No CrossRef data available.
Published online by Cambridge University Press: 22 April 2026
I intended a short, inoffensive article on ME ū, devoted chiefly to the Early American au-spellings. The present monograph has grown out of unavailing struggles to retain my original intention and the accepted phonological theory in face of the recalcitrant rhymes and spellings of Professor Miles L. Hanley's Wisconsin Collection. Various extended analyses of minor factors that figure largely in the following pages were needed, in the first place, to settle my own misgivings; they are retained in the hope that they may perform a like service for the reader. I am fully conscious of the many points on which either my imagination or my methodology may have failed me, and it is with certain inward doubts that I present my tentative conclusions for criticism, correction, and amplification.
1 Here and elsewhere a and c immediately preceding a date = ante and circa.
2 Miss Hall's collections include spellings from several unpublished letters as well as from the materials in E. C. Barker's edition (Washington, D. C., 1919, 1922; Austin, 1926).
3 All from the Wisconsin Collection (as of December 15, 1938). See §2 below. Since the completion of this Monograph, the Collection has been greatly enhanced in scope, and now (October 1, 1939) contains materials from 260 volumes of verse.
4 Many of the following works are discussed, for her own special purposes, by Mrs. E. K. Sheldon in her unpublished dissertation, Standards of English Pronunciation according to the Grammarians and Orthoepists of the 16th, 17th, and 18th Centuries (University of Wisconsin, 1938). For recognition of the importance of some of the minor English orthoepists, for my approach to Sheridan and Walker, and for a systematic preview of English orthoepy in general, I am greatly indebted to Mrs. Sheldon's suggestive study.
1 Cf. Harold Whitehall, Some Fifteenth Century Spellings from the Nottingham Records 66–8; in Essays and Studies in English and Comparative Literature by Members of the English Department of the University of Michigan (Ann Arbor, 1935).
2 Bellot (1580), Wallis (1653), Mason (1622), and Cooper (1685) indicate either [ɔu] or [əu]. The latter is probably the sound described by Wilkins (1668), by Tiffin (1751), and by Elphinston (1765). Noah Webster (Dissertations on the English Language 86, Boston, 1879) first favors [ɔu]; but in the American Spelling Book 14 the first element is ‘a sound peculiar to itself’ ([Λ] or [ə]), and in the Dictionary (1828) the diphthong has become a sound interpreted as [au]. For further details see Otto Jespersen, A Modern English Grammar 1.235 (Part 14, Heidelberg, 1928); G. P. Krapp, The English Language in America 2.192–201 (New York, 1925); and Eilert Ekwall, Historische neuenglische Laut- und Formenlehre2 65–6 (Berlin and Leipzig, 1922). More recently, both Ellis (On Early English Pronunciation, with Special Reference to Shakespere and Chaucer; EETS, ES, London, 1867–89) and Sweet have described the Rec. St. British sound as [əu], though the latter recanted in favor of [αu] in 1908 (The Sounds of English, an Introduction to Phonetics 71, Oxford). Daniel Jones regularly writes [au] (An English Pronouncing Dictionary, using the Alphabet of the International Phonetic Association2, London, 1926) and Jespersen 1.236 considers [au] ‘a fairly accurate rendering,’ but prefers to analyze the first element as 'intermediate between the [ₔ] of bird and the [Λ] of cut.' For the American [αu / au] (J. S. Kenyon, American Pronunciation6 206, Ann Arbor, 1935) no such difficulty seems to exist, yet a type [Λu] or [əu], described by Willard (General Class Book, Greenfield, Mass., 1840) and by E. F. Shewmake (English Pronunciation in Virginia, Davidson, N. C., 1928) still causes trouble: cf. Cabell Greet, A Record from Lubec, Maine, and Remarks on the Coastal Type, in American Speech 6.397 (1931). In discussing these various opinions I retain [u], used by the writers, rather than the specific [u].
3 Used suggestively, however, in Krapp 2.193.
4 Gill's geaun and gaun (1621) refer to the ‘advanced’ Northeast Midland sound, as do the au dialectal spellings in Shadwell and Otway. Cf. Harold Whitehall, Thomas Shadwell and the Lancashire Dialect 10.261–78; in Essays and Studies in English and Comparative Literature, University of Michigan Publications (Ann Arbor, 1933).
5 Cf. R. E. Zachrisson, The Pronunciation of English Vowels 1400–1700, 78 (Göteborg, 1913).
6 Anders Orbeck, Early New England Pronunciation (Ann Arbor, 1927); Claude M. Simpson, Early Rhode Island Pronunciation 1636–1700 (American Dialect Society Microfilm Monograph No. 1; Document 1013, published on 35 mm. film by Am. Documentation Inst., Washington, D. C., July, 1937); Mrs. Matthias' unpublished Mount Holyoke Dissertation is incorporated in Professor Hanley's Wisconsin Collection of Early American Rhymes and Spellings, mentioned below. These three works are used here principally on grounds of accessibility. There are several other meritorious but unprinted dissertations that might well have found use in this monograph.
7 H. C. Wyld, A History of Modern Colloquial English3 (London, 1936); Zachrisson 78–80; William Matthews, The Vulgar Speech of London in the XV-XVII Centuries, NQ January 2 to April 3 (1937); Sailors' Pronunciation in the Second Half of the Seventeenth Century, Anglia 47.193–251 (1935); Sailors' Pronunciation 1770–83, Anglia 49.72–80 (1937); Some Eighteenth-Century Phonetic Spellings, RES 12.1–31 (1936); Some Eighteenth-Century Vulgarisms, RES 13.1–19 (1937); Variant Pronunciations in the Seventeenth Century, JEGP 37.189–206 (1938); Harold Whitehall, A Short Study of the Vowels in the Language of the Shuttleworth Accounts, PQ 10.10–26, 269–76, 384–91 (1931); 11.293–302 (1932); WNR 3.61–71; Helge Kökeritz, The Phonology of the Suffolk Dialect (Uppsala, 1932).
I have used, more sporadically, Asta Kihlbom, A Contribution to the Study of Fifteenth Century English (Uppsala, 1926).
8 At the time of writing (December, 1938), the collection of English rhymes had been extended to a total of 470,000 rhyme-slips from 125 volumes, of which several had been covered in entirety. This extension was made possible by WPA assistance (Project No. 465–53–3-152. WP 7718–2). A large mimeographed volume containing a sample of the English rhyme-evidence together with American rhymes and occasional spellings is to be issued shortly by Professor Hanley under the title English and American Pronunciation, 16th to 18th Centuries.
9 For the analysis of these collections I am indebted to two M. A. dissertations written by my students at the University of Texas and as yet unpublished: The Phonology and Orthography of the Austin Papers (Austin, Texas, 1938) by Leila Maude Hall; and The Massie Papers c1800–1859: Linguistic and Historical Materials from a Virginia Family (Austin, Texas, 1938) by Lee Lloyd Pickett. Both collections of documents are among the Archives in the Library of the University of Texas.
10 Kenyon 207.
11 Cf. Karl Luick, Historische Grammatik der englischen Sprache, 564 anm. 4, and 562 (Leipzig 1929).
12 Ellis 1153. So apparently does Webster (American Spelling Book 14) when he analyzes the sound of ow or ou as composed of ‘a sound peculiar to itself ([Λ] or [ə]?) and that of oo.‘ But it is useless to speculate about the meaning intended by writers who have not considered the problem from this angle: the basic assumption, that the development is the development of the first element of the diphthong only, underlies all the arguments advanced by Wyld, Zachrisson, Sweet, and other writers, and it is notable that even so careful a work as Harold Orton's Phonology of a South Durham Dialect 233 (London, 1933), describing one of the earliest stages of diphthongization, omits any mention of the relative length of the two elements in the diphthong [öu].
13 Kökeritz 64 seems to be on the track of the hypothesis given here, but does not state it. His explanation of the development of ME ī, a parallel case, is undoubtedly correct.
14 R. E. Zachrisson, The English Pronunciation of Shakespeare's Time as Taught by William Bullokar 61–3 (Uppsala and Leipzig, 1927).
15 With his customary acumen Luick (Gram. 564) seems to have assumed, though he does not state, some such conclusion.
16 Wyld 231: ‘The other English grammarians of this century [i.e., the 16th] are so obscure on this vowel that it is a mere waste of time to try to wring some meaning out of their accounts.' It would be interesting to estimate how far the charge of conventional purism is based upon the Grammarians’ accounts of the sounds arising from ME ū and ME ī.
17 2.193.
1 Zachrisson 54 and 208.
2 Possibly patterned after abound vb.
3 John Jones, Practical Phonography (1701) 87, edited by E. Ekwall; Neudrucke frühneuenglischer Grammatiken 2 (Halle, 1907). Ekwall interprets the sounds as [o'u] and [əu].
4 These five and such forms as coult, boult, mould beside mold, poultice beside pultess, bowl, prowl beside boll, proll, chowse, chouse beside choose, etc., were apparently in good repute as conventional spellings from the 16th to the 18th century. They have often been construed as genuine phonetic spellings. They were, of course, semi-phonetic in origin. Houp, hout are historic forms.
5 This variation is specifically mentioned by Sharp (1767), Sheridan (1780), Perry (1788), and Nares (1784), as well as by earlier authorities and by Walker.
6 See Walker's Dictionary under lome, loam.
7 Wisc. has a strangely parallel a leudes ‘alludes’ 1756.
8 But cf. Zachrisson, Bullokar 125–50, where only dewe and hew of the Kihlbom-Binzel spellings are accepted. Tew and dewe possibly reflect the East Anglian and New England development to [ᵾ].
1 Regarded by Zachrisson (and Kihlbom) as Northern forms.
2 The orthoepists reproduce these forms: Elphinston (1765) noo, toon, doo, dook, etc.; Sheridan pre-zhoom, re-zhoom, etc.
3 All Kihlbom's examples are perhaps too early for consideration; but note that they occur in the texts side by side with the ou-forms given in §11 above.
4 Kihlbom mentions boold, hoolde, goold from the same source; other spellings in A. Binzel, Die Mundart von Suffolk in frühneuenglischer Zeit (Darmstadt, 1912). Simpson 38 does not list his oo-forms.
5 Possibly Kihlbom's foolkes (Paston L) might be considered with these.
6 I have not included chuse, etc. (Wisc., Simpson, Orbeck, Sharon, Austin, MVS, MSP7, Zachrisson, Wyld, etc.), nor the doubtful Remuffing, remuving, impruvement, impruffement (Sharon, Wisc., Simpson, Orbeck, MVS, and most other sources).
7 Zachrisson 131; Theo Spira, Die englische Lautentwicklung nach französischen Grammatiker-Zeugnissen (Strassburg, 1912). Particularly notable for their clear statement of this identification are Eliot (1593), WodRoephe (1625), and Sherwood (1632).
8 For Mulcaster and Butler, see below.
9 Cooper (45) suggests spellings boord [buᵊd], foorth [fuᵊθ], proov [pruv], stoop [stup] for the currently preferred forms board, forth, prove, stoup.
10 For would, should, could, bishop (booshop), and many other words.
11 Among others for crouch [krut∫], slouch [slut∫], and vouch [vut∫]. 12Especially Loughton (1735), Watts (1721), and Farro (1754); Sheridan has pre-zhoom, re-zhoom as preferred pronunciations, and Walker indicates loom [lum] for loam, troonk [trunk] for trunk in discussing local variants.
13 American Spelling Book 51 (Boston, 1798). To the usual wool, wood, good, hood, foot, stood, blood, flood, soot Webster adds the shortenings before -k.
14 On the development of ME u and ō 1 shortened, cf. Wilhelm Horn, Die Entwicklung des mittelenglischen kurzen u im Neuenglischen, in Englische Studien 60.128–9 (1925–6); Zachrisson, Bullokar 56–60; Zachrisson, Notes on the Essex Dialect and the Origin of Vulgar London Speech, in Englische Studien 59.346–60 (1925); Zachrisson, The Early English Loan-words in Welsh and the Chronology of the English Sound-shift, in Studies in English Philology, A Miscellany in Honor of F. Klaeber 306–7 (Minneapolis, 1929); Luick Gram. 656–63; Wyld SH 183–5; Kökeritz 164–6.
Some kind of unrounded sound, almost certainly [ɜ], is indicated by the three premier phoneticians, Wilkins, Tiffin, and Holder, but for the shortenings of ō 1 a long range of orthoepists, including Hart, Smith, Sherwood, Miege, Cooper, Butler, Jones, The Anonymous Instructor of the Palatines, and Greiffenhahn recommend either [u] or [u]. There is little doubt that [u] was an important variant, if not the principal form, until after 1800.
15 Charles Butler, English Grammar (1634) 28, edited by A. Eichler (Halle, 1910). For typographical reasons I have not joined Butlers' ligatures oo and ee nor reproduced his various crossed consonants. His inverted apostrophe used as a sign of length is given, although in most cases a final -e would have exactly the same effect.
16 The preceding passage, too long to be quoted here, may be summarized as a complaint against the use of ou in bloud, floud, courage, scourge, flourish, nourish, young, youth, woulf, double, trouble (for which Butler would write oo and u) beside loud, proud, cour, scour, mound, coul ‘cowl’, scoul, doubt, trout.
17 Cf. Jespersen 1.241. MnE ousel [uzl] is an outstanding survival of this tendency.
18 So Bullokar, Gill, Butler, Hodges, Cooper, Jones, Lediard, and some later writers.
19 Cf. Jespersen 1.236; Luick Gram. 567.
20 Luick Gram. 615 and Bullokar, Hart, Wallis, Butler, Cooper, Writing Scholar's Companion, etc.
21 See the list in Jones and cf. Luick Gram. 567, 569. Jones' list (85, oo-ou) is quite extensive.
22 Ekwall, Intro. 187, and Isaac Watts, List 8.
23 Jones 85 (oo-ou), and Jespersen 1.241.
24 Luick Gram. 570; Farro (1754) 76, etc.
25 Walker condemns gould and ould as vulgar; Isaac Watts definitely indicates [u] for both of them. Hart has kuld for what may be either could or cold.
26 Too much cannot be made of these unstressed syllables, but Jones, as late as 1701, notably indicates oo ([u] and [u]) in such endings.
27 Cf. especially Cooper (1685) 49, and Watts, List 8.
28 Kökeritz (161) makes a strong case, on chronological grounds, for disregarding u-spellings as indications of early shortening.
29 Cf. Max Löwisch, Zur englischen Aussprache von 1650–1750 nach frühenglischen Grammatiken 73–80 (Halle, 1899); Jespersen 1.330–5; Spira, Die englische Lautentwicklung nach französischen Grammatiker-Zeugnissen, under u; and §2814 above.
30 Kökeritz (165) summarizes the most recent theory. The modern Southern American development [ɜ] may either be labialized or be centralized to the point where phonetic compensation produces the same effect as labialization. It is an interesting fact that Wallis (1653), often hailed as the first to describe the unrounding of u in unambiguous terms, seems to describe a centralized vowel that possessed some measure of labialization. Possibly, as Ellis (175) believes, Wallis is mistaken in his observation, but if he is mistaken there is every likelihood that the sound was sufficiently centralized to produce the effect of a labialized vowel on his ear. Wilkins denies all motion of tongue and lips but otherwise indicates a high central vowel, and Holder's long description (Ellis 178–9) cannot be interpreted in any other manner. Miege identifies with Fr. eu, i.e., he coincides with Wallis.
31 Cf. Jespersen 1.333–5. Note that many of the commonest words in the ME u category, precisely those words, in fact, from which u-transferences were most likely to be made, retained [u].
32 An interesting orthoepic parallel is Butler's use of oo and u for ME u. Cf. A. Eichler, Schriftbild und Lautwert in Charles Butler's English Grammar (1633, 1634) und Feminin' Monarchi' (1634) 48 (Halle, 1913).
1 The Writing Scholar's Companion (1695) 28, edited by Eilert Ekwall (Halle, 1911).
2 Simon Daines' Orthoepia Anglicana (1640) 52–3, edited by R. Brotanek and M. Rössler (Halle, 1908). Daines was a Suffolk man, and often reproduces Suffolk forms.
3 Cf. also Ellis 103.
4 Cf. C. H. Grandgent, From Franklin to Lowell: a Century of New England Pronunciation, PMLA 14.217 (1909); and Krapp 2.132. There are several diphthongal variants with a centralized front element, but from the present discussion the basic phoneme would appear to have been [u].
5 Ellis 5.237–64: districts D.19, D.16, D.18, including Essex, Norfolk, Suffolk, Rutland and parts of Northants.
6 Cf. Kihlbom 160; R. E. Zachrisson, A Contribution to the History of the Early New English Pronunciation (especially in the 15th Century), Englische Studien 52.299 (1918); W. Dibelius, John Capgrave und die englische Schriftsprache, Anglia 23.361 (1907); H. M. Flasdieck, Forschungen zur Frühzeit der neuenglischen Schriftsprache 2.25 (Halle, 1922); Kökeritz 158–9 gives grounds for supposing Mn. Suffolk [u] is a comparatively recent shortening of [u].
7 Luick Gram. 586–91, and Untersuchungen zur englischen Lautgeschichte (Strassburg, 1896), especially §66. Luick mentions several ambiguous references by orthoepists additional to Gill's, notably Mason's (1622), and Wallis' (1653).
8 Kihlbom disagrees with Luick's view that East-Anglian [u] is from earlier [uə], and our evidence shows the necessity of revising Luick's conservative chronology. But Kihlbom's argument does not take into account the possibility that the early ou-spellings may represent [uə] rather than the [u] which she assumes, and Luick's theory may be factually if not chronologically correct. The examples quoted above from WSA (270, 272) and WNR undoubtedly express [uə].
9 No coherent explanation of this asymmetry has ever been advanced. The supplanting of [e] for ē 2 by [i] except in break, great, steak, is vaguely put down to Class or Regional dialect without attempted explanation. If the suggestion made here is valid, it may furnish an explanation of the great popularity enjoyed by ea- and oa-spellings during the 16th and 17th centuries. For ME ē 2 in East Anglia, see Kökeritz 133–9.
10 So interpreted by Orbeck, Krapp, and Simpson, possibly correctly, although from the argument in §§31–2 above, their interpretation must be recognized as containing an unproven assumption.
11 Krapp also gives orthoepistical evidence for the establishment of the modern sound at least as early as the second half of the 18th century and sporadic o-spellings (cf. Kökeritz 158) may point in the same direction. Practically all these American rhymes are paralleled in Butler, Pepys, Pope, Herrick, and G. Fletcher.
12 Kökeritz 38 has [u:] in ago, alone, coal, clothes, gofer, loaves, reproach, roach, sloe, soap, toe; [u] in bone, choke, cloak, goat, hope, loaf, moat, most, oath, oats, poach, post, smoke, stone, wholly.
Although in this discussion I have agreed with Ekwall's interpretation of the orthoepic evidence from works c1700, it is possible that the ‘(o) obscure like (oo)‘ of Writing Scholar's Companion may actually indicate an early development of the modern New England sound. The entire paragraph containing our examples (Writing Scholar's Companion 28) is ambiguously worded.
1 A pronunciation [uə] in these words seems to have been general until the last decades of the 18th century; it is noted especially in Jones (1701) and The Expert Orthographist (1704), but Sheridan (1780) condemns it as an Irishism. Kökeritz 37 records [u] as a Suffolk variant in board, force, fortnight, fourteen, sport. See also §33 for the list in Writing Scholar's Companion (1695).
2 Cf. Ellis 183.
3 This list is made up from words mentioned by Wallis 60, Jones 113, Writing Scholar's Companion 28, Cooper 49, Elphinston 1.33, and Walker's Dictionary. Cf. Ellis 172–84.
4 Possibly, however, an analogical pronunciation based on school.
5 Ekwall, Jones 113. Note that Jones has u, not ǔ, in these words.
6 Cf. Kökeritz 153. Kökeritz records the survival of the 18th century forms for o before nasal and in sovereign, cochineal, etc.
7 Noticed in MEV 10.
8 Zachrisson ES 346-60. Denied by Luick Gram. 358.
9 KöKERITZ 30, 41-54.
10 C. M. Wise, Southern American Dialect, American Speech 8.37-43 (1933).
11 G. E. Fuhrken, Standard English Speech 46-7 (Cambridge, 1932).
12 This is not to deny the development to [ɔ:] but simply to point out that it was not the sole development. For [ɔ:] early spellings like awffer, fauster, caust, pauster (cf. MSP7, MEPS) are conclusive.
13 See Luick Gram. 600.
14 Chapter V above.
1 Hodges (1643) has cool'd, could in his list of like words.
2 [kud], [kuld], and [kud], especially the first two, are given by practically all Grammarians down to Walker; the long vowel is clearly indicated by Gill, Butler, Hodges, Cooper, Jones, and by some later writers.
3 The Art of Reading and Writing English 698 (London, 1721).
4 Dictionary (1797) under Mould.
5 Thomas Smetham, The Practical Grammar 36 (London, 1774)
6 If Hart's [kuld] refers to cold and not to could it is almost contemporary evidence for this. Cf. Jesepersen 1.108.
7 Chapter VI above.
8 According to Kökeritz 192 the present Suffolk sound is [αu]; EDG has [ɒu] for most of the East Anglian area. Ellis (D.16, D.18. D.19) approximately concurs. ME o before -ld has fallen together with ME ō 2, however, in South Lincolnshire (under [uə]) and in several other neighboring dialects; cf. EDG 74,88.
1 Zachrisson 80.
2 Zachrisson 207; Bullokar 60–5.
3 Ellis 156. Wallis' further identification of the ū-words with would, should, could, court seems to me certain evidence for the correctness of my view.
4 For Butler, Luick concurs with the opinion expressed above, but from Butler's description alone it must be admitted that any safe conclusion is impossible. Cf. Luick Gram. 564 anm. 4, and Eichler, Butler 27–8.
5 Ellis 157: ‘U gutturalem, ante u Germanicum oo anglice exprimentum semper scribimus per ou; ut out ex; about circa . . .‘
6 Zachrisson 152, following Otto, interprets EU as e and u, perhaps correctly.
7 Ellis 1027. Rowel is a somewhat doubtful example, though a ME rouel from OF rouele is well attested, and a form ruel is current to the end of the 16th century.
8 Ellis 1057.
9 We shall have more to say later regarding Elphinston's remarkably accurate analysis of ou (ME ū).
10 Ellis 1051.
11 I quote from Krapp's citation (2.196) : '. . . if we said o-oo extremely quick, it discovers this Sound exactly; as louse, mouse, fowl, town, etc. which are sounded quick lo-oos, mo-oos, fo-ool, to-oon.' This seems to be the ‘fast-diphthong’ of Kenrick and Steele.
12 In the ‘abusive’ pronunciation gâoon cited by Krapp 2.195. Krapp's summary of the several positions taken up by Webster is enlightening.
13 R. Forby, The Vocabulary of East Anglia; an attempt to record the vulgar tongue of the twin sister counties, Norfolk and Suffolk, as it existed in the last Twenty Years of the Eighteenth Century, and still exists; . . . (London, 1930). I quote from Kökeritz 169. For the Mn. Suffolk [εu], [ᵌu] see Kökeritz 64–7, where, however, the second type is regarded as a mere variant of the first.
14 Text: 'long u, instead of open u.' Kökeritz emends to read 'lengthening out long a, instead of open a'; but reference to the description above shows that it should read 'lengthening out long u instead of open a.'
15 Most of the inconsistencies between Palsgrave, Smith, Bullokar, Hart, and the other 16th century observers disappear immediately when a diphthong of the type described here is postulated.
16 Jespersen 1.286–7.
17 The following labial may actually have prevented centralization, or alternatively, it may have caused the smoothing of a nascent diphthongization by overstressing the labial influence of the second element. The explanation given above seems the more likely.
1 See the summary in Luick Gram. 564, Ellis 156–60, and Zachrisson 207–210. Nothing important can be made of French-English and English-French Grammarians; but Wilkins' remark (I rely on Ellis' citation) that [æ] and [u] will not coalesce ‘into a plain sound’ seems significant, and Lediard's pairing of ou and the German au is to the point. Cf. Ellis 1045. That the North Midlands knew a sound very close to the MnE development is proved by Gill's transcriptions geaun, gaun, and by burlesque au-forms in such plays as Heywood and Brome's Late Lancashire Witches, Otway's Cheats of Scapin, Shadwell's Lancashire Witches, and Squire of Alsatia. Cf. Whitehall, Thomas Shadwell and the Lancashire Dialect, Michigan Studies in English and Comparative Literature 10.261–278 (1933); and Adolph Weiss, Die Mundart im englischen Drama von 1642–1800 (Giessen, 1924).
2 Zachrisson corrects his own examples in Bullokar 130–144, though occasionally from an excess of caution. I am moderately certain that the WNR examples, found contiguously in two short documents, are valid.
3 Wisc. gose, mon ‘moon’, rots ‘roots’, skol ‘school’, sone, son ‘soon’, spone, por ‘poor’, blod, fott, hok ‘hook’, loke, stod, tock, tok; Simpson behoue ‘behoof’, canno ‘canoe’, none ‘noon’, etc.; beside Wisc. approch, close ‘clothes’, cost ‘coast’, incroch, hom, hop ‘hope’, lod ‘load’, ocke ‘oak’, oth ‘oath’, polle ‘pole’, rod ‘road’, rod ‘rode’, rop ‘rope’, ros ‘rose’, ston ‘stone’, holl ‘whole’, wrot ‘wrote’. There is a special reason, of course, why o should be written for ō 2 in New England (Chapter V above), yet reference to the NED under the words given above will show similar spellings in the British Isles.
4 Wisc. copple, nombur, stomp, stob, sodon, thonder, troble, onckle, bolits; Sharon not, coten, rom, bokles, tob, gog ‘jug’, bots; Simpson trosted, costody, doble, hondred, hossey, nombur, ponctiall, ronne, stompe, socklin ‘suckling’, soden, somoned, yong; Orbeck troble, doble, yong, costady, hondred, don; Massie costom, costomery, somer, contry; Austin costom, incombeson, nomber, contry, doble, troble; MVS soper, Rottland, cott, rosett, Robyshe, copes, gotter; MEPS grombeld, botcher; WSA motten, honney, tofted, etc.
5 So Zachrisson 80; MVS (under ME ū, etc.).
6 This need not necessarily apply to British examples.
7 Cf. Luick Gram. 635–9.
8 Cf. blow/two (Pope 543.571), blow / too (Pepys 5.312.4), flow'd / blood (Butler 115.659), know / shew / ewe (Basse 186), soul / school (Butler 248.1), etc.
9 A. Gabrielson, Rime as a Criterion of the Pronunciation of Spenser, Pope, Byron, and Swinburne 86–8, 188–92 (Uppsala, 1909). The reference is to the evidence itself rather than to Gabrielson's statement of results; the latter, naturally, follows the conventional theory.
10 That is, in London and the South-East Midlands. A better case could be made out for the early development of the MnE reflex of ū in the North-Midlands; even there, however, a reappraisal of the evidence is necessary.
11 The evidence for the North-Midland development is chiefly: (1) Gill's eau-, au-transcriptions for the Lincolnshire sound in 1621; (2) au-forms probably written as au-forms in the Nottingham Records and some other North-Midland texts; (3) burlesque dialect forms in au that occur in certain Elizabethan and Restoration plays; (4) ‘advanced’ reflexes of ME ū in modern North-Midland dialect.
12 For the distribution of the [ǯu] type and for some indication of the relative length of its two elements—seldom noticed either by phoneticians or dialectologists—we must await the publication of the first volumes of the Linguistic Atlas. Some mention of its currency in the South, Maine, and Ontario is to be found in Cabell Greet, A Phonographic Expedition to Williamsburg, Virginia, in American Speech 6.165, 167 (1931); Cabell Greet, A Record from Lubec, Maine, and Remarks on the Coastal Type, in American Speech 6.397 (1931); E. F. Shew-make, English Pronunciation in Virginia (Davidson, N. C., 1928), where the pronunciation is described as uh-oo ([əu] or [Λu]). See also Krapp 192–6, where a useful summary tends to show that whether the [au] type was established earlier or not, the [ǯu] type was still the preferred type among certain 19th century orthoepists. A situation in Suffolk that almost exactly parallels that in Virginia may be inferred from Kökeritz 64, 168–70.
13 Cf. Borje Brilioth, Grammar of the Dialect of Lorton, Cumberland 52 (London, 1913). Brilioth's [uu], representing the first stage of diphthongization, has undoubtedly a dominant second element, but it tells us little of succeeding stages. Orton's [öu] (South Durham Dialect) is nowhere quantitatively analyzed.
14 E. A. Meyer, Englische Lautdauer, eine experimentalische-phonetische Untersuchung (Uppsala and Leipzig, 1903). Cf. also Jespersen 1.450.
15 Notes on the Length of Vowels, American Speech 12.128–34 (1937).
16 Quoted by Jespersen 1.450. Jespersen does not quote the second passage given here. Elphinston, of course, was the first observer to state the general law that all vowels are longer before voiced sounds.
17 Engelbert Müller, Englische Lautlehre nach James Elphinston 121–2 (Heidelberg, 1914). Cf. Jespersen 1.451.