I received very few answers on definition of TYPOLOGY

[Edith A Moravcsik]

As I see it, the concept of typology is a META-SCIENTIFIC concept: I think in all sciences, the goal is not only to describe the phenomena within a given domain but also to determine their distribution: under what conditions they occur. This requires finding answers to the following questions:

    (a) What are possible properties of the phenomena within that field and 
        what are impossible ones?
    (b) Given the possible properties, with what degrees of probability do 
        they occur (i.e., which are more likely to occur and which are 
        rare)?
    (c) From among the possible properties, which are not only possible and 
        not only probable but actually necessary?

In LINGUISTICS, there are various domains where these questions arise. Even if we describe a SINGLE language, the description will have to specify the distribution of structural properties in that language: which properties of sentences (or phrases; or discourses; etc.) are possible in that language, which are probable versus rare, and which are necessary. Some of the statements will be unrestricted (such as "In all sentences of this language, there must be (or is likely to be, or may be) a verb.") but most will be implicational - i.e., they are stated to hold under certain conditions (such as that "In interrogative sentences, the verb must (or is likely to, or may) precede the subject.")

Beyond describing individual languages, we are also interested in CROSSLINGUISTIC comparison; that is, in the distribution of structural phenomena across - rather than within - languages. This means we want to find answers to the three questions: what are possible structural properties of human languages as opposed to impossible ones; which are more likely to occur and which are less likely to occur; and which are necessary structural properties. And, again, our statements may be unrestricted ("All languages have (or are likely to have, or may have) Y.") or restricted ("All languages that have X, must (or are likely to have, or may have) Y.")

The sense in which the term "typology" has been ACTUALLY used in linguistics is limited in two ways:

   1/ The term is generally used for crosslinguistic typology (rather than for typologies of structures within individual languages).

   2/ The term is generally used for the study of those properties found to be shared by languages that are explainable not by genetic, areal, or cultural factors but by general physical and psychological properties of humans. This is so just to "make room" for historical and anthropological linguistics, which are concerned with historical and cultural explanations. 
 
In sum: as I see it, crosslinguistic typology is an instance of the meta-scientific research area of typology and more particularly, of linguistic typology. As Balthasar Bickel has repeatedly emphasized, crosslinguistic typology has to do with studying the distibution of structural properties among languages, with special focus on those shared  properties that are not due - or at least do not appear to be due - to historical and cultural factors. 

Best,

Edith Moravcsik


----- Original Message -----
From: "Yuri Tambovtsev" <yutamb at MAIL.RU>
To: LINGTYP at LISTSERV.LINGUISTLIST.ORG
Sent: Sunday, January 17, 2010 3:35:14 PM GMT -06:00 US/Canada Central
Subject: I received very few answers on definition of TYPOLOGY


Dear Typology colleagues, I am surprised. I received very few answers on definition on typology. Is it too trivial? At the same time it is quite obvious that many linguists have quite different understanding what typology is. I thought it is important to discuss this global problem on the Lingtyplist since everyday we discuss here the minute problems of typology. How is it possible to discuss anything in the branch of linguistics which is not properly defined? It is an enigma for me. Anyway, I am going to sum up the few answers which I received and publish it on this Lingtyplist. I am still looking forward to receiving you definitions of typology to yutamb at mail.ru   Remain yours sincerely Yuri Tambovtsev