Latin sync., ik(o) (was: The oddness of Gaelic words in p-)

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:

> On 2008-06-03 19:47, stlatos wrote:

> > You are the one who said "*ped-ka:- 'sin' > pecca:re", I simply said
> > that it was formed from an adj./n (many verbs in -a:- are formed like
> > this) with -ik- not -k-.

> I'd like to support Sean's idea. As a matter of fact, I intended to
> propose the very same thing myself, but he beat me to it. Here are more
> arguments in favour of this derivation:

Thank you for your words. Though we agree about the sound changes
concerning this word I don't agree with all the evidence you propose
as support, I'm afraid.

> An original *ped-ka:- is unlikely, as it would probably have undergone
> the "thorny" treatment of medial *-tk(^)-.

For the first set of changes I'd say that t>s/_k just as t>s/_t as
in *xYexYd+ >> e:sca (*xYed+ 'eat', xY = h1). A PIE form with a long
V would work just as well for this change.

> The assimilation *-T-K- >
> -KK- affects clusters of secondary origin. Latin has the noun <pedica>
> 'shackle, fetter', and a syncope of the *deksiteros > dexter type could
> easily have produced *pediko- > *petko- > *pekko-

I believe the VCVCV type was irregular, but

> a syncope of the *deksiteros > dexter type

was regular by Exon's Law (V-shortV-shortV-V > V-0-shortV-V) even if
it created clusters that weren't allowed (fixed by new sound changes
or metathesis).

In addition, it's not certain that *dekYsiteros ever existed in
Latin. There are forms without -i- (desnU) and the reason for that was:

*dekY+ 'be fitting'

*dekYixYnos 'right, correct'

*dekYixYwos '(moving) to the right'

An i disappeared between two palatalized C, among other places. The
resulting kYxY cluster could have been corrected by making xY+syl but
it didn't happen at that time. The real change was kYxY>kYsY with
other combinations changing in other ways. Later IE languages that
caused Ks>Ks. changed s.Y > s. + Y > s. + i for the most part.

> with a meaning like
> *'(someone) hampered by foot injury' (cf. <mancus> 'infirm, esp. in the
> hand' < *maniko-?),

I'd say that PIE *mYaNkos 'small, short, cut short' was the origin,
with specification to 'hand' by folk et. (since Latin was one of the
few to undergo changes that made these two words more similar).

> hence <pecco:> *'trip, stumble' > 'commit a fault,
> sin'. Something similar must have happened in <occa> 'harrow' (cf. Gmc.
> *aGiðo:), possibly from *h2ok^-i-táh2 > *okita: > *otika: > *otka >

occa

> and the associated verb <occo:>.


--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:
>
> On 2008-06-03 20:59, Brian M. Scott wrote:
>
> > By the way, Watkins derives Latin <pedica> 'fetter, snare'
> > from *ped-ika:, which suggests that your alternative
> > derivation of <pecca:re> wouldn't work anyway.
>
> That's too hasty. As a parallel case, we have <manica> 'handcuff' and
> <mancus> 'maimed in the hand'. Latin syncopes followed loose tendencies
> rather than fixed rules.
>
> Piotr


I'd say manica came from later analogy with pedica, which is formed
in an unrelated way. That is, *pet/ped+iko+ >> pecca:re but *ped+ik+
>> pedica.

The diminutive *ik+ had many uses, including forming the names of
small tools from verb roots. It had the nom. *i:x or *a:x and
sometimes IE languages mixed these together or with the weak stem in
*ik+. This is especially evident in dim. forming insect names (-ika:
/ -i:ka: / -i:k(s) / -a:k(s) / -ak / etc.).