Morgan vs. Dorsey on the Omaha Cross/Parallel Contrast : Theoretical Implications - article ; n°109 ; vol.29, pg 76-106

David B.Kronenfeld

Description

L'Homme - Année 1989 - Volume 29 - Numéro 109 - Pages 76-106
31 pages
Source : Persée ; Ministère de la jeunesse, de l’éducation nationale et de la recherche, Direction de l’enseignement supérieur, Sous-direction des bibliothèques et de la documentation.

Sujets

• Sociologie, société et politique

Informations

• Publié par: L-HOMME0
• Publié le: 01 janvier 1989
• Nombre de lectures: 6
• Langue: English
• Poids de l'ouvrage: 3 Mo

Extrait

David B. Kronenfeld
Morgan vs. Dorsey on the Omaha Cross/Parallel Contrast :
Theoretical Implications
In: L'Homme, 1989, tome 29 n°109. pp. 76-106.
Citer ce document / Cite this document :
Kronenfeld David B. Morgan vs. Dorsey on the Omaha Cross/Parallel Contrast : Theoretical Implications. In: L'Homme, 1989,
tome 29 n°109. pp. 76-106.
doi : 10.3406/hom.1989.369084
http://www.persee.fr/web/revues/home/prescript/article/hom_0439-4216_1989_num_29_109_369084o
David B. Kronenfeld
Morgan vs. Dorsey
on the Omaha Cross/Parallel Contrast :
Theoretical Implications1
their David with Theoretical the associated B. aid Kronenfeld, Implications. of formal kintypes techniques Morgan is — compared Lewis developed vs. Henry Dorsey with Morgan's that by on F. of Lounsbury the J. list O. Omaha Dorsey. of Omaha and Cross The A. Indian /Parallel K. comparison, Romney, kinterms Contrast reveals made and :
important systematic differences which derive from two different definitions of cross vs.
parallel relatives. Dorsey' s data entail a Dravidian-type cross/parallel feature while
Morgan's entail a combination of Iroquois-type and Dravidian-type. For a variety of
reasons it is concluded, for the Omaha, that Dorsey's data are correct. Possible bases for error are discussed, and related to the general issue of social situations in
which Iroquois-type terminologies, as opposed to Dravidian, might be expected to arise.
Lewis Henry Morgan's Systems of Consanguinity and Affinity (1871)
remains to this day, over a hundred years after its first publication, the major
compendium of Amerindian kinship terminologies. Its tables contain data from
more systems than any other source, and for each of those systems, with only a
very few exceptions, its data are more complete and more detailed than anything
that has appeared since. The lack of subsequent publication of any extensive
data relating kintypes to kinterms for most of these systems is a tribute to the
exhaustiveness of Morgan's endeavor. The lack, unfortunately, also has the
effect of making it very difficult for us to fully evaluate the detailed accuracy of
Morgan's data.
In this paper I want to use one of the few independent publications of data
that parallels Morgan's data in its completeness for a system also treated by
Morgan to examine the quality of his data. The comparison of Dorsey's
description of the Omaha terminology (1884) with Morgan's is especially
apropos since the two publications are roughly contemporaneous. The goal of
the comparison is not to « debunk » Morgan. His theoretical failings have long
been clear, if forgivable in their historical context, while the scope and — in
many ways — the quality of his data collection continue in many ways to tower
over all who have come since.
L'Homme 109, janv.-mars 1989, XXIX (1), pp. 76-106. Morgan vs. Dorsey 11
The comparison will reveal one major flaw in Morgan's data as well as some
minor imperfections (such as occasional sloppiness and a few bad analytic or
organizational decisions). Examples of include mixing up the
referents of some terms and sometimes being inconsistent in his phonetic
renderings, while bad decisions include representing terms phonetically without
indicating which alternates are variations of a single term and which similar but
different terms and, in a few places, dispensing with his generally tight
parent/child/sibling kintype specifications in favor of broader but vaguer ones.
These imperfections, while of historical interest, do not have any great theoret
ical import — even if rectifying them might somewhat improve our disciplinary
data base.
In contrast to such minor imperfections there is one major flaw which, while
hardly vitiating the general usefulness or importance of Morgan's data, has
important ethnological and theoretical significance. The flaw consists of
Morgan's imposition of an Iroquois-type cross/parallel distinction on a
terminology for which Dorsey correctly shows a Dravidian-type distinction. The
existence of the flaw will be demonstrated but the scope or breadth of its
occurrence (i.e. in the other terminological systems described in Morgan's
compendium) will be left much more in doubt by the lack of adequate
alternative data sets for other languages and cultures similar to that published
by Dorsey for the Omaha. This flaw, and the ambiguity concerning its scope,
show why, however much we respect Morgan and his work, we cannot afford
simply to rely on it without generating independent replications from other
sources.
I will first present a formal semantic analysis of the the consanguineal
kintypes of the Omaha kinship terminology (the real thing, not just the type) as
described by Dorsey (1884) and then a similar analysis of Morgan's (1871)
Omaha data. After explaining the form of the comparison and its rationale, I
will compare the two analyses to reveal an interesting and theoretically
important difference between the two. After a consideration of the relevance of
the comparison to other systems described by Morgan I will conclude with a
discussion of the implications of the comparison for Morgan's role in kinship
studies, for our comparative understanding of the social bases of kin
terminologies, and for the adequacy of our general kinterminological data for
basic ethnological purposes.
In this paper I will be distinguishing the Iroquois (vs. Dravidian or Omaha)
language, culture, or people from Iroquois-type (vs. Dravidian-type or Omaha-
type) terminologies and from (vs. Dravidian-type) cross/parallel
features or distinctions. The question of whether or not the Iroquois people
have an Iroquois-type kinship terminological system will be addressed later in
the paper ; it is not logically required that they have one. Iroquois-type and
Dravidian-type cross/parallel distinctions represent two alternative ways of
distinguishing cross from parallel relatives2. Iroquois-type terminologies are
distinguished from Dravidian-type terminologies by which cross/parallel DAVID B. KRONENFELD 78
Key for Figures 1 and 4
Morgan' s Dorsey' s My
Ref# Term Kernel Term Ref Kernel
wüiga" wee- te' -ga 1 a+m+m B same as Morgan
wee' wi>1an 2 -kä a+m+f
in- da -de 3 a+m Indadi A
e-nä' -ha 4 a+f lnnanha a
we-nis-se 5 m-m (see my # 22)
38 we-zhin-go f-m (see my # 22)
22 wee-Ehin-ga C a-m fof -m wij iñge
we-zhun' -ga 6 a-f Ö same as Morgan wij añge we- tush' wi^nepa D ,P " 7 -pä a-a-a wee-zhe' 8 -thä mome wij ince
wee- te' 9 -noo fome wi^inu 9 E ton' 10 -gä fofy, mof wilañge F
wee- ton- tha 42 f+fom-f-f (see my # 10, and cf. my # 11)
wee-Eon- thä Oí 11 fofe wijance
wee-sön' -gä 12 wisañga F aomy
we- ' -na 16 te a-m=f wilini Ó
we- ton- da d 17 a-f=m wi^ande
wee-toans ' -kä 18 wilucka I fom-m bob-m
19 (see my # 18) wi1ancka G mof -m
we- te' -Ehä 20 bob-f wiîujange ^) fora-f
wi ijan 21 (see my # 20) J mof-f
we-hun' -gä wihañ' ga 23 mom=f S same as Morgan " ' -kä 24 we -she fom=f wici^lan / wiláha" we- tä' " 25 -ha mo f-m a
' -me wee- tee 26 a+mof wilimi y
wee-nä' a+fom winegi H 27 -gee
wee- a' ** " wiegcañge 28 -gron-kä f-m wigaqca" * " wee - gö - thnough m=f 2? ashe-ah' 34 -ga m=f+m (see # 1) my gah - ' 35 ah m-f+f (see my # 2)
we - she ' - ä wici' e e " 30 f=mom
Notes :
▼ all males are given, erroneously, as #27 and all females as #20/21
-f- # 42 is possibly a typographical error for # 10
-| — |- presumably an error- -should be # 22 I
I
I
I
I
I
Morgan vs. Dorsey 79
8 21 19 2 26 3 6 22 21 19 4 2 4 271 10 12 2 II 4 27 10
2 2 2
A= O O= A AO EG0 29 23 28 25 16
k I k 6 22 6 25 23
=O A-O O=A 7 19 21 12 10 12 23 EGO29 l0 25 l2 l0
7 7 7 7 22 6 19 21 22 6 22 6 19 21 22 6 19 21
ÁO A 30 24 24 EGO 28
22 6 ft è ^ A-£
7 7 7 7 18 20 22 6 18 20 22 6 22 6
Fig. 1 . OMAHA KINSHIP TERMINOLOGY — DORSEY DATA
distinction they embody ; but any other type of terminology which involves a
cross/parallel distinction, such as an Omaha-type or a Crow-type terminology,
will have to embody some specific cross/parallel distinction-whether it be
the Dravidian-type distinction, the Iroquois-type one, or some other. The 80 DAVID B. KRONENFELD
interaction of type of cross/parallel distinction with skewing to define
terminological type is as follows :
Type of Cross/Parallel Distinction
Iroquois-type Dravidian-type Other
Unskewed ?
System System
Patri-skewing Omaha-type System
Matri-skewing Crow-type
TERMINOLOGIES
Figure 1 reproduces Dorsey's data on the Omaha Terminology (Dorsey
1884 : pi. xxxH and xxxin). The set of reduction rules3 given below will reduce
all consanguineal kinterm ranges to single kernel kintypes and account for all
the consanguine terminological equivalences shown in Dorsey's figure.
-> / + b/ Rulel. / + bob/ merging rule
-> /of-/ 2. / + mof-/ skewing rule