rank, hierarchy, and power

Mike Lieber (U28550@UICVM.BITNET)
Fri, 6 Jan 1995 14:22:34 CST

I appreciate very much Dwight Read's thoughtful post on the meanings of
"mankind" and on gender-specific and gender non-specific kin terms. I have
one quibble--actually more than a quibble--with the way that Dright, along
with most anthropologists, use the term hierarchy. Only in our field is the
term hierarchy (and its derivatives like hierarchical) used ambiguously to
denote two very different kinds of relationship. Given the politically and
emotionally charged uses of these terms and their connotations, I would like
to see anthropological usage change to reflect the use of hierarchy in the
sciences. Hierarchy in almost other field refers to a relation with a
whole-to-part ordering, the relationship of a whole to its parts being one of
logical type. The example that Dwight gives is a classical one:

/ €
male female

This is a relationship of category to its members, the one that Russell and
Whitehead used to construct the theory of logical types in _Principia
Mathematica_. The category (whole) is distinct from its members in that it
cannot be a member of itself. This difference of logical type characterizes
any hierarchy, e.g., the nucleus is part of the cell, the cell part of the
tissue, the tissue part of the organ, etc.. This is an example of a "nested
hierarchy" with each level nested in the one above it.

Hierarchy is quite distinct from an asymmetry of rank. The sargeant outranks
the corporal, i.e., has rights over the corporal (can do to the corporal) that
the corporal does not have over the sargeant. This is not a hierarchical
relation because both the sargeant and the corporal are of the same logical
type, both soldiers. More importantly, the corporal is not part of the
sargeant. Their relation is a part-to-part relation, not a whole-to-part
relation. The nature of hierarchical constaints are, ordinarily, quite
different from the kinds of constraints inherent in rank differences. The
term, power, then usually applies very differently to each.

An example of the difference between rank and hierarchy in power relations is
the secular chief and the priest on the small island, Kapingamarangi, that I
have studied for 30 years. The secular chief had authority over others, but
as first among his peers. He could represent the community to some kinds of
outsiders. The priest at times represented the community to the gods and the
gods to the community. At other times, in specific rituals, he embodied the
community. This embodiment is an entirely different kind of animal, and its
relation is not one of rank but one of whole-to-part. On some of the
Micronesian high islands, the political system was ordered by both rank and
hierarchy. The chiefs were ranked in a series from high to low, but there
were two levels of chiefs--paramount chiefs of "states", each of which was
subdivided into "sections" or wards (like counties) that had their own sets
of ranked chiefs, section chiefs. The relations among chiefs at each level
were those of differential rank. The relations between states and sections
was hierarchical--the section was part of the state, giving the highest
chiefs authority over the sections.

Why confuse relationships of very different ordering by the same term?
This is particularly important when the analytical focus is gender relations
where terms like "power", "dominance", etc. get thrown around craelessly to
begin with. If we're going to think, talk, and conduct inquiry about these
matters (rather than just scream at each other), I would think that the
clearer we are about the analytical constructs we use, the easier it is to
think, talk, and conduct inquiry. I realize that this goes against decades
of professional usage, but just because it's common doesn't mean that it's
useful. The only thing to lose by abandoning this older usage is a muddle.

Mike Lieber