Orthography in the Cely letters.
I will start this paper with a short presentation of the corpus and the differences between the original manuscripts and the edition used for the purposes of the analysis. Then, I will focus on the orthographic variation and selected dialectal features in the discussed letters, mainly on those variants containing the selected graphemes: p, 3, v, u, w, i, j, y, indicating the differences of usage among the authors. Finally, I will deal with homography in the text and the problems it may cause for morphological analysis.
1. The corpus -- the original manuscripts and the edition: summary of differences
The Cely letters is a relatively little known collection of early English commercial correspondence (1472-88). The most comprehensive edition of the letters -- by Alison Hanham (1975) -- is the basis for the present analysis. It contains 247 (1) letters (about 86,000 running words), varying in length (from over 100 to over 900 words), written by over 40 authors, mostly wool merchants. The original manuscripts include about 280 documents which are kept in the Public Record Office in London. Documents omitted in the edition are mostly lists of goods and bills (though there are also a few almost obliterated letters), and therefore their inclusion is not essential for the linguistic analysis.
In her edition Hanham provides the reader with both more and less extensive reconstructions and emendations (2) of words and phrases (obliterated due to damage). Such measures may be useful to a historian reading the documents, but they can be a handicap for a linguist. Therefore, I have also consulted the original manuscripts (3) at the Public Record Office and will refer to the results of this consultation below. An attempt has been made here to analyse the real, not reconstructed or emended language. Hence, all the fully or partly reconstructed forms (or those whose interpretation is impossible because the adjacent words have been obliterated) have been excluded from the analysis. This concerns clauses such as, e.g.,
(1) I hawe [reseyued a letter ffrom yowre] masterschypp (WLC :202: 676) (4)
(2) I haue sent Hayne wyth the wol flet, and (a) barell and a wyrkyn of befe and a barell candyll. (RCII 120: 557).
(3) Delowppys thow3te verylly as thys seson that he schul[d] ... delyuer hym be yowre masterschyppys commavndement a ... soo Wyllykyn hath ben at Calles all thys seson a ... (WLC 220: 1252)
The editorial changes introduced by Hanham concern mainly capitalization, punctuation, abbreviations, superscripts, and word division. (5) Capitalisation in the edition has been modernised. Capitals were used by the authors of the manuscripts only occasionally - usually at the beginning of the text and sometimes in the body of the letter, when the writer wanted to show his (or her (6)) reverence and friendly attitude towards the addressee (e.g., Brother in JD 18, RBC 15, 21). Moreover, proper names are most often spelt with an initial minuscule (e.g., inglond JD 18: 6, god GC 7: 334, thomas GC 124: 236, ihesus RE 152: 137 and 139). Also, the first person singular pronoun is normally represented by a minuscule i or, occasionally, y Hanham introduced capital letters in each proper name and also at the beginning of each sentence, sometimes simplifying double letters used in the original version, e.g., Fordyrmor instead of ffordyr mor (GO 178: 255), Fraunce instead of ffraunce (WLC 241: 1623).
Punctuation in the edited documents is also modern. In the manuscripts it is nearly non-existent: the authors do not use colons, semi-colons or full stops. Instead, single or double virgules (vertical slashes) are used to indicate the end of one sentence and the beginning of the next, but only GC, TK, RBC, RR, EB, RL, RT and TG employ them more or less regularly. In other authors' letters they remain exceptional (e.g., WM, RCI (7) and many less important writers do not use them at all). Past participles with the prefixes j or y are spelt with a hyphen in the edition and separately in the manuscripts, e.g., y wreten (WM 5: 7), y yeuen (WM 5:11), j ressayued (WM 9: 29), y promesed (WM 151:191), y clowsyd (GC 112: 215), j delt (GC 7:331), and I chosen (WLC 160:180). (8) Moreover, in the original letters there is no regular division into paragraphs. Usually, the body of the letter is a continuum and only in the case of lists (of priced goods) do the authors place each item in a new line.
Abbreviations are quite common in the original documents, and often include superscripts. The most frequent ones are: & (instead of and), (9) [p.sup.s] (pis), pt and [p.sup.t] (pat), wt and [w.sup.t] (wyth, with), and [y.sup.r] (your). Also a part of a non-abbreviated word can be superscript, e.g., [p.sup.c] and [y.sup.s] are very common in the letters. All these abbreviations and superscripts are expanded and written in italics in the edition. (10) To make the discussion of abbreviations complete, one also has to mention tachygraphs. These are more or less decorative flourishes or lines. They are commonly used, usually to represent particular letters (usually m, e.g., in cumforde RCII 19: 38, cumpany WLC 237: 1507 or recompens RS 244: 159) or sequences of letters, e.g., -er in letter, -es- in Jhesu, (11) or plural endings -ys, -is or -es. (12) The awareness of the fact that these plural endings were most often replaced by flourishes by the authors of the manuscripts and extended (not always consistently) by the editor saves the linguist from arriving at the false conclusion that these were dialectal variants.
Of the issues mentioned above, the differences in word division are probably of the greatest import for the linguistic (especially morphological and phonological) analysis. Although Hanham usually provides the reader with modernised spelling, she is not always consistent. This applies, i.a., to reflexive pronouns. In the original manuscripts the two elements of the pronoun are almost always written separately. (13) The author of the edition prefers joint spelling, but occasionally leaves the original unchanged, e.g., my selffe (GC 41: 69) and yt selffe (GC 109: 181). Such inconsistency could again lead to wrong conclusions on the part of a linguist relying too much on the edited version. (14) Editorial changes also concern the spelling of other words, e.g., prepositions, verbs, past participles, adjectives, conjunctions and relative adverbs, especially those whose first element is an unstressed prefix. They are often written jointly in the edition and separately (15) in the manuscripts, e.g., (16)
(4) Item, brother, as a pon the Sonday a fore the datte of thys letter my brother Rychard Cely and I wer at Pollys Crosse to here the sarmon (RBC 15: 18)
(5) he hayth won, the no xxxij, j sac di., xviij cl. of goode Cottyswolde a cordyng (17) to be price and ordenance for redy money, wych a montyd vnto xxv li. and od money (JD 18: 3)
(6) our brother has wrettyn for hys wyfe, and sche has a skewyshyd hyr that ther be so many Flemyng and Fraynchem a pon the see that sche dar not com (RCII 47: 134)
(7) be caus bat he was a str[aunger] and not knowen therfor be sayd Barkwey causeyd the sayd Wy[lliam] to make salle, wher for he is trobuld at this tyme (JE 246: 50f)
(8) I be seke yowr masterschypp and my mastyrs all that 3e whyll menystyr yowr law vn to me so that I shall nott nede to seke none hodyr ways (GC 7: 336)
Another problem in the discussion of word division is posed by the displacing of the word boundary. There are quite a few examples of this in the letters, e.g., a nanswer (JD 61: 90), a nox (RCII 119: 542), a ny (RCII 11: 434), a neynd (RCII 117: 500) or a nend (MC 222: 133). The word division in these words is retained by Hanham. However, it is usually modernised in the case of the word 'another', e.g., an nodyr (GC 41: 54), a noder (JD 51: 44), an other (RR 65: 176), a nodur (NK 159: 142), a noper (SRH 245: 194) and a noder (WM 69: 104) (Except for the last example, all are written jointly in Hanham's edition). The particularly unstable orthography of this word implies that what we perceive as one word today must still have been perceived as two words by the authors of the letters: an article and a determiner (or a pronoun). Yet, this hesitation concerning the word boundary may be a sign of the approaching morphological merger and, by the same token, the emergence of the determiner 'another'.
Some words are spelt jointly in the original letters and separately in the edition. These are mainly articles merged with the words following them, e.g., along senen (MC 222: 134), abill (RL 72: 14), apeny (WLC 237: 13). (18) Orthographic mergers (not indicated in the edition) apply also to other words, e.g.,
(9) he wyll dosoe I schall be ys good frende (RCI 11: 28)
(10) Y had west that ye would a taked so sor Y would nat a wreten so vn to you, nat and Y schuld agette therbey xx nobelys (WM 110: 131f) (19)
(11) and at your comyng ye shalbe trewlye payed therfore, whatsomeuer it cost (RT 123: 210)
(12) nomore vnto yowre mastyrschyppis at thys tyme, but allmyghty Jhesu preserue yow (WLC 237: 1518)
(13) ffarthermore pleesseyt yow to wette that I heue ressayuyd fro yow a letter wrette at Calles the xxx day of Octobor (RBC 15: 12)
All the examples of mergers quoted above are probably instances of orthography being influenced by the spoken language - the quoted writers must have perceived each sequence as one word and not two morphologically heterogeneous items.
An additional question that may be considered important by some researchers is the orthographic merger of p with y (20) in some of the letters, e.g., ye ('the') in RR 42, JD 44, 139, GC 124, NK 159, JE 246, yt ('that') in RR 42, JD 44, JE 246. In Hanham's edition p is employed in all these examples and no comment is made about the ambiguous status of the grapheme.
2. Orthographic variation and dialectal features
The original inventory of graphemes has not been changed by the editor. (21) It does not differ much from that used in Present-day English, apart from the two runic letters 3 and p, inherited from Old and Middle English. These two receding graphemes function in most cases as orthographic variants of y (or sometimes g) and the diagraph th, respectively. The other two sets of allographs that will be discussed below are a) i, j (22) and y, and b) u, v and w. (23) The grapheme z occurs only six times in: Zelond (twice, WLC 241 and 242), filz (a borrowing from French, twice, RR 42 and 65), poundez (once, YDC 1), and Callez (once, SJW 129).
Both 3 and p are still quite common in the letters (see Table 1). There are 270 (including 145 times pe and 57 times pat) occurrences of p in the letters of 23 writers and 586 (370 times 3e) of in the letters of 20 writers. Of the more important writers (those who wrote more than five letters) only TK does not use either character. Only 10 authors use both forms (WLC, GC, JR, RBC, RR, SRH, JSC, JPM, MC, SJW, and WA). WLC, JR and SRH are the only writers to use both graphemes with roughly the same frequency (the proportions are 117: 108, 2: 2 and 12: 9 respectively), whereas the remaining authors show a clear preference for one or the other letter, i.e., RR, SRH, JSC, JPM, MC, and WA for p, and GC, JD, RBC, and SJW for 3. In the case of JPM this preference is particularly marked, since the proportion between the two graphemes is 23: 1 in his letter (142, the only one he wrote; 462 running words). In the letters written by JC, WD, EB, JDC, JE, NK, RCD, RL, RS, TG, and AN only p has been found, and in the ones b y RCII, RCI, WM, HB, RE, R.A., RG, and VW, cnly 3. The usage thus varies widely among the authors.
The number of occurrences of 3 is high in RCI (53), and truly remarkable in RCII (273). Moreover, out of the 76 different forms containing the grapheme 3 recorded in the letters, 27 are to be found only in those written by RCII, e.g., 3er, 3eyr, bow3er, 3eystyr, and 3end. Even within the limited group of forms containing 3 and found only in his letters RCII's usage shows considerable variation, e.g., he uses five different orthographic forms to express the adjective 'young': 3enge (three times), 3eung (once), 3eunge (three times), 3ewng (once), and 3ewnge (six times). No variants of this adjective with an initial y have been found in his letters. The only other examples of the adjective 'young' in the whole collection are recorded in WLC (yong three times and yonge twice). RCII uses also three variants for the adverb 'yet': 3eyt (18 times), 3eeyte (once), and 3ehyt (once). In this case there are no variants with [y.sup.25] either. Other authors also employ forms with an initial 3 to express this word, e.g., WLC, GC and JR use 3ett (16, 10 times, and once, respectively), (26) RCI uses 3ete (seven times; and twice yete), RBC -- 3ette (once), RE -- 3yt (twice), and JR 3et (once). None of the last three authors uses any variants with an initial y in this word. In RCI's letters 10 idiosyncratic forms have been recorded, i.e. 3ete (mentioned above), 3eue, 3eve, 3eyng, 3eueyng, 3ere, 3erely, 3ongar, 3oure, and 3etys. A curiosity worth mentioning is SJW's use of the form 3er for the verb 'hear'.
As regards the usage of p, the number of forms (both types and tokens) containing this grapheme is much lower than the number of forms containing 3. WLC, with his 117 occurrences of p (80 of those in the word pe), is the unquestionable leader among the authors. This number seems small compared to the 1091 occurrences of the in the same letters. He also uses the form pat quite often, i.e. 22 times, but this number again seems low, compared to 534 occurrences of that in his letters. Most of the other writers use forms with p only occasionally, as variants of forms containing the diagraph th. This does not apply to the minor writers NK and RCD (each of whom wrote only one letter), in whose letters pe is the only exponent of the definite article. Yet, though RCD (60) does not use any form with an initial th, NK (159) uses thys once. In JPM there are as many as 23 forms with the initial p in a single letter (142). However, he is not consistent either -- three occurrences of the have been found in his letters.
Authors writing on Middle English and Early Modem English agree that v and u were allographs in complementary distribution (see e.g., Gorlach 1978: 50), i.e., v occurred word-initially and u word-medially, regardless of their phonetic realisation as either [v] or [u]. This regularity can easily be illustrated with examples from the letters:
(14) plese hytt yowre mastyrschyppys to vnderstond that I haue receyued yowre lettyr datyd at London the ffyrst day of September (WLC 235: 1443)
(15) Richard Stokys toke vpon vs for ... the wiche we toke in honde at the wrytteng and enstance of the Erle of W[orcester, Tresurer of] Englond, of whom we hadde wrytteng to haue sauyd vs harmeles (TK 219: 124)
Indeed, this rule is observed in the majority of the letters. However, apart from the graphemes v and u, also w, and occasionally f(f) occur as variants. Two sets of grapheme sequences have been chosen to check the differences of usage among the authors, i.e., vn-, un-, wn-, on-; have, haue, and hawe. The results of the search show that apart from the expected occurrences of v and u, some authors employed also alternative graphemes and diagraphs: w, (27) f and ff. This is illustrated in Tables 2 and 3, and in examples 16-23 below.
(16) But prevely kepe thys maters preve and lette me understand ys entent (RCI 85: 412)
(17) I haue resayuyd a letter from you, wrete at Bregys the xxiij day of October, the weche letter I haue wyl understand, and I tryste to God 3e be reqvrede and wyll amendyd (RCI 67: 354)
(18) I haffe wreton wnto hym to delyuer it yow (RL 72: 180)
(19) And for the love of Jhesu. forget not this, as ye wil have my servis whil I lif (RS 244: 163)
(20) Right whelbelouyd brother George, I recomend me to you, informyng you that I haue beyn to seke your blacke goune at Redhodys (RCII 32: 73) (32)
(21) I wndyrstond that ovr brother Robard and 3e and I ar sewyd at Westmyster (RCII 25: 65)
(22) I wnderstonde that yow haue ben sore seke ande now well rewiwid (TK 76: 23)
(23) 3e mwste remembyr to by canwas to packe in (RCII 197: 958)
Table 2 shows considerable polarisation among the authors. Vn- is the most common of the initial sequences analysed. WLC, GC and JD show particular consistency in their use of this form. The analysis of WM's letters yields one example of wn- (possibly a misspelling), but otherwise he consistently uses vn-. The other writers for whom this is the only sequence employed are JR and JPM. RCII definitely prefers wn- and RCI, un-. However, they are not consistent, i.e., RCII uses on- 14 times and vn- five times, and RCI uses vn- three times and on-twice. TK accepts both vn- and wn-; RL prefers wn-. The remaining authors shown in the table seem undecided as to which sequence to use.
Also, in Table 3, the differences among the authors are easily noticeable. The group of the most consistent writers includes: RCI, JD, TK, JC, JR, RE and RR using exclusively haue, WD, RS and VW using have, and RL and WMD using haffe. WLC prefers hawe, but haue occurs in his letters 11 times and haw twice. RCII favours haue in his early letters, but in the first half of the year 1482 he changes to hawhe and from that time onwards uses it almost exclusively. Occasionally, one finds hawe (six times), hafe (twice) and haw (once) in his letters. Another instance of hafe occurs in RBC, although otherwise he prefers the form haue. GC is also undecided, but he shows a rather strong preference for hawe. Finally, SRH hesitates between haue, hawe and haw.
To sum up, the differences among the authors are considerable. None of them strictly observes the aforementioned complementary distribution of the graphemes v and u, e.g., RCI prefers using u both word-initially and word-finally, (33) and RCII's usage stops conforming to the rule, with the adoption of hawhe.
The graphemes i, j and y are distributionally interrelated. In the manuscripts, i and j are apparently considered to be allographs of the same grapheme (no matter whether they correspond to (i), (j) or a numeral), i.e., a long stroke is used word-initially and word-finally and a short stroke in the middle of the word, and the grapheme i (a short stroke) never occurs word-initially or word-finally. This rule is seen especially in numerals, e.g.,
(24) Wrytten at London the xxiij day off Marche. Per Wylliam Cely (WLC 82: 20)
(25) I haue resayuyd from the ij letters be Wyll Maryon, j lett[er wryt] the xxiiij day of October, the weche j wyll understand, the toder letter wryt frysth day of Novembor (RCI 38: 226)
(26) Thomas Folborne, my prentys, freman of the Staple wytin iiij 3ere of ys terme (RCI 13: 53)
Similarly, the authors provided no indication whether a particular grapheme should be a capital or a minuscule. Thus, despite what one can see in the edited text, the first person singular pronoun 'I' usually does not differ graphically from the numeral 'one' (example 25 (34)) or from the first letter of jhesu, a word frequently used as part of the salutation at the beginning of a letter. In both cases the long stroke is used.
The letters j (a long stroke) and y are interchangeable as exponents of the old verbal prefix still occurring in past participles in a few documents, mostly those written by WM (23 instances) and occasionally by GC (two instances) and WLC (one instance), e.g.,
(27) Thomas Kesten hat y wreten vnto me that Y schuld fynd the way and the men to saff the cortte harmeles her (WM 5: 7)
(28) ye may say vnto hem that my maysterys youre modere haue j ressayued from John Cely ys viij sarpelerys of woll (WM 9: 29)
(29) Yowr wholl ys ally y clowsyd and nevyr a hone pakyd agen saweng that at ys ffowllest arayyd ys takyn togydyr (GC 112: 215)
However, the grapheme j (35) is much rarer in this function (four occurrences) than y (with 22 occurrences).
The graphemes j/i and y are interchangeable also in other sets of words, e.g, (h)yt and (h)it, hys and his, thys and this, byll and bill.
While analysing the orthographic variation in the Cely letters, I came across a few interesting dialectal features and individual authors' idiosyncrasies. Due to the mixture of different dialectal forms in the language of these letters, it is impossible to classify it as belonging to any particular dialect. Yet, although little information on the origins of the Cely family is available, they are often referred to as Londoners. (36) Samuels (1981: 51) hesitates how to classify their language in his typology of Late Middle English texts suggesting, nevertheless, that it might be an example of City English.
I will not try to present any new theory about the origins of the Cely letters as a whole. My aim is to provide a few more examples testifying to the degree of orthographic variation among the authors. One tends to notice mainly easily localisable features. Thus, northern forms in the letters attract attention. These include, e.g., qw(h)- and qu(h)- (37) forms found mainly in RCII (qw- also in RR) (38) (examples 30-32), haf(f)e (see Table 3 above and examples 18 and 33), haueff (JD, example 35), -f(f)yd (in RCII and JD, example 34), giff (RL, example 36), geffun, geyff (JD, example 35), thaym(e) (SJW, RR, example 37), and thay (mainly in RCII, HB, SJW, example 38). Forms characteristic of West Midlands can also be found in the letters, e.g., past participles with verbal prefixes (examples 27-29), wycche (WMD 192: 267), mycche (WMD 192: 269), (39) and onswer (40) (JD 140: 90).
(30) Syr, he has made howyr wyth Lombardys to Brogys apon [v.sup.c] li. qwherof he sendys to you iij lettyrs of exchonge, the qwyche he prays yow to schow them (RCII 121: 566)
(31) Sir, I haue ressauyd a letter from yow wrytt at Calleys the xij day of Nowembyr, wherby I wndyrstonde yowr greve, qwerof my godfather has wryttyn to yow (RCII 111: 425)
(32) And yf het ly in my power I schall do as moch that schal be vnto your plesure, as knoweth owre Lord, qwou send you good fortune wyth pe accomplichment off your goodly desyrys (RR 42: 161).
(33) and moste partte haue kepte my bedde, for I hafe ben so seke and sore that I goo wyth a staffe (RBC 102: 56)
(34) To my intierly beluffyd brother Gorge Cely, merchaunt at the Staple of Calles (JD 157: 131)
(35) I tolde hym he schulde geyff me redy money or he went owt of Gales or ellys he schulde haueff no wooll of me at thys tyme, and soo we departyd (JD 18: 5)
(36) I shall giff yow in euery li. acordyng as ye desir becawse pat master Levetenaunt callyth so fast apon for custem and subsete ye shall #haffe# in euery li. vj d. (RL 72: 174) (41)
(37) As thay sal se al the remedyis and demandys at the sayde Herron makys, I sende thayme be the sayde frere (SJW 129: 225)
(38) And I ansford them and thay wold geue me redy mony xxv s. iiij d., mony corant in Calles (RCII 34: 93)
Apart from dialectal features, several forms characteristic of individual writers have been found, e.g., gyu 'you', hett 'eat', inuiatory 'list' and whyt 'with' in RCII, iowr in GC, and also onyvn 'onion' in WLC.
The degree of homography in the letters is striking. Because of the number of homographs I will discuss the problem in the analysed documents on the basis of a few selected examples: a, an, her, the, they, and ffayer.
A is a typical example of a troublesome form in the letters. In most cases it represents the indefinite article 'a' (e.g., occurring 189 times out of 227 instances in the 67 letters written by WLC, see example 39). However, it also occurs as the preposition 'to' (from French a, 40), 'of' (41) and 'on' (42), the auxiliary 'have' (an unstressed form, 43), the third person masculine singular pronoun 'he' (44), the first person singular pronoun 'I' (45), indefinite pronoun 'all' (46), determiner 'all' (47), conjunction 'as' (48), and part of the adverb 'a little' (49).
(39) a lyttyll golde ryng wyth a lyttyll dyamond, and a typete of damaske (RCII 86: 319)
(40) a my welbeluffyd Gorge Cely, merchaunt at the Staple of Calles (JD 34: 44)-- preposition 'to' (borrowed from French)
(41) the sam day that Y departed fro Calles Y londed in the Dovnys at iij a clokt at afternon (WM 39: 37)
(42) now whyll I am a whryttyng of thys letter, Wylliam Mydwyttyrs mane ys com to fet mony and I pwynte my tyme wyth hym (RCII 148: 775
(43) y myght so a ben ffor that sesson ryght whell (GC 109: 174)
(44) a connott kepe the promysse that a made vnto yow (GC 41: 73)
(45) I haue payd and a mwste pay vyth#in# thys v days in parte of pyment of thes fellys (RCII 91: 337)
(46) I pray you to recomend me to my nostes and all goyd frendys a be name. (ROII 43: 122)
(47) I pray yow brynge wyth yow a the rekenyng that I am indettyd to you (ROII 95: 369)
(48) Furder syr, a tochyng the matter that yowre masterschypp wrote ^me^ of Thomas Whyte, meser (WLC 215: 1136)
(49) I hawe byn a lykull dyssesyd, but I thancke Godd I am amendyd and walkynge (WLC 201: 668)
Ambiguity can also be observed with regard to the form an, which can be the exponent of the indefinite article (see example 50), the auxiliary verb 'have' (in an unstressed form, see example 51), the preposition 'on' (52), and the conjunction 'and' (53).
(50) Allso ther ys an veryavns bytuyxt Kesten and John Vandyrhay ffor ix sarplers woll (GC 4: 7)
(51) Whe wolld an takyn dyvars whays wyth them but they woll no [oder] whay but as the wher payd last (GC 41: 64)
(52) he schall schypp be an sarten day, or e[ls ...] #ffall# yn an penwalte (WLC 160: 185)
(53) No more vnto yowre masterschyppys at thys tyme, but allmyghty Jhesu preserue an kepe yow (WLC 209: 938)
Her is also an exponent of several morphological categories. It occurs as the third person oblique sg feminine pronoun 'her' (54), the possessive determiner 'her' (55), the possessive determiner 'their' (56), the adverb 'here' (57), the preposition 'before' (58), and the verb 'hear' (59).
(54) sche has prayd me to whrayt to yow to by for her a for of calla[b]yr for to lay in the same gowne (RCII 165: 830)
(55) sche ys deyd and aull her whelpys (RCII 133: 627)
(56) Men ffeyr hytt wyll be nawght ffor ther ys but lycull ware heyr ffor men to bestow her mony apon (WLC 77: 10)
(57) Y schuld fynd the way and the men to saff the cortte harmeles her (WM 5: 7)
(58) my noncle ys exseketurs has promysyd me and Plomton be the faythe of ther bodys to be wyth howr father her Myhellmes (RCII 117: 500)
(59) Y can her nothyng therof (WM 39: 40)
Another homograph, the, occurs most often as the definite article (60), but it is not infrequent as the second person singular pronoun 'you' (42) (61), the third person plural pronoun 'they' (62), part of a relative marker (63) or a conjunction (64), and probably also an adverb (65 (43)).
(60) I had the wycwr of Awelay, and the preste of Awelay, and the preste of Berweke vyth #me# iij nytys, and dynyd and suppyd (RCII 19: 31)
(61) I pray the send me wrytyng as sone as ye can of youre avyse (RCI 20: 98)
(62) thyne we herd ij womon in the towne and the keped hyme at anoder howsse in the towne, and sow he ys deyd and theparded to God (JG 75: 62)
(63) I recommavnd me vnto wyth all my hart, desyryng to heer of youre welfare, the wheche I beseche Jhesus longe to preserue and kepe (RE 154: 149)
(64) I understonde that mater before youre wrytyng, for the weche I haue bogwyt not j sacke woll thys seson (RCI 31: 180)
(65) Ye shall not fawte of your mony in the Colde marte: the be such men as I shall haffe my mony of in that marte as I dare make me fast apon it is Jacob Yong Jocobson and Laurans Lambryghtson wyth Peter Martson: the be fast men (RL 72: 17Sf)
They occurs as the third person plural pronoun 'they', the definite article 'the' (only five times, in JD's letters, no. 28, 44, 51, 125, and 180), and the conjunction '(even) though' (example 66).
(66) therfor Y wold that he had the lengar day of the rest they yt be Mehellmas day (WM 151: 188)
Other homographs include, e.g., be (the preposition 'by', the infinitive, the present indicative plural and the present subjunctive forms of the verb 'be'), by (the preposition 'by', the present indicative plural form of the verb 'be', the infinitive of the verb 'buy'), hys (the possessive determiner 'his', and the present indicative third person singular form of the verb 'be'), owr (the possessive determiner 'our', and the noun 'hour'), pese (the nouns 'piece' and 'peace'), sawhe (the preposition 'except', the present subjunctive and infinive of the verb 'save'), sum (the determiner and pronoun 'some', and the noun 'sum'), ther (the adverb 'there' and the possessive determiner 'their'), were (the infinitive of the verb 'wear' and preterite indicative singular and plural of 'be', the adjective 'very', and the adverbial relative marker 'where'), whos (the relative pronoun 'whose', the preterite indicative first and third person singular form of the verb 'be'), ys (the third person singular of the verb 'be', th e possessive determiner 'his', the infinitive of the verb 'ease', and the demonstrative pronoun 'this').
There are cases where different parts of speech occur in the same form in the same sentence, e.g.,
(67) he was a thys syde Ypur wyth a grett ost off men (WLC 182: 402),
(68) they schuld a hadd a sparynge off a grett partt off ther costom (WLC 203: 704)
(69) thay toke hyt thankefully for thay had cwm a myle a fote that mornyng (RCII 165: 821)
(70) I pray yow by me a fayer and sewyr bycoket a standarde a payr scleuys a 3e haue and a fowld of mayll (RCII 114: 484)
(71) he was a thys syde Ypur wyth a grett ost off men (WLC 182: 402)
(72) and be that day yowre brodere porposed be the leue of God to be at Nortlacht, an at that day Wylliam Breten hat y promesed therfor to met hem (WM 151: 191)
(73) I grete the wyll, and I haue resayuyd from the a letter wryte at Caleys the xiij day of Auguste, the weche letter I haue wyll understand (RCI 31: 169)
(74) The sekenese raynyd sore at London, God sesyd wan ys wyll ys (RCI 56: 337)
Occasionally, homography may lead to ambiguity with regard to the morphological interpretation of word forms, e.g.,
(75) the costum of yowr iij M (1)C ffelles amovntys to xxv ii. vj s. viij d. ster. wherof ge most abat to pay at London of vj sarplers and a C ffelles hewry [v.sup.c] ffelles ys an sarpler (GC 247: 299)
(76) I schall send syche specyalltys of yowrys that be payabull yn thys martt be Wylliam Hy[ll] to receyue the mony for them, whyche amovntys an Cx li. ster. (WLC 216: 1170)
(77) John Dalton hath ben at a ffeyr yn Flaunders, and he hath bowght hym a ffeyr yonge horsse, and he standyss yn yowre stabull (WLC 162: 206)
(78) Ye shall not fawte of your mony in the Colde marte: the be such men as I shall haffe my mony of in that marte as I dare make me fast apon it is Jacob Yong Jocobson and Laurans Lambryghtson wyth Peter Martson: the be fast men RL 72: 175f)
In example 75 the form a may be interpreted as the indefinite article or the preposition 'of', in 76 an can be considered as either the indefinite article or the preposition 'on'/'to', (44) in 77 ffeyr can be the exponent of the adjective 'fair' or of the adverb 'fairly', and, finally, in 78 the can be the adverb 'there' or the pronoun 'they'.
The degree of homography is not the same for all the authors. This has been illustrated on the basis of three high-frequency forms: a, an, and her in the letters of those authors whose output exceeds 1900 running words, i.e., WLC, RCII, RCI, GC, WM, JD and TK. Tables 4-6 show the number of morphological interpretations for each of the forms in the letters of the above mentioned writers.
One could argue that the number of different interpretations of a particular form depends on the size of the analysed corpus. However, as we can see in the tables provided, the relation between the two is not directly proportional. From Table 5 we learn that RCII is the heaviest user of a. There are 236 occurrences of this form in his letters, compared to 227 occurrences in WLC, although WLC's corpus is almost twice as large. It is also in RCII's letters that the number of morphological interpretations of the form is the largest (10). In WLC and GC homography is also high -- a can serve as the exponent of 7 or 5 different parts of speech, respectively. In almost all the cases, a is the unstressed form of on, of have, he, I, all and as. A as the preposition 'to' constitutes an exception -- it is a full form of a, a borrowing from French. It is used as such only by four writers: WLC, RCII (21 occurrences), JD, TK and RBC.
A small number of occurrences of the form a functioning as the indefinite article in GC's letters can be explained if we take a look at Table 4, which shows this author's predilection for an in this function. In his letters, an seems 1:0 be an almost universal exponent of the indefinite article regardless of the orthographic and phonetic environment in which it occurs. The other writers use an much less frequently, in most cases before vowels. RCI employed an as the indefinite article only once and WM and JD did not use it in this function at all. This, however, is understandable in view of the fact that no word following the indefinite article in their letters starts with a vowel. Like a, an is used as an unstressed form of on and have, but its frequency in that function is much lower and the number of possible morphological interpretations is smaller.
The third form taken into consideration is her. It occurs quite often in the letters, especially those by WLC and RCII. It is also in their letters that the homography with regard to this form is the highest. Only RCI does not use her at all. Instead, he uses here 20 times (16 times as the verb 'hear', three times as the adverb 'here' and once as the third person singular feminine pronoun 'her'). (45) While the remaining authors use the form her as the exponent of, at most, two parts of speech (e.g., in WM it occurs 14 times as an adverb and three times as a verb), in RCII the number of interpretations is five, and in WLC it is four (with a noticeable preference for the possessive determiner 'her').
One can conclude that the degree of homography is the highest in RCII's letters. Interestingly, it is also in his letters that the orthographic variation seems to be most noticeable (see the previous sub-section). Homography is considerable in WLC and GC. The form yielding the greatest number of homographs is a.
In this paper I have presented a few important issues concerning the orthography of the Cely letters; viz., the problems resulting from the necessity to rely on the edited version of the documents (concerning, e.g., word division and abbreviations), orthographic variation among the authors, and homography and related problems.
The results of the research show that the level of orthographic standardisation in the letters is low. It is indicated by the high degree of homography and considerable differences among the authors with regard to orthographic variants used in their letters. Each of the authors has different orthographic habits, cf., e.g., GC's predilection for an as the exponent of the indefinite article, RCII's frequent use of <3>, dialectal features noticeable in the letters written by JD, RR, RCII, RL, RBC, SJW, and WMD. The writers are not consistent even within their own orthographic systems, which, in fact, seem only vaguely established.
Nevertheless, the difficulty in subsuming the language of the letters under any particular dialect must be an indication of a process of standardisation having already started (see Samuels 1981). Tendencies such as, e.g., the prevalence of haue and vn- with complementary distribution of the graphemes <v> and <u> being the rule for the majority of the writers, seem to confirm this claim. Apparently, the documents reflect an intermediary state of the language, which is on its way from dialectal and individual standards towards a national one.
Table 1 Occurrences of p and 3 (24) AUTHOR 'ETH' 'YOGH' WLC 117 108 RCII - 273 RCI - 53 GC 12 73 WM - 1 JD 10 18 TK - - RBC 3 10 JC 5 - JR 2 2 WD 2 - HB - 9 RE - 3 RR 11 7 SRH 12 9 AN 16 - Table 2 The distribution of variants vn-, wn-, un-, on- (28) according to authors (29) vn- wn- un- on- WLC 361 -- -- 9 RCII 5 91 -- 14 RCI 3 -- 76 2 GC 112 -- -- -- WM 123 1 -- -- JD 32 -- -- 10 TK 6 6 -- -- RBC -- 2 1 -- JR 4 -- -- 1 SRH 2 1 -- -- JPM 3 -- -- 4 RL -- 3 -- -- RG 3 1 -- -- Table 3 The distribution of have, haue, haw(e), hawhe and haf(f)e (30) according to authors (31) have haue hawe haw hawhe haffe hafe WLC -- 11 285 2 -- -- -- RCII -- 111 6 1 46 -- 2 RCI -- 139 -- -- -- -- -- GC 1 3 53 -- -- -- -- WM 3 45 -- -- -- -- -- JD -- 48 -- -- -- -- -- TK -- 15 -- -- -- -- -- RBC -- 16 -- -- -- -- 1 JC -- 7 -- -- -- -- -- JR -- 21 -- -- -- -- -- WD 5 -- -- -- -- -- -- HB 1 9 -- -- -- -- -- RE -- 9 -- -- -- -- -- RR -- 4 -- -- -- -- -- SRH -- 1 1 1 -- -- -- RL -- -- -- -- -- 4 -- RS 3 -- -- -- -- -- -- VW 1 -- -- -- -- -- -- WMD -- -- -- -- -- 5 -- Table 4 an TOTAL Indef Art. 'on' 'have' 'and' WLC 54 51 1 -- 2 RCII 26 25 -- 1 -- RCI 1 1 -- -- -- GC 63 56 2 5 -- WM 1 -- -- -- 1 JD -- -- -- -- -- TK 3 3 -- -- -- Table 5 a TOTAL Indef. 'to' 'of' 'on' 'have' 'I' 'he' Pron. Det. Art. 'all' 'all' WLC 227 190 1 4 2 29 -- -- -- -- RCII 236 177 21 4 10 7 2 10 1 1 RCI 111 108 -- -- -- 3 -- -- -- -- GC 19 7 -- 2 1 5 -- 2 -- -- WM 51 43 -- 1 -- 7 -- -- -- -- JD 14 11 3 -- -- -- -- -- -- -- TK 14 13 1 -- -- -- -- -- -- -- Conj. 'as' WLC 1 RCII 2 RCI -- GC -- WM -- JD -- TK -- Table 6 her TOTAL Pron. Det. Det. Adverb Prep. Verb 'her' 'her' 'their' 'here' 'ere' 'hear' WLC 22 6 3 12 1 -- -- RCII 25 4 11 -- 8 1 1 RCI -- -- -- -- -- -- -- GC 6 -- -- -- 4 -- 2 WM 17 -- -- -- 14 -- 3 JD 5 -- -- -- 5 -- -- TK 2 1 -- 1 -- -- --
(1.) Two of these written in French, and one written in Dutch, have not been taken into account.
(2.) See Hanham (1975: xxvii).
(3.) I have had the oportunity to consult over one hundred (i.e. nearly half) of the documents included in the edition.
(4.) In the examples, square brackets indicate reconstructions and round brackets - emendations (letters or words added for the sake of comprehensibility). The code and numbers refer to the name of the author, the number of the letter (according to Hanham's edition), and the number of the record in the electronic version of the letters prepared by the present author (the database has been prepared in Access 2000). See the appendix for the full names of the authors.
(5.) Additionally, two evident mistakes (misprints?) have been found: in one of RCII's letters Hanham uses ye instead of ge found in the manuscripts (RCII 47: 143), and in NK's letter (WA 166: 20) she uses that instead of pat.
(6.) In the whole collection there are two letters written by women, and only one of them is written in English.
(7.) They are found only in drafts 16 and 17 (AN), which, though attributed to RCI, are written in a different handwriting.
(8.) See below for a discussion of orthographic variation and dialectal features.
(9.) According to a different taxonomy the symbol & could also be treated separately, not as an abbreviation, but a word sign (see Fisiak 1996: 14, 16). Also other word signs which would be difficult to type here are frequently used in the letters to represent the words per and syr.
(10.) With the exception of &, which is sometimes retained in the edition, e.g., WLC 201: 659, 202:680 and 682.
(11.) This word is most often represented as jhu with a kind of macron above it. See below for the discussion of the graphemes I and j.
(12.) Flourishes used to represent the endings -er, -ys, -is and -es are so common that giving examples of particular documents where they occur seems pointless.
(13.) There are a few exceptions, e.g., myselff(JR 65: 175),yourselffe (NK 159: 143), hemselffe (WLC 241: 1623), mysellfe (WM 69: 103).
(14.) Cf. Rutkowska (1999: 160ff.).
(15.) This confirms the findings of the authors of A linguistic atlas of late medieval English (henceforth LALME), vol. 4: xiii, concerning other late mediaeval texts.
(16.) All this said, it has to be remembered that in certain cases the interpretation of given words as written jointly or separately (on the basis of the handwritten text) is not easy and free of doubt.
(17.) This word can be treated differently, because it was divided across the line.
(18.) In certain cases the joint spelling was retained by the editor, e.g., prayow in WLC 105: 1700, alantarn (RCII 37: 994), and amane (RCII 37: 994).
(19.) Interestingly, in the preceding two cases in the same sentence, the auxiliary and the past participle are written separately.
(20.) Such merger can also be observed in other fifteenth-century texts (see Scragg 1975: 2).
(21.) Apart from the tachygraphs and most of the word signs. See the previous sub-section discussion.
(22.) authors of the manuscripts did not use dots over i and j. The single exception to this rule is the letter written in French by Waterin Tabary (no. 62).
(23.) Other sets of allographs can also be found in the letters, e.g., sch and sh in words such as 'shall', but these will not be analysed here.
(24.) Only the authors of more than one letter are included in the table.
(25.) However, other words with the initial y occur quite often in RCII, e.g., you -- 123 times.
(26.) Still, while yett is the most frequent variant of the adverb in question in WLC (28 occurrences), for GC and JR the variant with g is the only one used.
(27.) Additionally, some interchangeability of w and v has been noticed, such as, e.g., in wyth and vyth (RCII 19).
(28.) On- in words such as, e.g., on, only, onny ('any') or onys ('once') has been excluded from the count, since for these words no variants with v, u or w occur.
(29.) The authors not included in the table use only the variant vn-.
(30.) One example of the form haueff (JD 18: 5) and one of affe (SJW 129: 224) have also been recorded.
(31.) The authors not included in the table use exclusively haue (except for SJW, see fn. 31).
(32.) Cf. gown (RCII 25: 62, 74: 212), gowyn (RCII 43: 112f, 114: 494), govnne (WM 40: 66).
(33.) Still, the impression of RCI's relative consistency fades in view of the fact that he uses medial -v- in all instances of reflexive pronouns found in his letters, e.g., youre selve (2: 12), hym selve (38:242).
(34.) Elsewhere in the examples, the modernised spelling of the pronoun 'I' adopted by Hanham is retained for the sake of comprehensibility.
(35.) Hanham uses i instead, probably to indicate the vocalic nature of the sound to which this letter corresponds.
(36.) See Nevalinna (1996).
(37.) Forms with qw(h)- and qu(h)- were also common in East Anglia.
(38.) RCII spent some time in York (see Hanham 1975: xiv-xv); RR came from Attleborough, Norfolk (see Hanham 1975: 259).
(39.) WMD lived in Northleach - see letter 192.
(40.) Although this form is typical of the West Midlands, it was also recorded in Leicestershire (LALME 392: 352), which may explain its occurrence in one of JD's letters (the Daltons, fellow staplers of the Celys, came from Leicester-see Hanham 1975: 253).
(41.) The character # before and after a word or phrase indicates interlineations.
(42.) Mainly in RCI's letters.
(43.) The in this example could also be interpreted as the third person plural pronoun 'they'.
(44.) Indefinite articles are quite common before cardinal numerals in the Cely letters, e.g., and [he] hath sent me an C li. Fl. Vppon rekenying (WLC 200: 1259), I am in whay wyth Gyshbryght Van Whennysbarge ffor an ij of yowr sarpleris (GC 109: 198).
(45.) Also other writers use this form, e.g., WLC, JD, WM, GC, and TK, usually (but not always) as the adverb 'here'.
1997 Early Modern English. (2 edition.) Edinburgh: Edinburgh University Press.
Benskin, Michael -- Michael L. Samuels (eds.)
1981 So meny people longages and tonges: Philological essays in Scots and mediaeval English presented to Angus McIntosh. Edinburgh: M. Benskin and M. L. Samuels.
Blake, Norman (ed.)
1992 The Cambridge history of the English language. Vols 1-2. Cambridge: CUP.
1996 A short grammar of Middle English. Warszawa: Wydawnictwo Naukowe PWN.
1978 Einfuhrung ins Fruhneuenglische. Heidelberg: Quelle & Meyer.
1998 "Functional selection and variation in Early Modern English", Linguistica e Filologia 6: 97-110.
1975 The Cely letters: 1472-88. London: OUP.
Klemola, Juhani -- Merja Kyto -- Matti Rissanen (eds.)
1996 Speech past and present: Studies in English dialectology in memory of Ossi Ihalainen. Frankfurt am Main: Peter Lang.
1933 Select studies in colloquial English of the late Middle Ages. Lund: Hakan Ohlsson.
1992 "Phonology and morphology", in: Norman Blake (ed.), 2, 23-155.
McIntosh, Angus -- Michael, L. Samuels -- Michael Benskin
1986 A Linguistic Atlas of Late Medieval English. Aberdeen: Aberdeen University Press.
1952 A handbook of Middle English. Baltimore: The Johns Hopkins Press.
Mustanoja, Tauno F.
1960 A Middle English syntax. Helsinki: Societe Neophilogique
1996 "Interpreting the orthographic token -vef(f) for -ve in the epistolary style of some [late] fifteenth century Londoners", in: Juhani Klemola -- Merja Kyto -- Matti Rissanen, 265-283.
Quirk, Randolph -- Sidney Greenbaum -- Geoffrey Leech -- Jan Svartvik
1985 Comprehensive grammar of the English language. London: Longman.
1999 "Pronouns in the Cely letters", Studia Anglica Posnaniensia 34: 147-169.
Samuels, Michael. L.
1981 "Spelling and dialect in the Late and Post-middle English period", in: Michael Benskin -- Michael L. Samuels (eds.), 43-54.
Scragg, Donald G.
1975 A history of English spelling. New York: Manchester University Press -- Barnes & Noble Books.
|Printer friendly Cite/link Email Feedback|
|Publication:||Studia Anglica Posnaniensia: international review of English Studies|
|Date:||Jan 1, 2001|
|Previous Article:||Metaphors we love by: on the cognitive metaphors of love from the 15th century to the present.|
|Next Article:||The construction of power and pride in the framework of political allegory in the Middle English Pride of Life. (Literature).|