A: The lengthening of short vowels before bi-consonantal sequences (CC) that shared a place of articulation.   

The traditional description of HL states that short vowels were lengthened before bi-consonantal clusters that shared a common point of articulation. Typically, the first consonant would be a sonorant, such as a liquid or rhotic, while the second was a voiced obstruent. Examples from OE include 'cild' (child), 'climban' (climb) and 'blind' (blind). This change is typically dated to the 8th-9th century, but full diffusion may have taken longer.

As is argued in CoNE, more recent investigations on this topic have argued that the traditional descriptions are inadequate. HL did not occur equally in all language varieites and in all VCC contexts. Lass et al (2013), for instance, has argued that HL is variable --- the low vowel /a/ being lengthened far less frequently than the high vowel /i/ --- and in some cases lexically specific. Moreover, differences of application of HL in /ind/ and /und/ sequences comprise one of the classic Ribble-Humber line isoglosses separating Scots and Northen English, which retained the short vowel, from Southern English varieties.  

To read more on this change, please consult CoNE and the references therein or those listed here.   

Jordan (1934: 39)

Minkova (2014: 165)

Aitken & MacAfee (2002: 6-9)

Johnston (1997a: 65-66)

(Lass et al. 2013)

Child

Climb

Blind

Long

O: Homorganic Lenthening (HL) refers to the lengthening of short vowels before clusters of two consonants with the same place of articulation, the first of which is a sonorant, the second of which is typically a voiced obstruent or nasal. This change is traditionally dated to the second half of the 8^th and first half of the 9^th centuries (Jordan 1934: 39, Minkova 2014: 165; see also Jespersen (DATE: 118-120), Lass (CHEL2, 1999: 71-72) and Luick (DATE: 474-477), and Aitken & MacAfee (2002: 6-9) and Johnston (1997a: 65-66) specifically for accounts of the change in the history of Scots).

The clusters ostensibly involved in HL were: ld (e.g. OE cild 'child'), mb (e.g. OE climban 'climb'), nd (e.g. OE blind 'blind'), ng (e.g. OE lang 'long'), rd (e.g. OE beard 'beard'), rl (e.g. OE eorl 'earl'), rn (e.g. OE bearn 'bairn, child'), rð (e.g. OE eorðe 'earth'). When a third consonant followed, HL was blocked (e.g. OE cildru 'children').

As is argued in CoNE (Lass et al. 2013), traditional descriptions of HL are not sufficient. Firstly, the change is not regular and appears to be lexically conditioned, with different words being affected in different dialects and with vowel length in particular words not matching in sources from different periods (suggesting variable lengthening with one or other variant winning out through time). Secondly, not all vowels were affected equally; this is especially the case for the low vowel a, which was hardly affected by HL (except before ld, on which see further below). Thirdly, the change was more likely before certain consonant clusters than others, with it being most consistent before ld, mb and nd, and with its application before rC clusters being questionable except in a few cases such as beard. Fourthly, evidence for the date of the change is often lacking (since the spelling may not indicate that any change in length has occurred, cf. Modern English wind (n.) and find (v.)).

Differences in the application of HL in the sequences ind and und constitute one of the classic Ribble-Humber Line isoglosses separating far northern English and Scots dialects (with a retained short vowel) and Midland southern English dialects (with a long vowel), as in blind, find and found, ground (Wakelin 1972: 103). But even here the developments are not regular, with the vowel in wind (n.) only lengthening sporadically in English dialects, and the vowels in kind, mind, pound, round and sound being lengthened everywhere.

Lass et al. (2013) note that evidence for HL before ng is slim. Nevertheless, two definite instances occur in Older Scots: laing 'long' (alongside usual lang) and taingis 'tongs' (the long vowel in this word also underlies traditional pronunciations of this word such as [tjɛŋz] in northeast England), though neither is recorded in the FITS database.

Lengthening of a before nd is absent in later records even though Lass et al. (2013) note that it is indicated by Orm; in any case there appears to be no evidence of it in Older Scots. Before ld, as in cold and old, the existence of the rounded vowel in English dialects south of the Ribble-Humber Line can be taken as evidence for HL (as otherwise there is no obvious motivation for the raising and rounding of original a in these words). Lass et al. (2013) note, however, that HL in these words cannot be proven in northern English or Scots dialects, as the resulting forms (e.g. cauld and auld) could have developed from either the original short vowel or from a long (HL) vowel. Nevertheless, the coincidence of the ARR isogloss and the auld/old isogloss in northern England suggests that HL applied in these words in the north too. See Maguire (forthcoming) for further discussion of the development of a before ld in English and Scots.

As noted, the traditional dating for HL is in the Old English period, in the 8^th-9^th centuries. Lass et al. (2013) review the evidence for the dating of HL and conclude that this early date for HL cannot be proven, and that the earliest secure evidence for the change is in Orm, in the late 12^th century. Since HL must precede ARR (given the presence of ARR in words such as comb and old), a change which is probably datable to the 12^th century, we can assume that HL had occurred by at least this date.

Evidence for HL in FITS ( FITS instances )