Published online by Cambridge University Press: 22 April 2026
[The historical status of MnE [i] is problematical. On presumptive and orthoepistical grounds, most authorities believe the sound was always [i]; some, notably Luick and Sweet, that an original [i], first partly lowered to [i] in late ME, was completely lowered in EMnE. Spelling and rhyme evidence, here first cited, demonstrates the retention of the stage [i], on occasion, until circa 1800. The uniform [i] of MnE seems to be due to the phonologization of one of two EMnE variants, probably as a result of phonemic proximity.]
1 American Pronunciation6 164 (Ann Arbor 1935).
2 E. A. Meyer, Englische Lautdauer (Uppsala-Leipzig 1903); C. E. Parmenter and S. N. Treviño, The Length of the Sounds of a Middle Westerner, American Speech 10.129-33 (1935); R-M. S. Heffner, Notes on the Length of Vowels, American Speech 12.128-34 (1937); H. A. Rositzke, Vowel Length in General American Speech, Language 15.99-109 (1939).
3 A Modern English Grammar on Historical Principles 1.63-6 (Pt. 14, Heidelberg 1928).
4 On Early English Pronunciation 1.107 (EETS, ES, London 1867-89).
5 Handbuch der mittelenglischen Grammatik 57-60 (Heidelberg 1925).
6 History of English Sounds 217-8 (Oxford 1888).
7 Historische Grammatik der englischen Sprache 1.137, 193, 378-9 (Heidelberg 1914-29).
8 Not necessarily good evidence; cf. Jespersen 1.114-5.
9 Franklin is practically always correct in any phonetic dictum, but his system was never completed to his own satisfaction. One can never decide, therefore, whether what he says represents his final view or whether it is merely an interim statement. Cf. Carl Van Doren, Benjamin Franklin 425-7 (New York 1938); Kemp Malone, Benjamin Franklin on Spelling Reform, American Speech 1.96-100 (1925). On Webster's notion of ‘quality’ cf. G. P. Krapp, The English Language in America 1.114 (New York 1925).
10 Sweet 217-8, Jespersen 1.63-4.
11 Dr. John Jones's Practical Phonography (1701), edited by E. Ekwall, Neudrucke frühneuenglischer Grammatiken 2. cxxxii-cxlv, ccv-ccvi (Halle 1907).
12 Jespersen 1.63-4; Ellis 1.105.
13 Charles Butler's English Grammar (London 1634), edited by A. Eichler, Neudrucke frühneuenglischer Grammatiken 4.13-4 (Halle 1910).
14 The Art of Reading and Writing English (London 1721); in the Works of Isaac Watts, edited by D. Jennings and P. Doddridge 4.697-8, Table VIII (London 1753).
15 Jespersen 1.63.
16 But cf. Ekwall cxxxvi-cxl.
17 Ellis 5.240.
18 F. T. Elworthy, A Grammar of the Dialect of West-Somersetshire (London 1877); Ellis 5.145-55; Ekwall cxxxviii-cxl.
19 By this argument, I do not wish to imply that every stressed short i in the 16th, 17th, and 18th centuries was a higher high-front vowel. Already c1500, Hymn to the Virgin indicates either two developments [i] and [i], or [і] and [i]. Jones seems to agree. Tiffin (1751) puts the existence of a generalized [i] beyond all doubt. Cf. William Matthews, William Tiffin, 18th Century Phonetician, English Studies 18.97-114 (1936). Every successive investigation of the 17th and 18th century status of the English vowel-system makes clear the range and multiplicity of the widely used variations, and these variations are not mutually exclusive.
20 Luick 376-9.
21 The number of EMnE i-spellings is so enormous that it gives the appearance of an almost complete raising of ME e. The change certainly seems to have occurred as commonly before consonants other than nasals as before nasals. Nasalization, however, may have helped to retain it in the dialects and in MnE.
22 As, for instance, in the a-forms for ME u.
23 I include in this statement the semi-stressed ee of thirteen, briganteen, etc.
24 Jespersen 1.239.
25 Joseph Wright, The English Dialect Grammar 34-5 (Oxford 1905).
26 And especially by Butler, who uses it as a phonetic symbol, by Jones, and by Watts.
27 Anders Orbeck, Early New England Pronunciation (Ann Arbor 1927).
28 Claude M. Simpson, Early Rhode Island Pronunciation 1636-1700; American Dialect Society, Microfilm Monograph No. 1; published on 35 mm. film by the American Documentation Institute (Washington D. C. 1937).
29 Helge Kökeritz, The Phonology of the Suffolk Dialect (Uppsala 1932) = Kökeritz; William Matthews, The Vulgar Speech of London in the XV-XVII Centuries, NQ January 2 to April 3 (1937) = MVS; Sailors' Pronunciation in the Second Half of the Seventeenth Century, Anglia 47.193-251 (1935) = MSP7; Sailors' Pronunciation 1770-83, Anglia 49.72-80 (1937) = MSP8; Variant Pronunciations in the Seventeenth Century, JEGP 37.189-206 (1938) = MVP; Some Eighteenth Century Phonetic Spellings, RES 12.1-31 (1936) = MEPS; South Western Dialect in the Early Modern English Period (unprinted) = MSWD; Harold Whitehall, A Short Study of the Vowels in the Language of the Shuttleworth Accounts, PQ 10.10-26, 269-76, 384-91, 11.293-302 (1932) = WSA.
30 For rhyme sources see my monograph, Middle English ū and Related Sounds: their development in Early American English; Linguistic Society of America, Language Monograph 19.10-2 (Baltimore 1939). The spelling sources of the Wisconsin Collection are given in Miles L. Hanley, English and American Pronunciation, 16th to 18th Centuries, mimeographed preliminary edition (Madison 1938). Other abbreviations, chiefly of texts in my own collections, follow: Bate = Blacksmith's Accounts of John Bate of Sharon, Conn. (from the MS in my possession); BdR = Records of Boxford, Mass.; CrR = Records of Colchester, Conn.; FgR = Records of Fitchburg, Mass. (Wisc. Coll.); HdPC = Records of the Particular Court of Hartford, Conn. (Wisc. Coll.); HdR = First Land Records of Hartford, Conn., in Conn. Hist. Soc. Collections 14 (Hartford 1912); HdV = Hartford Town Votes (Wisc. Coll.); HnCR = Indian Deeds of Hampden County, Mass.; IhR = Records of Ipswich, Mass.; LeR = Records of Lee, Mass.; MayD = Mayflower Descendant vols. 4-6 (Plymouth 1902-6); MdR = Records of Marblehead, Mass. (Wisc. Coll.); MHS = Maine Hist. Soc. Collections, especially vol. 3 (Portland 1853); MnR = Records of Milton, Mass. (Wisc. Coll.); MrR = Records of Manchester, Mass.; NER The New England Historical and Genealogical Register vols. 1-50 (Boston 1847-97); NkR = Records of Norwalk, Conn.; PhR = Records of Portsmouth, R. I. (Wisc. Coll.); PR = Parish Records (Wisc. Coll.); SnR = Records of Sharon, Conn.; SyR = Records of Salisbury, Conn.; TyR = Records of Tisbury, Mass.; YkD = Deeds of York County, Maine.
The spellings and rhymes are given here in bulk; I make no attempt to list alternative explanations of individual words. I am aware that weedow, speet, etc., may show the results of lengthening in an open syllable, eend, etc., the results of lengthening before the group -nd, seelectmen (cf. Bloch's fn. 36a below) the effects of stressing the first syllable. The point is that none of these examples is spelled or rhymed in a manner at all different from the very numerous other examples that cannot be so explained. Alternative explanations merely cover some few of my examples; the only explanation that apparently covers all of them—always, of course, with the possibility of later modification—is the one advanced below.
31 The NED has also frequent ee-forms, quoted without source: bill, chicken, chisel, cripple, grid-iron, grin, litch sb. 3, skid, skillion, spit, twist, etc.
32 Starred rhymes have parallel spellings in the preceding section.
33 For interpretation of the vowel in believe, receive, etc., as an ē 1 cf. Luick 556.
34 Johnson's English Poets (London 1790).
35 Shakespeare, edited by Kittredge (New York 1936).
36 From the American sources and in selection only.
36a [The spelling seelectmen may, on the other hand, indicate a pronunciation with stress on the first syllable. Map 443 of the Linguistic Atlas of New England shows three types with initial stress to be still widely current in northeastern New England:
, the most common pronunciation in a large part of Maine and New Hampshire;
, concentrated in the Merrimack Valley; and
, occurring sporadically in the neighborhood of Haverhill, Mass., and elsewhere. Manchester, the town from whose records the form in the text is cited, is in Essex County, where all three of these types are today competing with each other and with the standard pronunciation.—B. B.]
37 I give only such examples as can be substantiated immediately by parallel rhymes or i-spellings. My own collection and the Wisconsin Collection together have an enormous number of i-forms for practically all words in the ME e category.
38 So also Butler (1633-4).
39 Numerous other ee-forms, not cited here, lack substantiating i-forms.
40 Representing a very small selection from the rhymes available in the Wisconsin Collection. Starred rhymes have parallel spellings in the preceding section.
41 Robert Nares, Elements of Orthoepy 55 (London 1784).
42 James Adams, Euphonologia Linguae Anglicanae 66 (London 1794).
43 C. Cooper's Grammatica Linguae Anglicanae (1685), edited by J. D. Jones, Neudrucke frühneuenglischer Grammatiken 5.74 (Halle 1911).
44 Earlier students of English phonology seem largely to have neglected the ee-spellings. There has certainly been no serious attempt to collect them. Cf. Zachrisson, Wyld, Süssbier, et al.
45 Cf. Luick 393-409 for the gist of Luick's own theory and an excellent review of the literature.
46 Jespersen 1.114-5.
47 Elphinston's empiric observation was sufficiently acute to allow him to anticipate the conclusions of Meyer, Heffner, Rositzke, etc. See E. Müller, Englische Lautlehre nach James Elphinston 121-2 (Heidelberg 1914), and Jespersen 1.450. In estimating Elphinston's phonetic gifts, however, it will always be wise to wonder how far he had been influenced by his close London neighbor, Benjamin Franklin.
48 Whitehall, Development of ME ū 56-8.
49 Luick 555 note; Ellis 5.134-5; Ekwall cxxxviii-cxl.