Remerciez-le!

Remerciez @Admin pour avoir partagé cet document gratuitement, de la manière la plus simple, en partageant sur les réseaux sociaux.

SBORNf K PRACf F I L O Z O F I C K E F A K U L T Y BRNF.NSKE U N I V E R Z I T

SBORNf K PRACf F I L O Z O F I C K E F A K U L T Y BRNF.NSKE U N I V E R Z I T Y STUDIA MINORA F A C U L T A T I S PHILOSOPHICAE UNIVERSITATIS BRUNENSIS K 7 (1985) - BRNO STUDIES IN E N G L I S H 16 NOTES ON THE HISTORY OF WORD DIVISION IN ENGLISH Josef Hladky" Word division at the end of a line is met with in nearly every written or printed text and the rules governing it in English obviously differ from the rules in other languages. The present notes, however, are not concerned with the contemporary rules (see Hladky 1984). Instead, a few observations will be offered on word division in some older texts. (For the sake of brevity, the texts will be referred to by the year in which they were written or the year of publication). 970-990 The oldest text in which the boundaries of word division have been studied is the Exeter Book. When looking at the ends of lines, however, we must bear in mind the fact that word boundaries within the lines of the manuscript reflect the practices that differ from later usage. Compounds, prefixed and derived words are sometimes written as two words' (ael mihtig, sorg lease, a toyrgde, ge dringari). Grammatical suffixes are also found separate from the stems (us ic, dryht nes). On the other hand we can find prepositions, pronouns, the conjunction ond and the negative particle ne written together with the following word (wende toworulde Upon sewitfia song: hewaes upp hafen enfila faedmutn; Crist 650—1). This usage (though not systematically applied) and the fact that no hyphen was used mean that a case of division at the end of a line may not in fact be a case of division at all, because the word might well have been written in two parts even within the line. But as we are not concerned here with any statistical analysis of the frequency of word division in 970—990, we may still use all words divided at the ends of lines in the search for the boundaries of division. From what has been said above, it is clear that one of the main word-division principles in 970—990, both at the ends of lines and within the lines, is the morpho logical principle. Thus we find divisions like geworh-te, gehyrw-don, dryht-ne, sylf-ne, min-re (all these examples and the examples used further on have been drawn from division at the ends of lines only). The morphological principle, however, does not operate with all endings. We find divisions at the end of an open syllable (this rule will be referred to as the CV-CV rule), e.g. myce-le, gie-fe bto-was, and between two consonants (the C-C rule), e.g. lon-ge, gedrin-gan, bear- 73 -nes, singen-de, leofes-tan, stron-gum, swear-tan. There are cases where the morpho logical rule and the C-C rule overlap, e.g. the above-mentioned min-re, or seles-te, geslaeh-te. The C-C rule is not without exceptions either. If the second consonant is a liquid, the two consonants are not separated. Thus we get boundaries like sa-wla, fu-gles (but also fug-les), deo-fles, fro-fre, bi-trum, ty-dran, hleo-prade, fi-brum,fae-grestum (but also hleop-res,fip-rum,faeg-ran). The consonant + liquid pairs (the C L pairs, for short), are kept together even in groups of three consonants, e.g. eal-dre, wul-dres (but also, exceptionally, wuld-res), ear-gra, tem-ples. There is, however, a rule stronger than the C L rule in groups of three consonants: if there is an ng sequence, it is not separated, e.g. eng-la, tung-lu,fing-ras. Similarly, s and / are not separated: gefaest-nad, blost-mum, for-ste, dyr-stig. Two final points should be made about the operation of the CV-CV rule in 970—990: (i) the CV-CV rule applies even if the word begins with a vowel (we get a V-CV rule), e.g. o-per, a-num, ae-peling, and (ii) the CV-CV rules does not decide the division boundary if there is an x involved, e.g. weax-an. (Th^re is another case of division after x, wrix-lan, but that might be also explained by the C-C rule.) 970—990, like some other later texts, contains instances of division inside syllables: dre-am, bartholome-us. This division may indicate full pronunciation of the two vowels. 1483 1483 comprises six Latin letters exchanged between Pope Sixtus IV and John Mocenigo, the Doge of Venice, as printed by William Caxton. Although in Latin, the text of the letters deserves mentioning as the only available specimen of 15th-century printing (and, unfortunately, also the only specimen of texts printed by Caxton). Division is very frequent in 1483; depending on the space available, this is either shown by a division mark or simply not indicated. (It seems that the unequal spacing familiar in modern times was not technically possible in the early days of printing.) The rules in 1483 can be arranged in the following way: (i) do not divide before x (predix-imus, vex-et, max-ime, but also maxi-me according to rule (ii), the CV-CV rule); (ii) divide after an open syllable, even if not complete (the CV-CV rule: diui-na, a-lios); (iii) do not separate a liquid from the preceding consonant (the C L rule: cala-brie, perpe-trate, pro-fligatus, sa-cre, exem-plo, con-tra, nos-tri); (iv) divide between two consonants (the C-C rule: an-no, ur-bis, relic-to, res- -pondebas); there are a few infrequent exceptions to this rule (indi-gnatus, re-sponsione: the second may be a case of division at a morphological boundary). There are not enough divisions of three or more consonants in 1483 and so no other rules for consonantal clusters besides rules (iii) and (iv) can b; found. 1592 The first English printed book available for analysis is the 17-page story The Sea-Mans Triumph. The rules are the same as rules (ii) to (iv) in the Latin 1483, with an extension to rule (iii). The rules are stronger than any concern for a mor phological boundary: (i) open syllable: ano-ther,go-uerned, a-gainst, ta-king, excee- 74 -deth; (ii) preserve a C L pair and preserve ct and st: trou-ble, gen-tleman, hum-bly, re-steth, vi-ctory (the rule about ct and st may have a connection with ligatures used in manuscripts and then in printing); (iii) divide C-C: wil-ling, accep-ted. 1623 The First Folio of 1623 is a text long enough for a thorough analysis of word- -division boundaries. The rules for the determination of the boundaries can be summarized as follows (the sequence of rules is deliberate: it indicates a hierarchy whereby rule (i) does not give way to rules (ii) to (v), etc.): (i) do not split st: question, distance, distinguish, mistook, arrested; (iia) preserve the suffix -tion (ac-tion, expecta-tion) and the following prefixes: con-flict, dis-cover, ex-alted, in-vre, per-aduenture, respect, sub-urb; (iib) preserve recognizable words in compounds: know-ledge, with-out, vp-on, Eng-land (and Eng-lish); (iii) divide after an open syllable, even if not complete: lea-ther, o-uer, ano-ther or a-nother (rule (iib) does not operate here), ha-uing, obtai-ned, reaso-nable, assu-rance, roa-rers, dange-rous; (iv) preserve ct and C L : do-ctor, distra-cted, pra-ctised; no-ble, trem-ble, cir-cle, peo-ple, gen-tleman, Pem-broke, se-cret, chil-dren, de-gree, par-tridge; There are exceptions to this rule in 1623: pic-ture, perspec-tively and doub-let. They may be cases of division under rule (iia) but this explanation cannot be supported by any other occurrence of -ture, -live and -let as recognizable morphemes. (v) divide between two consonants: thin-king, han-ging (but also stink-ing, hang ing), defor-med, gol-den, of-ten, hus-band, bas-ket, and even rec-kon with a digraph. The above rules do not preserve any other suffix than -tion. Even with the most frequent suffixes like -ing, -ed, -able, the boundary in 1623 is decided by one of the three last rules. Not included in the survey is the division after w {draw-er, cow-ardise, stew-ard, but also ste-ward) because the occurrence of only three words does not guarantee a safe rule (there was no w in the pronunciation of these words at that time, and this may be reflected in the division). 1640, 1642, 1644, 1661 The rules in these short texts are similar to those in 1592: (i) divide after an open syllable, even if not complete {transla-ted, mee-ting, ri-chest, e-lection), (ii) preserve C L , ct, st, tw (resem-ble, bre-thren, do-ctor, master, be-tween), (iii) divide C-C {transpor-ted, stret-ched, indig-nation, lan-thorne). Owing to the limited number of cases of division it is impossible to decide whether di-gression, con-fer, com-pany, im-pair, dis-charge, obscene are cases of division made solely according to rules (ii) and (iii) or whether they are cases of division at the morphological boundary. There are no instances like in-vre in 1623 to confirm the existence of the morpho logical rule. A similar lack of clarity as to which of the rules is hierarchically higher prevails in the case of belong-ing and Eng-land: belong-ing is the only case among the -ing forms where the stem and the suffix are preserved (in contrast to mee-ting OT spea-king) and Eng-land 75 is the only instance of etymological division which does not coincide with points of division according to the C L and C-C rule uploads/Management/ word-division-in-english.pdf