Re: [tied] Re: word without root

There's a rather extreme Polish example involving the same root *jIm- 'take'. The Slavic prefixed verb *vUz-jIm-ti 'take, take up' (Russian vz'at', Polish wzia,c') used to form the Polish imperative wez'mi(j) < *vUz-jIm-i (Russian voz'mi). This word, however, lost the final vowel and changed into wez'm, which was finally simplified, resulting in the modern imperative wez'. I doubt if any Pole who is not a linguist realises that wez' is an original prefix, not a verb root, and that the only trace of the original root is the palatalision of the final consonant -- this despite the fact that the prefix w(e)z- still exists in Polish!

It so happens that the Polish reflex of *vy-jIm-ti 'take out' has not suffered any mutilation and is still wyja,c', but we have played havoc with *Iz-jIm-ti 'take down (from), take off', where the preservation of nasality in Polish (<a,> with an underhook = nasalised [o~] or [oN]) probably caused some articulatory discomfort and made expected *zn'a,c' (from *(I)zUn-jIm-ti with an intrusive nasal in the prefix) to give way to dissimilated zdja,c', hence such aberrant forms as imperative zdejm(ij) for historically expected *zen'm(ij).

The Polish parlament is called Sejm (Gen. Sejmu, borrowed also into Lithuanian: Lietuvos Respublikos Seimas) from *sUn-jIm-U 'assembly' which should have produced *S'niem, Gen. *Sen'mu or the like. I suppose the non-Nom.sg. case forms were dissimilated (*sen'm- > *sejm-), whereupon the Nom.sg., which became the "odd man out", adopted the same stem form.

*jIm- is so capricious in Slavic because it split into very dissimilar and very short allomorphs in different environments (e.g. Polish jm-, ja,-, jem-, 'm-, 'a,-, where <'> means palatalisation of the preceding segment). New generations of learners simply couldn't cope with this mess, and I can't blame them.

Piotr

----- Original Message -----

From: s.tarasovas@...

To: cybalist@egroups.com

Sent: Tuesday, January 16, 2001 9:23 AM

Subject: [tied] Re: word without root

--- In cybalist@egroups.com, Andrei Markine <andrey@......> wrote: > Could cybalist members help me in resolving a puzzle? > > It is a Russian verb " vynut' " ("to take out"). > vy- is prefix, -t' is infinitive ending. If we look at the verb's conjugation: > 1sg: vynu > 2sg: vynesh' > 3sg: vynet > -nu- looks like common verbal suffix (reduced in future forms to -n- ). > That leaves no space for the root. > > I could guess that original structure was vy-n-im-ti (with -n- inserted by > analogy with prefixes vUn-, sUn-), then by analogy with large group of verb > on -nu, -nim- was replaced with suffix -nu and thus the verb eventually > lost its root. Is it correct? If it is, are there other examples of > "rootless" words? > > Regards, > Andrei Your conjecture is correct, a Common Slavic form was *vy-(j)Im-ti (<*u:- 'out'+ im- 'take'), 1 sg. pres. *vy-(j)Im-o, (here I for jer', o, for nazalized o, j is a prothesis, analogically replaced by -n- in East Slavic). However, Russian -nu- (<*-no,-) is not a common verbal suffix (and is not reduced to -n- in future forms), this suffix marks in Common Slavic, eg, inchoative verbs (the verbs denoting the beginning of an action, state, or occurrence), the future form in Russian is formally the same as the present one, other markers make the difference (usually the verb's stem is changed). Considering a normal developement we should have in modern Russian the following (taking into account acute's generazation on vy-): 1 sg.(f.<p.) vyn'mu<*vy(n)Imo, 2 sg.f. vyn'meSH<*vy(n)Imes^i 3 sg.f. vyn'met<*vy(n)ImetI. The forms we really have might be explained as simplification of consonantal cluster -n'm->-n- influenced by analogy, as you noted. Sergei