Printer Friendly

The monophthongisation of diphthongs before dorsal fricatives: a corpus study.


The paper discusses the geographical distribution of the monophthongisation of (1) the diphthong [ei] (< [e:c]) before the palatal fricative [c] (i.e. [eic > i:c]) and (2) the diphthong [ou] (< [o:x]) before the velar fricative [x] (i.e. [oux > u:x] in words like high an bough. The resulting monophthongs became the input to the diphthongisation rule, a part of the Great Vowel Shift. On the basis of forty-nine Middle English poetic texts from the Chadwyck-Healey online corpus un effort is made here to establish temporal and dialectal conditioning of the change.

1. Preliminaries

The present writer's recent article (Welna 2004) on a similar topic which discussed the question of long vowel raising in loanwords like OF frere > frire (> friar) or in native words like OE (A) brer > brier (> MoE briar) contained a conclusion that in certain environments, especially before the liquid [r], the process was initiated us early as the end of the 13th century and its effects were visible in the following century, thus a long time before the late 15th century, a period generally believed to have generated the basic impulse of the long vowel raising and diphthongisation, labelled as the Great Vowel Shift. Recently a hypothesis has been launched that the two Early Middle English long low vowel raisings, [ae: > [??]:], as in OE sae 'sea', and [a: > [??]:], us in OE stan 'stone', can also be considered harbingers of the Great Vowel Shift (Lutz 2004). In what

follows attention is focussed on a completely different set of changes which increased the number of input forms to the 15th century diphthongisation of long vowels.

In Middle English, the raising of the long front close vowel [e:] to [i:] operated in the two main contexts: (a) before <z> representing the palatalised voiced velar fricative which later became the semivowel [j], as in e.g. dezen 'die', eze 'eye', fleze 'fly', leze 'lie', teze 'tie' v., etc. and (b) before <h> representing the palatalised voiceless fricative [c], as in e.g. heh 'high', neh 'nigh', peh 'thigh', and also sheh 'shy', sleh 'sly', slehp 'sleight', the last three poorly evidenced in the Middle English texts.

A parallel process affected the long close back vowel [o:] which, when followed by the velar fricative [x], was raised to [u:] in items like bozh 'bough', clozh 'clough', ynozh 'enough', plozh 'plough', slozh 'slough', tozh 'tough'. The subsequently narrowed [e:] and [o:] created the input [i:] and [u:] respectively to the diphthongisation rule which ultimately produced the contemporary output diphthongs [ai] and [au]. Since the changes before [j] and the velar fricatives, palatalised or not, went along different paths the present brief contribution examines the temporal and regional circumstances of the raising only in the context before the fricatives [c] (long [e:]) and [x] (long [o:]) in the two of the above sets of words, with particular attention paid to spelling variation reflecting the changes.

2. Traditional grammars on e- and o-diphthongisation/raising before <h>

In his account of the change, Jordan ([1925] 1974: 120) postulates that some time in Middle English "there developed a glide sound /i/ after /e:/, so that /e:i/ > /ei/ originated; heih, heigh /heic/ 'high', neigh 'nigh' ... sleigh 'sly' ...". The diphthongal forms, first recorded, according to him, in the Ancrene Riwle, were rare in other 13th century texts. "This ei only later, probably in the second half of the 14th cent., was simplified to i, so that now high, nigh, sligh arose." According to Jordan's (1974: 132) quasi-sociolinguistic hypothesis, the monophthongised forms "first arose in vulgar language, while ouh, eih were the conservative more aristocratic pronunciations." Simultaneously, forms like heegh 'high' found north of the Humber seem to show that the monophthong [e:] remained unchanged in that region (Brunner 1963: 23). Later on, Brunner hypothesises that in the continuations of words like heigh 'high' or theigh 'thigh' "the sound i ... could also have come from inflected forms which had voiced [z] instead of final voiceless [x'] before an ending beginning with a vowel...". But from Jordan's (1974: 132) account it follows that the evolution of the vowel in the sequence <oh> was parallel to that in <eh>, the former sequence yielding /oux/ which monophthongised as /u:x/ in the 14th century in items like bough /boux/ > /bu:x/, plough /ploux/ > /plu:x/, tough /toux/ > /tu:x/, inough> /i'nu:x/, clough> /klu:x/, slough> /slu:x/, etc. The earliest evidence of monophthongisation are rhymes in Sir Tristrem (c1300, Yorkshire; 3110 ynou[??]: hou)
1) 3110 And knewe it wele ynouz,
   3111 pat fre.
   3112 Hye seyd, "say me, hou ..."

and in Robert Mannyng: ynowe (< OE zenoh) : bowe (< OE buzan) (c1330). This development is also found in the area north of the Humber.

As usual, the most detailed description of this and other sound changes is that offered in Luick's Historische Grammatik, where the initial stage of the change, i.e. diphthongisation, is dated to the 12th century. In a general process of the elimination of close quality in short vowels the sequences [eic] (heigh) and [oux] (inough) developed to [[??]ic] and [??]ux] with initial open [[??]] and [[??]] respectively (Luick 1940: 422-424). Luick's interpretation of the change of the uninflected [e:c] and [o:x] can be presented as the following sequences:
2)  (a)  [e:] + [c]>  [ei(c)]<eigh> /> [i:]  heigh  > high
    (b)  [o:] + [x]>  [ou(x)]<ough> /> [u:]  bough  > b(o)ugh

where the spelling <gh> is a reflection of the palatalised or non-palatalised velar.

Simultaneously, the development of the inflected forms creates the following pattern:
3)  (a) [e:] + [3 (+V)]> [ej]  <eye> > [ei] > [i:]  hey(e) 'high'
    (b) [o:] + [3 (+V)]> [ow]  <owe> > [ou] > [u:]  bow(e) 'bough'

As regards changes under (2a, b), Jordan believed in the reality of such developments (diphthongisation followed by monophthongisation to [i:] and [u:] respectively), whereas Luick (1940: 430-431) claimed that the two long vowels in (2a, b) were the result of an analogical transfer from the respective inflected forms under (3a, b) (cf. Welna 1988). According to Luick, the change occurred in words like eye and bowes (pl. of bough), i.e. where the vowels stood originally before the fricative [y], earlier than before the fricative [x], which seems to be proved by the surviving spellings containing the cluster <gh>. Curiously, the rise of the diphthongs [ei] and [ou] in the sets above is termed "breaking" by Jones (1989) who sees analogy with the respective process operating in Old English.

The consequence of Luick's assumption of the lowering of the original short close vowels was his hypothesis that the monophthongisation [eic > i:c] and [oux > u:x] never occurred because the first element was lowered and thus the diphthong would have shared its development with that of the diphthong [ei] in nouns like eight, weight, where the sequence [eic] yielded [aic] (> MoE [ei]). The only way out is then to assume that analogy was the primary factor in the monophthongisation. However, the sequences [eic, oux] need not have contained a lowered variant and thus the subsequent monophthongisation to [i:] and [u:], as in heigh > high and plo(u)gh > plough (with [u:]) could be treated as an assimilatory process in which the diphthongs [ei, ou] became transformed into monophthongs.

Because the evidence of the change offered in classic historical grammars is scant, an attempt will be made here to supply a larger body of data. Consequently, the present study is based on a selection of 49 poetic texts from the Chadwyck-Healey on-line corpus of Middle English poetry, with the following distribution: North 15, East Midland 17, West Midland 9, Southwest 7 and Kent 1. Poetry seems to be more useful than prose as it can offer evidence of rhymes which may help identify pronunciations sometimes difficult to recover because of the guise of very often ambiguous spellings. Perhaps the evidence presented in the ensuing sections will make it possible to formulate at least a partial explanation of how, where and when the raising of [e:] and [o:] before palatal/velar fricatives took place.

The list of words whose spellings in Middle English texts are examined is short and includes a selection of items in which original [e:] and [o:] first diphthongised to [ei] and [ou] and later monophthongised to [i:] (4a) and [u:] (4b) respectively. These words are listed below with their contemporary spellings:
4)  (a) high, height, nigh, sly, sleight, thigh
    (b) bough, enough, plough (plow), slough, tough

Special attention will be given to high frequency items, i.e. high (height), nigh and enough, i.e. words which show a high degree of grammaticalisation. For the sake of simplicity the spelling variants will be represented by the following symbols:
EH     egh, eh, ez, (hegh, heght, negh, neh, slezt, etc.)
EIH    eigh, eih, eiz, eygh, eyh, eyz (heigh, heih, heiz,
       slezt, etc.)
EY(E)  ey(e), ei(e) (hey, heye, ney, neye, etc.)
I(E)H  igh, ih, ygh, yh, yz (high, nih, hygh, nyz, etc.)
Y(E)   y(e), i(e) (hye, hi, nye, nie, etc.)
OH     ogh, oh, oz (enogh, anoh, etc.)
OUH    ough, ouh, ouz, owgh, owz, (enough, ynouh, enowz, etc.)
OW(E)  ow(e), ou(e) (enow, plou, ynou, etc.)
UH     ugh, uh, uz (enugh, ynuh, pluz, etc.)
(NER)  er(e) (ner, nere, nerr)

Examples also include grammatical variants of the above, such as plurals, comparatives and superlatives, as well as segments of nominal compounds, while verbal forms are disregarded. Words in that set will be presented in the context of localised manuscripts. Wherever possible, time references reflect the date of the manuscripts rather than the date of the text composition. Because the Middle English period witnessed the rise of the form ner(e) 'near', etc., the symbol NER reflects the statistics of such forms, which very frequently coincide in the same texts with forms of the old adjective nigh.

Since, regretfully, the editors of Chadwyck-Healey on-line texts employ the grapheme <y> for <z> (yogh), as e.g. in inoyhe 'enough', for inozhe or boyhess for bozhess 'boughs' (Ormulum), such forms are interpreted in the statistics as containing the grapheme yogh.

3. North

The time range of the 15 texts from the North is around two hundred years, from c1275 to c1475. Almost all of them come from Yorkshire, the exceptions being the Cursor Mundi (Northumberland) and the Avowing of Arthur (Cumberland). The distribution of the relevant spellings is shown in Table 1 (in the Appendix).

The above can be exemplified by the following forms:

Surtees Psalter          zhegh, heghte, negh, slegh, sleght,
                         inoghe, slogh
Horn Childe              neie, nere
Castelford's Chr.        hegh, heght, neghum, ner, sleght, anogh, plogh
Cursor Mundi             hegher, heght, heye, heite, hight, negh,
                         neigh, nei, nigh, nere, slight, bogh, bugh,
                         inogh, inou, plogh, slogh, slught, togh
Richard Rolle            hegh, hyght, negh
Lay-Folks Mass Book      heghly, hight
Gast of Gy               hegh, hyght, nere
Northern Passion         heghest, high, heght, hight, negh, inogh
Sir Ysumbras             heghe, nere, ynoghe, ploghe, plowghe, sloghe
Ywain and Gawain         hegh, high, heght, hight, negh, nere, bogh,
Sir Eglamour             heghe, hye, ynoghe
Sir Perceval of Gales    heghe, highte, nere, enoghe
Avowing of Arthur        heghehest, heyer, neghe, nere, enughe
Sege of Melayne          heghe, highte, neghe, nere, boughe, ynoghe
Robert of Knaresborough  heghe, heght, enoghe, ploghe, toghe

A closer analysis of the statistical data under (6) reveals characteristic features of Northern texts. If only spelling evidence is considered one cannot fail to observe that the conservative EH forms are found throughout the whole two hundred-year period. Their numerous instances are present in the Surtees Psalter and Thomas Castelford's Chronicle, having no diphthongal rivals in the time span between Richard Rolle (c1340) and St. Robert of Knaresborough (c1475). The only text which contains a significant number of diphthongised EY forms (119) is the Cursor Mundi, with 4.5 EY to 1.0 EH ratio. Such forms are practically absent in the other manuscripts of the period. What surprises one is a very small number of the intermediate EIH forms containing a diphthong which coincide with forms containing the palatalised fricative.

As regards the monophthongised I(E)H forms they seem to represent a substantial part of all forms as they are present in several texts, frequently matching in number the non-diphthongised forms. In this respect the North differs from other dialects which, as will be seen, exhibit a substantial number of intermediate, diphthongal spellings. Likewise, the monophthongised forms with <igh> are sometimes matched by the conservative EH spellings in rhymes. Let us consider several instances from the Surtees Psalter, Thomas Castelford's Chronicle and the Cursor Mundi:
7)  (a) 175 And for pat, torne vpe in heghte.
        176 Lauerd demes folke righte.

        2995 pai thought and spake quedenes vnrighte;
        2996 Quedenes spake pai on heghte

    (The Surtees Psalter)

Rhyming forms identical with those in 175/176 are also found in the lines 3343/3344 and 4266/4267. However, other rhymes are perfectly matched, thus standing in contrast with those adduced above:
7)  (b) 3804 And .i. firstgeten sal set him reghte,
        3805 Forbi kinges of erthe on heghte
        5497 [??]ider sothlike vpstegh on heght
        5498 Kinde, kinde ofe lauerd right

     (The Surtees Psalter)

As regards other texts with apparently impure rhymes, Castelford's Chronicle contains the rhyme sleght (23043) : fight (23044), while single instances of such rhymes are found in the Cursor Mundi (heght 1339 : sight 1340) and in St. Robert of Knaresborough (syght 593 : heght 594).

As usual, in such cases three variant interpretations can be considered: (a) rhymes under (7a) are impure, i.e. the pronunciations match spellings literally; (b) IH spellings are realised as EH pronunciations, and (c) EH spellings correspond to IH pronunciations. Of these variants, hypothesis (b) is untenable since one can hardly imagine the pronunciation with long [e:] under the guise of the IH spelling. More realistic is the hypothesis involving impure rhymes which are used together with pure conservative rhymes like those under (7b), but a doubt arises why the scribe did not use EH spellings consistently. The most likely seems the hypothesis that the vowel in the derivative heght became narrowed to [i:], which does not mean that the same change affected other EH words, including the adjective hegh. If so, we deal with yet another instance of a change following the familiar route of lexical diffusion.

Because all texts, except the Cursor Mundi (Northumberland) and the Avowing of Arthur (Cumberland) come from Yorkshire, the large number of the diphthongised forms in the Cursor Mundi may reflect the specific character of the language of the region.

Due to the scarcity of data, very little can be said about changes of words representing the OH type. The survival of the OH type is evident as there are only four instances of the diphthongised forms. Worth mentioning is that three words with u-spellings reflect the process of monophthongisation. These are bugh (11721) and slught 'slough' (745; with excrescent <t>) in the Cursor Mundi, and enough in the Avowing of Arthur, again, curiously, non-Yorkshire texts.

4. East Midland

The dating of the manuscripts representative of the East Midland ranges from c1175 to c1460, the period covered embracing almost three hundred years. Geographically, there is much variation as these texts come from various areas of the region. Of the 16 texts in total, 5 come from Lincolnshire, 4 from London, 3 from Essex/Middlesex, 2 from Norfolk and 2 from Suffolk. The East Midland shows the distribution of forms as shown in Table 2 (Appendix).

The East Midland texts contain the following forms of the relevant words:
Ormulum                        heh, heyhest, neh, sleh, bozhess,
Bestiary                       heg, inog
Genesis & Exodus               hege, neg, buges, anog, ynug
Havelok the Dane               heye, ney, ner, sley, inow, plow
King Horn                      ney, nere
Guy of Warwick                 hygh, hye, neye, nere, ynogh, ynowe
Arthour & Merlin               heiye, neiye, sliye, sleiyt, ynouz
King of Tars                   heye, hiht, neye, ner
Mannyng's Handlyng             hegh, heyyest, hygh, hye, heghpe,
                               nygh, nye, nere, slygh, slye, sleghbe,
Mannyng's Story of England     heg, hey, heighte, neygh, nere, sleye,
                               sleigbe, sleytes, ynogh, ynough, inowe
Chaucer's Romaunt of the Rose  high, height, highte, neigh, nygh, nere,
                               slyghly, slye, bough, y-nogh, y-nough,
Gower's Confessio Amantis      heyher, heyeste, hyhe, heihte, nyh,
                               nerr, slyh, sleihte, bowh, ynowh, ynowe,
                               ynou, plowh, plouh, plowman
Seven Sages of Rome            heghe, hegy, high, hye, neghy, ney, ner,
                               slyeste, ynowgh, inow
Lydgate's Fall of Princes      hih, heihte, nyh, nerre
Hoccleve's Regement            hygh, hyge, heighte, highte, ney, nigh,
                               nere, ynow, plow, tow
Emare                          hygh, hye, nere, slye

The most conservative text, Genesis & Erodus, completely lacks modified forms, exhibiting the EH and OH forms (but also 2 UH forms). The chronologically earliest manuscript of the Ormulum is more innovative as it not only preserves monophthongal EH forms but also develops a substantial number of EIH forms, the ratio being 2 (EH) : 1 (EIH). Other texts from Lincolnshire contain mostly conservative EH monophthongs but, unlike the situation in the North, diphthongised EY forms are in consistent use in the East Midland from around 1300 onwards. The old EH forms are missing not only in Havelok, King Horn, Guv of Warwick, but also in the London poets Chaucer and Gower. Chaucer's The Romaunt of the Rose contains practically only monophthongised IH forms and diphthongised OUH/OW forms.

Several East Midland texts also contain what seems to be "impure" rhymes. In fact they are more numerous than those in the Northern manuscripts. They occur in Havelok, Guy of Warwick, Arthur and Merlin and in Robert Mannyng. In Havelok, hev (1071) rhymes with fri (1072), which seems to indicate the monophthongisation [ei > i:]. But rhymes of that kind are particularly revealing in Guy of Warwick; cf. the sample below:
9)  971  Then he harde horsys neve,
    972  Helmes he sawe bryght on hye.

and a set of examples involving the form heiye which stands in rhyme with:
10)  (a) heiye: seyye 2813, seye 5987, seiye 6191, 9372, cuntreie 8312
     (b) heiye: aspie 1151, 6794, vilainie 2355, curteisielie 5064,
         Vnplie 5064, compeinie 7822, 8207, 9321

Evidently, rhyme-words in (10a) indicate the diphthongal pronunciation of heiye, while rhymes in (10b) seem to indicate monophthongal pronunciation with long [i:]. Much more complicated is the interpretation of a small set of the two rhyming pairs listed under (11a, b):
11)  (a) 6845 Ac per no hadde non scaped oway,
         6846 Nadde chaunce comen fram Crist on heiye
     (b) 7167 For he smot him forbi,
         7168 & Wawain for gret heiye

     (Guy of Warwick)

If rhymes in (11) are impure, the rhyme oway : heiye seems to conceal the pronunciations [ai] : [ei], while the pair forbi : heiye contains, as it seems, the pronunciations [i:] and [ei]. But if these are true rhymes we should have [ai] rather than [ei] in (11a) because this would reflect a well-known Middle English development [ei > ai]. On the other hand, the rhyme in (11b) makes the pronunciation of heiye with long [i:] plausible as no one would expect that forbi contains final [ei], thus matching heiye.

Other apparently impure rhymes are found in the poem Arthur and Merlin which features the rhymes of arizt and sleizt (lines 5883-4, 6583-4, 9107-8). These rhymes seem to testify to the process of monophthongisation [ei > i:] before palatal fricatives, cf.:
12)  5883  Alle pailaiden doun riyt
     5884  & made ber swiche a sleit ...
     6583  per seyd Merlin anon riyt
     6584  To king Arthour al pe sleiyt ...

     9107  & made swiche defense & sleiyt,
     9108  pat y no may telle it ariyt

Likewise revealing are lines from Robert Mannyng's The Story of England which, apart from the regular rhymes, like fley : ney, also contains several curious rhymes of ney with by (8447), party (!!) (8546), maistri (12518), cry (13058), softely (!) (13544), blody (13906) and wery (15944); cf.:
13)  9791   Toward york Colgrym fley,
     9792   Arthur hym pursued faste & ney ...

     8447   Sire Eldok was euere hym by,
     8448   & other barons also were ney.

     8545   yyf Eldokes help ne hadde be ney
     8546   Hengist had had be better party ...

     12517  Arthur seide, "Godes help ys ney!
     12518  "He wot ho schal haue be maistri."
     13057  Non knew ober, bey were so ney,
     13058  But by [??]er speche or by ber cry ...

     13543  be batailes neyghed ney & nev,
     13544  Sadly passing, and softely.

     13905  "& pat bou come me so ney,
     13906  Caliborne to make blody!"
     15943  & when he was [??]e toun ought ney
     15944  He bad [??]em reste [??]at were wery ...

     (Robert Mannyng, The Story of England)

Rhymes of this kind seem to confirm the hypothesis that monophthongisation [el > i:] must have been quite advanced at the time when the manuscript of Guy of Warwick was produced and when Robert Mannyng wrote his Story of England. This process must have been well in progress in the East Midland, although its effects are not unambiguously reflected in texts from that area. There is no similar evidence of "impure" spelling rhymes as regards the respective monophthongisation [ou > u:]. But it may be a graphemic rather than phonological problem because the spelling <ou> could represent either the diphthong [ou] (eg. bowe n., now [bou]) or [u:] (bowe v., now [bau]).

5. West Midland

The texts from the West Midland (10 in total) range in time from the early 13th century to the mid-15th century, a span of around 250 years. They come from 5 counties, i.e. Lancashire (5), Staffordshire (3), Cheshire (1) and Herefordshire (1). The distribution of forms is shown in Table 3 (Appendix).

The above figures refer to the following forms of words examined:
Harrowing of Hell    heye, heiye, nere, ynoh
Stanzaic Life of     hegh, heye, hye, hie, heght, hight, negh, neye,
Christ               sleght, inogh, plogh
William of Palerne   heiyh, heiz, heye, high, neyh, neiz, ney, nyh,
                     sleiliche, thighs, bowes, inow
Sir Gawain           heye, highe, heyt, neghe, in-noghe, in-nowe
Piers the Plowman A  heichliche, heiy, heiye, neih, nere, i-nouh,
                     i-nowe, plouh, plouy, slouthe
Piers the Plowman B  heigh, heyy, hiegh, neighe, neiye, nyeghe, nere,
                     ynogh, ynowe, plough-, plow, plou
Destruction of Troy  hegh, heyer, high, hyest, heght, hight, negh, ney,
                     nighe, ner, sleghly, slighly, slyeste, sleght,
                     theghes, ynogh, ynugh, enow, bogh, bowes, ploghe,
Sir Amadace          nere, ynuyhe
Sir Torrent          heygh, hey, high, hyght, i-nough, i-nowe
Sir Degrevant        heghe, heghte, highte, nere, boghe, ynoghe,
                     i-nowe, ploughes

The statistics testifies to the survival of the original forms with long [e:] and, to a limited extent, forms with [o:] in Cheshire (The Stanzaic Life of Christ) and OH forms in Sir Gawain, while in other texts forms with the diphthong [ei] and [ou] predominate. As compared with the East Midland data, the evidence of the monophthongisation to long [i:] is rather limited, while monophthongisation to long [u:] is confined to UH forms in Sir Amadace. Curiously, The Stanzaic Life of Christ, the earlier mentioned text from the middle of the 14th century, contains a significant number of strongly contrasting monophthongal forms, with long [e:] and long [i:] (37: IH + YE). The most complicated distribution is found in the relatively late manuscript of The Destruction of Troy (c1400, Lancashire) where numerous conservative EH and OH forms coincide with the diphthongised EY, OW and monophthongised IH forms.

There is little evidence of line-final rhymes because the most important West Midland poets employed alliteration or, at best, techniques, like the author of Sir Gawain and the Green Knight. In the West Midland, the diphthongal forms were particularly frequent between 1350 and 1400.

6. Southwest and Kent

The Southern region is represented by 7 Southwestern poems (Worcestershire 3, Gloucestershire 2, Surrey 1, Devonshire 1) and one text from Kent by William of Shoreham. The time range is rather narrow as it only embraces the period from around 1200 to 1380, i.e. less than two hundred years (see Table 4, Appendix).

These texts contain the following forms of words subject to analysis in the present study:
Layamon                    heh, heihliche, heyes, neh, neih, nih,
                           nere, inoh, inowe
Owl and the Nightingale    heh, heie, neh, ney, inoh
Sir Orfeo                  heiye, nere
Castel of Love             heiynesse, hyhhe, neih, nyyh, sleih-, inouh
Myrour of Lewed Men        hegh, ynogh
Robert of Gloucester Chr.  heiy, heie, hey, neiy, nei, nye, inowe, plou-
Sir Ferumbras              heghe, heye, neghy, ney, ner, slegh, sley-,
                           sleythe, ynow
William of Shoreham        heiye, hey, hy, ney, nere, y-now

Only three manuscripts (Layamon, Robert of Gloucester and Sir Firumbras) contain sufficient data to make any generalisation possible. As could be expected, the earliest of them, Layamon, exhibits prevalence of the monophthongal EH and OH forms, while Robert of Gloucester's Chronicle and Sir Ferumbras chiefly contain diphthongised forms. Because the Southwestern texts are relatively early they show very few traces of monophthongisation to [i:]. The only Kentish text, the poems by William of Shoreham, exhibits a similar distributional pattern with preference for diphthongal EY forms.

The only rhyme which might stir up controversy is that found in Robert of Gloucester's Chronicle, where hey marches ysay. This may reflect the incipient stage of the lowering of [e] in [ei] already mentioned in the section on the East Midland; cf.:
16)  5745 po poyte him in is slep pat he an hey tre ysay
     5746 Stonde bere bi syde him & as he bi huld an hey
     (R. Gloucester's Chronicle)

Summing up, one can state that data from the South also confirm that the year 1300 separates the texts with the old monophthongs retained from those which contain diphthongs. The monophthongisation of the latter on a larger scale was practically completed in the 15th century when the long monophthongs [e:, o:] formerly standing before palatal and velar fricatives became [i:, u:], thus becoming an input to the Great Vowel Shift raising rule. The adverb enough, the adjective tough and the noun slough, now [sl[??]f], failed to diphthongise as their long vowel [u:] became shortened before the initial stage of the Shift.

7. Conclusions

The evidence of spellings from poetic texts representing all dialects allows one to formulate the following generalisations:

1) EY

North of the Humber the diphthongised forms EIH or EY are only found in the Cursor Mundi (Northumberland, c1340), being almost absent in poetic texts from Yorkshire. But EIH forms are registered very early in Lincolnshire (Ormulum, c1175, East Midland). Only at the turn of the 13th century (Havelok) and before mid-14th century (Arthur and Merlin, Robert Mannyng) the EY forms become frequent. After 1350, the EH forms become extremely rare in the West Midland, except in The Destruction of Troy (c1400, sLancashire), a poem which shows a large variety of all possible forms, with the prevalence of those containing the old monophthong. In the Southwest, the conservative forms prevail only in the early texts (Layamon, c1200, Worcestershire), but EY forms replace them from c1300 onwards (Robert of Gloucester, Sir Ferumbras). The only Kentish text examined, William of Shoreham (c1315) exhibits EY throughout.

2) EIH EY > IH

The monophthong in the North is rather rare, although the Cursor Mundi contains 13 tokens exhibiting monophthongisation. In the East Midland, the first more numerous instances of the monophthong [i:] are found in Robert Mannyng's Handlyng Synne (1338, Lincolnshire, 45 IH forms). Curiously, the data in Mannyng's Story of England are drastically different (117 instances of EIH EY). Monophthongisation is regular in Chaucer and to a large extent in Gower, Hoccleve and Lydgate.

3) OH > OUH

The evidence of the diphthongisation OH > OUH is rather meagre. Forms unambiguously confirming the existence of the diphthong [ou] in the North are lacking. More numerous examples are found in the works of poets from London (East Midland), especially Gower, as well as Chaucer, Hoccleve and Lydgate. In the West Midland their number is high in Piers the Plowman (Versions A and B, of 1367 and 1377 respectively). The Southwest exhibits larger numbers of such forms after 1320 (Robert of Gloucester and Sir Ferumbras).

4) OUH OW > UH

The evidence of this change in dialects is even less than meagre, considering the ambiguous spelling <ou> which could represent a diphthong or a monophthong. Transparent UH spellings are rare in the North and the East Midland. The only text featuring more than mere isolated cases of such spellings is Sir Amadace from the West Midland (c1420, Lancashire, 9 tokens). No such forms have been identified in the Southern texts.

doi: 10.2478/v10121-009-0013-6


Bauer, Gero--Dieter Kastovsky (eds.) 1988 Luick revisited. Tubingen: Gunter Narr

Brunner, Karl 1963 An outline of Middle English grammar. (Translated by G. K. W. Johnston.) Oxford: Blackwell.

Chadwyck-Healey 1992 English poetry full-text database. Cambridge: Chadwyck-Healey Ltd.

Jones, Charles 1989 A history of English phonology. London: Longman.

Jordan, Richard [1925] 1974 Handbook of Middle English grammar: Phonology. (Translated and revised by E. J. Crook.) The Hague: Mouton.

Luick, Karl 1940 Historische Grammatik der englischen Sprache. Leipzig: Tauchnitz. Lutz, Angelika

2004 "The first push: A prelude to the Great Vowel Shift", Anglia 122: 209-224. Welna, Jerzy

1988 "Historische Grammatik and the Middle English diphthongal systems", in: Bauer, Gero--Dieter Kastovsky (eds.), 419-433.

2004 "Middle English e-raising: A prelude to the Great Vowel Shift", Studia Anglica Posnaniensia 40: 75-83.

* This is a revised version of the paper delivered at 44th International Congress on Medieval Studies. Western Michigan University, Kalamazoo, May 7-10, 2009.


University of Warsaw
Table 1. The distribution of the relevant forms in the North

                                         EH   EIH   EY(E)   I(E)H

c1275 Surtees Psalter            (Yks)   98
c1320 Horn Childe                (Yks)                2
c1327 Castelford's Chr.          (Yks)   35
c1340 Cursor Mundi               (Nhb)   28    2     119     13
c1340 R. Rolle                   (Yks)   6                    1
c1390 Lay-Folks Mass Book        (Yks)   4                    1
14c Gast of Gy                   (Yks)   1                    1
14c Northern Passion             (Yks)   7                    8
a1400 Sir Ysumbras               (Yks)   7
c1400 Ywain and Gawain           (Yks)   7                    5
c1410 Sir Eglamour               (Yks)   3
c1410 Sir Perceval of Gales      (Yks)   6                    4
c1420 Avowing of Arthur          (Cum)   2            1
c1450 Sege of Melayne            (Yks)   6                    3
c1475 Robert of Knaresborough    (Yks)   7

                                 Y(E)   OH   OUH   OW(E)   UH   (NER)

c1275 Surtees Psalter                   4
c1320 Horn Childe                                                 5
c1327 Castelford's Chr.                 35                        3
c1340 Cursor Mundi                      39           2     2     57
c1340 R. Rolle
c1390 Lay-Folks Mass Book
14c Gast of Gy                                                    7
14c Northern Passion                    3
a1400 Sir Ysumbras                      6     1                   1
c1400 Ywain and Gawain                  8                        16
c1410 Sir Eglamour                1     4
c1410 Sir Perceval of Gales             1                         7
c1420 Avowing of Arthur                                    1      5
c1450 Sege of Melayne                   1     1                   1
c1475 Robert of Knaresborough           12

Table 2. The distribution of the relevant forms in the East Midland

                                           EH   EIH   EY(E)   I(E)H

c1175 Ormulum                (Linc)        68   38
c1220 Bestiary               (Nort)        2
c1250 Genesis & Exodus       (Nort)        18
c1289 Havelok the Dane       (Line)                    31
c1300 King Horn              (Ess/Middx)                3
c1314 Guy of Warwick         (Sut)                      1       1
c1330 Arthour& Merlin        (Ess/Middx)               49
c1330 King of Tars           (London)                  13       1
c338 Mannyng's Handlyng      (Line)        6            1      31
c1338 Mannyng's Story of
  England                                  1    20     97
c1400 Chaucer's Romaunt      (London)            2             62
c1400 Gower's Confessio      (London)           37      3      236
c1425 Seven Sages of Rome    (Ess/Middx)   9            3       3
c1438 Lydgate's Fall of
  Princes                    (Sut)               3             74
a1440 Hoccleve's Regement    (London)            1      1      11
c1460 Emare                  (Line)                             2

                             Y(E)   OH   OUH   OW(E)   UH   (NER)

c1175 Ormulum                       97
c1220 Bestiary                      1
c1250 Genesis & Exodus              8                  2
c1289 Havelok the Dane                           7            5
c1300 King Horn                                               1
c1314 Guy of Warwick          1     11           5           42
c1330 Arthour& Merlin                     2      1
c1330 King of Tars                                            4
c338 Mannyng's Handlyng       14                 7           11
c1338 Mannyng's Story of
  England                           2     1      6           55
c1400 Chaucer's Romaunt       1     1    11      3           24
c1400 Gower's Confessio                  97     15           11
c1425 Seven Sages of Rome     4           1      4           44
c1438 Lydgate's Fall of
  Princes                                                    44
a1440 Hoccleve's Regement                        9            1
c1460 Emare                   3                               1

Table 3. The distribution of the relevant forms in the West Midland

                                           EH    EIH   EY(E)   I(E)H

c1310 Harrowing of Hell          (Staf)                  2
c1350 Stanzaic Life of Christ    (Chs)     84            5       5
c1350 William of Palerne         (Heref)         26     20       3
c1360 Sir Gawain                 (Lane)     3           25       4
c1367 Piers the Plowman A        (Staf)           5     11
c1377 Piers the Plowman B        (Staf)          33      4       8
c1400 Destruction of Troy        (Lanc)    131           7      48
c1420 Sir Amadace                (Lanc)
c1435 Sir Torrent                (Lanc)           1      2       5
c1440 Sir Degrevant              (Lanc)     4                   10

                                 Y(E)   OH   OUH   OW(E)   UH   (NER)

c1310 Harrowing of Hell                 1                         1
c1350 Stanzaic Life of Christ     32    6
c1350 William of Palerne                             2           24
c1360 Sir Gawain                        11           1
c1367 Piers the Plowman A                    21      3            2
c1377 Piers the Plowman B               1     2     64           13
c1400 Destruction of Troy         7     38          29     1      5
c1420 Sir Amadace                                          9      9
c1435 Sir Torrent                             6      1
c1440 Sir Degrevant                     3     1      2            7

Table 4. The distribution of the relevant forms in the
Southwestern and Kentish

                                             EH    EIH   EY(E)   1(E)H

c1200 Layamon                      (Worc)    107    9      2       1
c1275 Owl and the Nightingale      (Sur)      6            4
c1300 Sir Orfeo                    (Glouc)                 4
c1320 Castel of Love5              (Worc)           5              1
c1320 Myrour of Lewed Men          (Worc)     8
c1330 Robert of Gloucester Chr.    (Glouc)          2     299
c1380 Sir Ferumbras                (Dev)      6           157
c1315 William of Shoreham          (KENT)                 24

                                   Y(E)   OH   OUH   OW(E)   UH   (NER)

c1200 Layamon                             24           2           12
c1275 Owl and the Nightingale             5
c1300 Sir Orfeo                                                     3
c1320 Castel of Love5                           2
c1320 Myrour of Lewed Men                 5
c1330 Robert of Gloucester Chr.     5                 32
c1380 Sir Ferumbras                                   42
c1315 William of Shoreham           2                  2           21
COPYRIGHT 2009 Adam Mickiewicz University
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:LINGUISTICS
Author:Welna, Jerzy
Publication:Studia Anglica Posnaniensia: international review of English Studies
Article Type:Report
Geographic Code:4EUUK
Date:Jul 1, 2009
Previous Article:Intertextual transactions in American and Irish fictions.
Next Article:Middle English medical recipes: a metadiscursive approach.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters