From: Piotr Gasiorowski
Message: 365
Date: 1999-11-30
----- Original Message -----From: Mark OdegardSent: Tuesday, November 30, 1999 1:06 AMSubject: [cybalist] Re: PIE stress and accent (5)
Pre-PIE 'Comparative' PIE Gloss pád- pad-á po:ts pedós foot Piotr is not quite at the point where he is going to discuss umlaut and ablaut, but I see it coming. In the case of English foot, German fuss, we have an example of umlaut where the plural form takes on a form of vowel harmony, along with the loss of an ending. The word 'umlaut' is as polysemous in English (and, so I'm told, in German too) as is the English word 'accent'. You tend to get blank looks when you mention the word 'ablaut'.
I won't steal Piotr's thunder, and will let him explain things. This set of postings is useful and informative.
Mark.
Mark,I can't explain it all in my 'stress and accent' postings, so here's a separate sketch.Ablaut means the morphological alternation of vowels, especially in PIE and its descendants. In PIE, the same root could occur sometimes with the vowel e, sometimes with o, and sometimes without a vowel. In the last case a consonant or glide adjacent to the lost vowel had to become syllabic to ensure pronounceability.For example, the 'foot' root, *ped- occurred also as *pod-, and sometimes even as *pd- [bd]. Another root, *leuk- 'light, shine' alternated with *louk- and *luk-. We refer to these alternants as, respectively, the e-grade, the o-grade and the zero grade. Furthermore, vowel lengthening could affect either of the forms containing a full vowel (that is, e or o); this happened, for example, if a consonant-final noun stem had a full vowel in the final syllable of the nominative singular (as in *p@...:r 'father', or *die:us 'Sky God'; the latter had a diphthong with a lengthened first element).The distribution of the grades was determined by morphological factors. For no particularly important reason, just in deference to tradition, the e-grade serves as the 'citation form' of PIE roots in etymological dictionaries and the like. To sum up, in the case of *ped- 'foot' we have the following alternants:*po:d- as in *po:ts (Nom. sg.) -- lengthened o-grade*pod- as in *podm (Acc. sg), *podes (Nom. pl.) or *podns (Acc. pl.) -- normal o-grade*ped- as in *pedos (Gen. sg.), *pedei (Dat. sg.), *pedom 'foothold, place' -- e-grade*pd- as in (hypothetical) *n-bdos 'footless'In verbs, the e-grade typically occurs in the present tense, provided that the root vowel is stressed. The o-grade is found in some derived categories such as the causative verb (even in the present tense), as well as in the root-stressed forms of the perfect. The zero grade tends to occur when the root is not stressed, especially in 'mobile' paradigms, but also in some static stems (thus commonly in the aorist, one of the past tenses). Here are some examples for the root *leuk- 'shine':*leuk- as in *leuketi 'shines' or *leukos 'shining, light-coloured'*luk- as in *elukom 'I shone' (aorist), *luktos 'lit up'*lu-ne-k- (with a 'nasal infix') as in *lunekti 'is throwing light'*louk- as in *lelouke 'has thrown light' (reduplicated perfect)Traces of PIE ablaut survive e.g. in some of the ModE irregular verb patterns. For example the ablaut series *ei ~ *oi ~ *i (> PGmc *i: ~ *ai ~ *i) accounts for drive ~ drove ~ driven, and *en ~ *on ~ *(syllabic) n (> PGmc *in ~ *an ~ *un) underlies drink, drank, drunk.
Umlaut is not really a form of vowel harmony. The latter term is reserved for systems in which a selected phonological feature of the root vowel(s) determines the vowel timbres that are permitted to occur in affixes. This is what happens in Hungarian, Finnish, Turkish and numerous other languages. In a typical vowel-harmony system affixes may exist in two variants, e.g. one with a front vowel (-üt) occurring with front-vowel roots (mes-üt), and the other with a back vowel (-ut) to be added to back-vowel roots (mos-ut). If there is a string of affixes, they will all be 'harmonised' with the root: mes-üt-ök : mos-ut-ok. (This is just an artificial example made up to illustrate the principle of frontness harmony.)
Umlaut is a phonological process which makes a class of vowels assimilate in some respect to the vowel of the adjacent syllable (but chain assimilation affecting two or more consecutive syllables is also possible). The assimilated feature may be frontness, backness, lip-rounding, vowel height or a combination thereof.The individual Germanic languages show several kinds of umlaut, the most famous of which is variously known as 'i-umlaut', 'front mutation' or 'palatal umlaut'. It consists in the fronting of a back vowel (often with the retention of lip-rounding) when the next syllable contains a high front segment (i or the glide j). In Germanic the umlauted vowel normally belongs to the root, whereas the one that triggers i-umlaut belongs to a word-forming or inflectional suffix. Since unstressed high vowels were often deleted in early Germanic, the conditioning segment is usually lost in the historically attested languages, and the fronting of the root vowel is its only trace.Thus, for example, Old English has fu:l 'foul' but fy:lth (OE y = German ü) 'filth', the umlauted form contains the pre-Old English suffix *-ithu, forming abstract nouns (cf. ModE long, strong : length, strength). The OE verb meaning 'come' was cumu in the 1st person sg., but the 2nd & 3rd sg. forms were cymst < *kum-isti and cymth < *kum-ithi, respectively.Nouns with umlauted plurals (like mouse : mice < OE mu:s : my:s) are remnants of the PIE root-noun class. The complete history of foot : feet from PIE till present is as follows:
PIE po:ts, pl. pod-es PGmc fo:t, pl. fo:tiz (after the operation of Grimm's and Verner's laws, and with the vowel of the Nom. sg. generalised analogically across the paradigm) Pre-Old English fo:t, pl. fö:ti (with the vowel of the N. pl. fronted but still rounded, and the umlaut-triggering suffix still present) Old English fo:t, pl. fö:t (Anglian dialects) or fe:t (West Saxon, with the vowel unrounded) Chaucerian Middle English foot [fo:t], pl. feet [fe:t] Modern English foot [fut], pl. feet [fi:t]. We still use Chaucer's spelling but the old quality of the vowels in question has been affected by the Great Vowel Shift in the 16th century ([o:, e:] > [u:, i:]) and the subsequent shortening of [u:] to [u] in foot.Piotr