LINGUIST List 5.1286

Sat 12 Nov 1994

Disc: Typological classification, Word order

Editor for this issue: <>

Directory

1. , Re: 5.1254 Typological classification
2. Bill Croft, Typological method
3. , Re: 5.1237 Sum: Basic word order (and remarks on typology)

Message 1: Re: 5.1254 Typological classification

For what it is worth, I disagree with Martin Haspelmath (and agree
with Fritz Newmeyer) about the problem of defining the concepts
with which typological work operates. But we need to make sure that
this does not become a political thing: I don't see any difference
on this point between the work of typologists/functionalists as
compared to that of formalists. EVERYBODY who tries to compare two
or more languages has these problems. For ex., in reference to the
basic word order question, I noticed a long time ago that some languages
which were claimed to have OVS as basic actually rarely had both O and
S at the same time in the same sentence, so I argued that maybe the
term "basic" should not be applied there in the way in which it applies
to English SVO patterns. There are many many examples where we compare
incomparables and do not compare comparables because our concepts are
vague and our terminologies are ambiguous. I have, for example,
published some papers documenting the confusions surrounding the
term 'topic' in the typological literature. There seems to be
a lot of confusion likewise about the concepts of 'ergative' and
'passive'. And it seems to me that typological categories such
as 'configurational' (or non), 'pro-drop' (or non), and so on, are
just as poorly defined and just as liable to lead to all kinds of
confusion.

Pace Fritz Newmeyer, there has been extensive discussion of the
methodological issues that he raises. (These remarks are independent
of any assessment of Johanna Nichol's book that Fritz referred to,
incidentally.)

 First, good typological work devotes a substantial amount of
effort to defining and delimiting what sorts of constructions
should or should not belong to a particular structural type. Good
examples are many of Dryer's papers (e.g. Dryer 1989a, 1989b, 1992),
Koptjevskaja-Tamm 1990:ch. 1-3 on nominalizations, Stassen
1985:ch. 2,4 on comparative and 'chaining' constructions, and
Haspelmath 1993:ch. 2 on indefinite pronouns. More generally, the
discussion of the proper typological markedness criteria in Croft
1990:ch. 4-6 is an attempt to provide some *general* structural criteria
for a large class of cross-linguistic phenomena (see for example the
formulation of the structural consequences of Hopper & Thompson
1980's transitivity hypothesis on p. 131, or the application of the
criteria to the syntactic category problem in Croft 1991:ch. 2-3).
Delimiting and classifying the phenomena to be analyzed is not a
simple descriptive task; in fact, it raises some of the most interesting
issues in typological analysis, and have hardly been ignored in the
typological literature.

 Second, with respect to basicness for a language type in particular,
in the passage from Croft 1990 that Fritz referred to (pp. 33-36), I
argued that typologists had moved from classifying languages
typologically to classifying constructions typologically, and that that
was a good thing. Nevertheless, I suggested some general criteria for
determining a "basic" language type. And despite my reservations
about identifying "basic" language types, it has to be said that in the
great majority of cases, defining a "language type" instead of a
"construction type" is not terribly difficult. Matthew Dryer (p.c.) has
observed that inspection of texts generally tells you very quickly what
the "word order type" of a language is. In my own experience, in a
study where I am comparing the syntax of prenominal and
postnominal modifiers, I am quite frustrated at how difficult it is to
find languages where a modifier freely occurs both pre- and
postnominally. Of course, identification of "basic" or other types
depends on what resources you have available to you (quality of
grammatical descriptions, availability of texts/language consultants
etc.) as well as what phenomenon you are studying (e.g. word order or
head marking [easier] vs. relative order of modifiers or the syntax of
nonrestrictive modification [harder]). But for most studies it is not an
insurmountable problem, although there always some problematic
cases.

 Third, the sampling problem *has* been discussed at some length
by typologists (besides Bell 1978's seminal article, see also Dryer
1989c, Perkins 1989, Croft 1990:18-25, Rijkhoff et al. 1993). The
issues are too complex to go into here at length. But it should be
pointed out that the issues Fritz raises apply to different sorts of
samples with different purposes in mind. The question of how
independent particular instances are have to do with probability
samples (see in particular the Dryer, Perkins and Croft references for
discussion of these problems). Note in particular that the stability of
the phenomenon being studied is an important factor: the less stable it
is, the more likely cases are historically independent. The question of
coming across rare types pertains to variety samples (see in particular
Rijkhoff et al.). Some studies (e.g. Tomlin 1986, Koptjevskaja-Tamm
1990, Haspelmath 1993) draw on a very large variety sample and then
select a stratified probability sample from it for the appropriate
generalizations. Dryer's sampling technique attempts to combine the
needs of variety and probability samples into a single sampling
procedure.

 I should also add that with the shift towards dynamic (diachronic)
interpretations of synchronic typological patterns, the examination of
cognate phenomena in related languages, and the comparison of
"nonbasic" as well as basic typological strategies, becomes another
important method of analysis. That means that to some extent, the
problems of basic vs. nonbasic types in a single language, and of
historical relatedness of data sets, can be avoided---in fact, exploited---
in modern typological analysis.

Bill Croft

References:

Bell, Alan. 1978. Language samples. Universals of Human Language,
Vol. 1: Method and Theory, ed. Joseph H. Greenberg, Charles A.
Ferguson and Edith A. Moravcsik, 123-156. Stanford: Stanford
University Press.

Croft, William. 1990. Typology and universals. Cambridge:
Cambridge University Press.

-. 1991. Syntactic categories and grammatical relations: The
cognitive organization of information. Chicago: University of Chicago
Press.

Dryer, Matthew. 1989a. Plural words. Linguistics 27.865-95.
-. 1989b. Article-noun order. CLS 25.84-97.
-. 1989c. Large linguistic areas and language sampling. Studies in
Language 13.257-92.
-. 1992. The Greenbergian word order correlations. Language 68:81-
138.
Hopper, Paul and Sandra A. Thompson. 1980. Transitivity in grammar and
discourse. Language 56.251-299.

Koptjevskaja-Tamm, Maria. 1988. A Typology of Action Nominal
Constructions. Stockholm: University of Stockholm.

Perkins, Revere D. 1989. Statistical techniques for determining
language sample size. Studies in LanguageA013:293-315.

Rijkhoff, Jan, Dik Bakker, Kees Hengveld & Peter Kahrel. A method
of language sampling. Studies in Language 17.169-203.

Stassen, Leon. 1985. Comparison and Universal Grammar. Oxford:
Basil Blackwell.

Tomlin, Russell. 1986. Basic Word Order: Functional Principles.
London: Croom Helm.

Dept of Linguistics, U Manchester, Oxford Rd, Manchester M13 9PL,
 UK
w.croftmanchester.ac.uk FAX: +44-61-275 3187 Phone: 275 3188

Message 3: Re: 5.1237 Sum: Basic word order (and remarks on typology)

Fritz Newmeyer raises some important issues about typological research
as instantiated in Nichols' work and elsewhere, which call for some
comments. First, the notion of language type is an idealization:
this is often made clear in typology courses, where non-existent
artificial languages are considered as logically possible or
impossible language types, if not always in the published literature
(though the point is clearly made in Comrie's textbook, for example,
by the discussion of morphological typology). Similarly, Fritz is
quite right to point out that the assumption of no areal or genetic
bias, to the extent that it is made at all, is a counterfactual one.
What I would ask is whether such idealizations differ in kind from
those, many of them counterfactual, which underlie generative grammar.
The justification is similar in both cases: methodologically, one
cannot deal with all the relevant variables all the time, and
some of the concepts used are logically independent of the extrinsic
variables, e.g. one can investigate consistency with proposed
implicational universals regardless of the statistical bias in one's
sample.
The term `shaky typological pigeon-holing' is particularly inapt
as applied to Nichols' distinction between head-marking and dependent-
marking languages, which is quite explicitly a matter of degree
(see her 1986 paper where the degree of HM vs. DM is quantifified
in terms of the number of patterns instantiating each type).
If Nichols refers to a `head-marking language' this is shorthand
for a quantifiable tendency, just as we might describe Italian as a
`pro-drop language' without implying that there is a binary distinction
here (in fact there are degrees of pro-drop, e.g. some languages
allow null expletives but not null referential subjects, etc).
 Another point which I think all typologists would accept is that
languages which don't appear to match an established type are typological
data - for example, colloquial French as discussed recently on LINGUIST
is a language which might reasonably be described as having no basic
word order (I believe that Nichols classified it as VSO in one of her
tables, but that is again shorthand for a more complex situation).
This is an explanandum for which Nichols' 1986 paper offers an
explanation: head-marking facilitates word order freedom, and
colloquial French instantiates clausal head-marking in Nichols' sense.
 To sum up: pace widespread opinion, typologists' goal is not
pigeon-holing ("taxonomy" is also widely used in this context!)
but the investigation and explanation of patterns of variation.

Steve Matthews