[Lingtyp] Integration of postposed vowel-initial vs consonant-initial morphemes

[VG]

Hi Martin,

I am not familiar with this distinction/generalization. As far as I 
know, there is a difference between "stress-neutral" and 
"stress-shifting" suffixes in Germanic languages (we discuss that in 
König/Gast 2018, Ch. 3). Some suffixes are just added to their bases and 
form a phonological domain of their own, e.g. "likeli-hood", 
"partner-ship", "Wahrscheinlich-keit", "Ungereimt-heit". They are 
sometimes just analysed as phonological words (and they are typically of 
Germanic origin). Other suffixes either attract stress ("defizit-är", 
"Stabil-ität") or are integrated into the phonological domain of the 
host ("solid-ify", "solid-ity"), leading to "regular" stress assignment 
(to the extent that this exists in English). They typically seem to be 
of Latinate origin.


There may be a certain correlation between the segmental form of a 
suffix and its phonological behaviour; all the German stress-shifting 
suffixes that we list in our chapter start with a vowel. But English 
also has "-s", for instance ("semantics"; well, perhaps that was 
originally a plural). Stress-neutral suffixes often, but not 
necessarily, start with a consonant ("Ungereimt-heit", "likeli-hood" but 
also "Lobpreis-ung", "apprais-al"). There is probably a connection 
between the presence of a vowel, which facilitates 
integration/resyllabification, and phonological integration; but I don't 
think that the rule can be formulated in terms of the opposition between 
vowels and consonants. (And I'm not aware of a different treatment for 
'-ig' and '-lich'). But perhaps you have a different notion of 
integration in mind? (Btw I think Nanna Fuhrhop has worked on this.)


Best,
Volker


On 27.08.2020 11:44, Martin Haspelmath wrote:
> Interestingly, the phenomenon described by Hiroto Uchihara occurs in 
> German: vowel-initial suffixes (e.g. -ig) are typically described as 
> "integrated in the prosodic word", while consonant-initial suffixes 
> are described as "non-integrated". For example:
>
> /Farbe/ 'colour'
> /farb-ig/ 'colourful' (resyllabified)
> /farb-lich/ 'colour-related' (with devoicing: [farp-lɪɕ])
>
> If by "integration" we mean syllabification, then this makes very good 
> sense, of course. But by "prosodic word", many authors mean a more 
> important domain – one that is potentially relevant to a range of 
> different phenomena (e.g. stress, assimilation, ...).
>
> The trouble is that different prosodic word criteria do not always 
> give the same results (see Schiering et al. 2010: "The prosodic word 
> is not universal, but emergent"). So testing such claims is very 
> difficult.
>
> It seems to me that in addition to the prosodic structure, it is 
> simple length (in terms of number of segments) that plays a role: 
> Longer forms have a greater tendency to remain independent, while 
> shorter forms have a greater tendency to "attach" to a host in some way.
>
> Martin
>
> Am 27.08.20 um 04:13 schrieb Tim Zingler:
>>
>> Hi,
>>
>>
>> my dissertation looks at wordhood (or rather, the problems with it) 
>> cross-linguistically, and the facts you report are among the kind of 
>> phenomena that I was looking for in grammars. My sample contains 60 
>> unrelated languages, but I do not recall a single grammar discussing 
>> such an issue at any length. So, I would venture to say that they are 
>> not commonly reported, although I should also highlight that 
>> theoretical approaches to prosodic wordhood will cite sporadic 
>> examples of this kind. (The dissertation should be done later this 
>> fall. I would be happy to send out the final version).
>>
>>
>> More generally, to the extent that the situation you describe falls 
>> within the domain of syllabification, that is a phenomenon that gets 
>> surprisingly little attention as an indicator of wordhood, both in 
>> grammars and in theoretical works.
>>
>>
>> Best,
>>
>>
>> Tim
>>
>>
>> ------------------------------------------------------------------------
>> *From:* Lingtyp <lingtyp-bounces at listserv.linguistlist.org> on behalf 
>> of Hiroto Uchihara <uchihara at buffalo.edu>
>> *Sent:* Wednesday, August 26, 2020 6:29 PM
>> *To:* Linguistic Typology <lingtyp at listserv.linguistlist.org>
>> *Subject:* [Lingtyp] Integration of postposed vowel-initial vs 
>> consonant-initial morphemes
>>
>> *  [EXTERNAL]*
>>
>> Dear all,
>>
>> I'm aware of the asymmetry between the preposed and 
>> postposed morphemes in terms of their integration into the prosodic 
>> constituent with the stem (Himmelman 2014; Asao 2015), but is anyone 
>> aware of the difference in the level of integration between the 
>> vowel-initial vs consonant-initial postposed morphemes (suffixes or 
>> enclitics)?
>>
>> I have been observing that this might be the case in a couple of 
>> languages, including Teotitlán Zapotec and Alcozauca Mixtec. For 
>> instance in Teotitlán Zapotec, vowel-initial enclitics are clearly 
>> within the domain of syllabification, while consonant-initial 
>> enclitics are not. In Alcozauca Mixtec, it might be the case that 
>> vowel-initial enclitics are incorporated into the prosodic word, 
>> while consonant-initial enclitics are not. Is this something commonly 
>> reported in the literature?
>>
>> I would appreciate any insights.
>>
>> Best regards,
>> Hiroto
>>
>> Asao, Yoshihiko. 2015. /Left-Right Asymmetries in Words: A 
>> Processing-Based Account/. Ph.D. dissertation, SUNY Buffalo
>> Himmelmann, Nikolaus. 2014. Asymmetries in the prosodic phrasing of 
>> function words: Another look at the suffixing preference. 
>> /Language/90(4). 927–960.
>>
>> _______________________________________________
>> Lingtyp mailing list
>> Lingtyp at listserv.linguistlist.org
>> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
>
> -- 
> Martin Haspelmath (haspelmath at shh.mpg.de)
> Max Planck Institute for Evolutionary Anthropology
> Deutscher Platz 6
> D-04103 Leipzig
> &
> Leipzig University
> Institut fuer Anglistik
> IPF 141199
> D-04081 Leipzig
>
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20200827/709f1e4d/attachment.htm>