Re: LANGUE NATURELLE: Les groupes des verbes en Français (Re: TECH: Official languages of the list)

On Thursday, August 19, 2004, at 09:45 , Mark J. Reed wrote:

> On Thu, Aug 19, 2004 at 09:10:03PM +0100, Joe wrote:
>> Trebor Jung wrote:
>
>> The '-er' verbs mostly come from Latin '-are' and, I think,
>> 'ére'(Hence Spanish 'cantar' to French 'chanter'),
Nope - Not from -ēre; all from -āre

> And It. cantare, all from Latin cantāre.
Yep - to which you can add: Portuguese & Catalan 'cantar', and Romanian
'cînta' /k1n'ta/

>
>> the '-re' verbs come from '-ere', I think
>> (Spanish 'vender', French 'vendre').
>
> Both from Latin "vendere", which was usually encountered in CL via its
> passive form "venire".
I'm not sure what you mean by 'usually encountered in CL via its passive
form "venire"'. In any case, I think it'd be helpful to make it clear that
the 'passive' you are speaking about is 'vēnīre' (to be sold) with long-e
in the first syllable, and not the far more common 'venīre' (to come) with
short-e in the first syllable.

In fact 'vēndere' (to sell) is found more often in Classical sources than
'vēnīre' (to be sold). It's true that in the 'golden' Classical period of
Cicero and company, the passive of 'vēndere' is not found and is supplied
by the verb 'vēnīre'. But from the time of Seneca (mid 1st cent CE)
onwards, regularly formed passives of 'vēndere' are used and 'vēnīre'
doesn't seem to have survived to the 2nd century (tho I'm sure instances
will be found in archaizers of later periods).

Perhaps before getting back to the matter of verb groups in French and,
now, other Romancelangs, I should perhaps explain that the oddity of
active 'vēndere' ~ passive 'vēnīre' is explained by the fact that _both_
verbs are compounds:
vēndere <-- vēnu(m) dare (to give for sale - the longer form 'vēnumdere'
is also found in early Classical Latin)
vēnīre  <-- vēnu(m) īre   (to go for sale)

But this is not relevant as far as Vulgar Latin & the Romance languages
are concerned; only 'vēndere' survived in VL.

> Classical Latin had four verb conjugations, seem to have collapsed to
> fewer in the Romance languages.
In Iberian Romance, yes - elsewhere, no.

> The four CL conjugations were:
>
> 1st: āre        (long) A-stem
> 2nd: ēre        long   E-stem
> 3rd: ere        short  E-stem
> 4th: īre        (long) I-stem
>
> The 3rd conjugation was further subdivided, as some verbs regularly
> inserted an -i- between the stem and a certain subset of the endings.
Indeed, so that in reality there are 5 groups. Furthermore, both 'stare'
(to stand) and 'dare' (to give) are odd in that the -a- is short not only
in the infinitives but also in most other forms of the verbs where other
1st conjugation verbs have long-a. This explains why both 'stare' & 'dare'
  normally weaken to -stere & -dere & join the 3rd conj. when in compounds
(e.g. vēndere): the short-a weakens to short-e.

The Classical system is, more precisely:

1a. -āre		long-a stems (nearly all traditional 1st conjugation)
1b. -are	short-e stems (stare, dare)

2a. -ēre		long-e stems (the traditional 2nd conjugation)
2b. -ere	short-e stems (the traditional 3rd conj. of the 'mitto,
mittere' type)

3a. -īre 	long-i stems (the traditional 4th conjugation)
3b. -ere	short-i stems (the tradition 3rd conj. of the 'capio, capere'
type)

> In Spanish, the 1st conjugation verbs became -ar verbs; the
> 2nd and 3rd conjugation verbs became -er verbs; and the 4th
> conjugation verbs became -ir verbs.  Which seems eminently logical.
Rather, on the Iberian peninsular the short vowel (b) sub-groups didn't
survive, the verbs being reformed according to the more common long vowel
equivalents, leaving Iberian Romance with the three groups corresponding
to the three I listed above.

> The French, of course, Had To Be Different. ;-)
Not merely French; Italian & Romanian behaved similarly.

> While the fourth
> conjugation verbs went to -ir, as in Spanish,
Yes, that's my 3a above - true of all Romancelangs.

> the other three were
> treated differently.
Nope - the 1st conjugation (i.e. both my 1a and 1b) also remained as a
group in all the Romance langs.

> French has no -ar verbs; the 1st and 2nd
> conjugation verbs merged in -er,
No, no, no - a study of Old French will soon show that -er derived, by
_regular_ sound change from Vulgar Latin -áre (i.e. stressed -a-, there
being no phonemic distinction of length in VL). Latin _stressed_ -a- in
unblocked syllables regularly became -e- in Old French (with other
modifications if palatalized or nasalized), e.g. mare (sea) --> mer, quale(
m) (which) --> quel, tale(m) (such) --> tel, ma.tre(m) --> *medR@ --> mère
etc.

The Old 2nd decl. --ēre became, as one would expect, -oir in Old French.
Verbs with infinitives in -oir remain to the present day, e.g. avoir <--
VL *abére (CL. habēre).

In fact French, like Italian & Romanian, has retained _four_ groups,
corresponding (approximately) with the 4 of the traditional Latin grouping,
  with infinitive endings thus:
Classical Latin		French		Italian			Romanian
	-A(:)RE			-ER			-ARE			   -A
	-E:RE			-OIR		-ÈRE (stressed e)  -EA
	-ERE			-RE			-ERE			   -E	-I:RE			-IR			-IRE			   -I

> while the 3rd became -re.  (In both
> cases, the final form is due to a reduction from short E to zero in part
> of the ending: -e:r(e) -> -er, -(e)re -> -re.)
--(e)re -> -re is correct, but -e:r(e) -> -er is mistaken. Stressed long-e
of CL normally became -oi- in Olf French, and did so in the infinitives.
All the French -er infinitives are derived from -are.

What has made the picture confusing in French, Italian and Romanian is
that quite clearly there was much condfusion, due to phonetic weakening of
endings, between the traditional 2nd & 3rd conjugations (my 2a, 2b and 3b)
; in all three languages the long-e stems gave way, for the most part, to
the short-e type (which also absorbed the short-i types), so that in the
modern languages the -OIR/-ÈRE/-EA group is much reduced.

But even this is an over-simplification. Neither I nor Marcos have, for
example, mentioned that French -ir verbs come in two varieties, i.e.
finir ---> je finis, nous finissons
partir --> je parts, nouns partons

But that is another story. In fact the development from VL to the modern
Romance verb groupings is a complicated matter, way beyond the scope of a
single email. There was a great deal of confusion, as endings weakened,
more irregularities created by phonetic developments and subsequent
ironing out of irregularities by analogical remodelling, i.e. real
natlangs behaving like real natlangs  :)

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com    (home)
raymond.brown@kingston-college.ac.uk   (work)
===============================================
"A mind which thinks at its own expense will always
interfere with language."         J.G. Hamann, 1760

Reply