[Lingtyp] odd clitic behaviours

[Martin Haspelmath]

In order to "evaluate whether an element is a clitic in the first 
place", one needs a way to recognize clitics uniformly: a definition.

Here's a proposal for a definition of "clitic":
A clitic is a bound morph that is not an affix ("affix" is defined in 
Haspelmath 2021).

On this definition, most Indo-European prepositions and particles would 
be clitics, which perhaps does not correspond to our intuitions. But a 
short term like "clitic" should probably have a short definition, so I 
kind of like this proposal.

Clitics are often said to be "phonologically dependent" on their host, 
but the notion of "phonologically dependence" is notoriously vague, so 
it can hardly serve as a basis of a clear definition.

At the end of their 2012 book on clitics, Spencer & Luís say: "while the 
category of clitic may not exist, some sort of concept of clitic remains 
ubiquitous [...] as an umbrella term. [...] the term usefully points to 
elements which cannot easily be classified as normal affixes or normal 
function words."

This quotation shows how little basis there is for any "natural" 
definition of "clitic". This leads me to propose the novel definition as 
a non-affix bound morph (where "bound" is meant in the Bloomfieldian 
sense, not in some vague phonological sense). The crucial difference 
between affixes and clitics is thus that clitics are promiscuous, but 
affixes always occur on forms of the same class (noun, verb, or adjective).

Best,
Martin

Reference:

Haspelmath, Martin. 2021. Bound forms, welded forms, and affixes: Basic 
concepts for morphological comparison. /Voprosy Jazykoznanija/ 2021(1). 
7-28 (https://zenodo.org/record/4628279)

Am 06.12.21 um 07:09 schrieb Alexander Rice:
> thanks to all for the comments, they are well taken, I'm going through 
> them and I may follow up later with some replies
>
> for now though, as some have pointed out, it's worth reevaluating 
> whether these clitics are even clitics in the first place, I'll have 
> to look into that
>
> On Tue, Nov 30, 2021 at 11:49 PM Alexander Rice <ax.h.rice at gmail.com> 
> wrote:
>
>     Dear typologists
>
>     I'm working with a variety of Quechua, I have a set of three
>     morphemes. They and their equivalents in related varieties are
>     traditionally analyzed as evidential enclitcs or suffixes.
>
>     However in some data that I've been working with recently I've
>     noticed a couple of interesting behaviours of these enclitics:
>
>     1) They sometimes manifest as pro-clitics but only on the copular
>     verb and in a much more phonologically reduced from
>
>     2) At least one of the three appears to manifest as a
>     phonologically independent "word'. A native speaker with whom I
>     work sometimes transcribes the clitic as a separate word, and upon
>     my review of the recordings, many of these do appear to be
>     phonologically independent from what would usually be the
>     phonological host, and in some instances, they occur at the
>     beginning of an intonational unit.
>
>     I wonder if any of you have encountered or know of similar
>     phenomena, any references would be most appreciated.
>
>     Best,
>     Alex
>
>
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp

-- 
Martin Haspelmath
Max Planck Institute for Evolutionary Anthropology
Deutscher Platz 6
D-04103 Leipzig
https://www.eva.mpg.de/linguistic-and-cultural-evolution/staff/martin-haspelmath/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20211206/202ff274/attachment.htm>