comprise instances in optative, conditional, concessive, exceptive and temporal clauses, as well as clauses complementing verbs of wishing, commanding, doubting and hoping, and generalised relative clauses (see also Nevalainen 2000: 348).

Table 3 shows the distribution of the plural indicative are in our data. In the first half of the sixteenth century, the proportion of are already exceeds the one of be in prayers (63.7%), religious biographies (62.5%), and sermons (52.2%), whereas the letters clearly lag behind in period 1 (30.4%). At first sight, this delay of letters is rather unexpected since the replacement of be by are has been described as a change from below, spreading via everyday spoken interaction, rather than originating from above, that is, from institutional centres such as the church, the chancery, and the printing press (Nevalainen & Tieken-Boon van Ostade 2006: 294). [20]

However, there are at least three factors which may make the early rise of are in the three religious genres plausible. First, there seems to be a link between the occurrence of are and oral delivery. Nevalainen (1987) and Nevalainen and Tieken-Boon van Ostade (2006: 294) point out that are predominates in The Book of Common Prayer, especially in those sections which were intended for public delivery, and in the King James Bible, another book which was meant to be read out. Against this background, the early spread of are in prayers and sermons, both of which were designed for oral delivery too, seems hardly surprising. Also, long stretches of direct speech can be found in some of the early religious biographies (for example, The first examinacyon of Anne Askewe and A brefe chronycle concernynge the examinacyon and death of […] syr Iohan Oldecastell). Thus, the high percentage of are in the religious genres might well seem plausible.

Secondly, a look at the distribution of are across the social ranks in personal letters is quite instructive. Nevalainen (1996: 67) points out that in the sixteenth century the incoming form are was most advanced among the upper gentry (63%) and the upper clergy (50%). The higher percentage of are-forms in sixteenth-century religious genres may therefore reflect a predominance of writers from the upper ranks, especially the upper clergy, such as the bishops Richard Fitzjames, John Fisher, Cuthbert Tunstall and Hugh Latimer, but also the upper gentry, as for example Sir Thomas More. Thirdly, be might have been retained longer in private letters, next to the incoming variant are, because private writing is slower to adopt features of the evolving standard variety and therefore tolerates more non-standard variation (Nevalainen & Raumolin-Brunberg 1989: 97). [21]

Table 3 shows that in the first half of the seventeenth century are-forms increase steadily in all genres and outnumber be-forms with proportions between 86.9% and 95.2%. Be-forms preserved in the seventeenth-century texts prevail particularly in fixed expressions, for example, in existential there-clauses (see example (14)) and, in catechisms, in questions following a formulaic pattern (see example (15); note here, in particular, the are-form in the child’s answer).

(14)

There be some things oft repeated; yet because either they are uttered with some variety, and were things that it seems her heart was much carried forth to mention; therefore I have set them down …

(James Fisher, The wise virgin, 1653)

(15)

Father. Which be the proper and essentiall attributes of God?
Child. The proper attributes of God are vbiquitie, eternitie, vnitie, simplicitie, omnipotency…

(Arthur Dent, Pastime for Parents, 1606)

Generally, we may conclude that are-forms predominate over be-forms in all four religious genres, with personal letters at first lagging behind. Therefore, it is fairly unlikely that this feature should be a property which is particularly characteristic of religious genres and their supposedly archaic nature. Rather, the religious genres seem to follow the general development of the language here as well.

3.4 which vs. the which

The fourth change to be discussed is the generalisation of the relative pronoun which over the which. One might expect to find the more archaic variant the which retained longer in religious than in non-religious texts. However, the data in Table 4 show that the use of the which is not at all characteristic of religious prose.

Already in period 1 we find high proportions of which, that is, 79.7% in sermons, 96.5% in prayers and 97.7% in religious biographies. By the end of the seventeenth century, the relative pronoun which occurs with proportions between 97.4% in catechisms and 100% in sermons. This development is in line with the development found in letters, where the relative pronoun which is also well established in the early sixteenth century (with a proportion of 90%) and further increases to 99 per cent in the second half of the seventeenth century.

When taking a closer look at the data, Raumolin-Brunberg’s (2000) findings about the distribution of which and the which can largely be confirmed. Raumolin-Brunberg points to the fairly random distribution of the two variants with regard to various extralinguistic variables and argues for their status as free variants (2000: 221-222). However, she mentions a trend towards a grammatical specialisation of the which, as this variant comes to be restricted to prepositional phrases in the course of the sixteenth century. This trend may also be observed in our data (see examples (16) and (17)).

(16)

…if you refourme fyrst your lyfe to the rules of the Canon lawes / than shall ye gyue vs lyghte (in the whiche we maye se what is to be done of our part)

(John Colet, The sermon […] to the conuocacion at Paulis, 1531)

(17)

O how glorious is that kingdom in the which all Saints rejoyce with Christ!

(Church of England, Primer, or Office, 1658)

Moreover, Raumolin-Brunberg describes an interesting regional diffusion: the which, which is originally of northern origin, spread to London in the latter half of the fifteenth century, where it was favoured particularly by one group of people, namely London wool merchants. Despite this tendency, the which was not generalised and came to be replaced by which in the course of the sixteenth century (Raumolin-Brunberg 2000; see also Nevalainen & Raumolin-Brunberg 2000b). This regional preference for the which in northern England and London is also reflected in our sixteenth-century data. For example, in period 1 the highest proportion of the which occurs in sermons by preachers from London, such as John Colet (80%) and William Peryn (50%), as well as in sermons by John Fisher (17.7%), who was born in Yorkshire and lived in London throughout most of his lifetime. Interestingly, Cuthbert Tunstall, Bishop of Durham, used the southern form which exclusively. This shows that regional distribution alone cannot account for the variation between which and the which. In period 2, the highest percentage of the which is found in John Whitgift’s sermon (18.8%). The fact that the Archbishop of Canterbury was actually born in North-East Lincolnshire may explain his prolonged use of the which. In period 2, no instances of the which are found in sermons by preachers with no direct connection to northern England or London, such as Roger Edgeworth, John Jewel and Henry Smith.

In sum, considering the overall high percentage of which in all four religious genres, it is safe to say that with regard to this feature religious language cannot be considered more conservative than other non-religious genres. Here too, the four genres seem to follow the general development of the language.

4. Conclusions

The aim of this study was to get a first answer to the question to what extent Early Modern English religious prose was a conservative register and whether a uniform description as archaic could be applied to all religious genres to the same extent. The analysis of the four morpho-syntactic features which reflect the development towards modern Standard English suggests that most religious genres have followed the general development of the language, though sometimes later and to a lesser extent. Thus, one could say that religious language as a whole is only “more or less” conservative.

The comparison with the analysis of the Corpus of Early English Correspondence has also revealed that only few features are clearly diagnostic of religious language (for example, thou and -th), whereas others are obviously neutral (for example, the which vs. which). It seems that only few religious genres have preserved these archaic diagnostic features exclusively (Bible translations and prayers). As a consequence, the archaic character of the other religious genres stems from the Bible and prayer sections contained in them, not from other, independent features that may belong to the genres themselves.

Thus, religious language does not form a solid archaic block but a continuum, with key genres and more peripheral genres. The diagnostic features studied seem to arrange the four genres in a relatively stable pattern, with prayers and religious biographies forming the end points.

These results certainly raise doubts about the established view which sees religious language in toto as a conservative, archaic and even unintelligible register (for example, Crystal & Davy 1969). On the other hand, they confirm the special status of Biblical language and prayer, which seem to be responsible for the “religious colouring” of the other genres in the domain. [23]

Seen from a more general point of view, this investigation has also shown the need for more comparative data. Apart from the analyses based on the Corpus of Early English Correspondence, there are hardly any corpus-based studies which would offer hard and fast data on the distribution and spread of the morpho-syntactic features which reflect the development towards the standard. The position of individual features in this process and their stylistic associations are often fairly difficult to determine (see, for example, the discussion of be vs. are). Thus, the present results should be corroborated by further investigations which should focus on more features and should also include more genres, both from the religious and from other secular domains. This would allow us not only to locate religious language within the context of Early Modern English but also to trace morpho-syntactic standardisation across different Early Modern English domains and genres.

Notes

[1] See http://www.helsinki.fi/varieng/CoRD/corpora/COERP/index.html.

[2] Be vs. are is, however, discussed in Nevalainen (2000). But the numerical data offered there only reaches until 1639.

[3] On a pragmatic model for the analysis of the religious domain see Kohnen (2010).

[4] We would like to thank Kirsten Gather for her help with the tables and figures and for valuable discussion.

[5] On the compilation guidelines for the Corpus of English Religious Prose see http://www.helsinki.fi/varieng/CoRD/corpora/COERP/guidelines.html.

[6] Main raw frequencies for all figures and tables can be found in the appendix.

[7] The decision if you-forms addressing the reader in religious biographies were considered singular or plural was based on the immediate context. If a single reader was addressed explicitly, either in the preface or elsewhere throughout the text, we interpreted all you-forms as singular forms referring to this particular reader. If no single reader was specified in the text, all you-forms were interpreted as plural forms. This decision is based on the fact that reading in the sixteenth and seventeenth centuries was largely a public undertaking, and usually involved the person reading (aloud) and a number of listeners (see Collinson, Hunt & Walsham 2002: 61-66 for details).

[8] In this and the following presentation of the data on thou vs. you, the sermon by Thomas Doolittle (A call to delaying sinners, ca. 3,800 words) was excluded. This sermon contains an exceptionally high number of thou-forms, in particular primary items (102 out of 117, or 87 per cent in period 4). These primary items are addresses to a “sinner”, who, on the one hand, is clearly present in the audience, on the other, obviously fictional, since the whole audience would not normally be addressed with a singular address term.

[9] It may be argued that in the address to a single member from the congregation, a singular thou-form is less ambiguous than a singular you-form, which might refer to a single member as well as the whole congregation.

[10] Here it should be recalled that in prayers thou does not occur as a secondary item since it is always used in the direct address to God.

[11] Arthur Harris’s Life of Arthur Lake (1629) contains nine of the eleven primary items found in period 3, a frequency which is extremely unusual for a single text in the seventeenth century.

[12] This is due to 21 primary items of you in two texts (John Bale, A brefe chronycle concernynge the examinacyon and death of … syr Iohan Oldecastell (1544) and John Bale and Anne Askew, The first examinacyon of Anne Askewe (1544)), where primary you is only used by John Bale.

[13] The only exception is found in period 3, where the proportion in catechisms exceeds that of religious biographies by 1%.

[14] The small number of thou-forms in periods 2 and 4 used by the pupil occur in loose paraphrases of the Bible; they may actually be considered secondary items.

[15] The proportions for letters in this table are based on the numerical data given in Nevalainen and Raumolin-Brunberg (2003: 220).

[16] In this text the occurrence of the s-form is probably due to its northern origin (see also Nevalainen & Raumolin-Brunberg 2000a: 236).

[17] Note in example (10) that says God does not belong to the Bible quotation.

[18] How late exactly – in comparison to other genres – religious genres are in adopting the s-form should be examined in more detail. In Kytö’s (1993) study of s-forms in six genres of the Helsinki Corpus only private letters (79 per cent), official letters (28 per cent) and trials (28 per cent) show significant proportions of s-forms in the period between 1570 and 1640 (compare period 3 in our data). Thus, sermons and biographies may, in fact, not be so late. But a disadvantage of Kytö’s study is the relatively low number of tokens (except for letters and sermons).

[19] The proportions for letters in this table are based on the numerical data given in Nevalainen (2000). Since this data only reaches until 1639, no proportion is given for period 4.

[20] Change from above and below may refer to the levels of social awareness as well as the origins of diffusion in terms of socio-economic hierarchy (see Nevalainen 1996: 15; Labov 1994: 78).

[21] Here it should also be recalled that not all groups of letter-writers had adopted the are-form to the same extent. The class of the upwardly mobile, for example, had a very low share (7 per cent) of are-forms (Nevalainen 1996: 67).

[22] The proportions for letters in this table are based on the numerical data given in Nevalainen and Raumolin-Brunberg (2003: 222).

[23] On the special status of religious language, in particular Biblical language and prayers, in news discourse see Kohnen (2009).

Electronic references

Corpus of Early English Correspondence (CEEC). 31st March 2011. http://www.helsinki.fi/varieng/domains/CEEC.html.

The Helsinki Corpus of English Texts. 31st March 2011. http://www.helsinki.fi/varieng/CoRD/corpora/HelsinkiCorpus/index.html.

References

Barber, Charles. 1997. Early Modern English. Second edition. Edinburgh: Edinburgh University Press.

Collinson, Patrick, Arnold Hunt & Alexandra Walsham. 2002. “Religious publishing in England 1557–1640”. The Cambridge History of the Book in Britain, Vol IV, 1557–1695, ed. by John Barnard & D. F. McKenzie, 29-66. Cambridge: Cambridge University Press.

Crystal, David & Derek Davy. 1969. Investigating English Style. London: Longman.

Kohnen, Thomas. 2000. “Explicit performatives in Old English: A corpus-based study of directives”. Journal of Historical Pragmatics 1(2): 301-321.

Kohnen, Thomas. 2009. “Religious language in early English newspapers?” Early Modern English News Discourse. Newspapers, pamphlets and scientific news discourse, ed. by Andreas H. Jucker, 73-89. Amsterdam: John Benjamins.

Kohnen, Thomas. 2010. “Religious discourse”. Historical Pragmatics (= Handbooks of Pragmatics 8), ed. by Andreas H. Jucker & Irma Taavitsainen, 523-547. Berlin: Mouton de Gruyter.

Kytö, Merja. 1993. “Third-person present singular verb inflection in early British and American English”. Language Variation and Change 5: 113-139.

Labov, William. 1994. Principles of Linguistic Change, Vol. 1. Oxford: Oxford University Press.

Nevalainen, Terttu. 1987. “Change from above: A morphosyntactic comparison of two Early Modern English editions of the Book of common prayer”. Neophilologica Fennica, ed. by Leena Kahlas-Tarkka, 295-315. Helsinki: Société Néophilologique.

Nevalainen, Terttu. 1996. “‘Introduction’ and ‘Social Stratification’”. Sociolinguistics and Language History, ed. by Terttu Nevalainen & Helena Raumolin-Brunberg, 3-18; 58-76. Amsterdam/Atlanta: Rodopi.

Nevalainen, Terttu. 2000. “Processes of Supralocalisation and the Rise of Standard English in the Early Modern Period”. Generative Theory and Corpus Linguistics: A Dialogue from the 10 ICEHL, ed. by Ricardo Bermúdez-Otero, David Denison, Richard M. Hogg & C.B. McCully, 329-371. Berlin and New York: Mouton de Gruyter.

Nevalainen, Terttu. 2006. An Introduction to Early Modern English. Edinburgh: Edinburgh University Press.

Nevalainen, Terttu & Helena Raumolin-Brunberg. 1989. “A Corpus of Early Modern Standard English in a socio-historical perspective”. Neuphilologische Mitteilungen 90: 67-103.

Nevalainen, Terttu & Helena Raumolin-Brunberg. 2000a. “The third-person singular -(E)S and -(E)TH revisited: the morphophonemic hypothesis”. Words: Structure, Meaning, Function. A Festschrift for Dieter Kastovsky (= Trends in Linguistics. Studies and Monographs 130), ed. by Christiane Dalton-Puffer & Nikolaus Ritt, 235-248. Berlin: Mouton de Gruyter.

Nevalainen, Terttu & Helena Raumolin-Brunberg. 2000b. “The changing role of London on the linguistic map of Tudor and Stuart England”. The History of English in a Social Context: A Contribution to Historical Sociolinguistics, ed. by Dieter Kastovsky & Arthur Mettinger, 279-337. Berlin and New York: Mouton de Gruyter.

Nevalainen, Terttu & Helena Raumolin-Brunberg. 2003. Historical Sociolinguistics: Language Change in Tudor and Stuart England. London: Longman.

Nevalainen, Terttu & Ingrid Tieken-Boon van Ostade. 2006. “Standardisation”. A History of the English Language, ed. by Richard Hogg & David Denison, 271-311. Cambridge: Cambridge University Press.

Raumolin-Brunberg, Helena. 2000. “Which and The Which in Late Middle English: Free variants?” Placing Middle English in Context, ed. by Irma Taavitsainen, Terttu Nevalainen, Päivi Pahta & Matti Rissanen, 209-225. Berlin and New York: Mouton de Gruyter.

Appendix