Germanic: comparative inflection, strong/weak contrast

[freknu]

Looking at Old Norse adjectival inflection there is only "weak" inflection for comparatives (it's actually neither strong nor weak, but uses the weak stem inflection), and looking at Proto-Germanic there doesn't seem to be a strong/weak contrast either.

Could this be a trait left over from PIE, or could there be a different, general, reason for why comparative adjectives do not seem to follow the pattern of strong/weak inflection as with positives and superlatives?

[Daniel]

Just adjectives? I don't know. I know the nominal paradigms can roughly be traced back to PIE. But you probably know that already as well.

[freknu]

Comparatives and numerals seem to lack the strong/weak pattern, I'm unsure about the participles (I think they could be inflected as well, just as in descendent languages).

Also, the terms strong and weak for adjectives simply means that they are inflected as strong and weak stems, the actual meaning is one of definity: strong adjectives are indefinite, weak adjectives are definite.

[Daniel]

Hm, I don't think I can help. I'll keep skimming the thread if you find more info though.

strong adjectives are indefinite, weak adjectives are definite.

This reminds me of something found in Slavic languages. I don't know if that helps you much. But there are predicative and attributive forms; there's a specific term for this and I can't remember specifically what it is. Sometimes they encode definiteness, so you could call them definite and indefinite forms, but that's not always accurate, and I don't think that's the main term for them.

Here's a discussion you might find relevant:
http://forum.wordreference.com/showthread.php?t=2294862

I don't know if that implies that they go back to PIE or perhaps there was contact or perhaps they just are different/separate developments. I also don't know if they are actually similar in any of the details.

Edit: Aha! It's just "long" and "short" adjectives. I knew it was a generic term, but with a specific technical meaning.
Here's one reference on it:
https://benjamins.com/#catalog/books/la.153.02bab

More info:
http://forum.wordreference.com/showthread.php?t=1481230
http://en.wikipedia.org/wiki/Old_Church_Slavonic_grammar#Adjectives

[lx]

If you find an answer as to why, let me know! 
I always thought the comparative must imply some implicit definite reference that rendered the strong comparative obsolete, but there are instances where you use weak/strong in the superlative that don't imply definite reference and you could use the comparative instead and it seems that in that case there is no argument to be made for the absence of strong comparatives based on this reason alone.

[freknu]

Besides curiosity, there's a more practical reason that I'm asking.

In northern Ostrobothnian (roughly above Vasa; compared to southern below Vasa) there is still an archaic, seemingly strong comparative, hvítaran, which also seems to appear in a reduced form in the southern hvítán.

Compare the conventional inflection regimes with the hypothetical strong comparative — using the masculine inflection of hvít (white).


Positive, strong

	PG.	ON.	dial.
nom.	*hwīt-az	hvít-r	hvít
acc.	*hwīt-an-ōN	hvít-an	hvít-an
dat.	*hwīt-amm-ai	hvít-um	?
gen.	*hwīt-as	hvít-s	hvít-as

Positive, weak

	PG.	ON.	dial.
nom.	*hwīt-ô	hvít-i	hvít-a
acc.	*hwīt-anuN	hvít-a	hvít-a
dat.	*hwīt-ini	hvít-a	hvít-a
gen.	*hwīt-iniz	hvít-a	hvít-as

Comparative, strong

	PG.	ON.	dial.
nom.	*hwīt-ōz-az	*hvít-ar-r	? > hvít-án
acc.	*hwīt-ōz-anōN	*hvít-ar-an	hvít-ar-an > hvít-án
dat.	*hwīt-ōz-amm-ai	*hvít-ar-um	?
gen.	*hwīt-ōz-iniz	*hvít-ar-s	hvít-ar-as > hvít-án-as

(NOTE: -án- reanalysed as the comparative stem suffix)

Comparative, weak

	PG.	ON.	dial.
nom.	*hwīt-ōz-ô	hvít-ar-i	*hvít-ar-a
acc.	*hwīt-ōz-anuN	hvít-ar-i	*hvít-ar-a
dat.	*hwīt-ōz-ini	hvít-ar-i	*hvít-ar-a
gen.	*hwīt-ōz-iniz	hvít-ar-i	*hvít-ar-as

Superlative, strong

	PG.	ON.	dial.
nom.	*hwīt-ōst-az	hvít-ast-r	hvít-ast
acc.	*hwīt-ōst-an-ōN	hvít-ast-an	hvít-ast-an
dat.	*hwīt-ōst-amm-ai	hvít-ast-um	?
gen.	*hwīt-ōst-as	hvít-ast-s	hvít-ast-as

Superlative, weak

	PG.	ON.	dial.
nom.	*hwīt-ōst-ô	hvít-ast-i	hvít-ast-a
acc.	*hwīt-ōst-anuN	hvít-ast-a	hvít-ast-a
dat.	*hwīt-ōst-ini	hvít-ast-a	hvít-ast-a
gen.	*hwīt-ōst-iniz	hvít-ast-a	hvít-ast-as

Note that this could very well be an innovation rather than a conserved feature. What is also interesting is that the -án- form does not show any sign of retroflection. Both regions with and without retroflection show a plain nasal -án [ɑːn] rather than retroflexed [ɑːɳ].

The change /VrC/ > /VːCʳ/ (where C is /n, t, (d), s, l/, V is a low vowel (especially /ɑ/), and /ʳ/ indicates retroflection) is both productive and strong, so it's not only unusual, but very puzzling, to see hvítaran change into hvítán without retroflection.

At least if you consider it to be a modern innovation. If it was indeed an archaic feature (before retroflection began to affect words) it would be easier to see the change *-aran > *-arn > *-án.

So I'm kind of scratching my head bald at the moment.

It doesn't necessitate an archaic strong inflection of comparatives, but it's kind of ends up weird without one. Because there's so little documented material on Ostrobothnian (and Nylandian, the both are very probably closely related) it's even tougher. I also don't have any extra-Ostrobothnian material showing similar features/changes, so I don't have much to compare it with either

[freknu]

Hm, I don't think I can help. I'll keep skimming the thread if you find more info though.

strong adjectives are indefinite, weak adjectives are definite.

This reminds me of something found in Slavic languages. I don't know if that helps you much. But there are predicative and attributive forms; there's a specific term for this and I can't remember specifically what it is. Sometimes they encode definiteness, so you could call them definite and indefinite forms, but that's not always accurate, and I don't think that's the main term for them.

Here's a discussion you might find relevant:
http://forum.wordreference.com/showthread.php?t=2294862

I don't know if that implies that they go back to PIE or perhaps there was contact or perhaps they just are different/separate developments. I also don't know if they are actually similar in any of the details.

Edit: Aha! It's just "long" and "short" adjectives. I knew it was a generic term, but with a specific technical meaning.
Here's one reference on it:
https://benjamins.com/#catalog/books/la.153.02bab

More info:
http://forum.wordreference.com/showthread.php?t=1481230
http://en.wikipedia.org/wiki/Old_Church_Slavonic_grammar#Adjectives

The use of the 3rd person jь is reminiscent of the Old Norse definite article and suffix: (h)inn, -inn; but is unfortunately not relevant to the issue at hand

[freknu]

Actually, it probably doesn't require a strong comparative at all.

It just as well have come from *hwītōzanuN → *hwītaʀan- → hvítaran → hvítán. I guess I was too hung up on Old (West) Norse, and not realising that some peripheral dialects could just as well have preserved an archaic form that was lost in later languages.

It probably even makes more sense since it doesn't require a reanalysis of already well-established theories.

*hwīt-ōz-ô → *hwīt-aʀ-a-
*hwīt-ōz-anuN → *hwīt-aʀ-an-
*hwīt-ōz-ini → *hwīt-aʀ-an-
*hwīt-ōz-iniz → *hwīt-aʀ-an-

Not to mention it explains why I can't find a seemingly weak adjective stemming from Old Norse hvítari (cf. Sv. vitare).

Seems I was barking up the wrong tree, again!