30 
David B. Kronenfeld 
Anthropos 86.1991 
father (or real mother) from whom one may get 
something (such as an inheritance or a pre-death 
gift), and one’s father’s brother (or mother’s sister) 
from whom one is unlikely to get anything of the 
sort. I suspect that maame spreads a little more 
widely as a label than does papa because there is 
no relevant inheritance distinction among sisters 
and because a set of sisters is quite solidary and 
thus - for both of these reasons - more likely 
to be a unit, even from the point of view of 
the respective women’s children, than is a set of 
brothers. 
In this paternal (or parental) inheritance, re 
al siblings (G°) are part of the set of potential 
inheritors, while relatively extended siblings are 
not, and therefore there is some tendency to use 
the English sibling terms. The reason for the rar 
ity of English in G“ 1 is not clear to me, unless 
the particularization of nuclear family relations 
is achieved indirectly by the use of individual 
names instead of by the use of a new English 
kin term. In any case, the usage in this system 
of English borrowings is consistent with the usage 
within the set of traditional kernels — where G _1 
is less divided than G +1 - and is consistent with 
Greenberg’s general findings concerning marking 
relations (1966: Ch. 5). 
As I have said elsewhere, the use of English 
kin terms does not seem to me to be evidence 
that the expansion-reduction rules of the unskewed 
variant of the basic Fanti terminology are an En 
glish loan. These borrowings do perhaps indicate a 
shift in the kernel paradigm from a semi-Hawaiian 
(or generational) type of system to an Eskimo-like 
system. I infer the direction of change from the 
use of loan words, and from my impression that 
old people used these loans much more rarely 
than did young people. Unless the matrilineages 
or matrilineal inheritance breaks down (of which 
I see no evidence in the village) one might expect 
the Fanti terminology eventually to evolve into a 
kind of skewed Eskimo system. 
Conclusion 
Kinship terminological systems can be formally 
or logically analysed without regard for other as 
pects of the culture or language in which they are 
embedded. At the same time, such purely formal 
analyses of such a narrowly delimited domain are 
of only limited interest of students of language 
or culture. If we are interested in the social or 
economic conditions that occasion such terminol 
ogies, or in the human cognitive abilities that shape 
and constrain them, then we need more informa 
tion about the cultures (and languages) in which 
they are embedded, and we need to examine in 
detail the relationship of the axioms of our various 
formal analyses to the relevant attributes of the 
cultures in question. I have extensively addressed 
such functional questions in my previous work on 
Fanti kinship. But, beyond such attention to the kin 
terminology proper, we have also to see what other 
semantic and cultural systems routinely interact 
with kinship terms (narrowly defined) to produce 
the conceptual units (Frakean “segregates”) that 
native speakers think in terms of and that form 
part of their routine reasoning about such units. It 
is to this latter task that the present paper has been 
addressed. 
For Fanti kinship in particular, we have ex 
plored the extra-kin-terminological definitions of 
siblingship that distinguish common lineage-mem 
bership from lineage-relevant relations from other 
usage, the effects and usage of sets of optional but 
ubiquitous seniority and sex markers, the complex 
system of inheritance rights, based on siblingship, 
which both partially occasions the use of the se 
niority markers and which biases the assessment 
of seniority. We have concluded with an exami 
nation of the specific ways in which changes in 
the system of inheritance rights which structures 
much of the terminology are being reflected in the 
terminology. The topics treated in this paper are 
of only limited interest by themselves, in isolation 
- their importance lies in the role they must play 
in any full understanding of the functioning of the 
Fanti kin terminology that we may aim at. 
The extra-kin-terminological features discuss 
ed here are specific to Fanti kinship. The general 
issues suggested by this discussion include deter 
mining what are the comparable ancillary attri 
butes of other kin terminological systems, and then 
beginning the process of developing a compara 
tive ethnological framework for more effectively 
studying the functioning of kinship terminologies. 
The research on which this paper is based was supported 
by Public Health Service Fellowship No. Fl-MH-34, 
116 and Grant No. RG4-MH13769 from the National 
Institute of Mental Health. Research assistance was also 
provided by the Academic Senate of the University of 
California. 
I wish also to thank Andrew B. Ashun, Hugh Glad 
win, Judy Z. Kronenfeld, and Naomi R. Quinn for their 
help with the original research and Eugene N. Anderson, 
Alan R. Beals, and Judy Z. Kronenfeld for their helpful 
critical comments on the present paper. I am grateful,