True meaning of morphosyntacitc categories

Thu Jan 19 15:33:41 UTC 2006

Dear Tanya,

the question you raise is highly important (and likewise often 
neglected) in (General) Linguistics.  The best 'field' to look at is 
Cognitive Linguistics, because you ask for the 'meaning' of *linguistic* 
categories, that is for their cognitive 'reality'. On the one hand, you 
should recall that 'linguistic' terms and the categories labeled by them 
have been set up during the curse of linguistic research since the times 
of Ancient Greek, Classical Arabic, and Sanskrit grammar traditions. 
Many of them have been handed down by generations of linguists and 
(before) philosophers of language) - hence it is not surprising that 
their 'semantic value' is everything but 'secure'. In general, it is 
highly problematic to derive the functional value of a morphosyntactic 
category from its traditional label. You will probably know - taken as 
an example - the history of the 'accusative', which has come about 
because of a false translation of the underlying Greek terms by Latin 
grammarians. The same holds for instance for the Latin term 'kasus' > 
'Case'. On the other hand, we should be aware of the fact that 
linguistic labels have a basically heuristic value. Most of them have 
been set up to denote a class of common behavior, properties or the 
like, described from the perspective of a specific theoretical or 
practical point of view. Nevertheless, some of the categories formulated 
in this type of Basic Linguistic Theory (BLT, to take up a term by Bob 
Dixon) seem to include more than just an arbitrary name for a 
*linguistic* category. Here, both folk linguistics (and the schooling 
tradition) and serious analytic considerations have contributed to the 
assumption that even *linguistic* labels tend to mean something. In 
order to approach this layer, we should start from the hypothesis that 
cognitive categories (in terms of disjunctive, radial, or prototypical 
categories, family resemblance, output of metaphorization processes 
etc.) in fact *may* be mirrored in 'linguistic' categories. In other 
words: The meta-description (linguistic categories) is said to reflect 
(in parts) a cognitive reality. It is one the ambitious goals of 
cognitive linguistics to disclose this correlation. Still, one should be 
very cautious: Not all linguistic categories really have 'meaning' (in 
the sense of cognitive semantics), and others may have a meaning which, 
however, is *not* reflected by the term used to label the category. If 
you browse through the present-day literature on Cognitive Linguistics 
(especially in the field of Cognitive GrammarS (!)) you will find a 
plentitude of examples that discuss the 'semantic' layer of a given 
category. However, note that quite often linguistic labels have to be 
'translated' into the meta-language(s) of Cognitive Linguistics before 
further studies become possible. For instance, it comes clear that from 
the point of view of Cognitive Linguistics, prepositions and case 
categories are subsumed under the label 'cognitively relational' 
(relating them to verbs), whereas postposition and verbal agreement are 
subsumed under the label 'cognitively referential' (relating them to 
nouns etc.). Plurals represent a derivational feature (reflecting the 
different types of referentiality) rather than an inflectional one, even 
if they fuse with (relational) Case. The best way to check whether a 
'linguistic' term matches a cognitive reality (or: has a 'meaning') is 
to analyse the functional and semantic domains expressed ALL the members 
of the corresponding category  and to see whether their prototypical, 
radial, or hyperonymic 'basis' is in accordance with the 'name' of the 
linguistic category. Recall that this procedure - that can be reinforced 
with the help of studies in the grammaticalization of those morphemes 
that are involved in the production of a category - firstly is 
language-dependent. Only VERY few linguistic categorial labels hold as 
universals, and if they do, their semantics is often strongly bleached 
or underdetermined. To sum up (and this is what I always tell my 
students): Take the names of linguistic categories just as what they 
are: Arbitrary (and traditional) names of common paradigmatic, 
structural, or constructral behavior (forming paradigms). They hardly 
ever tell us about the cognitive 'semantics' (or: functions) of the 
given paradigm etc., but they motivate researchers to look for just the 
semantics or functions from a cognitive point of view. It's always in 
cognition, where semantics takes place (be it lexical, grammatical, 
phonological, pragmatic or what so ever), but not in the act of naming 
linguistic paradigms from a mere descriptive (and traditional) perspective.
Best wishes,
Wolfgang

tanya at ruc.dk schrieb:

>Dear Funknetters,
>usually just a silent but interested reader in the linguistic debates of this
>forum, I now ask for your help. More specifically I would appreciate any
>references on investigations into the meaning/functional content of
>morphosyntactic categories, such as tense, case, number, gender, etc. I find
>plenty of material on syntactic categories and the like, but precious little on
>the 'real' meaning of case and others. Danish linguist Louis Hjelmslev
>contributed greatly to this field of study, but I would very much like to know
>if you are familiar with any other theories/investigations/approaches...
>
>The question within this line of thought, is whether the established terms for
>morphosyntactic categories are truly appropriate for the 'job' they do: the
>most obvious question seems to be whether the different genera really have to
>do with different genders, biological or cultural, as the terms 'masculine' and
>'feminine' would suggest; and if not, what _do_ they mean? But also the term
>'plural' in the category of number is misleading: are there really more than
>one item when we talk of _trousers_, _scissors_, _glasses_ ? And even more
>complex, I would say, what about mood and case?
>
>I will of course post a list of references to anyone interested.
>
>Eagerly awaiting your replies,
>
>Tanya Christensen,
>graduate student,
>Copenhagen, Denmark
>
>
>
>
>
>  
>

-- 
#############################
Prof. Dr. Wolfgang Schulze
Institut für Allgemeine und Typologische Sprachwissenschaft  (IATS)
[General Linguistics and Language Typology]
Department für Kommunikation und Sprachen / F 13.14
Ludwig-Maximilians-Universität München
Geschwister-Scholl-Platz 1
D-80539 München
Tel.:     ++49-(0)89-2180 2486 (secretary)
             ++49-(0)89-2180 5343 (office)
Fax:     ++49-(0)89-2180 5345
E-mail: W.Schulze at lrz.uni-muenchen.de
Web: http://www.ats.lmu.de/index.php