New conjugations in IE languages

[Otto Kretschmer]

Have any IE languages innovated a new verb conjugation for person/number by any way like grammaticalization of pronouns?

I know many IE languages that gained new TAM features but haven't heard of an iE lang regaining actual conjugations

Any ideas,?

[Travis B.]

It is not a new verb conjugation per se - but, if I recall correctly, the German -st 2S ending was formed by joining a 2S ending -s with the consonant of a postposed du.

[zompist]

Have any IE languages innovated a new verb conjugation for person/number by any way like grammaticalization of pronouns?

Yes, French.

Don't be misled by the archaic spelling. The conjugation of (say) donner is more or less

ʒdɔn
tydɔn
idɔn/ɛdɔn
nudɔnõ
vudɔne
idɔn/ɛdɔn

In addition to person-prefixing, French also shows incorporation of objects, negatives, and tense into the verb, e.g. ʒənləlyiedɔne. The overall structure is reminiscent of Swahili.

This may surprise you, but again you have to ignore the spelling— the spaces are not a linguistic fact. That these verbal complexes are words is supported by their very strict formation— you generally can't place other material in between the morphemes.

You may also find the Portuguese personal infinitive of interest.

[Otto Kretschmer]

Very interrsting stuff, thanks!

How os the origin of Armenian forms like sirum em, sirum es etc ?

[Nortaneous]

Have any IE languages innovated a new verb conjugation for person/number by any way like grammaticalization of pronouns?

I know many IE languages that gained new TAM features but haven't heard of an iE lang regaining actual conjugations

Any ideas,?

The personal endings of the Tocharian languages don't faithfully reflect any inheritance from PIE, but their development isn't understood

[KathTheDragon]

Have any IE languages innovated a new verb conjugation for person/number by any way like grammaticalization of pronouns?

I know many IE languages that gained new TAM features but haven't heard of an iE lang regaining actual conjugations

Any ideas,?

The personal endings of the Tocharian languages don't faithfully reflect any inheritance from PIE, but their development isn't understood

I don't know where you're getting your information, but this is wrong. The personal endings are by-and-large transparently inherited. See Melzahn's "The Tocharian Verbal System" for a comprehensive overview.

[Nortaneous]

I don't know where you're getting your information, but this is wrong. The personal endings are by-and-large transparently inherited. See Melzahn's "The Tocharian Verbal System" for a comprehensive overview.

TB 1sg -u: "there are a lot of different views about the source of the w/u-element in Tocharian B, cf., e.g., the statement by Hackstein, 1995,
151f., fn. 7 that its derivation “bleibt m. E. eine offene Frage”. For earlier approaches, see VW II/2, 261."

TA/B 2sg -t: "According to many scholars, (*)-tä is to be derived from the PIE Middle II/perfect ending *-tHa (see, e.g., Krause, 1951, 151f.; TEB I, 258, § 462,1; VW II/2, 261ff.; Pinault, 1989, 154; Adams, 1993, 20). There is, however, the problem that PIE *a was expected to result in PT *-a, as has indeed been the case in the respective preterit ending *-stā, as already pointed out by Pedersen, 1944, 5. ... There is, of course, the further problem that one does not expect a Middle II/perfect ending to show up in active present/subjunctive paradigms. This problem has been explicitly addressed by G. Schmidt, 1995, 71 (see above sub 3.2.1.1), and can, of course, easily be solved within the general framework of Jay Jasanoff." Bringing in Jasanoff's theory isn't very appealing, though, and neither is the transfer of one personal ending from the middle/perfect to the present - especially since the development isn't phonologically regular. Postposition of a reduced (short) pre-PToch *tu > PToch *-tä as per Meillet seems preferable, at least on phonological grounds.

TA 3sg -ṣ and TB 3sg -ṃ: too much to excerpt here, but many proposals, often involving either postposition of particles or hypercorrection.

So none of the basic singular endings in TB are transparently inherited.

[Moose-tache]

Slavic has some endings that areoriginally adjectival and declined like adjectives, and so vary by number and gender, that are syntactically treated more like verbs. I think some Slavic languages do this with the reflexive/middle -sja, but they all do it with the innovated past tense-from-participle -l. When speaking Russian, you want to know if someone who is doing something now is 1st/2nd/3rd, but if they're doing it in the past you only care if they're masc/fem/neuter, which always felt like a fun shift in perspective to me.

To some extent, we may be able to include some periphrastic constructions in English. "I've run the whole way" only displays overt TAM marking on a clitic attached to the pronoun. I can imagine a future English in which the perfect and pluperfect are regularly indicated with -v and -d respectively on the subject.

Zompist: are there any other clues that these French verbal complexes are in fact single words? Not being able to interupt them doesn't sound like a strong argument on its own, but maybe there's more to it than I realize.

[Otto Kretschmer]

Yes. This is the case in Czech and my native language Polish - the past tense is a fusion of Proto Slavic resultative participle and the verb to be.

In Czech you have jsem byl, in Polish both forms have fused and created a new verb tense byłem (był jeśm > byłem)

[zompist]

Zompist: are there any other clues that these French verbal complexes are in fact single words? Not being able to interupt them doesn't sound like a strong argument on its own, but maybe there's more to it than I realize.

What do you think a word is?

"Words" are not really easy to define, if you don't let the writing system do it for you. They seem to be a combination of several concepts, as I noted on my Yingzi page:

(1) a phonological unit-- e.g. something with one stress accent or one pitch contour; or a unit within which intervocalic stops get voiced.
(2) an element which can stand alone (e.g. in response to a suitably chosen question), as suffixes or bound morphemes cannot.
(3) a morphological unit you can't insert other morphemes into (e.g. black dog is not a word since you can change it to black, tired dog; but you can't turn blackbird into blacktiredbird)
(4) the abstraction underlying a set of morphological forms (e.g. write underlying write, writes, writing, written, wrote).
(5) an expression with a conventional meaning-- something that has to be defined in the mental lexicon (this sense is also called a lexeme).

Note that there is no guarantee that all these criteria coincide, which is why we have edge cases like clitics.

If you're looking at a particular utterance, (4) and (5) are of no help, as they're really defining "lexeme" instead.

The French verbal complex meets all of (1) (2) (3).

There may be another criterion— that you can't leave out constituent morphemes without it being a different word. E.g. Spanish "pienso" cannot be abbreviated "piens" or "o". French meets this criterion: you have to say "je pense". You can't call a blackbird a "black". (This is a bit tricky since you can call it a "bird", but I'd argue that that is substituting a different word— "bird" is not a form of "blackbird" any more than it's a form of "robin".)

Finally, I'd add that morphological (word-building) rules may differ from syntactic (sentence-building) ones. This is again true of French: the constituent order differs (SVO in syntax, SOV in morphology); you can rearrange a sentence ("Moose, je lui a répondu, moi") in ways that you cannot with the verbal complex (*lui ai répondu je).