KaiTheHomoSapien wrote: ↑
06 Apr 2020 02:49
Where does the Latin word "senex" come from? By which I mean, how come the nominative singular ends in -x when the stem of all the other case forms of this adjective is simply sen-, whence genitive singular "senis" and nominative plural "senēs" (and not *senicis, *senicēs). So why the -x in the nominative singular? Are there other nouns and adjectives that exhibit this pattern?
In Latin, in fact, probably in many languages with case, the nominative singular is often statistically salient enough that it can develop or retain an irregular stem that is at odds with the rest of the declension.
Here, the expected nom. sg. *senis seems to have been influenced by various words for professions formed with compounds of two roots that create an athematic stem that incidentally ends in [ɛk], like jūdex 'judge' (< iūs 'law' + dīc(ere) 'say' + nominative -s) or haruspex 'diviner that reads animal guts' (< har(u)- 'guts' + spec(ere) 'look at' + nominative -s). The [k] also makes an appearance in the common derived word senectūs 'old age'.
It appears there are only two other examples of weird longer nominative singulars in Latin:
- nix nivis 'snow' (the nom. sg. retains a previous -g- that is dropped in the declension, cf. the verb ning
it 'it snows', Irish sneach
- Juppiter Jovis 'Jupiter' (the nom. sg. is a compound of Jov- + pater 'father')
The language has more weird nominative singulars, but they're not longer than the declining stem (grūs gruis 'crane', bōs bovis 'cow/bull/ox' with genitive plural boum, pār paris 'pair', carō carnis 'meat'). An interesting example is supellēx supellēctilis 'furniture', where the -ilis suffix that derives adjectives (agilis 'agile') has been attached in the declining stem. For IE studies, also iter itineris 'trip, journey' and jecur jecinoris 'liver', which reflect an old Indo-European -r-/-n- alternation in declension (cf. the evolution peth2r pth2en-s 'wing' into Greek pterón with -r- and Latin petna > penna with -n-), and femur feminis 'thigh' (which, having no cognates, may even be a borrowing shaped into an IE pattern). We may also mention ego vs. the m- stem in its declension and derived terms (meus).
Some examples of weird nominative singulars in Old French would be emperere empereor [empəˈrerə empəreˈɵr] (< imperātor imperātōrem, although this particular -ere/-eor suffix is common), cuens conte (< comes comitem), seindre seignor [ˈseindrə seˈɲɵr] (< senior seniōrem). In Old Occitan, enfas enfan [ˈenfas enˈfan] (< īnfāns īnfantem, expected nom. sg. *enfantz *[enˈfants] < *infantis), molher molher [ˈmoʎər moˈʎɛr] (< mulier mulierem), senher senhor [ˈseɲər seˈɲor] (< senior seniōrem). In PIE, see wédo:r udn-és 'water' (with the -r-/-n- alternation mentioned above), with a relatively strange long [o:] (many other such words too though, e.g. uksé:n uksn-és 'bull, ox').
Interestingly, I can't think of any example in Standard Arabic. It likely helps that in that language the nominative is always marked with a vowel-initial suffix just like in the other cases (whether it's -us or -u in the two most common declensions for common nouns, or dual -aani, or masc. plural -uuna and fem. and abstract plural -aati, or -in in C-C-j nouns, or...), unlike old Indo-European and its nominative singular ending in -s or nothing (as opposed to other cases formed historically with -o-s, -o-m, -o:s, -eh2, -eh2s, -osyo, etc.).
McFadden, Thomas. 2018. "*ABA in stem-allomorphy and the emptiness of the nominative". Glossa
https://www.glossa-journal.org/articles ... /gjgl.373/
It discusses possible hierarchies of stem irregularity by case in a discussion of Tamil, Finnish and Latin, with further examples from a few other languages.