06fo-raw

06fo, PDF
Alexandra Fodor

A corpus study of and as a subordinating conjunction from Middle English to Early Modern English

1. Introduction Conjunctions are usually divided into two subgroups: coordinating and subordinating conjunctions. Although this categorisation is obviously valid, there are overlaps and – as Kortmann (1997) asserts – we can even talk about polyfunctionality among conjunctions. Thus (adverbial) subordinators can function as coordinators (e.g. weil ‘because’ in colloquial German), and coordinating conjunctions can introduce subordinate clauses as well. During the history of the English language and functioned not only as a coordinator but also as a conditional subordinator. The aim of the present paper is to study the use of and as a subordinator in conditional clauses. The linguistic data will be drawn from two corpora, the PennHelsinki Parsed Corpus of Middle English (PPCME2) and the PennHelsinki Parsed Corpus of Early Modern English (PPCEME). The texts considered are presented in detail in the Appendix.

2. Background In this section, I discuss and as a subordinating conjunction in the light of a number of previous studies. Section 2.1. focuses on the origin and lifespan of and ‘if’, section 2.2. on possible contact influences giving rise to the use as a subordinator. 2.1. Origin and lifespan According to the Dictionary of Old English (DOE) (s.v. and conj.), the beginnings of the conditional and can be placed in the Old English (OE) period, cf. gelīce and… meaning ‘like as if’. A number of scholars see Middle English (ME) as the period when the use of and in the meaning ‘if’ emerges. Mitchell (1985:§3668) seems to agree with the Oxford English Dictionary (OED) which places the first occurrence of conditional and into the Early Middle English (EME) period: the OED cites two examples from Layamon’s Brut as the first relevant instances (OED s.v. and, conj. formerly prep. IV.C.). The Middle English Dictionary (MED) cites an example from the EME Lambeth Homilies as the earliest instance of conditional and. (MED s.v. and (conj. (& adv.)) 5). Fischer (1992:348), who sees the subordinator use as a development from the coordinator and, regards and ‘if’ as a later ME addition to the list of subordinators. If most scholars see and ‘if’ as a ME innovation, how close to the present time does its lifespan extend? Several scholars discuss the use of and in the meaning ‘if’ in Early Modern English (EModE). Kortmann (1997:293) gives a list of adverbial subordinators which include an(d) if in ME an if and an(d) in EModE. Visser (1966:§880) observes the use of conditional an both in ME and in EModE, with possible variants and and an if; for EModE he also gives the aphetic form nif. Rissanen (1999:281) points out the decline of and (or an) as a conditional subordinator in EModE as compared to ME. The Even Yearbook 7 (2006), Department of English Linguistics, Eötvös ISSN 1218–8808, http://seas3.elte.hu/delg/publications/even, © 2006, Alexandra Fodor

Loránd

University,

Budapest

A corpus study of and

The conditional and can be found even in PresentDay English (PDE), not only in Irish or Scottish English but also in colloquial PDE. (Kortmann 1997; Rissanen 1999:2812). For American English, Curme (1931:318323) mentions the use of and or its variant an in the meaning ‘if’ in certain dialects today: “An is still to be heard in our southern mountains and here and there in New England”. It is worth pointing out that the relative frequency of and ‘if’ over the long diachrony has not received a systematic discussion in the literature so far. There is, then, all the more reason to apply quantification to the data examined in this essay. 2.2. Is the rise of the subordinator use of and due to contact influence? It is possible that the subordinator use of and arose in English as the result of a language internal development. As Rissanen (1999:281) puts it, “[t]his conditional/concessive use of an(d) may have arisen from a simplified correlative use in which an(d) loosely expresses various relations between two clauses.” It has, however, also been suggested that language contact might have triggered the subordinator use of and. The OED puts forward Old Norse enda as a possible source for the conditional use of and. Klemola and Filppula (1992) propose two factors where and ‘if’ can derive from. According to one hypothesis, it might have Latin origin, but the difficulty with this theory is that “Latin models typically lack overt subordinators” (315). According to another hypothesis the conditional and could stem from Celtic languages, where subordinate clauses introduced by a subordinator meaning ‘and’ are rather similar to subordinate and clauses in ME and EModE. Here is an Old Irish example from Klemola and Filppula 1992: “do ∙bertis cech nolc from osmesse oc taircitul cech maith dóibsom ‘they used to inflict every evil on me, though I was (lit. and I) prophesying every good to them’” (31516). Celtic languages continue using the subordinating ‘and’ constructions. Here is a comparable example from Irish English, cited from Klemola and Filppula 1992. In such a ‘Celtic’ variety of English, contact influence from Irish is very plausible. It is to be noted, however, that the semantics of the nonfinite and clause is not conditional, or concessive as in the previous example, but it rather expresses temporal simultaneity (Klemola and Filppula 1992): “Well, I seen the time you’d buy a farm for…five or six hundred… Seen farms selling and I a young lad” (316). It is also worth noting that Kortmann 1997, when discussing the polyfunctionality of conjunctions, points out that “a coordinating conjunction introducing an undoubtedly subordinate clause is a wellknown phenomenon in Celtic languages and…was clearly not uncommon in earlier stages of other Western European languages” (61). The main aim of this essay is not to try to find a solution to the question whether the conditional and in English is a native development or due to contact influence. However, the fact that in my ME corpus, the conditional and is relatively more frequent in West Midland texts than in texts representing any other dialect would fit the Celtic contact hypothesis quite well (see further sections 3.2. and 6.1. below).

The Even Yearbook 7 (2006), © Alexandra Fodor

A corpus study of and

3. Research: methods and process My specific target in this paper is to examine the use of the conditional subordinator and in Middle English and in Early Modern English. I will thus take prose texts from both periods and analyse them. The chronological coverage is roughly from 1150 to 1710. I wish to find out the exact rate of occurrences of and ‘if’ conditional clauses (and possible spelling variants) in these two periods. In addition to the diachronic analysis, I will also look at diatopic variation. This is naturally only possible with texts for which dialect is given as a separable parameter. As already pointed out in section 1, two corpora will be utilised in my study. Both the PPCME2 and the PPCEME are based on the relevant parts of the Helsinki Corpus. As the two corpora contain different numbers of prose texts, and as those texts differ in length, also relative numbers will be taken into consideration. In both parts all the texts containing and conditionals will be considered, and analyses according to diatopic and diachronic variation will be carried out. The lack of sufficient number of examples in certain dialects and/or periods might be due to the relative limited number of manuscripts available in the parsed version of the PennHelsinki Corpus (as compared to the Helsinki Corpus, for instance). 3.1. Types of conditionals We can divide conditionals into “positive” and “negative” types. These types can be realised by both if and and. Table 1 is based on my corpus material. For a brief discussion of the ME subordinators, see also Mustanoja 1960:469.

Positive 1a. if/and 2a. and if1 3a. what if/and 4a. no conjunction inversion

Negative 1b. if/and … not 2b. but if/and (=unless) 3b. unless 4b. no conjunction inversion Table 1.

This study only concentrates on the positive and types, excluding those instances where no subordinator introduces the conditional clause (4a). During the analysis certain problems arose, however. The first complication concerns subcategory 2b where the but if/and sequence occurs: there are many instances where – seemingly – this combination occurs, but where but itself functions as a coordinating conjunction separating two clauses, one of which is a conditional clause introduced by and. In such cases, the item is categorised as an example of subcategory 1a and is obviously counted. Another problem is found in clauses from 1a: it often happens that and instead of introducing a conditional clause introduces a reported question. These clauses are not included in the analysis.

The OED uses the expression “strengthening effect” when and and if stand together (OED s.v. and, conj. formerly prep. IV.C.1.b.). Kortmann (1997), on the other hand, says: an(d) ‘if’ + if ‘if’ = an(d) if; this produces a polymorhemic subordinator where both elements express the same function/meaning. “There is an element of redundancy involved here.” (312) The Even Yearbook 7 (2006), © Alexandra Fodor

A corpus study of and

3.2. PennHelsinki Parsed Corpus of Middle English The Middle English section of the Diachronic Part of the Helsinki Corpus of English Texts formed the basis of the text samples in the PPCME2. It includes almost 1.2 million words of running text in 55 text samples. Table 2 lists the distribution of word count according to both diatopic and diachronic distribution. In accordance with the Helsinki Corpus there are four subperiods (ME1, ME2, ME3 and ME4), and five dialectal areas in Middle English (Kentish, Northern, Southern, EastMidlands, and WestMidlands). ME1

ME2

ME3

ME4

(11501250)

(12501350)

(13501420)

(14201500)

Total

Kentish Northern Southern EastMidlands WestMidlands

4316 130.804 116.802

51.914 45.035

18.470 104.179 207.831 81.092

11.070 43.834 178.972 162.152

56.230 29.540 148.013 562.642 360.046

Total

251.922

96.949

411.572

396.028

1.156.471

Table 2. The 55 parsed prose texts were put under scrutiny in order to explore the regularity (if possible) of andclauses, and the contrast in the rate of occurrence between the two types of conditional clauses. From this examination it emerged that the ME part of the PennHelsinki Corpus contains 100 instances of and used in the sense of ‘if’. Those instances appear in 23 texts (in addition to the regular ifclauses). There is one text (CMINNOCE), where, interestingly enough, the only conditional clause is introduced exclusively by and, “And he breke them he is sharpely correctyd” ‘If he breaks them, he will be sharply corrected’. Considering the other types of conditionals there is another text (CMREYNAR) where the combination of the two subordinating conjunctions, and if, occur, “But and yf he wolde haue comen hyther he myght haue ben here” ‘but if he would have come … , he might have been here’. The combination what and is not represented among the ME instances. The first conditional and can be found in ME1 in The Lambeth Homilies (West Midland dialect). Like ME1, ME2 has only one example of and ‘if’ (Kentish dialect). A sudden increase can be seen in ME3 with 22 instances in eight texts. In ME4, however, there is an abrupt rise in the number of andoccurrences: 76 andconditionals divided between 13 texts. Thus the number of andconditional clauses reaches its peak during the ME4 period; a considerable number of andinstances can be observed in the West Midland dialect: 79% out of all andclauses occurred there in ME4. Table 3 shows the distribution of andconditionals according to both diachronic and diatopic distribution.

The Even Yearbook 7 (2006), © Alexandra Fodor

A corpus study of and

Period/Dialect Kentish East Midland

ME1 Ø Ø

ME2 ME3 CMKENTSE: 1 Ø Ø CMCLOUD: 11 CMWYCSER: 1 CMMANDEV: 1

West Midland

CMLAMBX1:1

CMBRUT3: 1 CMEDVERN: 3

Southern

CMAERL3: 2 CMHORSES: 2 CMPOLYCH: 1

Northern Total

ME4 Ø CMAERL4: 1 CMCAPCHR: 1 CMINNOCE: 1 CMJULNOR: 1 CMKEMPE: 4 CMREYNAR: 3 CMVICES4: 1 CMMALORY:50 CMMIRK: 7 CMSIEGE: 3 CMGREGOR: 1 CMROYAL: 2 CMROLLEP: 1 76

Table 3.

Diagram 1. When diatopic criteria are taken into consideration, it turns out that there is one dialect area, viz. the West Midlands, where in ME4 the andclauses are in substantial majority – considering all types of conditionals. Because of this rather unexpected result it would be beneficial to examine texts originating from the same dialectal area in EModE as well. It is regrettable, however, that in the PPCEME the diatopic distribution of the manuscripts is not available. As to genre distribution, the following observations can be made: the majority of and conditionals occur in Romance (53%), then in Religious Treatise (14%); the distribution The Even Yearbook 7 (2006), © Alexandra Fodor

A corpus study of and

between the text types is, however, rather wide – as seen in Table 4. Table 4 presents absolute numbers; relative numbers will be considered in a later study. Diagram 2 represents the cumulative percentages from ME1 to ME4.

ME1

ME2

ME3

ME4

Homilies 1% (1) 1% (1) Religious Treatises Sermon Travelogue History Handbook – Medicine Fiction Romance Rule Table 4.

14% (14) 1% (1) 1% (1) 2% (2) 2% (2) 2% (2)

8% (8) 10% (10) 2% (2) 3% (3) 53% (53)

ME1ME4

Homilies (2%) 2% 2%

22 %

Religious Treatises (22%) Sermon (11%) Travelogue (1%) History (4%)

53 %

11 % 1% 3%

2% 4%

Handbook – Medicine (2%) Fiction (3%) Romance (53%) Rule (2%)

Diagram 2.

3.3. PennHelsinki Parsed Corpus of Early Modern English The PennHelsinki Parsed Corpus of Early Modern English consists of almost 1.8 million words. The corpus itself is divided into three subcorpora: the Helsinki part consists of roughly 573,000 words; the Penn1 part consists of roughly 615,000 words; the Penn2 part consists of roughly 606,000 words. The two Penn parts are supplements to the Helsinki part; they mostly contain text samples written by the same authors as in the Helsinki part. Penn2, however, contains more new material than Penn1. In accordance with the Helsinki Corpus itself all the three directories are

The Even Yearbook 7 (2006), © Alexandra Fodor

A corpus study of and

divided into subperiods, viz. E1, E2, and E3. Table 5 presents the distribution of word counts in both subcorporal and diachronic respects. Helsinki 196,754 196,742 179,477 572,973

E1 (15001569) E2 (15701639) E3 (16401710) Total

Penn1 Penn2 194,018 185,423 223,064 232,993 197,908 187,631 614,990 606,047 Table 5.

Total 576,195 652,799 565,016 1,794,010

Due to the twofold supplementation in the PPCEME, the number of the texts analysed, 192 in all, is almost four times as high as in the Middle English part. Interestingly enough, PPCEME also contains 100 instances of and used in the sense of ‘if’ (this already suggests a decline in the number of occurrences). The distribution of those examples is, however, slightly different from that in PPCME2: andconditionals can be detected in all subperiods in EME, as shown by Table 6. Those instances, however, appear in 13 texts (E1:3; E2:9; E3:1). E1 And

Table 6. 90

Number of instances

80 70 60 50

And

40 30 20 10 0 E1

Subperiods

Diagram 3. As far as the distribution in Table 1 is concerned, all types of combination with and were found in the corpus; thus the strengthening effect of if beside and (= and if) was represented in PPCEME as well, e.g. “for so they get more and if they went together” (HARMAN). The what if… combination also occurred in the form of what and …, as in “what and it hadde beene any other man, and not your good dames husbande” ( HARMAN), or “what and she come not” (UDALL). The spelling variants are not as manifold as in the Middle English part; and, however, also had two spelling variants in the form of ampersand (&) and the phonologically reduced

The Even Yearbook 7 (2006), © Alexandra Fodor

A corpus study of and

form an. As mentioned above, the diatopic analysis is unfortunately not possible in this period since only those texts were put into the corpus which represents the standard British dialect. Still, another form of classification is feasible, namely the analysis according to genre. A considerable number of text types is accessible in the PPCEME, from biblical texts to private letters. Table 7 shows the actual genredistribution of andconditionals in each subperiod. Diagram 4 represents the cumulative percentages from E1 to E3.

E1 Fiction Biography Drama Comedy Handbook Letter – nonprivate Letter – private Proceedings, trials Sermon Philosophy Travelogue

12% (12) 16% (16) 38% (38) 3% (3) 2 % (2) 3% (3) 8% (8) 1% (1) 2 % (2)

4% (4) 3% (3) 1% (1) 1% (1) 3% (3) 1% (1) 1% (1) Table 7.

E3 1% (1)2

E1E3

Fiction (12%)

1% 8%

Biography (20%) 1%

Drama Comedy (42%)

12 % 20 %

Handbook (4%) Letter – nonprivate (3%) Letter – private (6%)

Proceedings, trials (8%)

42 %

Sermon (1%) Philosophy (3%) Travelogue (1%)

Diagram 4. 2

According to the syntactic tagging of the PPCEME, there should be no subordinating and instances in E3. There is, however, one clear example of conditional an in this subperiod: “… and so your Honour wou’d have said, an you had seen how the poor thing stuck’t.” I am grateful to Dr Matti Kilpiö for pointing out this example to me. The Even Yearbook 7 (2006), © Alexandra Fodor

A corpus study of and

And meaning ‘if’ occurs in ten text types; the majority can be found in Dramas, in E1. The number of andinstances in Biography and in Fiction is also noteworthy. In E2, however, there is a drastic fall in the use of andclauses that ends in the almost total demise of and subordinators in E3.

4. Diachronic variation from ME1 to E3 In the two corpora analysed both the rise and the decline of the subordinator and can be clearly seen. The first andinstance is found in The Lambeth Homilies in ME1. This is in agreement with the evidence of the MED (see section 2.1. above). From my analysis of PPCEME it appears that at the end of the EModE period the conditional and (or its spelling variants, an or &) was still in use – very infrequently, however. What happened between ME1 and E3? This question will be answered with the help of relative numbers. (The importance of having a look at the relative numbers lies in the deficiency of the number of available texts in the two corpora as well as on the fact that these texts are varying in length.) Table 8 shows the relative frequencies per 10,000 words. In case of andconditionals the relative frequency does not reach 1/10,000 from ME1 to ME3; in ME4, however, it exceeds that number, and the relative frequency is almost 2/10,000. Only a modest fall comes in E1, in E2, however, a drastic decline can be observed; this leads to an almost complete disappearance of and conditionals in E3. Diagram 5 shows the relative frequencies of and ‘if’ in all the subperiods studied.

And

ME1

ME2

ME3

ME4

0,039

0,103

0,534

1,919

1,475

0,214

0,018

Table 8.

Relative frequencies of and 2 1,5 1 And 0,5 0

ME1 ME2 ME3 ME4

Subperiod

Diagram 5.

A corpus study of and 10

5. Questions and tentative answers 5.1. Why the rise? It is interesting why and ‘if’ was the dominant subordinating conjunction in one subperiod and in one genre, and not in others? And why did it emerge in the first place? The rise (and spread) of the conjunction can be due to contact with the Celtic languages, as mentioned above, in section 2. If the usage already existed in any of the Celtic languages and came into contact with any English dialect, then it could have influenced that dialect. As far as the spread of the conditional and is concerned, the increase in the number of instances might not show the situation de facto. Laing (2000) mentions the possible role of some scribes: during the copying procedure it might have happened that the scribe arbitrarily converted the texts, or even translated them into his/her own dialect. Thus it can happen that the distribution of the andconditionals was more uniform in each dialect than the present results show. My own results concerning ME show that and ‘if’ is frequent in the West Midland dialect area. This tallies well with the Celtic contact hypothesis, given the geographical adjacency of this dialect area with areas where a Celtic language was spoken, like Wales. Accounting for the occurrence of the conditional and in seven East Midland texts in terms of direct Celtic influence is problematic: perhaps we could hypothesise that the contact influence leading to the use of and as a subordinator arose in the West Midland dialect area (cf. the early Lambeth Homily instance and the Layamon’s Brut instances cited by MED), and spread to other dialect areas, particularly the East Midland area. There is, however, room for more work here. 5.2. Why the decline? In connection with the decline of conditional and, the hypothesis of Culpeper and Kytö (2000) should be mentioned. They are of the opinion that the occurrence of andconditionals was first minimalised and then disappeared in the EModE period. The reason for this might be that the use of the conjunction and became restricted: thus “other conjunctions might have been used instead of and for particular functions” (309). So after the period of Middle English the usage of both subordinating and coordinating conjunctions became more and more specified: and occurred less in the role of a subordinator until it became almost completely extinct. On the other hand, the process of standardisation has most probably contributed to the ‘tidying up’ of the system of connectors in EModE, and the prescriptive tendencies of the 18th century must have worked in the same direction.

6. Conclusion This paper aimed at examining the use of and as a conditional subordinator from the beginning of Middle English to the end of EModE. It also set out to prove that and ‘if’ played an important role in conditional clauses and thus it should not be overlooked when analysing such subordinate clauses. In the process of analysis it turned out that the use of and instead of if introducing conditional clauses was constantly increasing, especially towards the end of the ME period, while a continuous decrease characterised the EModE period.

A corpus study of and 11

With the examination of the two corpora studied, a quantitative diachronic survey could be presented, which also paid attention to diatopic and text typological factors. In order to get more precise and more reliable results, however, it would be advisable to look at other corpora from both periods as well, and study subordinate andclauses from a variety of angles not touched upon in this essay.

References Bosworth, J. and Toller, T.N. 1955. An AngloSaxon Dictionary. Oxford: OUP. Clark Hall, J. R. 1898. A Concise AngloSaxon Dictionary. London: Swan Sonnenschein. 1962. Concise AngloSaxon Dictionary. Cambridge: CUP. Culpeper, J. and Kytö, M. 2000. “The conjunction and in early Modern English: Frequencies and uses in speech related writing and other texts”. In Ricardo Bermúdez Otero et al. (eds.) Generative Theory and Corpus Studies. Berlin, New York: Mouton de Gruyter, 299326. Curme, G. O. 1931. A Grammar of the English Language in Three Volumes: Volume III. Syntax. Boston: D.C. Heath. Fischer, Olga 1992. “Syntax”. In Norman Blake (ed.) The Cambridge History of the English Language, Volume II: 10661476. Cambridge: CUP, 207409. Klemola, J. and Filppula, M. 1992. Subordinating uses of and in the history of English. In Rissanen et al. (eds.) History of Englishes. Berlin, New York: Mouton de Gruyter, 310318. Kortmann, Bernd 1997. A Typology and History of Adverbial Subordinators Based on European Languages. Berlin. New York: Mouton de Gruyter. Laing, Margaret 2000. “Early Middle English – the EastWest divide”. In Irma Taavitsainen et al. (eds.) Placing Middle English in Context. Berlin, New York: Mouton de Gruyter, 97114. Mitchell, Bruce 1985. Old English Syntax, Volume II. Oxford: Clarendon Press. Mossé, Ferdinand 1952. A handbook of Middle English. Baltimore: The Johns Hopkins Press. Mustanoja, Tauno. F. 1960. A Middle English Syntax. Part I. Parts of speech. Mémories de la Société Néophilologique de Helsinki, 23. Helsinki: Société Néophilologique. OED= The Oxford English Dictionary. 1961. Vol. 1. Ed. by James A. H. Murray – H. Bradley – W. A. Craigie – C. T. Onions. Oxford: Clarendon Press. Rissanen, M. 1999. “Syntax”. In Roger Lass and Richard Hogg (eds.) The Cambridge History of the English Language, Volume III: 14761776. Cambridge: CUP. Visser, F. Th. 1966. An Historical Syntax of the English Language. Vol. II. Leiden:Brill.

Access to electronic corpora and dictionaries: The Dictionary of Old English. Fascicle F and Fascicles AEnglish (with revisions). CDROM Version 1.0 The Middle English Dictionary: http://www.ets.umdl.umich.edu/m/med/ PPCME2: http://www.ling.upenn.edu/histcorpora/PPCME2RELEASE2/index.htm PPCEME: http://www.ling.upenn.edu/histcorpora/PPCEMERELEASE1/index.htm

Texts cited Middle English CMAELR3, CMAELR4: Aelred of Rievaulx’s de Institutione Inclusarum. Early English Text Society, 287. ed. J. Ayto and A. Barratt. London, 1984. PP. 32.270 39.532 (MS VERNON, i.e., MS BODLEY Eng. Poet. A.1., S. C. 393842). CMBRUT3: The Brut or the Chronicles of England, Part I. Early English Text Society_{O.S. 131. ed. F. W. D. Brie. London, 1960 (1906). CMCAPCHR: Capgrave’s Chronicle. John Capgrave’s Abbreuiacion of Cronicles. Early English Text Society, 285. ed. P. J. Lucas. Oxford, 1983.}

A corpus study of and 12

CMCLOUD: The Cloud of Unknowing. The Cloud of Unknowing and the Book of Privy Counselling. Early English Text Society, 218.ed. P. Hodgson.London, 1958 (1944). CMEDVERN: The Mirror of St. Edmund (Vernon Ms.). Horstman, C. 18951896. Yorkshire writers: Richard Rolle of Hampole. London: Swan Sonnenschein & Co. CMHORSES: Late Middle English Treatise on Horses. Sockholm Studies in English, XLVII. ed. A. C. Svinhufvud. Stockholm: Almqvist & Wiksell International, 1978. CMGREGOR: Gregory, William. Gregory’s Chronicle. The Historical Collections of a Citizen of London in the Fifteenth Century. Camden Society, N.S. XVII. ED. J. GAIRDNER. WESTMINSTER, 1876. CMINNOCE: In Die Innocencium. Two Sermons Preached by the Boy Bishop, at St. Paul’s Temp. Henry VII, and at Gloucester, Temp. Mary. Camden Society Miscellany, VII. Camden Society, N.S. XIV. ed. J. G. Nichols. London, 1875. CMJULNOR: Julian of Norwich. Julian of Norwich’s Revelations of Divine Love. The Shorter Version ed. from B.L. Add. MS 37790. Middle English Texts. ed. F. Beer. Heidelberg: Carl Winter Universitaetsverlag, 1978. CMKEMPE: Kempe, Margery. The Book of Margery Kempe, Vol I. Early English Text Society, 212. ed. S. B. Meech and H. E. Allen. London, 1940. CMKENTSE: Kentish Sermons. Selections From Early Middle English 11301250, Part I. ed. J. Hall. Oxford: The Clarendon Press, 1963 (1920). CMLAMBX1: The Lambeth Homilies. Morris, Richard. 1969. Old English homilies and homiletic treatises. Part I. EETS O.S. 29, 34. New York: Greenwood Press. Originally published by Trübner (London, 1868). CMMALORY: Malory, Thomas. Morte Darthur. The Works of Sir Thomas Malory. ed. E. Vinaver. London: Oxford University Press, 1954. CMMANDEV: Mandeville’s Travels. Hamelius, Paul. 19191923 (for 1916). Mandeville’s travels, translated from the French of Jean D’Outremeuse. EETS O.S. 153, 154. London: K. Paul, Trench, Trübner & Co. CMMIRK: Mirk, John. Mirk’s Festival: A Collection of Homilies, by Johannes Mirkus (John Mirk), Part I. Early English Text Society, E.S. 96. ed. T. Erbe. London, 1905. CMPOLYCH: Trevisa, John. Polychronicon Ranulphi Higden, Monachi Cestrensis, Vols. VI, VIII. English translations of John Trevisa and of an unknown writer of the fifteenth century. Rolls Series, 41. ed. J. R. Lumby. London, 1876, 1882. CMREYNAR: Caxton, William. The History of Reynard the Fox. Translated from the Dutch original by William Caxton. Early English Text Society, 263. ed. N. F. Blake. London, 1970. CMROLLEP: Richard Rolle, Epistles (Ego Dormio, The Commandment, The Form of Living). Allen, Hope E. 1931. English writings of Richard Rolle, hermit of Hampole. Oxford: Clarendon. CMROYAL: Middle English Sermons, edited from British Museum MS. ROYAL 18 B. XXIII. Early English Text Society, 209. ed. W. O. ROSS. London, 1940. CMSIEGE: The Siege of Jerusalem in Prose. Memories De La Societe Neophilologique De Helsinki, XXXIV. ed. A. Kurvinen. Helsinki: Societe Neophilologique, 1969. CMVICES4. The Book of Vices and Virtues. A Fourteenth Century English Translation of the Somme le Roi of Lorens D’Oreleans. Early English Text Society, 217. ed. W. N. Francis. London, 1942. CMWYCSER: English Wycliffite Sermons, VOL. I. ed. A. Hudson. Oxford: Clarendon Press, 1983.

Early Modern English BOETHCO: Bax, Ernest Belfort (ed.). 1897. Boethius’ Consolation of Philosophy, translated from the Latin by George Colville, 1556. The Tudor Library, V. London: David Nutt. BOETHEL: Pemberton, Caroline (ed.). 1899. Queen Elizabeth’s Englishings of Boethius, De Consolatione Philosophiae, A.D. 1593, Plutarch, De Curiositate, Horace, De Arte Poetica (Part), A.D. 1598. EETS, O.S. 113. London: Kegan Paul, Trench, Trübner.

A corpus study of and 13

COVERTE: Coverte, Robert. 1971 (facsimile). A trve and almost incredible report of an Englishman, 1612. The English Experience, 302. Amsterdam: Theatrum Orbis Terrarum and New York: Da Capo Press. EBEAUM: Macray, William D. (ed.). 1884. Beaumont papers. Letters relating to the family of Beaumont, of Whitley, Yorkshire, from the fifteenth to the seventeenth centuries. London: Nichols. ELIZ1570.E2: Butler, G.G. (ed.). 1913. The Edmondes papers. A selection from the correspondence of Sir Thomas Edmondes, envoy from Queen Elizabeth at the French court. London: Nichols. EVERARD: Searle, Arthur (ed.). 1983. Barrington family letters 16281632. Camden Fourth Series. Vol. 28. London: Offices of the Royal Historical Society, University College London. GIFFORD: White, Beatrice (ed.). 1931. A dialogue concerning witches and witchcraftes, 1593. By George Gifford. With an introduction by Beatrice White. Shakespeare Association Facsimiles, 1. London: Oxford University Press. HARLEY: Lewis, Thomas Taylor (ed.). 1854. Letters of the Lady Brilliana Harley, wife of Sir Robert Harley, of Brampton Bryan, Knight of the Bath. Camden Society, 58. London: [no publisher]. Reprinted 1968 (New York: AMS Press). MIDDLET: Middleton, Thomas. 1969 (facsimile). A chaste maid in Cheapside, 1630. Menston: Scolar Press. MOXINDEN: Gardiner, Dorothy (ed.). 1933. The Oxinden letters 16071642. Being the correspondence of Henry Oxinden of Barham and his circle. London: Constable. PERROTT: Rawlinson, Richard. 1728. The history of that most eminent statesman, Sir John Perrott, Knight of the Bath, and Lord Lieutenant of Ireland. London: [no publisher]. WOLSEY: Ellis, Henry (ed.). 1824. Original letters, illustrative of English history; including numerous royal letters. Series 1, vol. 2. London: [publisher unknown]. Alexandra Fodor Eötvös Loránd University sanfodor2@yahoo.com