Full text of "Journal of ethnobiology."

See other formats

VIC j
5ILY

JOURNAL (of) ETHNOBIOLOGY

Maya knowledge and “science wars”

— Anderson

Eastern Sumbanese bird classification
— Forth

Knowing, gathering and eating:
knowledge and attitudes about wild food
in an Isan village in Northeastern Thailand

— Somnasang & Moreno-Black

The jaguars of Altar Q, Copan, Honduras:
faunal analysis, archaeology, and ecology
— Ballinger & Stomper

Perception and management of cassava
(Manihot esculenta Crantz) diversity
among Makushi Amerindians of Guyana

— Elias et al.

The historical ecology of Southeastern
longleaf pine and its southernmost
expression — Walker

A cladistic approach to comparative
ethnobotany — Hart & Cox

Volume 20, Number 2. Winter 2000

JOURNAL STAFF
EDITOR: ng sca “ie 1 Oi of Anthropology, Tulane University, New Orleans, LA 70118

deere penn ‘Spanish Alejandro de Avila B., A. P. 533, Oaxaca, Oaxaca C.P. 68000, MEXICO

(serbo@antequera
aici si vk STANT: mei Dudley, Department of Anthropology, Tulane University, New Orleans, LA
mul abot Dalen br edu)
BOOK REVIEW Michael K. Steinberg, Department of Geography-Anthropology, University of Southern

aine Gian ME 04038 (mstein@usm. vate edu)

SOCIETY OFFICERS
PRESIDENT: Deborah M. Pearsall, American Archaeology Division, University of Missouri, Columbia,
ee ae

PRESIDENT-ELECT: Karen R. Adams, Crow Canyon Archaeological Center, Cortez, CO
SEC CRETARY/TRE mnie aaa Virginia Popper, UCLA Institute of Archaeology, Box 951519 Fowler A-210, Los Angeles,

erratic cia ee Mollie S. Toll, Museum of New Mexico, Office of Archaeological Studies, Box 2087,
Santa Fe, NM 87.

BOARD OF TRUSTEES
M. Kat Anderson, University of California, Davis, CA
Enrique Salmé6n, Fort Lewis College, Durango, CO
Melinda A. Zeder, Smithsonian Institution, Washington, D.C.

Ex officio
Past presidents: Steven A. Weber, Amadeo M. Rea, Elizabeth S. Wing, Paul Minnis, Cecil Brown, Catherine S.
Fowler, ae seach Panne Permanent board member Steven D. Emslie. The editor, president,
president-e d conference coordinator.

at iP

EDITORIAL BOARD

Eugene N. Anderson, University of California, Riverside, CA: ethnobotany, China, Maya.
Scott Atran, CNRS, Paris, FRANCE: vecaanng ag ci cea aa coagi slow science, eal
Brent Berlin, University of Georgia, Athens, GA: ethnobiological classificat
Robert A. Bye, Jr., ay Botanico, Universidad Satan Aut6noma de ac Aides, DE. ek ethno-

otany, Mex
H. Sorayya Carr, El ann California: zooarchaeology.
Nina Etkin, University of Hawaii, Honolulu, HI: medical ethnobotany, the Pacific.
Gayle J. Fritz, Washington University, St. oats MO: eae hae
Terence E. Hays, Rhode Island College, Providence, RI: ethnobiology, Papua New Guinea.
Chris Healey, Northern Territory University, Darwin, AUSTRALIA: ethnozoology, Australia and New Gui
Timothy Johns, Macdonald College of McGill whites Quebec, CANADA: chemical ecology, a East
Africa

Harriet V. Kahuilediy, McGill ei Quebec, CANADA: ethno/human nutrition, First Nations cf Canada.

David L. Lentz, New York Botanical Garden, Bronx, NY: paleoethnobotany, Mesoamerica, Central A

Brien A. Meilleur, Center for Plant eon game Missouri Botanical Garden, St. Louis, MO: mca plant
conservation, aR

Naomi Miller, University of P srl wet Philadelphia, PA: paleobotany, ethnobotany, Near mssrhg Archaeology.

Gary Nabhan, Arizona Scie Cnet useum, Tucson, AZ: ethnobiology, Sonoran desert cultu

Darrell A. svi Oxford Centre for the Environment, Ethics, and peo Oxford University, esteae ENGLAND:

tatural resource management, ethnoecology, tropical cultural ecolo

Amadeo M. Rea, San Diego, CA: cultural e per zooarchaeology, ieee

Elizabeth J. Reitz, University of Georgia, Athens, GA: zooarchaeolo

Mollie S. Toll, Museum of New Mexico, Santa re, NM: prehistoric and historic ethnobiology.

The Journal of Ethnobiology is published semi- annually. Manuscripts for publication and book review sections should

sent to the appropriate editors as listed on the inside back cover of this issue.

a Drawing of Sumba Hornbill by Chris Healey, Visiting Fellow, North Australia
esearc

enna ant cn tery pie

Journal of
Ethnobiology

VOLUME 20, NUMBER 2 WINTER 2000

Si vners sete.
pie a

Aiea

CONTENTS

ETHNOBIOTICA iv

MAYA KNOWLEDGE AND “SCIENCE WARS”
Anderson 129

EASTERN SUMBANESE BIRD CLASSIFICATION
Forth 161

KNOWING, GATHERING AND EATING: KNOWLEDGE AND ATTITUDES
ABOUT WILD FOOD IN AN ISAN VILLAGE IN NORTHEASTERN
THAILAND

Somnasang & Moreno-Black 197

THE JAGUARS OF ALTAR Q, COPAN, HONDURAS: FAUNAL ANALYSIS,
ARCHAEOLOGY, AND ECOLOGY
Ballinger & Stomper 223

PERCEPTION AND MANAGEMENT OF CASSAVA (MANIHOT ESCULENTA
CRANTZ) DIVERSITY AMONG MAKUSHI AMERINDIANS OF GUYANA
Elias et al. 239

THE HISTORICAL ECOLOGY OF SOUTHEASTERN LONGLEAF PINE AND
ITS SOUTHERNMOST EXPRESSION
Walker 269

A CLADISTIC APPROACH TO COMPARATIVE ETHNOBOTANY
Cox

Hart & 303

BOOK REVIEWS 159, 193, 217, 237, 266, 300

ETHNOBIOTICA

Darrell Addison Posey
1947-2001
Darrell Posey, 53, who had a celebrated thnok ist, activ-

ist, and specialist on Amazonia, died at Oxford, England on March oe 2001 eae inoper-
able brain tumors that had been diagnosed a few months before. Posey probably will be
most remembered for his research on Kayap6 Indian resource management and environ-
mental knowledge together with his notable activism in defending the Kayap6 and other
peoples against the Xingu Hydroelectric Project of the late 1980s. On both academic and
darts fronts, Hoey Freres seeds su doet from — loyal colengnes and friends as
assor nemies. A fi , Posey
ponies! to thrive at the center of controversy. It must be that his personal drive and many
friendships weighed more heavily in the end than the = ee —— = ond cal
saries and his disappointments in life. Other than h
Posey had attended the second meeting of the Society of Ethnobiology as well as many
meetings after that and was a member of the Editorial Board of the Journal of Ethnobiology
since the founding of the journal with volume 1, number 1 in May 1981 until his death.

Born and raised in Henderson County, Kentucky, Posey never lost the midland accent
characteristic of that region. He earned the B.S. in entomology at Louisiana State Univer-
sity in 1970, the M.A. in geography and anthropology at Louisiana State University in 1974,
and the Ph. we in anEOPEIEY at ahs sis cicn ee of ieaincer in 1979. Posey was an os
ciplinary scl linary programs in
research institutes in \ the 1980s and 1990s. He was also z a prolific writer. In terms of his pub-
lished output to date, dents authored or co-authored Eee books, edited or co-edited four
books, wrote 154 articles chapters, and produced 22 bool , some of which were
reprinted in one form or another. He directed two museum exhibits on Kayap6 resource
ponalaa ec and so eee the catalogs that accompanied them. He was also involved
in the p tary films and videos, mainly on Kayapé ethnobiology
and resource management. At the time of his death, Posey had two authored books and
three edited books in press.

Among the numerous awards Posey received for his k and activism were the “Chico
Mendes Award” for Extraordinary Courage in the Defense of Nature, given by the Interna-
tional Sierra Club in 1989 and the United Nations “Global 500 Award” for Outstanding
Achievement in Service to the Environment, bestowed on him by the U.N. Environmental
Programme in 1993. At the time of his death, he was coordinator of ECOS (Ethnoecology:
The Ecological and Social Dimensions of Well-Being) at the Institute for Social and Cul-
tural Anthropology of the University of Oxford. He also held other positions at Mansfield
College, Oxford and the Federal University of Maranhdo, Brazil. He was elected Fellow of
the prestigious Linnean Society of London in 1999 and he maintained professional affilia-
tions with Linacre College (Oxford), St. Anthony’s College (Oxford), the Institute of
Ethnobiology of the Amazon (which he had founded at Belém, Brazil), and the Carnegie
Museum of Natural History. He held numerous advisory and editorial positions with vari-
ous organizations, including the Society of Ethnobiology, the International Society of
Ethnobiology, the Global Environment Facility, and the Indigenous Peoples Media Center.

Posey rather en eny ae the ethnobiological study of Amazonia and its
peoples. Already well-known as an }

on Kayap6 folk entomology (among
other findings, his research in this area showed that the Kayap6 recognized more species
of wasps than Western taxonomy in regard to the Kayapé area, and that, indeed, one of the

species so recognized was new to science) and native resource management, having pub-
lished his research in numerous peer-reviewed journals, Posey in 1988 found himself at
the center of a professional and political crisis that would catapult him eventually to glo-
bal renown. In 1987, while serving as Director of the Ethnobiology Program at the Museu
Paraense Emilio Goeldi in Belém, Brazil, net was asked by the then Director of mak mu-
seum, Dr. Guilherme de La Penha, to und tal and social i
of indi lations in the area of influence of the Xingu Hydroelectric Project, with
unding for the shaky to be provided by Hietronorte, i a state-run electrical paver company.
Posey later said he believed he could
then in the late stages of planning, by working against it from the aide The dam project
itself, once completed on the lower Xingu River, would flood a large area of indigenous
lands and reserves, including part of the lands pertaining to the Kayap6o of the village of
Gorotire, which i is where hidressds nee ceed: out seletie of his ethnobiological research since
the time of his d 1 many friends. By early 1988, Posey
realized that his work from within would come to naught; he called it an “aborted effort.”
Later that year, he traveled with two Kayap6 Indian leaders to Washington, D.C. in order
to denounce the dam project in the offices of the President of the World Bank, who was
then intending to disburse half a billion dollars in loans to Brazil in order to finance that
project. As a result of this visit, and of the negative publicity that attended it, the World
Bank soon suspended payment of ne funds needed to drive the dam project forward. (In
spite of Posey’s efforts, the dam f been removed from national energy plans
in Brazil).

Upon his return to Brazil, Posey and the K 5 lead ted and finger-
printed on charges of harming the reputation abroad of Brazil. Soon released and awaiting
trial on the charges, and also warned by the United States Embassy in Brasilia that his life
was now in danger, Posey continued with plans to host the First International Congress of
Ethnobiology, which was held in July, 1988, at Belém, Brazil. The congress included hun-
dreds of scholars from thirty-five countries and proved to be a resounding success, in spite
of the presence of undercover federal intelligence agents, who mingled with the crowd.
The Declaration of Belém, which called for protection of native knowledge, use, and man-
agement of biological resources as well as human rights of native peoples, was one of the
results of this congress, largely thanks to Posey’s efforts. While Posey was enjoying wide-
spread endorsement for his activism by many colleagues in the international arena, the
charges against him and the two Kayap6 leaders were quietly dropped.

Posey lived mostly in Germany during 1989-91 as a Humboldt Fellow; he also founded
INEA (Institute of Ethnobiology of the Amazon), an NGO headquartered in Belém. He
eventually vacated his research position at the Museu Goeldi and became Senior Associate
Fellow at Oxford in 1992, where he remained in one capacity or another until his death.
Having confronted the Brazilian federal machinery over the Xingu Hydroelectric Project
in 1988 (and it must be said in sacs that some of Posey’s supporters were Brazilian fed-
eral and state officials who th shared sympathy for his cause), Posey became Spe-
cial Advisor to the Brazilian Special Secretary on Internal Affairs and Indigenous Peoples
during 1992. In that same year, he was Convenor and President of the Earth Parliament at
the Earth Summit in Rio de Janeiro.

From about that time on, Posey’s work was primarily aimed at promoting the intellec-
eas proper’) spa (IPR) = ect peoples while ees ef pen ~~ threats to

ti g from int d other

een interests. In 1992, Eugene Parker, who had been one : of Posey’s aaicaions.
published a scathing review of Posey’s work with the Kayap6, calling into question, in

particular, Posey’s findings with regard to Kayap6 management and fostering of the forest
islands _ — — hese eure savanna pscveaie tah Posey had claimed in
Parker argued,
essentially, that the forest islands were naturally occurring phenomena and that Posey’s
methods in determining them to be otherwise were flawed and sloppy. The debate over
the validity of Posey’s data was reminiscent of the Redfield/Lewis and Mead/Freeman
controversies from earlier generations of anthropology. Posey replied in the American An-
i = since dbaccad did not conduct his interviews in the Kayapo language, he
to negate Posey’s conclusions about the efficacy
of native seers in remaking the jandacupe under question. Parker then supplied a
rejoinder in the same journal, reiterating what he had asserted to be questionable science
on Posey’s part. Posey responded, in his defense, one more time to this critique in a chap-
ter published in W. Balée (ed.), Advances in Historical Ecology (1998). It seems likely that this
controversy between Parker and Posey will remain unsettled until further, scientifically
sound archaeological and ethnolinguistic research is carried out to determine the origins
and development of forest islands in the cerrado country of the Gorotire Kayap6.
Regardless of this inflammatory exchange in the pages of the American Anthropologist,
Posey by 1992 was devoting much more of his time to IPR. Indeed, he published three major
books on that subject in 1996 (D.A. Posey and G. Dutfield, Beyond Intellectual Property; D.A.
Posey, Traditional Resource Rights; and D.A. Posey and G. Dutfield, Indigenous Peoples and
SHS te all = — were favorably cciisihaaes a the very end of gs life, peeey s un-
published I to hi the K
The two books authored . Posey and still i in press at the time oe his death are entitled
Ethnobiology of the Kayapo Indians of Brazil and E ete ee ita nee
A full account of Darrell Posey and the infl ayapo
studies, native resource t, and activism for native land rights and IPR remains
to be written. For those who knew him i in life, Darrell Posey will be remembered for his
courage in the face of danger; his loyalty as a friend and colleague; his charismatic effect
on researchers, young and old, around the globe; and his folksy, Kentucky wit and humor
that never seemed to fail him, even in his darkest trials. He will be missed.

Journal of Ethnobiology 20(2): 129-158 Winter 2000

MAYA KNOWLEDGE AND “SCIENCE WARS”

E. N. ANDERSON
Department of Anthropolo
University of California, Riverside
Riverside, CA 92521-0418

ABSTRACT. e-RnOWIEdge! is sociatly constructed, i ha homanis succeed i ee ‘knowing
a great deal al t “science”
involve extreme positions, from claims that all science is arbitrary to claims that
science is somehow a epriiinged body of truth. : Something may be learned by
considerin ver witha long record
of high piiaeton: _ Yucatec Maya cuenchiainay agrees with contemporary
g y rere tS, almost all of them highly specific,
1. It differs in ma ts, most of them highly
inferential and cys Lape One may centatively. conclude that common
observation of everyday matters is more directly affected by interaction with the
- Tae | s bead yee pelle | Bed 4 1 4 41 4 nt o. 4

operate at all levels.

a1 Lee 1
ovr

Key words: Yucatec Maya, science wars, philosophy of science,

Yucatan Peninsula

RESUMEN.—E] El conocimiento es una construccién social, pero los humanos

proponen que “ciencia” es verdad absoluto. Seria posible conocer mucho si

investiguemos el conocimiento biolégico de una cultura, muy diferente, con una

ees larga de alta en as snclaes wine etnobioldgico de los Yucatecos
orme

en pene Gerivedas de 3 I gases pragmatica. Pero, el es deferente en otros

recnectons logical. S

tentativamente que la observacion de fenémenos concretas es mas afectada por la
interaccién con el medio ambiente que por el razén deductivo, pero que factores
sociales influyen el pensamiento en todos niveles.

—La connaissance est construée socialement, mais, aussi, les hommes
apprendrent beaucoup de leurs environs. En les debats reciens sous la nature de
“science” il y a positions extremes. Les uns propose que “science” est des chose
arbitraires; les ities Age are science’ “est la vérité absolue. C’est possible a
savoir plus de ces logique d’un culture
different—un culture que tient une hae durée” civilisée. Les Maya yucateque
possédent un sae biologique que ressemble . — ne la science

4 s ] o. hI © ch. ‘; rva ati n
, selon Derrida, “differances”) sont
des choses ‘aia ou cosmologiques. On peut concluir, tentativement, que la
observation de phenoménes concrétes est plus afectée par l’interaction avec
l’environment que le raison logical, mais les influences sociales existent en touts
niveaux

gi rt
ae as Tee Asée. /

130 ANDERSON Vol. 20, No. 2

SCIENCE WARS

Anthropology has recently seen debates concerning the nature and value of
“science.” These debates are part of a wider challenge to canons of truth, of liter-
ary quality, and indeed of all those matters that anthropologists regard as part of
culture.

As is frequent in academic conflicts, the debate over “science” has quickly
escalated, with the most famous participants being those who take the most ex-
treme positions. This has led to the term “science wars” (see the excellent account
in Hacking 1999). However, there are serious questions under the rhetoric. Lead-
ing philosophers of science disagree profoundly—though, of course, less
profoundly than the extremists of the semi-popular media—over the nature and
practice of science.

It is obviously impossible to summarize this debate here, even at a superficial
level; the present article merely makes a small contribution to the knowledge base
that underlies one aspect of the controversy, the debate on how much of science—
in this article, more specifically the classification of living things—is social
construction, and how much is based on a reality out there in the world.

Loosely arrayed on one side are those such as Imre Lakatos (1976), Thomas
Kuhn (1962), Ian Hacking (1999), and Paul Feyerabend (1987), who hold various
positions that give social construction a large role in scientific practice. They are
not a uniform group. Kuhn sees the social organization of science as structuring
the quest for truth, but is not ready to write off either the search or the goal as
hopeless. (In spite of certain claims to the contrary, Kuhn clearly states that he
regards some paradigms as more correct than others, and he sees progress in sci-
ence over time.) Hacking, also, explicitly distances himself from those who see
“science” as the construction of arbitrary nonsense, though he sees social con-
struction as important and sometimes overriding the truth. Feyerabend seems to
hold a more radical position, at least for debating purposes; he appears to see
science as a social belief system, no more believable on the face of it than witch-
craft or flying-saucer lore.

There are those who think—following Foucault (e.g. 1971) but going far be-
yond anything Foucault actually said—that, since we cannot know external truth,
all of the claims of science must be false, and must be made simply to keep elites in
power, as “truth” was constructed in Orwell’s 1984 (Orwell 1948). This is the “vul-
gar Marxist” version of Marx’ claims about religion, expanded to cover the field
that many people see as the “new religion” of “20 century” people (on these
matters see Hacking 1999).

This position depends on an inconsistency: people are seen to be living in a
completely solipsistic world in relation to the natural environment, yet to havea
perfect grasp of the realities of interpersonal power. Thus, this position, like other
radically “culturological” and culture-essentializing positions in anthropology, is
deeply incoherent. People are hypothesized to have a mystical, virtually perfect
grasp of their culture, such that insiders participate in a perfect unity that is un-
fathomable to outsiders—yet somehow this perfect learning does not extend to
any phenomena other than social or cultural ones, and somehow the visiting eth-
nographer has no way of contacting that mystic participation. The natural world,

Winter 2000 JOURNAL OF ETHNOBIOLOGY 131

in particular, is apprehended only via this mystically absorbed cultural percep-
tion. The individual humans who are so superbly good at learning from their elders
are incapable of learning from their observations.

Arrayed on the other side are a number of philosophers who see science as a
way of getting at real truth about the environment—an — that can be sub-
verted or mistaken, but, when done right, gives us pragmatically consistent and
useful data. Leaders of this general view include Philip Kitcher (1985, 1993), Larry
Laudan (1996), Alexander Rosenberg (1992), Lewis Wolpert (1993), and many more.
These too are a diverse lot, but they all agree that science is a search that produces
ever more accurate data and theories, not just a social game that produces ever
more complex arbitrary representations.

However, and notably, all these writers have abandoned the naive positivist
positions so popular in the early 20 century. No current philosopher of science
(so far as Iam aware) continues to defend the near-religious regard for “covering
laws,” “falsification,” and “objectivity” that dominated science, and confined it in
narrow channels, through much of the Tan Hacking and Philip Kitcher
provide especially sober and thoughtful critiques of this position (rather unfairly
blamed on Karl Popper, who advocated such procedures but was not so naive as
to think they defined all science; see Hacking 1999; Laudan 1996).

It may be remembered that Francis Bacon, in his original definitions of the
scientific enterprise, was not only aware of all these problems but was more sensi-
tive to them than are some modern philosophers. He defined the
observation-experiment method and warned his readers of the “Four Idols”—the
biases we would now call “social construction” or “cultural baggage”—that can
blind the unwary and unaware (Bacon 1901, orig. 16" cent

As noted above, Hacking, in the most recent salvo in the long and confusing
“science wars,” concludes that science is somewhat socially constructed, some-
what factual (1999:99). This seems to me to be rather an evasion.

The present paper obviously cannot even begin to summarize the literature
on science wars.” It merely makes a single point: science is 100% socially con-
structed, but usually an accurate representation of the world in spite of that.
Evidence is supplied from a comparison of Maya and biological classifications of
birds. Maya ornithological taxonomy maps fairly well onto biological taxonomy,
but there are major differences. This disproves both simple realism (the Maya see
the natural distinctions just as the biologists do) and extreme social construction-
ism (the Maya system must be totally different from the biologists’, since the
societies are so different).

What has been missed, in the “science wars,” is the fact that society does not
necessarily get things wrong. To say something is socially constructed is not to say
it is inaccurate. After all, people have to learn their social constructions from each
other. If they can learn their culture through interaction, why can they (and, thus,
their culture) not learn about the environment from interaction, and then teach
each other in further interpersonal interactions?

Anthropologists have turned their ethnographic gazes onto the actual prac-
tice of science in dozens of societies. Beginning with traditional small-scale societies,
they have expanded their gaze to encompass modern laboratories. Particularly
noteworthy for its impact on the intellectual field is the work of Laura Nader and

132 ANDERSON Vol. 20, No. 2

her associates (Nader 1996). Nader has long encouraged research on the movers
and shapers of contemporary society, including scientists. Her group has thus stud-
ied modern laboratories and university halls, often comparing them with her
alternative study area, the highland Zapotec world, which has its own science
(Gonzalez 1998; Nader 1996).

Roberto Gonzalez, in particular, has provided some very thoughtful insights
into Zapotec traditional agricultural science. He sees it as definitely a science (tak-
ing “science” in sense #1, below). He analyzes it in terms of “assumptions” —folk
theories—that hold together a body of empirical, pragmatic knowledge. He shows
that these work like the theories of modern international science: they are basic,
largely counterintuitive ideas, extracted from experience, and used to generate
new practices and to explain and justify old ones. Some of them are highly ques-
tionable, but so are some assumptions of modern science.

Serious studies of nonwestern science go back to the dawn of anthropology.
One recalls Frank Cushing’s researches on Zuni agriculture and food, not pub-
lished in book form until 1920 (Cushing 1920) but carried out in the early 1870s.
Malinowski also produced classic studies in this area (Malinowski 1935), as did
his students such as Raymond Firth (1957) and Audrey Richards (1948). A self-
conscious movement to study “ethnoscience” arose in the late 1940s, largely among
students of George Murdock at Yale, working in Oceania (Goodenough 1953;
Conklin 1962; Frake 1980). Many of the earlier ethnoscience studies seem to the
contemporary anthropologist rather naively positivistic and formalistic, paying
rather little attention to such ideas as did not fit well into a “Western” scientific
framework. Ironically, this was not true of Cushing’s (or, to a somewhat lesser
extent, Malinowski’s) work, which should have served as examples.

Closer to the area of this paper, Scott Atran (1999) has analyzed Itz4 Maya
“folkbiology” from a similar point of view, analyzing knowledge and its linguistic
recognition.

In any case, this large body of research established “ethnoscience” or “folk
science” as something to study. Ethnographers came to see traditional knowledge
as worthy of serious, detailed attention. They were exhorted to understand it in its
own terms (“emically”), rather than merely comparing it (usually unfavorably)
with “Western” science. In spite of rearguard action by opposing scholars like
Marvin Harris (1968), studies of traditional knowledge grew and flourished apace.
It seems only natural—in fact, surprisingly long in coming—that ethnographers
should turn their attention on contemporary university laboratories.

SCIENCE

This, of course, brings us up against the question of defining “science.” Much
recent writing conflates several different things under that label. The following
seem to me to be quite separate phenomena:

1. Science as search for truth—for accurate data about the world, and for theo-

ries and hypotheses that model that world in ways that guide further searching
and understanding. Wolpert (1993) notes that these latter are critical, and that
science depends on fearlessly generating and testing even the most counter-

Winter 2000 JOURNAL OF ETHNOBIOLOGY 133

intuitive of proposals. It is understood that the search for truth often takes
wrong turns, as in the famous cases of “phlogiston” (see Kuhn 1962) and static
continents (Oreskes 1999), but that is in the nature of a search. The search is
seen as resulting in a body of facts, or at least pragmatically useful data, held
together by a framework of higher-level representations that can be called
“theories” and “axioms,” or, with Gonzalez (1998), “assumptions.” To extreme
social constructionists, this framework might be seen as merely “worldview”
or “cosmovision.

Science as a repository of True Facts or of Absolute Truth. Although still mili-
tantly upheld by some champions (e.g. Sokal 1997), this position is no longer
really tenable. Contemporary scientific practice can get us to the truth about
many things when ordinary observation will not, but that does not make Sci-
ence a God-given repository of infallible wisdom. In fact, philosophers of
science now hold that a genuine search for truth must take wrong turns. Oth-
erwise it is not a search—merely a repetition of the obvious. Humoral medicine,
alchemy, phlogiston, and many other theories were good ideas in their time,
probably the best that could have been done with the data at hand. The
disproofs of these theories signalled advances in the methods and techniques
of scientific practice, and, following those, the improvement of theory. Science
advances not by learning ex cathedra truth but by providing better and better
theories, as Kuhn (1962) pointed out and as most scientists now agree. Possi-
bly a subvariant of the old science-as-God’s-truth view is the popular
conception of “science”: Flashy technology. This is the concept we find in mass
media and Star Wars. (Wolpert [1993] overmakes the distinction between sci-
ence and technology, but has some valuable comments on the issue.)

Science as one specific form—the modern Western form—of the search for
accurate knowledge. There are two major contenders for the form. First, there
is the self-conscious “science” that began in perhaps the 14" century and was
formulated in the 16'" and 17" centuries in the writings and work of Francis
Bacon, Galileo Galilei, René Descartes, and, later, men like Robert Boyle and
Isaac Newton. This is a science defined (ultimately) as working from observa-
tion to inductive and then hypothetico-deductive theories, and testing these
(and the observations also) by experiment. It contrasts this search with re-
ceived wisdom, bias, and popular belief. This science did indeed break radically
with earlier ways of knowing. If it is not the only “science,” it at least deserves
some sort of terminological marking; Randall Collins’ “rapid discovery sci-
ence” (Collins 1998) is a good choice. The second is “science” as defined by
the logical positivists in the mid-20" century, with its formal operations, cov-
ering laws, emphasis on verification and/or falsification, and formally
(=mathematically) stated theoretical models. (This is so restrictive that it has
been abandoned by most current authorities.) This type of definition has the
nee of cutting off one specific type of truth- search, but it has the disad-

wag oh t 7
science aad other knowledge traditions.
Science as “what scientists do.” This allows us to include the faked data, char-
latanry, and vendettas that sometimes characterize scientific practice. It also

134 ANDERSON Vol. 20, No. 2

directs us to look at scientists as whole people—with their own eating habits,
daily lives, paranoias, and so forth.

5. Science as a social institution. This, of course, does not exist in most tradi-
tional societies—even China, with its unquesti y great scientific tradition,
did not have a concept of “science” or a “science establishment” (until mod-
ern times). Traditional societies usually have a term for “knowledge” but not
one for “science” as opposed to other types of knowledge. Chinese xue, Arabic
‘ilm, and, of course, Greek/Latin scientia included philosophy, literary stud-
ies, history, and other humanities. By contrast, in the United States and other
rich modern countries, “Big Science” now has a life of its own, institutional-
ized in such organizations as the National Science Foundation.

Evidently, the first three of these approaches characterize science as a special
kind of truth-seeking activity. The second pair treat science as a part of social ac-
tion in general—as a social construction. The third approach above is somewhat
intermediate, in that it regards science as a social construction—but a superior
one, one that inevitably leads to truer and better knowledge.

Champions of science see science as a truth-seeking activity, and see faked
data, vendettas and the like as “bad science”—alien contamination of the enter-
prise. The attackers and critics of science see it as a part of social action, and thus
see it in terms of 4 and 5 above. For many of them, the “bad science” is just as
scientific as the “good,” and phlogiston is just as real as any other scientific con-
clusion (presumably including well-demonstrated things like the laws of
thermodynamics).

There are thoughtful reasons to see science in all the above ways—so long as
they are kept analytically distinct.

If one looks at institutions, modern American “Big Science” is a totally differ-
ent thing from the tiny and scattered band of experimenters, often working in
terror of religious persecution, who created European “modern science” in the
1500s. They are linked by being related to the search for truth and understanding,
but they are not linked by similarities in institutional or political forms.

Conversely, if one is looking at the accumulation of accurate data, one can
reasonably look at Assyrian medicine, Chi gricultural i ts, and Maya
bird lore along with Nobel Prize experiments. One will not, however, be terribly
concerned about the personal lives of the Assyrian or Chinese scholars.

If one sees science in a broader and more sociological sense—science as the
activities of people who want to know something about the world, beyond what
intuition and received wisdom tell them—then the personal lives of the Assyrian
and Maya scholars become more interesting. This is the position of most histori-
ans of science. It has the major advantage of allowing all human societies into the
club, rather than defining “science” so as to restrict it to one cultural tradition.
Given the high prestige of the word “science” in today’s world, there are obvious
political ramifications to these alternate courses of action.

Certainly the most reasonable of the restrictive definitions would confine the
term “science” to post-1600 Baconian-Galilean science. This would reduce to a
historical footnote the Assyrians, Chinese, Mayas, and indeed well over 99% of
the human species’ long quest for understanding.! If “science” is limited to the

Winter 2000 JOURNAL OF ETHNOBIOLOGY 135

institutionalized rules of methodology that the positivists and scientific philoso-
phers of the early 17" century (let alone the 20" century) invoked, then by definition
there was no science before 1600 (or 1900 in the case of the positivists). Moreover,
much of modern science does not count. Astronomy, astrophysics, paleontology,
historical geology, most of ecology, and most of behavior biology are basically
observational sciences, rather thar being based on controlled experiments. Not
only Maya bird lore, but even 5 | research, rarely conforms
to the full Popperian or Hempelian canon (see Kitcher 1993 for the best discussion
of these issues). Above all, and most directly relevant to the present paper, tax-
onomy is not a hypothesis-and-lab-experiment science, though modern cladistics
is beginning to change that.

Much of the rhetoric in the “science wars” of the late 20" century has been
associated with a disregard for the above distinctions. Sometimes the disregard
appears to be willful, but often it is simply careless. In any case, what has often
happened has been an all-out attack on the entire search for knowledge and un-
derstanding, justified by the failings of some scientists (some do fake their data,
many are biased in one way or another). Conversely, some champions of science
have failed to make the necessary distinctions, and have talked as if an attack on
the current social i tion of science in the United States was an attack
on all attempts to know anything. Sokal (1997), in particular, seems to be peril-
ously close to taking such a view.

Of course, in the real world, it is impossible to have a search for truth that is
completely disinterested, wholly objective, and uninfluenced by social attitudes
and institutions. We have known this since at least the day of David Hume. Even
after half a century of critical theory, C. Wright Mills’ book The Sociological Imagi-
nation (1950) remains probably the best statement on the subject in social science.
The best the scientist can hope for is to understand biases, adjust them in a moral
direction, and compensate for them by seeking verification or disproof of findings
from other investigators from other schools or laboratories (Bacon 1901; Kitcher
1993).

At this point, it may be interesting to turn to a different tradition. If two ut-
terly different societies, with utterly different scientific traditions, come to similar
conclusions from similar data, perhaps there is objective truth lurking behind the
cultural screen. If and when two such societies differ totally in the way they con-
struct the world, then science may not exist at all, and the social construction of
knowledge may truly be said to be an arbitrary and solipsistic activity. To some
extent, the degree of “social construction” in science is an empirical question.

THE MAYA OF QUINTANA ROO

For the last ten years, in collaboration with Mexican (including Maya) and
United States colleagues, I have been carrying out on knowledge of plants
and animals among the Yucatec Maya of the “Maya Zone” of Quintana Roo. This,
the central part of the state, is the area that was never truly reconquered after the
Maya rebellion of 1846-48 known as the “Caste War” (Bricker 1981; Dumond 1998).
The Maya in the present Yucatan state were crushed in 1848, but in what is now

136 ANDERSON Vol. 20, No. 2

Quintana Roo they remained independent until 1901, and in the remote west-cen-
tral interior they were never really subdued. There was fighting as recently as
1934 in Dzula, the community next to my own base in Chunhuhub. Current in-
habitants of Dzula do not admit defeat. Alfonso Villa Rojas, ethnographizing the
area eastward in the mid-1930s, encountered much hostility and some personal
danger. The Maya, unsubdued, have continued to preserve a cultural tradition
that is at least five thousand years old in the area (Redfield & Villa Rojas 1934).

Like all cultures, Yucatec culture has profoundly changed over time, and is
rapidly changing today; but Yucatec biological knowledge retains much knowl-
edge of respectable antiquity. This is shown by archaeology, which discloses five
millennia of milpa fields and cropping patterns not dissimilar to many one sees
today (see Sharer 1994). It is also shown by colonial documents, which, from the
16 century onward (Landa 1937; Alvarez 1997; Arzépalo Marin 1987, 1996; analy-
ses in Anderson and Medina in prep.), record biological and medical lore close to
today’s. The Maya, as everyone knows, created one of the greatest, most brilliant,
most innovative, and most original civilizations the world has ever seen (Sharer
1994). The modern Yucatec of Quintana Roo are one of the several successor groups
of the Classic Maya. It is probably safe to assume that much of their biological
knowledge is derived from a Classic Maya base, given the consistency in usage
since the very earliest dictionaries (Alvarez 1997; Anderson and Medina in prep.).
This base has been greatly supplemented in more recent centuries by Spanish (in-
cluding Moorish) lore and international biological science. The modern Maya are
not some sort of living fossil, preserving for us the mysteries of the Classic Maya;
nor are they a tiny isolated group. They are bearers of the elaborate and expert
science of a long-lived, populous, brilliantly successful, constantly evolving civili-
zation.

Maya languages have a written tradition going back 1600 years, at first in
hieroglyphic and syllabi ipts, later in Spanish letters. Written transmission has
been a small but significant part of cultural transmission for a very long time. In
Yucatan, for instance, we have such examples as the Rituals of the Bacabs (orig. ca.
1600; see Roys 1965, Arzapalo Marin 1987), which records magical and medical
lore from the earliest part of the Colonial period.

Such a huge tradition is far from homogeneous or uniform (see e.g. Hervik
1999), and has its own self-reflexive turn (Sullivan 1989). This article focuses on
knowledge recorded in and around Chunhuhub, Quintana Roo.

Chunhuhub is a large farming town of some 5,000 people, occupying an ejido
(communal landholding) of 14,330 ha. All are Yucatec Maya except for a few ad-
ministrators and technicians, and a small number of in-migrants from central
Mexico. Al t is bi

ilinoial Most m3 sia 4517]

2x J raise maize, beans, squash,
chilies, and other crops by slash-and-burn cultivation of tracts ranging from 1 to 4
ha. Yields reach a ton per hectare or more. Every family has its dooryard garden;
many of these are large and contain up to 90 species of useful plants. Herbal medi-
cine is commonly practiced. Some g is still obtained, but hunting in recent
years has depleted game stocks (Anderson and Medina in prep.). Seventy percent
of the ejido is covered with forest, all of it in various stages of regrowth from past
cultivation. Some logging is carried out, but valuable woods were depleted in the

Winter 2000 JOURNAL OF ETHNOBIOLOGY 137

early 1990s. Stockrearing and beekeeping are important. The vast majority of the
population is highly knowledgeable about forests, fields, wild and tame animals,
medicinal herbs, insects, and indeed all aspects of the environment. Given the
solidly agricultural nature of the community, this knowledge is of a pragmatic,
experiential type, fitting well into the wider model of “ecology of practice” devel-
oped by Nyerges (1997).

Research in Chunhuhub lasted for six months in 1991 and six more in 1996,
with almost annual visits during intervening years. I was joined in the field by
Eugene Hunn during a month in 1991; he introduced me to Felix Medina Tzuc,
who became my collaborator and field assistant. Dr. Hunn al ded bird voices
in the field for Maya experts to identify (Hunn 1992) and worked with Don Felix
and myself on seeing how far Maya could go in identifying birds from pictures in
Peterson and Chalif’s guide to Mexican birds (Peterson and Chalif 1989). Other-
wise, research consisted primarily of walking through Chunhuhub and neighboring
ejido and ranch lands, observing birds in the field and obtaining Maya identifica-
tions. I also listened to a great deal of Maya conversation about birds and other
biota, including a great deal of discussion and argument over just what to call a
particular bird. Since I was studying “referential practice” (Hanks 1990), rather
than in the psychology of classification, I found it expedient to spend a great deal
of time in the field listening to actual practice, and made minimal use of formal
eliciting techniques beyond the frame interviewing described by Frake (1980). Thus,
the following data refer strictly to name usage in ordinary conversation. I did not
carry out experiments of the sort done by Atran (1999) and others, since I was
interested, at this stage of research, in different questions (see Hanks 1990 for dis-
cussion and justification of the referential-practice approach in studying Maya;
however, experimentation will be carried out in future research, opportunity per-

The Maya do noth concept of “science” in the modern international sense.
They do, however, have a reasonable equivalent. It is based on the core term k’aj
“to know.” Connected to this is the complex word ool, which means “heart,” and
by extension “knowledge, will, condition” (and sometimes also “lungs” and other
internal items near the heart). Uniting these, we get k’ajool, “to know something,
to recognize,” and thus the verbal noun k’ajoolal “knowledge.” This is as near as
we can get to “science.” It is not a far reach; k’ajoolal focally signifies practical
working knowledge.

This article focuses on classification and uses of birds, with some comparisons
to bird representation in contemporary international biological science.

MAYA BIRD CLASSIFICATION

Classification is often described as “carving nature at the joints.” This, of course,
assumes that nature has joints. How similar are Maya bird taxa to those of con-
temporary biological science?

In ethnoornithology, as in science wars, there is a range from social-construc-
tionist to realist positions. No one is as extreme as Feyerabend (1987), but Ellen
(1993), Forth (1996), and to a degree Bulmer (1967) stress social factors, and Ellen

138 ANDERSON Vol. 20, No. 2

has been sharply critical of narrowly realist models. Conversely, Boster (1987; Boster,
Berlin and O’Neill 1986; Boster and d’ Andrade 1988) and Hunn (1977) seem more
prone to assume people recognize categories that are real in the sense of evolu-
tionary biology. Atran (1990) and Berlin (1992) take a relatively strong position:
people are mentally programmed to recognize the multistranded similarities that
evolutionary relationships provide, and thus do carve nature at the joints. Atran’s
later position seems considerably more qualified and nuanced, due to his pro-
longed study (including use of psychological experiments, in collaboration with
psychologists) of Itzaj Maya classification (Atran 1999).

e Yucatec Maya data are consistent with the position that the Maya recog-
nize groupings that are natural in the sense of evolutionary biology. However, use
and other cultural and social factors enter into and shape the classification system.
The system can be understood only by taking both culture and nature into ac-
count.

Maya bird names are mostly at a level that Brent Berlin (1992) calls “folk ge-
nerics.”* These are usually one-word names. They contrast with each other; to
place a bird in one folk generic means it is not in any of the others. They are some-
times broken down into “folk specifics,” which are normally formed by adding an
adjective to the generic. Thus ch’om means “vulture”; chak pool ch’om, “red-headed
vulture,” is the Turkey Vulture (Cathartes aura). Maya, English, and Latin, like most
languages (Berlin 1992; his usage is followed here, rather than that of Atran 1999,
more for convenience than because of any deep theoretical reason), use the classic
pattern in which a folk-generic name is modified by an adjective to produce a
specific. (The Greco-Latin genus name Cathartes, roughly “one who cleans up,”
covers one or two other vulture species; aura comes from a Native American name
for this bird.) Latin terminology has many higher categories—the familiar phyla,
classes, orders, and families of Linnaean taxonomy. Maya terminology has only
one: the unique beginner ch’ich’ “bird.” Maya also has very few folk specifics.
Almost all classifying of animals and plants is done at the folk-generic level. (This
is true in most Native American systems.

Of the 89 named terminal taxa (folk genera not broken down, or folk species)
listed in the Appendix below, 63 have a one-to-one correspondence with the spe-
cies recognized by international ornithology. Ten are focus-and-extension names:
a focal species whose name is extended, more or less often, to other birds that are
seen as distinct but are not named. In 9 cases, a terminal taxon is a Linnaean genus
(4 cases), part of a genus (2 cases), or a group of closely related genera (5 cases). In
3 cases, a terminal taxon names a whole family, and in one case a name covers two
unrelated but very similar families (kusuun: swifts and swallows). One name—
only one—is a broad, vague category without Linnaean counterpart.

In two cases, a folk generic is broken down into folk specifics, all of which
have a one-to-one correspondence with the international ones. One of these folk
generics corresponds to a Linnaean family, one to a pair of closely related Lin-
naean genera.

In addition, I identify 13 groups, loosely named or named by extension of the
name of one of their species (see below). Of these 13 larger groups (“folk fami-
lies”), one corresponds to two (Linnaean) orders, two to an order, six to a family,
two to part of a family, and two to a genus.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 139

Many small birds are not idered important enough to have names of their
own. These are lumped into broad, vague categories that may or may not resemble
international scientific taxa.

These groupings are of two kinds. First, there are some genuine categories
that are well-bounded, well-recognized, and correspond loosely to international
taxonomic units. An example is provided by flycatchers of the Linnaean family
Tyrannidae. These are divided into three groups in Maya: takay “large yellow-
bellied flycatchers,” juiiro “large flycatchers that have a loud call that sounds like
juiiro,” and yaj “small flycatchers.” These groups are seen as related, as is proved
by the fact that juiiro (a rather exotic term) can be lumped with either takay or
yaj. It is explicitly recognized that these groups are diverse. Felix Medina Tzuc,
for instance, pointed out to me the only pair of Piratic Flycatchers that we saw in
our many months of co-work, and explained: “That takay is taking over the nest
of those orioles.” It was, indeed, doing that, but not so obviously that Don Felix
could observe it on the spot; he relied on his knowledge of the bird. The Piratic
Flycatcher (Legatus leucophaius) is a rare bird in Yucatan, and only a person with a
great deal of field knowledge would realize that it is a special sort of yellow-bel-
lied flycatcher that takes over the nests of other birds rather than building its own.
It does not have a special name in Yucatec Maya, but it is recognized nonetheless.

To some extent, there is a “focus and extension” semantics here. Takay most
commonly refers to the Couch’s Kingbird (Tyrannus couchii). Yaj has a definite
focus: the small Myiarchus flycatchers. These birds have a miserably mournful-
sounding call, like a child whimpering “yaj!” (“pain!” or “I hurt!”). No other small
flycatchers call like this, so the name qualifies as an extension. Another type of
broad category is much vaguer. “Little brown birds” are all vaguely lumped as
yankotij, a name which properly belongs to the Tropical House Wren. (This is
clear from its literal translation: “The one under the wall.” Only Tropical House
Wrens forage and nest in the stonewalls of the Maya house compounds.) “Little
yellow birds” are all vaguely lumped as chinchinbakal, a name that has no gener-
ally agreed focal referent. It covers goldfinches, warblers, small yellow-marked
tanagers, and much more.

Several other names can be extended ad hoc. The ones that can be extended
are known and constantly used to label some unknown bird. Other names are
never extended. Unknown medium-sized red birds, such as migrant red tanagers,
are lumped as chakts’its’i (“the red bird that says ts’its’i” )—a name that properly
belongs to the Northern Cardinal (Richmondena cardinalis). By contrast, sojlin “ant-
tanager” is not normally extended; if it is used for anything but an ant-tanager, the
extension is regarded as a mistake. Ts’‘apim “saltator” (Saltator spp.) is extended to
any medium-sized brownish bird of unknown identity. K’ok’ “Clay-colored Robin”
(Turdus grayi) is extended to cover any robin-like bird, such as wintering thrushes
from North America. Pich’ “Melodious Blackbird” (Dives dives) is the name used
for unknown birds that are smallish and black.

A very different type of extension is the use of one common name to cover a
natural group. In these cases, the name contrasts at two levels: (1) in its normal or
proper referential usage, it applies to one species; (2) in its extended usage, it ap-
plies to that species and the natural group it is in. Acommon case is t’uut, properly
the White-fronted Parrot (Amazona albifrons—by far the commonest parrot in the

140 ANDERSON Vol. 20, No. 2

area). This name is extended to cover all parrots (though not parakeets). In par-
ticular, the Yucatan Parrot (Amazona xantholora), ek’xikin “black ear” in Maya, looks
very much like the White-fronted and often travels with it; the two species are
collectively t’uut to everyone, unless and until the distinctive black earpatch can
be seen. Similarly, woodpeckers can be collectivized under the term k’olonte’ (or
sometimes che’hun); quail under bech’; hawks under ii’ or chuy; and a few others
as noted in the Appendix.

Hofling and Tesuctin (1997), in their dictionary of Itzaj Maya (which is very
close to Yucatec), treat these generalized terms as higher-level taxa that might be
called “folk families.” Thus, they treat ixt’ut (=t’uut) as a general term for parrots,
with the several folk generics (including t’uut in its more restricted sense) grouped
under it. Yucatec does exactly the same. The Itzaj use ixpaloomaj (the Spanish
word paloma, Mayanized) for pigeons and doves; Yucatec has a similar way of
labeling pigeons by extending the term ukum. Hofling and Tesuctin (1997) also
introduce a range of gender and environment categories that seem to cross-cut
rather than structure the Maya general purpose taxonomy. This is problematic for
the comparative nomenclaturist. In particular, their separation of tame and wild
birds under totally different headings is certainly not the Yucatec pattern. How-
ever, in general, Hofling and Tesuctin’s Itzaj classification is very close to Yucatec,
though their lumping of blackbirds and anis seems definitely not a Yucatec view,
and their lumping of quails and tinamous ina “covert category” of “ground birds”
(1997:76-77) seems rather ad hoc.

Atran’s excellent work on the Itzaj (1999) has gone into a different realm: cat-
egories psychologically real to his specific consultants, as shown by tests in the
field. These categories include “fish-eating water birds,...edible fruit-eating ground
birds,...edible fruit-eating tree birds,...inedible flesh-eating birds,...inedible fruit-
eating birds,” and “inedible blood-sucking birds [i.e., vampire bats]” (Atran
1999:172-174). None of these have emerged as categories from any work done by
me or others in Yucatec. It is notable that the category of “edible fruit-eating ground
birds” has a very different composition from Hofling and Tesuctin’s similar cat-
egory, though the same people were talking about the same general set of birds.
Similarly, “inedible flesh-eating birds” includes groups that Hofling and Tesuctin
and the Chunhuhub Yucatec both separate into a “hawk” group and an “owl”
group. The other assemblages found by Atran are even larger and less well de-
fined, and nothing like them emerges from Hofling and Tesuctin’s data or from
mine; they appear to be categories arrived at by testing for psychological similar-
ity, and are certainly not part of a linguistic taxonomy.

Hunn (1977) treats Tzeltal bird names similarly, recognizing “groups” that
are, de facto, folk families—natural groupings as recognized by the Tzeltal, but
not named as formally as the folk genera are. These, again, are similar to Yucatec
and to Itzaj (Hofling and Tesuctin 1997), but also include several other sets that he
calls “complexes. Most of these are the same, or much the same, as Yucatec (hawks,
vultures, doves...). Others include montane Chiapas species outside the knowl-
edge of Yucatec observers. However, some groupings psychologically real to the
Tzeltal would seem exceedingly far-fetched to the Yucatec, e.g. the link of squirrel
cuckoos with quail (Hunn 1977:153-5) or of trogons and motmots (Hunn 1977:169-
170). Hunn found the wide groups of waterbirds and black birds that Hofling and

Winter 2000 JOURNAL OF ETHNOBIOLOGY 141

Tesuctn found and that seem nonexistent for the Yucatec (except in so far as the
latter use the general descriptive term ch’ich’ ha’, “water bird”—without any im-
plication of real relationship).

Such groups blend into the “covert categories” of Berlin and his students. I am
very loath to invoke covert categories without proof that the people in question
really do think that a group is a real category. I think that wider-than-generic cat-
egories are clearly shown by extension of terms, if reliable and predictable—not
purely ad hoc like the extensions of ts’apim and chakts’its’i. But one must work
constantly in the field, with consultants, to make these distinctions, and even then
they could be challenged. I have done it in the appended table, but I have done it
with great care—only when a group is explicitly and reliably named by an ex-
tended term, and I have independent interview data suggesting that the group is
seen as a natural one. The extension of terms like t’uut and k’olonte’ does most
certainly show that the Maya recognize the parrots and the woodpeckers as natu-
ral categories. The extensions are thus of considerable interest.

All this reveals a pattern (the Yucatec one is very similar to the Itzaj one de-
scribed by Atran 1999). Big, obvious, or useful birds have their own names, which,
though “folk generics,” correspond with the species of Latin taxonomy. Small,
rare, or unobtrusive birds are referred to by names that are also “folk generics,”
but that do really correspond to genera or even families. Very small, insignificant
birds are simply lumped with the most convenient and well-known small bird of
the same color.

Consider the guild of woodpeckers and teank foragers: :

The area’s five common species f d t, obtrusive, noisy,
confiding, and impossible to miss. They are . parceled out under three names (two
almost identical species being lumped as kolonte’, and two as che’hun; either is
sometimes extended to cover woodpeckers in general).

Woodcreepers, though equally diverse in the area, are much less common,
less easy to observe, and dull in color. They have only one name, tatak’ che’, cor-
responding exactly with the Linnaean family Dendrocolaptidae.

Small trunk-foraging birds (such as the Plain Xenops, Xenops minutus) are rare
and obscure. They have no names at all, but, when noticed, are lumped under the
garbage-can category created by extension of yankotij.

Similarly, all game birds have their own names, but various non-eaten birds of
equal size and obviousness are lumped into broad categories. Hawks are lumped
into form-classes: each group with a distinctive flight profile, or appearance in
flight, has its own name. This causes some interesting debates about e.g. the posi-
tion of the White-tailed Kite (Elanus leucurus), which has pointed wings like a falcon
and thus could be a k’eenk’eenbak’, but is large and heavy-bodied and pale like an
ii’ (focally the Gray Hawk, Buteo nitidus) and thus could be in that category. Maya
discussions of such issues while away many a sleepy hour, and remind the visit-
ing ethnographer of debates among ornithological taxonomists.

One significant observation is that none of the wintering birds from North
America is named. Though Yucatan is vitally important as a major wintering ground
for many midcontinent species, with Chunhuhub alone playing host to thousands
of birds, not one has a Yucatec Maya name. (One, the Indigo Bunting Passerina
cyanea, has the Spanish name azulejo. In other areas of the Peninsula, migrant war-

142 ANDERSON Vol. 20, No. 2

blers are collectively referred to by the onomatopoeic word ts’ip, but I have not
heard this word used by Chunhuhub Maya.) Instead, the migrants are the major
beneficiaries of the loose extension of words like yankotij and chaktsi’tsi’.

In short, nature has joints, but society sometimes sees every reason not to rec-
ognize them. When birds are useful or too obvious to ignore, they get their own
names, which cover exactly the same space as a Linnaean species. To the degree
that birds are useless and otherwise nonsalient, they are lumped into progres-
sively wider and vaguer categories. Most of these categories correspond to the
larger Linnaean taxonomic units: genus, family. Then, as terms are extended out
to birds that are not only insignificant but do not even breed in Maayab (“Maya
land”), the terms cease to have any relation to Linnaean categories. Instead, they
lump birds roughly by size and color. (As a matter of fact, the same was true of
early European taxonomy, and Linnaeus himself did some broad lumping.) How-
ever, all of them have a focal exemplar that is a real, well-recognized Linnaean
species or tightly-knit group. The only exception is the catchall term chinchinbakal.

In other words, almost all Maya taxa, when not loosely extended, correspond
exactly with Linnaean taxa—at the species level, if the bird is salient; otherwise at
the genus level (but only if the genus is tightly knit, with all local species similar)
or at the family level. The less salient the birds, the more wide the Linnaean group
that equates to the labeled group in Yucatec. Some families (hawks, flycatchers)
are parceled out in ways not like those of international ornithology, but the parcel-
ing does make a great deal of sense in terms of the realities of Chunhuhub. They
accurately label natural-seeming groups, united by appearance and voice—even
when they cut across Linnaean taxa (as they sometimes do—but only in marginal
extensions of the terms).

One concludes that classification is a social construction, but one that must
take account of real natural differences if it is to be of any use at all (cf. Atran 1990;
Berlin 1992; and literature reviewed therein). Since the Maya and contemporary
international biologists are both trying to find useful labels that represent some
sort of external reality, there are many similarities in the two systems. Since the
uses in question are not the same, there are also differences—largely at the level of
“lumping.” The Maya lump species that are unimportant to them. The biologists
find all species equally important—at least in the Class Aves. However, biologists
too lump things they do not find salient. I am told by colleagues that the few
thousand recognized species of nematodes could probably be split into hundreds
of thousands (if not millions) of species, if nematode taxonomy were as developed
as avian taxonomy. Thus, one does not expect, and does not find, quite so good a
fit as one would expect from some of the work of Boster (1987; Boster, Berlin and
O'Neill 1986; Boster and d’Andrade 1989) or of Atran’s earlier theorizing. Maya
extension of terms fits well with Boster’s findings that broad visual similarities
serve as primary markers of relationship, and also with Boster’s observation that
Native American peoples are prone to name birds from their vocalizations. This
affects classification; flycatchers, for instance, are broken down as much by vocal-
ization as by appearance. The Maya also consider behavior and habitat in making

identifications and classifications. The term pujuy, for instance, is extended to birds
that act like the focal pujuy.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 143

MAYA BIRD USE

Knowledge of the uses of birds is straightforward, but not without interest in
the present connection. The most important use is as food. In addition to domestic
fowl (chickens, ducks and turkeys), several wild species are hunted—especially
quail, tinamous, chachalacas (bach, Ortalis vetula), and the very few larger game
birds still found in the area. Wild birds are also kept as pets, especially parrots,
parakeets, doves and pigeons, and-rarely now—large game birds.

Birds for food are usually shot with shotguns or rifles. (Maya hunters wingshoot
quail with ancient .22s, a feat that would awe any Anglo-American shooter.) How-
ever, small birds, and all birds wanted as pets, are caught with traps and snares.
Most common is a simple box trap, usually used by boys to get pets. Small birds
are baited in, and the boy pulls a string that removes a twig holding up a small
box. It falls over the birds. This is sometimes used more seriously, to get quail for
food. Nooses, snares, and sticky materials are occasionally used to catch small
birds. Sometimes a batea is staged: a hunt in which men form a long line and beat
the bush for game. Birds, however, are not successfully hunted this way, since
they fly off.

Birds are occasionally used as indicators of time or the like. For instance, the
Bright-rumped Attila (Attila spadiceus—one of the juiiro flycatchers) is sometimes
called the pak’sak’al, “plant-the-brushfield,” because it sings loudly at the time of
year when a farmer should be doing that. The noise of feeding birds can attract
one to wild fruit. Last of all, some birds, especially parrots, parakeets, and jays, are
often pests of the milpa fields. They must be controlled by scaring them away,
and—in desperate cases—by traps, slingshots, and guns. The Maya of Chunhuhub
love and cherish birds, and will not kill a pest bird unless its depredations become
devastating.

To this extent, knowledge is highly pragmatic. Social construction enters the
picture to the extent that only the larger and tastier birds are defined as edible; no
one would eat a hawk, toucan, or other large but non-choice species unless hun-
ger was serious.

However, a different kind of knowledge exists. Many birds are associated with
various sorts of dark powers. These fall into two categories: Ominous birds and
birds used in magic.

Ominous birds are the nocturnal species, considered unlucky through both
indigenous Mexico and traditional Spain. The Barn Owl (xooch’, Tyto alba) is par-
ticularly feared; its loud and hideous shriek presages death. Even the common
little pujuy (nightjar or pauraque, Nyctidromus albicollis) is worrisome. When it
calls and jumps up after insects, it presages death. Since hundreds of pujuy call
and jump all night, every night, in Chunhuhub, one would expect many deaths—
and, sure enough, every day, several people die in Mexico. Since any death,
anywhere, counts as a “hit,” the predictive value of the pujuy is confirmed. Some
Maya also believe the loud, wild call of the peppershrike (ch’uyin, Cyclarhis
Sujanensis) i is, ominous.

seer t y birds, but other areas of the Yucatan’s
Maya world have nee such animals. From Chan Kom, the most intensively
ethnographized community in the peninsula, we hear of the purple taankas par-

144 ANDERSON Vol. 20, No. 2

rot (Redfield and Villa Rojas 1934; taankas means any mental problem from numb-
ness to frenzy) and the eagle witch (way kot; Re Cruz 1996), apparently a witch
that can transform into an eagle. I have heard of a bird similar to the taankas
parrot in other nearby communities.

Other birds are useful in working magic. Magical practices are carefully and
sharply distinguished from standard medical and pragmatic knowledge, by being
labeled with the Spanish words magico or secretos. A specific magical practice is a
secreto (and it is said that “women know seven secretos to men’s four”). This word
is not used for practices, even esoteric ones that do not involve dealing with dark
or suspect supernatural powers.

Commonest of secretos is the use of hummingbirds, takay, and pujuy for love
magic, in that order of abundance. The bird is caught on a Tuesday or Friday—
these being the “bad days” of the Catholic church calendar—and dried. Dried
hummingbirds are carried in the pockets of young men for love magic (as is true
throughout Mexico). The powdered head of a dried hummingbird or takay, thrown
on a girl as she enters church, makes her fall madly in love with the thrower—at
least, if he knows and uses the right charms. A pujuy head is sometimes so used,
or the pujuy powder can be used in cursing. The t’unkiya (another nightjar,
Caprimulgus salvini) can also be used for this. Some magical medical charms in-
volve these birds.

Other witchcraft includes the magical introduction of live scorpions and the
like into people, but birds seem not often involved in these practices.

There were times when avian magical medicine was much more important.
The Rituals of the Bacabs, a collection of long ritual curing chants that seem to be
pre-Columbian, includes many birds. Jays, woodpeckers (kolonte’, che’hun), and
above all macaws (moo’, Ara spp., now extinct in the Yucatan) were particularly
important. They are invoked in many of the long chants. Often, they are associ-
ated with fire, and with the curing of insanity. The extremely arcane and difficult
Maya of the chants almost preclud i tudy at this time, but there is a major
future study of Bacab biology to be done.

Obviously, we have been moving farther and farther from anything that could
be verified by actual experience with nature. The edibility of birds is common
experience. Their value as pets is a matter of opinion, cultural as well as personal;
at the least, some of them can be tamed and make affectionate housemates. The
omen value of night birds is clearly nil to the outside scientific observer, but the
Maya can point to an almost perfect correlation between nocturnal calls and some-
one dying somewhere. This, then, is a judgment call in which “social construction”
really does allow many different results. Love magic is more difficult to defend,
since everyone admits that it rarely works.

Finally, we enter a realm in which no amount of personal experience can al-
low entry. It is quite safe to say that no one has really seen a purple taankas parrot
or the flaming jays of the Bacab songs. No one (to my knowledge) believes in them
in Chunhuhub; Chunhuhub is a particularly pragmatic, down-to-earth place. But
Chunhuhubians believe in other supernaturals that would seem to the outsider to
be just as difficult to observe, such as the aluxoob (tiny beings, usually thought to
be ancient Maya sculptures that can be animated by the right methods). Some

Winter 2000 JOURNAL OF ETHNOBIOLOGY 145

have seen the xtabay (the demon woman). However, it is admitted that one usu-
ally sees the xtabay only after consumption of a large amount: of alcohol. This is

traditionally thought to be because she finds drunkards vulnerable to
her evil charms, but skeptical Maya are quite aware of the obvious alternative
explanation.

In short, there is a realm, marked off in Maya thought, in which social con-
struction has really run far beyond any observable or verifiable reality. This is a
realm in which love, death, and fear are paramount. There are countless anthropo-
logical theories of magic, and it would be tedious and irrelevant to catalog them
here. Suffice it to say that almost all agree that, in these areas, human fears and
desires press irresistibly hard against the boundaries of observable reality. It is by
no means clear if any culture, including the culture of professional psychologists
and GOCtOrS; has ie solution to the problems of rains love and predicting death.
This t st st people from believing they can “have dominion over Judg-
ment Day” (as the ctaditionéd blues line has it), or at least over love. Exploring
these issues is outside the realm of this paper.

There is no explicit body of theory holding Maya bird knowledge together,
but one could, with Gonzalez (1998), formulate assumptions. First, it is assumed
that birds that look alike and sound alike are natural categories. If the birds are
essentially identical, they must be in one category, and if lumped they are lumped
with similar birds. No Maya, and probably no one on earth, would classify king-
birds, horned owls and cormorants in one group opposed to another group made
up of small flycatchers, barn owls and grebes. Social construction does not work
that way. Second, there is an assumption that all things are potentially useful for
filling material needs, and that all things large enough to be interesting should be
explored for their value in these areas. This assumption has led to the accumula-
tion of a great deal of lore about birds as food and as pets, and how to obtain them.
Third, there is an assumption that love, harm, and some kinds of fate can be con-
trolled by use of secretos, and that birds are useful in this enterprise. Certain birds
are earmarked for the tasks of magic.

DISCUSSION

Culturally standardized, traditional knowledge is, by definition, 100% socially
constructed. However, as Marx said of history: “Men make their own history, but
they do not make it just as they please” (Marx 1986:277). Observed external reality
provides constraints that cannot always be ignored. One cannot indefinitely be-
lieve in the safety of consuming deadly poisons, or walking off cliffs. Even if an
individual did so believe, a culture would not encode the belief. Experience to the
contrary would be too commonly observed.

The Yucatec Maya live as subsistence farmers in a harsh environment. They
survive only through having a literally encyclopedic knowledge of soil, water,
useful plants and animals, and useful farming techniques. Unlike academics at
prestigious universities, they do not have the luxury of believing anything they
wish or of dismissing the real world. Instead, they must constantly interact with
nonhuman reality. They walk a razor edge; the least mistake, the least failure to
invoke the correct strategy, can mean death.

146 ANDERSON Vol. 20, No. 2

Accurate knowledge does matter. Chunhuhub residents tell that some years
ago, two young Dutch hikers got lost in the woods near Chunhuhub. They died of
thirst in the waterless bush. The forest where they died was festooned with wild
grapevines (saya ak’, Vitis spp.). Every experienced Maya knows these grape-
vines, and knows that they store water, containing up to a cup or more of clear,
pure water per linear meter of vine. The reason why this knowledge is so wide-
spread is grimly obvious from the fate of the unknowing Dutch youths.

e more one knows about farming and about baalche’ (“things of the trees” —
wild animals), the better one lives. The forest provides, for those that truly know it
well, a good living, and even a few luxuries such as pet birds. Moreover, the Maya
yield to none in their enjoyment of the wild birds. They love the songs and color as
much as medieval European poets seem to have done. Enjoyment, too, leads to
knowledge and to its social construction. Many a Maya bird taxon appears to be
widely recognized simply because the birds in question are so amusing, or beauti-
ful, or delightful. This, too, is a use of nature, and a socially constructed one; but it
requires the existence of the birds, and the potential to enjoy them.

Interaction with nonhuman lives should not surprise those who believe in
“the social construction of reality” (Berger and Luckmann 1967). After all, social
construction can only arise from people interacting and discussing. It cannot exist
unless people actually do see and respond to an external reality—the reality of the
others they meet and the communication transactions they experience in dealing
with those others. If people are interacting with each other and learning from that,
it seems hard to deny that people interact with birds also, and learn something of
the avian world.

So ecological knowledge, like other knowledge, arises from practice (Nyerges
1997). It arises from interaction between people who are interacting with the non-
human world. It is phenomenological, but a phenomenology based on sensory
experience (Abram 1996).

en “nature fights back,” refusing to let people ignore it, society can con-
struct knowledge only within strict limits. If people want to use birds, the need for
an adequate classification system is strongly felt. This is a place where Nature
really has joints, if not always clear and obvious ones. People need to “carve Na-
ture at the joints” if they are to deal effectively with birds and communicate
effectively about them. As a result, classification systems from around the world
look somewhat alike. On the other hand, society and history play a role in deter-
mining which birds are used, which are held salient, which are ignored. Social
construction determines which are recognized as species, and which are lumped
into broad vague categories.

Berlin (1992) has demonstrated the similarity of classification systems around
the world, and the similarity of many systems to modern scientific taxonomy. This
he ascribes to a tendency of humans to perceive certain sorts of discontinuities
and continuities in nature. It is perhaps more accurate to say that people perceive
all sorts of things, but interact with humans and with other lives so much that
everyone, eventually, tends to realize that some differences matter and some do
not. The differences between different quail species are real, and matter to the
Maya. The differences between small flycatcher species are equally real to a biolo-

Winter 2000 JOURNAL OF ETHNOBIOLOGY 147

gist, but are of no special consequence to the Maya, who therefore ignore them (cf.
Boster and d’Andrade 1989). Overall, the Maya data fit much better with the find-
ings of Boster and his associates than with those of more social-constructionist
scholars. However, it is noteworthy that the latter (e.g. Bulmer 1967; Ellen 1993;
Forth 1996) have often been those who carried out research in east Indonesia or in
Papua-New Guinea, areas where systems may be genuinely very different from
both Maya and western models.

However, different Maya groups, and even different Maya consultants within
the same group, obviously classify birds in different ways. This is not so much a
matter of failing to perceive relationships as of devising classifications that fit one’s
own referential and ecological practice (Hanks 1990; Nyerges 1997). In particular,
birds are lumped ad hoc if there is no special better reason to lump them, or if
there is no pragmatic reason to see them as deeply and basically separate.

Words, after all, are to talk with, and there is no sense providing a verbal label
for something one does not talk about. Conversely, “utility” in the narrow sense
originally adduced by Hunn (1982) did not exhaust the reasons why people might
want to talk about something. They might want to talk about it only because it is
common and has a pretty song, and is thus hard to ignore if one loves birds as
much as the Maya do; thus there are not one but two names for the singularly
“useless”—but pretty and songful—Yellow-green Vireo (Vireo flavoviridis).

Even classification systems get confused with power relations, as Foucault
(1971) showed for the Linnaean system; one need only look at its hierarchy, with
“Kingdoms,” “Orders,” and “Families” duly arranged by relations of inclusion. I
find no evidence that the Maya system was concocted with one eye to the State,
even though the ancient Maya did have states. But one cannot be sure. If relations
with the natural world and with fellow farmers are clearly reflected in the system,
relations with the hierarchy may also be. The weird birdlore reflected in the Ritu-
als of the Bacabs may well have a great deal to do with politics. We do not know.

Moreover, as belief gets uncoupled from immediate observation, society can
construct with a much freer hand. In international biological science—and, even
more, in high-energy physics and in astrophysics—much high theory is purely
speculative. Theorizing runs far ahead of observation. Conversely, sometimes a
new theory is irrationally rejected for decades, until the buildup of supporting
facts is so overwhelming that no one can deny it any more (see Oreskes 1999).

Of course, the ideal of testing it is there; but by the time a theory is adequately
tested, theorists have already gone on to even wilder flights of imagination. It
should, then, surprise us not at all that the hardheaded and pragmatic Maya farm-
ers believe some very improbable things about birds.

To the dispassionate anthropologist, the mistakes people make seem remark-
ably similar. As our felt needs for knowledge outrun our possibility of checking,
we come to believe some very improbable things. In so far as a whole society is
made up of people with such needs and such biases, a whole society can construct
a whole system of knowledge that is far from observed reality. This is as true of
20 century scientists as of Maya farmers. Against the Maya use of birds in love
magic, we can set the enormous amount of speculation on love that fills rack after
rack in any bookstore. Much of this material seems to the uninitiated to be as far

148 ANDERSON Vol. 20, No. 2

from any observed reality as are the beliefs about the use of powered takay heads.
It is not only the Maya whose need to deal with love runs far beyond their ability
to understand it.

In Mayaland and in the modern laboratory, observations are usually good
and accurate, unless driven by powerful antecedent beliefs. This is because people
can check their observations against reality, on frequent occasions, and thus are
disabused of the minor errors that derive from unquestioned assumptions, sheer
ignorance, and mistake. Interpretations and explanations, in so far as they are
decoupled from direct observation, are increasingly tentative. Accordingly, they
must be more and more self-consciously tested against reality. At no stage is the
process free of bias and social construction, but at no stage is the process so re-
moved from reality-testing that it is pure construction in a vacuum.

CONCLUSION

There is, then, a universal search for truth. We can use the term “science” for
this worldwide search for more and more accurate data and understanding.

However, every culture, every society, has its own unique form of “science,”
and systems of knowledge are indeed socially constructed, in a very literal sense.
It would thus be possible to limit the term “science” to the activity defined by
Bacon, Galileo, Boyle, et al; however, the restriction of the term to contemporary

ig Science is absurd, and the restriction to formal, positivist work
(a restriction still made by e.g. Cronk 1999) is not only absurd but eae vio-
lated by almost all working scientists (Hacking 1999; Kitcher 1993; Kuhn 1962).

Even the limitation to post-Baconian experimental practice may be seen as
arbitrary and Eurocentric. “Science” is a highly prestigious label in modern soci-
ety. Refusing to use Egyptian, Greek, Chinese, Near Eastern and Maya traditional
knowledge systems seems undesirable, not only because it would add to the al-
ready great amount of bias in the world, but also because it might lead
contemporary scientists to slight traditional knowledge.

owledge is socially constructed, but it is through the very process of social

construction—inevitably involving interaction, checking, and feedback—that accu-

rate, empirically useful knowledge can be increased, refined, corrected, and made more

aluable. Mistake-making is an inevitable cost of this system. Science flourishes in

so ie as people keep interacting with the world, to verify or disprove the specula-
tions they have entertained and the conclusions they have reached.

Because of differences in this and in entire social contexts, knowledge systems
in different cultures can look very different. They can also look similar, especially
when they are under constant tight control _ feedback from the actual “world
out there.” The degree ystems, and the degree of arbitrari-
ness that enters into socially constructed knowledge systems, are matters for
empirical investigation.

NOTES

‘ Lavoid the term “western science,” because contemporary biology i is an piniernational, not
a western, project; Chinese, African, Indian and oth

tions to it. “Western” science, conversely, a includes a great deal of lore aig as wn

Winter 2000 JOURNAL OF ETHNOBIOLOGY 149

humoral medical theory, still common in western folk societies) that is no longer part of
formal oo science. And, anyway, the Maya live west of Europe. “Western” rr
not to say prejudicial, term y's

J
global society. Writers such a as Wolpert (1993) restrict the term “science” to the west, usu-
ally through ignorance of what other cultures are doing. Wolpert, for instance, states that
“the Chinese, often his ay of as scientists, were expert eis but made negligible
contributions to science. .. (Wolpert 1993:xii).”
Even given Wolpert’s seclrintive definition of “science” Geaicaiy; post-1600 western ex-
perimental science, but extended to include ancient Greek speculation and modern non-
laboratory sciences), this statement is absolutely wrong, and a ein complete igno-

rance of Chinese science and philosophy—ar y Wolpert’s
wildly inaccurate discussion of nine (1993: 46-47). His opinion of all ’ ‘primitive’ and
nonwestern traditions is summed u ..for thousands of years the mythology and cos-

mology of almost all cultures entertained neither a critical tradition nor curiosity about
nature (Wolpert 1993:54).” He equates nonwestern know lodge -ereking: inchiding Chinese
and Islamic science, witha (1993:26).
Yet—as an educated Englishman—he adulates the ancient Greeks, crediting them with the
full Baconian-Galilean approach; this is, again, not accurate. It is surprising and depress-
ing to find that claims of this sort can still be published in an academic work. Wolpert’s
work is also confused and inconsistent. He defines science in various ways, loosely classi-
fiable into a broader definition and a narrower one. By Wolpert s broader definition
(thoughtful observation leadin intuitive” meaning,
loosely, “consistent with everyday air IRE Ad and all”), all societies have science.
(This is not helped by Wolpert’s lack of clarity about just what is counterintuitive.) By his
narrower one, only certain post-1600 sciences count. The latter definition would rule out
taxonomy—contemporary biological as well as Mayan.

2 There is no previous systematic account of Yucatec Maya bird names. Existing accounts
such as those of Pacheco Cruz (1958) and Hartig (1979) are incomplete, out of date, and
seriously compromised by major errors. (Pacheco Cruz does include a great deal of cul-
tural material that is of great value—including a very large amount of magic and folklore,
— beyond any eng I encountered, ) pAtzaj | Maya, which i is virtually a dialect of Yucatec,
Atran (1993,
1999) has provided lists of terms, and Charles Hofling, with F. F. Tesuctin (1997), have
provided an entire dictionary. This dictionary gives a list of bird names (pp. 72-77). This
list breaks up the bird names into various categories, including use-categories, and pro-
vides a number of different sorts of higher-level taxa that might be sone “folk families”
(see above). One or two of these groupings seem highly idi
psychologically there for the Yucatec. For instance, the Yucatec saeeyy not group anis with
blackbirds. However, most of them are the same as the Yucatec groups. I have been more
cautious in listing groups. For example, their category of ground game birds—named in
Spanish but not in Itzaj—is probably real, in some sense, to the Yucatec too, but I have not
listed it because it is not a Yucatec-named group. Presumably all of the groups listed in the

nd are certainly not

ent from anything familiar in Yucatec. Many names, too, have quite different usages from
those common in Yucatec; for instance, ts’apim refers to orioles instead of saltators. They
also use the diminutive ix- (equivalent to modern Yucatec x-) wherever it is commonly
used in speech; but the diminutive is actually an optional addition to the name, so I have
not indicated it.

150 ANDERSON

Vol. 20, No. 2

ACKNOWLEDGEMENTS

I gratefully acknowledge, above all, the bisik of Eugene Hunn, Felix Medina Tauc, and

the people of Chunhuhub. William Balée and ti

editorial input.

dv aluable
i ©

RATURE CITED

LITE
ABRAM, DAVID. 1996. The Spell of the
Sensuous. Pantheon, New York.
ALVAREZ, CRISTINA. bes een
a Maya

L Me ico.
ANDERSON, E. N., and FELIX MEDINA
TZUC. In prep. Animals and the Maya
in Central Quintana Roo. Ms.
pec MARIN, RAMON (editor).
6. Calepino de Motul. UNAM,
he co.

~ 1987. a Ritual de los Bacabes.

UNAM, Mex
ATRAN, SCOTT. 1993. Itza Maya
agroforestry. Current Anthropology

. 1990. Cognitive Foundations of
Natural History. Cambridge University
Press, a ridge.

Sessile 9. Itzaj Maya folkbiological
Hasecies Co ognitive universals and
cultural particulars. Pp. 119-204 in
Folkbiology, Douglas Medin and Scott
Atran (editors). MIT Press, Cambridge.

BACON, FRANCIS. 1901 (orig. 16" cent.).
Novum Organum. P. F. Collier and Son,

York.

TE and THOMAS
KMANN. 1967. The Social
Construction of Reality. Anchor Books,

New York.

BERLIN, BRENT. 1992. On Ethnobiological
Classification. Stanford University
Press, Stanford.

BRICKER, VICTORIA. 1981. The Indian
Christ, the Indian King. University of
Texas Press, Austin.

BOSTER, JAMES. 1987. Agreement between
biological classification systems is not

O’NEILL. 1986. The correspondence of
Jivaroan to scientific ornithology.
American Anthropologist 88:569-583.

, and ROY D’ANDRADE. 1989.
i and human sources of cross-
cultural agreement in ornithological
lassificati American Ant! logist

91:132-142.

BULMER, RALPH. 1967. Why is the
cassowary not a bird? A problem of
zoological taxonomy among the Karam
of the New Guinea Highlands. Man 2:5-

25.

COLLINS, RANDALL. 1998. The Sociology
of Philosophies. Harvard University
Press, Cambridge.

CONKLIN, HAROLD. 1962. Hanunoo

CRONK, LEE. 1999. That Complex Whole.
Westview Press, Boulder.

CUSHING, FRANK. 1920. Zuni Breadstuff.
Museum of the American Indian, Heye
Foundation, New York.

DUMOND, DAN. 1998. The Machete an
the Cross. University of Nebraska Press,

Lincoln.

ELLEN, ROY. 1993. The Cultural Relations
of sgpreawey Cambridge University
Press, Cambridge.

FEYERABEND, PAUL. 1987. Farewell to
Reason. Verso, London

FIRTH, RAYMOND. 1957. We the Tikopia.
(second edition). George Allen and

Unwin, London.

FORTH, GREGORY. 1996. Nage birds:
Issues in... the... analysis of
pie eneeg Er classification.
Anthropos 91:89-109.

FOUCAULT, cL ay tke Order of

Things. Pantheon, New
FRAKE, CHARLES. 1980. es and
Cultural Description. Stanford

University Press, Stanford.
GONZALEZ, ROBERTO. 1998. are
science: Farming and food in

Northern Sierra of Oaxaca, Mex
Ph.D. dissertation LORIE AS
University of California, Berkeley.

Winter 2000

GOODENOUGH, WARD. 1953. Native
Astronomy in the Central Carolines.
University of Pennsylvania, Museum o
Anthropology, Philadelphia.

HACKING, IAN. 1999. The Social
Construction of What? Harvard
University Press, Cambridge.

HANKS, WILLIAM. 1990. Referential
Practice. University of Chicago Press,

hic

HARRIS’ MARVIN. 1968. The Rise of
pases Dees eo Science. Thomas Y.
Crowell, New York.

HARTIG, HELGA-MARTA. 1979. oe —
de Yucatan: Nomenclatura e
Espaniol- Ae rdienos Fondo Satria
de Yucatan, Mexi

HERVIK, PETER. 1999. Maya People within
and beyond Boundaries. Harwood
Academic Publishers, Amsterdam.

HOFLING, CHARLES ANDREW and
FELIX FERNANDO TESECUN. 1997.
Itzaj Maya-Spanish-English Dictionary.
oe 8 aa Salt Lake City.

HUNN, E 1977. Tzeltal Folk
Zoology. a Press, New Yor

_—___—— . 1982. The utilitarian factor in folk
She cl classification. American
Anthropologist 84: 7.

. The use of scare recordings
as voucher eli mare and stimulus
materials in ethnozoological research.
Journal of Ethnobiology 12:187-

KITCHER, PHILIP. 1985. Vaulting
Ambition. MIT Press, Cambridge.

—_—___—. 1993. The Advancement of
Science. Oxford University Press, New

ork.
KUHN, THOMAS. 1962. The Structure of
Scientific Revolutions. University of
Chicago Press, Chicago.
LAKATOS, IMRE. 1976. Proofs and
Refutations: The Logic of Mathematical
Discovery. Cambridge University Press,

Cambridge.

LANDA, FRIAR DIEGO DE. 1937 (Spanish
orig. 16 century). Yucatan Before and
After the Spanish Conquest. (translated
2s Scat Gates). The Maya Society,

EAUDARN, “LARRY. 1996. Beyond
wheasteoaae and Relativism. Westview
Press, Boulder

MALINOWSKI, BRONISLAW. 1935. Coral
Gardens and Their Magic. American
Book Co., New York.

JOURNAL OF ETHNOBIOLOGY 151

MARX, KARL. 1986. Karl Marx: A Reader.
(edited by Jon. Elster). Cambridge
University Press, Cambridge.

MILLS, C. WRIGHT. 1950. The Sociological
Imagination. Oxford University Press,

ew York.

NADER, LAURA (editor). 1996. Naked

Boundaries, Power and Knowledge.

u York.

NYERGES, ENDRE (editor). 1997. The
Anthropology of Practice. Gordon and
Breach, Amsterdam

ORESKES, NAOMI. 1999, The Rejection of
Continental Drift. Oxford University
Press, New York.

ORWELL, GEORGE. agi ace iv
Nineteen Eighty-F
University Press, New Yo

PACHECO CRUZ, SALVADOR. 1958.
Diccionario de la Fauna Yucateca.

a, Mérida.

PETERSON, ROGER TORY and EDWARD
CHALIF. 1989. Aves oe México: Guia de
Campo. Diana, Mexi

RE CRUZ, ALICIA. 1996. "The Two Milpas

re)

REDFIELD, RC
VILLA ROJAS. 1934. Chan Kom, A
Maya Village. Carnegie aoe of
Washington, Washington,

RICHARDS, AUDREY. 1948. pias and
Work in a Savage Tribe. Free Press,

lencoe.

ROSENBERG, ALEXANDER. 1992.
Science of Diminishing Returns?
Uinivebaity of Chicago Press, Chicago.

ROYS, RALPH. 1965. Ritual of the Bacabs.
University of Oklahoma Press, Norman.

SHARER, ROBERT. 1994. The Ancient
Maya. Stanford University Press,

Stanford.

SOKAL, ALAN. 1997. Impostumes
Intellectuelles. O. Jacob, Paris

SULLIVAN, PAUL. ntiisieed
Conversations. University of California
Press, Berkeley.

WOLPERT, LEWIS. 1993. The Unnatural

tu Science. Harvard University

Press, Cambridge.

152 ANDERSON Vol. 20, No. 2

APPENDIX.—Chunhuhub Yucatec Maya Bird Names and Their Correspon-
dence with Linnaean Nomenclature

Maya is transcribed according to the system recently standardized and accepted
for Maya languages. This system is still unfamiliar in Yucatan, but is winning rapid
acceptance and is used in the newer literature. Only common, well-identified names
are given. I have recorded several others that are either vague or need more re-
search. Only the commonest Spanish names are provided. Unlike Hofling and
Tesuctin, I have not bothered to respell Spanish names in Maya transcription (see
e.g. “ixpaloomaj” above). The people of Chunhuhub are bilingual, and usually
pronounce the Spanish without any Maya accent. To respell Spanish names seems
pedantic. Like other tropical American peoples, the Maya are fond of naming birds
from their call. In the forest, birds are far more often heard than seen. Often, indi-
viduals do not even know the appearance of a bird well known by voice.

Class Aves: ch’ich’ “bird”
Arbitrarily arranged in Linnaean order; no obvious order or high-level groupings

arise from the data. Vague and tentative “covert” or ad hoc categories are often
proposed, but I prefer to be conservative, staying with unquestionable data.

Tinamou cluster: Non

Mankolom. Great Tinamou, Tinamus major. Does not occur locally, but known to
locals who who have been farther south.

Non (nom). Rufescent tinamou, Crypturellus cinnamomeus. Common; a game bird,
but not often obtained because of its extreme wiliness.

Ke’el non. Little tinamou, Crypturellus soui. In spite of a name that makes it sound
like a subcategory of the foregoing, this is recognized as a different bird.

Kamacho, Olivaceous Cormorant, Phalacrocorax olivaceous. Spanish-sounding ex-
tension of mach, the more general Yucatec name. Extended to the Anhinga, Anhinga
anhinga. I have heard the cormorant called jichkal but this seems nonstandard.

Kuts ha’ “water turkey.” Muscovy Duck, Cairina moschata. (More commonly just
called pato, the Spanish for “duck.” To distinguish it from the rarely found domes-
tic mallard, it is called pato criollo “native duck.”) Common domestic and rare wild
bird. Used for food and as a pet.

Pijije. Black-bellied Whistling-duck, Dendrocygna autumnalis. Echoic. This is the
Spanish name too, but it was probably borrowed from a Maya language or from
Nahuatl.

Ch’om “vulture.” One of the few named groups in which a true folk generic is
broken down into folk specifics:

Batab ch’om “chief vulture.” King Vulture, Sarcorhamphus papa.
Box pool ch’om “black-headed vulture.” Black Vulture, Coragyps atratus.
Chak pool ch’om “red-headed vulture.” Turkey Vulture, Cathartes aura.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 153

Hawk group: All hawks are recognized as related, as is shown by the fact that they
can all be covered by widely extending the terms ch’uy and ii’ and by the Spanish
aguililla or gavilan. Careful speakers, however, break out several other categories.

Ch’ uy “large hawk.” This term can cover any large hawk.
Ek’pip. Black alate Spizaetus tyrannus. Extended to other hawk-eagles and
large impressive hawks

Ti’. Focally the Gray Hawk, Buteo nitidus, but used for any medium-sized hawk,
especially pale-colored ones.

Sak ii’ “white hawk.” The Gray Hawk in particular—or, sometimes, any light-
colored hawk—as opposed to other ii’.

Jonkuuk. Harpy Eagle, Harpia harpyja. Now extinct in the area, but the name is
well known to local residents, who remember the bird and describe it accurately.
It also appears in the Colonial dictionaries, with unmistakable descriptions.

Koos. Laughing falcon, Herpetotheres cachinnans. Unlike the other hawk names,
this one is not often extended.

K’eenk’eenbak’. Small falcons. Apparently the most typical, or perhaps even focal,
one is the Bat Falcon, Falco rufigularis. However, the term is extended to any small-
ish, pointed-winged bird of prey, up to and including the White-tailed Kite, Elanus
leucurus, which is also called ii’ and sak ii’. The male bat falcon is called kiris or
kiklis.

Bach. Chachalaca, Ortalis vetula. There is some possibility that the chachalaca is
seen to be related to the following four, but I have no evidence of it. Common;
food item but usually too wily to kill. Also called kobi or koba.

Kox. Crested Guan, Penelope purpurascens. Food. Now very rare.
K’ambul. Curassow, Crax rubra. Faisan in local Spanish. Food and pet. Rare.

Kaax. Domestic fowl, Gallus domesticus. Name derived from Kastelan “Castilian,”
a recognition of the introduction of the bird by the Spanish. A rooster is t’eel, which
must once have meant a male bird or male game bird in general.

Bech’ group:

Bech’. Yucatan Bobwhite Quail, Colinus nigrogularis. Common. Potentially a food,
but in practice too small and wary to be worth the trouble of hunting it.

Chibilub. Singing Quail, eee thoracicus. Rare; potential food, actually too
rare and well-hidden to hunt

Turkey group: unlabeled but eee recognized, and terminologically united by
sharing special terms for tom and hen.

Uulum. Domestic Turkey, Meleagris gallopavo. Echoic name. Common; important
food resource. Tom is tso’, hen is tuux

154 ANDERSON Vol. 20, No. 2

Kuts. Ocellated Turkey, Agriocharis ocellata. Formerly important game bird, now
almost exterminated by overhunting. I believe the tom and hen are labeled as in
the preceding.

Gallinola. Gray-necked Wood-rail, Aramides cajanea. Also Northern Jacana, Jacana
spinosa (when it is not given its proper Maya label). Extended to any other rails
present (the only common one is the Sora Porzana carolina, a winter visitor). Water
birds are so rare in interior Quintana Roo that Maya names have usually been
replaced by Spanish ones—as in this case.

Correa. Limpkin, Aramus guarauna. Spanish name; probably a variant of the com-
moner Spanish carao.

T’eel ha’ (“watercock”). Jacana, Jacana spinosa. Probably extends to similar birds.
Pigeon cluster; all lumped as ukum or under the Spanish term paloma.

Paloma. Rock Dove, Columba livia. Common tame bird. Since it is a Spanish intro-
duction of no great age in the area, it has no Yucatec name.

Chuukij. Scaled pigeon, Columba speciosa.

Ukum (ukuch). Red-billed pigeon, Columba flavirostris. Echoic. This is the common
pigeon of the area, and its name is routinely extended to mean “large pigeon in
general,” i.e. to cover the preceding species. Also called kukut’kib, which name is
also extended to the foregoing. Used for food, but rarely taken.

Sakpakal. White-winged Dove, Zenaida asiatica.

Tsutsuy. Leptotila doves and similar doves. Common is the White-tipped Dove,
Leptotila verreauxi. Other species occur and are not distinguished terminologically,
except for the Ruddy Quail-dove, Geotrygon montana, which is chak (red) tsutsuy
or k’aankab (red-dirt) tsutsuy.

Mukuy. Ground doves. Probably echoic. Three species:
Chak mukuy “red ground-dove.” Ruddy Ground-dove, Columbina talpacoti.
Abundant; occasional pet.
Sojol mukuy “leaf-litter ground-dove.” Common Ground-dove, C. passerina.
Rare.

Tuch mukuy “ground-dove that calls tuch,” thus part-echoic. Blue Ground-
dove, Claravis pretiosa. Common but shy and seldom seen.

Parrot group: recognizable by being lumped collectively as t’uut.

T’uut. White-fronted Parrot, Amazona albifrons. Common; frequent pet. Also a fre-
quent pest of milpas, eating maize, fruit, and almost anything else well above
ground level.

Ek’xikin “black ear.” Yucatan Parrot, Amazona xantholora. The Maya name hits
home—it points to the one field mark reliably distinguishing this uncommon bird
from the preceding.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 155

Kocha’. Red-lored Parrot, Amazona autumnalis. Pet, but rare in the area. Also called
kulix.

Taadi’. White-crowned Parrot, Pionus senilis.

K’ili’. Aztec Parakeet, Aratinga nana. Abundant, and a very serious pest, descend-
ing in flocks on maize and fruit. Sometimes shot with slingshots when caught in
the act. (Chunhuhub Maya do not usually, otherwise, kill even the worst pests.)
Kili’ are never called t’uut.

Baakenchulul. Pheasant cuckoo, Dromococcyx phasianellus. Extended to cover the
Lesser Roadrunner Geococcyx velox, rare and probably a recent arrival in the area
(coming with large-scale clearing of forest).

Kipchoo’. Squirrel cuckoo, Piaya cayana. Echoic.

Chikbu’ul. Groove-billed Ani, Crotophaga sulcirostris. Echoic, but folk-etymologized
in that bu’ul means “beans,” and anis often hide in bean vines.

Owls would seem a natural cluster, and this may be shown by their uniformly
ominous significance, but they are always kept terminologically distinct, so far as
I have heard.

Xooch’ (or xiich’). Barn owl, Tyto alba. Echoic. A bird of very bad omen; if it shrieks
over a house, an inhabitant or relative will die. This common European belief may
have been introduced by the Spanish.

Tunkuruchu’. Great Horned Owl, Bubo virginianus. Echoic. Also called bujk’aanij
and xo’chikin. A bad omen.
Kulte’. Mottled Wood-owl, Ciccaba virgata.

Chaxnuk (from chak xnuk, “little red old man”). Ferruginous Pygmy Owl,
Glaucidium brasilianum. Also a bad omen, but so common and tame that no one
takes it very seriously. Name extended to other small owls. Also called koak’ab,
“the one who goes ko at night,” which is, obviously, a part-echoic name.

Nightjar cluster: Identifiably a cluster because they are covered by the well-known,
widely used Spanish term tapacamino.

Pujuy. Paraque, Nyctidromus albicollis. Probably echoic.

T’unkiya. Salvin’s Nightjar, Caprimulgus salvini. (Probably also covers the rare
Yucatan Will, Nyctiphrynus yucatanicus.) Echoic.

Jaap. Common Potoo, Nyctibius griseus. Echoic.

Ts’unuun. Hummingbirds in general. The many species found in Chunhuhub are
not terminologically distinguished. Apparently echoic of flight sound.

Uulum k’aax “forest turkey.” Trogons, Trogon spp. Echoic; name from similarity of
call to turkey’s common note. Several species occur and are seen as different, but
they are not terminologically recognized. People in other areas say that kux is the
correct name for the trogon.

156 ANDERSON Vol. 20, No. 2

Juj. Blue-crowned Motmot, Momotus momota. Echoic.

Toj. Turquoise-crowed Motmot, Eumomota superciliosa. Echoic. The Spanish name,
often used, is pajaro reloj—“ clock bird”—because this motmot regularly swings its
long, pendulum-like tail from side to side.

Toucan cluster; recognized because the name panch’el is used for both species.
Panch’el. Collared Aracari, Pteroglossus torquatus.

Pitoreal or tucan. Keel-billed Toucan, Ramphastos sulfuratus.

Woodpecker cluster; collectively called either che’hum or kolonte’.

Che’hun. Golden-fronted Woodpecker, Melanerpes aurifrons. Often extended to the
Yucatan Woodpecker Melanerpes pygmaeus and sometimes to other species.

Chi’pirix. Ladder-backed Woodpecker, Picoides scalaris. Name—or, usually, just the
pirix—sometimes extended to the Yucatan Woodpecker (which looks like a Golden-
fronted but is smaller, about the same size as the Ladder-back). Name also extended
to the male genitalia, as is the Spanish picocarpintero (“woodpecker”) in Mexican
folk speech.

Kolonte’. Lineated Woodpecker, Dryocopus lineatus, and Guatemalan Ivorybill,
Campephilus guatemalensis. These two woodpeckers are very similar and tend to
occur together. Even those who see that they are separate species tell me that the
birds are too similar to be worth distinguishing! Probably an echoic name.

Tatak’che’ (tak’ak’che’). Woodcreepers, family Dendrocolaptidae. A collective term.
It is extended to cover the Smoky-brown Woodpecker, Veniliornis fumigatus, which
looks and acts more like a woodcreeper than a woodpecker—though it is some-
times called che’hun, too. The several species of woodcreepers are uncommon and
hard to spot, and—again—even those who see they are different see no reason to
recognize that fact terminologically. Echoic, but of the birds’ pecking, not of their
calls.

Sob (or, more rarely, pu’). Barred Antshrike, Thamnophilus doliatus.

Flycatcher group: united by loose and sloppy use of the following three names—
especially the first and last—to cover the whole group.

Bech’ lu’um. Black-faced Antthrush, Formicarius analis. Also called tsiimink’aax
(“forest horse”)—a name also used, formerly at least, for the tapir. The reason for
this naming is hard to imagine.

Takay. (This name is always spoken with the diminutive suffix x: Xtakay. It is given
that way in other works.) Large yellow-bellied flycatchers, focally the Couch’s
Kingbird, Tyrannus couchi, but including many species. Echoic.

Juiiro. Medium-sized brown forest flycatchers. Echoic; these all have calls that
sound like juiiro.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 157

Yaj. Small flycatchers. Echoic. Focal is the Olivaceous Flycatcher, Myiarchus
tuberculifer, whose mournful whistle does sound absurdly like a small child call-
ing “Yaj!” (“Thurt!”).

K’eo. Masked Tityra, Tityra semifasciata. Echoic. Extended to other tityras and similar
birds. Name sometimes extended to peelank’eolij.

Kusuun (kusaam). Swallows and swifts, collectively (families Hirundinidae and
Cypseluridae).

Pa’ap. Brown Jay, Psilorhinus morio. Echoic. Never called ch’el or linked with
ch’eloob in any way, so far as I can tell.

Jay group: Ch’el. These could be thought of as two “folk species” of a “folk ge-
neric,” or as two very closely related folk genera united in a broader group. In
spite of its name, the aracari toucan does not seem to be regarded as a ch’el.

Ya’ax ch’el. Green Jay, Cyanocorax yncas.
Ch’el. Yucatan Jay, Cyanocorax yucatanica. Probably echoic.

Yankotij. Wrens, and, by extension, all small brown birds. The focal one is the
Tropical House Wren Troglodytes musculus, which is literally the “one under the
wall” (see main text). Many other species occur but are not named separately.

Po’okin. Black Catbird, Melanoptila glabrirostris.
Chiik. Tropical Mockingbird, Mimus gilvus. Echoic. Often Hispanicized to chica.

K’ok’. Clay-colored Robin, Turdus grayi. Echoic. By extension, any medium-sized
brown bird that is at all similar, such as wintering thrush species from North
America. Hispanicized to coquita.

Ooxil. Yellow-green Vireo, Vireo flavoviridis. Name means “the one in the bread-
nut tree.” Also called ts’ i‘kalants’i’, which is echoic of the bird’s commonest song
phrase. One of the few cases of a bird with two names.

Ch’uyin. Rufous-browed Peppershrike, Cyclarhis gujanensis. Echoic. Extended to
other birds with songs vaguely like “chuyin.

Sojlin. Ant-tanagers, Habia rubica and H. fuscicauda.

Ts’apim. Saltators, Saltator spp. Possibly echoic. Two species occur but are not dis-
tinguished. Name routinely extended to unknown birds that look even vaguely
like saltators.

Ya’ax bech’ lu’um (“green ground-quail”). Olive Sparrow and Green-backed Spar-
row, Arremonops rufivirgatus and A. chloronotus. These two virtually identical birds
are not distinguished. They are not regarded as related to the Black-faced Ant
thrush, in spite of the similarity in name.

Azulejo. Indigo Bunting, Passerina cyanea. A Spanish name; there is no Yucatec
Maya name, as is usual with winter visitors.

158 ANDERSON Vol. 20, No. 2

Chinchinbakal. Any small yellowish bird, including goldfinches, warblers, tana-
gers with yellow underparts, etc.

Pich’. Melodious Blackbird, Dives dives. Extended to other blackbirds that may
occasionally appear.

K’aaw. Great-tailed Grackle, Quiscalus mexicanus. Echoic. Almost always said with
the diminutive: xk’aaw.

Ts’iu. Red-eyed Cowbird, Molothrus aeneus. Echoic.

Yuyum. Large orioles, focally the Alta Mira Oriole, Icterus gularis. Often Hispani-
cized to yuya.

Jonxa’anij (“the one who nests in palmettos”). Smaller orioles, focally the Hooded
Oriole, Icterus cucullatus, which is the one that really “nests in palms.” There are
controversies about where the rarer orioles fit, but usually they are called yuyum.

Mut’. Yellow-billed Cacique, Ambl holosericeus. This name appears to be the
Yucatec reflex of the widespread Maya root mut “bird.” (Ch’ich’ is a Yucatec form
that may reflect an ancient alternate root or may simply be onomatopoeic.) I do
not know why the Cacique is “the” bird par excellence, but perhaps it is related to
the tight pair-bonding of the birds (they always answer each other—the Maya
assume one of the pair has died if a call is not answered). There are other indica-
tions that this is a very important mythic bird; see Anderson and Medina Tzuc,
forthcoming.

K’uubul. Wagler’s Oropendola, Psarocolius wagleri. Echoic.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 159

Ethnobotany: A Reader. Edited by Paul Minnis. University of Oklahoma Press,
Norman. 2000. $18.95 (paper). ISBN 0-806-13180-2.

Edited by Paul Minnis and intended as a review of important works in ethno-
botany, this book aims to give the reader a broad impression of what the discipline
encompasses. Designed as a reader to accompany textbooks for upper division
coursework, the fourteen chapters in this volume come from articles previously
found in the Journal of Ethnobiology. The editor highlights contemporary studies in
ethnobotany to acquaint the readers with current methods and findings in the
field. To this end, the book is organized into four sections, ethnoecology, folk
classification, food and medicines, and agriculture, each with a short introduc-
tion.

In the first section, ethnoecology is introduced in three chapters as the study
of human perceptions and management of ecological phenomena, a discipline with
roots in human ecology and cultural ecology. A particularly strong theme in the
chapters by Janis Alcorn, Gary Nabhan, and Kat Anderson is the concept of adap-
tive management. In each of these studies ecological perturbations of plant
environments are driven by human needs, but the outcomes for ecological com-
munities as a whole are generally positive. In the language groups studied, new
resource use patterns lead to greater richness in biological system diversity at the
level of both plants and other organisms, such as birds and mammals.

The folk classification section focuses on the ways that people in different lan-
guage groups construct vocabularies related to plants, with particular emphasis
on critiquing or adding to the approaches of Brent Berlin. In the introduction
Cecil Brown presents Berlin’s most important ideas, folk classification as basically
an intellectual exercise of organizing plants into series of hierarchical categories in
the Linnean tradition. In the following two chapters, Brian Morris and Nancy
Turner question the premise of intellectual rather than utilitarian motivation for
the naming of plants in their respective studies of the Chewa in Malawi and Salish
language groups. Eugene Hunn and David French find that Sahaptin hunter-
gatherers do not organize their plant vocabulary into hierarchical series but
coordinate organism names into focused groups or peripherally to similar organ-
isms.

The longest section of the book relates to food and medicines and focuses on
human interactions with wild plants. Much ethnobotanical literature is oriented
to the study of plants as products that can be used as commodities, for example in
the pharmaceutical industry. Contrarily for this book, the authors are more inter-
ested in studying the human-plant relationship to illuminate questions of
conservation. Paul Minnis looks at famine foods in the desert borderlands be-
tween the United States and Mexico and how ethnic conceptions of these plants
change as new foods are adopted. Robert Bye examines the ethnoecology of main-
taining early-succession plant the t f edible wild greens
among the Tarahumara Indians. Eugene Hunn critiques representations in
Murdoch's 1967 Ethnographic Atlas by presenting examples of female plant-gath-
ering activities in the Columbia Plateau area. Jan Timbrook and Robert Voeks
examine tl f human interaction with medicinal plants. Timbrook
notes how certain plants move into or out of favor with Chumash people as a

160 BOOK REVIEWS Vol. 20, No. 2

result of active cultural exchange with other language groups, and Voeks describes
how the African diaspora into Brazil resulted in the use of Yoruba religious deities
to inform the medicinal uses of local flora.

The agriculture section deals with the ethnoecology of plant modification and
domestication. Gary Nabhan shows how plant domestication is an ongoing pro-
cess among several aboriginal tribes in the southwestern United States. The
maintenance of potato diversity in Peruvian highlands in the face of agricultural
modernization leads Stephen Brush to conclude that we cannot conceptualize the
Green Revolution simply as the replacement of pre-existing agricultural diversity
with new domesticated varieties. George Estabrook employs the concept of invis-
ible technology to explain why fuel-wood choices are specific to certain tasks in
central Portugal.

The main criticism of this book stems from its relatively narrow geographic
focus. Of the fourteen chapters in the book, ten relate to North American lan-
guage groups and only two treat groups outside of the Americas. This is
unfortunate given the detailed and important coverage in the ethnobotanical lit-
erature of diverse groups in Africa, Australia, and Asia. The absence of any
discussion on the medicinal knowledge of China is a particularly glaring omis-
sion in the section on food and medicines.

The strength of this volume lies in its explicit treatment of hybridized ethnobo-
tanical systems in situations where western influences lead to profound changes in
the ways that humans interact with the botanical environment. Ethnic vocabularies
and interactions with the plant world are not perceived as static, but as evolving
and changing with the needs and experiences of the groups involved in their cre-
ation. This approach is useful in conceptualizing how human perceptions of plant
resources are changing, and in turn how conservation might best be realized.

J. Anthony Abbot
Department of Geography
University of Minnesota

Journal of Ethnobiology 20(2): 161-192 Winter 2000

EASTERN SUMBANESE BIRD CLASSIFICATION

GREGORY FORTH
Department of Anthropology
University of Alberta
Edmonton, Alberta, Canada T6G 2H4

To Oemboe Hina Kapita and to the memory of Louis Onvlee (1893-1986)

ABSTRACT.—In regard to ethnozoological classification, the Austronesian-
speaking area of insular Southeast Asia is one of the least documented parts of
the world. Dictionaries of the language of eastern Sumba by Kapita (1982) and
Onvlee (1984) include over fifty names for kinds of avifauna with glosses in
Indonesian (Bahasa Indonesia) and Dutch as well

mostly From septiiaiy pegmange by the naturalist Dammerman in the 1920s.
G

author

in the domain of Rindi, the eastern Sumbanese classification of birds is discuseed
with regard to nomenclature, internal structure, and its relation to a general
ethnozoological taxonomy. On the basis of recent ornithological studies of this
part of Indonesia, the association of Sumbanese categories with scientific taxa is
also reviewed. Finally, the prominence of certain bird categories in the symbolic
idioms of ritual speech, myth, and augury is considered as a factor hypothetically
linked with eastern Sumbanese ethnoornithological classification.

Key words: eastern Sumba, Rindi, naming and classification of birds,
ethnozoological taxonomy, symbolism.

RESUMEN.—E] area de habla austranesia del sudoeste insular de Asia es una de
las partes merioe: spcamentades del mundo en on a la clasificién

Ka ita

(1982) y Onvlee (1984), incluyen alrededor de cincuenta nombres de clases de
aves, con términos en scree ape Bahasa Indonesia) y holandés, asi como

tambié t el trabajo de campo
ade Pee 4 Haweieues Dammerman en los ‘afios1920s. ‘Combinando estos
datos con la inf ica recopilada por el autor de este articulo

en el area de Rindi la dace de } pajaros del sumbanés del este es discutida
en relaci6n a la nomenclatura, estructura interna, y su relacién a una taxonomia
etnozooldégica general. Sobre la base de recientes estudios ornitologicos en esta
parte de Indonesia, también se discute la asociacién de categorias sumbanesas
con taxa cientifica. Pinannente, la prominencia de ciertas categorias de pajaros en
va ritual, el mito ne spanceeaieas apnepcammgoiets como

Pa 4 } BS

del Este.

RESUME.—La classification ethnozoologique des Austronésiens de la partie
insulaire de l’Asie du sud-est est l’une des moins connues dans le monde. Les
dictionnaires de Kapita (1982) et Onvlee (1984) sur le langage du Sumba de I’est
contiennent plus de cinquante noms d’espéces aviaires traduits en indonésien et

162 FORTH Vol. 20, No. 2

en hollandais, ainsi que leurs identifications scientifiques, dont la plupart
p . sa Pa Pan ae I 1 ‘, iots Dammerman. dans les années
1920. A partir de ces données et de l'information ethnoornithologique recueillie

nsi

par l’auteur dans le domaine du Rindi, la lature, la structure interne, ains
que relation de la classification des oi du Sumba de |’est avec une taxonomie
ethnozoologique générale sont examinées. L’association des catégories

banai t des taxa scientifiq t égal t 4 la lumiére de récentes
études ornithologiques effectuées dans cette partie de |’Indonésie. Enfin,
l’importance de certaines catégories d’oiseaux dans l’expression symbolique du
discours rituel, du myth et de l’augure est considérée comme un facteur
hypothétiquement relié a la classification etl ithologique du Sumba de l’est.

INTRODUCTION

To date no special study has been conducted into ethnozoological classifica-
tion on the Indonesian island of Sumba. The several Sumbanese languages belong
to the Austronesian family and more specifically to a Central Malayo-Polynesian
grouping (Blust 1979). As part of general ethnographic research carried out in the
eastern Sumbanese domain of Rindi in 1975 and 1976 (see Map 1), I recorded a
number of names for bird kinds. Most of these, and some others besides, appear in
dictionaries of the main eastern Sumbanese dialect of Kambera compiled by
Oemboe Hina Kapita (1982), the principal Sumbanese expert on the culture and
languages of the island, and by the late Louis Onvlee (1984), a linguist and Bible
translator and the main Western expert on Sumbanese languages. Onvlee con-
ducted research jointly with Kapita between 1926 and 1955.

e objective of this paper is to review all information recorded so far con-
cerning eastern Sumbanese ethnoornithological classification. In view of the
common appearance of birds in symbolic idioms, for example in Sumbanese ritual
language and myth, this topic is relevant to rather more than questions of
ethnotaxonomy and nomenclature. Since Onvlee wrote in Dutch while Kapita’s
work is in Bahasa Indonesia (the Indonesian national language), a particular pur-
pose is to make these data available to a wider ethnobiological, linguistic, and
anthropological audience not familiar with these languages.

Another relevant source, also in Dutch, is the writings of the naturalist K.W.
Dammerman (1926a, 1926b). On the basis of research conducted on Sumba be-
tween 14 March and 26 May 1925 (Monk et al. 1997:882), Dammerman published
32 bird names in eastern Sumbanese (Kambera) and the same number in the west-
ern Sumbanese language of Laura (transcribed by Dammerman as ‘Laora’).
Dammerman’s work is germane not only because he was the first Western scien-
tist to identify ornithological species present on the island, including two
Sumbanese endemics, but also because Onvlee, especially, relies heavily on
Dammerman (1926a) in defining eastern Sumbanese categories. Onvlee’s dictio-
nary thus includes just five scientific names that appear to be drawn from
(unspecified) sources other than Dammerman. Although both Onvlee and Kapita
significantly extend the list of Sumbanese taxa recorded by Dammerman, one un-
fortunate result of the two lexicographers’ reliance on the early naturalist is that
many of the Latin binomials incorporated in their dictionaries are now superseded.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 163

Nevertheless, modern binomials are for the most part readily inferred from a re-
cent comprehensive ornithological study of Sumba and other eastern Indonesian
islands by B. Coates and K.D. Bishop (1997).

Another pitfall of the two » lexicographers' reliance on : Dammerman i is of course

lity that tl twas mistaken i in associ taxa wit

particular scientific species and genera. Modern iio tateae on bird kinds present
on the island, however, supported by information drawn from Onvlee and Kapita,
as well as linguistic and ethnographic data compiled by the present author, sug-
gest that Dammerman’s identifications, scientific as well as ethnozoological, were
largely accurate. In this regard as well, the work of Coates and Bishop (1997) has
proven especially useful. In identifying eastern Sumbanese bird taxa I have also
been able to draw on information provided by several eastern Sumbanese I ques-
tioned in 1999 in Kupang, the provincial capital located in western Timor.! With
the aid of Coates and Bishop’s field guide, the Kupang informants were able to
clarify a number of important issues concerning the identity of birds named by
Sumbanese terms.

In spite of possible remaining gaps and ambiguities, the data compiled here
appear sufficiently complete to analyze eastern Sumbanese bird classification and
to establish the general outlines of the system. A point of some relevance in this
regard is the comprehensive nature of Onvlee’s dictionary. As a student of
Sumbanese languages who resided on the island for over 20 years, it cannot easily
be assumed that Onvlee would have been unfamiliar with many terms referring
to birds, even if he had been unable to identify them scientifically. Also notewor-
thy is the circumstance that Kapita, although a native Sumbanese and probably
better able to draw on a local knowledge of birds, does not record any names that
do not also appear in Onvlee’s dictionary. In fact, Onvlee’s work is slightly more
complete. This is not to suggest that further research could not uncover additional
names for birds. Indeed, three terms encountered in the course of my own enqui-
ries, in Rindi and among Sumbanese in Kupang (landu witu, mabihi, rawa kawi,
see Table 1), are not listed by either Onvlee or Kapita, nor can they obviously be
accounted for as local variants of terms the two lexicographers do record. Even so,
the total of nearly sixty-odd I discuss below (56 plus as many as ten unnumbered
productive binomials, which total however includes a number of synonymous
usages) compares favourably with the 66 bird names (or 59, if probable synonyms
and terms with non-empirical referents are excluded) recorded for the Nage of
Flores (Forth 1996, 1999), and the 54 terms recorded for the Nuaulu of Seram (Ellen
1993b). The figure for Nuaulu should be taken in the context of a total of 195 spe-
cies recorded for the island of Seram (Ellen 1993b:57). The comparable total for
Sumba as a whole is 161 (see Appendix 1; cf. Monk et al. 1997:354, who list 103
breeding species for Sumba).° In view of environmental differences between east-
ern Sumba and the more heavily forested and better watered western part of the
island, moreover, it can safely be assumed that the number for eastern Sumba is
rather lower than this.

However the results of the present enquiry are judged, one may hope that
they will be of sufficient interest to spur others to advance the study of eastern
Sumbanese ethnozoological classification. Indeed, publishing information on the

164 FORTH Vol. 20, No. 2

ethnology of Sepa birds i is at present a matter of some seca Owing 1 in
part to the advance of th national language, |

names for natural kinds i is reported to be decreasing, especially 2 among younger
people. In addition, as a result of increased hunting (nowadays particularly with
air rifles), trapping to supply the export trade in cage-birds, and the destruction of
natural habitats caused by human population increase, clearing of forests for agri-
culture and lumber, and modern development efforts, in some cases the species to
which the indigenous names refer are themselves probably in decline (see Monk
et al. 1997:821-35; Coates and Bishop 1997:39-40).*

as
ie SiGe pee
BALI pe
ee BLEN
TIMOR
10°} Rindi Ku oo
SEMAUJ?

SAVU,
o> ,
ROTI

1 18E 118° 120° ie ro
FIGURE 1.— The Lesser Sunda islands showing Sumba and Rindi

BIRDS IN SUMBANESE ETHNOTAXONOMY

main product of the present exercise is a list of Sumbanese names for

birds with provisional identifications. These are set out in Table 1. From linguistic
and etl compiled by the present author in 1975-76, supple-
mented by. Kapita (1982) and Onvlee (1984), it is possible to locate these bird taxa
within a broader context of folk taxonomy (see Figure 2). The Sumbanese classifi-
cation of animal kinds can readily b th Jel of ethnobiological
taxonomy developed by Berlin and his associates feel Berlin 1992). Berlin’s terms—
unique beginner, life-form, intermediate, folk generic, and so on—thus provide a
useful means of analyzing local categories and exploring their mutual relation-
ship. The nature of eastern Sumbanese bird nomenclature also lends support to
Berlin’ S approach SDORRE as it — a Classification based primarily on observ-
eatures including vocal behaviour) rather than

on non-perceptual criteria. All the same, ‘the Sumbanese categories are further re-
lated in other, non-taxonomic ways, thus participating in a separate, symbolic
classification (cf. Forth 1998b:190-91). Non-taxonomic articulations of bird catego-

Winter 2000 JOURNAL OF ETHNOBIOLOGY 165

ries are discussed after a review of nomenclature. In the concluding section I briefly
consider the possible influence of symbolic value on the ethnoornithological tax-
onomy in general.

Whether tern Suml tl 1 | classification i tes a term
se Ai designating ‘animal(s) ‘thus a named “ unique beginner” (Ber-
lin 1992:15)—is somewhat debatable. The main candidate is the expression
makayidi-yadaku, ‘things that move,’ which both Onvlee and Kapita further gloss
as ‘the whole of creation’ (or ‘all creatures,’ semua makhluk, Kapita 1982 s.v.
kayidiku). The phrase is based on the compound yidi-yada, COmprininy we
roughl ‘to move,’ and producing an allit
symbolism comparable to English ‘topsy-turvy’ or ‘twist (and) turn’ and Bahasa
Indonesia gerak gerik, which indeed translates the Sumbanese pair. In combina-
tion the affixes ka- and -ku lend a repetitive or continuous quality to the basic
verbal compound, while ma- (‘that, what, that which’) renders the nominal sense.
Although logically makayidi-yadaku could include Homo sapiens, the phrase is
not normally applied to human beings and is thus comparable, for example, to the
BAG Adama English use of _— bid apete the word contrasts with ‘man, human.’
The sense of the Sumbanese phrase, that
is, he Bel that it appears mostly to be used as a reference to ‘animals’ in general
rather than to single individuals. Consistent with this, Kapita does not gloss
makayidi-yadaku as binatang (‘animal,’ Bahasa Indonesia), nor does Onvlee trans-
late it with Dutch dier or beest. (Also, the last author gives the phrase as da
makayidi-yadaku, thus incorporating the plural article da.) Nevertheless,
makayidi-yadaku is a term Sumbanese regularly employ to refer to animals, not
an expression constructed in response to lexicographical questioning. That they
possess a category of ‘animal’, moreover, is indicated by the numeral classifier
ngiu (tail), which is used when ennumerating animals but not humans or other
living things.°

Although makayidi-yadaku applies to all non-human animals, its focus ap-
pears to be undomesticated kinds. Interestingly, yada, the root of yadaku, can
mean ‘wild, untamed, difficult to tame,’ as well as ‘to move, be capable of move-
ment’ (Onvlee, Kapita s.v. yada). Yet this sense—probably involving a metonymy
whereby a propensity to movement connotes the opposite of tameness—is not
clearly decisive for its incorporation in the longer expression.® Domestic animals
are collectively called banda. This however is a secondary meaning of a word, the
main sense of which is ‘goods, possessions, wealth’ (cf. Bahasa Indonesia benda).
As this derivation may suggest, the term moreover refers particularly to large live-
stock, a principal form of wealth in the Sumbanese traditional economy. Some of
my Rindi informants claimed that banda could be understood in the wider sense
of ‘animal’ (Bahasa Indonesia binatang), with wild animals then being specified
as banda matamba (wild banda); but neither Onvlee nor Kapita record the latter
phrase and I suspect that, even at present, it is not a widespread or standard us-
age. Whatever the case, and regardless of the extent to which makayidi-yadaku
and banda may share common referents, the two categories are not obviously re-
lated by taxonomic inclusion. By the same token, banda can be characterized as a
utilitarian category, referring mostly if not entirely to a class of economic values,

166 FORTH Vol. 20, No. 2

while makayidi-yadaku is a descriptive phrase naming a category of living things
distinguished explicitly on behavioural, and implicitly on morphological, grounds.

Since makayidi-yadaku refers to an ability to move, it is significant that the
taxon subsumes two major categories both of which are denoted by phrases refer-
ring to specific kinds of movement. Both constitute “life-form” taxa, in the sense
defined by Brown (1979) and Berlin (1992:15ff). One is mabei, ‘things that creep,
crawl,’ a large and internally diverse taxon that can include insects, arachnids,
reptiles, amphibians, and even fish. The other is mahawurungu, ‘things that fly.’
Not surprisingly, birds are focal to mahawurungu. Contrary to what the name
would suggest, not every sort of aerial creature is included in the taxon. For ex-
ample, the ‘Flying dragon’ (probably Draco volans, cf. Dammerman 1926a:218, in
Rindi called kumbu lai hawurungu, ‘flying lizard’) is reckoned not to belong to the
mahawurungu but rather to the mabei, together with other lizards. Some Rindi
thought that flying insects—such as houseflies, wasps, and bees—should be
counted as mahawurungu. However, because they crawl as well, the creatures are
also—and probably more usually—classified as mabei. It almost goes without say-
ing that bats are classified as mahawurungu. Since the term translates exactly as
‘flying things’ it may be questioned whether Sumbanese, like most folk zoolo-
gists, actually classify bats as kinds of ‘birds.’ That they do so is indicated by the
fact that Rindi mentioned bats with birds when listing names of ‘flying things,’ as
well as by a local belief that bats lay eggs.

The derivation of mahawurungu requires comment. Hawurungu (to fly) com-
prises a fused prefix, ha-, and wurungu, evidently a cognate of Malay (or Bahasa
Indonesia) burung (bird) and, following some authors, a reflex of an Austronesian
protoform referring to birds in general (see Dempwolff 1938; Lopez n.d., cited in
Wurm and Wilson 1975). The same sources give no indication that wurungu re-
flects protoforms meaning ‘to fly,’ and it may therefore be supposed that the
Sumbanese word derives from a term that once denoted birds but has since, and
with the addition of the prefix ha-, acquired a verbal sense which, by way of fur-
ther prefixing, has as it were reverted to its original meaning.” There is however
an alternative interpretation. Both Kapita and Onvlee indicate a derivation of
hawurungu (to fly) from wuru, denoting the sound produced by the wings of a
flying bird, or by an object that is thrown into the air. In that case, the eastern
Sumbanese word for ‘to fly’ may even be based on an onomatopoeia, and its re-
semblance with burung (bird) may be coincidental.®

Connected with the opposed modes of locomotion to which the terms refer,
mahawurungu and mabei comprise creatures that typically occupy areas located
respectively above and below the human domain. Indeed, it is partly in this re-
gard that one can comprehend the inclusion of fish (iyangu) among the mabei. For
although fish do not actually ‘creep’ or ‘crawl, living in water they inhabit a re-
gion beneath the space inhabited by humans. In addition, fish resemble other mabei,
particularly reptiles, in several obvious morphological respects, thus suggesting a
switch from behavioural criteria as a basis of classification.” Although they consti-
tute two major named life-form taxa within th kayidi-yadaku (moving things),
mabei and mahawurungu obviously do not exhaust the Sumbanese category of
‘animals.’ Most noticeably excluded are mammals, both wild and domesticated.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 167

These are designated only by individual terms (for example, ruha ‘deer;’ buti ‘mon-
key;’ ringu ‘dugong’) which in most cases refer to terminal taxa coinciding with
scientific species. The only distinction made within these categories is wild
and domesticated varieties of what are idered single kinds. Thus,
the wild cat and wild swine are called meo rumba and wei rumba (rumba is ‘grass,
bush’) while their domestic counterparts are usually referred to simply as meo
(cat) and wei (pig). As can be seen from Table 1, something similar is done when
distinguishing wild and domestic kinds of fowls (manu) and ducks (rendi).

2a (Tree Diagram)

{unique beginner]
makayidi-yadaku

[life-forms]
awurungu mabei (other animals, e.g. mammals)
(flying things) (creeping things)

[intermediates]

rawa iRitu mdanginu (other birds

(or 'Columbiformes') (small passerines)

[folk generics] i

rawa kamukumukitu marakuku manginu (other named

rawa tana etc. (=Lonchura, etc.) generics)
[varieties]

manginu uhu, manginu wataru, etc.

FIGURE 2.—Bird taxa in eastern Sumbanese ethnozoological classification

168 FORTH Vol. 20, No. 2

Since Sumbanese are quite clear that all bird inst f makayidi-yadaku,
this category and mahawurungu are unambiguously related by taxonomic inclu-
sion (see Figure 2). For the most part, the ‘bird’ taxon immediately includes a series
of terminal generic taxa (using “generic” in the sense proposed by Berlin 1992:53-
53) that are not further divided into named sub-types or varieties. The resulting
structure of three levels is of course quite typical of folk classifications of living
things.

At the same time, the names recorded in Table 1 include some 28 binomials,
and of these 16 appear to be productive in the sense that the second component
specifies the referent as an instance of the class labelled by the first term. Thus,
rawa tana, for example, specifies a member of the more inclusive category rawa.
Other binomials are unproductive. Several comprise two words describing a fea-
ture of the bird (e.g., katua wei, laleba nggangga, landu witu, mbera wurungu).
The name totoru laka is also descriptive, combining a probable instance of ono-
matopoeia with a colour term (laka, see Table 1). The six names that include kulu-,
while formally similar to productive binomials, are not actually productive. De-
spite the probable connection between kulu- and words in other Austronesian
languages that mean ‘bird’ or a kind of bird (e.g., Nage/central Flores kolo ‘dove’
[Forth 1996:106]), tl I termd t denote any particular class of bird.1°
As can be seen from Table 1, the birds thus named are quite various, and indeed,
Rindi denied that they were related to one another, or were variants of a single
kind. What is more, kulu- also occurs as a component of terms designating natural
kinds other than birds (see kulu-kengu, millipede; kulu-nderi, a kind of grass; see
also kulungu, a small sort of mouse and kulu, a kind of breadfruit, Artocarpus com-
munis [Onvlee 1984]). In this respect, the morpheme is reminiscent of central Flores
(Nage, Ngadha) kaka, which similarly forms part of bird names but further oc-
curs in terms for other natural kinds (regarding Nage, see Forth 1996:101; also
Verheijen 1963 on the same element in Manggarai).

Most productive binomials form pairs, thus distinguishing just two taxa within
the more inclusive kind (e.g., rendi manu and rendi yalangu). By far the most promi-
nent instance of productive binomials are the six terms denoting kinds of rawa, all
of which refer to members of the Columbidae (pigeons and doves). Not all mem-
bers of this family, however, are specified with compounds of rawa. Others,
particularly it seems smaller members, are named kulu-ndiha or mbara, a word
that is itself modified to produce terms distinguishing different kinds of dove (see
mbara manu and mbara nggela). None of the available evidence indicates that
either mbara or kulu-ndiha is straightforwardly included within a broader taxon
labelled rawa. At the same time, some of the referents of rawa, applying at the
folk generic level, may overlap with those of mbara. Onvlee thus gives Treron
teysmanni as the referent of both mbara and rawa ratu. (Kapita’s gloss, by con-
trast, links only Streptopelia chinensis with mbara, whereas Onvlee, rather

I ts thi ies Wi Sumk term.) Onvlee further equates
this small dove is a probable

. gly, co this species with
rawa kakoruku with Geopelia maugei, while in Rindi
referent of mbara nggela.

The terms rawa and mbara are also comparable insofar as both are used with-
out qualification to refer to particular kinds further named by compounds. For
example, rawa refers especially to Ducula aenea (rawa kamukumu), the Imperial

Winter 2000 JOURNAL OF ETHNOBIOLOGY 169

pigeon, a bird that in Rindi at least is further called rawa manu and is the largest
of the Sumbanese Columbiformes. " The appearance of mbara in compound ex-
pressions denoting creatures that g rice crops (see Table 2) strongly
suggests that this term specifies Streptopelia and Geopelia, although the cuckoo-
dove (Macropygia ruficeps) also consumes rice (Coates and Bishop 1997). Larger
pigeons, including both the genera Ducula and Treron, are by contrast fruit-eaters.
Hence, it is conceivable that Sumbanese apply rawa generally to Columbiformes
while reserving mbara for particular instances, distinguished either absolutely or
situationally, according to their dietary behaviour—a matter that bears on an ob-
vious utilitarian or practical consideration.

Whatever the exact relation between rawa and mbara, the former term can be
seen as labelling an “intermediate” grouping, falling between the levels of life-
form taxa and folk genera (Berlin 1992:139-160; see Figure 2). On the other hand,
this status might more accurately be assigned to a larger, unnamed class of
Columbiformes, most of which—but not all—are designated as rawa while others
are called mbara (or by terms including these) or kulu-ndiha. In other words, all
Columbiformes may be treated as a single, distinct—though strictly-speaking co-
vert—intermediate category in eastern S | taxonomy
(cf. Forth 1996). Another candidate for intermediate status is a group comprising
most or all of the diurnal raptors. As noted in Table 1, Falconiformes are generally
labelled as ikitu, although the term’s primary referent appears to be the Brahminy
kite, a bird more specifically named as ikitu marakuku (‘White necked/throated
ikitu’). Consistent with this broader use of ikitu are three western Sumbanese
(Laura) terms for diurnal raptors listed by Dammerman (1926a), all of which in-
clude the cognate wikita. These are wikita liza, wikita rewa koko, and wikita
labo, and correspond respectively to eastern Sumbanese kapaha, ikitu/ikitu
marakuku, and_mbaku/mbaku tehiku.

ikitu

ikitu tolungu (‘meat hawks’)

ikitu mafakuku
(Brahminy Kite) | mbaku kapaha tariku, etc.

mbaku mbaku tehiku

FIGURE 3.—A Rindi classification of diurnal raptors (ikitu)

170 FORTH Vol. 20, No. 2

According to the detailed report of a Rindi informant, ikitu, understood as a
general term for diurnal raptors, comprises two divisions. One includes a single
kind, the Brahminy kite (marakuku). The second, labelled as ikitu tolungu (roughly,
‘meat hawks’), admits other named kinds (kapaha, mbaku, tariku; see Table 1),
including members of both the Accipitridae and the Falconidae (see Figure 3).!
The same source specified the following order of size among members of the sec-
ond division: mbaku tehiku (Haliaeetus leucogaster, the largest), mbaku, kapaha,
tariku (the smallest).!> Unfortunately, I was unable to determine how far this clas-
sification of diurnal birds of prey is shared among eastern Sumbanese. Essentially,
it separates the Brahminy kite (Haliastur indus), the raptor most closely associated
with the name ikitu, from all other members of the broader taxon identified with
the same term. Since Haliastur indus also eats flesh, the contrasting term “meat
hawk’ does not illuminate the basis of the division. It is however possible that the
bird’s greater reliance on scavenging rather than killing may be relevant.'* Also,
while the Brahminy kite is often sighted inland—where, like other raptors, it is
perceived by Sumbanese as a threat to poultry—it is more commonly found on
the coast. This of course is also the preferred haunt of the sea-eagle (mbaku or
mbaku tehiku). But the sea-eagle is nevertheless distinguished from the kite inso-
far as, according to Rindi at least, the former is the one raptor that does not steal
chickens.

As elsewhere in Indonesia, on Sumba large raptors play a prominent role in
creation mythology. In parts of Flores, it is the Brahminy kite in particular that is
associated with a creator deity (Laubscher 1975; see also Freeman 1960 and Metcalf
1976, regarding the Brahminy kite in Bornean cosmology and augury). In eastern
Sumba, by contrast, this part is taken by Haliaeetus leucogaster, the White-bellied
sea-eagle (see below). Yet Rindi descriptions of the mythical bird sometimes sug-
gested features of the Brahminy kite. Relevant to a possible conflation of the two
birds which this suggests is Onvlee’s gloss, which indicates that the large kite may
be classified as a kind of mbaku, more particularly the ‘White-throated mbaku’
(mbaku bara kuku, see Table 1), a situation that recalls the sort of classificatory
overlap already evidenced with regard to Columbiformes. Effecting an inversion
of the relation of inclusion implied by Onvlee, my Rindi informant listed both
mbaku and mbaku tehiku under ikitu tolungu. However, his description of mem-
bers of the first taxon (mbaku) did not unambiguously indicate the Brahminy kite,
and it may well be that, in Rindi at any rate, this term is applied to a variety of
larger eagles and hawks. Regional variation in folk classification may be relevant
in resolving these issues. So too may colour phases and environmental contexts of
large raptors. For example, Haliastur indus (the Brahminy kite) may be classified
as a kind of mbaku specifically when encountered near the sea (cf. Ellen 1993a).

Another folk taxon represented as comprising a number of distinct types is
manginu. Like Onvlee (1984), Rindi informants claimed there were numerous
‘kinds’ (Bahasa Indonesia macam, jenis) of manginu.!© Mentioned among these
were manginu uhu (‘rice manginu’), manginu wataru (‘maize manginu’), manginu
tana (‘ground [-dwelling] manginu’), and manginu kulu-kataitaku; but there were
reckoned to be many more besides, whose names were not known. With the ex-
ception of kulu-kataitaku, a term used alone to label a generic taxon (see also

Winter 2000 JOURNAL OF ETHNOBIOLOGY 171

manginu kadu, Table 1), these several qualifiers however suggest simple descrip-
tors serving merely to distinguish varieties of a single basic kind. In its most focal
sense, manginu denotes Estrildine finches, especially munias (genus Lonchura) and
similar small birds that do damage to rice and other cereal crops. Thus, the com-
pound mbara manginu, ‘doves and munias,’ is a standard expression for birds
that destroy ripening cereals. Yet in a more inclusive application the term further
refers to a large variety of generally small passerine birds (or “dicky-birds”; see
Figure 2), in which context it suggests an intermediate grouping comparable to
ikitu and rawa (cf. the Nage term peti, ana peti, Forth 1996, which is similarly
paired with a term for Columbiformes, kolo, to refer to crop pests).!” At the same
time, in its more inclusive sense manginu differs from these insofar as it appears
not to be simultaneously identified with a single, undivided folk generic (as ex-
emplified by ikitu marakuku and rawa kamukumnu). This broader usage of manginu
would moreover account for the apparent absence from the eastern Sumbanese
ethnoornithological lexicon of special (that is, folk generic) terms for small birds
such as flowerpeckers, flycatchers, honeyeaters, sunbirds, titmice, wagtails, war-
blers, white-eyes, and whistlers (see Appendix 1).'° Apart from the focal finches
and other small birds designated by special names, I would estimate that as many
as 25 species listed in Appendix 1 could be classified simply as manginu.

Eastern Sumbanese categories include two other candidates for intermediate
status. One is an implicit taxon comprising the two black birds named nggangga
(Large-billed crow) and laleba nggangga (a drongo, probably the Wallacean
drongo). That these form a set is suggested not only by their physical resemblance,
particularly in regard to colour, but by the designation of one as the ‘sister’s child’
(laleba) of the other (regarding the use of kin terms as evidence of covert interme-
diate taxa, see Berlin 1992:145). The other instance of a possible intermediate taxon
is panii. Usages recorded by Onvlee indicate that this term serves both to name
the Flying fox and as a label for a more inclusive class that also includes much
smaller bats (e.g., those called pahomba in Rindi). By contrast, evidence from Rindi
suggests that panii may there refer only to Flying foxes (Pteropus spp.) while smaller
members of the Cheiroptera may consistently be named with other terms.

NOMENCLATURE

As mentioned above, several names comprising two lexemes refer to empiri-
cal or reputed characteristics of the birds so named (e.g., katua wei), as do
components of productive binomials (e.g., ikitu marakuku). Where the meaning
of a name, or part of a name, is analyzable, this is indicated in Table 1. The six
names comprising kulu- all have analyzable second elements. Of these, four refer
to visible features (e.g., -kadu, ‘horn, horn-coloured’); one is onomatopoeic (-
kawaki); while the other possibly refers to some general quality attributed to the
bird (-ndiha, ‘good, attractive’). In contrast, among names constituted of single
lexemes, only two—kola and mbaku—possesses an independent meaning that
describes a morphological or behavioural feature of its avifaunal referent. Another
possibility is nggokaria (heron), insofar as this may be a variant of nggokaru, ‘to
stretch, crane the neck,’ which Onvlee further lists as the word for ‘heron’ in the
Lewa dialect of eastern Sumba. (Alternatively, nggokaria may comprise two ele-

172 FORTH Vol. 20, No. 2

ments, nggoka and ria, the first of which recalls central Flores gako, ‘large heron,’
see Nage gako tasi, Forth 1996.) Laleba nggangga indirectly refers to a morpho-
logical feature insofar as it alludes to a resemblance with the Large-billed crow,
nggangga. By the same token, it is the only name that includes a kin term (laleba,
‘sister’s child’). (Ana in ananjaki cannot be construed as ‘child’ in the sense of a
relationship term.) About 17 of the names in Table 1 describe physical features of
their referents. Others refer to i tal iati (see landu witu, mbaku
tehiku, rawa tana).

Only a minority of names—about seven—are locally recognized as wholly or
partly onomatopoeic (koka, kui, kulu-kawaki, nggangga, rawa kamukumu, rawa
kakoruku, tutuku). Judging from reported vocalizations, another eight are pos-
sible onomatopoeia (see kahuhu, kahiku, kaluki, kutuku, kuu, pipi, pirihu, totoru
laka). The large number of avifaunal names beginning with ka-, a fused prefix,
reflects a general lexical feature of Sumbanese languages. However, in some cases
(notably kamukumu and kakoruku, but see also, e.g., kahiku, kahuhu) the prefix
specifies something producing a sound denoted by the root (mukumu, koru; see
Table 1). In other instances, the fused suffix -ku appears to effect a similar result
(see kutuku, tutuku). The bl bet ggokaria and nggonggali is prob-
ably superficial.

Of the nearly sixty terms listed in Table 1, 33 or about 60 per cent, appear to
name single scientific species. A large majority of these are the single representa-
tives of their genera present on Sumba. If Dammerman and the lexicographers are
accurate, a further three probably refer to just two members of the same genera
(ananjaki, kalewaru, powa). Eight names, including three designating taxa of in-
termediate status, apparently refer to three or more species of the same genera or
indeed two or more genera (ikitu, karata, manginu, mbara, mbera wurungu,
nggokaria, pipi, rawa). In the remaining cases, information is insufficient to make
a determination.

How far names for folk generic and intermediate taxa are further applied, or
“extended” (Ellen 1993b), to species and higher order groupings associated with
scientific kinds listed in Table 1 cannot be decided without more research into
Sumbanese knowledge of birds. There are, for example, five species of Hirundo
(see mbera wurungu) on Sumba, numerous members of the Ardeidae (including
Ardea, Egretta, and representatives of six other genera; see nggokaria), and at least
eight species of the Scolopacidae (sandpipers, snipes and allies; see pipi, kahuhu).
Yet one cannot know a priori how many members of these ornithological families
and genera Sumbanese would regularly identify with the indigenous terms. It is
similarly unclear whether less common Columbiformes (such as the White-throated
pigeon, Little cuckoo-dove, Nicobar pigeon, and two species of Ptilinopus, or fruit-
doves) would be classified as rawa or named with other, thus far unrecorded,
terms. In accordance with the classification illustrated in Figure 3, it is a reason-
able surmise that ikitu can be applied without further qualification to
Falconiformes, including five members of the Accipitridae recorded on Sumba
but not indicated as possible referents of terms listed in Table 1. In regard to size,
these are either comparable to the Brahminy kite (ikitu marakuku) or White-bel-
lied sea-eagle (mbaku tehiku, the largest Sumbanese raptor) or fall somewhere
between the two.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 173

These sorts of questions of course turn partly on the completeness of the bird
names listed in Table 1, an issue addressed earlier. As suggested, it is probable that
a large number of small passerine species are simply classified as manginu. Com-
parative evidence supports this. For example, Ellen, writing on the Nuaulu of
Seram, refers to an “under-differentiation of passerines” that is “quite astonish-
ing” (1993b:79). (A more general “under-differentiation of avifauna”, or

classificatory “lumping”, Ellen partly att and straggling.) Given
that eastern Sumbanese are not a maritime people, one should also not be ‘Sur-
prised that seabirds and species i I ts are not

well represented in Table A. a Larger passerine birds recorded on Sumba which
are not obv with known indigenous terms, but which one might
expect to be. separately named, include bee-eaters, cuckoos, cuckoo-shrikes, and
the dollarbird (see Appendix 1). Yet even some of these could conceivably be named
by employing already attested categories. Smaller cuckoos, for example, could be
classified with the koel (kutuku), as might the larger Channel-billed cuckoo
(Scythrops novaehollandiae), while bee-eaters and other larger passerines are argu-
ably not so large that they could not be included in the category manginu.*! A
comparable case are small kingfishers (Alcedinidae), if these are not classified with
the significantly larger Halycyonidae as kahiku. (Also not to be ignored is the fact
that Sumbanese has a single life-form label—mahawurungu—that can simply be
applied to identify any bird not further classifiable.) In marked contrast, the names
listed in Table 1 include no fewer than five terms referring to parrots (kaka, kariku,
katala, pirihu, wowangu). Not only do all of these apparently denote single scien-
tific species, but the five kinds exhaust parrot species occurring on Sumba. Thus,
even though they do not compose a distinct emu as grouping —or at least
not one that is named—the Sumbanese Psittacida
elaborated to an even greater degree . diurnal raptors (ikitu) and
Columbiformes (rawa).

TABLE 1.—Eastern Sumbanese bird names

1. ananjaki (or ana njaki). One or more species of Anthus (D 12: Anthus rufulus). Two
present on Sumba include A. novaeseelandiae (Richard's pipit) and A. gustavi (Pechora
pipit). As in other eastern Indonesian languages ana, ‘child’, can also denote a small
version or instance of something (cf. Nage ana go, probably A. novaeseelandiae, Forth
1996). Njaki apparently has no independent meaning.

ikitu Haliastur indus intermedius, Brahminy kite (D 30, Haliastur intermedius). In Rindi
the term refers more generally to Falconiformes. Kapita glosses it as BI elan

(‘hawk’) and ‘Palconidae’ (sic).

ikitu marakuku. Also a reference to H.indus. oe means ‘white-neck(ed),
throat(-ed).’ The bird is also called by this term alon

kahiku. Halcyon chloris (D 19). The only other len kingfisher present on Sumba
is H. australasia (C&B). Whether the Sumbanese term might also refer to other
kingfishers is not known. In view of one cry of H. chloris reproduced as ‘kick kyew’
(C&B), the name, particularly the root hiku, is probably onomatopoeic.

kahuhu. A small shorebird, sandpiper. Kapita identifies the bird with the species
Dammerman lists for pipi (see below). The word is probably unrelated to huhu in the
sense of ‘breast, milk’ and is possibly onomatopoeic (see ‘teu-hu-hu’, the call of the
Common redshank, Tringa totanus, C&B).

Hod

174

hw
-

~_
—

[RS
14.

FORTH Vol. 20, No. 2

kaka. Cacatua sulphurea citrinocristata, Yellow-crested cockatoo (D 24: Cacatua
citrinocristata). Onvlee also lists kaka ratu, or ‘royal cockatoo’, as a reference to a
cockatoo ‘with red eyes’, but this appears not to refer to a separate species.
kalewaru. Collocalia spp. (D 18), swiftlets. Species occurring on Sumba include the
Edible-nest swiftlet, C. fuciphaga, and the Glossy swiflet, C. esculenta. Nests of the
former species are collected for sale and eventual export. The local name kalew

(cf. Manggarai lawar, ‘swiftlets and swallows’, Verheijen 1963) is related to BI kelawar
or kelelawar, referring to a small bat. It is therefore of some interest that both swiftlets
and bats characteristically roost in caves, often the same caves. Onvlee glosses the
almost identical name kaliwaru as ‘swallow’ and, referring to Dammerman (D 17),
as Hirundo. Dammerman in fact gives only mbera wurungu (see below) as the eastern
Sumbanese name for swallows, though for western Sumba he lists the cognate
kalewara for both Hirundo and Collocalia

kaluki. Megapodius reinwardt reinwardt (D 38: Megapodius duperreyi), Reinwardt’s
scrubfowl. The name plausibly imitates part of the bird’s call, given by C&B as *kli-
au-kau’ (see wundu, below).

kapaha. A small falcon, kestrel (D 29: Cerchneis occidentalis). Following C & B and the
descriptions given by Dammerman (1926a:214, 1926b:22), the term is likely to
include Falco moluccensis, the Moluccan kestrel or Spotted kestrel. (Another small
falcon is F. longipennis, the Australian hobby, but this has only once been recorded on
Sumba.) Kapita similarly describes the bird as a small Falconiforme with speckled
plumage. Whether the name is related to pahangu (=paha + -ngu), meaning ‘to drop
(trans.), let drop’, is not indicated by the lexicographers.

kapi padangu. Recorded in Rindi, the term does not appear in either Onvlee or
Kapita. Kapi means ‘to flicker (of a light or fire)’; padangu is ‘plain, pasture.’ In view
of kapi mama, ‘firefly’, the term possibly refers to an insect rather than a bird.
kapiru. Described by Onvlee as a “small red bird, a ground thrush” (Dutch
grondlijster) which lays eggs in “a hole in the ground”. Kapita’s gloss “Pittadae” (sic)
indicates a pitta. Although not actually in holes, Pittas in general do nest on or near
the ground (C&B). The only pitta recorded on Sumba is Pitta elegans maria, the
Elegant pitta, which has a red belly and vent and is described as “locally common or
moderately common” on Sumba (C&B). Kapita’s further gloss, kutilang, an Indone-
sian (BI) name for a kind of bulbul (none of which are natural to Sumba), can
probably be ignored.

. karata. Terns (Laridae, sub-family Sterninae), probably including the Gull-billed

ci Gelochelidon nilotica, and one or more species of Chlidonias and Sterna. Rindi
ormants stated there were two kinds of karata but could not distinguish them by
name. A similar account was given by a Kupang informant, who described a ‘pure
white’ variety that occurs near inland lakes and paddy fields and another sort,
cream-coloured with dark marks on the back of the head, which is found on the
coast. Onvlee and Kapita appear mistaken in identifying karata with, respectively,
gulls (Dutch zeemeeuw) and Jaegers (BI burung camar). There are no gulls (Larinae)
on Sumba
kariku. Eclectus roratus cornelia (D 26: Eclectus cornelia), the Eclectus parrot. Rindi
distinguish red and green varieties (kariku rara and kariku muru). As Dammerman
(1926a:213-14) notes, these are actually the female and male of the same species, a
fact recognized by Rindi people as well.
katala. Tanygnathus megalorynchos sumbensis (D 27), Great-billed parrot.
katua wei. The name means ‘companion of the pig.’ Kapita further glosses it as ‘pig
spirit’ (BI roh babi). Both he and Onvlee identify this as an owl whose call indicates
the presence of wild pigs, a notion encountered elsewhere in Indonesia (Forth
1998b). Information is insufficient to determine whether katua wei refers to a

Winter 2000 JOURNAL OF ETHNOBIOLOGY 175

i
o

_
>

~
™~

=
~

~_
K=)

)
=

No
—

particular species of owl or whether this is one of the Strigidae or the Tytonidae (cf.
wangi below). Among the Strigidae, C&B report just one, unidentified, member of
the genus Otus on Sumba, as well as Ninox rudolfi, the endemic Sumba boobook.
Quite possibly, katua wei denotes one or more varieties of nocturnal sound, rather
than a particular kind of visible owl.

koka. Philemon buceroides neglectus (D 11; not cited by Onvlee), the Helmeted
friarbird, the only Philemon species present on Sumba. The name, also rendered
contextually as nggauka, is described as onomatopoeic.

kola. A diurnal raptor, listed by Onvlee as ‘Astur torquatus’, distinguished by a ‘neck
ring of feathers’ (cf. Latin ‘torquatus’, ‘wearing a twisted collar or necklace’). The
name probably derives from kola in the sense of ‘speckled, flecked.’ Kapita glosses
the term as ‘falcon, Peregrine falcon’ (BI alap-alap; Falco_peregrinus).

kuu. Both Onvlee and Kapita identify this as a kind of kite (Dutch bastaard-wouw)
and specifically as ‘Elanus hypoleucos’, an apparent reference to the Black-winged
kite, Elanus caeruleus hypoleucos. Neither the name nor the bird is listed by
Dammerman. As Onvlee notes, the call of this bird, usually heard in the evening, is
considered inauspicious. In Rindi, it is more particularly regarded as a manifestation
of a witch. Consistent with this association are the partly crepuscular habits of Elanus
caeruleus (C&B 1997:247), since Sumbanese witches, too, are believed to zs Sey
active at twilight. On the other hand, Rindi described kuu as a ‘black’, o

coloured, bird, as did informants in Kupang. That this does not entirely cand with
the plumage of the Black-winged kite may be explained by the bird’s nocturnal
associations; indeed on this account Kupang informants claimed never to have
clearly seen the bird. Alternatively, kuu might refer to another bird altogether. In
Umalulu, a domain immediately to the north of Rindi, I was shown a photograph of
a dead bird identified as a kuu which had been killed because of its inauspicious
calling after dark and which, I recall, resembled a female koel (see kutuku, below).
Kruyt (1922:559), apparently inaccurately, describes kuu (transcribed as ‘koeoe’) as
another name for the owl otherwise called wangi (see below). He also characterizes
the name as onomatopoeic, as did one Kupang informant. Noteworthy here is the
arguable resemblance of kuu to a nocturnal call of the koel (‘kooeei or ko-el’, C&B).
However, another cry of the bird, rendered by Kupang informants as ‘yeep’ or
‘weep’, corresponds with vocalizations reported by C&B for the kite, Elanus
caeruleus

kui. A bird named after its call (Onvlee), but otherwise unidentified. Kapita de-
scribes it as having green plumage and feeding on ripe mangoes and other fruits. If
this description is accurate, a likely candidate is the glossy green Short-tailed
starling, Aplonis minor, which is reported as ‘moderately common’ on Sumba (C&B p.
468) and consumes fruit (see also Mason and Jarvis, 1989:47, regarding the papaya-
eating habit of Balinese Aplonis panayensis). (Contrary to Forth 1981:113, kui seems
not to be associated with witches. This idea apparently reflects confusion with kuu
[see above].)

. kulu-kadu. Saxicola caprata (D 9: Pratincola caprata), Pied bushchat (or Pied Chat,

C&B). Kadu, ‘horn’, alludes to the largely black plumage of the male. A term
mentioned by Kupang informants, manita watu (watu is stone, cf. BI batu), is
possibly a dialectal name for the bushchat (cf. BI kucita batu

kulu-kakuta. Oriolus chinensis broderipii (D 14: Oriolus broderipi), Black-naped oriole.
Meaning ‘resembling betel (kuta),’ kakuta seems to refer to the bird’s appearance,
although its plumage is predominantly yellow and blac

. kulu-kanuhu. Terpsiphone paradisi sumbaensis (D 16: Terpsiphone sumbaensis), Asian

paradise-flycatcher. Kanuhu, which refers to the horse colour called Isabella, may
describe the male bird’s predominantly off-white plumage. Onvlee, however, further

176

nN
os)

bh
to

~
wn

i)
SN

nN
XI

i)
Go

~)
s©

FORTH Vol. 20, No. 2

glosses the term as ‘rolled combed cotton’, which recalls the bird’s exceptionally long
tail feathers. A Rindi myth recounts how these were derived from strands of cotton.

. kulu-kawaki. A rail. Given by Onvlee as Rallina fasciata (Red-legged crake), but this

species has not been recorded on Sumba (C&B). In respect of its attributed vocaliza-
tions, the bird is possibly Gallirallus philippensis, the Buff-banded rail, or Amaurornis
phoenicurus, the White-breasted Waterhen. The second element of the name is
onomatopoeic (Kapita). Reputed to be especially vocal just before the onset of the
southwest monsoon, the bird is significant in Rindi as an index of seasonal change.

. kulu-ndiha. A sort of pigeon (Kapita: BI merpati, burung dara, both general terms

for ‘pigeon, dove’). The term is at least partly, and perhaps exclusively, applied to
domestic pigeons. Onvlee gives as western Sumbanese equivalents rawa and rawa
tana (see below). Dammerman (D 35) gives kuru ndasi (cf. Nage kolo dhasi, Domes-
tic pigeon, Forth 1996:106) for ‘Turtur tigrinus’, or ‘turtle dove’, referring apparently
to Streptopelia chinensis (tigrina is the name of the sub-species of S. chinensis found in
Wallacea; see mbara). The term kuru ndasi is not found in either Onvlee or Kapita.
The second element of kulu-ndiha possibly derives from ndiha in the sense of ‘good,
tive.’

attractiv:

. kulu-taitaku or kulu-kataitaku or manginu kulu-kataitaku. Recorded in Rindi. Not

listed by Onvlee or Kapita, though Onvlee gives manginu taitaku (see below).
Onvlee glosses taitaku as ‘to walk, run quickly’ (cf. Kapita who provides the same
gloss for kataitaku), and describes manginu taitaku as a bird that moves by jump-
ing. The most likely referent is Rhipidura rufifrons sumbensis, the Rufous fantail. This
is supported by Dammerman (1926a:210, 1926b:79), who gives the Laura (western
Sumbanese) name for ‘Rhipidura semicollaris’ (semicollaris is the name now given to
the sub-species of R. rufifrons that occurs on Flores) as kela kataga, ‘dancing bird’, a
term evidently cognate with kulu-kataitaku.

kutuku. Eudynamys cyanocephala, Australian koel (D 21: Eudynamis everetti). Accord-
ing to C&B, this is the only member of the genus occurring on Sumba. The Common
koel, E. scolopacea, is not listed by these authors, who however note that it may be
conspecific with E. cyanocephala (1997:353). A possible application of the term kuu to
the koel was noted above (see entry no. 17). Since the suffix -ku can indicate redupli-
cation, the name kutuku is conceivably explained as a reference to a bird that calls
‘kutu kutu’ (cf. tutuku, below).

. laleba nggangga. Kupang informants identified this bird as a drongo. It is most

probably the Wallacean drongo, Dicrurus densus. The name literally means ‘sister’s
child of the crow (nggangga).’ Accordingly, Rindi describe laleba nggangga as like a
crow only smaller. Onvlee (s.v. nggangga), citing Dammerman, lists the term as a
reference to a ‘small black bird’ which he identifies as Corvus enca. However, C. enca
does not occur on Sumba; nor in fact does Dammerman (1926a:208) link this spe-
cies—which can hardly be called ‘small’—with the Sumbanese name.

. landu witu. From illustrations, Kupang informants identified this as a species of

nightjar, probably Caprimulgus affinus. Their descriptions of the bird’s appearance
and habits also support this identification. The term is translateable as ‘sign of (from)
the long grass’ (see Onvlee, s.v. landu, which also means ‘crest’). The name does not
appear in either Kapita or Onvlee. According to Kupang informants, the distinctive
nocturnal cry of a nightjar, reproduced as ‘cheri-ki-ki-ki’, indicates the presence of a
thief, an idea that may illuminate the sense of ‘sign’ (landu) in the bird’s name.

. mabihi. Unidentified. An eastern Sumbanese name reported only by Kupang

informants, who gave quite various accounts of the bird’s appearance. The name is
not found in Onvlee or Kapita. The variant in the dialect of Mangili is mabahi.
manginu. Small birds, kinds of Munia (D 13; cf. Lonchura). The focus of the
Sumbanese category comprises several of the Estrildine finches, especially ones that

Winter 2000 JOURNAL OF ETHNOBIOLOGY 177

Ww
bd

ny
y

Se)
i)

35.

do damage to crops (see the Dutch name rijstdiefjes’ [Dammerman, Onvlee], ‘little
rice thieves’). C&B record six species on Sumba, including four munias (Lonchura
molucca, L. punctulata, L. quinticolor, L. pallida), the Red avadavat (Amandava
amandava), and the Zebra finch (Taeniopygia guttata). The names of other small birds
classified as manginu are qualified, as follows

manginu kadu. A black, horn-coloured manginu ‘which however does not eat rice’
(Onvlee). Very probably another name for kulu-kadu (see above).

manginu kani. Passer montanus, Tree sparrow (Kupang informant; kani, or uhu kani,
is the cereal Panicum viride). According to C& B, the species was first recorded on
Sumba in the 1940

manginu taitaku (see kulu-taitaku, above)

. manu. Gallus gallus, Domestic fow]; distinguished from manu tata (below) as manu

mopu, ati fowl.’ Several varieties are distinguished according to colour, size or
derivatio

. manu pati Gallus varius (D 37), Green junglefowl, sometimes simply called tata (cf.

Nage kat

mbaku. Haliaeetus leucogaster, White-bellied sea-eagle (D 31: Haliaetus occas
Probably further applied to other large eagles and hawks. Mbaku also

float, soar, glide, hover’ (Onvlee; see also luku mbaku, ‘mbaku river’, as ao name of
the Milky Way).

mbaku bara kuku. Described by Onvlee as an eagle with a white neck, smaller than
the sea-eagle (mbaku), and otherwise ‘orange’ in colour. This however describes the
Brahminy kite (see above: ikitu, ikitu marakuku. Marakuku is synonymous with
bara kuku).

mbaku tehiku. Given by Onvlee as ‘large sea-eagle’, the designation appears to be
merely a more elaborate name for Haliaeetus leucogaster. On the other hand, it could
conceivably include another large coastal raptor, the osprey (Pandion haliaeetus),
which is peer 2 in size between the larger sea-eagle and the Brahminy kite.
Tehiku is ’

. mbara. A de According to Kapita, my own field notes, and identifications

provided by Kupang informants, the term denotes Streptopelia chinensis, the Spotted
dove, and possibly also Geopelia maugei (but see rawa kakoruku, below). Following
Dammerman, Onvlee associates the name with Treron teysmannii (D 32: Osmotreron
teysmanni), the endemic Sumba green pigeon. However, partly in view of Onvlee’s
further identification of the scientific name with rawa ratu, this appears mistaken.
mbara manu. A larger kind of mbara, the size of a domestic fowl (manu; Onvlee s.v.
mbara). The term was also recorded in Rindi, where it probably specifies Streptopelia
chinensis.

mbara nggela In Rindi, a smaller kind of mbara, possibly Geopelia maugei. Onvlee
(s.v. ngguku, nggela) describes mbara nggela as a dove smaller than a similar kind
called ngguku nggela (see below). Both of the latter terms mean ‘to move up and
down, to nod’ and refer to the Columbiforme habit of bobbing the head. Kapita and
Onvlee further gloss ngguku as ‘to coo.’ In Sumbanese songs and narratives, the call
of the mbara is rendered as turu tu tu (Wielenga 1909).

mbera wurungu. Swallows, Hirundo spp. (D 17). The name translates as ‘broken
fragment of a pot’ (Onvlee, Kapita) and thus is evidently not related to hawurungu in
the sense of ‘to fly.’ It is a reasonable surmise that mbera wurungu is further applied
to Wood swallows (Artamus leucorynchus) and swifts (Apus spp.; see Appendix 1), but
this cannot be known from the available sources

nggaha. Domestic goose (recently introduced). From BI angsa (goose).

nggangga. Corvus macrorhynchos (D 8), Large-billed crow. Like its many cognates in
Austronesian languages, the name is locally recognized as onomatopoeic.

FORTH Vol. 20, No. 2

37. nggokaria. Herons, Ardea spp. (D 40). The term probably also refers to Egrets

vo
%

es)

rs
-

rs
ot

(Egretta spp.) and other members of the Ardeidae. Rindi informants stated there
were two kinds of nggokaria but did not know names for these. A Kupang informant
suggested that Australian pelicans (Pelecanus conspicillatus), occasional visitors to
Sumbanese inland lakes, might be classified as nggokaria. Probably relevant here is
their watery habitat, as well as the white and black plumage, shape of their heads
and—as the informant himself pointed out—large bills.

nggonggali. Rhyticeros everetti (D 20: Rhytidoceros everetti), Sumba hornbill. The
species is a Sumbanese endemic.
. ngguku, ngguku nggela. Apparently the same bird otherwise designated as mbara or
mbara nggela (see above). In Rindi, ngguku nggela, which refers to Columbiformes’
habit of head-bobbing, was a war cry uttered periodically by victorious warriors,
particularly it seems when returning to their village with enemy heads, and while
preparing the heads for ritual use (see Kruyt 1922:561). A Kupang informant from
the eastern Sumbanese domain of Lewa reported the name ngguku kulungu, which
he identified from illustrations in C&B as Geopelia maugei (see mbara).
. pipi. One or more kinds of sandpiper (Tringa spp.; D 41: Tringoides hypoleucus).
Dammerman’s reference particularly suggests Actitis hypoleucos, the Common
sandpiper, whose call C&B describe as ‘a piping tii-tee-tee’ (C&B). Onvlee and Kapita
give the equivalent term in the Mangili dialect of eastern Sumba as ahu ramuku,
‘pool dog’ (ramuku is ‘pool, pond, wallow’). Pipi may be synonymous or overlap
with kahuhu (see above).

pirihu. Trichoglossus haematodus (D 23), Rainbow lorikeet. The endemic Sumbanese
sub-species is T. h. fortis. The name bears some resemblance to one of its calls,
reproduced by C&B as ‘peaow, peaow, peow’, and so is possibly onomatopoeic.

powa. Coturnix spp., quails (Onvlee). Two species occur on Sumba, C. ypsilophora
pallidior and C. chinensis, the Brown quail and Blue-breasted (or Chinese) quail.
Whether the term also applies to buttonquails (Turnicidae), two species of which
(including one endemic) occur on Sumba, is not known.
. rawa, rawa kamukumu. Ducula aenea (D 33: Carpophaga aenea), Green imperial
pigeon. Dammerman’s description clearly fits D. aenea. The Sumbanese sub-species
is D. a. polia. As both Onvlee and Kapita note, the bird is also simply denoted by the
generic term rawa, thus indicating that the species is the focus of the Sumbanese
category. Kamukumu is onomatopoeic.

rawa kakoruku. Geopelia maugei (D 36), Barred dove (Onvlee). Kapita associates this
term with BI perkutut, which McKinnon (1991), referring to Jawa and Bali, identifies
as Geopelia striata. Rindi claim that if this bird calls at night it presages a death. The
second element of the name is onomatopoeic (cf. koru, ‘to coo (of a dove)’; ka- and -
ku are fused affixes).
. rawa kawi. A kind of pigeon, otherwise unidentified. The name was recorded only
in Rindi. Informants were unable to explain kawi, a word appearing in neither
Onvlee nor Kapita, though they described rawa kawi as intermediate in size
between rawa manu (see below)—the largest rawa—and rawa tana.
. rawa manu. A kind of large pigeon, about the size of a domestic fowl (manu). The
name was recorded in Rindi, where informants’ descriptions suggested it may be a
local designation for the bird otherwise known as rawa kamukumu (Ducula aenea).
rawa ratu. Treron teysmannii, Sumba green pigeon (following Onvlee, who cites D
32, probably in regard to the western Sumbanese name ‘rawa ratoe’; cf. mbara
above). The species is a Sumbanese endemic. Ratu refers to a high-ranking ceremo-
nial leader. In view of its use in similar contexts (see e.g. ularu ratu, the Reticulated
python; also kaka ratu under kaka above), the term probably alludes to the bird’s
colourful plumage.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 179

48. rawa tana. Chalcophaps indica (D 34), Emerald dove. In Rindi I was told that the bird
is so named because it remains on or close to the ground (tana, also ‘land, earth,
soil’) and never alights in trees. It is also described as silent or rarely vocal, unlike all
other rawa. (This notion may derive from the circumstance that C. indica usually
does not call when on the ground; see C&B 1997:316.) Owing to its reputedly quiet
nature and iridescent green plumage, in ritual language the bird’s name symbolizes
favourable qualities and spiritual beneficence.

49. rendi. Anas spp. (D 42), ducks. Wild ducks and introduced, domestic varieties are
distinguished respectively as rendi matamba and rendi mopu, ‘tame ducks.’ Accord-
ing to Onvlee, domestic varieties are further distinguished with reference to their
origins, as for example rendi jawa and rendi manila. The most common species of
wild duck, according to Dammerman, is Anas superciliosa, the Pacific black duck.
According to C&B, the only other species of Anatidae found regularly on Sumba is
Anas gibberifrons, the Sunda teal. However, a member of the Dendrocygnidae,
Dendrocygna arcuata, the Wandering whistling-duck, is also present and locally
common.

rendi manu. Recorded in Rindi. A kind of wild duck (manu, ‘domestic fowl’)

rendi yalangu. Recorded in Rindi. A kind of wild duck, smaller than rendi manu. Yalangu
is evidently a different word from yalangu, which Onvlee records as the name of a
kind of eucalyptus tree (Melaleuca leucadendron).

50. tariku. A kind of falcon or small kite (Onvlee, Kapita); otherwise unidentified. An
apparent synonym is taripu

51. totoru laka. Centropus bengalensis, Lesser coucal (D 22: Centropus javanicus). This is
the only coucal that occurs on Sumba. The first part of the name is probably ono-
matopoeic (cf. ‘totok, totok, totopuk, totopuk, totopuk’ C&B). Laka, denoting a red
colour and a tree, the leaves of which are used to produce a red nail varnish,
evidently refers to the bird’s rufous wings. Totoru laka does not appear in Kapita.

52. tutuku. The name of the coucal given by Kupang informants (cf. totoru laka), who
explained it as deriving from an imitation of its the bird’s call, rendered as ‘tutu
tutu.’ The fused suffix ~ku indicates repetition of the preceding element. Tutuku,
however, is not listed as the name of a bird by either Onvlee or Kapita. In view of the
relatively close relation and similarity of size, form, and colour between coucals and
koels (kutuku, see above), the phonological and morphological resemblance of the
names tutuku and kutuku is a point of some interest. Another is names for the Lesser
coucal in dialects of the Manggarai language of western Flores, where it is called
kotok or totok (Verheijen 1963).

53. wangi. Tyto alba sumbaensis, Barn owl (D 28: Dutch kerkuil, Strix flammea.) Another
Sumbanese species of Tyto is T. longimembris, the Eastern grass ow]. Dammerman
states that the bird is considered to manifest a punitive spirit, and that its ‘croaking’
or ‘cawing’ forebodes sickness or death. Rindi people associate wangi with witches
and particularly with the spiritual essence of a witch, called wandi. A connection

tween this term and wangi is however improbable (see Forth 1991:109 n6, 445). A
more likely cognate of the bird’s name is suangi, a word that in Malay and other
western Indonesian languages refers to a maleficent, nocturnal spirit (cf. eastern
Indonesian Malay suangi or suanggi, ‘witch’). A Kupang informant saat wangi
as designating all owls ‘that are seen’, thus implicitly contrasting the
katua wei (see above). Even if this is correct, however, the available slp still
seen to Tyto alba as the focus of wangi.

nner Geoffroyus geoffroyi floresianus (D 25: Geoffroyus floresianus), Red-cheeked

parro
55. Aen Asynonym of kaluki (see above)

180 FORTH Vol. 20, No. 2

56. yapi. Gallinula spp. (D 39: Gallinula frontata; Onvlee, incorrectly citing Dammerman,
gives ‘Gallinula phunicura’). Two species of Gallinula occur on Sumba, G. tenebrosa and
G. chloropus, the Dusky moorhen and Common moorhen. For Sumbanese the bird’s
most distinctive feature is the bright red bill, which appears in both species. Another
red-billed water bird that occurs on the island is Porphyrio porphyrio, the Purple
swamphen. From illustrations, Kupang informants identified both the latter and G.
tenebrosa as yapi. Onvlee describes the bird as inhabiting paddy-fields and doing
damage to rice and tubers.

ADDENDUM: BATS

pahomba. A — bat (possibly sant iste sp.), about the size of a swallow or large
butterfly. The creatures are said to roost in the tops of banana trunks where, Rindi claim,
they lay eggs about the size of a age s. Apparently referring to the same small bat,
Onvlee (s.v. pani=Kapita’s panii) lists the terms pani ru_kalu (‘banana leaf bat’) and pani
palinju wiki (‘bat that fouls itself’). Sometimes pahomba bats enter houses. In Rindi,
they are considered a manifestation of spirits associated with clan shrines, also called
pahomba. Being mystically powerful, the creatures are also able to assume human form.
panii (pani, Onvlee). Flying fox (Pteropus spp., following Dammerman). The term also
means ‘to talk, speak’, and probably alludes to chattering noises made by large fruit-
bats. Rindi people reported two kinds which they distinguished only as ‘large’ and
‘small’ (panii bokulu, panii kudu). These may correspond to the two species recorded by
Dammerman (1926b:22) as ‘Pteropus morio’ (P. alecto morio?), described as almost entirely
black, and ‘Pteropus gilvus’ (Dobsonia peronii?), which is yellow-brown in colour (see
Forth 1998a, regarding two kinds of Flying fox distinguished by the Nage of central
Flores). On the other hand, the reference may be to Flying foxes and much smaller bats
respectively, especially in view of indications that panii also serves as a general term for
bats (see pahomba above).

panyonga makaweda. Recorded in Rindi. A small bat. The name means ‘tricks, fools
elderly people.’

KEY:

Onvlee=Onvlee 1984, Kapita=Kapita 1982.

BI=Bahasa Indonesia, the Malay-based national language.
C&B=Coates and Bishop 1997.

D=Dammerman 1926a except where otherwise indicated.

Numerals after ‘D’ indicat identified | erman (1926a). Latin
names given by Dammerman, many of vahich 2 are now superseded, are not italicized. Where only a
number is given bates ‘D’, the scientific name provided by Dammerman coincides with the one
aco at prese

‘Kupang informant(s)’ refers to Sumbanese consulted in Kupang, the capital of Nusa Tenggara Timur
province, in 1999.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 181

SYMBOLIC ASSOCIATION AND METAPHORICAL PAIRING

Especially the composition of life-form taxa (named with reference to modes
of locomotion), and of apparent intermediate taxa, indicate that men Sumbanese
ethnozoological classification is primarily based on
criteria. At the same time, several categories of birds are linked in non-taxonomic
ways that may generally be called symbolic. Several ethnoornithological catego-
ries are paired in the parallelistic idiom of ritual language. For example, the
conjoined phrases pirihu pauli, kaka makanguhuru, ‘assembled parrots, collected
cockatoos,’ refers to any large group of people who unite to expedite a ritual or
some other customary undertaking. A number of pairings are listed in Table 2.
Together these include 13—thus nearly a quarter—of the categories appearing in
Table 1.

Although the list is meant to be illustrative rather than exhaustive, a glance at
Table 2 suggests that the three names associated with intermediate taxa (ikitu,
manginu, rawa) are particularly salient in ritual speech. It may be no coincidence,
moreover, that the most prominent of these, rawa, a term that pairs with at least
four others (including plant and insect categories), is the focus of one of the most
elaborate areas of eastern S The most
frequently named bird in ritual idioms is in fact the domestic fow] ees Manu,
however, is mostly coupled with wei, denoting the domestic pig, and the refer-
ence is then to the value of both as domesticates, sacrificial victims, and (with

TABLE 2.—Bird names included in ritual speech couplets

manu//rendi, domestic fowl/ /duck

manu/ /wei, domestic fowl/ /(domestic) pig
mbara//manginu, dove//munia (or ‘small seed-eating bird’)*
nggangga/ /ikitu, crow/ /hawk (or specifically, Brahminy kite)
nggauka/ /rawa, friarbird (=koka)//pigeon

pirihu/ /kaka, lorikeet/ /cockatoo

rawa//moha, pigeon//moha tree (unidentified)

rawa/ /ngginggi, pigeon/ /spider

rawa/ /ngguku, pigeon (Imperial pigeon) / /dove

rawa tana/ /ngginggi rara, Emerald dove//red spider
tariku/ /ikitu, falcon//hawk (or specifically, Brahminy kite)

*AS a general reference to ) crop pests, these two terms are : simply oh acre In the same context,
ea with the pal ‘small mice and ra

ale itt

182 FORTH Vol. 20, No. 2

regard to their intestines and livers respectively) as auguries revealing the will of
ancestral spirits. The pairing of rawa and nggauka—a ritual language substitute
for koka (friarbird)—evidently reflects the association of these two birds in cre-
ation mythology (see below), in which context the first term refers specifically to
the Imperial pigeon (rawa kamukumu).” By contrast, in one pair of phrases link-
ing rawa with ngginggi (spider), the former is specified as rawa tana, the name of
the Emerald dove (Chalcophaps indica), which is then more exactly paired with
ngginggi rara (red spider), denoting a particular non-poisonous arachnid. In yet
another pairing ngginggi is specifically linked with rawa muru, ‘green pigeon,’
which, though not a standard ethnoornithological fen probably also refers to
the Emerald dove. On the other hand, in th , pakameli
mohangu, ‘cooing like a pigeon, trembling like leaves of the moha tree’ (a refer-
ence to spirits of the dead, Onvlee s.v. rawa), the Columbiforme in question is
evidently the Imperial pigeon, since it is specified as making the sound kamu
(=kamukumu

Where Sumbanese ritual speech conjoins Hw bird categories, the names gen-
erally denote kinds th ork lly similar. For example,
Rindi explain the pairing of ‘duck’ and ‘domestic fow!’ (rendi and manu) with
reference to their characteristically large broods. In fact, the only obvious excep-
tion is the pair rawa and koka. Expressions featuring the couplet ikitu//tariku
associate the two Falconiformes with contrasting environments—land and sea, or
earth and sky. Yet the metaphorical import of these expressions turns less on the
opposing environmental associations than the identical predatory habits of the
two raptors. Indeed, since it is tariku rather than ikitu (a term most closely linked
with the Brahminy kite) which is linked with the sea, the environmental associa-
tions are contrary to ornithological fact. A similar complementarity of dry land
and water is also implicit in the ritual speech coupling of domestic fowls and ducks,
though in this case of course the environmental contrast is valid.

Most of the ethnoornithological terms included in Table 2 designate kinds
which are also prominent in Sumbanese myth. Among these are the Imperial pi-
geon (rawa, rawa kh A ) and the friarbird (koka), who dispute over the length
of day and night and the mortality of mankind (Forth 1992). Especially important
in creation mythology is the White-bellied sea-eagle (mbaku). In parallel form
named as i Mbongu i Mbaku (mbongu, ‘mist, dew;’ mbaku, ‘to float, glide, hover’),
the sea-eagle flaps his great wings causing the waters of the primeval flood to
subside and the dry land—specifically the island of Sumba—to appear. Also men-
tioned in myths of creation are the crow (nggangga), cockatoo (kaka), and fantail
(in Rindi called kulu-kataitaku or manginu kulu-taitaku), as well as a spider
(ngginggi), a non-ornithological taxon which, interestingly enough, also appears
in Table 2. The three bird kinds figure as well in an origin myth of the Rindi clan
Kanatangu, which further recounts how various species acquired their character-
istic plumage. Other birds appearing in this narrative include the oriole
(kulu-kakuta), Asian paradise-flycatcher (kulu-kanuhu), Eclectus parrot (kariku),
and Spotted dove (mbara). Several narratives recorded by Wielenga (1909)—mostly
fables featuring animal kinds—similarly feature three Columbiformes (mbara,
rawa, and a pigeon identified as ‘kuru ndiha,’ cf. kulu-ndiha), three parrots (kaka,
kariku, pirihu), the crow, the friarbird, the junglefow] (manu tata; see also Wielenga

Winter 2000 JOURNAL OF ETHNOBIOLOGY 183

1913), a heron (nggokaria), and ‘small birds’ (manginu). In the mythological genre
called analalu—or ‘orphan’ tales—a dove, specified as mbara, sometimes appears
as a messenger who, in a song, reveals the identity and relates the tribulations of
an orphaned hero.”4 From these several traditions, it can be seen how mythically
significant avifauna—including a large raptor, Columbiformes, and small birds
classified as manginu—coincide to a significant degree with kinds composing hy-
pothetical intermediate taxa, just as do the bird categories of parallelistic ritual
speech. Also noteworthy in this regard are the several parrots and bird kinds whose
names include the component kulu- (although, as demonstrated, kulu- does not
actually label a divided taxon, nor for that matter a class of any kind).

Another symbolic value attaching to several bird kinds concerns their signifi-
cance as omens. The augural value of the ow] called katua wei, whose cry is thought
to reveal the presence of wild pigs (see Table 1), of course pertains to hunting.
Vocalizations of the kingfisher (kahiku) heard near a settlement indicate to Rindi
that a thief is about. A similar significance is attributed to the distinctive nocturnal
cry of the nightjar (landu witu, see Table 1). The cries of two other birds are espe-
cially ominous because Sumbanese regard them as manifestations of witches.
Variously described as birds ‘ridden by’ witches or as physical forms taken by the
malevolent spirit (wandi) of a witch, these are the Large-billed crow (nggangga)
and the bird called kuu (provisionally identified as the Black-winged kite, see Table
1; Forth 1981:113; Kruyt 1922:559). If either of these birds calls near a house where
someone is ill, then it is a sign that the illness is caused by a witch, and that the
person will likely die. By the same token, a large flock of crows is a more generally
negative portent. The owl called wangi is also considered an embodiment of a
witch’s wandi (see Table 1 regarding the resemblance between these two terms),
an idea that possibly illuminates the common belief that if an owl alights on a
papaya tree, the tree will die.’ Because of their association with witches, one should
not verbally abuse or throw stones at crows, owls, or the kuu. Nevertheless, it is
permitted to kill a kuu if it makes too much noise after dark.

According to Kruyt (1922:558-9), the cries or behaviour of several birds were
formerly significant in the context of warfare and headhunting. An outgoing war
party would rejoice if they encountered crows (nggangga) cawing. However, if
they came across crows perched silently, this was considered an ill omen, and the
party should return home. It was similarly inauspicious if either a crow or a bird
of prey (ikitu) flew across the path of a group of warriors, as this would portend
casualties and defeat. On the other hand, if birds of either kind followed a war
party, then they could be confident of success. If a rawa—which from Kruyt’s
description appears to refer to an Imperial pigeon (rawa kamukumu)—was heard
calling in a village at night, this indicated that an enemy was preparing to attack.
Similarly, if a mbara (glossed by Kruyt as ‘wood pigeon’) called in the evening or
early in the morning—something the bird does not usually do—then people should
not to leave their homes the following day. Were they | to do so, they would suffer
some (unspecified) loss. Kruyt adds t f the mbara might
alternatively indicate the presence of thieves in the vicinity, as might the noctur-
nal screech of an ikitu (diurnal raptor). In Rindi I encountered a similar belief
concerning the Columbiforme identified as rawa kakoruku (see Table 1, nos. 43,
33, s.v. mbara), whose nocturnal cry can presage a death. Although the interpreta-

184 FORTH Vol. 20, No. 2

tion cannot be fully developed here, it seems there may be a more general connec-
tion between Col y ones designated as ngguku and nggela
(see Table 1, s.v. mbara, ngguku)—and headhunting. Apart from the fact that they
commonly fall victim to birds of prey, this could be motivated by their apparently
flexible necks, evidenced by their habit of bobbing their heads up and down, and
the relative disattachment of head and body which this might suggest.7°

Among birds whose cries or behaviour Sumbanese consider ominous, the oc-
currence of neither owls nor the kingfisher has any particular relevance for their
taxonomic status: both are simply called by terms occurring at the level of folk
genera. On the other hand, Columbiformes and diurnal raptors, thus birds associ-
ated with named intermediate taxa, figure quite prominently in this symbolic
domain, as they do in myth and ritual speech idioms. Evidently the most ominous
bird of all, the Large-billed crow (nggangga), moreover, participates in a hypo-
thetical intermediate category together with a similarly dark bird, the drongo
(laleba nggangga), which, being specified as its ‘sister’s child,’ shares part of its
name.

CONCLUSION

Despite the provisional nature of several ornithological identifications of
Sumbanese bird taxa given in Table 1, I have argued that the information available
here is sufficient to offer a general characterization of their folk classification. Within
the entire domain labelled by the term makayidi-yadaku (animals), the classifica-
tion isolates two named life-form categories— flying things’ and ‘crawling things.’
Members of these two categories are the most strongly contrasted in terms of
morphology, behaviour, habitat, and symbolism. Other animals, notably mam-
mals, then figure as a residue of this binary partition. They are, to be sure,
makayidi-yadaku, yet they occupy no separately named or otherwise clearly dis-
tinguished grouping within this inclusive class. Relevant here may be the
circumstance that there are relatively few mammals, especially large mammals,
on Sumba, while those that are present (deer, wild pigs, macaques, several ro-
dents, the palm civet, and a wid a) are quite various. The most numerous

species ar are classified as—or with— birds.’
In other folk systems, ‘fish’ sometimes appear as another named life-form taxon
(Brown 1979). But while Sumbanese possesses a general term for ‘fish’ (iyangu),
and while they have names for over one hundred kinds (Forth 1981:429, note 28),
Rindi classify these and other water creatures as instances of mabei (crawling or
creeping things).

Intervening between the life-form taxon mal g things) and the
numerous folk generic taxa, eastern Sumbanese bird chansiticating: reveals three or
more identifiable “intermediate” classes. Two, which partly at least are identified
with the names rawa and ikitu, respectively comprise Columbiformes and
Falconiformes. A third, labelled manginu, includes a variety of small passerine
birds that have as their focus several species of Estrildine finches. Remarkable in
this regard i is the overall resemblance between eastern Sumbanese bird classifica-

nand the eth of the Nage people of central Flores
(Forth 1996). Largely from evidence provided by free recall lists, I have previously

Winter 2000 JOURNAL OF ETHNOBIOLOGY 185

Pa a re eG |

shown how N. ral intermediate taxa. Most
prominent among these, both in terms of the number of named folk generics they
include and the priority usually given to them in free recall, are mostly covert
categories coinciding with the scientific groupings labelled ‘Falconiformes’ and
‘Columbiformes.’ In addition, the Nage category ana peti closely parallels
Sumbanese manginu, being especially associated with the genus Lonchura and re-
lated birds of very small size but contextually including many other small kinds
besides, even to the extent that ana peti is sometimes used as a general term for
‘bird.’ (Nage, it should be noted, have no term exactly corresponding to Sumbanese
mahawurungu which unequivocally includes all birds). Another similarity with
Nage concerns the relatively low number of onomatopoeic terms in Sumbanese
bird nomenclature, which even including unconfirmed instances amount to no
more than 25 per cent of names. For Nage the figure is approximately 30 per cent,
which also appears quite low in comparison with some other languages (Forth
1996:103).

Sumbanese treatment of Columbiformes and Falconiformes, especially, bears
on another general issue of ethnoornithological classification. Not only are these
two groups taxonomically salient; they are equally prominent in the symbolic
genres of parallelistic ritual speech, myth, and bird augury. With regard to ap-
proaches Berlin (1992:143,149) calls utilitarian or functionalist, as well as his own
observation that intermediate taxa may sometimes be grounded in “cultural” as
opposed to morphological or behavioural factors, one needs to consider whether
these two facts may be related. Relations between symbolism and taxonomy can-
not be treated conclusively here, not least of all because it has not been possible to
review all data relevant to the Sumbanese valorization of birds. In closing, how-
ever, I would register several points.

First, even where named by terms further associated with hypothetical inter-
mediate taxa (e.g., rawa, ikitu), most—perhaps all—symbolically significant birds
are actually identifiable with folk generics (e.g., rawa tana, ikitu marakuku), nearly
all of which appear to correspond to single scientific species. It is therefore the
most basic categories rather than more inclusive ones that possess distinctive sym-
bolic value; hence it cannot be claimed that this sort of value crucially informs
intermediate or higher groupings. The use of general names in symbolic contexts
even where more specific designations are available might be ascribed to
performative requirements of parallelistic language and traditional narrative, both
of which favour single lexemes. A similar consideration could apply to standard
admonitions regarding omen birds. That is, where a particular member of a di-
vided class has a negative association, it might be thought prudent to extend this
to all, empirically similar, members of the same class. It is just conceivable that
distinctions may be made within a class in order to deny an omen contextually or
to restrict its range absolutely (cf. Bulmer 1968:637-8, regarding a similar interpre-
tation of New Guinean transformation beliefs). This, however, is merely a
hypothetical possibility, and there is no obvious indication that such symbolic
motivation of class division has been operative in eastern Sumbanese.

A second consideration is that named intermediates are not only mythically,
metaphorically, and augurally significant; they possess more pragmatic—func-
tional or utilitarian—kinds of value as well. This should not be surprising, as both

. ere |
J dit TpPiicr uly

186 FORTH Vol. 20, No. 2

symbolic and utilitarian values attaching to avifauna typically relate, more or less
directly, to perceptible physical properties, including the apr
of particular birds. With one exception, all diurnal raptors (ikitu) thus prey on
domestic fowls. Columbiformes, both rawa and mbara, are also common victims
of raptors, while at the same time, among wild birds, pigeons and doves are espe-
cially valued as food, and so fall prey to human hunters as well. Mbara, or smaller
doves, further draw practical attention as crop pests, as do many small birds clas-
sified, at the folk generic level, as manginu (a label further applying to another
hypothetical intermediate taxon).”” It comes as no surprise, therefore, that ikitu,
rawa, and mbara are categories of Sumbanese bird augury, especially in the con-
text of warfare, and thus in relation to human competition and violent aggression.

As the foregoing should suggest, while various sorts of interest in birds may
affect the attention given to particular kinds in folk taxonomy, distinctions mani-
fest in Sumbanese symbolic usage, like those pertaining to utilitarian concerns,
are consistent with observable morphological and behavioural differences exist-
ing independently of those interests. Where natural kinds differ from or resemble
one another in regard to symbolic value or practical utility as well as empirical
features, therefore, one cannot readily argue that one of the three sorts of factors is
more determinant of their linguistic and cultural recognition than the others. In-
deed, it is probable that the three interact in complex ways. Nevertheless, in view
of their relative ontological independence, morphological and behavioural con-
siderations are, in the long run, likely to prove the most important. Supporting
this in the present case is the fact that the majority of Sumbanese birds are not of
any practical value, nor are they symbolically significant.

My final point concerns the very terms of the contrast implicit in the issue of
symbolism. Contrary to what is often implied in debates over the relative impor-
tance of perceptual versus non-perceptual factors in ethnobiological classification,
the symbolism of natural kinds is not necessarily more culturally specific or con-
tingent than is the taxonomic recognition (or mental representation) of salient
natural features.”® This is indicated, for example, by the fact that ‘hawks’ and ‘doves’
form a metaphorical contrast in English-speaking cultures as well as in eastern
Sumba, and by the extraordinarily widespread conception of owls and crows as
birds of ill omen. Again, this undoubtedly stems from objective physical features
of the kinds in question. Yet the fact that their metaphorical value (if not their
specific interpretations) is cross-cultural attests as much to universal properties of
human cognition as it does to the universal availability of physical avifaunal fea-
tures for symbolic deployment. These observations tend to suggest that “symbolic
value” and “psychological salience” are not as distinct or separate as is often sup-
posed. For this reason alone, the question of whether bird symbolism exercises a
decisive influence on aspects of ethnotaxonomy, in eastern Sumba or in general,
must remain moot.

Ol DEllIdvigUl

NOTES

1 Hereafter designated as ‘Kupang informants,’ these included Ibu Djukatana, who is a
daughter of Oemboe Hina Kapita; her husband, Bapak Drs. Ng. Djukatana (Oemboe Juka
Tana); Bapak Minggus Osa and Bapak Thomas Tagudodu, both associated with Nusa
Cendana University; Bapak Petrus Yiwa, an Instructor in the Law Faculty at Artha Wacana

Winter 2000 JOURNAL OF ETHNOBIOLOGY 187

University; and several eastern Sumbanese undergraduate students of the latter univer-
sity. am most grateful to all of the foregoing for their help, and to Dr. Tom Therik, the
Rector of Artha Wacana, who sponsored by visit and kindly assisted me in a variety of
ways while in Kupang.
2Sumbanes t ibed ding to th hy used by Kapita and Onvlee.
Single ietees ae represent sounds similar to their English denotata. The /a/ (see
ananjaki, Table 1) variously denotes the schwa or a sound like the /u/ in ‘duck.’ The /w/
is otter closer to English ep v/, especialy in medial positions (see rawa). The /b/ and /d/
y contrasting with /mb/ (see mbaku) and /nd/ (see kulu-ndiha).
I follow Kapita i in employing ii and uu to indicate vowel lengthening (see panii and kuu),
whereas Onvlee places an acute accent above the vowel in question.

3 From lists provided by Verheijen (1963), dialects of the Manggarai language of western
Flores would each appear to possess perhaps as many as 100 names for birds, including a
fairly large number of productive binomials. The area however is relatively rich in avi-
fauna; Verheijen’s estimate is 180 species, which he claims probably coincide with all those
found on Flores island (1963:678). A count of species described in Coates and Bishop (1997)
yields a total of 232 for Flores. For the Tobelo of the large Moluccan island of Halmahera,
P.M. Taylor (1990) similarly records 111 bird categories, which also include a number of
productive binomials.

4 Reporting on his 1925 visit, Dammerman (1926b:24) remarked how cockatoos were al-
ready extinct in the vicinity of Waingapu, eastern Sumba’s main port, owing to the large
numbers caught for export to Java.

5In his Dutch-Kambera wordlist, Onvlee (s.v. dier) refers to bohu, a word meaning ‘thiev-
ish, voracious,’ and more specifically to mabohu, which he translates as ‘wild animals.’ As
I know partly from my own experience in Rindi, however, the phrase refers not to wild
animals in general but more specifically to relatively large and dangerous animals. Ac-
cordingly, Onvlee exemplifies his initial gloss with ‘wild pig, snake, crocodile,’ while Kapita
translates bohu with Bahasa Indonesia buas, a word that means ‘cruel, savage,’ and ‘wild’

nly in this restricted sense. Analyzable as ‘what is wild (savage, cruel)’, mabohu thus
does not of itself necessarily denote animals. For example, tau bohu denotes a (human)
thief.

6 According to Onvlee, yada more specifically refers to a quick movement. He further glosses
the word as ‘to teem, swarm’ and ‘to wriggle, fidget.’ Similarly, he describes yidiku as
denoting a movement slower than yidi.

7 An evident borrowing from Malay, mburungu occurs in compound expressions where it
refers specifically to bird figures found, for example, on European coins (Onvlee 1984).

8 Worth noting here is the form wururu, which also refers to the sound of bird’s wings or
something being thrown, but further denotes a cry uttered at the beginning, or between
segments, of ritual song, including mortuary song (cf. Onvlee 1984; Kapita 1982). When
performed, the element ru is repeated not just twice but numerous times, thus effecting a
sort of whirring sound. Whether this is meant to replicate the sound of a bird in flight, or
anything else in particular, I was unable to establish from questioning.

188 FORTH Vol. 20, No. 2

9 Illustrating uses of makayidi-yadaku, Onvlee (s.v. yidi) records the phrase makayjidi-
yadaku la wai, “everything that teems in the water”. While evidently including fish, and
thus in a sense cross-cutting mabei, the expression differs from the two life-form terms by
its inclusion of the term for ‘animal(s).’

10 Fernandez (1996:146) lists the Proto-Flores term for ‘bird’ as *kolon.

11 The use of manu, ‘domestic fowl,’ to make distinctions within named ethnobiological
categories appears to be quite widespread in eastern Indonesian languages, and not only
to apply to birds. For example, among the Nage of central Flores, smaller varieties of the
Green tree viper and the Monitor lizard are respectively called hiku manu and ghoa manu
(Forth 1995:53, 66 n.4). On Sumba and elsewhere manu further appears in the names of
plants, but this is mostly explained by resemblance to parts of domestic fowls. For ex-
ample, pelu manu, ‘wattle (of a fowl),’ and tara manu, ‘cockspur,’ designate respectively a
kind of wild grape and a thorny plant.

12Two other raptors listed in Table 1, the kola and kuu, were not mentioned in this context,
but were not explicitly excluded either.

13 Kupang informants described kola, which was not included in this series, as signifi-
cantly larger than kapaha and tariku and “almost as large as mbaku.” (Although Onvlee
and Kapita describe the kola as relatively small, their wording suggests that the compari-
son is with the Brahminy kite or another large hawk.) Also missing from the Rindi com-
parison is the kuu, which Kupang sources described as the smallest diurnal raptor.

14 Coates and Bishop (1997:247) describe Haliastur indus as a scavenger, and with regard to
diet mention only “carrion, insects, fish, etc.” MacKinnon (1991:84) says that it feeds on
“almost any animal material, dead or alive,” and that it “catches small animals, steals larger
prey and scavenges along waterways...delicately picking up floating debris.” Referring to
Bali, Mason and Jarvis (1989:45) note that the bird has a “reputation as a stealer of chick-
ens,” but otherwise characterize it as a “general scavenger.” As regards Haliaeetus leucogaster,
the same authors describe the sea-eagle as snatching fish and sea snakes from the water,
occasionally feeding on carrion, and sometimes hunting fruit-bats.

15 Major settlements in eastern Sumba are mostly located in proximity to rivers, and most
are within several kilometres of the coast.

16 Speaking their own language, eastern Sumbanese use banjaru, ‘row, line, group,’ to ex-
press the idea of ‘(natural) kind.’ Ngia, ‘place, position,’ subserves the same function in
some contexts. Thus, hangia hangia (ha-, ‘one, each’) means ‘all kinds (of things)’ (see
further Onvlee, s.v. ngia).

7 Also worth noting is Wielenga’s (1917:33) comparative listing of mango, a cognate of
manginu, as the term for ‘bird’ in the western Sumbanese language of Lamboya. That
manginu might include more than passerine birds was suggested by a Kupang informant,
a young man from Lewa, who identified kahuhu (sandpipers) as an instance of the cat-
egory.

18TIn the report of his expedition, Dammerman (1926b) describes a honeyeater (Myzomela
sp., p-79), a fantail (p. 79), a whistler (p. 24), and the yellow wagtail (p. 36). Although he

Winter 2000 JOURNAL OF ETHNOBIOLOGY 189

observed the honeyeater and fantail in eastern Sumba, in his shorter article (1926a)
Dammerman does not list eastern Sumbanese names for these species, only a western
Sumbanese name for the fantail and another for a sunbird (Cinnyris buettikoferi, apparently
Nectarinia buettikoferi). The whistler (identified as ‘Pachycephala fulviventris’) is evidently
Pachycephala pectoralis, the Common golden whistler, the only whistler recorded on Sumba.
Dammerman describes the bird as very common.

1° Nggonggali probably reflects of Proto-Austronesian *’enggang (the initial /e/ represents
e schwa; cf. Bahasa Indonesia enggang), ‘hornbill’ (Dempwolff 1938:49; cf. Wurm and
Wilson 1975:104). Interpretations of nggokaria (heron) were mentioned earlier.

20 None of categories in Table 1 apply exclusively to seabirds or species found only on the
coast. From experience in Flores, I would guess that cormorants and grebes can be classi-
fied with ducks as rendi. The classification of birds like plovers and pratincoles (see Ap-
pendix 1) with sandpipers (kahuhu, pipi) would also not be surprising.

21 In 1975-76 I sometimes observed flocks of bee-eaters (Merops_spp.) when travelling be-
tween Rindi and the port town of Waingapu, but unfortunately never in the company of
informants. A Sumbanese man in Kupang gave kahi as a term for bee-eaters, which he
identified from illustrations and descriptions. But this name is probably not distinct from
kahiku, a reference to kingfishers, which eaters in appearance, diet, and nesting
behaviour. Among the Nage of Flores, who do have a special term for bee- eaters, speci-
mens viewed at a distance, flying high overhead, are sometimes identified with the term
for swallows and swifts (Forth 1996:92). Bee-eaters and the dollarbird are both non-breed-
ing migrants on Sumba.

2 The pairing occurs in the expression ka, hibu rawa, ‘nest of the friarbird, nest
of the pigeon,’ a euphemism for the male genitalia (Onvlee, s.v. nggauka). If the specific
referent is the male scrotum and testicles (tilu, also meaning ‘egg’), then evidently the nest
of any bird could serve as me — vehicle. OWNER, ane fact that these two kinds
in particular are selected can be ascribed to t of koka and rawa, as
the two birds that contested 6 over such matters as whether or not humans should die and
give birth, in order to replace themselves (see Forth 1992). Their association with repro-
ductive organs in this ritual speech context is thus intelligible with reference to their asso-
ciation in myth.

23 Several popular similes feature yet other birds. Onvlee records the following: paana
kalukingu, ‘to have (raise) children like a kaluki;’ pari kalukingu, ‘as strong as a kaluki;’
paren ngandu yapingn, ‘to have 2 mouth as red as (the bill of) a yapi’ (said of youngsters
‘to cluster, huddle together like quails.’
The first expression alludes to the fact that Megapodes incubate their eggs under a large
pile of debris, and therefore appear not to look after their offspring to the same extent as
other birds.

24 More often, songs of this sort are performed b haracters themselves. In one
case (C. Forth 1982: 93-92), the dove cara takes the place of the hero, assuming his
identity in the lyric. In another myth, the human hero is born in the form of a dove (mbara;
Wielenga 1909).

190 FORTH Vol. 20, No. 2

25 This idea, recorded in Rindi, is also mentioned by Kruyt (1922:559-60). Kruyt describes
the ow] as the “personification of the witch” and as the Sumbanese “bird of sickness an
death.” An owl alighting on the roof of a house sreuegenate corte death for the —— =
the Kambera region, when its nocturnal cry is heard,

be placed on their side. One should also not comment on or respond to the sound, for this
could result in death, construed as an indication that the person had “answered the call.”

26 Kruyt (1922:559) says that owls were not —— Significant in times of war. How-
ever, he also notes that if warriors he hing they should go under a tree and
make an offering of betel and areca to sti le “the evil associated with this bird” from
following them.

27 The lumping of numerous small birds in the taxon labelled manginu might at first ap-
pear to be an instance where utilitarian concerns have a preponderant influence on the
emergence of an intermediate taxon. This is contradicted, however, by the inclusion in
manginu of ornithological kinds that are known not to infest crops, but which especially in
regard to size and form d ble the crop pests. Also, of course, by no means all avifau-
nal destroyers of crops are classified as manginu.

28 A similar argument could be made in regard to the opposition of empirical (or intellectu-
alist) and utilitarian values. A more comprehensive, cognate question is whether symbol-
ism is to be understood as an intellectual activity or product or, as Malinowski argued, a
kind of utilitarianism.

ACKNOWLEDGEMENTS

The author’s research in Sumba was conducted for two years in 1975-76 and was
sponsored by the Indonesian Institute of Sciences (LIPI) and Nusa Cendana University in
upang. The research was funded from a grant awarded by the former Social Science
Research Council of the United Kingdom to Professor Rodney Needham. My visit to Kupang
in 1999 was funded from a grant awarded by the Social Sciences and Humanities Research
bo of Canada. I am grateful to all these bodies for their support, as I am to Professor
ham.

LITERATURE CITED

BERLIN, BRENT. 1992. Ethnobiological | COATES, BRIANJ.and K. DAVID BISHOP.
Classification Prin of 1997. A Guide to the Birds of Wallacea:
Categorization of Plants and Animals in Sulawesi, the Moluccas and Lesser
Tradi al ieties. Princeton Sunda Islands, Indonesia. Do
University Press, Princeton. Publications, Alderley, Australia.

BLUST, ROBERT. 1979. Proto-Western DAMMERMAN, K.W. 1926a.
Malayo-Polynesian Masini Bijdragen Soembaneesche dieren- en
tot de Taal-, Land- en Volkenkunde

plantennammen. Tijdschrift voor
135(2-3):205-51.

BROWN, C.H. 1979. Folk zoological life-
forms: Their universality and growth.
American Anthropologist 81:791-817.

BULMER, R.N.H. 1968. Worms that croak
and other mysteries of Kalam natural
history. Mankind 6(12):621-39.

Indische Taal-, Land- en Volkenkunde.

66:205-239.

. 1926b. Een cre naar Soemba.
uygrok & Co., Bat.

DEMPWOLFF, OTTO. 1938. Vergleichende
Lautlehre des Austronesischen
Wortschatzes. Dritt Band:
Austronesisches Worterverzeichnis.
Dietrich Reimer, Berlin

Winter 2000

ELLEN, ROY. 1993a. The Cultural Relations
of Classification: An Analysis of Nuaulu
Animal Categories from Central Seram.
Cambridge University Press,
Cambridge

.1993b. Nuaulu Eoorodiogy, A
Systematic Inventory. CSAC
Monographs 6. Cae for Social
Anthropology and Computing,
University of Kent, Canterbury.

FERNANDEZ, INYO YOS. 1996. Relasi
Historis Kekerabatan Bahasa Flores:
Kajian Linguistik Historis Komparatif
Terhadap Sembilan Bahasa di Flores.
Penerbit Nusa Indah, Ende, Indonesia.

FORTH, CHRISTINE. E. sin An analysis
of traditional narrative in eastern
Sumba. Ph.D. dissertation, Uninenity of

ford.
FORTH, GREGORY. 1981. Rindi: An
Ethnographic Study of a Traditional
main in Eastern Sumba.
ie acorenpea te van het Koninklijk
Instituut vo Taal-, Land- en
Volkenkunde No. 93. Martinus Nijhoff,
e ree
1992. The pigeon and the
friarbird: The mythical origin of death
and daylight in eastern Indonesia.
Anthropos 87:423-41.
ae Ethnozoological
classification and _ classificatory
anguage among the Nage of eastern
Indonesia. Journal of Ethnobiology
15(1):45-69.

. 1996. Nage birds: Issues in the
analysis of ethnoornithological
classification. Anthropos 91:89-109

98a. On deer and dolphins:
Nage ideas regarding animal
transformation. Oceania 68(4):271-293.
—_______. 1998b. Things that go ‘po’ in the
night: The classification of birds, sounds
and spirits among the Nage of eastern
Indonesia. Journal of Ethnobiology
18(2):189-209.
1999. sae Saison oie on
Nage bird classificatio and
ethnoornithology. pe easy 94:568-

FREEMAN, J. D. 1960. Iban augury. Pp. 73-
98 in The birds of Borneo, B. Smythies
— ep and the Edinburgh.

KAPITA, . 1982. Kamus Sumba/
idea . Percetakan
Arnoldus, Ende, papecenny

JOURNAL OF ETHNOBIOLOGY 191

KRUYT, A.C. 1922. De Soembaneezen.
Bijdragen tot de Taal-, Land- en
Volkenkunde 78:466-608.

LAUBSCHER, MATTHIAS SAMUEL. 1975.

Gottesnamen in_ indonesischen,
vorzugwei ostindonesischen
popegr acreage pe 209-29 in Der

n Stietencron

(editor). Me ia ok Patinos Verap.
LOPEZ, Studies on
Dempwolff’s Vergleichende Lautlehre
des Austronesischen Wortschatzes.
Summer Institute of Sanne
Philippines. Mimeographed (cited in

Ae bie) 8
, VICTOR and FRANK JARVIS.
1989. Birds of Bali. Periplus hie
Berkeley, California and Singa
MACKINNON, JOHN. 1991. Field Guide
to the Birds of Java and Bali. Gadjah
Mada University Press, Yogyakarta.
METCALF, PETER. 1976. Birds and deities
in Borneo. Bijdragen tot de Taal-, Land-
en Volkenkunde 132:96-123.
MONK, KATHRYN, YANCE DE FRETES,
and GAYATRI REKSODIHARJO-
LILLEY. 1997. The Ecology of Nusa
Tenggara and Maluku. The Ecology of
Indonesia Series Volume V. Periplus

Soembaas)- -Nederlands Woordenboek.
Foris Publications Holland, Dordrecht.
TAYLOR, PAUL MICHAEL. 1990. The folk
biology of the Tobelo people.
Smithsonian Contributions to
Anthropology Number 34. Smithsonian

Institution Press, Washington, D.C.
VERHEIJJEN, J.A.J. 1963. Bird Names in
rai, Flores, Indonesia.

-718.

WIELENGA, D.K. 1909. Schets van een
Soembaneesce spraakkunst (naar ‘t
dialect van Kambera). rina hiaagial
Genootschap' van tisk
ree Bata

CIA 3. Soem’ caneeselie verhalen in
' sets van Kambera, met vertalingen

- 1917. _ Vergelijkende woordenlijst

Saeubs en eenige Soembaneesche
spreekwijzen. Verhandelingen van het
Bataviaasch Genootschap van Kunsten
en Wetenschappen 61(5):1-96.

192 FORTH

pleonps es Pacific Linguistics Series

Department of Linguistics,
sat pier of Pacific Studies, The
Australian National University,
Canberra.

Vol. 20, No. 2

Winter 2000 JOURNAL OF ETHNOBIOLOGY 193

The Cultural Relations of Classification, an Analysis of Nualu Animal Catego-
ries from Central Seram. Roy Ellen. Cambridge University Press, Cambridge.
1993. Pp. 315. $80.00 (cloth). ISBN 0-521-4311-4.

In The Cultural Relations of Classification, Roy Ellen has compiled an impressive
collection of ethnozoological data complete with taxonomic charts, reviews of past
theoretical approaches, appendices with local glosses and ecological zones, and
his own definition of cognition as it relates to culture and classification. Although
written in 1993, the book is still a relevant and important resource today with a
resurgence of interest in the fields of ethnobiology and ethnoecology. The data for
the book are drawn from a series of field visits to the Nualu of south central Seram,
an island in eastern Indonesia. It is mteresting to note that although classic
ethnobiology has been generally perceived as “scientific,” Bien describes his i im-
pression of ethnobiology in the late 1970s dal, linking
with cognitive processes. He states these processes are ‘intriguing to him because
they link “socio-linguistic positivism with interpretive post-modernism” (1993:2).
Given this description it is easily ani sotOod how Ellen aivespes from classic
ethnobiological approaches. He focuses on t
with fluid boundaries, overlapping classes and fuzziness, while many
ethnobiologists would most likely be appalled at any discussion of the post-mod-
ern in relation to ethnobiology.

Throughout this work, Ellen both heeds the accomplishments of his predeces-
sors and contemporaries in ethnobiology while he simultaneously criticizes the
methodological approaches and analysis of some of those same colleagues, par-
ticularly Berlin, Raven, and Bisediove (Ellen 1993, see also Berlin 1992).

Ellen’s primary tl tical concern has been with classification as “situationally
adapted and dynamic devices of practical importance to their users, reflecting an
interaction...between culture, psychology and discontinuities in the concrete world;
a lexical and semantic field firmly embedded in a wider context of beliefs and
social practices” (1993:3).

Two areas where Ellen diverges from traditional ethnobiological approaches
are: 1) methods of data elicitation and 2) the interpretation of the language of clas-
sification. The etic versus the emic view has been the subject of much debate in
anthropology. Many believe that researchers tend to view cultures from an etic
perspective even if they are skilled ethnographers. Ellen contends that the use of
classic ethnobiological systematic methods, without unstructured ethnographic
interviewing and observation may lead to faulty conclusions regarding classifica-
tion and taxonomies. Furthermore, he believes that a predisposed belief in
hierarchical structures such as the Linnaean system biases some ethnobiologists
to assume that the societies they study do indeed possess a hierarchical system of
classification. Ellen’s discussion of the language of classification attempts to pro-
vide the reader with background on etl hes of the past and at
the same time critiques these approaches in comparison with his own. One prob-
lem with this critique is that his use of terms is often an inaccurate representation
of the works he cites. Ellen’s main point in offering this background for the reader
is both to acknowledge the vast body of ethnobiological research that has pre-

194 BOOK REVIEWS Vol. 20, No. 2

ceded The Cultural Relations of Classification, and at the same time provide the
groundwork for his approach to ethnozoology and classification of the natural
world.

Ellen provides several specific cases that are intended to illustrate the biased
methods he finds problematic. Here, I provide a brief description and counter-
explanation of a few cases. Ellen gives an example (assuming he is replicating a
biased classic ethnobiological approach) of the way in which the question “what
is nakatua wekae (red-sided eclectus parrot) a kind of?” encodes an answer because
the answer is nakatua. (1993:25). What Ellen fails to acknowledge is that in a sys-
tematic ethnobiological study, the question may not be asked in that way, because
a specimen could be pointed to without using the name nakatua at all (Brent Ber-
lin, personal communication 1995, Berlin 1992). Another way of asking the question
would be, “Does nakatua have any relatives?” Moreover, Ellen provides another
example of what he considers a culturally inappropriate elicitation method with
the question (again, assuming a hierarchically biased classic ethnobiological ap-
proach), “’What is asu (dog, Canis familiaris) a kind of?’” (Ellen 1993:25). He points
out that this is culturally inappropriate because a dog is not a kind of anything,
except maybe an animal. At this point the reader is lacking information about
whether or not there is more than one type of dog on the island of Seram. Later
Ellen describes dog, asu, as one of Berlin’s unaffiliated generics, whereas in Berlin’s
more recent approach, if there is only one type of dog, then the dog may be con-
sidered by ethnobiologists to be a monotypic genera, that is a generic class with
only one member (Berlin 1992:33).

Other terms that are confused with current uses in ethnobiology are produc-
tive, optional uninomial and binomial. It is entirely possible that Ellen was unaware
of, or lacked access to, other ethnobiological research that occurred simultaneous
to the time of his writing. Regardless of the reason for this divergence of terminol-
ogy, it is important that the reader is aware that Ellen often provides misleading
secondary explanations of other ethnobiological researchers’ usages of classifica-
tion language. Another possible reason for differences in perspectives and
approaches is that Ellen’s work is ethnozoological, whereas much of the work of
ethnobiology has been concerned with ethnobotany.

Ellen’s attempt to link the ideational and operative is indeed different from
the ethnobiological studies with which he compares his work. Ellen (1993), Hunn
(1985), Rappaport (1979), Nazarea-Sandoval (1995), and Bellon (1995) all attempt
to link the cognitive processes with behavior and decision-making. Different ob-
jectives seem to have motivated the various research approaches to which Ellen
refers.

Early pioneers in the fields of ethnoecology, Frake (1962) and Conklin (1969),
were interested in both classification and behavior through a cultural relativist
approach. Some later ethnobiologists were inspired by their work, but pursued
another direction, in search of universal compartments in the human mind. With
the search for a universal understanding of how humans order their natural world,
the debate between utilitarian and intellectual basis for classification of plants and
animals was spurred in the late 1960s and early 1970s. Ellen’s perspective clearly
leans toward a utilitarian approach (Hunn 1985), but rather than rely solely on

Winter 2000 JOURNAL OF ETHNOBIOLOGY 195

usefulness, and/or cultural salience in considering the utilitarian categories, Ellen
adds the dimensions of habitat, behavior, sound, smell, religion, and cultural con-
text to a traditional cognitive categorization primarily based on perception of
morphological qualities (and occasionally sound, as in the case of birds (Berlin
1995)). This tendency toward a more holistic ethnoecology is a valuable advance
in the field. I find Ellen’s attempt to include habitat and biotopes in his analysis
especially useful when considering the field of ethnoecology in relation to deci-
sion-making regarding natural resource use.

Ellen seems to follow Rappaport’s (1979) earlier attempts to include an analy-
sis of ideational and operational basis for classification, a multilevel decision
making process. In doing so he develops the theory of “prehension” which stresses
the situational bias of classification. He believes that the kinds of cognitive pro-
cesses that he has proposed are apparent in the “social construction of categories.”
It is this combination of approaches in which I find both the strength and weak-
ness of Ellen’s viewpoint. While the inclusion of cultural basis and a broader range
of contributing factors in classification scene: of Asha lence bomction liberates "
from a rigid ethnobiological approach, it also d |
that examines the way the human brain fund: It is clear that Ellen believes in
a heavy influence of the sociocultural construction of cognitive processes. Ellen
acknowledges some type of hierarchical ordering is necessary in the human stor-
age of knowledge, but he proposes a system of overlapping levels of hierarchy,
and clearly states that some forms of classification among the Nualu are varied,
therefore they fit more than one level at a time.

I find the book to be a useful resource for ethnoecology, especially because
Ellen has explored territories with limited previous research, but I propose cau-
tion regarding comparisons of his work with other ethnobiological works because
of the differences in terminology and Ellen’s sometimes misrepresented re-expla-
nation of certain terminology and concepts. I also noted that Ellen’s references
include few sources of publications from the 10 to 15 years prior to publication of
the book; he seems to be responding to and drawing mostly from works of the
1970s and early 1980s.

One should consider the purpose of Ellen’s work along with Bellon (1995),
Dove (1993), Frechi et. al (1989), Gragson and Blount (1999), Posey et al. (1984),
Nazarea-Sandoval (1995), Nazarea (1999) and, Warren, et. al (1995), to name only
a few, as examples of an integrative approach to understanding the cultural basis
for classification and its implications for resource use. In these contexts, The Cul-
tural Relations of Classification provides a functional reference for those interested
in ethnoecology and ethnobiology, but readers should refer to other current re-
search in cognition and ethnobiology (see Atran 1999) if they are more concerned
with cognitive taxonomic classification.

196 BOOK REVIEWS

Vol. 20, No. 2

LITERATURE CITED

, SCOTT and DOUGLAS L. MEDIN
(editors). 1999. Folkbiology. MIT Press,
Cambridge.

BELLON, MAURICIO. pies Farmers’
knowledge and ustainable

agroecosystem ee ment: an
operational definition and an example
from Chiapas, ie Human
Organization 54(3): 264-

BERLIN, BRENT. 1995. Ethnobiological
classification, principles of
categorization of plants and animals in
traditional societies. Princeton

1969. An
pera see ntie geal to shifting
agriculture ent an
Cultural iain AP. Ve (editor).
The Natural History Press, Garden Ci

relations in Pakistan. Population and
Environment: ournal
ee Studies, 15 (2).
ELLEN, ROY. 1993. The cultural relations
of wrpemmann an analysis of Nualu

animal categories from central Seram.
Cambri egy University Press,
Cambridge.

FRECHIONE, J.D. et al. 1989. The

Amazon, an ethnoecology of Lake
Coari. Advances in Economic Botan

FRAKE, CHARLES O. 1962. Cultural
Ecology and Ethnography. American
Anthropologist 64 (1).

GRAGSON, TED L. AND BEN G. BLOUNT

ns, Ga.
HUNN, EUGENE. 1985. The utilitarian
factor in folk biological classification in
Directions in Cognitive Anthropology.
J. Dougherty (editor), The Cie. of
Illinois Press, Urbana and Chica
NAZAREA, VIRGINA D. (editor). “1999,
Ethnoecology, situated knowledge /
located lives. University of Arizona
Press, Tucson.
NAZAREA-SANDOVAL, VIRGINIA. 1995.
Cognized models: ethnoagronomy and
ethnogastronomy. Pp. 102-141 in Local
Knowledge and Agricultural Decision
Making in the Philippines: Gender,
Class, and Resistance. Cornell
University Press, Ithaca and London.
POSEY, DARREL A. et al. 1984.
Ethnoecology as applied anthropology,
Amazonian developm n
eels 43 (2):95-107.
RAPPAPOR . 1979. On Cognized
eas in Ecology, Meaning, and
Religion. North Atlantic Books,
Richmond.
WARREN, D. MICHAEL et al. (editors).

systems. Intermediate Technology
Publications, London.

Rebecca L. Austin
Ph.D. candidate
University of Georgia
Research Associate
Fort Lewis College

Journal of Ethnobiology 20(2): 197-216 Winter 2000

KNOWING, GATHERING AND EATING: KNOWLEDGE AND
ATTITUDES ABOUT WILD FOOD IN AN ISAN VILLAGE IN
NORTHEASTERN THAILAND

PRAPIMPORN SOMNASANG
Department of Community Medicine
on Kaen University
Khon Kaen 40002, Thailand

GERALDINE MORENO-BLACK
Department of Anthropology
University of Oregon
Eugene, Oregon 74403

ABSTRACT. — In societies undergoing economic and social transformation, the
level of knowledge about local resources, aie attitudes toward them, can be
affected by a variet lidentity, i with local
pean ne cultural transformations, economic s status, education level, age and
gender. To address the topic of what factors affect traditional wild plant use and
knowledge in Northeastern Thailand we interviewed and tested male and female
villagers in different age categories. Each interviewee was asked to identify and
provide. information about photographs of 40 wild food items. Additional

information g grap hin the village. The results
indicate that th gender diff in th and procurement of wild food
items. However, most vill , despite their economic level, still prefer wild food

over cultivated food. Using a knowledge and attitude test, it was also found that
women are better able to identify items correctly, but that there are no gender
differences in knowledge of other factors pertaining to the wild food items.
Additionally, adults carmen oned ee | than children. Consequently age,
and thus indirectly, exposure to t practices, also i is —— to
knowledge about wild faodis resources. Our It t wild
food is being lost within the community despite retention of the taste for these
items and desire to retain them in the diet.

Key words: Thailand, wild food, food preference, a age.

RESUMEN.—En sociedades sometidas a t f y sociales,
el nivel de conocimiento sobre recursos locales y actitudes sobre ella pail ser
afectadas por una variedad de factores, entre ellas la identidad cultural,
identificacién con tradiciones locales, transformaciones culturales, el estado

econémico, nivel de educacion, edad sexo. Ent y varios
1-¢ 1 1 £
4 > hd 1 ] > L. 4 1 ‘ e 1 . me ]
afectan el uso tra y p j a region ——
de Tailandia. Se el pidié a cada aldeano identifi tents foloe des

de comida salvaje. Obtuvimos informacién adicional através de investigaciones
etnograficas hechas en el mismo pueblo. Los resultados indicaron que existen
diferencias en el uso y obtencidn de articulos de comida salvaje. Sin embargo, la

198 SOMNASANG and MORENO-BLACK Vol. 20, No. 2

mayoria de los aldeanos, a pesar de su nivel econdémico, prefieren comida salvaje
a comida cultivada. La nica diferencia que se revel6 usando pruebas de
conocimiento y aptitud fue que las mujeres ec Bees — resultados
identificando los articulos de comida salvaje

adultos dieron resultados mas altos que a nifios. La edad, e indirectamente la
exposicién a practicas tradicionales de subsistencia, también estan relacionadas
al conocimiento de comida salvaje se pierde ante la comunidad, a pesar de la
retencién del saber de éstos articulos y el deseo de mantenerlos en la dieta.

RESUME.—Dars les sociétés qui entreprennet des transformations, le niveau de
connaissance sur les ressources locales et les attitudes envers ces ressources, sont
affectés par une variété d’agents inculant l’identité culturelle, l’identification avec
des traditions locales, des transformations culturelles, le statut économique, le
niveau d’éducation, l’age et le sexe. Pour aborder le sujet des agents qui affectent
l’usage ee des plantes sauvages at os: capaci re planites ong nord-
est de la Thailande,
des villages d’ges “dispetente On a demandé a aie ‘oe d’identifier et de
donner des renseignements de quarante aliments sauvages qui o ont été
photographié. Les informations supplémentaires ont été obtenu a travers la
recherche ethnographique dans le village. Les resultants montrent qu’il y a des
differénces entre les sexes dans l’acquisition et l’usage des aliments sauvages.
Cependant, la plupart des habitants du village, malgré leur statut économique,
préfeérent les aliments sauvages aux aliments cultivés. Employant un examen de
connaissance et d’attitude, on a également noté que les femmes sont meilleures a
identification précise des articles mais il n’existe pas une differénce de
connaissance entre les seces concernant = ‘atte forteury a propos des articles
sauvages. De plus, les marques des adulte hautes que
celles des enfants. Par consequent, l’age, parent ainsi que exposition aux
practiques traditionelles de subsistance, fait également partie de la connaissance
sur les ressources de la nourriture sauvage. Les resultants suggérent que la
connaissance des aliments sauvages est en train d’étre perdue en dedans de la
communauté en dépit du fait qu’ils retiennent le godt pour ces articles et leur
désir de les retenir dans leur régime.

INTRODUCTION

Human-plant interactions are one aspect of a society’s construction of nature
and the landscape. Recently, researchers have begun to explore the factors that
influence the knowledge people have about their environment. In particular, re-
searchers have begun to ask questions that highlight why people know about some
plants and not others (Nolan 1998) and what determines these perceptions. Cul-
tural factors often take precedence over other considerations such as species
availability or abundance. The process of choosing and obtaining specific plants
or animals speaks to how s aepe Seals view themselves, each other and the environ-
ment. Thus, social rel about the environment are enacted
in the procurement of wild food resources.

In this paper we explore the relationships among people, the natural environ-
ment and the ways individuals preserve and transform their culture and
environment. We specifically focus on knowledge of wild or semi-domesticated

Winter 2000 JOURNAL OF ETHNOBIOLOGY 199

plants and animals and practices concerning the use of these resources because
they create an intensive interaction with the physical and social environment.

Since consumption usually occurs in the home it has often been assumed that
women control the production of consumption as well as the food habits of the
family (McIntosh and Zey 1989; Levin 1943). However, non-domesticated /gath-
ered and semi-domesticated food enters the household through a variety of
channels. Traditional foods, which still form an important part of the diet, may be
procured by the female head of household, husband, male and female relatives,
friends, neighbors and children. Within a community, knowledge, patterns of re-
source use, and the landscape, are affected by gender, class, economic level and
personal life experiences. Consequently, theories about the patterns of decisions
and actions that occur must take these differences into account. Complex local
histories of resource use will do much to increase our understanding of the ways
that local systems of resource management transform themselves in response to
global processes; in particular they can help us evaluate the ways in which the
sustainability of rural production practices are ensured or undermined (Collins
1991). One of the goals of this paper is to describe the variation in knowledge and
resource use within a community. We specifically focus on gender; however, since
knowledge of the environment is embedded in culture, social and economic fac-
tors also are addressed.

DESCRIPTION OF THE STUDY SITE

Physical landscape. —Northeastern Thailand, also called Isan, provides an excel-
lent setting to observe variation in knowledge of the environment and resource
use in the context of gender roles and social relationships. Geographically, Isan is
set off from other regions by mountains and is characterized by erratic rainfall and
poor soils. This gently sloping plateau of undulating hills, terraces and flood plains
also includes a zone of hills and upland areas in the west and the south (Hafner
1990) that extract moisture from the southwest monsoon airstreams. Thus, while
contributing to the biodiversity of the region, these hills also make the area more
Sate: to alvesaec gle ee ey of erratic rainfall, nutrient poor soils
withp moistur water combine to make
the ee difficult for wet-rice panels [ia some of the earliest archaeo-
logical sites in Asia with evidence of agriculture, pottery and bronze work are
located in the Northeast (Higham 1982; Solheim 1968). The semi-arid environ-
ment greatly influenced the traditional subsistence system and other adaptations
to the habitat.

Cultural landscape. —Nature’s impact on culture and personal identity has been
great in Isan. Regional identity, which involves a sense of belonging and pride, as
well as in-group and out-group categorization, is strong and tied to the landscape
and wild resources, especially those that are involved in the cuisine of the region.
Traditionally, the people in the Northeast adjusted to variability in these habitat
factors through the development of a combined subsistence system, in which they
complemented their reliance on the staple glutinous rice and other subsistence

200 SOMNASANG and MORENO-BLACK Vol. 20, No. 2

FIGURE 1.-The landscape surrounding the village. Paddy fields (background) and
sugar cane cultivation area (foreground).

crops with large inputs from wild food (Moreno-Black 1994; Phongphit and
Hewison 1990; P et al. 1986; Somnasang et al. 1988; Tontisirin et al. 1986).
Isan people have a great deal of traditional knowledge concerning the environ-
ment, wild plant and animal resources. They are also knowledgeable about
predicting climatic npg cropping practices, green manuring, and energy ex-
traction (Rambo 1991).

The rich flora and fauna in the Northeast provide a vast array of edible and
useful plants and animals that are gathered for a wide variety of purposes, includ-
ing food, building material, crafts, medicinal use, and religious purposes. These
indigenous practices, and the knowledge that they represent, have been acquired
over many generations and are deeply ingrained in regional Thai culture
(Phithakpol 1990). The diet, characterized by a staple core of glutinous rice, fish,
and fish products, is supplemented by a wide variety of local wild and semi-do-
mesticated plants and animals (Moreno-Black et al. 1996; Moreno-Black 1994;
Ngamsomsuke et al. 1987; Ngarmsak 1987; Pradipasen et al. 1986; Somnasang 1996;
Somnasang et al.1998, 1988). These important items —collected from forests, up-
land fields, rice paddies, gardens, house areas, canals, ponds, swamps, rivers, and
dam areas —contribute valuable nutrients. Coupled with a variety of cooking
methods, they add diversity to a potentially monotonous diet. Northeastern food
has been one regional specialty that has begun to gain popular interest and de-
mand in Bangkok (Van Esterik 1992). In fact, Isan food was characterized and
romanticized in popular folk media and has acquired both national and interna-
tional recognition with the initial publication, gain in popularity, and wide
distribution of the novel Child of the Northeast (Boonthawee 1976).

Winter 2000 JOURNAL OF ETHNOBIOLOGY 201

: aa eee

; Ai fae ES : SO Be
FIGURE 2.—-A scene from the village showing houses and dirt road.

Life-ways are changing rapidly in Isan and villagers are being affected in terms
of economics, social relationships and culture. At the local level, a growing influ-
ence of the nationalized central Thai image, an increase in development projects,
and a rising monetization of the economy are beginning to influence the regional
lifestyle in a variety of ways. Local resources, which used to dominate home use,

7 Ceo ; os
~~. ied %

FIGURE 3.—A gathering party.

202 SOMNASANG and MORENO-BLACK Vol. 20, No. 2

FIGURE 4.—A village woman on her way to gather red ants.

are being used as cash generators. At the same time the abundance of these re-
sources is declining due to national economic and forestry programs that do not
emphasize the local species. The local economy has expanded to emphasize cash
cropping, wage earning and both temporary and permanent out-migration of
adults.

RESEARCH METHOD AND DESIGN

Selection of research village. — The research village was selected from twelve poten-
tial villages in the province of Khon Kaen. All of the potential sample villages
were chosen using secondary data obtained from the Khon Kaen Policy and Plan-
ning Division, The Khon Kaen Governor’s office and district-level extension offices.
Additional information about the villages was obtained through interviews with
government officials at provincial, district, and sub-district government.

Rapid rural appraisals (RRA) (Chambers 1980; Lovelace, Sukesine and Sugin
1988) were conducted in the twelve villages in the province of Khon Kaen. The
headman in each village was interviewed in order to obtain information concern-
ing the general profile of the village, the local environment, occupations of the
residents, problems of concern within each village and the use of local wild food
resources. Additionally, two or three households in each village were selected on
the basis of economic level (poor, middle, rich). Semi-structured interviews
(Grandstaff and Grandstaff 1987) were conducted with the individuals from each
of these households to acquire information about wild food utilization.

Analysis of the headman interview and villager interview data enabled us to
select one village site for an in-depth study. The chosen village was selected be-

Winter 2000 JOURNAL OF ETHNOBIOLOGY 203

cause it represented an average sized village, was characterized by high wild food
use, and villagers obtained wild food from three main sources (forest, local water
resources and paddy fields). Additionally, the villagers were not heavily involved
in selling wild food at the market in the Khon Kaen. Finally, the village was mod-
erately accessible all year round, but was not located on a paved highway.

In-depth village study. — This part of the research involved intensive participant
observation and in-depth interviewing in the study village (Figure 1). Our map-
ping of the village showed that 105 of the 112 houses that were present were
occupied. Census information was obtained from all of the 105 households. The
census interview included demographic information as well as questions concern-
ing household economics, agricultural practices, wild food use and transplanting
activities. The environment surrounding the village was also surveyed in order to
record the sites of forest, cropland, and water resources. The study sample house-
holds were selected by stratified random sampling based on economic stratification
developed in conjunction with the headman. The 64 households in the sample (3
high income, 38 middle income and 23 low income) represent 60% of the total

FIGURE 5.—Separating red ant
eggs from red ants.

204 SOMNASANG and MORENO-BLACK Vol. 20, No. 2

households in the village. Individuals from these households were interviewed
about wild food gathering practices, the types of wild food that are gathered, and
how the food items are used. Participant observation, focusing on food procure-
ment, processing and preparation activities, was used to amplify the interview
data. The interviews were used to obtain information about gathering practices;
knowledge of wild food habitats; the ways wild foods are used; and specific wild
food management practices such as transplanting and propagation techniques.
A “Knowledge” test and an “Attitude” test were administered to adults from
the study sample and children who attended the one school in a nearby village
village. Sixty-four adults (32 women and 32 men) and forty-one children (21 girls
and 20 boys) were interviewed using a set of laminated photographs of 40 wild
food items selected from information obtained during interviews and participant
observation (Table 1, Appendix A). The forty food items were divided into four
categories: plant (32 items), fish (5 items), insect (2 items) and snail (1 item).

TABLE 1.—Age and gender distribution of “Knowledge and Attitude Test”
participants.

Age Female Male
10-11 Zt 20
15-25 8 8
26-35 8 8
36-45 8 8
46-55+ 8 8

otal 53 oz
Mean age of school children = 11 years

jean ag! f wy © years

Each adult was asked to identify the item (identification test). These data were

used in the odds ratio analysis and the scoring system described below was then
utilized to compute the I score component of the knowledge test. The knowledge
test was composed of several parts and individuals were scored on: ability to iden-
tify an item (I score); culinary information, such as how to eat, prepare, and cook
the food as well as its taste (C score); non-culinary uses (O score); horticultural
practices, such as transplanting, maintaining or propagating the item (TP score);
and knowledge of procurement practices (P score). Each person could score up to
ten points per item (Table 2) and a total of 400 points for the complete test of 40
items.
For the attitude test individuals were also asked about their attitude toward
each item especially in terms of taste qualities and consumption preferences. The
children were given the same set of laminated pictures of the wild foods; however,
the questions were simpler and shorter since they were only asked to identify
each item and express their knowledge of and attitude toward the item.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 205

TABLE 2.—Criteria and scoring system for the Wild Food Knowledge Test.”

CRITERIA NUMBER OF POINTS
No Knowledge Incomplete Knowledge Knowledge

Item Identification (I score) 0 1 2
Culinary Knowledge (C score) 1 2
Procurement Knowledge (P Score) 0 1 2
Other Use Knowledge (O score) 0 1 2
Horticultural oe ee Score) 0 1 2

Incomplete knowledge = = Intervi ither has i plete knowledge of the it

Knowledge = Int qd trat full k led in t f all criteria

dge
RESULTS AND DISCUSSION

In the Northeast, resources are recognized and perceived as useful within the
context of the environmental and social reality of the icapie Rnowledge of local
plants and animals, especiall y tl that consumed as food,
accumulates over generations and reflects the way people learn from and about
their environment. The villagers relied heavily on a variety of non-domesticated
plants and animals. A diversity of habitats was utilized, including paddy fields,
upland areas, forests, ponds, streams, swamps, rivers, and other water reservoirs.
Commonly utilized items include leafy algae, green plants, fruits, mushrooms,
amphibians, crustaceans, fish, birds, reptiles, insects and mammals. Wild foods
were also used as condiments and often contribute to the distinctive flavor of Isan
food. Wild food entered the household in a variety of ways: 1) production through
family agricultural activities; 2) procurement through gathering, fishing and hunt-
ing by household members; 2) gifts from relatives and neighbors, 3) exchange with
other individuals in the village or nearby villages; and 4) purchase. Consequently
many individuals can contribute to a household’s food consumption and men,
women and children all have some involvement with wild food beyond consum-
ing it.

Shared traditions, beliefs and attitudes ing wild food use. —The majority of people
in the village reported eating or using wild food at the present time or in the re-
cent past. The majority of villagers indicated they liked to eat wild food, often
preferring wild food to cultivated food and food from the market. Only 3 people
(5%) reported they did not like wild food, while only 3 people felt that they like
wild food as much as cultivated food. The villagers believe that wild food is nec-
essary and that it is the most important food for everyday life. Analysis of the
in-depth interviews revealed that males and females did not differ in their atti-
tude ing wild food and both men and women preferred to consume wild
food over cultivated food.

206 SOMNASANG and MORENO-BLACK Vol. 20, No. 2

TABLE 3.—Reasons for preferring wild food over cultivated food (n = 64)

REASONS FOR PREFERRING WILD FOOD RESPONSES
Female Male Total

Taste Factor

Taste: delicious, taste better, etc. 19 15 34
Higher nutritional value, less fat 6 7 13
Fresh, better than market food 6 6 12
Natural 5 5 10
Total 36 33 69

Ease of Procurement and Safety
Less poisonous agents, fewer chemical,

less harmful 17 17 34
Easy to obtain, can gather on their own 9 7 16
Cleaner than market foo 6 3 9
Total 32 27 59
Economics

No need to buy food, save money 15 14 29
Total 15 14 29

Note: Maximum answers = 3 per person

The villagers gave a variety of reasons why they preferred wild food over
cultivated food (Table 3). The most frequent reasons were related to qualities con-
cerning: 1) taste; 2) ease and safety; and 3) economics. The concept of taste is
obviously very important and included a number of important components. First,
villagers often mentioned the fact that wild food is more delicious and tastes bet-
ter than cultivated food. They also thought wild food was fresher and tasted more
naturally sweet than market food. This is partially because of the inherent quali-
ties of the food as well as the fact that villagers can gather wild food and consume
it soon afterwards. Villagers also indicated that wild food was better because it
grows naturally, has more nutrient value, especially vitamins and protein, and
less fat. These specific nutrition-related characteristics, which some individuals
are now ascribing to wild food, are most likely derived from information obtained
from government sponsored health and nutrition education programs.

The second group of reasons for preferring wild food revolved around the
concept of ease and safety. Villagers preferred wild food because it was easy to
obtain or they did not have to spend time or energy cultivating it; it is natural and
grows by itself. Villagers consider wild food safe in part because they assume only
cultivated food would be contaminated with fertilizers or insecticides. However,
it is likely that many of these items, especially those growing in paddy fields and
gardens, are contaminated with fertilizers and pesticides. Wild food is also con-
sidered to be clean because it comes from the natural environment. It is not mixed
with food that can become dirty at the market in town.

The last set of reasons is related to economics. The villagers considered wild
food to be good because they did not have to pay money for it, thus they were able
to save money by eating wild food. Many villagers felt that poor people needed to

Winter 2000 JOURNAL OF ETHNOBIOLOGY 207

FIGURE 6.—A typical dish made with red ant eggs.

rely on wild food because they do not have much money to spend on food. How-
ever, the villagers did not look down on individuals who relied on wild food.
Many people also recognized that even rich people utilize wild food as a way to
save on spending money for something that is available without monetary cost.
Consequently, wild food was not stigmatized as a “food of poverty.”

Procurement patterns were affected by time factors. Gathering is as commonly
done in conjunction with other activities, such as gardening, agricultural wage
labor or tending cattle, as it is done as its own activity. Gathering patterns are to a
large extent dependent on both the seasonal availability of the food and seasonal
workload of the villagers. In the rainy season two factors impact the amount of
time spent gathering. First, because wild food is abundant, less gathering time is
required. Secondly, in this season villagers are actively engaged in rice agricul-
tural tasks so they tend to choose to gather food that is close at hand. Similarly for
those households that are moving into cash oriented agriculture, such as growing
asparagus or cucumbers for the commercial market, the opportunity to gather
wild food is limited. Thus, it was not surprising that in the rainy season 53% of the
women spent an hour or less gathering. On the other hand, in the cool season,
when there are fewer agriculture-related demands, 64% of the women spent one
to two hours gathering. In the hot season, however, the scarcity of wild food, more
than time constraints from other activities, results in the village women spending
more time gathering wild food, and 56% spent two or more hours in gathering
activities.

The selection of food is influenced not only by time constraints and preference
but also by attitudes about the identity of individuals who are known to consume

208 SOMNASANG and MORENO-BLACK Vol. 20, No. 2

the food. Personal identity and food consumption are linked in powerful ways.

The villagers in this study expressed a variety of opinions

who consume wild food. These associations consciously and unconsciously affect
the selection of these foods within the village context. Villagers indicated that they
believed that wild food is essential for the poor. They indicated that the poorer
people in the village have to depend on wild food, have more experience obtain-
ing it, and are very knowledgeable about how and where to procure wild food.
The villagers also expressed the sentiment that the poor also exchange wild food
for other food items such as rice. However, villagers with all these opinions also
indicated that they did sas eee down on individuals who gathered wild food;

instead, they were and hardworking and often a degree of
admiration was voiced.

Conflicting beliefs were expressed when others in the village indicated they
believed rich people consumed more wild food because they had money to buy it
when it was not easy to obtain or when they were too busy to gather it themselves.
A few individuals indicated they believed rich people ate more wild food because,
unlike the poor people who had to engage in wage labor, wealthy individuals had
time to obtain non-domesticated food. In these individuals’ minds, the poor people
in the village relied on cheap, prepared foods from the village shops. For this group,
purchased food was associated with poverty and was considered to be poor qual-
ity foods.

When asked how urban people view wild food, the villagers were of the opin-
ion that town people like non-domesticated food and are often eager to purchase
these items for ingredients in specific dishes, or as snacks. Villagers believed that
wild food has a high market value, sells better than domesticated food and thus
brings them a better earnings than domesticated food. A very small number of the
interviewees (10 individuals) differed from the general opinion of the rest of the
sample by thinking that towns-people had a poor attitude toward wild food, look
down on village people who eat wild food or think that wild food, especially in-
sects and some animals are not clean and are disgusting. These villagers also
thought that the urban people may look down on villagers who eat this type of
food.

Variation in wild food knowledge. —Variation in knowledge of wild food was evi-
dent in the study sample, although the majority of villagers stated a preference for
wild food and continued to seek out and procure wild food items. The knowledge
test was used to measure recognition, culinary knowledge, consumption, procure-
ment, other uses, and transplanting and horticultural techniques (Table 2).

An odds ratio (Agresti 1990) which interprets differences between two popu-
lation proportions or possibilities, was employed to estimate how frequently one
population identified a wild food correctly compared to the other (Identification
test). We first compared men and women and then compared adults and children
in terms of their ability to identify items in the wild food test (Table 4). In this
analysis we grouped items into categories (plants, fish, and insects). The odds
ratio analysis showed that, compared to men, women were more likely to be able
to identify plants (1.38 times), insects (1.38 times) and fish (1.07 times).

Winter 2000 JOURNAL OF ETHNOBIOLOGY 209

TABLE 4.—Gender based comparison of likelihood of correct identification of
wild food.

Gender Female Male Odd’s Ratio
I N I N

Adult

Plant Group 626 1006 552 1014 1.38*

Fish Group 59 129 64 145 1.07

Insect Group 45 63 40 62 1.38*

Total Wild Foods 730 1230 656 1253 RS

Children

Plant Group 175 326 155 312 ee a

Fish Group 34 76 42 73 0.60

Insect Group 16 2 13 32 1.80*

Total Wild Foods 225 445 210 429 1.07

Note: I = Total correct identifications
N = Number of valid samples
* = significant difference at p _.05

Among children, girls were more likely to identify wild plant species correctly
(1.17 times), and insects correctly (1.80 times), than boys. However, girls recog-
nized fish less often (0.60 times). When the children and the adults were compared
concerning their ability to identify the wild items, the odds ratio showed that adults
were 1.31 times more likely to identify plants and 2.35 times more likely to cor-
rectly identify insects. But the adults were less likely to identify fish correctly (odds
ratio = 0.78) (Table 5). A total of 25 plants were used in the “attitude and knowl-
edge” test. Eleven plants were correctly named by 80% of the children, but there
were 16 plants that 20% or fewer of the children could name. However, a very
large majority of the students (78%) said they consumed between 13 and 15 of
these plants, and they also indicated they liked wild foods.

In general, adult women and girls out-performed their male counterparts in
identification of wild food items. Girls were 1.07 times more likely to identify the
total three groups of wild food than boys, and adult women were 1.34 times more
likely to identify all three groups correctly than the men (Table 4).

TABLE 5.—Age based comparison of likelihood of correct identification of wild
food (“identification” test).

Wild Food Type Adult (N=64) Children (N=41) Odd’s ratio
I N I N

Plant group 1178 2020 330 638 1.31"

Fish group 123 274 76 149 0.78

Insect group 85 125 29 61 23557

Total Wild Foods 1386 2419 435 848

Note: I = Total correct identifications
N = Number of valid samples

= significant difference at p _.05

210 SOMNASANG and MORENO-BLACK Vol. 20, No. 2

The wild food knowledge test was also used to compare knowledge of wild
food by comparing mean scores on the components of the knowledge test: identi-
fication (I score), culinary knowledge (C score), procurement methods (P score),
other uses (O score), transplanting and horticultural information (TP score) and
total score (S score) (Table 2). The mean scores of each food group were calculated
and compared between males and females and among the age groups using an
en see of covariance: The analysis of covariance showed that there was no sig-

der in the I, C, P OTP and S sores. Although women
tended to have higher scores in each component, the differences were not statisti-
cally significant. However, there were significant differences between the age
groups for both males and females for all five categories of the knowledge score: I
(p = .001), C (p = .009), P (p = .004), O (p= .001) TP (p = .001), and S (p = .001). Thus,
it is clear that the older individuals have more knowledge about plants than the
young. However, the age factor was not significant in terms of knowledge of in-
sects or fish.

Retention and loss of knowledge aus local d ticated resources -* an
issue ¢ 1 with tai
siteetvel diversity, and ethnobiological knowledge, and those interested in under-
standing how culture is expressed and rendered meaningful. An examination of
our data shows that gender and age are factors related to differences within this
community in terms of knowledge of wild food. This finding is similar to research
done by Wester and Yongvanit (1995). When they compared individuals from dif-
ferent villages and levels of education in Isan, they also found that in almost all
age groups women scored slightly higher than men on a test of wild plant knowl-
edge. In their sample of 795 males and females (10 - 99 years of age), they also
found that there was a general tendency for scores to increase with subject age
until about 70 years, when scores of men showed a sharp decline. They further
found, in marked contrast to scores of village populations, students, all of whom
were younger than 30 years old, scored low. However, Ogle (1984) in her research
in Swaziland, found that children recognized a large number of different wild
species and reported high consumption of the items. Consequently, she concluded
knowledge was not being lost to the extent feared.

Differences in knowledge, both within and between communities and groups
of individuals is to be expected, since men and women utilize and define the envi-
ronment differently. The division of labor that occurs in agricultural communities
leads to differentiation in work patterns, contact with resource areas and procure-
ment of resources themselves. That children can have less familiarity or knowledge
than elders is also not surprising since they have had less time to accumulate in-
formation, fewer life experiences, fo are often not as involved as adults are in
resource utilization. Add s adults are pulled into wage earning activities
outside the village, opportunities ie communication of information also decrease
or vanish completely.

Differences in knowledge should not be accepted or dismissed lightly since
the loss of traditional knowledge among many subsistence-oriented communities
in many parts of the developing world has been noted. For example, Anderson
(1993) has expressed concern about the perpetuation of such information among

Winter 2000 JOURNAL OF ETHNOBIOLOGY 211

the hill tribes of northern Thailand and Works (1990) recorded that elders in a
Peruvian community frequently lament that younger individuals are no longer
interested in plants or gardens. Similarly, Maikhuri and Gangwar (1993) observed
that knowledge of plants in younger individuals in the Khasi and Garo tribes of
Northeastern India was judged to be poor. Similarly, the older villagers in our
study often expressed concern about their perceptions of the loss of knowledge
among the younger generation. Elders complained about their children and grand-
children not being interested in learning about the procurement and preparation
of wild food.

CONCLUSIONS

Loss of traditional knowledge among agricultural communities as they expe-
rience the effects of globalization has been noted in many parts of the world.
Nonetheless, many individuals remain connected to local practices, at least to some
degree. In Isan, the strong connection between ethnic identity and cuisine encour-
ages the preservation of local resources that imbue their taste, texture and odor to
Isan cuisine. Many of these items are wild plants and animals. Additionally, many
of the plant food items are also integrated into the local belief system regarding
health and serve as medicinal plants. The continued reliance on important wild
resources has led to items being utilized by different people at different times.
Consequently the distinct interests that different segments of society have in the
rural productive environment influence resource use and preservation (Collins
1991). However, as suggested BY. neste and sone vanit (1995), es iaacioange or vio
of traditional practices and | conscious choic
It is sometimes an incidental result of new patterns of living.

Adults are increasingly participating in the new economic patterns, which,
rather than being embedded in the mixed subsistence practices, more and more
involve cash cropping of introduced and domesticated crops as well as working
for wages and both temporary and permanent out-migration. Children and young
adults are also affected. Young adults are increasingly attracted to the trappings of
the urban, cosmopolitan culture. Children who attend school, especially the higher
grades outside the village, are occupied with studies that keep them from partici-
pating in subsistence actives to the extent that children did in the past. It also
exposes them to different ideas and activities. In the study reported here, the con-
nection between land and life that was deeply embedded in daily survival in the
past appears to be in the process of being altered as a result of the adoption of
urban values and goals. Individuals are also absent from the locale during the
periods of their life when they would haye been participating in activities that
necessitated the accumulation of local I the d wild
food resources. Although they still enjoy and prefer the tastes that these wild food
items give to local dishes, and indeed connect them to local ethic identity, they are
not as knowledgeable about them as the elders are.

The information obtained in this study highlights the fact that there are vital
connections between gender and age (and thus work patterns, division of labor,
and participation in the wider national and global processes) and knowledge of

212 SOMNASANG and MORENO-BLACK Vol. 20, No. 2

local resources. Women were 1.34 times more likely than men to identify plants,
insects, and fish correctly. The knowledge scores showed that women have greater
knowledge than men in terms of recognition, gathering knowledge, preparation
and consumption knowledge, and uses of wild food. However, the scores from
the total | ge test indicate that there were ificant gender differences.
There were significant differences in plant knowledge among “the different ages,
with the older individuals scoring higher than young adults and children. These
differences, which may be indicators of erosion in the local knowledge base, are
potentially being exacerbated by both temporary and permanent out-migration,
which results in fewer opportunities to learn about their native environment and
the resources that are utilized.

It is clear that factors such as gender roles, division of labor and the economic
and social impact of nationalization and globalization determine the breadth and
depth of knowledge about the habitat and local indigenous resources. These fac-
tors are probably more important than resource abundance and diversity in the
contemporary conditions of rapid economic and cultural change. The fact that
wild food knowledge is largely a construct of such factors has important implica-
tions for the survival of both knowledge about these resources, and ultimately the
resources themselves. Villagers in the study are aware and concerned about the
potential loss of knowledge. They frequently expressed a desire to preserve wild
food for the younger generations. They mentioned that some types of wild food
were rare, and they feared their children or grandchildren would not get to enjoy
them. Villagers often proposed ideas to enhance retention, such as transplanting
species, decreasing the amount of wild food gathered for sale, working with offi-
cials to develop programs, willingness to comply with laws that would protect

and habitats, and seeking to develop educational programs directly
cngied at children.

Our findings indicate that younger individuals have less knowledge than the
elders in the village. It is important, especially from the perspective of the villag-
ers, that local knowledge about wild plants and animals be maintained and
documented. We believe government policymakers and development workers
should consider ways to integrate local knowledge into the formal education sys-
tem and incorporate it into relevant projects whenever possible. Efforts should be
made to work with village elders to develop resource material such as illustrated
pamphlets or monographs about local wild food resources for schools and public
libraries.

We also believe that villagers, forestry experts, and government officials could
work together to determine ways to preserve indigenous species while still en-
abling villagers to utilize these resources. Additionally, villagers should be
encouraged to develop management strategies that will limit harvesting practices
that are currently straining local resources. At the same time, it is vital that the
concept of sharing be maintained, so that villagers can procure and consume wild
food together, thus enjoying both the social and nutritional benefits that charac-
terize [san cuisine and culture.

Our study indicates that it is very important to begin obtaining dynamic ac-
counts that relate women’s and men’s activities to the process of ecological change

Winter 2000 JOURNAL OF ETHNOBIOLOGY 210

and the continuity of local traditions and knowledge. The preservation of knowl-
edge about wild food plants appears to be the result of the continued connection
with the local geography, and personal identification with the village or region.
This is linked to specific food items or taste qualities, economic pressure, and in-
trinsic family interest and motivation to preserve traditional patterns. In particular
it is important that we recognize the ways in which different interests in the envi-
ronment are structured by gender, age, class, caste, or ethnicity in order to identify
individuals who are knowledgeable about local habitats and resources. Identifica-
tion of important cultural influences in the construction of local knowledge will
be invaluable for the design and application of conservation programs.

ACKNOWLEDGMENTS

This study was part of a larger project focusing on the marketing and use of non-
domesticated, indigenous plants and animals in the northeastern part of Thailand. The
project was — by grants from The Wenner Gren Foundation for Anthropological
Research (G. M Foundation (G. M-B), a Margaret McNamara
Memorial arg yeaa (P.S.), and a Ford Research Grant, NW Regional Consortium of
Southeast Asian Studies (P.S.).

We thank the staff of the National Research Council of Thailand for their assistance
and willingness to allow this project to continue over an extended period of time. We are
especially grateful to the villagers who participated in the project for their patience, help,
information and hospitality. They were very generous with their time and their knowledge
and openly shared their experiences with us. Special thanks go to Drs. G. Fry, C. Silverman
and H. Wolcott for their helpful comments during the development of this paper. We also
thank the anonymous reviewers. Their positive feedback is greatly appreciated. They also
provided invaluable comments on this article. Finally, we are especially grateful to our
families who willingly and patiently endured our absences, our work schedule and who
supported our efforts throughout the project.

LITERATURE CITED
ANDERSON, EDWARD. 1993. Plants and COLLINS, JANE. 1991. Women and the
People of the Golden Triangle: environment: Social reproduction and
Ethnobotany of the Hill Tribes of sustainable development. In Women and
Northern Thailand. Dioscorides Press, International Development Ann
Portland. R.Gallin and A. Fergurson aie), Vol.

AGRESTI, ALAN. 1990. Models for binary 2. Westview, Boulder.
response variables. Pp. 79 - 129 in GRANDSTAFF, _ 5S.W. ‘
Categorical Data ee John Wiley GRANDSTAFF. 1987. “semi- structured
and Sons, Ne

BOONTHAWEE, K. 1976. Child of the _ in RRA. Pp. 129 - 143 in Proceedings of

Northeast. Duang Kamol, Bangko the 1985 International Conference on

CHAMBERS, ROBERT. 1980. Rapid Rural Rapid Rural Appraisal. Rural Systems

Appraisal: Rationale and Repertoire. IDS Research and Farming System Research

Discussion Paper No. 155. Institute of Projects, Khon Kaen University,
Thailand.

Development Studies, Brighton.

214 SOMNASANG and MORENO-BLACK

HAENER, J. A. 1990. Forces and policy
issues affecting forest use in ee
Thailand 1900 -1985. Pp. 69 - 94 i
Keepers of The Forest, M. Poflenberser
(editor). Kumarian Press, West

Hartford.

HIGHAM, C.FE.W. 1982. Prehistoric rice
cultivation in Southeast Asia. Scientific
American 250:138-146.

LEVIN, KURT. 1943. een jee et

habits and methods of c e. Pp.
65 in The Problem of cana Food
Habits, Bull. No. 108. National Academy
of Science, National research Council,
Washington, D

LOVELACE, G., Ss. SUKESINE, and S.
SUJIN. 1988. Rapid Appraisal In
Northeast Thailand: Case Studies. KKU-
Ford Rural Systems Secerich i
Khon Kaen University, Khon Kaen,
Thailand.

psi tin R.K. and A. K. GANGWAR.

a Garo tribes of Meghalaya,
Northeast India. Economic Botany 47:
345 - 357.

McINTOSH, ALEX and MARY ZEY. 1989.
Jape nen of food
Pp flepnied id Ha
and acta 3(4):317 - 332.
MORENO- BLACK, CERALDINE. 1994.
Gathered food and culture change:
traditions in transition in Northeastern
Thailand. J. Home and Consumer
Horticulture ies 3):93-107.
rE MNASANG, and S.
~THAMATHAWAN. 1996. Cultivating
continuity and creating change:
Women’s” garden practices
Northeastern Thailand. acl and
Human Values 13(3):3-12
NGAMSOMSUKE, K., KAMOL, ¥
SAENCHAL, P. PROMBUROM, and B.
SURAPORN. 1987. Bases attitudes
toward forest plantation an
conservation farming in ealketed
villages of the Phu Wiang valley, Khon
Kaen. FAO/UNDP rite Integrated
Development of t u Wiang
Watershed Project, "hor Kaen
University, Khon Kaen, Thailand.

Vol. 20, No. 2

NGARMSAK, TIPWANNA. 1987. Status

geogra Ry and
knowledge in the Oza
highlands. Journal of éhucbioloay
18(2):249 - 269.

OGLE, BRITTA A. 1984. gee plant
resources. Cares 101: 38 -

PHONGPHIT, SERI ey KEVIN
HEWISON. 1990. Thai Village Life:
Culture and Transition in the Northeast.
Village Foundation, Bangkok.

PHITHAKPOL, BULAN. 1990. Preface. In
Plants from The Markets of Thailand, C.
Jacquat (editor). D. K. Bookhouse,

angko

PRADIPASEN, MANTANA,

Dietary Survey. Institute for Population
and Social Research, Mahidol

University, i a a Thailand.
—— O, TERRY. an Ecology
Rural resource Ménagatiaht in
oilinaont Thailand. Farming Systems
a Project, Khon Kaen University,

n Kaen, Thailan

SOMNASANG, PRAPIMPORN. 1996.
i se: Gender issues in
h.D.

dissertation, University of Oregon,
Eugene

' GERALDINE MORENO-
~~ BLACK, and KUSUMA CHUSIL. 1998.
Indigenous knowledge of wild food
hunting and gathering in Northeast
Thailand. Food and Nutrition Bulletin
19 (4): 359 - 356.

AGARAT RATHAKETTE and
SUMALEE RATHANAPANYA. 1988.

(editors). KKU-Ford Rural Systems
Research Project, Khon Kaen University,
Khon Kaen, Thailan

SOLHEIM, W.G. IL. 1968. Early bronze in
northeastern Thailand. Current
Anthropolology 9:59-62.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 215

VAN ESTERIK, P. 1992. From Marco-Polo
to McDonald’s: Thai cuisine in
transition. Food and Foodways 5: 177-

TONTISIRIN, K. , J. YOUNG-AREE,

PTONTIWATTASATION, K.SRIANUJATA,
. SUKANGPLENG. 1986. Annual
Raa 1985/1986: Thailand Food
Habits Project. ASEAN Subcommittee
on Protein: Good Habits Research and
Development. ASEAN-Australian
nomic Cooperation Program,
Bangkok, Thailand.

WESTER, LYNDON and SEKSON
hear cicconale 1995. Biological diversity
and community lore in northeastern
Thailand. aod of Ethnobiology 15:71

WORKS, MARTHA. 1990. Dooryard
gardens in Moyobamba, Peru. Focus: 40:
12-17.

216

APPENDIX A.—Wild food items used in the Knowledge and Attitude Test*

SOMNASANG and MORENO-BLACK

Botanical name Local name
Plants

Amaranthus gangeticus phak kaenkhom
Amorphophallus spp. erok
Antidesma acidum

Butomopsis latifolia
Calamus spp

Centella asiatica
Cratoxylon formosum
Curcuma parviflora
Cyclea peltata
Dioscorea alanta

mark mao

pinoy

whai (rattan)
mark muay (fruit)

dork grajeaw (flower)
kruamanoi
mon liam (tuber)

Dioscorea hispida kloy (tuber)
Emilia sonchifolia phak linpi
Garcinia cowa somong
Hydrocharis morsus-ranae yopae
Irvingia malayan bak bok (fruit)
Limnoccharis flava phak kanjong
Limnophilia aromatica phak kayang
Marsilea crenata phak waen
Monochoria vaginalis ehin
Ottelia alismoides obab
Oroxylum indicum linfa (pod)
Polygonum odoratum phak paew
Sauropus androgynous kantong
Solanum trilobatum mark kheng
Stephania brevipes huabua
Thyrsostachys siamensis normai huak (bamboo shoot)
Wolffia globosa kipum
— mark pipuan
— liumpua
Fungi
Careya sphaerica hed phungtam
Fresh water fish
Cirrhinus jullieni pla khona
Cirrhina microlepis pla suit
Cultrops siamensis pla pap
Solidago polyglossa pla khao huangleang
Snail
ai hoi sai
ects
Brachytropes portentosus
ate) maeng huakuai

*Dpl ee

Poe % Ant A}
7
Homchoen at Khon Kaen University.

consultation with Dr. Sompong Thamathawan and Dr. Samang

Winter 2000 JOURNAL OF ETHNOBIOLOGY 217

Feeding the World. Vaclav Smil. MIT Press, Cambridge. 2000. Pp. xxviii, 360,
many text figures, bibliography, index. $32.95 (cloth). ISBN 0-262-19432-5.

This book is an absolute must for all interested in the world food situation and
its future. It is the best up-to-date source I have seen on the subject. Smil defines
his subject in the widest possible way, and brings together an amazing range of
information on all of it. One stands in awe of his data retrieval system. (This is his
15'* book, and the others are comparably data-rich.

The best feature of the book, and intended so to be, is his focus on efficiency as
the best way to deal with food shortages in the near future. Much of the book is a
long litany of wastes: water and fertilizer wasted in the field, grain lost to weevils
and rats in storage, grain fed to livestock that could eat grass, foodstuffs unneces-
sarily thrown away in processing, and on to plate waste. This last grows ever
worse as people live on takeout junk-food and no longer save leftovers. As he
points out, we already have the world food problem solved, as far as production
goes. We are producing enough food for all humans alive today, and could even
provide for the entire population expected to be with us in 2050. The problem is
all that loss. And his figures on that are conservative. He believes loss in storage
to be around 10 percent to 15 percent; there are much higher estimates.

Smil is a cautious optimist. He tries to steer a course between “cornucopians”
and “catastrophists.” He dismisses the former (such as the late Julian Simon) with
a single tart line: “If the global grain output were to continue growing only as fast
as it has done during the 1980s (almost 2 percent a year), the annual harvest of
cereals would surpass the Earth’s mass in less than 1,500 years...(p. xii).” The
catastrophists are much more formidable foes. Much of the book is taken up by
debates with Paul Ehrlich and Lester Brown, whose concerns cannot be written
off. To be sure, their dire predictions have been wrong; he gives a series of Brown's
now-invalidated ones (p. 12). But one might argue that the reason these predic-
tions were wrong is that they scared people: into action. After all, they were usually
couched in terms of “if nothing is done..

In any case, something was done and people are now better fed than ever
before in history. About 1.2 billion people are hungry, but 1.2 billion are
overnourished, so it balances out; the problem is clearly one of efficient allocation,
not absolute shortage. (A more politically liberal observer than Smil might say
that there is a bit of a redistribution problem there, too.) “If the rich world’s food
losses could be held to 20 percent of the overall supply, the annual savings...would
be equivalent to...nearly half of all cereals on the world market (p. 210).”. Mom
was right to tell us to “think of all the starving people in Asia”— though, even at
the age of eight, I wondered how my eating too much and getting fat was helping
them. Smil has it right: we should stay thin and let the price of food fall.

The bulk of the book consists of a truly incredible assemblage of information
on the state we are in — so cebaiuiae fertilizers, crops, land base, soil erosion, stor-
age, and on to proces , human nutritional needs, and plate waste.
He notes, for ies that some of the gloomy Seedicione of the 1960s were based
on assessments of human protein needs that we now know were far too high. We
can get by on a little plant protein; we don’t need all that meat. But, if we want

218 BOOK REVIEWS Vol. 20, No. 2

meat, some meats (chickens, pigs) are far more efficiently produced than others
(America’s adored beef is the worst). Eggs and milk are better still. The greatest
value of this book to ethnobiologists, after its basic message of efficiency, is its use
as a reference work; it is encyclopedic in coverage of a vast and often obscure
literature on agriculture and food.

o one human can bring so much together without making some dubious
claims, however. Predictably, most of Smil’s are in the optimistic direction, but he
has also missed some cheering thoughts.

To begin with the over-optimism, Smil accepts the current projections (by the
United Nations and other agencies) that world population will level off around 10
billion in the next couple of generations. I do not believe this. The easy battles
have been won: Europe is down to ZPG, East Asia is near it, and some other well-
organized, highly educated countries have made a start. The rest of the war is
going to be a great deal harder. Birth rates are falling slowly in South Asia and
Latin America, but so are death rates. Birth rates are not falling, or not by much, in
Africa and the Middle East. In these areas, little or nothing is being done to reduce
population increase. China’s one-child rules are cracking and the system may
crumble. Even the United States continues to grow rapidly, and current govern-
mental policies are increasingly antithetical to demographic leveling off. Absent
the most horrific of Malthusian checks, we will probably see rapid population
growth throughout Latin America, Africa, and west and south Asia for the rest of
this century at least, and appreciable growth in the United States.

This is debatable. Much less debatable — indeed, a clear mistake — occurs on
page 194 where Smil claims that “diets of several hundred million people are ap-
preciably enriched by consumption of hunted and collected wild animal species”
(apart from fish). Alas, overhunting and habitat destruction have made this state-
ment obsolete. Only in the most remote and thinly populated areas — the Subarctic,
the Australian outback, the inner Amazon — do people get significant game meat
today. Much more typical is the Yucatan Peninsula where game was a staple food
as recently as a generation ago, but now is virtually nonexistent.

Another place where one might question Smil is his section on desertification.
He says “...there is little doubt that virtually all early estimates have greatly over-
estimated the impact of desertification, mainly because they mistook the cyclical
nature of these changes for steady degradation.... Desert margins contribute rela-
tively little to global food supply...” (p. 76) There are problems with both these
claims. The cyclic waxing and waning of the Sahara against the Sahel was under-
estimated in the late 20h century, but this does not greatly change the estimates.
Overgrazing, deforestation, and overcultivation have been devastating. There are
too many thousand photographs of “climate change” stopping short at a barbed-
wire fence or a reserve border (see e.g. the magnificent collection in Jacobs 1995, or
Charco 1999) to make “climate change” a believable explanation of the world’s
desertification. I have personally seen thousands of cases, on four continents, of a
desert landscape giving way—at a fence or other barrier—to a three-foot stand of
lush grass or a dense brushland. This can take place on large scales: Even the
rather thin protection that Israel gives the Negev has now made Israel's national
border quite visible in satellite photographs.

Winter 2000 JOURNAL OF ETHNOBIOLOGY ae

The same could be said for erosion. Smil correctly celebrates the really amaz-
ing strides against soil erosion that have been made in the United States, Europe,
and some other places, and concludes that soil erosion is not of major concern. Yet
he is surely familiar enough with China to know the catastrophic state of erosion
there. Perhaps he is less familiar with India and the dry parts of Africa. Clearly he
is less familiar with Mexico and Latin America. Of course, there is very little good
information on many of these areas, but what we have — and what anyone can
see on the ground or from the air — is quite disturbing.

Here, as in some other cases, Smil tends to assume that “no news is good
news.” This assumption stands on somewhat believable ground when Smil notes
that many national statistics understate production. (Maya subsistence farming
in western Quintana Roo produce tens of thousands of tons of maize and fruit a
year, none of which gets counted in national statistics.) However, the assumption
is hard to credit when soil erosion is at issue. Here, the human tendency is the
other way: to overlook and underestimate. Satellite pictures could improve our
understanding, if we knew enough about interpreting them.

So much for over-optimism — Smil’s one lapse into under-optimism is in pre-
cisely our area. Smil seems only slightly aware of the enormous potential of
underutilized and under-researched crops and cultivation systems, to say nothing
of wild plants that could be cultivated. All readers of this journal will have their
own pet examples and the cumulative total thereof (were we to pool our knowl-
edge) would surely be enough to feed the world a few times over. Possibilities for
expansion range from relatively well-known systems like Maya mixed orchards
and Spanish olive groves to exotic potential crops like California’s tarweeds and
meadowfoam, and from well-known but undervalued animals like guinea pigs to
outside cases like oryx and addax antelopes.

But is there a chance that all these measures will be adopted? Is Smil right, or
will the catastrophists prove all too correct in the end?

It is well to remember that, although “on average” the world is doing well, the
most dreadful fantasies of the catastrophists are now the reality in many coun-
tries. These include Ethiopia, Eritrea, Sudan, Somalia, the entire Sahel, and several
other African countries, as well as Afghanistan, Uzbekistan, parts of Kazakhstan
and Ukraine, North Korea, Haiti, and many more. These suffer from dense and
fast-growing populations, collapse of food production, desertification (not caused
by climate change!), lack of education and research, and, often, other environmen-
tal catastrophes, from endemic warfare to Chernobyl. It is noteworthy, but hardly
surprising, that the most environmentally devastated countries are also the coun-
tries with the worst food problems.

Conversely, the successes of Europe have gone beyond the wildest dreams of
the optimists. Zero population growth and lavishly abundant food are accompa-
nied by rapid improvement in the environmental outlook, as green consciousness
spreads across the continent. Spain and Portugal are notable among countries
that have seen explosive growth in production and income while actually improv-
ing (at least locally) their environments.

What accounts for these differences? The conventional wisdom provides us
with three possibilities, all obviously wrong. First, most common among environ-

220 BOOK REVIEWS Vol. 20, No. 2

mentalists and still not uncommon among developers, is the idea that “capital-
ism” is the source of all evil. This clearly does not account for the above picture.
Among other things, many environmentalists seem to think that the rich nations
are environmentally more trashed than the poor ones, because of high consump-
tion. This is not the case. To see really ravaged environments, one needs to go to
such localities as Uzbekistan, Ethiopia, or China. It is the most productive, food-
exporting countries that also do the best by their environments. Second, there is
the reverse view: Capitalism is what the world needs; socialism is the evil. This
fails to account for ongoing and worsening problems of countries that have enthu-
siastically bought into capitalism and accepted IMF discipline, including most of
Latin America and southeast Asia. Third is the idea that dependency and global-
ization are the culprits. If this were so, we would expect to find countries very
tightly enmeshed in the dependency end of the global economy, such as Mexico,
Taiwan, South Korea, and Thailand, to be the poor ones. They are not; it is the
most isolated countries, such as Ethiopia, Somalia, Bhutan, and Laos, which are
the worst off.

There is one simple predictor. Strong yet d ith a strong
tradition of accountability are always associated with progress in both ea pro-
duction and environmental awareness. Scandinavia and the Low Countries are
examples. Spain, Portugal, Greece and Hungary — the point by rapidly de-
veloping food production and (except possibly inG l awareness
in the wake of democratization. Weak yet authoritarian governments are at the
opposite pole, characterizing the sad examples listed above. A change from de-
mocracy to chaos or authoritarian rule accompanies environmental decline
(Malaysia, Indonesia, Pakistan), which eventually must lead to food production
failure (Guatemala, El Salvador, and elsewhere). It would seem that this correla-
tion should be studied systematically by those interested in the problem. It has
more to tell us than either the global optimists or the global catastrophists.

Smil ends his book with a look at China, a country he knows extremely well.
Here he crosses swords with Lester Brown whose book Who Will Feed China? Wake-
up Call for a Small Planet (1995) made the catastrophist case in lurid detail. They
both work from very similar factual bases, and (as I know from my own indepen-
dent research on China) they are both quite reasonable in their interpretations of
the data. The difference between them is really over something they never dis-
cuss: The leadership China has and will have. Brown assumes that China’s
ee will continue to be as it i snow. At present, the leaders do not oneal
1 problems that might limit future food product nd the
routinely i imprison those who raise the issue. Smil evidently hopes and sass that
a new generation will have new ideas.

Indeed, it appears that Smil’s cautious optimism, Brown’s worries (see also
Brown et al. 2000), or the extreme optimism of the irrepressible and delightfully
outrageous Libertarian Ronald Bailey (2000), have much less to do with the facts
than with their take on human nature. They all say surprisingly similar things
about what we are doing, what we can do, and what we need to do, technologically.
They even have similar political views, seeing governments as far too prone to
hinder rather than help. The difference is that Brown sees governments as inevi-

Winter 2000 JOURNAL OF ETHNOBIOLOGY 223

table and frequently prone to act as they do in Sudan and Afghanistan. Smil looks
at the governments of Europe and Canada, and hopes. Bailey dreams of abolish-
ing government altogether.

One wishes Bailey were right. If only people, released from the bonds of the
State, would work together for love and profit. Unfortunately, the world is the
way it is. I know Bailey is wrong. I hope Smil is right. But if I had to bet money,
I’d bet with Brown.

LITERATURE CITED
BAILEY, RONALD (editor). 2000. Earth , CHRISTOPHER FLAVIN, and
Report 2000: “pains. the True State "HILARY eee “grace 2000. State
of the Planet. McGraw-Hill, New York. of the World 2 W. W. Norton for
BROWN, LESTER. 1995. Who Will Feed Worldwatch emia, New York.
China? Wake-up Call for a Small Planet. CHAKECO) TESUS- (1999. El Bosque
W. W. Norton for Worldwatch Institute, Mediterraneo en el Norte de Africa.
New York. Agencia Espanola de Cooperacién
Internacional, Madrid.

JACOBS, LYNN. 1995. Waste of the West.
Author, Tucson.

E. N. Anderson
Department of Anthropology
University of California Riverside

Journal of Ethnobiology 20(2): 223-236 Winter 2000

THE JAGUARS OF ALTAR Q, COPAN, HONDURAS: FAUNAL
ANALYSIS, ARCHAEOLOGY, AND ECOLOGY

DIANE A. BALLINGER
Dallas VA, North Texas Health Care System
Department of Physical Medicine and Rehabilitation
UT Southwestern Medical Center
D

JEFFREY STOMPER
College of Lake County
Grayslake, IL

ABSTRACT.—An excavation at Copan, Honduras, a Late Classic Maya site,
revealed the ritual cache of bones of at least fourteen big cats associated with
Altar Q. Several of the large cats were identified as jaguar, Panthera onca.
Preliminary analysis showed that the animals were in good health at the time of
their deaths. All but one were adults. Tail fans of several species of birds
accompanied the feline bones. oe with od crypt of the felines were the
smaller tomb burials of two macaws, Ara sp. Th

that all the jaguars were procured locally because of environmental constraints.

Keywords: Maya archaeology, zooarchaeology, jaguar, Copan, ritual

RESUMEN.—La excavacion del escondite des huesos felinos en el sitio Maya con
la fecha de eke tardio en re Honduras, ha revelado un entieramiento des
losh a ,y otros gatos grandes, asociados con
el Altar Q. El analisis ha mostrado que los animales estaban saludables quando
murieron. Todos menos uno que eran adultos. Los huesos de los jaguares estaban
acompanados por saat abanicos de cola de varios pajaros. Los huesos de dos

han cerca dela crinta

arcangas Arasp.,

x:
de los j jaguares. Los autors concluyen que no es possible que los jaguares fueron
obtinidos en el valle de Copan porque del constre(imientos cercania.

RESUME. Redan i Copan Honduras un site Maya de Classique Tardif ,a
revelé Panthera onca, quatorze
chats grands, associés avec l’Altar oe sprees des ossments a monstré que les
animaux etaint en bon santé au temps du mort. Il y a un jeune et trieze adults
dans |’assemblage. Les ventails de la queue de les oiseaux a accompagné les
ossments des chats grands. Associés avec le crypt des les animeaux il y ont deux
petites enterrements de les deux oisieux, Ara sp. Les auteurs ont conclué que ce
n’etait pas possible pour les jaguars Itre obtiner dans le vallée de Copan parce que
des contraintes d’environnment.

224 BALLINGER and STOMPER Vol. 20, No. 2

INTRODUCTION

The discovery of an offertory cache of feline bones yielded an exciting glimpse
of ancient Maya ritual behavior to archaeologists. Feline bones filled the masonry
crypt to the east of Altar Q at Copan, Honduras, an altar whose sides bear high
relief portraits of the sixteen rulers of Copan’s Classic Period dynasty. The cache
at the foot of Altar Q contained the remains of at least fourteen large felines, two
macaws, and the tail fans of seven other birds. This is a report of the discovery and
preliminary analysis of the offertory complex associated with Altar Q and Struc-
ture 10L 16 in the Acropolis area of the Main Group. Jeffrey Stomper, then a graduate
student under the supervision of Dr. William Fash, the Director of the Copan
Acropolis Archaeological Project instituted under the auspices of the government
of Honduras, excavated the offertory complex in 1988. The offering consisted of a
sealed crypt containing the remains of at least 14 large felines, including Panthera
onca, and macaws (Ara sp.) associated with a stone altar on a low, round platform
that was placed on the central axis of the western side of Structure 10L 16 in the
Main Group at Copan.

Until now, the evidence that large felines were involved in Maya ritual has
consisted of skeletons of single animals and portions of skeletons of cats buried in
ritual contexts (Pohl 1983, 1990). Landa (Tozzer 1941) discussed animal sacrifices
stating that animals were used in rituals and were sacrificed. They were presented
to the gods either alive or not, sometimes dismembered but also whole (Tozzer
1941). The cache at Copan is the largest cache of felidae bones discovered by Maya
archaeologists to date.

JAGUAR SYMBOLISM

Maya art and iconography are rich with depictions of jaguars. The same is
true of other Mesoamerican civilizations. Copan, Palenque, Uxmal, and Tikal are
among the lowland Maya sites where jaguars are featured prominently (Morley,
Brainerd, and Sharer 1983; Spinden 1975; Tozzer and Glover 1910). Spinden (1975)
considered them second only to the snake in symbolic importance to the Maya.
Coe (1972) pointed out the links between rulers and jaguars and the separation of
king and commoner in native Mesoamerican religions. He further remarked on
the antiquity of the jaguar as a religious icon and its ties to Mesoamerican reli-
gions and the ruling lineage (Coe 1972).

Images of jaguars frequently decorate ceramics recovered from Maya sites,
appearing on a variety of vessels. On these, men often wear clothes with jaguar
markings or jaguar pelts (Spinden 1975: 149). Many times jaguar symbolism is
part of a ritual or religious context on vessels, such as one from Altar de Sacrificios.
On this vessel, the ruler of Yaxchilan is dressed in jaguar skin trousers, mitts made
of jaguar paws complete with claws, and a jaguar headdress. A second figure in
jaguar regalia is near that has the arms, hands, tail and feet of the jaguar on a
human body (Saunders 1989: 146).

The jaguar epitomized two different kinds of strength to the Maya. The jaguar
motif was associated with the underworld and its supernatural power and also
with physical strength. The association with physical strength derived from the

Winter 2000 JOURNAL OF ETHNOBIOLOGY 225

fact that jaguars are powerful, nocturnal hunters. Ideas of the supernatural power
of the jaguar arose from the early totemic tradition of Mesoamerica and the jaguar
nagual’s (spiritual co-essence) relationship to the shaman’s power. Possession of
supernatural power and physical strength was important to Maya rulers because
their political power rested on their ability to act as a priestly bridge between the
ancestors, the underworld, and the living world (Schele and Freidel 1990).

Maya artists juxtaposed rulers and warriors with jaguar symbolism in art.
Strength and prowess of combat were required of both the ruler and the warrior
but Span of the ruler. A fearsome beast of the tropical forest, the jaguar per-
sonified a dual control of the supernatural, a necessary power for Maya
kings, and the physical prowess needed by the successful warrior (Saunders 1989;
Hassig 1985).

Mesoamerican cultures from Olmec to Aztec revered the jaguar. The Olmec
associated jaguars with shamanic power and filled their art with images of were-
jaguars. The Aztecs linked the jaguar to war, sacrifice, and royalty (Saunders 1989:
150; Hassig 1985). Furthermore, they associated jaguars with jade, rain, and fertil-
ity (Saunders 1989). At the time of the Spanish conquest, Aztec (Mexica) traders
routinely transported pelts and live animals from outlying parts of the empire to
the capital. Pictures of both live animals and pelts appear on trade and tribute lists
of goods moving from Soconusco, the colonial province located on the coastal
plain of the state of Chiapas, Mexico, to the Aztec Empire (Voorhies 1989).

Relicts of this symbolism are found today in remote areas of Central America,
Mexico, and South America. Jaguar symbolism is most visible in masks and cer-
emonies performed in isolated villages. Wearing a jaguar mask transforms the
wearer into a new creature that combines animal, human, and | quali-
ties (Saunders 1989). Thus, the jaguar is an ancient and potent symbol permeating
native cultures of Mexico, Central, and South America.

ARCHAEOLOGY

As part of the first season’s work of the Copan Acropolis Archaeological Project,
William Fash conducted preliminary investigations of Structure 10L-16 and the
adjacent Plaza areas in 1988 (Fash 1991; Agurcia, Stone and Stomper 1989). During
the spring of that year, excavations at the site of the western base of Structure 10L-
16 and Altar Q were in progress, supervised by Jeffery Stomper and veteran local
excavator, Ismael Gonzalez. Stomper concentrated the initial excavations on the
area in front of and beneath Altar Q, anticipating finding the dedicatory offering
or other remains of associated rituals at that locus. Other examples of dedicatory
caches had been found either directly in front of or underneath altars and stelae
elsewhere at Copan (Stromsvik 1941). An area was marked off directly west of the
altar and excavated to a depth of 120 cm. Fearing for the integrity of the trench if
he dug between the stone supports that uphold Altar Q, Stomper elected instead
to tunnel beneath the altar from the eastern sidewall of the same pit. This mini-
tunnel excavation beneath the altar produced no evidence of a cache, nor was
there any evidence that the area under the altar had been disturbed.

226 BALLINGER and STOMPER Vol. 20, No. 2

4 em

1g.

rf
sa te
a; Lh)

>>)
1h

Wi, :
O—_—
=

| {

P ; orte (today)
‘

Corte (c.1900)

iS,

FIGURE 1.—The East Court, Main Grou
h

p, Copan, Honduras. Supplied by William and
Barbara Fash.

Winter 2000 JOURNAL OF ETHNOBIOLOGY yf

Stomper began another pit that mirrored the location of the previous one.
Immediately, he found two features. The first was a plain, round altar, CPN 13470,
with a diameter of 36 cm. Altars such as these were used as stands for incense
burners elsewhere in Copan during the Late Classic period (Fash 1983:464). The
second feature was what appeared to be several rough slabs of stone, Feature 8, in
a line below the level of the last plaza floor. The excavation was expanded to un-
cover the extent of this feature.

The capstones were aligned in a north-south direction, sloping from east to
west, and covering the masonry crypt that contained the animal bones. Probing
between the stones revealed a hollow area approximately 1.3 m deep and at least
1m wide. When Stomper removed the capstones, he found that debris from places
where the walls had caved in filled the crypt. The crypt, measuring 131 cm long
and 48 cm wide, had walls of eight courses of finely cut building stone rising to a
height of 117 cm.

Within the crypt, the excavators found animal bones. The removal of the first
level of bones revealed an irregular intrusion of lime plaster in parts of the cist.
Excavation of this layer revealed that more bones were embedded in the layer of
plaster with still more bones below it. These layers were removed in the plaster
matrix in large sections and kept separated. Stomper continued excavations into
the fill under the cist to a depth of 3 m under the plaza level. He recovered only 7
fragments of ceramics from under the cist. Approximately 319 cm below the origi-
nal cist floor, the excavators uncovered an earlier plaster floor of the West Court.
Excavations were terminated at this point.

Two small shafts adjacent to the cist were also uncovered just outside the north-
east and southeast corners of the cist. The inside of the roughly square shafts
measured 22 cm on each side. At the bottom of one shaft were nine whole pris-
matic obsidian bladelets and parts of two others. Located 25 cm above the obsidian
were the bones of a medium-sized bird. The second shaft also contained the bones
of a bird but no obsidian. Nothing else was found.

Stomper excavated the entire area between Altar Q and Structure L10-16, re-
vealing a round platform of small, faced stones two courses high in some places.
Unfortunately, the whole platform was not preserved. Located in the center of the
platform was an oval stone with a smooth upper surface, repeatedly charred by
fire. Resting on the top of the platform was a small, highly polished fragment of
jade. The area around the platform yielded incensario fragments, a ceramic in-
cense burner. In an adjacent area to the west of the platform just below the level of
the last plaza floor, a small cache of obsidian lancets was found.

FAUNAL ANALYSIS

During the summer of 1988, the faunal material was identified and recorded,
and a preliminary report was filed at the Central Office of the Project in Copan
(Ballinger 1988). The bones were counted, sorted into skeletal elements, and ex-
amined for pathological conditions: and é anomalies. i Eenpaet used weight-bearing
bones of th to detern umber of individuals (MNI)
because weight-bearing bones have large areas of tee compact bone, and, hence,
tend to be better preserved (Brain 1981). Left and right elements were identified

228 BALLINGER and STOMPER Vol. 20, No. 2

and counted. Ballinger matched the bones by size or age to determine if they be-
longed to a single individual (Chaplain 1971). Juveniles were identified by the
lack of epiphyseal closure. The analyst ignored most of the fragments because
time was limited. The largest MNI of the elements was then determined as the
MNI for the species (Klein and Cruz-Uribe 1984). Actual bone counts were made
at Copan. Photographs of the best preserved crania were taken for later identifica-
tion. Final identification to species level was made in the Zooarchaeology Lab at
Indiana University, where a comparative collection is housed.

Analysis was done under field conditions without the use of a comparative
collection or manuals commonly used by faunal analysts. Neither a comparative
collection nor library resources were available in 1988 at the Copan laboratory to
aid in identification. Ballinger took notes supplemented with photographs and
sketches. The initial identification of the bones as feline resulted from her exami-
nation of the teeth, crania, scapulae, and femora while in Honduras.”

Condition of the Bones.—Burial in a closed crypt resulted in good preservation. Al-
though the bones were subjected to natural decay, they were protected from some
of the taphonomic processes that radically change relationships between skeletal
elements (Lyman 1982). Thus, while many of the bones were not articulated, they
were close to the position in which they were placed in the crypt. The crypt walls
kept them from being dispersed after burial. They were also protected from water
and heat, the two most powerful agents in bone dissolution (von Endt and Ortner
1980). The bones were dry and chalky but retained their shapes, so skeletal ele-
ments were easily identified. Exfoliation was present on some of the bones but
most of them were well preserved.

An assessment of pathological lesions was made on the bones that had mini-
mal flaking on them. The bones of the second level, excavated as a unit and curated
as a unit, allowed Ballinger to determine the position of the jaguars of the second
level in the crypt by the association of skeletal elements. The heads of some had
been laid over the rear legs and feet of others.

Results of the Faunal Analysis—The MNI of the felines was at least 14. This number
is conservative because the analyst was unable to perform a complete examina-
tion of all the faunal material. Ballinger found that many of the bones found in the
crypt were Panthera onca. Differences in jaguars and puma lie in cranial morphol-
ogy. According to Olsen (1968), jaguar crania have a sagittal concavity on the
superior aspect of the cranium that rises to a pronounced lambdoidal crest that
gives a slight s-shaped curve to the jaguar’s skull. The posterior crest gives the
jaguar cranium a squared off appearance and the skull appears longer and more
rectangular. The puma, however, has an oval skull lacking the massive, posterior
cresting. Crania of jaguars and puma are similar in their anterior aspects and teeth
but differ in the posterior aspect. It is the posterior features on the crania the dis-
criminate between the species. The more complete crania from the crypt have
distinctive nuchal robusticity and a more elongated architecture of the jaguar. Al-
though there is not enough cranial material remaining to account for 14 jaguars,
an estimated 6 animals in the assemblage were Panthera onca. The others are fe-
lines but remain to be positively identified as jaguar. The remaining bone in the
assemblage is bird bone and a few intrusive rodent bones.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 229

The bones of the felines are very similar in size, indicating that the animals
were similar in age and sex. They were also healthy animals. Assessment of indi-
cators of general health routinely surveyed by investigators: cortical thickness,
osteoporosis, and the frequency and severity of periosteal reactions, lesions of re-
active bone growth resulting from localized and systemic infections, demonstrated
that the cats were free from these pathological conditions that commonly mark
the bones and indicate a decline in health status.

Large cats are prone to bone diseases in captivity. Osteoporosis, thinning of
the bones, is a problem for large felines in zoos. It results from inactivity, too little
protein, old age, and metabolic disorders (Fowler 1986). Caged felidae are also
highly susceptible to metabolic bone disease that results in a rickets-like bowing
of the long bones (Fowler 1986). The cortices of the bones were thick and only one
periosteal reaction was found. The single periosteal reaction was on the hind foot
of one animal and had fused two metatarsals. Appendicular bones were robust
with prominent muscle markings, and had no sign of rickets-like bowing.

No cut marks, dismembering marks, or skinning marks were found on the
feline bones that were examined. The removal of pelts results in a characteristic
pattern of skinning marks. Skinning for pelts often makes cuts ringing the lower
metapodial where the knife has circled the ankle or wrist to loosen the skin. Car-
pals, tarsals, and the bones of the digits are then removed with the pelt and, thus,
are missing from the assemblage. Ballinger found no evidence that the pelts were
removed after death. Further work, however, should include careful examination
of the bones for skinning and other butchering marks.

In addition, the vertebrae must be examined. Ballinger concluded that crania
may have been disposed of separately for some of the cats. The paucity of cranial
fragments compared to the amount of post-cranial material leads Ballinger to the
conclusion that crania may have | disposed of separately. Further analysis may
reveal cut marks on the first or second cervical vertebrae if the heads were re-
moved before burial.

The avian bones in the assemblage were collected from two places, the small
shafts adjacent to the crypt and from the crypt itself. The bones of an adult macaw,
Ara sp., were found in each shaft. Both of the macaw skeletons were missing the
pygostyle, the bone to which the tail fan attaches. Two pygostyles of the correct
size were recovered from the crypt, possibly the ones missing from the birds in the
shafts. The pygostyle is a fragile bone, however, and may have been destroyed by
postmortem diagenesis. There are seven other pygostyles from other unidentified
birds in the crypt. Three of these probably are the same species. The other four are
two matching sets and may represent two more species. Lack of a comparative
collection precluded the complete identification of the avian bones. In all, nine tail
fans were buried with the jaguars. Other bird bones buried in the crypt included a
caudal vertebra, a fragment of a tarsometarsus, a fragment of a proximal radius, a
rib fragment, and two phalanges. All of these remain unidentified. There were no
signs of rodent gnawing or cut marks on the avian bones.

Preservation of the macaw bones differed for the two shafts. Burial 1 lacks
cranial bones. The head may have been removed at the time of death but the re-
mains were too broken to properly check for cut marks. The remains of more than

230 BALLINGER and STOMPER Vol. 20, No. 2

one bird may be present in the second shaft. The bones removed from the second
shaft have more fragments of long bone and cranium than Burial 2. Differences
between the two shafts’ contents, however, may have been affected the microenvi-
ronments of each shaft. Ballinger identified the macaws by the presence of the
beak in Burial 2 and a comparison of the two birds. It was clear that post-cranially
they were the same (Hargrave 1970). However, all of the avian bone should be
reanalyzed with a comparative collection at hand.

In summary, a MNI of at least fourteen large felidae, some of them Panthera
onca, were identified from the bones in the crypt. The animals were healthy with
no indications of longterm protein deficiency or inactivity. Accompanying the fe-
lines were the tailfans of nine birds, including two macaws. No evidence of
butchering or mode of death was found in this analysis. A full study of the feline
bones may reveal sex and age differences, less obvious evidence of disease, and
the mode of death of the animals. Similarly, comparison of the avian material to
comparative skeletal collections may reveal the species of the unidentified mate-
rial.

DISCUSSION

Natural History and Ecology of the Jaguar—The jaguar (Panthera onca), is the largest
of three species of spotted cat native to Central America (Burton 1987). Classed as
a big cat, the jaguar has a shoulder height ranging from 110-155 cm (Burton 1987).
The jaguar prefers a forest habitat but can live in savanna environments if there is
enough brush cover. They also live in slightly arid areas. Good swimmers and
climbers, jaguars are primarily nocturnal feeders, sleeping during the middle of
the day (Burton 1987).

Few investigators have studied the natural history of the jaguar in the wild so
most information comes from zoo studies. Many of the details about breeding
patterns come from studies of zoo animals. Jaguars are very easy to raise in cap-
tivity. They breed easily and can live on two or three pounds of meat a day,
according to David Ruhter,’ former Curator of Large Animals, Houston Zoo (Per-
sonal communication via telephone, 1994). Less is known about their behavior in
the wild. Large felidae are long-lived animals whose prey is normally about half
their body weight (Sunquist and Sunquist 1989; Gittleman 1984; Packer 1986). They
usually need large areas in which to forage and, except for lions, are solitary.

Today, the jaguar has adapted to more crowded habitats in Central America.
In 1983, Rabinowitz and Nottingham (1986) tracked the movements of nine jag-
uars in Belize. The authors found that the home ranges of the jaguars overlapped
but that as long as prey was available, they remained on them, avoiding each
other. Analysis of scat determined that in Belize jaguars feed on seventeen species
of prey (Rabinowitz and Nottingham 1986). This is consistent with reports from
other areas to the south that jaguars feed on diverse prey: capybara, fish, peccary,
and alligator (Ewer 1973). Generally, jaguars feed nocturnally but the female in
the Belize study changed her feeding pattern to the daytime in order to exploit
cattle as prey. This study suggests that the jaguar is flexible and able to adopt new
habits as needed to survive (Rabinowitz and Nottingham 1986).

Winter 2000 JOURNAL OF ETHNOBIOLOGY 231

Naturalists do not know much about the behavior of the puma in Central
America. Felis concolor occupies a wider range of habitats than the jaguar, ranging
from mountains to jungles to deserts (Ewer 1973). The puma is slightly smaller
than the jaguar. Puma diet is omnivorous in tropical climates where they consume
several types of rodents, fish, and other small game. The main component of their
diet is deer and domesticated farm animals, although they have demonstrated
behavior flexibility in the selection of prey species.

The jaguar and the puma are in almost direct competition in Central America
because of the similarity in size, habits, and diet. Puma are also nocturnal feeders
unless forced to feed at other times. They are known to live near home ranges of
jaguars (Rabinowitz and Nottingham 1986) in Belize, but in the North American
west, their ranges are larger and do not overlap.

The ecology of the jaguar has important implications for explanations of how
the Maya acquired at least fourteen big cats for ritual use. Rue (1987) states that by
the Late Classic, the valley was heavily deforested resulting in a loss of habitat for
tropical forest animals. If, however, enough brushy areas still existed in the valley,
jaguars could have survived there, given the presence of adequate prey. In gen-
eral, species that live in savanna or parkland areas have larger populations than
forest living animals (Berkoff et al. 1984), thus deforestation could lead to a slight
increase in the population of cats in the valley. Jaguars have demonstrated their
ability to survive crowded conditions and the ability to change feeding schedules.
It is highly probably that they survived in a mosaic environment of parkland and
brush.

Ultimately, however, access to prey governs population size. A deforested but
still brushy area could have supported sufficient deer and other species of prey to,
in turn, support a small jaguar population. Similarly, a mosaic of brush and cleared
spaces will support a white-tailed deer (Odocoileus virginianus) population whereas
closed canopy forest will not (Smith 1975). Deer have small home ranges, often
living near humans and thriving. Historically, deer have been an important part
of Maya subsistence (Mandujano and Rico-Gray 1991) and only recently have de-
clined in importance as numbers decreased. Both brocket deer and white-tailed
deer feed in cleared areas of new growth near fields because tender shoots are
present there. There also may be fewer insects in young forest (Mandujano and
Rico-Gray 1991). Mandujano and Rico-Gray (1991) remarked that the decline of
Yucatan’s deer population was directly attributed to over-hunting and loss of habi-
tat, a relatively new situation. Population decline was exacerbated by the loss of
native farming methods that provided browse each season after the fields were
burned, cleared, and farmers severely trimmed brushy growth in and around the
milpas. This growth sprouted anew with the rains. Pohl (1994) proposed that
Copanecos were very likely to have raised deer in and around their homes and
fields in the valley.

Other animals available as prey are Brocket deer (Mazama sp.) and peccary
(Tayassu tayacu). Both would have done well in this open, brushy habitat. Brocket
deer browse on the same kinds of plants as white-tailed deer: tender twigs, shoots,
and leaves of a variety of herbaceous plants and fruits. The peccary is omnivo-
rous, living in a variety of tropical habitats, including forests and dry savannas

22 BALLINGER and STOMPER Vol. 20, No. 2

(Lawlor 1979). Smaller prey, such as paca, agouti, and armadillo are native to the
area and can live in brushy environments.

The Copan Valley could have supported a small population of big cats even if
it were partially deforested. Whether it could have supported a population large
enough to provide at least 14 adult big cats from local sources is another question.
The Copan Valley is comprised of 26 square miles of territory. Of that, the heavily
populated Copan pocket would have been too urban-like for jaguars or pumas to
live there. Game such as deer would have been drawn to the areas surrounding
fields further out rather than the kitchen gardens closer to the central site. Big cats
would have lived farther out from the Main Group at Copan.

Normally big cats have large home ranges of 15 square miles. If we assume
that crowding has limited the size of the home range a home to 5 square miles, the
Copan Valley would only provide ranges for five jaguars at a time. Avoidance
behavior as demonstrated in Belize would not make it possible for more animals
to survive unless even with prey populations of very high density. Thus, the large
number of felines found in the crypt most likely resulted from a combination of
trade, hunting, and hand raising.

THE OFFERTORY COMPLEX AND THE JAGUAR BURIAL

Yax Pasah, the last ruler of Copan, had begun an aggressive new building
program by raising Temple 11 and making significant additions to the West Court,
where Altar Q is located including Altar Q (Fash 1991). The altar is an illustration
of the succession of the last ruling dynasty of the polity. Along the sides of the
altar the names sixteen rulers are inscribed beginning with Yax Ku’k Mo’ (Blue or
First Quetzel Macaw) and ending with Yax Pasah. Yax Pasah, by his choice of
iconography for the West Court, indicated its relationship with the underworld,
Xilbalba, and, thus, to his ancestors (Schele and Freidel 1990).

Yax Pasah faced serious problems during his reign, including deforestation, a
population that was shrinking, and a diminution of his power (Fash 1994). Schele
and Freidel (1990) note that most of his monuments state the cosmic sanction of
his rule, and hypothesize that his reign was marked by crises. His impressive sac-
rifice and building program may have been an attempt to restore Copan to its
previous place in the hierarchy of major sites.

Structure 10L-16 was the final product of the remodeling in the West Court.
The jaguar burial occurred at this time as a ceremonial cache for the dedication of
the last version of Structure L10-16. The Copenecos placed Altar Q at the base of
this structure. Yax Pasah buried at least 14 big cats in the crypt associated with
Altar Q. Once the people of Copan placed the felines in the crypt, tail fans of birds
were placed there, and the crypt was then covered with three capstones. A small,
round, carved altar was placed next to it directly between the two shafts. A final
plaza floor was then laid, covering the capstones. Jaguars and other spotted cats
had an unknown but important ritual significance at Copan. Pohl (1994) remarks
that while skeletons of spotted cats were often cached in ritual contexts by the
Maya, Copan has more such caches than any other site. All of these have so far
been found in elite contexts. Pohl (1994) considers them a measure of the high

Winter 2000 JOURNAL OF ETHNOBIOLOGY Zao

position of the site of Copan in the Maya hierarchy. The inclusion of macaws was
ritually significant but their meaning is less clear. The name of the dynastic founder
was macaw. Macaws occur frequently in the decoration of public ritual space at
Copan, especially on the ball court.

Questions of where the jaguars came from and how they were killed remain
unanswered. The ecological conditions and size (over 75 km) of the Late Classic
Copan Valley could have supported a small population of jaguars or pumas even
if partially deforested. It was not, however, large enough to provide 13 adult and
one semi-adult large cat. David Ruhter (Personal communication 1994) suggested
that the easiest way to assemble that many jaguars would be to gather kits in the
spring and raise them by hand. There is evidence that jaguar kits were given as
gifts. A vase from Tikal shows a turbaned figure holding a kit and presenting it to
the ruler (Culbert 1993).

The feline bones, however, presented no evidence of long-term inactivity or
protein deficiency, indicating that the Copanecos had taken good care of the ani-
mals and that their caged time was short. At this time, it is impossible to determine
whether the animals were raised by hand or captured as adults.

We can only speculate about how the jaguars met their fates. Preliminary ex-
amination revealed no cut marks on the bones to indicate that the jaguars had
been skinned or dismembered. A more complete analysis of the bones may reveal
skinning marks leading to determination of their modes of death. The presence of
crania and numerous tarsals and metatarsals, carpals and metacarpals indicate
that some heads and feet were not removed. Often on ceramic vessels, individuals
are portrayed dressed in jaguar pelts with heads and paws still attached (Coe 1973;
Schele and Miller 1986; Kerr and Kerr 1989). It appears that the valuable and ideo-
logically significant skins were left on the animals.

CONCLUSION

Jaguars could have lived in the Copan Valley even after deforestation if a
brushy, mosaic environment was present. The size of the valley, the amount of
tilled fields, and even with deer management, the amount of prey limited the num-
ber of jaguars and pumas that were living there. The flexibility of feeding patterns

d the presence of small home ranges as demonstrated by modern jaguars in
Belize indicate that jaguar populations in the past could have adjusted to smaller,
overlapping ranges close to humans populations without undue stress. Jaguars
would have been in direct competition with the Maya for scarce protein given
their reliance on a maize-based diet.

The lack of disease and stress indicators on the bones of the jaguars indicates
that the bones are the remains of generally healthy animals that had not been
captive for a long time. These animals had no chronic met that cause
bowing of the leg bones, no osteoporosis, and only one periosteal reaction. Fur-
thermore, all but one of the animals were adults with full epiphyseal closure. This
indicates that healthy, adult cats were chosen for the cache. The cats could have
been kept in good condition for as long as four to six months by being fed two or
so turkeys each day (David Ruhter, Personal communication 1994).

234 BALLINGER and STOMPER Vol. 20, No. 2

Hopkins (1992), in her work on the animals bones removed from the Cenote
of Sacrifice at Chichen Itza, argues against the assumption that the jaguar bones
found at Chichen were solely the result of sacrifice because there are too many
other reasons that the bones could have been deposited in the cenote. At Copan,
the bones clearly are in a ritual context, one that is repeated many times at the site.
The cache of the feline and bird bones represented a significant gift to the people
of Copan. By placing his sacrifice in the West Court with all of its symbols of the
underworld, Yax Pasach hoped that a continuous connection with the ancestors
and the jaguars’ supernatural power would be forged. Thus, homage was paid to
garner the aid of ancestors and gods to keep the polity strong and growing when
many Maya cities were falling into decline. It was to no avail. Shortly after this,
the people of Copan ceased to erect dated monuments and the population of the
central valley declined.

Future work should include analysis of the assemblage of feline bones. Mea-
surements of the bones and assessment of small differences between the jaguar
and the puma should be made. The analysis should also contain reconstruction of
crania and pelves, where possible. Finally, a close examination for cut marks and
any other butchering marks should be carried out.

NOTES

1 The jaguar bones are curated at the Central Laboratory in Copan Ruines, Honduras. This
is also where the field reports, preliminary reports, manuscripts, theses, and dissertations
are housed. The archives at Copan include drawings, photos, and a small library and are a
good resource for investigators.

2 Two students of Wm. Fash have examined the collection of animal bones but I have not
been able to contact them by publication time.

3 David Ruhter was curator of large animals at the Houston Zoo when I contacted him. He
has since moved out of state and the personnel department at the Zoo would not release
his address to me.

ACKNOWLEDGEMENTS

Barbara and Bill Fash provided the site map, read the draft, criticized constructively,
and secured the permission to publish from IHAH. Many thanks. The reviewers made
very good suggestions on improving the manuscript. Any errors are the authors’ own.
Further thanks go to Rebecca Storey who first took me to Copan and permitted me to share
housing with her in 1988. She has also helped me fulfill the governmental regulations that
are necessary to work in another country. Ballinger: s } dissertation research at Copan in
1988 was funded by a grant awarded by the D gy and the Graduate
School, Indiana University. My thanks also go to Dick Adams who gave me lab space and
let use the comparative collection in the Zooarchaeology Laboratory, Indiana University,
several times.

Winter 2000

JOURNAL OF ETHNOBIOLOGY 235

LITERATURE CITED

STOMPER. 1989.
rieiiie saat of Structure 10L-16,
Copan, Honduras. Presented at the 55th
annual meetings of the American
Anthropological Association,

AGTRE RICERDO. DONNA STONE
nd JEF

Washington D.C.
BALLINGER, DIANE A. 1988. Preliminary
a: oe Waa 1 Sat 1, f£ Altar

Q. MS on file at the Proyecto Copan
Laboratorio, Copan Ruines, Honduras.
BERKHOFF, MARC, THOMAS J. DANIELS
and JOHN L. GITTLEMAN. 1984. Life
history patterns and png comparative
social ecology of carnivores. Annual
Review of Ecological Systeins, 15:191-

232,
BRAIN, C. K. 1981. The Hunter or the
caste University of Chicago Press,

Chica

BURTON” JOHN A 1987. The Collins Guide
to the Rare Mammals of the World. The
Stephen Green Press, Lexington.

CHAPLAIN, R. E. 1971. The Study of
Animal Bones from a ae ta
Sites. Seminar Press, New

COE, MICHAEL D. 1973. The Maye Scribe
and his World. The Grolier Club, New

. 1972. Olmec jaguars and Olmec
kings. In The Cult of the Feline, E.
Benson (editor). Dumbarton Oaks
Research Library and Collections,
Washington, DC.

CULBERT, T. PATRICK. 1993. Maya
Civilization. Smithsonian Books,
Washington, D

EWER, R. F. 1973. The Carnivores. Cornell
University Press, Ithaca,

FASH, WILLIAM L. 1983. Reconocimiento
y excavaciones en el valle. Pp. 229-469
in Introducci6n a la Arquelogia de
Copan, C.F. Baudez (editor). Volume 1
Secretaria de Estado en el Despacho de
reall 2 Turmiso, Tegucigalpa,
Hondur

ees 1991. Scribes, We
Thames and — London

. 1994, Changing perspectives on
Maya civilization. Annual Review of
Anthropology 23:181-208.

FOWLER, MURRAY. E. 1986. Zoo and Wild
Animal Medicine. 2nd edition. W. B.
Saunders ap aon Philadelphia.

d Kings

GITTLEMAN, JOHN L. 1984. The
behavioral ecology of carnivores. Ph.D.
dissertation, University of Sussex,
Brighton, England.

HARGRAVE, LYNDON. 1970. Mexican
Macaws: Comparative Osteology and
messi
University of Arizona Press, Tucson.

HASSIG, ROSS. 1985. Trade, Tribute and
Transportation: The Sixteenth Century
Political Economy of the Valley of
Mexico. University of Oklahoma Press,

Norman.

HOPKINS, MARY R. 1992. Mammalian
remains. In Artifacts from the Cenote of
Sacrifice, Chichen Itza, Yucatan: textiles

Memoirs of the Peabody Museum of
Archaeology and Ethnology, Harvard
University, 10, (3): 369-385.

KERR, JUSTIN and BARBARA KERR. 1989.
The Maya Vase Book. Volume 1. Kerr
Associates, New York.

KLEIN, RICHARD and KATHRYN CRUZ-
URIBE. 1984. The Analysis of Animal
Bones from Archaeological Sites.
University of Chicago Press, Chicago.

LYMAN, R. LEE. 1982. Archaeofaunas and
subsistence studies. P 31-393 in
Advances in Archaeological Method
and Theory, Volume 5, M. Schiffer
(editor). Academic Press, New York.

soto ot NO, 5. and V. RICO- ee

Fics of the white-tailed dene
(Odocoleus virginianus roc by the
Maya of central Yucatan, Mexico.
Journal of Ge itors ree 175-183.
MORLEY, S. G., G. W. BRAINERD and
ROBERT J. SHARER. ong pan Ancient
Maya. Stanford University Press,

Stanford.
OLSEN, STANLEY J. 1968. Mammal

Harvard university, Cambridge.

236 BALLINGER and STOMPER

PACKER, C. 1986. The ecology of sociality
in felids. Pp. 429-451 in aa a

>
WD
'S

io)

=n
WM.
°

Gi,
=
ies]
<

ae
fr

oo
_

ie)

oe Princeton University Press,
Princeton.
POHL, MARY D. 1994. Late Classic fauna
an

oo — consumption. Pp. 459-
rest, and W; L
Museum of peenese is and Ethnology,

Pp. 146-174 in Excavations at Seibal,
Volume 3, G.R. Willey (editor). Papers
of the Peabody Museum of Archaeology

and Ethnology , Harvard University,
Cambridge.
: 1983. Maya yrenest faunas:

s, cahces,

x d the vin elo

Pp. 55-103 in Civilizations in the ancient
Americas, R. M Leventhal and A.L.
Kolata (editors). — of New
Mexico Press, Albuque

HAM. 1986. cious and
behavior of the jaguar, Panthera onca, in

Belize, Central America. Journal of
Zoology 210:149-59.

RUE, DAVID J. 1987, Early p.snipiies and
early postclassic Maya occupation in
western Honduras. Nature 326:285-86.

SAUNDERS, NICHOLAS J. 1989. People of
the Jaguar: The Living Spirit of Ancient

merica. Souvenir Press, New York

SCHELE, LINDA and = DAVID
FREIDEL.1990. A Forest of Kings: The
Untold Story of Ancient Maya. Wm.
Morrow & Co., or.

and MARY | ELLEN MILLER.
1986. The Blood of Kings: Dynasty and
Ritual in Maya Art. Fort Worth Kimball

Middle
pi ie exploitation of animal
populations. Papers of the Museum of
peeneeobes No. 57, University of
Michigan, Ann Arbor.

Vol. 20, No. 2

SPINDEN, Bile 1975. A Study of
aya art. Reprint. Dover Publications,

New Yo “

STROMSVIK, G. 1941. Sub-stelae Caches at
Copan and Quirigua. Carnegie
Institution Washington,
Contributions of Anthropology and
History, No ca arnegie Institution,

Washington

SUNQUIST, M. E and F. C. SUNQUIST.
1989. Ecological restraints on predation
by large felids. In Carnivore Behavior,
Ecology and Evolution, J. L. Gittleman
(editor). Cornell University Press,

Ithaca.

TOZZER, ALFRED M. 1941. Landa’s
Relacién de las cosas de Yucatan. Papers
of the Peabody Museum, 28. Harvard
University, Cambridge.

—__——. and M. ALLEN GLOVER. 1910.
Animal figures in the Maya codices.
Papers of the Peabody Museum 4(3).
Harvard University, repre =

VON ENDT, D. W. and, D. J. ORTNER.
1980. Experimental sr of bone size
and temperature on bone diagenesis.

Journal of Archaeological Science 11:

9-53.
VOORHIES, BARBARA. 1989. Ancient
Trade and Tribute. University of Utah
Press, Salt Lake City.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 237

El Bosque Mediterraneo en el Norte de Africa. Jestis Charco. Agencia Espajfiola
de Cooperacién Internacional, Madrid (Avenida Reyes Catélicos, 4. 28040
Madrid, Spain). 1999. Pp. 370. Many color and black-and-white photographs,
maps, figures, tables. No price quoted. ISBN 84-7232-825-2.

This magisterial work is far enough outside the normal scanning pattern of
Western Hemisphere readers that it needs some attention here. Charco has pro-
duced one of the most splendid studies of a Third World region’s forests and its
conservation problems. North Africa still has large forested areas in the Atlas
Mountains and a few scraps remain in the coastal lowlands of Morocco, Algeria
and Tunisia. However, the vast majority of the historic forest area is gone, almost
all the rest is degraded, and deforestation is proceeding.

The forests of the Atlas are largely evergreen (cedar, pine, a few fir) higher up
and largely oak lower down. The oaks are diverse, each species having its preferred
habitat. S if at low altitudes; Tetraclinis articulata, a strange
cypress-like plant, covers thousands of hectares. Fragments of wild olive forests
and riparian groves exist. Among the most fascinating trees is the argan (Argania
spinosa), endemic to Morocco (almost exclusively in the southwest), valuable for
timber, fuel, forage, and the high-quality oil of the fruit. Its value has protected it,
but not well enough; like all other trees, it is under assault. It grows in a strange
subtropical forest at the west end of the Atlas, where its usual cohorts are other
endemics or are Canary Island species that still have a tiny mainland foothold.

The book is arranged by forest type, as defined by dominant tree. More spe-
cifically, the system is the “phytosociology” of Braun-Blanquet and followers, which
is almost universal in European botany though relatively little known in the West-
ern Hemisphere. For each forest type, common species are indicated — not only
other trees, but also shrubs, herbs, and fauna. Rare and endangered plants and
animals receive attention. Ethnobotanists will find full accounts of the uses of the
trees and of the varying levels of preservation that they have received.

Northwest Africa has known agriculture and stock raising for 7,000 years. For
3,000 of those years, states and cities have been present. The impact on the forest,
especially the lowland forest, has been among the most severe found anywhere in
the world. The Roman Empire exploited the forests and exterminated the native
elephant. After its fall, unstable governments and an emphasis on animal hus-
bandry led to progressive decline. The twentieth century saw the extermination of
the lion, the Nile crocodile, and most other large animals, though a few leopards
may survive. Deer have been kept available by stocking. The larger birds are simi-
larly impacted. Even small animals are often endangered.

On the other hand, Cl : the lands in cues
tion give us a more hopeful picture than the earlier wexaeees works of Mikesell (1961) and
McNeill (1992). He and ff-girt areas
forests still cover thousands of hectares. Astonishing relictive forest turning up
in dofd (Dracaena draco), previously believed endemic to

the Canary Islands, was found in southwest Morocco as recently as 1995.

Charco is emphatic in pointing out that human agency has done the damage.
On the other hand, local systems have developed reasonably good ways of man-
aging much of the landscape. Cultivation between standing trees, nondestructive

238 BOOK REVIEWS Vol. 20, No. 2

cork extraction from cork oaks, rational though thorough utilization of argan trees,
careful terracing of slopes, and many other devices reduced the human impact.
Religious beliefs and sanctions preserved forests in many places. The problem
today is that very rapid population growth has gone along with equally rapid
modernization, complete with roads, big dams, unregulated logging, and the rest
of the litany. Yet economic growth, which would have brought pressures for more
rational exploitation of resources, remains slow and uneven in the region. In addi-
tion, Algeria has been racked by warfare.

Recently, a revisionist view blames desertification on climate change. Charco
provides many photographs of forests surviving in cemeteries, sacred sites, and
reserves, and regenerating with wild abandon in fenced-off plots. It is clear that
the damage is indeed the result of logging, firewood collection, and, above all,
stock rearing — not climate change. The result, in loss of valuable plant products
alone, has been catastrophic. To this one may add the loss of topsoil; much former
forest is now eroded to bare rock. Water runs off where once it seeped into ground-
water. Wildlife and wild herbs are gone. The economy has lost an incalculable
amount of wealth. It is truly staggering to contemplate how much richer the three
countries would be if they had been able to exploit the forests in anything like a
sustainable manner. The time is not too late; protection would allow the forests to
recover, eventually, in much of their former habitat.

To a Californian, this book is thought-provoking (at the very least). The south-
ern half of California is a near-perfect geographic match for northwest Africa. We
too are using our environment in a destructive, non-sustainable way. The same
desertification that affects North Africa is not far off, unless dramatic changes oc-
cur. One wonders if North Africa’s poverty and instability will be duplicated here.

This book is illustrated with I hs, charts, diagrams, and tables
Several color photographs are presented for every forest type. The quality of these
pictures is high. They are taken to show the vegetation and its fate, not just to fill up
space with something “pretty” (as is the case in all too many books). The book lacks an
index, but excellent organization and the many visual aids make up for that.

In short, this is a book that everyone interested in forest use and management
will want to see. Have no fear if your Spanish is limited; the photographs speak
for themselves.

This and many other ecological and ethnobotanical works are available from
the Libreria Agricola, Fernando VI, 2. 28004 Madrid — a bookstore worth know-
ing, especially if (like me) you can order books for your university library.

LITERATURE CITED

MCNEILL, J. 1992. The Mountains of the MIKESELL, MARVIN. 1961. Northern
editerranean World. Cambridge Morocco: A Cultur raphy.
University Press, Cambridge. University of California Press, Berkeley.
University of California Publications in

Geography, 14

E. N. Anderson
Department of Anthropology
University of California Riverside

Journal of Ethnobiology 20(2): 239-265 Winter 2000

PERCEPTION AND MANAGEMENT OF CASSAVA (MANIHOT
SCULENTA CRANTZ) DIVERSITY AMONG MAKUSHI
AMERINDIANS OF GUYANA (SOUTH AMERICA)

MARIANNE ELIAS

cefe-cnrs
1919 Route de Mende
Montpellier cedex 5, Ftance

LAURA RIVAL
Department of Anthropology
Eliot College, University of Kent
Canterbury, Kent CT2 7NS, United Kingdom

DOYLE MCKEY
cefe-cnrs
1919 Route de Mende
Montpellier cedex 5, Ftance

ABSTRACT.—This article presents the ways in which Makushi subsistence
economy and its farming practices, food preparations, cultural knowledge and
social processes have all played a role in bringing cassava varietal and genetic
diversity into existence. After comparing cassava varietal diversity among the
Makushi of Guyana with that found in the rest of Amazonia, the authors discuss
the genetic implications of traditional management and show that evolution in
populations of domesticated cassava results from the combined action of natural
and human sElecHen. Various socio- cultural factors CRSTVIEsnE eve Pree,

particu
aan of seedlings and Sent coming from vegetative propagation, are
examined. The approach adopted, which integrates indigenous botanical
knowledge, elements of plant genetics and ecology, a that diversity
cannot be reduced to a finite stock of well defined, varieties,
but is, rather, a fluid and evolving process by which farm-grown varieties are
continuously gained and lost.

Key words: Cassava, M thot lenta, Makushi A indi Guyana, traditional

agriculture

RESUMEN.—Este articulo presenta la manera en que la economia de subsist
de los Malushi y las practicas de cultivo, la prepararion de las comidas, el
conocimiento cult ido a una importante
diversidad genética y varietal de la yuca. Después de un andlisis comparativo i
la diversiddd de la yuca cultivada por los Makushi de Guyana con la que
encuentra = el resto de * Amazonia, ie articulo saci sobre la aiotleaacile
genétic uestra que que la evolucién
de las sb lackcints de yuca cultivada es la consecuencia de la accién combinada

240 ELIAS et al. Vol. 20, No. 2

de la seleccién natural y la selecci6n humana. Varios factores socio-culturales
implicados en la oe en nl Sage anand los intercambios de estacas entre los
euliivadores se je co planta nacida de semilla y la penis
; siraekee Ea dak gracié

indigenos, de elementos de genética y de ai muestra que la diversidad,
lejos de estar oe aun cuyanty t = de entitades determinadas llamadas
variedades, es a dinamico, por el cual las variedades
cultivadas son continuamente adquiridas y perdiddas.

RESUME.—Cet article présente la facon dont l’économie de subsistance des

Makushi ainsi que les pratiques de culture, la préparation des mets, les

connaissances culturelles et les mécanismes sociaux ont tous contribué a une

importante iia variétale et génétique du caress Apres une analyse
le

d
ariétale c u Nakiwuchi de Guyana avec

celle que l’on trouve dans le reste de l’Amazonie, Varticle discute les implications
geénetiques on pratiques de culture traditionnelles et montre que l’évolution de

manioc domestiqué est la conséquence de l’action combinée de la

sélection naturelle et la sélection humaine. Plusieurs facteurs socioculturels

impliqués dans la sélection, en particulier les échanges de boutures entre les

cultivateurs et Yopposition Sonceptiate entre plante issue de graine et plante

L’approche adoptée, qui intégre

les connaissances botaniques indigénes, des éléments de génétique et d’écologie,

souligne que la diversité, loin d’étre réduite a un stock fini d’entités déterminées

appelées variétés, est au ore un processus fluide et dynamique par lequel

les variétés cultix t acquises et perdues.

INTRODUCTION

While present concerns about the conservation of genetic resources have led
to the reassessment of the human side of plant/people interactions, the question-
ing of the division between natural and artificial mechanisms of selection has
prompted a growing number of researchers to pay particular attention to the ways
in which people affect plant genetics (Salick 1995). These developments, which
have given rise to a new type of economic botany that integrates ecological and
evolutionary dimensions in the reinterpretation of plant domestication, cultiva-

ion and management, have resulted in renewed interest in traditional farming
systems, known for maintaining high levels of biological diversity (Boster 1983;
Salick and Merrick 1990; Salick 1992a; Brush et al. 1994). Traditional farming sys-
tems have so far been studied by geneticists, who have attempted to assess the
level of genetic diversity found in Pont ecHOns | of cultivated plants, incding vari-
eties, and by social scientists, who have car e g urveys.
However, only a few studies have tried to connect both genetic and ethnobiological
aspects of the maintenance of varietal diversity (e. g., Quiros et al. 1990; Viinerer
1991; Salick 1992b; Louette et al. 1997; Souza et al. 1997; Emperaire et al. 1998).

The overall objective of our investigation of cassava (Manihot esculenta Crantz,
Euphorbiaceae) is to apply an integrative approach combining ethnological, eco-
logical and genetic data in order to understand how the biological diversity of this
cultivated plant is traditionally managed and maintained. For this, we have cho-

Winter 2000 JOURNAL OF ETHNOBIOLOGY 241

sen to study bitter cassava cultivation in a Makushi community of the North
Rupununi savannas of Guyana, South America (Figure 1). Some of the results of
our genetic studies are now available (Elias et al. in press); ecological studies are
still in progress. The specific objective of this paper is to document the cultural
knowledge and practices that have brought the genetic diversity of cassava into

N
Fa
4
Pa
2nN7 parima
/
. ATLANTIC OCEAN
‘ Baran
ios
VENEZUELA /
%,, fPrv02
Gp ~ Georgetown
Mazaryp,
f ALinden
\ & £
‘ e ¢
4 - Xe)
‘\ S 9 BY Q
ge ee | = g
‘N | 2
/ } & fi rupuka
/ rs spa@ug
Ti geet *»
\.? Taushida \
A Toka Apoteri * SURINAM
(e)
: \ 6 Annai @
a Mgssar¢ SRewa bs
ne a D as
a *, -y am
\ a ee rrr | hoe
&ethe
Boa Vista ! s
O . 8 § )
\ & iS a i)
oer ks FA wo6
\ © 3
( coe es
BRAZIL Ne as \ vatican
‘ ~ F ie !
‘ os if
i ‘
Pe ia 7a, .
*, * Area where the study was conducted 100 km
Zags s

FIGURE 1.—Map of Guyana, showing the location of the Makushi communities where
the study was conducted.

242 ELIAS et al. Vol. 20, No. 2

existence. This information on the cultural environment, combined with informa-
tion on ecology, is essential to the interpretation of population genetics and
evolutionary patterns (Elias et al. in press).

Bitter cassava, characterised by high cyanogenic-glucoside content of the tu-
berous roots, is the staple crop in the Amazon, where it has been cultivated for
more than 3000 years (Renvoize 1972). Most cassava research to date has been
conducted in the Northwest Amazon (see for example Hugh-Jones 1978; Boster
1984a; Dufour 1985, 1988; Chernela 1987; Van der Hammen 1992; Salick et al. 1997),
where the cultivation and processing of (mainly bitter) cassava have been meticu-
lously described. Rival (in press) presents a comparison of cassava cultivation and
processing between northwest ian groups and Amerindians of the Guyana
shield.

METHODS

Site of study.Our study was mainly based in Rewa, a small Makushi community
of the North Rupununi in Guyana, situated at the confluence of the rivers Rewa
and Rupununi (forest area), 50 km WSW of Apoteri (4°02’ N, 58°35’ W). Marianne
Elias spent 9 months over 3 field sessions in the Rupununi (April 1997, February —
May 1998, September — December 1998); Laura Rival spent 5 weeks (2 field ses-
sions in April 1997 and April 1998); and Doyle McKey spent one month (one field
session in October 1998). In April 1998, there were 162 people living in Rewa (in-
cluding three Wapishana men, six Wapishana women, and two Patamona women),
forming 27 households. Because of its small size and relative isolation, Rewa is
less subject to ial and political r than sava ities. Other
Makushi villages, located in the savannah area, were also visited. Concerning the
farming system, no major differences were found between these villages and Rewa,
which can thus be considered representative of the Makushi farming system in
this region.

Collection of ethnographical data. —Our methods involved observations, participation
in farming practices and food processing, open discussions, structured and semi-
structured interviews and questionnaires. Symbolic and other cultural data were
collected as part of a participatory research programme led by Laura Rival and
involving the close collaboration of two remunerated Makushi women research-
ers. We tried to work with as many farmers as possible; however, three female
farmers (ages: 22, 35, 52) and two male farmers (ages: 41, 50) were identified as
our main informants. As almost everybody in Rewa is fluent in English or Portu-
guese; discussions and interview were conducted in these languages, including
some Makushi words for critical points. Bilingual villagers helped us with people
who spoke only Makushi. One villager in Rewa and two villagers in the savannah
area helped us with Makushi spelling.

Assessment of varietal diversity and of its distribution.—Varietal diversity was assessed
by first asking farmers to bring leaves of every kind of cassava they cultivate, then
by visiting farms and interviewing farmers. To estimate the equitability of diver-
sity distribution among cassava farmers we used the index of equitability, which

Winter 2000 JOURNAL OF ETHNOBIOLOGY 243

is often applied to estimate the “evenness” component of the diversity of biologi-
cal assemblages.
dpm p,

n, om
It is calculated as follows : F = ee , Where P; ao 3 te mn ,and n,;

are the numbers of households that cultivate variety i (respectively, the number of
varieties owned by household 1).

MAKUSHI CASSAVA FARMING TODAY

The Makushi, a Carib-speaking group, live in the Rio Branco-Rupununi re-
gion, which is a region politically divided between Brazil (Roraima State) and
Guyana (Region 9). There are approximately 20,000 Makushi today, of whom 7,000
live in Guyana (CIR 1993). Historical records of Makushi presence in the Rio Branco-
Rupununi region date back to the early part of the 18th century (Riviére 1963;
CIDR 1989; Farage 1991; Hemming 1994, 1995; Santilli 1994). With cattle ranching
expanding at the turn of the century, they experienced increased land shortages
and pressure to work as domestic servants or cowboys, and many of them left the
Roraima hills in Brazil for the Rupununi savannas of what is now Guyana (Farage
1991), where they can still be found, living in relatively small communities, headed
by a captain or “toushau.” Makushi families historically traded manufactured
goods for cassava and cassava derivatives, particularly farine, a meal made of
dried granules of grated cassava (Farabee 1924; Diniz 1966; Farage 1991). White
settlers needed the Makushi as much for their agricultural products as they did
for their labour force, and there is little doubt that such demands modified the
indigenous agricultural system. Such influence accentuated the importance of
horticulture over hunting and gathering, and reinforced the tendency towards
sedentarisation in nucleated villages. If, like = sigs Amerindian groups, the
Makushi had traditionally produced cassava o prepare fermented drinks
for festive occasions, the condemnation of such Solition-Hietal activities by mis-
sionaries, the new trade opportunities, and additional factors linked to interethnic
contacts, led to the utilisation of cassava surpluses to make farine for sale, a trend
which has influenced the choice of cultivated varieties.

Bitter cassava cultivation is central to Guyanese Makushi slash-and-burn ag-
riculture, which is still directed toward subsistence, rather than market, production.
Many varieties of bitter cassava are maintained (Makushi Research Unit 1996),
and the diet is supplemented by other starchy crops such as yams (Dioscorea spp.),
corn (Zea mays L.), sweet potato (Ipomoea batatas (L.) Lam), sweet cassava (Manihot
esculenta), plantains (Musa paradisiaca L.), or vegetables such as pumpkin (Cucurbita
maxima Duch. ex Lam.), shallots (Allium cepa ‘aggregatum’ L.) and onions (Allium
cepa ‘cepa’ L.), or fruits such as watermelon (Citrullus lanatus [Thunb.] Mansf.), as
well as fish and meat (game and cattle).

Agricultural activities are conditioned by climatic and edaphic factors. Wet
(from May to September) and dry (from November to March) seasons are pro-
nounced and interspersed with brief transitional periods. Two factors make
savannah soils particularly unsuitable for crop cultivation. In addition to being
particularly infertile, they are subjected to weather extremes; they are almost en-

244 ELIAS et al. Vol. 20, No. 2

tirely flooded during the rainy season, and parched under a scorching sun during
the dry season. This explains why people traditionally lived in forest galleries
along main rivers, where soils remain moist throughout the year, and used the
savannah mainly for seasonal hunting.

With the rare exception of plantain and cotton farms, Makushi farms are, on
the whole, bitter cassava farms. cree cassava, igh sweet potatoes, plantains,
pumpkin or wat | ter-cropped with bitter cassava, but these
crops are rather marginal. Bitter cassava is consumed daily as farine (u’wi), cas-
sava bread (kai), fermented drinks (parakiri, kasiri and wo’), casereep (kumast),
a black, thick paste used throughout Guyana to cook meat and fish, and, finally,
starch tapioca (imu yanasa) or porridge (Figure 2).

ach year during the dry season, households clear new farms (ranging from
400 m_ to more than one hectare) in old fallows or secondary forest, that is, on
land that has already been used in the past to cultivate cassava. As the main prob-
lem in farm management is protection of crops from predators and fluctuating
climate, risk aversion, particularly the risk of flooding and of leaf-cutter ant inva-
sion, seems to play a greater role than soil fertility in the choice of farm sites.

Clearing a new farm (mi? ya’ti) is considered hard work, and is exclusively a
male activity. It was traditionally performed by a man with the help of his rela-
tives and friends (mayu), in exchange for food or drinks (parakiri or wo’).
Cooperation in forest clearing now tends to be restricted to immediate kin, for
example, father and son. New farms are burnt (mii po’ti) in March or April, just
before the first rains. Fire is propagated with lit kokerite (Attalea regia [Anderson])
fronds, a process difficult to control and which may lead to forest fires and crop

pulled

see white or creamish
roots
scraped
put in water
led
aa sie ne CASERERE
oe kumasi
mixed water
squeezed i
aerate aE
; Starch
night en i arched
" ——» -TAP
water + sqWeezed sifted mu yainasa

starch backed and

away — 2g Ps pieces \
soaked in put i a bucket
ater of water
hed

CASSAVA PARAKARI wo’ drink
EAD arakiri wo’
i

FIGURE 2.—Processing and products made of cassava by Makushi Amerindians.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 245

losses when weather conditions are particularly dry. Ground cleaning (mi? kui
ma), the third stage in preparing a new cassava farm, is undertaken by the entire
household, children included. The male farmer hoes (ya nu mii) the field, and
shapes the soil into mounds (locally known as “banks”), in which cuttings (locally
called “cassava sticks”, kisera in Makushi) are planted (soil mounding, a practice
rather in Amazonia, is also found in the Upper Xingu [Carneiro 1961]).

Female farmers, sometimes accompanied by their children, divide the long
stems of freshly harvested cassava into cuttings. Although they store long stems
in shady places or keep them for weeks, slightly buried in the ground before using
them as planting material, farmers prefer to prepare cuttings directly from a living
plant. Moreover, the selection of cuttings depends on whether a plant is healthy
and whether it has produced a generous harvest of roots. Cutting the stems is a
woman’s work, but planting is usually done by men, although this is not always
the case in other villages (Mary Riley, University of Illinois at Chicago, USA, per-
sonal communication). Three to five cuttings of the same variety are planted
together in each mound, pointing to the west. Although most people are aware
that cuttings planted eastward grow as well as those pointing westward, they still
maintain that a westward orientation protects the plant from the sun and favours
its growth. There may be a connection between this belief and the common Makushi
practice of burying corpses with the feet pointing east, head west, and the face
looking to the rising sun (data collected by Laura Rival in Massara, April 1997).
Another common practice observed by Laura Rival in the field, and aimed at en-
couraging growth and tuber production, is to plant a magical plant (generically
known as bina in Guyana) in the middle of the field, or to chew the root of such a
plant, and spit it over the mounds (see also Makushi Research Unit 1996).

The time it takes a farmer to plant a new cassava farm depends on his avail-
ability, and on how many stems the household has in stock; it thus varies greatly
from farm to farm, and from individual to individual. Some farmers plant cassava
every day for a week, and then stop for two or three weeks before continuing. As
a result, several months can separate the youngest and the oldest plants in a single
farm. This practice, as well as the planting of varieties with different maturation
rates, ensures a constant supply of harvestable roots throughout the year. Cassava
roots are harvested on average nine months after planting (range: 4 months to 2
years), and the crop requires little care during this period. A first crop does not
generally require weeding more than three times, but women may perform spot
weeding every time they harvest roots (both weeding and harvesting are female
tasks).

Roots are harvested when and as they are needed. Stems of freshly harvested
cassava are sometimes replanted in mounds hoed again for what the Makushi call
the “second crop.” More commonly, they are taken to a new farm, where the best
cuttings are selected and planted. The surplus is stored, hidden under kokerite
fronds, or planted in tight bundles at the edge of the farm. Some may be given to
neighbours, relatives or friends. Unused stems stored in bundles also grow and, if
the weather is not too dry, develop tuberous roots nine months after. These roots
are harvested, but the stems are usually discarded, being used as planting mate-
rial only if the farmer is desperately short of cuttings.

246 ELIAS et al. Vol. 20, No. 2

After harvest, first crop farms are planted with a second crop of cassava. Sec-
ond crop farms require greater weeding efforts, and their productivity is lesser.
When harvested, they are not generally hoed again and replanted, except if the
parcel cleared for the first crop is located in primary rainforest or old secondary
forest, where the rich, deep forest soil allows for a third, and even sometimes a
fourth crop. Abandoned farms are soon covered with fast growing pioneer veg-
etation. Old farms are left fallow for two to twenty years, or more. Although farm
land is not privately owned, old fallows are customarily left for the use of the
families who first cultivated them.

As in most Amazonian societies, cassava processing is the responsibility of
women, and food intimately connected with gender specialisation. However, it is
not uncommon to see a man helping his wife making edible products out of bitter
cassava, which contains a high concentration of cyanogenic glucosides and must
be detoxified before consumption. Roots are scraped, and the skin and inner peel
(except in the preparation of parakiri and wo’) removed. They are then washed
and grated. The grated pulp is squeezed in the long plaited cassava sleeve press,
known in Guyana as matapi (tinki). To make farine, women first mix fresh grated
cassava with roots that have fermented for three days, and then squeeze the mix-
ture. The half-dry pulp is then used for preparing the final products (Figure 2).
This process ensures that most of cyanide is removed (Dufour 1989, 1995)

To sum up, the farming system so far described is broadly similar to those
found among other Amazonian bitter cassava farmers (see in particular Dole 1963,
1978; Yde 1965; Diniz 1966; Grenand and Haxaire 1977; Hugh-Jones 1978, 1979;
Carneiro 1983; Dufour 1983; Chernela 1987; Mowat 1989; Van der Hammen 1992;
Grenand 1993; Emperaire et al. 1998). We are fully aware that the farming system
we have studied may be a recent development linked to sedentarisation. We sus-
pect, for instance, that the present shortage of farmland has resulted in sharp
reduction of fallow periods. This shortage, dramatic in the savannah, where farm-
ers have to walk for hours or paddle for several days to reach their farms, is
beginning to affect Rewa as well. Another consequence of sedentarisation is that
farms must last longer, so people tend to select varieties that can stay in the ground
longer, such as, for example, the popular “white man stick” that remains harvestable
two years after planting.

CASSAVA DIVERSITY AND ITS PERCEPTION BY THE MAKUSHI

Cassava diversity in one Makushi village — Cassava varieties are distributed in farms
either in a structured pattern (i.e. planted in contiguous monovarietal patches), or
at random, the latter distribution being more frequent. Every variety has a Makushi
name, of which the English translation is usually known. In Rewa, we collected 86
different names of cassava varieties (Table 1), corresponding to 76 varieties (while
some varieties given different names by different farmers, f knew
several varieties under a single name). Varieties that farmers were unable to name
were not included in the survey. Each household owns on average 16 varieties
(Table 1). Varieties differ in their frequency of representation among households.
While some are shared by almost all farmers (for example, paranakiri piye and

Winter 2000 JOURNAL OF ETHNOBIOLOGY 247

kuraatuma piye), others are owned by only a few, or even by a single person (for
example, oronki piye and dominko ye). The equitability index for cassava variet-
ies is 0.63 (it would be equal to 1 if all varieties had the same frequency in the
village), and 0.52 for households (it would be equal to 1 if all the households had
the same number of varieties). No one in Rewa has an accurate or exhaustive know]-
edge of the varietal diversity present in the village.

Sweet cassava is planted either in the farm, together with bitter varieties (but
the two types of roots are not mixed, as sweet cassava is not cultivated for the
same purpose), or in the kitchen garden next to the house. Farmers say that sweet
cassava roots develop more slowly than bitter ones, and that the plant grows taller.
Four different types of sweet cassava, all called kana as opposed to kise (bitter
cassava), are grown in Rewa. They all have white roots, and are generically re-
ferred to with English names (“brown stick,” “four months,” “eighteen months,’
or “white stick”). The lack of a proper Makushi name confirms that sweet cassava
is not regarded as “real” cassava; like yam or sweet potatoes, its place in Makushi
diet is secondary (Makushi Research Unit 1996). Semi-quantitative tests based on
a colorimetric method using alkaline picrate (Williams and Edwards 1980) con-
firmed that all varieties designed as bitter by the Makushi have a high cyanide
content, whereas so-called sweet varieties have a low cyanide content (Elias un-
published data, see also Dufour 1988).

Perception of bitter cassava diversity. Makushi lore includes several myths referring
to cassava and its origin, as well as to Cassava Mother (kisera yan), the master
spirit that owns the plant, looks after its well being, and ensures good harvests
(Rival in press). Although myths are complex cultural representations that cannot
be reduced to one single dimension or message, Makushi myths relating to cas-
sava clearly convey the idea that this plant exists first and foremost as a cultivated,
that is, a domesticated or cultural plant, whose inalterable blueprint is the master
spirit Cassava Mother, and whose origin relates to the transformation of a human
body. Wild cassava! is represented as a degenerated cultivar escaped from gar-
dens, which has stopped producing tuberous roots, because it now grows in
non-cultivated (i.e. non-cultural) spaces, such as hill tops in the savannah.

Taxonomy.—Savanna-dwelling farmers of mixed origin (i.e. of Makushi and Black,
locally known as “dougler”), and Makushi who have no direct knowledge of cas-
sava cultivation, recognise only three kinds of cassava: yellow, cream and white
varieties. They differentiate them by color and use, as cassava bread is made with
the roots of white varieties, and farine with the roots of yellow varieties. The first
morphological trait mentioned by a farmer is the color of the root, and its inten-
sity. Three main categories are actually differentiated: yellow types, creamy types
(which non-farmers confuse with yellow types), and white types. This classifica-
tion, also found among other groups (Emperaire et al. 1998), plays a determinant
role in evolutionary terms. This may explain why Makushi people tend to under-
estimate their varietal diversity, and why outsiders are only aware of differences
in root color.

Named varieties constitute the second, and more ambiguous, level of classifi-
cation. Varieties are named after animals, plants, objects, dishes, qualities, and

Makushi names

English names

AM NE WE VA DP JM TA RE JE MA AE DH FW JA CS EK NI WA JH PH CH LE ZP HS

ainis piye
aknes piye
akuriu ye
amilton ye
amo’ko piye =
danal piye
amuru piye
ankela piye
anra piye
dominko ye
eti piye
eni piye
esekwipo ye
isman piye =
sandra piye =
amuru piye ?
kaima piye =
eri piye
kanaima ye
karmani piye
kasiri piye
kediam piye

kini’ piye (2 types) =

reni piye
ko’ko piye
kraiwa piye
kari’na piye
kompani piye
krompi piye
kumia ye

Inez stick
agnes stick
agouti stick
Hamilton stick
grand father st. =
Danal stick
thick stick
Angela stick
crane stick
Domingo stick
Eddie stick
Eni stick

thick stick ?
pumpkin stick =
Ely stick

jombie stick
Carmani stick
kashiri stick
Kediam stick
dry stick =

Reni stick

grand mother stick

brazilian stick
Carib stick
company stick
?

fish stick

x
x
4
x x
xX x x
x
»«
x
x
x x x
x
x
x x i &
»4
x

Meron em

‘saljtuey FZ Suoure UOTNQL]sIp I1ay} pure “eMay UT puNo}j saateA PARSSeD— [ FIAVL

Makushi names English names AM NE WE VA DP JM TA RE JE MA AE DH FW JA CS EK NI WA JH PH CH LE ZP HS
kunani piye fish poison st. x x x
kuraatuma piye caiman stick

2 types) ok Se eX ee x oF — 8 Se ee eS Cee ox x
kurari piye curral stick x ox x x x =x x * i ee a
kuraswa piye = Crash Water st =
selia piye = Celia stick =

waimko piye Waimko st. ae. x x xX oR x x x x x x
lio piye Lio stick x x
mai piye bitter stick x x x x x x

maka piye maggah stick x

makarpon ye Makar pond stick x

marasi piye marudi stick x x x x

mauri piye Mauri stick x x x
meekoro piye black man stick x x x ek xX x x x

mepriko ye Elfrida stick x . *

omano piye Omano stick x

oronki piye bamboo stick x

paapa ye father stick x * x

pakaima ye buffalo stick x x

pali piye = Bali stick =

kaiwan piye fat boy st 5 ata aS x x x xo oe x Be
papiro ye Pablo stick x x Xx x Xx x x x x
parakiri piye parakari stick x x x x

paranakiri piye white man stick X XX MOK KX KX KM. K Mee MM ROR ee ke ey lk
paranakiri piye white man

itakon ye = stick cousin =

naman piye Naman stick x x < oN x x x x x
paranakiri piye white man

perurupe stick seedling x x
pinkiu ye peccarie stick ae x Xx Xx x x
piraun piye Brown stick x

pirikwa piye = bird stick =

ipo ye sweet stick x << xs x ee x x x x “os x

(panunuod) T aqeL.

AM NE WE VA DP JM TA RE JE MA AE DH FW JA CS EK NI WA JH PH CH LE ZP HS

Makushi names English names
pirori piye Pearl stick x
prona piye brown stick x x
reki piye thick stick *
rikitun piye black stick x
rora piye green stick x Xx x
sakari piye Zaccharie stick Xx
saketa piye Sagda stick x
sapri piye fine fish stick x x se x
seruak piye 3 months stick x x
siya piye Shea stick x x x x x x Xx
siment piye cement stick pe Seay X X Xe ex KUM MS Se ee x x
siwal piye Sea Wall stick x »
sona piye Jona stick x x x
supra piye cutlass stick Xx x x ee ae Mook ee x
tare’kiya pimoi water turtlle egg
piye = suyu ye = stick = yellow stick =
five month stick = 5 months st =
u’wi piye ? farine stick ? ee ae cae x * a x x
tari ye black potatoes stick Xx
tikiri piye ? x Xx
toni piye Tony stick x
urakasa ye pigeon stick x x
usariu ye deer stick x
u’wi piye farine stick x MM. Xx x Rk Ko x Xx x
uyara piye macaw stick
(cassava heart) x x
waakiri piye I love stick x
walakise piye Walax stick x
wo’ ye drink stick Lk eee ae ¥

‘(panuryuod) [ a[qey,

total number of varieties : 76

10) 4250 4 16s 23

14. 22. oe 22. 12. 21. 9. 2 ee a I

oat 30 2 Oe

Winter 2000 JOURNAL OF ETHNOBIOLOGY Zl

according to the place of origin or the person from whom cuttings were obtained.
Names derived from animals, plants, objects, dishes, and qualities are said to be
more authentic and traditional than those based on Christian names and toponyms,
with which they often overlap. For instance, the real name of “Ely stick” is “pump-
kin stick,” but as the farmer who received it from Mrs. Ely had forgotten the real
name, she chose to call the new variety after Mrs. Ely. Almost all the inhabitants of
Rewa speak both Makushi and English, but bilingualism does not seem to affect
the naming system, given that all Makushi names have an English translation. We
were told that it is the husband’s, rather than the wife’s, responsibility to give
names to new varieties. On the whole, the Makushi naming system is similar to
those used by many other traditional Amazonian groups (Salick et al. 1997;
Emperaire et al. 1998).

Visits to various farms with different informants allowed us to confirm that
farmers use the taxonomic classification outlined above fairly consistently. Eight
times out of ten, they agree on what name to give to a particular variety. However,
since the varietal make-up of each farm is different, a farmer visiting a neighbour’s
farm may be unable to name some of the varieties. We also detected several cases
of synonymy and of homonymy (Table 1). The first genetic analyses, which are the
best tools to detect such events, have confirmed these observations (Elias et al. in
press).

Accuracy of taxonomy is of great importance in determining evolutionary
pressures on cassava, since individual human selection acts on taxonomic units,
locally identified as varieties. The fact that generic terms are increasingly used (at
least in our informants’ perception) may lead to an impoverishment of the highly
specific variety vocabulary. For example, a really yellow variety is now commonly
referred to by the generic term “farine stick” and a variety hard to grate by the
term “dry stick”, while an increasing number of varieties are called by the names
of those from whom they were obtained. As a result, a farmer may have two or
three es of “farine stick,” “dry stick,” or “Bali stick”. However, most farmers
can still differentiate among homonymous varieties. They can, for example, dis-
tinguish the “dry stick” with broad leaves from the “dry stick” with narrow leaves,
and explain these differences accurately to the botanist. But we are not entirely
certain that they are able to distinguish two morphologically close varieties with
exactly the same properties, or coming from the same place, if they call them by
the same name. The consequences of such taxonomic impoverishment are dis-
cussed below.

Recognition process.-The Makushi seem to identify their varieties all at once, just as
they or we identify people we know. Several farmers told us that they cannot name
a variety before checking the appearance and quality of the roots. This instanta-
neous and integrative recognition process, which appears to depend entirely on
the plant phenotype, was first identified by Shigeta (1996), who calls it “face-to-

face recognition
Our hypothesis i is s that the qecognition process operates on the basis of an un-
conscious hi rs. This is best illustrated when a plant

is hard to identify, because of phenotypic plasticity. The overall aspect of the stem
is the first character observed, and the most important for defining the identity of

252 ELIAS et al. Vol. 20, No. 2

the plant. The next salient traits are, in decreasing order of importance, the aspect
of the leaves, the color of the petiole, wee — aspect and color of the root, and,
finally, the flowers. These ol t with those of Boster (1985)
and of Emperaire et al. (1998).

While certain varieties are preferred for particular uses, there is much func-
tional redundancy among varieties, and there appears to be no practical reason
for having so many varieties of bitter cassava. The Makushi seem to enjoy diver-
sity for its own sake, and more for aesthetic and recreational reasons, or out of
scientific curiosity, than for security purposes. Our obser te those
of Carneiro (1983), who noted that the Kuikuru (Brazil) cantink explain why they
have so many varieties; they just have them. Makushi farmers also say that they
do not know, that they simply like seeing many different types of cassava in their
farms. Like collectors, they seek to cultivate as many varieties as possible. They
are driven by a deep-seated curiosity that pushes them to acquire new types con-
tinuously, “to try them out.” This readiness to experiment has real implications
for the structure and dynamics of cassava diversity.

THE DYNAMICS OF DIVERSITY

Cassava diversity is not static, and farmers experience intentional or uninten-
tional losses of varieties, for which they compensate by acquiring new types. As
stated before, acquiring new types can also be motivated simply by curiosity. We
review here environmental and human selective factors involved in the dynamics
of diversity.

Environmental selective factors.-Severe droughts and floods are recurring climatic
conditions which can lead to the loss of varieties, and to which Makushi people
have learned to adapt. According to our informants and our own observations,
exceptionally dry weather such as that caused by El Nifio from August 1997 to
April 1998 kills recently planted and growing cassava alike. Drought affects root
production, and can destroy all the propagation material. Varieties present in low
densities, or owned by a few farmers only, are more likely to disappear. The
Makushi have therefore developed special planting strategies to protect the cut-
tings from drought, and, to a lesser extent, flood. Farmers, for example, are careful
to keep in the soil cuttings which, at first sight, look dried-out but which will
almost certainly sprout again with the first showers. Mounds are reshaped with
taller profiles, and replanted with six or eight (instead of four) cuttings to maximise
the chances of having at least two well-developing plants in each. The general
strategy is to prioritise the long-term reproductive cycle. Rather than saving
harvestable roots, farmers try to preserve the stems as “germplasm banks”. For
this, they look for swampy areas, which are under normal circumstances flooded,
and transport their best stems -sometimes over great distances- from their farms
to the swamps, to stock them in large bunches there until it rains (Rival 1998).
Whereas cassava varieties in many parts of South America and Africa are seri-
ously affected by diseases (cassava bacterial blight, Boher and Verdier 1995; African
cassava mosaic virus, Fauquet and Fargette 1990), varieties in Rewa do not seem
to suffer much from disease. However, in 1998, a new plague locally called “white

Winter 2000 JOURNAL OF ETHNOBIOLOGY 253

disease” (turere) caused by a small white insect (Aonidomytilus albus, Hemiptera:
Diaspididae) which kills all affected plants struck all the high forest farms located
three hours downriver from the village. One family with downriver farms lost all
its cassava varieties.

Farmers seem to have many problems with, and often complain about, herbi-
vores (Table 2). Leaf-cutter ants Atta spp. (“acoushi ants”, kuinan), which often
destroy entire farms, and are very difficult to get rid of, are identified by farmers
as the most dangerous predator of cassava. Farmers may decide to abandon a
farm altogether, rather than try to eliminate ants. Other herbivores feeding on
leaves usually act sporadically, and rarely kill plants. But their predatory activities
itla aff fect plant vigour and root , even caus-
ing roots to become watery. Plants that are frequently attacked by herbivores may
not be multiplied. The “carelessness” of peccaries and tapirs, which often trample
and destroy everything when feeding in a farm, may also lead to variety loss.
Farmers do not seem to be aware of any particular herbivore/variety association,
and information collected from different farmers regarding perception of varietal
resistance was contradictory, which could reflect either strong soil/ genotype in-
teraction for resistance, great variability, or lack of interest in this feature.

Human selection.—As a cultivated crop, cassava is under intense human selective
pressure. Vegetative propagation allows for the instantaneous selection of pheno-
types of interest. Farmers, who can decide whether or not to keep a variety, to a
certain extent also control the relative frequency of each variety. If there is no con-
scious selection for resistance te PRES, diseases, or climatic conditions, some
varieties will , whereas others will decrease in frequency.
As far as we were able to observe, there i is no directional selection on maturity rate
either, since farmers need both precocious and late-bearing varieties in order to
minimise risk and secure a continuous supply of harvestable roots. As scientists,
we were even asked jokingly to invent a new and improved variety that would
have such a broad maturity rate spectrum that there would be no need to grow
any other variety.

TABLE 2.—Main wild herbivores feeding on cassava in Rewa.

class of species Makushi attacked importance
herbivore name parts of damage
Insects Atta spp. (leaf-cutter ants) kuinan leaves +++
Erinnyis ello (larvae of asphinx moth) —arari leaves ++
Birds unidentified anakwa leaves +
Mammals Mazama americana (brocket deer) usari leaves +
Dasyprocta agouti (red rumped agouti) akuri roots +
Dasyprocta prymnolopha (black rumped
gouti akuri roots +
Agouti paca (paca) urana roots +
Tayassu pecari (white-lipped peccary) pinki roots +++
Tayassu tajacu (collared peccary) piraka roots +++

Tapirus terrestris (tapir) waira roots the

254 ELIAS et al. Vol. 20, No. 2

Taking up Boster’s (1984a) hypothesis, McKey and Beckerman (1993) suggest
that the productivity criteria include many other agronomic features. Human se-
lection for productivity would combine with natural selection for survival and
resistance. Like other Amazonian cassava cultivators (Boster 1985; Chernela 1987;
Salick et al. 1997; Emperaire et al. 1998), the Makushi consciously select varieties
primarily for their productivity. For instance, every household in Rewa has the
paranakiri piye (“white man stick”) variety, which produces high yields. It is the
most frequent variety in 17 of the 24 farms included in our census. Yield is, how-
ever, highly variable among farmers and crops. Some varieties can be very
productive in the first crop, but do poorly in the second, whereas others, which do
not have a very high production in the first crop, do well in the second.

The color of the root is another important criterion for selection. The main
dish in traditional Makushi diet is cassava bread made with white varieties. But
when ranchers settled in the Rupununi in the 1930s, demand for farine, a staple
with which they fed their workers, soared. Farmers responded to the demand by
preparing increasing quantities of farine, which is made with yellow varieties,
and they also increased the proportion of farine in their own diet, thus multiply-
ing yellow varieties at the expense of white ones. This shift from cassava bread to
farine, and from white to yellow varieties, is widely recognised as a recent phe-
nomenon (see in particular Yde 1965). People nevertheless say that they like farine,
which is easy to store and process, and that they want yellow forms, hence their
eagerness to acquire yellow varieties. Four “really yellow” varieties are found lo-
cally, of which at least one is cultivated in great quantities.

Varieties are also selected for their culinary or “processability” properties, but
these criteria vary greatly from farmer to farmer. Some women we interviewed
prefer watery roots, which are easier to grate, whereas others liked dry roots, richer
in starch. Aesthetics also play a role in selection, and a variety with an unusual
combination of morphological characters undeniably arouses the farmer’s curios-
ity and interest.

In addition to selecting varieties according to these criteria, farmers also pro-
tect rare varieties, thus encouraging frequency-dependent selection. A variety is
rarely discarded, even if it is not very productive. Low-yielding, rare varieties are
simply kept at low density (i.e. one or two mounds per farm), and this is consid-
ered enough to prevent their loss. Farmers explain that they do not like losing
varieties, for a “bad” variety sometimes becomes “good” under different condi-
tions. Rare varieties, however, become more vulnerable to loss by direct or natural
selection. Finally, farmers lose varieties, not only because of environmental pres-
sures or deliberate choice, but also because of bad space management. As each
farm is limited in size and contains a finite number of mounds, there is sometimes
no place left for experimental planting, and cuttings of marginal varieties may no
longer be available when space becomes available again. If this occurs, the farmer
has lost this variety. In sum, conscious human selection on cassava varieties is
rather lax; it acts primarily to preserve diversity, rather than to maintain or aug-
ment some particular desired traits.

The exchange of cuttings and the t of volunteer plants grown from
seedlings, which represent two means of acquiring new varieties or recovering
lost ones, greatly contribute to the maintenance of cassava diversity.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 255

Exchange networks. Farmers can recall the origin of all the varieties they have
in their farms, no matter how far back they have to go. In former times, when the
rule of uxorilocality and bride service was more generally applied, a young hus-
band was expected to cultivate with his wife the farm of his parents-in-law, until
his father-in-law gave him the permission to set up his own farm. The young hus-
band would then receive a share of cuttings from his father-in-law, a stock which
he complemented with cuttings brought from the farm of his own parents. The
young wife would receive cuttings from her mother, sisters, mother-in-law and
sisters-in law. The couples we interviewed always specified which varieties were
brought into the conjugal farm by the husband, and which by the wife.

It is perhaps because a transfer of cuttings is initiated by request from a farmer
in need that farmers so accurately remember the origin of the cuttings they culti-
vate. The most general term used to express the idea of transfer is “borrowing.” A
farmer short of planting material or willing to try a new variety asks another farmer
for stems, which he or she will “repay” at a later stage with stems of a different
variety, or with farine or parakiri made with roots from the borrowed varieties.
Delayed reciprocity has been observed among other Amerindian groups, such as
the Aguaruna (Boster 1986). There is no specified rule on how long after having
“borrowed” a farmer must “give back.” The only time when transfers are not re-
ceiver-initiated, and when there is no pay back, is when a farmer donates cuttings
to compensate damages, such as the destruction of part of a neighbour’s crops
through uncontrolled fire. The exchange of cuttings between closely related kin
and neighbours whose farms are contiguous and who still practice a form of shared
labour (mayu) is less formal and generalised, in the sense that strict reciprocity is
not an issue. Only very o lly will a farmer acquire cuttings of a rare variety
by paying with cash, a cutlass, a bicycle or any other trade item. This type of ex-
change is more likely to occur between unrelated farmers who do not live, and
have no family connection, in the village.

Two types of exchanges, one “massive” and the other “occasional,” can be
distinguished. Farmers who need large amounts of planting material for a new
farm borrow cuttings from just a few varieties, which they usually already pos-
sess, but not in sufficient quantity. Those who want to try out a variety they do not
have in their farm ask for a stem or two while visiting neighbours, friends or rela-
tives. Farmers, who are always keen to acquire new varieties, carefully multiply
the trial cuttings until reaching the desired density. The closer the locality from
which a new variety originates, the more likely the retention of its real name will
be. In contrast, a variety coming from a distant locality may be called after a top-
onym, for example “Shea stick” or “Crash Water stick.”

In contrast with Chernela (1987), who has reported exchanges over an area
465 km wide, most Makushi cutting exchanges (and all massive exchanges) occur
within the village community, in particular among immediate neighbours, who
are usually related through descent and marriage (unpublished data). Asa result,
some families never interact, while others are continuously exchanging cuttings.
Frequent exchanges between the same families are expected to lead to taxonomic
homogenisation (Boster 1986). According to informants, cuttings were not traded
across ethnic boundaries in the past, and even inter-village exchanges were -and
still are- restricted, which, of course, does not exclude acquisition through war-

256 ELIAS et al. Vol. 20, No. 2

fare, looting, and the taking over of another tribe’s old farms. Some farmers care-
fully avoid asking for cuttings, as they derive great prestige from relying on their
own stocks, while lavishingly giving away to borrowers. Others, who keep their
farms at a distance, never invite visitors, and share only reluctantly, are consid-
ered “stingy.” Yet others do not dare asking for coveted cuttings, which they
surreptitiously pick from the farms of neighbours or hosts (a behaviour which
may lead to generalised theft in times of drought and starvation). To summarize,
while some families always experience a deficit of planting material, others al-
most never borrow cuttings. And whereas some share their surpluses generously,
others prefer to let their unused bundles of cuttings dry out. When examining the
social networks underlying ive exchanges, we soon identified “source” fami-
lies, whose efficient management of cassava production not only prevented them
from experiencing shortages, but also ensured that they usually disposed of suffi-
cient planting material to give away. “Sink” families, in contrast, managed their
stocks poorly, often ran out of cuttings for their new farms, and heavily depended
on borrowing. “Source” farmers are proud and respected community members,
with a higher status than “sink” farmers.

Tepuru piye. The cassava plant grown from spontaneous seedlings. New vari-
eties are often acquired through the incorporation of volunteer cassava plants
grown from seedlings, the products of sexual reproduction, that appear spontane-
ously in farms (Makushi Research Unit 1996). Although cassava has long been
propagated vegetatively, it has retained its ancestral capacity for sexual reproduc-
tion. In Makushi farms, most varieties produce flowers, and these produce fruit
which dehisce at maturity, dispersing their seeds before the plants are harvested.
Seeds are projected on the ground by exploding capsules, and then secondarily
dispersed by ants (Elias and McKey in press) and perhaps by other mechanisms.
Although seed physiology is poorly documented, variable dormancy has been
reported (N. Morante, Centre Internacional de Agricultura Tropical, Cali, Colom-

ia, persona ication), and scattered anecdotal observations seem to show
that preserving a vegetation cover prevents seed germination.

Seeds germinate whenever a new farm is cut in an old fallow. Spontaneous
seedlings, tepuru piye (from tepuru, seed) are found in places that were cultivated
as long as 35 years ago. One farmer even found a tepuru piye plant in a farm he
had just cleared in a high forest location three hours down river by canoe from
Rewa, which was, according to oral tradition, last cultivated by a group of Caribs
more than fifty years ago. He decided to multiply it, and called the new variety
kari’na piye, “Carib stick.” The surprisingly long dormancy suggested by this
observation and by Amuesha comments (Salick et al. 1997) can be explained by
the presence of seed banks along with seed propensity for long survival (Elias and
McKey in press). Dormancy of seeds from a seed bank must be broken by cues.
Many light-demanding tropical pioneer species respond to changes in light qual-
ity or soil temperature regimes following removal of vegetation. In others,
germination is enhanced by the direct (high temperature) or indirect (higher con-
centration of minerals in ash) consequences of fire (Garwood 1989). Some of the
cues that break dormancy of wild Manihot species and cassava seeds have been
identified; they involve high temperature (Ellis et al. 1982; Nassar and Teixeira

Winter 2000 JOURNAL OF ETHNOBIOLOGY 257

1983) and scarification. As seedlings usually start growing before the first cuttings
are planted, they do not suffer much competition from planted cassava. We have
yet to gather information on the survival rate of seedlings. Up to 400 seedlings
have been found in one young crop of 425 m_ (Elias and McKey in press), which is
the highest density ever reported in the literature (Emperaire et al. 1998 report one
or two seedlings per farm).

Farmers pay special attention to plants grown from spontaneous seedlings,
which they have no difficulty identifying. Although they may compete with, and
affect the growth of, planted cuttings, seedlings are only rarely weeded, and usu-
ally left to grow until they reach maturity. Like common varieties, they are harvested
and their roots processed. If the farmer is satisfied with the yield, the color of the
root, and any other characteristic, the stem is divided into cuttings which are soon
replanted and multiplied in a specific location. If the tuberous roots are found
unsatisfactory, the stem is generally discarded, although farmers keen to experi-
ment clone it, hoping that it will become more productive after one cultivation
cycle. Tepuru piye are thus multiplied over generations of planting, and, since they
often present novel combinations of morphological characters, are usually treated
as new varieties. Naming these new phenotypes is not easy. Audacious farmers
create appropriate names, such as, for example, Mr Nathaniel Edwards in Rewa,
who, without hesitation, named a “very yellow” variety from a spontaneous seed-
ling “I love stick” (waakiri piye). Most farmers, however, take the view that all
varieties are pre-existent, each with a fixed name, and SO prefer not to invent new
names. Consequently, they keep the non-sp 1 piye, which
they apply to different phenotypes.

When the characteristics of a seedling closely match those of a known variety,
the seedling, assumed to have grown from the seed of this particular variety, is
assimilated to, and named after it (adding the term perurupe, such as in paranakiri
piye perurupe, “seed of white man stick”). Seedlings which are considered en-
tirely identical to a known variety are treated exactly like any other member of
this variety. In the field, we were able to document four cases of such pseudo-
reappearance of a known variety from a seedling. However, we suspect the
frequency of complete assimilation of seedlings to known varieties to be much
higher. Whereas it is easy to put one cutting aside in order to multiply a given
variety, this is not the case when a great number _ cuttings from the same variety
have to be distinguished, hence tl the onfusion and unintentional mix-
ing of cutti rom seedlings with cuttings from pre-existing varieties.
Farmers may not tii the seedling origin of particular stems, if the morpho-
logical characteristics of the latter are not sufficiently distinctive. Some of the
farmers we interviewed were sacra? aware of making mistakes when identify-
ing varieties, and of t including new seedling pl thin known
varieties. The assimilation of erik from seedlings into known aeetern is one of
the reasons for intravarietal genetic diversity in cassava grown by the Makushi
(Elias et al in press), which is associated with intravarietal phenotypic diversity.
Adding to the frequency of such confusion between similar varieties is the fact
that the phenotypes of a given variety also vary with environmental factors.

Obtaining quantitative data on the incorporation of sexually produced indi-

258 ELIAS et al. Vol. 20, No. 2

viduals into the stock of material for vegetative propagation was extremely diffi-
cult. One farmer, a highly respected widow with a level of education above average,
told us that in the high-forest farms cultivated by her husband and herself over a
period of 15 years, cuttings from seedlings came to represent 30 per cent of the
varieties grown, against 15 per cent for the varieties with which they planted the
first farm in the area, and 55 per cent for the varieties they acquired through ex-
change with neighbours and relatives. We do not know whether this case is
exceptional or representative. One would expect to find many varieties derived
from seedlings in every farm, given the high density of seedlings growing in cas-
sava farms, and the information volunteered by farmers, who say they multiply
most of the tepuru piye they find in their farms. However, when asked for the
varieties previously multiplied from tepuru piye, farmers only name on average
one or two (the maximum was eight). Such a contradiction can be explained by
the aforementioned phenomenon of assimilation of tepuru piye into other variet-
ies. Another possible explanation is that these varieties, which are represented by
a small number of individuals in the first generations, are easily lost. Besides, their
identification requires a special memorisation effort on the part of farmers.

edlings are not only the sole source of novel genotypes; they are also, at
times, the only source of cuttings. In 1950, as reported by many old people, farm-
ers in the savanna experienced a very severe drought, which lasted almost two
years. Their crops dried out, and they soon ran out of planting material, which
they recovered by clearing old farm locations, where seeds germinated, providing
them with new stocks of cuttings.

Local wisdom, according to which cuttings produce individuals simil hose
from which they come (i.e. identical genotypes), is contradicted by the behaviour
of tepuru piye, which rarely reproduce the characteristic features of the particular
varieties from which the seeds originate, but display instead novel or unusual
combinations of traits, due to recombination in a genome that is very heterozy-
gous (Colombo 1997; Lefévre 1989), a property enhanced by allogamous
reproduction (Rogers 1965). Makushi farmers, who find the unusual trait combi-
nations of seedlings quite puzzling, explain them by presuming that a seed from a
given variety, because of its small dimensions, is influenced by other varieties af-
ter falling to the ground, and denatured.

This perception may in turn explain why they never plant cassava seeds, but
propagate clones of tepuru piye, despite their knowledge of seed planting, a tech-
nique commonly used to grow crops such as papaya and corn. It may also explain
why farmers never give away cuttings from spontaneous seedlings, but only cut-
tings from third or fourth generation clones, which they have replanted in various
types of soil, and observed. In any case, cuttings from tepuru piye are greatly ap-
preciated for their “vitality,” their “youth.” As one farmer told us, normal cuttings
(i.e. clones) get too “accustomed to the soil,” they are too “tamed,” and end up
producing less and less, while cuttings from spont seedlings often produce
increasingly better yields (see Rival [in press] for further disetiseicns af cultural
representations associated with tepuru piye).

Winter 2000 JOURNAL OF ETHNOBIOLOGY ooo

DISCUSSION

Makushi cassava diversity i in ~~ nEZON context.—Bitter cassava is the staple crop of
most lowland A , and the Makushi, who possess at least 76
varieties, are no exception. This high varietal diversity is comparable to the diver-
sity encountered in other groups. Grenand (1993) counted 31 bitter varieties in a
Wayapi community in French Guiana, and Carneiro (1983) 46 among the Kuikuru
of Brazil; Emperaire et al. (1998) found 61 bitter varieties in one caboclo village,
and a total of 140 names among all the villages they studied; Dufour and Wilson
(1996) found more than 100 varieties among Tukanoans from Yapu in Colombia,
and Chernela (1987) collected 137 different names among four communities along
the Uaupés river in Colombia. A comparable level of diversity is encountered
among sweet cassava farmers: 117 names were collected by Salick et al. (1997) in
16 Amuesha communities, and more than 100 by Boster (1984b) among the
Aguaruna in Peru.

Genetic implications of traditional management.—Despite its peculiarities, the Makushi
farming system, which shares many features with other native Amazonian cas-
sava farming systems (slash-and-burn agriculture, strong ooiasapuepeia=iim pressure,

conservative human selection, planting system, of seedlings, and so
forth), can be used as a model to study the genetic ‘consequences of traditional
management.

sava, which grows in an ecosystem shaped by both environmental and
cultural factors, is the target of two interacting types of selection, natural and hu-
man. Human selection is either conscious or unconscious. Genotypes constitute
ultimately the units of selection in both natural and unconscious human selection.
In the case of deliberate human selection, however, the units of selection are the
taxonomic units. One taxonomic unit can include several genotypes, given the
assimilation into a single variety of different clones with similar phenotypes and
of plants of sexual origin. If a farmer decides to multiply such a heterogeneous
variety, she or he may multiply not only one genotype, but several, while, at the
same time, taking the risk of losing other genotypes, because (1) genotypes shar-
ing the same phenotypes are indistinguishable, and (2) no special effort is made to
maintain all the intravarietal phenotypic variation (Boster 1985).

Although human and natural selection have distinct origins, they neverthe-
less act as similar agents of evolution (Salick 1995), and may even interact. Human
deliberate selection on features such as productivity can be superimposed on, and
reinforce, natural selection. In a given environmental context, selection for pro-
ductivity can constitute selection for resistance to drought, herbivory, or other
ecological factors (Johns 1990; McKey and Beckerman 1993). To understand the
observed patterns of varietal diversity, therefore, both natural and artificial selec-
tion must be taken into consideration.

What are the consequences of natural and human selection, and of their inter-
action, on genetic diversity? Pressures may either increase or decrease varietal
diversity. At the farm and community level (intervarietal diversity), environmen-
tal pressures may lead to local or short-term decreases in varietal diversity. Varieties

260 ELIAS et al. Vol. 20, No. 2

can be lost, for example, during a flood or after ant attacks. Human selection may
also lead to decreased diversity, when a variety is accidentally lost because of lack
of suitable space to plant it, or when it is deliberately discarded. Since criteria for
discarding a variety are mainly based on productivity, human selection acts, in
this case, in the same direction as natural selection.

Human selection can also increase varietal diversity. Makushi farmers tend to
favour new varieties that are more liable to disappear, for they are represented by
a few individuals only. Since the environment is highly variable and unpredict-
able, natural selection will not always favour the same genotypes. Makushi
appreciation of diversity for its own sake leads them to keep many varieties, even
if they are not currently very productive. Although this is not an explicit, deliber-
ate strategy, the result is possession of a varietal pool globally resistant to many
kinds of pressures.

Unlike most tuber crops, cassava is propagated by stem cuttings, a material
different from the edible part. The absence of use competition between material
for propagation and for food enables farmers to be more selective about what is
planted. With exchanges of planting material occurring widely among villagers, a
variety accidentally lost in a farm is easily recovered, and the risk of losing variet-
ies greatly diminished within the community. Such exchanges also allow farmers
to acquire new varieties and increase their own varietal diversity. Furthermore,
exchanges taking place between different communities lead to increased diversity
at the village level. Intervarietal diversity is further increased with the multiplica-
tion of volunteer plants grown from seedlings, tepuru piye, which create new
genotypes. Aside from somatic mutations, probably of rare occurrence, tepuru piye
are the only source of original diversity.

Compared to many other Amazonian groups, the social context of Makushi
farming is favourable to a high degree of varietal diversity. Firstly, in the Makushi
farming system, each household manages varietal diversity on its own farms. In
contrast, the Amuesha (Salick et al. 1997) delegate the management of diversity to
their shamans, who not only must grow many more varieties than other farmers,
but who are also responsible for experimenting with the plants grown from seed-
lings. Such a centralised system based on a reduced number of specialists may
have the advantage of ensuring a more homogeneous redistribution of diversity,
but it is far more vulnerable. Secondly, and although they are said to be accultur-
ated, having been converted to Christianity (but see Butt-Colson’s [1967] work on
Pemong and Akawaio millenarist movements) and having lost many of their cus-
toms and traditions, the Makushi have preserved a strong sense of collective
identity and have maintained their traditional subsistence agriculture. Makushi
choices on whether or not to plant a given variety, and on the frequency of plant-
ing each variety, are still personal choices which are not yet dictated by the market
economy or any other external factors. Their situation differs from that of
detribalised and socially destabilised Tukanoans described by Grenand (1993), who
have lost their knowledge of cassava cultivation and processing, a loss that has
led to a dramatic impoverishment of cassava diversity in Tukano land. Thirdly,
Rewa and other Makushi forest villages are not subjected to market forces. Vari-
ous authors (Salick et al. 1997; Emperaire et al. 1998) have shown that markets

Winter 2000 JOURNAL OF ETHNOBIOLOGY 261

push farmers to increase productivity and to cultivate a restricted range of com-
mercial varieties, furthering the tendency to grow rine Ae varieties at
the expense of rarer varieties that may, as a result, get lo los

We have presented the ways in which intervarietal tic diversity is shaped
by environmental and human pressures. But what are the genetic consequences at
the intravarietal level? Scientists studying cassava genetic diversity commonly
assume that varieties are clonal, because they are propagated vegetatively. How-
ever, the incorporation of sexually produced cassava in a variety, and more
generally, the confusion between two types of cassava that have similar pheno-
types, lead to intravarietal genetic diversity (Colombo 1997, Elias et al. in press).
Intravarietal diversity is observed despite the fact that each variety is subject to a
phase of intense genetic bottleneck, when cuttings are selected for the next gen-
eration. Since an adult plant can provide from two to fifteen cuttings, a farmer
often has too many cuttings to plant, except when he or she is planting a new
farm. Surplus cuttings may be given to neighbours or relatives who need them,
pe the bulk is simply planted in large bundles at the edge of the farm. These

ttings are functionally ‘dead,’ since they will either die out, or grow without
being replanted. In other words, only a small prepernen of individuals in each
variety fF tes in producing {1 5 be elimi-
nated in this way through genetic drift, leading to a decrease of intravarietal genetic
diversity at the farm level. Observed intravarietal diversity (Elias et al. in press) is
thus the result of the balance between the frequency of confusion of plants sharing
similar morphological characters, and the strength of genetic bottlenecks.

CONCLUSION

The results presented here illustrate the ways in which selective pressure is
exercised by various socio-cultural factors. This study should be regarded as a
first step in the proposed integrative approach, which will be used as a background
for understanding forthcoming results of genetic analyses and ecological experi-

ments.
NOTES

1 The put wild relative of cultivated cassava Manihot escul bsp. flabellifolia (Olsen
and Schaal 1999) is not found around Rewa, but small sh aa exist around other vil-
lages in the savannah area (at least 75 km away from Rewa). Wild cassava (kwana) cut-
tings are not planted, nor are the tuberous roots used, although old people in the savanna
report having eaten mixtures of wild cassava and cultivated caspave roots during times of
starvation in the 1950s. Genetic data (Elias et al., in f flow from wild
cassava is not a significant evolutionary factor in cultivated etn eee in Guyana, and
that the local wild gene pool has contributed little to the diversity of the cultivated pool.

ACKNOWLEDGEMENTS

We wish to thank th tofG and ially the Ministry of Amerindian
Affairs and the Environmental Protection Agency, for vessel us to carry out our fieldwork
in Guyana. This study was supervised by the European Commission (DGVIII, Avenir des

Peuples des Foréts Tropicales Programme), which provided funds for fieldwork, and also

262 ELIAS et al.

Vol. 20, No. 2

supported by grants from the French government (Ministére de la Recherche et des

Techniques, Bureau des Ressources Génétiques and Centre National de la Recherche
Scientifique). We acknowledge the National Agronomic Research Institute (Georgetown,
Guyana), the Amerindian Research Unit of the University of Guyana, the Conselho Indigena
do Roraima, Ari Weiduschat, Dr. Carlos Borges and Victor Luria for their help. Many thanks
to two anonymous reviewers, whose comments greatly improved this manuscript. Our
greatest debt of gratitude is to the Makushi Amerindians of Rewa, Massara, Annai and
Toka, for their hospitality and their kind participation in our study.

LITERATURE CITED

BOHER, B. and V. VERDIER. 1995. Cassava
bacterial blight in Africa. The ee of
knowledge and _ implicati for
designing control ec nae Aidican
Crop mene Journal 2:1-

BOSTER, J. S. 1983. A comparison of the
diversity of Jivaroan gardens with that
of the tropical forest. Human Ecology
11:47-68.

. 1984a. Classification, cultivation,
and selection o cultivars of
Manihot esculenta (Euphorbiaceae).

. Advances in
. The New York
setae eplan ee York.
ee b. Inferring decision making
from (erases and behaviour: =

selection. Human Ecology 12(4): 343-358,
Ee. SeleCUON Or Aig we
distinctiveness: Evidence

Economic gees! 39: 310-25
Exchange of varieties and
information be tween Aguaruna manioc
cultivators. American Anthropologist
88(2):428-436.
BRUSH, S., R. KESSELI, R. ORTEGA, P.

Conservation Biology 9(5):1189-1198.
BUTT-COLSON, A., 1967. The birth of a
religion: The origins of a semi-Christian
religion among the Akawaio. In Gods
and Rituals: Readings in Religious Belief
and Practice, J. Middleton (editor).
Natural History Press, New York.

CARNEIRO R. L., 1961. Slash-and-burn
cultivation among the Kuikuru and its
implications for cultural development
in

Consequences, J. Wilbert (editor).
Sociedad de Ciencias Naturales La Salle,
Caracas.

. 1983. The cultivation of cassava
among the Kuikuru of the Upper Xingu.
Pp. 65-111 in Adaptive Responses of

ative Amazonians, R. B. Hames and
W. T. Vickers (editors). Academic Press,
New York.

CHERNELA, J. M. 1987. Os cultivares de

Histérico-Antropolégica 1. Centro de
Informacao Diocese de Rordima, Boa
Vista.

CIR. 1993. Os indios no futuro de Roraima.
Conselho Indigena de Roraima, Boa
Vista.

COLOMBO, C. — bite de la diversité
génétique aniocs américains
(Manihot Souinte Crantz) par les
marqueurs moléculaires (RAPD et
AFLP). Unpublished PhD dissertation,

NSAM, Montpellier.

DINIZ, E. S. 1966. O perfil de uma situacao
interétnica. Antropologia. Boletim do
Museu Paraense Emilio Goeldi 31:1-31.

DOLE, G. E. 1963. Techniques of preparing
manioc flour as a key to culture history
in ps nee Spain Pp. 241-248 in Man

nd Cult , A. Wallace (editor).
Canvegay OE Pennsylvania Press,
Philadelphia.

Winter 2000

—___—. 1978. The use of cassava among
the Kuikuru: Some interpretations. Pp.
217-247 in The Nature and Status of
Ethnobotany, R. I. Ford (editor).
Anthropol. Pap. Mus. Anthropol. Univ.
Michigan 67.

DUFOUR, D. L. 1983. Nutrition in the
Northwest Amazon: Household pee
intake and time-energy expenditure. Pp.

Native: Amazcnanis, &. B. H
W.TA ). Academic Press,

New York.

5. Manioc as a dietary staple:
implications for the budgeting ut =
and en n the northwest Am

Schwerin (editors). Gordon and Breach,
New wee

8. Cyanide content of cassava
are esculenta, Euphorbiaceae)
cultivate use sed by fukanoan Indians in

42: epi
. 1989. Effectiveness of cassava
detoxification techniques used by
indi in Northwestern
Amazonia. Interciencia 14(2): 88-91.
aS els i A closer look at the
nutritional li

Laooa

of Amazon

(editor). The University of Arizona
Press, Tucson & London
and W. M. WILSON. 1996. La

douceur de Pamertume: une
P|

par les Indiens Tukano ‘ ‘Amazonie. Pp.
1231-1238 in L’alimentation en Forét
Tropicale: Interactions Bioculturelles et
Pp, as eS my, 1 4 —

M. Hladik, A. Hladik, H. Pagezy, O. F.
Linares, G. J. A. sob a ey Froment
(editors). UNESCO, P.

ELIAS, M., and D. Mc KEY. th Press. The
unm managed reproductive ec ecology o
domesticated plants i in traditional

cassava “and a call for data. Acta
Oecologica

JOURNAL OF ETHNOBIOLOGY 263

ELIAS, M., O. PANAUD, and T. ROBERT.

ess. Assessment of genetic

variability in a traditional cassava

(Manihot esculenta Crantz) farming

markers. Heredity.
D. G and E.

ROBERTS. 1982. An investigation of the
influence of constant and alternating
temperature on the germination of
cassava seed using a two-dimensional
temperature gradient plate. Annals of
Botany, 49:241-246.

EMPERAIRE, L., F. PINTON and G.
SECOND. 1998. Une gestion dynamique
de la diversité variétale du manioc en
Amazonie du Nord-Ouest. Nature,
Science et Société 6(2):27-42.

FARABEE, W. C. 1924. The Central Caribs.
Museum of ee Publications
10. University of Pennsylviana,
Philadelphia.

FARAGE, N. 1991, As muralhas dos sertoes.
Os povos indigenas no Rio Branco e a
colonizacao. Paz e Terra, Sa6 Paulo.

FAUQUET, C. and D. FARGETTE. 1990.
African cassava mosaic virus: etiology,
epidemiology and control. Plant Disease
74:404-411.

GARWOOD, C. N., 1989. Tropical seed
banks: a review. Pp. 149-209 in Ecology
of Soil Seed Bank, M. A. Leck,
Parker and R. L. Simpson (editors).
Academic Press, London.

GRENAND, F. 1993. Bitter manioc in the
lowlands of tropical America: from
myth to commercialisation. Pp. 447-462
in Tropical Forests: People and Food, C.
a a A. Hladik, H. Pagezy, O. F.

G. J. A. Koppert and A. Froment

(eto. ee Paris.
sarin HAXAIRE 1977.
Monograph oi abattis Wayapi. Jatba
24(4):28
EM aa rs ae a ree! we and
a. Pp. 39-

ea a

Roraima, P. Furley (editor). Routledge,

New York.

. 1995. Red Gold: The Conquest of
the Brazilian Indians. Revised edition.

Harvard University Press, Cambridge.

264 ELIAS et al.

pee ones C. 1978. Food for thought.
tterns roducti and

pene in Pira-Parana society. Pp.
41-66 in Sex and Age as Principles o
Social Ddifferentiation, J. Lafontaine,
(editor). Academic Press, sccm

. 1979. From the Milk River: Spatial
and Temporal Processes in Saahles:
Amazonia. Cambridge University Press,

Cambridge.

JOHNS, T. 1990. With Bitter Herbs They
Shall Eat It. Chemical Ecology and the
Origins of Human Diet and Medicine.
University of Arizona Press, Tucson.

LEFEVRE, F. 1989. Ressources Génétiques
et Amélioration du Manioc, Manthot
esculenta Crantz, en Afrique. Thése,
IN aris.

LOUETTE, D. A. CHARRIER and J.
BERTHAUD. 1997. In situ conservation
of maize in Mexico: genetic diversity

and maize seed management in a
traditional oo Economic
Botany 51(1):20-

USHI RESEARCH UNIT. 1996.
Makusipe Komanto Iseru (Sustaining
oo Way of Life). (Ed. J. Forte).
Report of a _ research project
soya eee for sven
International Rainforest Program
North Rupunini District Development
Board, Georgetown, Guia

MCKEY, D. and S. BE CKERMAN. 1993.
Chemical ecology, plant evolution and
traditional caceava cultivah ra

Pp. 83-112 in L’alimentation en Forét
Tropicale: Interactions Bioculturelles et
D, eet 1 Ti 1 4 cc

M. Hladik, A. Hladik, H. Pagezy, O. F.
Linares, G. J. A. Koppert and A. Froment
(editors). UNESCO, Paris.

MOWAT, L. 1989. Cassava and Chicha.
Bread and Beer of the Amazon Indians.
Shire Publications, Aylesbury.

NASSAR M. A. N. and R. P. TEIXEIRA.
1983. A cuca da dorméncia da
semente das espécies selvagens da

andioca, Manihot spp. Ciéncia e
Cultura co 630-632.

OLSE B. A. Sage aioe
Selec. on he ori cassa
Phylogeography of Manihot ayy
Proceedings of the National Academy
of Science 96:5586-5591

Vol. 20, No. 2

QUIROS, Cc. E., $B. BRUSH, D.S.
DOUCHES, K. S. MERER and G.
HUESTIS. 1990. Biochemical and folk
assessment of variability of Andean
cultivated potatoes. Economic Botany

-266

RENVOIZE, B. S. 1972. The area of von
of ss a as a crop plant —a
review vidence. Eeaciotiae
Botany 26: 352-360.

RIVAL, L. 1998. El Nifio Crisis in the North
Rupununi Savannas of Guyana. APFT
Briefing Note 18. European Union, DG
VILL, sr

i aN press: Seed and clone. oa

oe cultivation. In rarely as
Visible d the Material: the
Amerindianization of Society in the
Work of Peter Riviere, L. Rival and N.
Whitehead (editors). Oxford University
Press, Oxford.

RIVIERE, P. 1963. An ethnographic survey
of the Indians on the divide of the
Guianese and Amazon river systems.
Unpublished PhD thesis, University of
Oxford, Oxford.

ROGERS, D. J. 1965. Some botanical and
ethnological considerations of Manihot
esculenta. Economic Botany 19:369-377.

SALICK, J. 1992a.

ana fo = 0 2
Cmsarviiies of Neotropical Forests:
Working with Traditional Resource Use,
K. Redford and C. Padoch peered
Columbia ose Press, New

92b. Crop domestication pus
the evolutionary npoaed of cocona
(Solanum _ sessiliflorum Dunal)
ceoner ers Biology 26:247-285.

—_—__—. 1995. Toward an integration of
evolutionary ecology and economic
botany: Personal perspectives on plant/
people interactions. Annals of Missouri
Botanical Garden 82:25-33.

—_——— ., N. CELLINESE and S. KNAPP.
1997. Indigenous diversity of cassava:
Generation, maintenance, use and loss
among the Amuesha, Peruvian upper
Amazon. Economic Botany 51(1):6-19.

Winter 2000

—__——.. and L. MERRICK. 1990. Use and
maintenance of genetic resources: Crops
and their wild relatives. Ae es in
Agroecology, C. R. Car
Vandermeer and P. M. Ribcset (editors).
McGraw-Hill, New York.

SANTILLI, “s pages 1994. As Fronteiras
da Repub Historia e Politica entre o
Macuxi no Vale do Rio Branco. NHU-
USP, FAPESP, Sao Paulo.

SHIGETA, M. 1996. Creating landrace
diversity: the case of the Ari people and
Ensete (Ensete ae in Ethiopia.
Pp. 233-268 in Redefining Nature:
Ecology, Culture and Domestication, R.
F. Ellen and K. Fukui (editors). Berg,
Oxford.

SOUZA, Vij: J: BAIN; Cx ‘SILVA;: VV:
BOUCHET, A. VALERA, E. MARQUEZ
and E...: EGUIARTE.....1997,
ee Do agricultural
practices modify the population
structure of the nitrogen fixing bacteria
Rhizobium etli biovar phaseoli? Journal of
Ethnobiology, 17(2):249-266.

JOURNAL OF ETHNOBIOLOGY 265

VAN DER HAMMEN, M. C. 1992. El
nejo del mundo: naturaleza y
sociedad entre los Yukunas de la
Amazonia Colombiana. Estudios en la
Amazonia Colombiana IV. PhD
dissertation. Tropenbos Colombia,
ogota.

WILLIAMS, H. J., and T. G. EDWARDS.
1980. Estimation of cyanide with
Alkaline Picrate. Journal of Sciences of
Food and Agriculture 31:15-22.

YDE, J. 1965. Material Culture of the Wai-

a
ZIMMERER K. i 1991. The regional

in hi ighlan e Peru.
Biogeography, 18:165-178.

Journal of

266 BOOK REVIEWS Vol. 20, No. 2

Utilizacién y conservacién de los ecosistemas terrestres de México, pasado,
presente y futuro. Antony Challenger with the collaboration of Javier Cabal-
lero. Comisién para el Conocimiento y Uso de la Biodiversidad; Instituto de
Biologia, Universidad Nacional Auténoma de México; Agrupacion Sierra
Madre, S.C. 1998. 847 pp., maps, diags., photos, refs., and index. Paper. ISBN
970-9000-02-0.

In the preface of this book Antony Challenger explains how he set out to write
this enormous tome. As he describes it, only someone young (he was 26), enthusi-
astic, and inexperienced would have accepted the challenge to write a single boo
that was both sufficiently detailed and broad enough in scope to serve as a basic
text on conservation in Mexico. It took him seven years and 847 pages. The result
is an extensive and invaluable overview of the political and cultural ecology of
biodiversity in Mexico.

The book consists of an introduction, a four-chapter section on the history of
human uses of ecosystems, six chapters covering the biogeography, ecology, and
human uses of the main terrestrial ecosystem types, and a conclusion addressing
sustainable development in Mexico.

the introductory chapter, Challenger e> thei f biodiversity,

<pla
summarizes the threats to biodiversity in Mexico, /and describes how indigenous
and traditional farmers continue to make important contributions to Mexican
biodiversity through, Se pia processes _ fee Sil ges. and adaptation of exist-
ir lg crop varieties to d human needs. Specific topics include
species relationships, co-evolution, ial: selection and the major activities threat-
ening biodiversity in Mexico. In contrast to the destruction of Mexican ecosystems
under modern production techniques, indigenous agricultural activities enhance
biodiversity through ongoing processes of domestication and diversified produc-
tion systems maintaining a heterogenous landscape.

Chapter Two covers “primary production and the conservation of natural re-
sources” from the Pleistocene to the Spanish Conquest, including the origins of
agriculture in Mexico, and the agricultural basis of several pre-Columbian civili-
zations in Mexico. Chapters 3, 4 and 5 address the Colonial period, independence,
the Porfirian dictatorship, and modern Mexico from 1910 to 1994. Each of these
chapters concludes with a map of Mexico illustrating the general distribution of
agricultural, grazing, forestry, mining, and industrial activities and the ever-grow-
ing extent of alterations to natural ecosytems.

These chapters take a cultural ecology perspective in which society and na-
ture are interwoven. For example, the culture of the Maya encompasses their
agricultural systems and settlement patterns, and these have repercussions on
vegetation and wildlife. In the same way, environmental Ananee alec syed
and a number of theories for the collapse of M s hin
ronmental change. In his discussion of the conquest, Challenger dieactibes a process
of cultural and ecological transformation in Mexico following the introduction of
new species, including human pathogens like small pox and scarlet fever. In the
chapter covering modern Mexican history, as addresses the environmen-
tal implications of topics such as land reforms, import substituti

Winter 2000 JOURNAL OF ETHNOBIOLOGY 267

tropical colonization and the promotion of cattle ranching, agricultural policies,
the debt crisis, and finally neoliberalism.

Historically and currently, Mexican d ] tst ies are Pi amt
Furthermore, they are actively destroying the bases from which Mexico could con-
struct a sustainable development alternative. One of the greatest ‘eipailies of the
Spanish Conquest and colonization, for example, was the loss of indigenous tech-
nical knowledge, agricultural techniques, and perhaps crop varieties. These
disappeared with the drastic re-organization of production around European plants
and animals, mining, and export crops like dyes and sugar, and the collapse of
indigenous populations. Currently, the North American Free Trade Agreement
continues the process by pitting Mexico’s small-scale indigenous maize farmers
against enormous and heavily subsidized agro-industries in the United States and
Canada.

After setting out the human history of land use in the first 250 pages of the
book, Challenger is ready to describe the current situation of primary production
and conservation and vegetation types. Chapters 6 through 11 focus on specific
ecological zones. Topics include the sources of various taxa, profile drawings of
different forest types, and descriptions of the main ecological processes shaping
these ecosystems, such as forest succession. Challenger provides long lists of cur-
rent and suggested protected areas in each of the five major ecosystems. Most
importantly, each chapter mapas indigenous and modern systems of manage-
ment in each ecosystem

The final chapter cakes the value of indigenous technical knowledge and
traditional environmental management systems. In contrast to green revolution
agricultural techniques that homogenize the environment, rely on unsustainable
chemical and fossil fuel energy inputs, deplete soil and water, poison people and
face real risk of collapse, traditional agricultural systems maintain environmental
heterogeneity and often have proven sustainability. Ethnoecology, therefore, has
an important role to play in the search for sustainable environmental manage-
ment.

Challenger argues that sustainable development in Mexico requires the con-
servation and intensification of traditional farming systems, including serious
efforts to rehabilitate soils and forests. This, in turn, requires a radically different
development strategy, one based on food self-sufficiency, not the current “market
logic” of importing grains. Sustainable development requires a paradigm shift, a
la Kuhn. A drastic change in thinking should accompany a global redefinition of
— — to pensiautonemous national economies trading as equal
partners. In Mexi t t would also require land reform, es-

scala directed sans cattle ranches i in tropical zones.

n his final paragraph, Challenger writes that the biggest challenge is not
how to manage natural resources, but rather how to manage ourselves. This is the
weakest aspect of the book. The power relations behind the market structures and
government policies he criticizes remain opaque. In a brief section on plantation
ba. for example, Challenger does not address the issue of how these subsi-

benefit t lintegrated pulp and paper companies
and fail to “reach the small-farmer actors he expects to use them to restore de-

268 BOOK REVIEWS Vol. 20, No. 2

graded environments. Similarly, he is very optimistic about the ability of biotech-
nology to play a role in improved traditional systems, without considering the
goals of the actors behind most biotechnology research.

In another example of the book’s relative weakness on social issues, Chal-
lenger calls for the establishment of more parks and the funds to delineate and
patrol them as part of a broader process of ecological zoning and land use plan-
ning. The United States park system remains a model suitable for emulation (p.
785). He does not address the state of conservation within existing Mexican parks,
nor the compelling and difficult issues of social justice, cross boundary, and other
people management issues in Mexico’s parks, many of which are inhabited and
nearly all of which are heavily used.

Conversely, though, - description of Maxine’ s ecosystems, their natural his-
tories, human impacts, an strategies omprise the strength
of the book. Challenger ae ‘the value of ethnoecology for sustainable
development, clarifies the cultural and ecological destructiveness of past and
present development policies, and identifies potential techniques for better man-
agemen

Unfortunately, the book’s size and organization dilute these compelling mes-
sages. The book would have benefited from better editing. The conclusion, for
example, contains sections on human impacts on ecosystems and potential areas
for parks that should have been included in the respective ecosystem chapters
where there! is a material. The conclusion also contains lengthy discussions
of , and organic coffee, but these
too should have been contained in the ecosystem chapters with similar material.
Similarly, a few tables could have replaced pages of description of protected areas
throughout.

Even with such editing, the book should be printed in two or more volumes.
Incredibly, this 850-page book is published in a single paperback volume. This is a
book that needs to be in every Mexican college library and accessible to every
organization addressing conservation and development issues in Mexico. Unfor-
tunately, even moderately-used copies will rapidly fall apart; my copy already
has a cracked spine. It is too bad the book is not itself more sustainable.

Despite the length, a thorough index and a detailed table of contents make the
material accessible. Challenger maintains an author-date citation style through-
out with references following each chapter, and this guarantees the utility of the
book as an obligatory desk reference for all Latin Americanists with interests in
the intersections between biodiversity, people, politics, and history. We are fortu-
nate Challenger took on a Herculean task, and executed it so well.

Daniel Klooster
Department of Geography
Florida State University

Journal of Ethnobiology 20(2): 269-299 Winter 2000

HISTORICAL ECOLOGY ve THE SOUTHEASTERN
LONGLEAF AND SLASH PINE FLATWOODS: A SOUTHWEST
FLORIDA PERSPECTIVE

KAREN J. WALKER
Florida Museum of Natural History
Randell Research Center
PO Box 117800, University of Florida
Gainesville, FL 32611

ABSTRACT.—Before E A ttl t of th tl t US., longleaf
pine (Pinus palustris Mill.) was present and largely dominant on an estimated 85
percent of all upland area within the longleaf’s botanical range. Today, longleaf is
present on only about 2.6 percent of those uplands. In addition, uplands forested
wee Bipeh eg Saud vince sneer) have been reduced from a pre-
t 1993). This dramatic landscape
change is a ae of long-term relations saibah yeas the ee prosystems and human
activity. Understory plants, soil moisture, and periodi 1 factors
while domestic animals, agriculture, the naval-stores an lumber industries, and
fire reduction were human-related factors. Some of the S t’s last old-growth
pine forests were logged in south Florida during the 1920s, 1930s, 1940s, and 1950s.
Mostly of the pine flatwoods type, these were the southernmost forests in the
longleaf pine’s range and they included both longleafs and the south Florida
variety of slash pine (Pinus elliotti var. densa Little & Dorman). In southwest
Florida’s Lee County, historic and oral-historic research focused on the pine
Seheronds near Fost Myer, north and south off the Caloosahatchee River. South of
the ri major component
of oe 1924-1944 logging operation that greatly impacted the flatwoods of both
areas. The results of this historical ecology research illustrate the heterogeneous
process of landscape change at regional (Southeast U.S.), subregional (south
Florida), and local (southwest Florida) scales.

ey words: historical ecology, longleaf and south Florida slash pines, southwest
Florida, oral history, archaeology.

RESUMEN.—En el sudeste de los Estados Unidos, y antes de la colonizacion de

los euro-americanos, la presencia de los pinos de hoja larga (Pinus palustris Mill)

dominaban en gran parte el terreno elevado que quedaba dentro del area

demarcada botdnicamente para este tipo de pino. Se estimaba que el area

compendia un ochenta y cinco (85) por ciento del terreno. Hoy dia, tan solo el dos

pine? seis (2. 6) pors ciento de los pines de hoja larga estan gepresentados dentro
Ademias, las areas en

mas altos dond 1entran los pinos cortados (Pinus elliottii Englem) han sido

reducidas a atro (0.4) por ciento, en comparaci6n a el tres punto tres
Se 2) Leedo ciento que e existia durante la epoca pie euro-americana or 1993). roe

durante mucho tiempo entre el aurea tate de los pinos y la actividad
humana. Est q medad en terreno, y los incendios

270 WALKER Vol. 20, No. 2

que occurieron periddicamente, fueron los factores ecolégicos. Los factores
humanos que contribuyeron a éste cambio fueron, los animals domésticos, la
agricultura, las tiendas de tipo marino o nautico, las industrias de madera, y la
reduccién de los incendios. En el sudeste se encuentra el crecimiento de algunos
de los bosques de pino mas viejos y que fueran registrados en el sur de la Florida
durante los afios 1920, 1930, 1940, y 1950. En su mayoria, los pinos del tipo se
encuentra en las areas de terrenos llanos estan en la parte sur de Florida y son de
hoja larga, asi como la variedad de pino cortado (Pinus elliotti var. densa Little &
Dorman). En el Condado de Lee, que se encuentra en el sudoeste de la Florida,
hay estudios histéricos y de historia oral donde se enfoca el tema de los pinos en
las llanuras cerca de Fort Myers y en la parte norte y sur del Rio Caloosahatchee.
En un estudio arqueolégico que se realizé al sur del rio, se hizo possible el
documentar los restos de uno de los componentes principales en la operacién de
la extracci6n de madera durante los aioe de 1924 a 1944 y lo que caus6 un gran
impacto en los bosques q en los terrenos llanos de ambas 4reas.
Ea una inspeccion de tipo pintinico-ecoligion se ude documentar el Lpmecnse

raAienie

dela region (sudeste de los Estados l Unidos), la sub-region (sur de la Florida), y la
parte local (al suroeste de la Florida).

RESUME.—Avant l’imp] ti iro-amé€ricaine dans le sud-est des Etats-Unis,
le pin des marais (Pinus alias Mill.) était courant et prédominait largement
avec une estimation de 85 pour cent de tout le haut pays classé dans la variété
botanique du pin des marais. Aujourd’hui, le pin des marais se trouve seulement
dans 2,6 pour cent du haut pays. De plus, les hautes terres boisées de pitchpins
américains (Pinus elliottii Bes ose ont été réduites d’un pourcentage preéuro-
améri pour t 0,4 pour cent (Gel de 1993). Ce changement
dramatique de paysage est le résultat de relations a longs termes entre les
sia acini du pin e Tachvie humaine. Les pane des sous-bois, l/’humidité
s facteurs écologiques alors
que les animaux domestiques, agriculture, l’€quipement naval et les industries
du bois, la réduction de feu, furent les facteurs relatifs 4 l‘homme. Certaines des
derniéres anciennes foréts de pins du sud-est furent abattues dans le sud de la
Floride dans les années 1920, 1930, 1940 et 1950. Saved sap a? ties de pins, on
les trouvait le plus au sud sous la variété de
a la fois les pins des marais et la variété de pitchpins américains (Pinus elliotti var.
densa Little & Dorman) de Floride du sud. Dans le County Lee de Floride du sud-
ouest, la recherché historique et orale historique s’est focalisée sur les foréts de
pins pres de Fort Myers, au nord et au sud de la riviére Caloosahatchee. Au sud
de la riviére, un étude archéologique a révélé les restes d’un élément majeur de
l'opération de 1924-1944 sur l’exploitation du bois qui a grandement influencé les
foréts de chacune des régions. Les résultants de cette recherche écologique
historique illustre le processus hétérogéne de changement de paysage a l’échelle
régionale (le sud-est des Etats-Unis), sous-régionale (le sud de la Floride) et locale
(le sud-ouest de la Floride).

INTRODUCTION

Historical ecology, as defined by Crunaley (1994a, 1994b, 1998) and: others (Balée
1998; Headland 1997; Winterhalder 1994), i

Winter 2000 JOURNAL OF ETHNOBIOLOGY 271

of the dialectical relations between people and the physical environment. This
approach views the cause of cultural and ecosystem change as interactive rather
than deterministic. Crumley (1994b:6-7) states that “long-term sequences may be
traced through the study of changing landscapes, defined as the material manifes-
tation of the relation between humans and the environment.” Examination of
landscape change at more than one temporal and spatial scale is crucial to the
analysis because the process of change at le may not be the same at another
scale (Marquardt and Crumley 1987:2-9). Combinations of archaeology,
ethnohistory, ethnography, ethnoecology, ethnobiology, history, geography, and
the environmental sciences are appropriate to the integrative study called for by a
historical ecology approach. Ethnobiology, for example, focuses on the relations
between people anid pa and animals but does not emphasize the historical
(including arcl continuum or landscape elements other than plants and
animals. Historical ‘ecology is broad in scope, potendally. eee passing the
multiscalar past and present, and multiscalar landscape elements such as climate,

fire, geomorphology, soils, plants, animals, and humans.

Employing the approach of historical ecology, I examine a landscape change
that occurred across the Coastal Plain region of the U.S. Southeast—the greatly
diminished forest ecosystems of the longleaf pine (Pinus palustris Mill.) and the
slash pine (Pinus elliottii Engelm.) (Little 1971; Wunderlin 1998)—but with a local-
scale focus on the pine flatwoods of southwest Florida’s Lee County. Longleaf
pines were once so abundant in the Atlantic and Gulf Coastal Plain states that
they and their plant and animal associates composed one of the dominant forest
ecosystems of the region (Frost 1993; Wahlenberg 1946; Walker 1991). Old-growth
longleaf and slash pine forests greeted early European and EuroAmerican explor-
ers, travelers, and settlers to the Southeast; these once-seemingly endless forests
were described as open stands of pines towering over a low understory often domi-
nated by grasses or saw palmetto (Serenoa repens W. Bartram). Deforestation of the
region’s old-growth pine forests was a long process encompassing several hun-
dreds of years but intensifying primarily Curing | - eighteenth through twentieth
centuries. Frost (1993) presents an j tal history of the
longleaf at this long-term regional scale.

Examining the longleaf pine from a south Florida perspective is also important
because this subregion supported the po ihenrnnost forests of longleaf and their
penetration into Florida’s sul ted or understood. Longleaf
pine forests in south Florida were and are of the flatwoods type, the land generally
being too low to support the sand-hills longleaf forest type. In addition, the
subregion’s slash pine is Pinus elliottli var. densa (Little & Dorman), distinct from the
typical northern variety, Pinus elliottii var. elliottii; the former has some characteris-
tics similar to longleaf (Abrahamson and Hartnett 1990:112; Moyroud 1996-1997:11;
Small 1930; Snyder et al. 1990). Earlier in the twentieth century, the south Florida
slash pine was thought to be Pinus caribaea Morelet, the Caribbean pine (e.g.,
Harshberger 1914; Small 1930:42). Indeed, the southernmost slash pinelands (e.g.,
Everglades National Park) exhibit a distinct assemblage of plant taxa owing to their
subtropical location (Snyder et al. 1990). The distributions of longleaf and south
Florida slash pines overlap at least in the northernmost areas of south Florida (e.g.,

272 WALKER Vol. 20, No. 2

Atlantic
Ocean

Highlands Co.
, Glades Co.

ooo "Noel and es \

aan of Mexico

FIGURE 1.—Map showing estimated pre-EuroAmerican Southeast U.S. range of longleaf
pine (Pinus palustris Mill.) and botanical ranges of typical slash pine (Pinus elliottii elliottii
Engelm.) and south Florida slash pine (Pinus elliottii densa Little and Doonan). The longleaf
range is generally based on Frost’s t 1umMerous sources
(1993:Figure 2). The more detailed southwest Florida range is Solid on Harshberger (1914)
and Sudworth (1913:Map 35). The slash pine ranges are from Little (1971, 1977). Inset m map
is of Lee County showing the Hickey Creek Mitigation Park (HCMP) and the Cape Coral
and Hickey’s Creek/Lehigh components of the McWilliams /Dowling & Camp logging
system (as reconstructed by James Pickens from 1944 aerial photographs), Slater Mill, and
other locations mentioned in the text. The shaded areas are hypothesized to have been
forested with a mix of old-growth south Florida slash pines and longleaf pines.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 273

Lee County, Highlands County). A third native pine, the sand pine (Pinus clausa
Chapm.), is found in the western half of Lee County with its southernmost occur-
rence in extreme western Collier County (Harshberger 1914:Map; Little 1978:Map
4). South Florida was one of the last areas of the Southeast to experience intensive
EuroAmerican settlement. Thus, the subregion’s pine forests were some of the last
of the Southeast’s old-growth pine forests to be logged. Clear-cut logging, the final
phase in the Southeast’s pine-deforestation process, did not begin in southwest
Florida until the 1920s (Zeiss 1983), continuing to as late as 1956 (Tebeau 1957). Thus,
a few individuals with first-hand knowledge of south Florida’s old-growth forests
and their destruction are still living today, representing valuable oral-historic re-
sources. And logging-related features and archaeological deposits are still evident
on the landscape, allowing documentation.

THE PRE-EUROAMERICAN PINE FORESTS

A Problematic Documentation.—Researching historical documents concerning the
southern pines is problematic (Frost 1993:18; Wahlenberg 1946:268). This is largely
due to the botanical similarity of the four yellow pines—longleaf, slash (two vari-
eties), shortleaf (Pinus echinata Mill.), and loblolly (Pinus taeda L.) (Wunderlin
1998:62)—and their often overlapping distributions (Little 1978:Maps 5, 6, 8, and
10). Of these four pines, however, only the south Florida slash and longleaf pines
are native in south Florida. Even so, whether referring to south Florida or areas to
the north, early land surveys, maps, and travelers’ accounts seldom specify the
species of pine recorded.

A plethora of common names have been used at any one time for these pines,
all described as “yellow” because of their wood color (Record and Hess 1943).
Wahlenberg (1946:268) explains that patterns of geography and sometimes eco-
nomics (i.e., pine products) could be observed in the variation of names. For
example, longleaf pine was generally called “fat” pine in the deep south,
“longleaved” and “longstraw” pine in the Atlantic states, “turpentine” and “rose-
mary” pine in North Carolina, “brown” pine in Tennessee, and “orchard” pine in
Texas. Loggers and lumbermen variously called longleaf “yellow” pine, “heart”
pine, “southern” pine, “hard” pine, and “pitch” pine (see also Mohr 1896:28;
Panshin and deZeeuw 1980; Record and Hess 1943). Mohr (1896:28) lists “slash
pine,” “swamp pine,” “bastard pine,” “meadow pine,” and “she pine” as com-
mon names used for slash pine.

Most bothersome to researchers, the word “longleaf” was sometimes used in
the lumber industry to indicate any of the yellow pines that met lumber standards
of high quality. To Wahlenberg (1946:268), this confusion was understandable from
a lumberman’ 8 pot of vEW asics the yellow pines that are easily distinguished

hed (anatomically) by their wood (see

also “Record and Hess [1943] and Panshin and deZeeuw [1980] for examples of

osperm keys that reflect this problem). In addition, early forestry surveys

often combined longleaf and slash pines in a category called “turpentine pines”

when reporting acreages (Wahlenberg 1946:xiii-xiv, 1), in part due to intergrading
(i.e., mixed stands) of the two species.

274 WALKER Vol. 20, No. 2

Archaeological and paleoecological documentation of the pines is equally as
problematic, if not more so, as historical and ethnobotanical documentation. The
most frequently recovered archaeological plant remains are in the form of small
fragments of charred wood, often found in great quantities. Although « charred wood
fragments often can be identified to species under , wood
anatomists and archaeobotanists are not able to distinguish between the southern
species of Pinus (Panshin and deZeeuw 1980; Record and Hess 1943). Unfortu-
ss this aAsity also apples to preserved, waterlogged wooden artifacts.

y of pine pollen is also limited to the genus level, as pollen from
the various species are “difficult or impossible” to distinguish (Watts 1993:15).
However, one promising, indirect, avenue of identifying past longleaf woodlands
is the determination of the mass and relative abundances of associated understory
plants based on phytoliths recovered from soils (Kalisz et al. 1986:187).

cone of Acreage and Range. = Despite the difficulties of historical research, recon-
g PE ted for pre Puronmencen pos,
especially for longleaf pine. Reported esti for tl
forests range from 50 to 92 million (e.g., Frost 1993; Landers et al. 1995; Wahlenberg
1946:8; Walker 1991:128). For example, Frost (1993) calculates that 92 million acres of
the region’s woodlands included “ some longleaf pine” and of that acreage perhaps
roughly 74 milli f-dominated woodlands. One writer reports that
only .002 percent of the co. forests remains (Winn 1996:15). Estimates for cur-
rent acreages range from 1 to 5 million (e.g., FCMP 1995; Landers et al. 1995; Longleaf
Alliance n.d.). A 1995 systematic inventory by county of longleaf pine (comprising
more than 50 percent of the tree cover) resulted in an estimate of 2.95 million acres
(Outcalt and Sheffield 1996:2). Of the current longleaf acreage, Virginia has none and
Florida has the most, almost one million acres (Outcalt and Sheffield 1996:20).
Estimates of longleaf’s pre-EuroAmerican areal distribution also vary.
Wahlenberg (1946:46) distinguishes between a botanical range (potential range)
and a commercial range (range of exploitable forests), pointing out that most re-
constructed distributions were probably based on commercial (i.e., exploitable)
ranges, resulting in conservative boundaries (e.g., Mohr 1896; Sargent 1884). Thus,
he concludes that the pre-EuroAmerican longleaf-forest boundaries lay somewhere
between the two ranges. Frost’s (1993:18) recent reconstruction of longleaf’s range
may be the best to date at the regional scale because it is a synthesis of the major
studies published between 1861 and 1971. But it does not depict the true nature of
longleaf’s southernmost distribution. Wahlenberg (1946:49-50) notes that longleaf
is restricted in it by snow, which is dangerously heavy when
accumulated on the tree’s long needles. However, competition from deciduous
species may be a more important factor. Generally, longleaf pine extended across
the Coastal Plain (Figure 1), from southeastern Virginia across to portions of Loui-
siana and a small area of eastern Texas (Frost 1993). Distribution maps also
consistently show that longleaf pine was found throughout Florida’s panhandle,
and its north and central peninsular regions. Typical slash pine had a more re-
stricted, even more southern range, generally distributed from southern South
Carolina to central Florida and west to southeast Louisiana (Little 1971; Figure 1),
often characterized as concentrating along the coastal areas (e.g., Sargent 1884:520).

Winter 2000 JOURNAL OF ETHNOBIOLOGY 275

The southern longleaf boundary may be the more difficult of the two to recon-
struct because the distribution of the south Florida slash pine overlaps with the
southernmost longleafs (Figure 1) and the similarity of the two yellow pines has
resulted in an often ambiguous historical record. Apparently. it is increased soil
moisture that marks the longleaf’ tl and Hartnett
1990:111-112; Peet and Allard 1993:61). Like typical slash pine, south Florida slash
pine is more tolerant of poorly-drained soils and as a result is the more dominant
pine across south Florida. The majority of pre- and post-EuroAmerican maps de-
pict longleaf’s range as halting northwest of Lake Okeechobee in south-central
Florida and at the Caloosahatchee River in southwest Florida, limited to the main-
land (e.g., Frost 1993:18; Little 1978:Map 8; Schwarz 1907; Wahlenberg 1946:44). It
may be that these south Florida boundaries were “commercially drawn,” as
Wahlenberg called it, and therefore are conservative. (This is certainly the case
with the forest-survey maps of Mohr [1896] and Sargent [1884].)

For example, University of Florida herbarium records document scattered
longleafs in the Estero area (FLAS 120603, collected 1975) of southwestern Lee
County and an “extensive open stand of [longleaf] trees” on Pine Island (FLAS
82831, collected 1961), west of mainland Lee County (Figure 1 inset). Outcalt and
Sheffield’s (1996:19) inventory shows acreages of longleaf-dominated forest in two
south Florida counties, Glades and Highlands, west and northwest of Lake
Okeechobee (Figure 1). Frost’s (1993:18) reconstruction includes this Okeechobee
locale, depicting it as part of a division called “scattered longleaf pine in slash
pine areas transitional to south Florida communities.

otanist John Harshberger (1914:89) traveled through Lee County
(including what is today Collier County) early in the last century and reported
that “on the west coast, south and north of the Caloosahatchee River, the slash-
pine mingles with the long-leaf pine, Pinus palustris Mill.” and in another entry,
“scattered growths of longleaf-pines, Pinus palustris Mill., continue south of the
Caloosahatchee River into Lee County on the authority of J. A. Davison, an engi-
neer, as far as Surveyor’s Creek, and the tree has been reported at Henderson’s
Creek, but it is not an important element of the forest, which consists of the slash-
pine, Pinus caribaea Morelet [today known as Pinus elliottii var. densa] and associated
species.” Surveyor’s Creek, today known as the Imperial River (Grismer 1982:330),
is located in southernmost Lee County (Figure 1 inset). Henderson’s Creek is lo-
cated even farther south, between Naples and Marco Island, in today’s Collier
County. Sudworth’s (1913:Map No. 35) botanical range for longleaf pine includes
most of Cape Coral. It also extends south of the Caloosahatchee River including a
locale overlapping eastern Lee County and western Hendry County, a band along
the river, Pine Island, and a locale in the Estero area of southern Lee County. Based
on Harshberger and Sudworth, Frost’s presettlement transitional mixed longleaf-
slash zone should be extended to include parts of southern Lee County in order to
depict more accurately longleaf’s southernmost botanical range, as I have indi-
cated in Figure 1.
Longleaf and Slash Pine Forest Ecosystems. —Eighteenth and nineteenth-century ac-

counts of travels through the Southeast paint images of extensive open forests of
tall pines (e.g., Bartram 1791:43, 186, 191; Brinton 1869:95, 104; Romans 1775:14-

276 WALKER Vol. 20, No. 2

17; Vignoles 1823:86-87). One could see for a great distance into the forests. It was
thus also easy to travel through them and to hunt game animals, described as
abundant. In some cases, longleaf was clearly the dominant tree being described
(e.g., Bartram 1791:33, 52; Romans 1775:16) but more commonly, only t the generic
“pines,” “pinelands,” “pine flatwoods,” etc. indicated. Similarly, t
(1914:90) and Small (1930) described the south Florida slash pine forests as “an
usually open” with an unobstructed view, and “endless.” These early accounts
and others suggest that pre-EuroAmerican slash- and especially longleaf-domi-
nated forest ecosystems may have been characterized by a lower understory than
most pine forests of today. The interpretation is far from certain (Myers 1990:182),
however, because by the eighteenth century, feral and free-ranging European-in-
troduced hogs and cattle were abundant in the pine forestlands, grazing and
foraging in the understory (e.g., Romans 1775:16).

General characteristics of mature longleaf-dominated pine ecosystems include:
low longleaf stand density; minor hardwood component, mostly oaks; grass-domi-
nated groundlayer; high plant species richness; frequent surface fire; occurrence
across a wide geomorphic and hydrologic gradient (although well-drained sandy
soils are most common); and stands of uneven-aged trees (Landers et al. 1995:40;
Palik 1995:6; Schwarz 1907:3-17). An important difference between longleaf and
slash forests is the much slower rate of longleaf growth while in the seedling stage,
leading to the undeserved reputation of being slow to reach timber size (Franklin
1997:5; Landers et al. 1995:42). Longleaf forests are often vieibty. aut =m slash
pine forests in that bunch grasses (especially the ta Michx.
in the north and Aristida beyrichiana Trin. & Rupr. in the southernmost areas) are
the dominant understory plant of the former while saw palmetto and to a lesser
extent gallberry (Ilex glabra L.) typically dominate in a slash pine forest. However,
recent studies recognize a wide diversity of longleaf ecosystems based on vegeta-
tional composition and soil moisture (e.g., Harcombe et al. 1993; Peet and Allard
1993), including a longleaf system with saw palmetto along the northern Gulf
Coastal Plain (Peet and Allard 1993:57, 58). Most of Florida’s longleaf forests of the
Gulf Coastal Plain, including those of southwest Florida were or are probably of
the “southern longleaf flatwood” type, described as often including slash pine
and saw palmetto in the relatively wetter areas (Peet and Allard 1993:61, 65). What-
ever the dominant pine, “natural” flatwoods generally are highly stratified with a
high tree canopy (pines drop their lower limbs, sometimes a result of fire) and a
low plant understory.

Longleaf pine itself is most readily distinguished from other southern pines
by its long needles, 10 to 15" (25-38 cm), and large cones, 6 to 10” (15-25 cm) (Harrar
and Harrar 1962:51-60; Little 1980:291; Wahlenberg 1946:3). Longleaf has the po-
tential to live 500 years or more but usually trees are victims of storms, if not
humans, long before reaching such an age (Bengtson et al. 1993; Landers et al.
1995:39-40). Compared to other southern pines, longleaf is the most resistant to
disease, insects, and rot, adding to its value as timber wood. South Florida slash
pine is less resistant than longleaf but more resistant than typical slash pine.

Longleaf pines are intolerant of competition but remarkably tolerant of sur-
face fire; thus, frequent—at least once a decade and optimally every 2-3

Winter 2000 JOURNAL OF ETHNOBIOLOGY 2ry

years—low-intensity fires are the key to controlling the growth of competitors
such as hardwoods and even slash pines (Abrahamson and Hartnett 1990:132;
Landers et al. 1995:40; MacLaren and Stevenson 1993:407; Rebertus et al. 1993).
South Florida slash pine is less fire resistant than longleaf but more fire resistant
than typical slash pine (Abrahamson and Hartnett 1990:112, 131; Snyder et al.
1990:259). Along with fallen pine needles, highly flammable wiregrasses (Moore
1996a:18; Peet and Allard 1993:46-47) and saw palmetto (Arahamson and Hartnett
1990:129) provide fuel for the fires, usually ignited by lightning strikes. In the ab-
sence of human influence (either Indian or EuroAmerican), fires would have been
seasonal, primarily limited to the summer lightning season of April to mid Au-
gust (Myers 1990:185). In pre-EuroAmerican times, a single-ignition fire could burn
extensively without the limitations of roads and other human-made barriers. The
pines themselves withstand fire in part because of their multi-layered fire-resis-
tant bark (Snyder et al. 1990:259). Longleaf seedlings also regularly survive fire;
the seedlings of south Florida slash pine have a lower survival rate yet fare better
than those of typical slash pine (Small 1930:42; Snyder et al. 1990:259). Thus, longleaf
and South Florida slash flatwoods are especially fire-maintained and fire-depen-
dent. A high frequency of 2 to 3 fires a year would enhance and expand longleaf
stands (Rebertus et al. 1993) and slash pine stands as well. In addition to reducing
woody competitors, fire contributes to the germination of seeds (especially of
longleaf and the understory grasses) by producing appropriate soil conditions; to
turnover of litter, humus, and nutrients; and to increased vigor of some species
populations (Abrahamson and Hartnett 1990:129; Myers 1990:178).

In addition to wiregrasses, a high diversity of fire-adapted groundcover plants
in both longleaf- and south Florida slash-dominated flatwoods sustains a diverse
animal life (Abrahamson and Hartnett 1990:116; Engstrom 1993; Guyer and Bailey
1993; Johnson 1995; Moore 1996b:19). This is in part because many of the fire-
adapted plants produce new growth, providing food, soon after a fire has burned
through the forest. Pine seeds also provide food for many birds and small mam-
mals (Frost 1993:31; Wahlenberg 1946:179). Gopher tortoise (Gopherus polyphemus),
box turtle (Terrepene carolina), eastern diamondback rattlesnake (Crotalus
adamanteus), black racer (Coluber constrictor), pine woods tree frog (Hyla femoralis),
great horned owl (Bubo virginianus), bobwhite quail (Colinus virginianus), red-
cockaded woodpecker (Picoides borealis), turkey (Meleagris gallopavo), fox squirrel
(Sciurus ete feria nd eel er deer (Odocoileus virginianus) are some of the ani-

oods. Most, if not all, benefit from periodic fire. For example,
gopher else more typical of high pinelands (Myers 1990:186) but also present
in the drier flatwoods and scrubby flatwoods ( and Hartnett 1990:119),
cannot survive dense woody vegetation. The underground burrows of tortoises
serve as fire refuges not only for the tortoises but also for over 300 other vertebrate
and invertebrate animals (e.g., Dodd 1995; Folkerts et al. 1993:165-166, 181-182;
Myers 1990:186). Early EuroAmerican observers also recorded bison, black bear,
panther, red wolves, and even elk in the longleaf forests of the Southeast (Engstrom
1993:128).

Today there is general agreement among researchers that pre-EuroAmerican
pine forests differed from most present-day ones in that they had higher fire fre-

278 WALKER Vol. 20, No. 2

quencies, more uneven age structure, and a more open understory with greater
grass components and less shrub plants (Abrahamson and Hartnett 1990:104).
Researchers also agree that the reduction of fire frequency may be responsible for
much of the difference (Abrahamson and Hartnett 1990:104; Frost 1993:21, 34-35).

The American Indian Factor —The occurrence and distribution of woodlands (pine
and mixed hardwood) and other plant communities and how they changed
throughout pre-EuroAmerican history are increasingly being linked to human in-
fluence, and in particular to human use of fire (Delcourt and Delcourt 1997, 1998;
Delcourt et al. 1998; Pyne 1998). While there are many reasons recorded in historic
accounts (Wagner in press) for why American Indians set fires in eastern North
America, perhaps two of the most important ones were to sti ] plar its
for attracting wildlife and to drive game. While traveling in north Florida, Bartram
(1791:139) stated that “fires are set almost every day throughout the year in some
part or other, by the Indians, for the purpose of raising the game, as also by the
lightning.” Attracting wildlife may have been the primary reason for American
Indian management of Florida’s pine flatwoods, especially in pre-agricultural times
(generally before A.D. 1200 in north Florida). In south Florida where crop agricul-
ture was not practiced, attracting wildlife and improving visibility for hunting
undoubtedly would have been the primary reason for setting fires. White-tailed
deer and other game animals of the pine forests are highly visually oriented, need-
ing to see their surroundings (Johnson 1995:29).

To date, little research has focused on American Indian use of woodland fire
in Florida. One study by Kalisz et al. (1986), however, identified a spatial correla-
tion between archaeological sites associated with non-agricultural Indians and
present-day longleaf stands (occurring as “islands” in a landscape of predomi-
nantly sand pines) in north-central Florida’s Ocala National Forest. In addition,
quantification of wiregrass phytoliths in the soils beyond the present-day longleaf
stands strongly suggests that they were once more extensive. Kalisz et al. (1986:191)
hypothesize that “the longleaf pine islands were maintained through annual or
frequent burning by early humans; longleaf pine islands are prehistoric cultural
features.” Change in the natural fire regime of either sandhill pine stands (as in
the Ocala case) or pine flatwoods due to an increase in the number of fires and the
addition of a second burn season (winter dry season), if maintained, would have
resulted in forest expansion, especially where longleaf.
were present.

Fifty-six years earlier, in his study of south Florida slash pine “islands” within
the Everglades (“Everglade Keys”), botanist John K. Small (1930:41-42) hypoth-
esized about the ecological influence of American Indians:

or south Florida slash pines

...when the aborigines first occupied the Everglade Keys, they doubtless
found them clothed with hammock. ...without doubt, the aborigines pur-
posely set fire to the hammocks in order to drive the game into the open
places, thus facilitating their primitive means of hunting game. ... But there
had been developed plants that were fire-proof, so to speak, just for such
regions...the Caribbean-pine (Pinus caribaea)...the seedling pine-trees after
several years of uninterrupted growth will survive fire, and when a little

Winter 2000 JOURNAL OF ETHNOBIOLOGY 279

older they are normally perfectly fire-proof. Thus the pinewoods were
developed and have spread as the hammock retreated.

DEFORESTATION OF THE SOUTHEASTERN AND SOUTH FLORIDA PINE
FLATWOODS

Human-related factors involved in the complex process of pine deforestation
include American Indian agriculture, introduced European animals, the naval-
stores and logging industries, EuroAmerican agriculture, and reduction of fire.
Both American Indians and EuroAmericans contributed to the landscape change,
although clearly the latter played the greater role. Importantly, African Ameri-
cans, masked by the EuroAmerican economic histories, comprised the greater
percentage of the labor force for EuroA g | produc-
tion and for the naval-stores and logging industries.

American Indian Agriculture—In prehistoric times, ES ee Serica Indians
may have cleared portions of the more fertile ts have been
made to estimate how much pineland was impacted i in the southeast 1 region. Stud-
ies that identify and estimate agricultural lands surrounding large
Mississippian-period population centers are on the increase, but so far these have
focused on areas outside of the longleaf and slash pine ranges (see summary in
Wagner in press). Frost (1993:28) notes that Alabama Indian farmers may have
cleared much longleaf pineland for their ext Itural fields. Many towns
of these Mississippian-period agriculturalists were palisaded, representing an
additional impact on forests (Wagner in press), possibly including pinelands.
Moreover, palisades were replaced, sometimes several times over the occupation
of a site. Alabama’s Moundville, for example, was palisaded at least three times
using a minimum of 10,000 logs each time (Scarry and Steponaitis 1997). Maize
agriculture spread throughout north and central Florida after about A.D. 1200 and
may have impacted pinelands of these subregions. South Florida’s prehistoric In-
dians, on the other hand, did not practice agriculture.

Introduced European Animals.—Except for the possibility of Indian agriculture as a
significant factor, one might argue that the pine deforestation process, in a broad
sense, began in 1539 with the Spanish expedition led by Hernando de Soto (Smith
1968). De Soto entered the Southeast in west central Florida and brought with him
a food supply that included droves of pigs (Milanich and Hudson 1993:38), herd-
ing them along the exploration route throughout much of the southeastern range
of longleaf and slash pines. In addition to the resultant, immediate short-term
foraging and grazing of the forest groundlayer, unknown numbers of pigs are
believed to have escaped, forming the basis for a non-native feral population, one
that proliferated in the pine flatwoods. The use of pinelands for grazing contin-
ued and diversified when an area’s first EuroAmerican settlers learned—perhaps
from American Indians—that purposely set surface fires in the woods, especially
those with longleafs and south Florida slash pines, reduced the shrub layer (saw
palmetto, etc.) and produced new grass forage for their grazing animals. Romans
(1775:16) wrote of the north Florida longleaf forests “that immense stocks of cattle

280 WALKER Vol. 20, No. 2

are maintained, although the most natural grass on this soil is of a very harsh
nature, and the cattle not at all fond of it, it is known by the name of wire grass;
and they only eat it while young...the woods are frequently fired, and at different
seasons, in order to have a succession of young grass.” The periodic burning of
the forest floor by Indians and EuroAmericans benefitted the forests as did fires
ignited by lightning, and especially in the case of longleaf and south Florida slash
pine forests, perhaps even expanded them if their burning episodes represented
ane increase in fr Overs fire hin, wea? eee there was generally an important

Indian and E forest t. Prehistoric and
many eee pease Indians “fire-managed” pine forests primarily to increase
the abundance of native wildlife which ey | hunted for food. Hie inae th grasses,
wildlife, and fire were elements of the native flatwoo tem, EuroAmerican
livestock was not. Feral and domestic hogs and cattle and even sheep and goats
(in some areas), free from fencing as late as the 1950s in south Florida, fed on the
many grasses and pine seedlings in these open woodlands (Sargent 1884:492).
Departing from the pattern, however, historic-period American Indians in Florida,
notably the Seminole, also engaged in cattle-raising on the open range, first in
north and central Florida and later in south Florida. Great numbers of feral cattle,
many from Spanish origins, roamed the pinelands free for the taking.

The feral hog population had reached a saturation point across most of the
longleaf range by 1850, and probably earlier although pre-1840 documentation
doesn’t exist (Frost 1993:32). While the grasses may have benefited from hog and
cattle grazing, the collective rooting, grazing, and trampling of the non-native
animals proved to be too much for the pine seedlings, especially those of the slow-
growing longleafs. It is reported that a single hog in one hour can root as far as 30
feet, eating some eighty starch-laden longleaf seedlings (Walker 1991:129, 192-193).
Thus, feral hogs, in particular, were responsible for the destruction of countless
longleaf seedlings, preventing forest regeneration (Frost 1993:30-34; Schwarz
1907:94; Wahlenberg 1946:178-179). In addition, soil compaction and trampling
caused by these animals contributed to the inability of seedlings to survive
(Abrahamson and Hartnett 1990:146).

South Florida was still in many ways a frontier during the first half of the
twentieth century. For example, many south Florida cattlemen continued centu-
ries-old burning practices in the pine woods so that their stock could graze on
new grass growth (Akerman 1976:246-247; Franklin 1997:19; Zeiss 1983:118-119), a
practice that was compatible with pine forests if seedlings survived their fire-in-
tolerant stage. Wild pigs, on the other hand, still very populous in the 1940s and
1950s in south Florida, continued to consume pine seedlings in massive quanti-
ties, significantly ae the region’s source of forest regeneration.

Naval Stores and Logging. American settl realized more lucrative uses
for the longleaf and slash pine forests. The naval-stores industry faced trees (as
many as three or four sides of mature trees) and attached cups or boxes to collect
resin that was used for the production of rosin, pitch, tar, and turpentine (Butler
1998; Frost 1993:24-27; Mohr 1896:69; Wahlenberg 1946; Walker 1991:77, 146-151).
The first three products were enormously important to the shipbuilding industry,
while the numerous uses for turpentine varied from lamp oil to laxatives. The

Winter 2000 JOURNAL OF ETHNOBIOLOGY 281

1834 introduction of the copper still for turpentine distillation resulted in a prolif-
eration of turpentine operations (Butler 1998:72-73; Frost 1993:26-27). The still
allowed the resin to be reduced to turpentine at the extraction sites and thus saved
significant shipping costs. Mature stands of longleaf often produced for only about
four years (Mohr 1896:70). Pine trees tolerated extraction of resin but were weak-
ened significantly and thus became more vulnerable to fire, insects, and storms
(Mohr 1896:61, 72).

The most lucrative and most destructive of all the pine industries was timber-
ing. The tall, straight longleaf pines with their rot- and insect-resistant wood, for
example, made excellent ship masts, long-lasting dock pilings, and when milled,
made beautiful homes. Southern longleaf pine, in general, had the reputation in
European, Caribbean, and South American markets of being North America’s stron-
gest wood due to its density (Mohr 1896:53). Initially, transporting longleaf and
slash pine logs to the mills was a a slow and difficult task. Logs were floated via
natural and human-excavated waterways thus, the area of forest that could
be logged was limited to that which had access to the waterways.

That limitation vanished with the nineteenth-century arrival of the steam-
driven locomotive and railroads to the southeastern forests. In addition to the
locomotive, steam-powered log skidders, sawmills, and circular saws contributed
significantly to the new logging technology. Almost as soon as the main rail lines
were laid by railroad companies, lumber panies leased logging rights or bought
extensive acres of forested lands adjacent to the lines. Logs were taken from the
woods to the sawmills by railcars pulled by a steam locomotive. Due to this accel-
eration of the logging industry based on steam technology, most of the region’s
remaining old-growth longleaf and slash pines forests were clearcut between 1870
and 1920 (Frost 1993; Wahlenberg 1946). Just as EuroA ttl t had been
late coming to Florida, especially the southern half of the peninsula, the state was
late in receiving attention from the railroads. The logging of south Florida’s pine
forests began in the 1920s. Old-growth pines were still being logged in this subre-
gion in the 1950s although much of the focus had shifted to cypress in the Big
Cypresss Swamp and Fakahatchee Strand (Tebeau 1957).

EuroAmerican Agriculture. Bagi broad-scale logging, many pinelands were
I fields

cleared by E the purpose of
Much of the landscape across the region was converted to cotton plantations in
the 1800s. Later, especially after 1940, many logged pinelands and old plantation
lands were planted in slash or loblolly pines. Slash (primarily the typical P. elliottii
elliottii) and loblolly were thought to be fast-growing (due to their early rapid
growth) compared to the longleaf, and thus were considered more economical to
grow, ignoring the higher quality of longleaf wood. Dense plantations of slash
and loblolly, with trees planted in neat rows, became the accepted management
approach in forestry practices on public-, industry-, and other private-owned lands.
In still other areas of the Southeast, including parts of south Florida, citrus groves
and non-woodland cattle pastures replaced the old-growth flatwoods.

Reduction of Fire-—The reduction of fire frequency in the Southeast’s pine forests
intensified with the progression of EuroAmerican settlement. As roads and agri-

282 WALKER Vol. 20, No. 2

cultural fields became more numerous, the pine woodlands became more frag-
mented, requiring a higher frequency of fire ignitions to burn large areas. Prior to
fragmentation, a single lightning ignition could burn extensively across the land-
scape. With fragmentation, fire was effectively eliminated from many parcels of
pine forests (Frost 1993:34). Thus, early fire suppression was perhaps an uninten-
tional result of EuroAmerican settlement. Many of the logged Southeast lands,
including longleaf woodlands, experienced serious erosion and flooding (Walker
1991:170-175). This situation, combined with poor agricultural practices, led to the
Southeast’s navigable rivers being muddied and even clogged. As a result, the
federal government began in 1911 to buy the logged lands to protect the Southeast’s
watersheds. In this manner, over 10 million southern acres were added to the Na-
tional Forest system, and trained foresters took on their management (Walker 1991).
Nonetheless, many millions of acres, especially those forestlands that supported
longleaf pine, were not allowed to regenerate naturally.

Perhaps the most critical barrier to regeneration was what might be called the
“Smokey Bear Myth.” Although purposeful fire-suppression steadily followed the
progression of EuroAmerican settlement, the U.S. Forest Service’s Smokey Bear
campaign, culminating in the 1950s, left no doubt in the minds of Americans that
all forest fires were destructive and dangerous, and were not to be allowed under
any circumstances (Landers et al. 1995:41; Moore 1996c:22; Walker 1991). Because
foresters did not understand the beneficial role of frequent surface fires (e.g., Mohr
1896:62), they unknowingly contributed to the degradation of the pine forests.
Without frequent surface fire, the forest floor became thick with pine needles and
cones and the shrub layer grew dense, all providing fuel for highly destructive
fires when fires did occur. Without fire, the longleaf pines were eventually out-
competed by other pines and hardwoods, the slash pines were often out-competed
by hardwoods, and the various understory plants and animals specifically adapted
to the longleaf and slash forests declined in abundance (Peet and Allard 1993:46).
Even in the relatively remote rock pinelands of today’s Everglades National Park,
twentieth-century fire suppression resulted in a reversal of Small’s (1930) hypoth-
esized scenario in that a succession toward hardwood hammock has occurred
(Hofstetter 1974:203).

DEFORESTATION OF SOUTHWEST FLORIDA'S PINE FLATWOODS

Archaeological Survey and Historical Research of the HCMP.—During January of 1996,
Robin Denson (Gulf Archaeology Research Institute) and I conducted an archaeo-
logical survey on a tract of Lee County-owned land just south of the Caloosahatchee
River and east of Fort Myers in southwest Florida (Figure 1 inset) (Walker et al.
1996). Prior to and during this same time, I also conducted historical research and
a series of interviews with long-time local residents. Much of the area today is
characterized by seasonally wet south Florida slash pine/saw palmetto flatwoods
and dense saw-palmetto prairies. The county property, known as “Hickey Creek
Mitigation Park” (HCMP), was named for Hickey’s Creek (after nineteenth-cen-
tury settler Dennis O. Hickey) which runs through it toward the Caloosahatchee.
In part, the park is intended to be a preserve for gopher tortoises in perpetuity to

Winter 2000 JOURNAL OF ETHNOBIOLOGY 283

offset tortoise habitat destroyed elsewhere in southwest Florida (Roger Clark, per-
sonal communication, 1996; Riley et al. 1993), hence, the use of the word
“Mitigation.”

Our archaeological survey documented five American Indian archaeological
sites on the park property. Artifact collections include primarily a few pottery
sherds, one bone pin, and one bone point; no other faunal remains or other date-
able organic materials were found. The sherds are all of the Sand-tempered Plain
type, also known as “Glades Plain,” and are only roughly diagnostic of time pe-
riod. Because they are not very thick, a post-A.D. 500 date is suggested. These are
all small sites and four are associated with the banks of Hickey’s Creek (Walker et
al. 1996). One site tenuously was based on a single chert flake likely produced
from working or reworking a projectile point. Unlike other sites, it is located in the
middle of today’s slash pine/saw palmetto flatwoods. Larger sites are reported
for the mouth of Hickey’s Creek on the Caloosahatchee River and just to the south
of the park property on Hickey’s Creek. The latter produced a relatively large
sample of pottery sherds that suggest a post-A.D. 500 habitation, more long-lived
than the small sites within the park. It is possible that all sites are contemporane-
ous. The two large sites may have been the main habitation villages for the area
while the smaller kside sit y] | short-term hunting / fishing camps.
The chert flake may have been lost during a hunting episode in the flatwoods. The
bone point also suggests food procurement, associated with either fishing or hunt-
ing. Freshwater and periodically estuarine fishes would have been available in
Hickey’s Creek and white-tailed deer, raccoon, opossum, turkey, gopher tortoise,
quail, and other game animals would have inhabited the flatwoods, all offering
substantial food resources for the Indian residents.

The EuroAmerican homesteader of the Hickey’s Creek area was Dennis O.
Hickey (Little in Walker et al. 1996:Appendix A) who during the post-Civil War
decades farmed, growing “large crops of cabbage, eggplant and squash” (Grismer
1982:109), “raised” cattle in the woodland tradition and also operated a store in
Fort Myers (Little in Walker et al. 1996:Appendix A). Also, during the period of
1870-1926, cattle drives (Dodrill 1993:10), some led by Hickey, regularly pushed
through both the Hickey’s Creek (Little in Walker et al. 1996) and Cape Coral (Zeiss
1983:26; 111-113) areas grazing and trampling in the pine woods on their way to
Punta Rassa where the animals were then shipped to Cuba. Hickey’s descendent,
Mrs. Beverly Little, believes the location of Hickey’s home, however, was beyond
the boundaries of our survey parcel.

The park includes the archaeological remains of a logging rail system, two
logging camps, and associated refuse dumps, all dating to the 1930s and 1940s.
Only the younger of the two camps had been recorded with the Florida Site File.
Our primary informant, Mr. Dan Garner (Figure 2), told us that an earlier, 1930s
camp and rail line had existed, and he took us to this location in what today is a
dense, high saw-palmetto prairie (Walker et al. 1996: Appendix D). There, the sur-
vey crew found a few surface artifacts reflecting the decade.

We soon learned that the Hickey’s Creek area was an important part of what
once was an extensive logging network run first (1924-1929) by the J. W. McWilliams
Lumber Company and later (1929-1944) by the Dowling & Camp Company (Pickens

284 WALKER Vol. 20, No. 2

in Walker et al. 1996: Appendices B and C). Two brothers, William and James, were
sons of Thomas Dowling, who ran a logging operation in north Florida along the
Suwannee River centered at Dowling Park (Anonymous 1988; Doris Dowling
Crews, personal communication, 2000). Vaughn Camp was of the Camp family,
which centered its extensive operations in Franklin, Virginia (Rouse 1988). Dowling
and Camp’s logging network included what are today two major population ar-
eas of Lee County—Cape Coral and Lehigh Acres—the former located on the north
side of the Caloosahatchee and the latter located just to the south of Hickey’s Creek.

At the beginning of the HCMP survey project, we assumed that only south
Florida slash pine had been logged from the Hickey’s Creek and Cape Coral areas.
But after our initial historical and oral history work, we began to consider that the
logged forest adjacent to and south of Hickey’s Creek also may have included
significant longleaf pine and wiregrass components (Walker 1997; Walker et al.
1996). Both areas share in large part a common soil association, the Pineda-Boca-
Oldsmar, which falls into the category of nearly level, poorly drained, deep sandy
soils with a pine flatwoods association (USDA SCS 1984).

Despite the poorly drained soil association, the land south of the river was
recorded by Vignoles as “high pine” land on his 1823 natural history map of Florida.
He typically used “high pine” to refer to longleaf pinelands similar to its use to-
day (Myers 1990:153, 174). Botanist John Harshberger (1914) was more explicit
when he stated that longleaf occurred mixed with the more dominant south Florida
slash pine, both north and south of the Caloosahatchee. Efforts to locate company
records that might more clearly identify the species of logged pines—through

FIGURE 2.—Author’s primary informant, lumberman and cattleman Mr. Dan Garner of
Alva, Florida, was interviewed in January of 1996 near Hickey’s Creek.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 285

Dowling relatives and other avenues—were unsuccessful. A Fort Myers 1926-1927
telephone directory listed the McWilliams Company as manufacturers of “Rough
& Dressed South Florida Dense Long Leaf Yellow Pine” lumber (Walker et al.
1996:47). During the time of this listing, the McWilliams Company was logging
Cape Coral (Zeiss 1983:98-108). Interpretation of the listing is not straight-forward.
Both south Florida slash and the longleaf pine produced wood that was more
dense than the northern slash pine. Might McWilliams have been advertising both
south Florida slash and longleaf pine with no distinction, in light of Harshberger’s
mixed longleaf/slash record for Lee County and Wahlenberg’s (1943:268) point
that high quality pine wood was often sold as “longleaf,” regardless of species?
The name “longleaved yellow pine” is listed by Mohr (1896:13) as one of the com-
mon names for longleaf and thus suggests that at least some of the Cape Coral
pine was longleaf. And furthermore, Sudworth’s map records the presence of
longleaf in the Hickey’s Creek area but only slash pine to the south in the Lehigh
Acres area, suggesting a longleaf component there. It is probably no coincidence
that the areas targeted for logging first by McWilliams and later by Dowling and
Camp were areas indicated on Sudworth’s U.S. Forest Service map as including
longleaf pine.

A lifelong local resident, lumberman/cattleman Dan Garner (Walker et al.
1996:Appendix D) (Figure 2), during the archaeological survey, described the old-
growth forest just south of the Caloosahatchee at Hickey’s Creek:

That was the most beautiful pine you ever seen in your life. You just go
out there, and you could see a turkey and anything else...there weren’t no
weed, no palmettos, no nothing. Heart pine. Big heart pine. ...it wasn’t
near this rough [with high saw palmetto like today]...lots of tortoises and
hogs out here...deer, turkey, bobcat...when I was a boy, this was the best
place in the world...you could kill all the game in the world.

Unfortunately, Mr. Garner and other local residents of the area knew these
pine trees only as “heart pine” or “yellow pine.” (Mr. Garner intended to take me
to see a mature pine forest in the Lake Okeechobee area—one that is very similar
to how he remembered Hickey’s Creek’s old-growth forest, but he died before we
could go.) The same common-language problem is true of the Cape Coral oral
histories that Zeiss (1983) collected. The phrases “virgin pine,” “first growth trees,”
and “heartwood” appear throughout—but no mention of species. Most of Cape
Coral was described as “high pineland” by early residents (Zeiss 1983:180).

The pre-logged areas of Cape Coral and Hickey’s Creek may be characterized
best by seasonally wet mixed south Florida slash and longleaf flatwoods. The wetter
Lehigh Acres locale probably supported mostly slash pine and both the Hickey’s
Creek and Lehigh Acres locales included some areas of pond cypress wetland.
Based on Sudworth (1913) and Harshberger (1914), Lee County’s pre-logged
flatwoods, while many included longleaf, were dominated by the south Florida
slash pine. This characterization is a revision of my earlier hypothesis of longleaf-
dominated flatwoods for northern Lee County (Walker 1997; Walker et al. 1996).

Cape Coral and Hickey’s Creek/Lehigh Acres Logging Operations.—Typically, lumber
companies clear cut southeastern old-growth pine forests, moved on to the next

286 WALKER Vol. 20, No. 2

FIGURE 3.—Photograph taken in north Florida of a mule-drawn high-wheeled log cart
(with a longleaf pine log) fitting cattheman Mark Bateman’s description of the circa-1940-
1944 carts at Hickey’s Creek.

area to be logged, and sold the logged land as soon as possible. The Cape Coral
and Hickey’s Creek/Lehigh Acres operations followed this pattern. Once the At-
lantic Coast Line completed a line to Fort Myers and later the Seaboard Air Line
Railway Company completed a line from Fort Myers east into interior south Florida
(Grismer 1982:233-234; Turner 1999:33-36; Walker et al. 1996: Appendices B and Kk),
nearby pinelands were purchased or leased by the lumber companies. McWilliams
began logging in 1924 and cut pine to build a large sawmill and houses at Slater
(Zeiss 1983:99) in what is now North Fort Myers (Figure 1, inset). McWilliams
and, after 1929, Dowling & Camp logged the pine flatwoods of Cape Coral. Dowling
& Camp later logged the pine flatwoods of the Hickey’s Creek and Lehigh Acres
areas from 1932 to 1935 and 1940 to 1944 (Walker et al. 1996). The intervening
years were spent logging an area in neighboring Hendry County also on the south
side of the Caloosahatchee River. This may be in part a locale depicted by Sudworth
(1913:Map 35).

Company rail crews laid “spurs” into the pine flatwoods; the rails were laid
on ties hewn from pine. A logging crew of about 100 men cut 100,000 board feet a
day, about 800 to 1,000 trees a day in the Hickey’s/ Lehigh area (Garner in Walker
et al. 1996:56:Appendix D). Estimates of 50,000 to 120,000 board feet a day are
reported for Cape Coral (Board and Bartlett 1985:115; Zeiss 1983:100). Trees were
felled by axe or reciprocating saw and logs were chained to high-wheeled carts,
and then pulled, dragging one end, by teams of mules to the rail spur (Figure 3).
Cattleman Mr. Mark Bateman (Walker et al. 1996: Appendix H), a local resident,
recalls the scene from his youth:

Winter 2000 JOURNAL OF ETHNOBIOLOGY 287

What I was impressed with — because as a young kid, watching the mules
pull the logs out to the road...all the leather and chains [of the big-wheel
log carts] and everything going together and hearing the mule skinner
with the whips and what have you. ... they had the big chain wheels, you
know, so high...sand wheels and they’d back over the logs. ... They had
the steam engine. ... But they snaked everything to the edge with mules. ...
That was something to see. I can hear it and see it just as plain as you and
I talking right now.

The 1940s logging episode saw the addition of at least one Caterpillar tractor,
operated by Mr. Garner, to the Hickey’s/Lehigh operation (Garner in Walker et al.
1996:Appendix D); one artifact collected during the archaeological survey is a ca.
1940s Caterpillar clutch disc. Mules, however, continued to be the primary haul-
ers of logs out of the woods. Mr. Garner also noted the addition of an electric saw
toward the end of the operation, ca. 1943-1944. Logs were loaded onto flatcars
using steam-powered draglines and a company-owned steam locomotive (fueled
by pine slabs) then pulled the logs to Slater Mill. At one time, Dowling & Camp
operated with ten locomotives. In 1944, seven remained. Engine #103 (Figure 4)
was used to remove logs from the Hickey’s/Lehigh area, taking them to Slater.

As soon as an area was “cutover,” rail crews picked up the iron spurs and re-
laid them in new, uncut areas of forest (Garner in Walker et al. 1996: Appendix D;
Zeiss 1983:102). The railroads and their rail spurs, even when taken up, left visible
grades, especially in south Florida where beds often were raised to avoid the sea-

FIGURE 4.—Dowling & Camp’s Engine 103 hauled pine logs from the Hickey’s Creek
operation to the mill at Slater. Photo courtesy of James Pickens.

288 WALKER Vol. 20, No. 2

sonally flooded lowlands. The grades are usually paralleled by excavation ditches
as is the case at Hickey’s Creek. Lost railroad spikes and spent ties are often found
in the ditches. In addition, as the logs were dragged from the woods, they left
linear “scars” in the ground, all leading to the closest rail spur. Studied from aerial
views, the spurs and log scars can be traced, revealing dendritic or feather-like
patterns (Pickens in Walker et al. 1996: Appendix B). A series of aerial photos taken
in 1944 covering the two Lee County areas documents the spurs and log scars,
which allowed Mr. James Pickens to reconstruct the logging system (Figure 1, in-
set). The feathery patterns show the two major components of the system. The
eastern Hickey’s/ Lehigh component is the smaller of the two. The larger, western,
Cape Coral component originated at Slater Mill where logs from both areas were
milled until the mill and all logging closed down in 1944 (Board and Bartlett
1985:115; Godown and Rawchuck 1975:108; Walker et al. 1996:Appendix F; Zeiss
1983:99).

Cape Coral and Hickey’s Creek Logging Camps.—Temporary camps for the logging
and rail crews and their families were established in the woods. Typically, only
one woods camp would exist at a time. Zeiss interviewed several individuals who
remembered various camps in the Cape Coral area. Locations for at least four
camps were described (Zeiss 1983:103, 105). One of these consisted of “shacks”
and others used boxcars or railroad passenger cars for housing. Detailed memo-
ries of the Hickey’s/Lehigh logging operation and its camps come from Mr. Garner

FIGURE 5.—Photograph taken in a west Florida longleaf forest shows a boxcar logging
camp similar to those described for Hickey Creek circa 1932-1935 and 1940-1944 (mules
were used at Hickey’s Creek instead of oxen). Photo courtesy of Florida State Archives,
Tallahassee, FL.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 289

(in Walker et al. 1996: Appendix D). When about 10 years old, Mr. Garner frequented
the ca. 1932-1935 camp and while in his teens he worked with the logging crew of
the ca. 1940-1944 camp. Both camps consisted of railroad boxcars serving as year-
round, portable homes for the logging and rail crews and their families. Both crews
of both camps were African American. The crew supervisors were EuroAmerican
and lived in the nearby town of Alva and elsewhere. Mr. Frank Gay at one point
supervised the Hickey’s Creek crew (Mrs. Serena Gay, personal communication,
1996). A photograph taken in a northwest Florida longleaf forest shows a boxcar
logging camp (Figure 5) similar to the one described for Hickey’s Creek except
that mules were used instead of oxen. The camps had outhouses, and although
temporary, the 1940s camp had substantial government [WPA]-built privies with
cement foundations.

Both camps had commissaries for purchasing groceries, dry goods, and per-
sonal items. The 1930s commissary, like the workers’ homes, was a boxcar. This
was probably also the case with the Cape Coral camps since there was a large
commissary not far away at Slater. The 1940s Hickey’s Creek commissary, on the
other hand, was a substantial one-story structure built of “heart pine” lumber.
Workers were paid with company “scrip” and aluminum tokens, a common prac-
tice among logging companies, particularly during the Depression era. Children
rode a bus to attend school in Fort Myers. A medical doctor visited once a week
from Fort Myers and administered medicines contained in bottles such as those
found during our survey (Walker et al. 1996). Mr. Garner describes camp life with
images of children, baseball, sour-orange wine, whiskey made from cane-skim-
mings, and “good times.” Vegetable gardens, i y pork and beef, and local
wildlife including gopher tortoises, raccoons, and fish, were central to the diet of
the woods community.

Post-Logging Decades.—Like many of the Southeast’s pine flatwoods, those of the
Cape Coral and Hickey’s/Lehigh areas were clear cut. In addition to removing
the seed source, the logging activity greatly disturbed seedlings that were present,
along with the seedbed itself. Combined with the destructive feeding behavior of
feral pigs, the competition from fast-growing oaks, the reduction of fire, the intro-
duction of citrus and other agricultural crops, and open-range cattle grazing, the
mixed longleaf/slash pine forests had little opportunity to regenerate. The remain-
ing old-growth pine stumps at Cape Coral were extracted from the land and
transported to Mississippi and to Brunswick, Georgia for use in naval-stores prod-
ucts (Zeiss 1983:180). At some point, stumps at Hickey’s Creek also may have been
taken out (Roger Clark, personal communication, 1996; Riley et al. 1993:22); we
observed telltale depressions in the ground during our survey. Taking advantage
of the highly desired dense pine to the very end, landowners salvaged the lumber
out of the old Hickey’s Creek commissary building during the 1950s (Crawford in
Walker et al. 1996:Appendix I) to use elsewhere.

For a while, the land that today is the county’s HCMP and is largely in south
Florida slash pine, scrub oaks, and saw palmetto, was used for cattle grazing by
cattlemen, including Mr. Garner. Through the late 1940s, the 50s, 60s, 70s, and 80s,
Garner and others conducted burns in order to provide new grass growth for their
cattle, a longstanding woodland-grazing tradition (Garner in Walker et al. 1996:Ap-

290 WALKER Vol. 20, No. 2

pendix D). Despite these burns, however, the Hickey’s Creek pine forest only par-
tially recovered from the clear-cut logging. During these same decades, feral pigs
were still in abundance and citrus groves were planted in some of the area (Little
in Walker et al. 1996:Appendix A).

During the 1940s and 1950s, cattle also continued to be an important element
of the Cape Coral and Lehigh Acres landscapes but this use of those logged lands
came to an end during the latter part of the 1950s. Lee County’s human popula-
tion increased dramatically in the post-war years, a time of housing shortages.
And many WWIIservicemen who had been stationed in Fort Myers returned with
their families to establish new homes. So, not surprisingly, most of the cleared
land in the Lehigh Acres locale, first transformed into ranchland, soon (by 1954)
ended up under the ownership of a development firm initially called Lee County
Land and Title Company, and later, Lehigh Development Corporation (Dodrill
1993:6). The developers’ marketing strategy to lure families to Lehigh Acres in-
cluded a 1961 promotion in which a new home was offered as Grand Prize on the
TV show “The Price is Right” (Figure 6) (Board and Bartlett 1985:186). Similarly, in
1958, a massive housing development was initiated in the western sector of Lee
County’s logging system (Dodrill 1993; Zeiss 1983). Today Cape Coral (Figure 1,
inset) has become, landwise, the second largest city in area in the south next to
Jacksonville, Florida (Gainesville Sun, Sept. 11, 2000).

FIGURE 6.—In 1961, TV game show The Price is Right offered as Grand Prize a new
south Florida (Lehigh Acres) home located in the logged pinelands that were part of the
Hickey’s Creek/Lehigh pine logging system. Photo from Board and Bartlett (1985).

Winter 2000 JOURNAL OF ETHNOBIOLOGY 291

SCALAR PERSPECTIVES

Following the introduction of European domestic animals, the degradation of
old-growth forests accelerated and largely followed the transgression of non-Span-
ish EuroAmerican settlement (Frost 1993; Wahlenberg 1946; Walker 1991). The
process was slow at first, in the eighteenth century, and intensified with the ar-
rival of railroads in the nineteenth and tietl turies that were pushing farther
and farther into the southern states. Thus, while southeastern Virginia was the
first subregion to lose the longleafs on a massive scale, mostly in the eighteenth
century, south Florida was the last, losing its old-growth longleafs and south Florida
slash pines in the 1920s through the 1950s.

The local historical ecology of Lee County’s pine forests at Cape Coral and
Hickey’s Creek/Lehigh Acres may be largely typical of the process of landscape
change that occurred with other south Florida pine forests. One important differ-
ence, however, stands out. The majority of Lee County’s pre-EuroAmerican pine
flatwoods may have been characterized by a mixture of south Florida slash and
longleaf pines, with longleaf representing the southernmost limit of its range. It
may be more appropriate to conceive of two south Florida subregions in terms of
pine forests. One is the transitional south Florida where longleaf diminishes in
dominance, mixes with south Florida slash until a point is reached when only
slash occurs. The latter situation of “pure” south Florida slash flatwoods is the
second south Florida subregion.

The reconstructed pre-E American position of south Florida’s pinelands
is in reality nothing more than a reconstruction of one ecological episode in the
historical continuum. Representing today’s southernmost extent of longleaf, this
marginal subregion is the ideal area to test for the long-term climatic episodes—
commonly known as the Roman Optimum, Vandal Minimum, Warm Medieval
Period, and Little Ice Age—of the past two millennia. During the cooler/drier
periods (VM, LIA), longleaf may have expanded farther into south Florida while
during the warmer /wetter episodes (RO, WMP), longleaf may | treated north
The nineteenth and early twentieth centuries correspond with the end of the LIA;
thus our perception of a reconstructed pre-EuroAmerican forest might better be
situated in the LIA, an episode of hypothesized longleaf expansion. And our con-
cept of a twenty-first-century range of south Florida pine flatwoods and their
composition might better be situated in our current warm and wet trend with a
hypoth 1 retreating longleaf distribution. Unfortunately, tracking of fluctuat-
ing pine forest composition through time awaits the development of methods to
identify the archaeological and paleoecological remains of longleaf versus slash

At the local scale, the historical ecology of Lee County’s pinelands is generally
similar to that of the greater southeastern Coastal Plain region, but important dif-
ferences exist. First, although it remains to be tested (if possible), I hypothesize
that like many subregions of the Southeast, Lee County’s American Indians fire-
managed their local pine flatwoods. However, whereas in other areas of the
Southeast, clearing for agriculture by American Indians may have been a factor, it
wasn’t in south Florida. Rather, here the purpose likely would have been for main-
taining game populations, especially those of white-tailed deer.

292 WALKER Vol. 20, No. 2

The impact of cattle and especially pigs may have been longer, extending later
in time in south Florida than in other southeast subregions. Still the frontier in the
nineteenth and early twentieth centuries, south Florida continued its open-range
tradition (historic American Indian and EuroAmerican) and cattle drives
(EuroAmerican) through both the Cape Coral and Hickey’s/Lehigh locales on the
way to Punta Rassa for shipment to Cuba as late as the 1920s. Even so, it might be
argued that cattle and pinelands may have been compatible rather than in conflict
with the pinelands. Although today most of the cattle industry exists to the east in
interior lands of south Florida, feral pigs are still a challenge for management and
restoration of the HCMP pinelands.

One el tin the I J le Sout} t trajectory that may be largely miss-
ing from Lee County’s forest history is naval-stores production. Despite the fact
that the Hickey’s Creek property at one time was owned by Consolidated Naval
Stores Corporation (Walker et al. 1996:69-70), I found no record or memory of ac-
tual turpentining. Zeiss’s (1983:98) explanation for the absence of turpentining in
the Cape Coral area is that the pines were more valuable as lumber for building
material because it “was loaded with pitch, which served to protect it against dry
rot as well as from invasion by termites.” Perhaps by the time broad scale indus-
trial exploitation of forests reached south Florida, the importance of turpentining
had faded and clear-cut logging was economically more desirable.

Another difference is that fire suppression may not have been as important a
factor in the inability of the Cape Coral and especially Hickey’s Creek forests to
regenerate: during the post-logging years, the area’s cattlemen regularly burned
the logged lands. For Cape Coral and Hickey’s Creek, the impact of clear-cut log-
ging (destruction of the seed source) followed by the pressures of cattle and
feral-hog grazing, trampling, and rooting, the introduction of citrus trees and other
agriculture (in some areas), was too great for any remaining longleaf seedlings.
Furthermore, Cape Coral and the southern portion of the Hickey’s Creek logging
system (Lehigh Acres) were quickly transformed by developers into residential
communities. During this time (1950s), however, a close watch was kept and fires
were suppressed in the Cape Coral area, as more and more new residents arrived.

Another difference, more temporally related, is that logging everywhere in
south Florida was more rapid than in more northern subregions. This was due, at
least in part, to advances in logging technology during the 1940s. The operation at
Hickey’s Creek combined the old ways—axe, mule and cart, railroad, locomo-
tive—with some of the new ways—electric saw, tractor—although the old still
dominated. By the 1950s, trucks were regularly replacing the need for railroads
and locomotives in some areas of south Florida (e.g., Collier County). Again, we
see two south Florida logging histories, one characterized by a transition in tech-
nology, the other by an essentially modern technology.

CONCLUSION

What became of the once extensive old-growth southeastern pine forests in-
volved a long process of dynamic interplay between numerous environmental
and cultural factors possibly beginning as early as A.D. 800 in some parts of the

Winter 2000 JOURNAL OF ETHNOBIOLOGY 293

region. Generally (i.e., at the long-term regional scale), the same process of land-
scape change happened across the entire Southeast distribution of longleaf pine.
A historical ecology approach to southeastern pine deforestation, however, con-
tributes to the recognition of heterogeneity within the process of this broad-scale
landscape change. In particular, examination of the southernmost margin of
longleaf clarifies the extent of its pre-EuroAmerican penetration into southwest
Florida (at least for the LIA). The study of Lee County’s pine flatwoods from a
historical ecology approach has resulted in the hypothesis that longleaf pine was a
component of Cape Coral’s and Hickey’s Creek’s pine forests. Intergrading with
south Florida slash pines, these longleafs would have been the southernmost of
their range. Harris (1999) makes the point that south Florida’s tropical forests have
been under documented and thus under appreciated. The same can be said for
south Florida’s pine forests.

Southwest Florida experienced the longest history of pre-logging pine defor-
estation with perhaps one of th t of logging It was a subregion
of transition where longleaf and south Florida slash pines intergraded and where
old and new logging technology came together, but also an area where the
longstanding tradition of compatible fire-managed woodland grazing persisted
into modern decades. To cap the processual continuum, large portions of Lee
County’s logged old-growth pinelands were transformed into two of the earliest
post-war massive suburban housing developments, the beginning of a new era
for south Florida—one of enormous human migration to the Florida’s subtropics.

Estimates for upland landscape changes are presented by Frost ( 1993:19-20).
An astonishing 85 percent of the Coastal Plain’s pre-EuroAmerican uplands in-
cluded longleaf pines (71 percent consisted of longleaf-dominated uplands). Slash
pine, on the other hand, is estimated to have characterized only 3.3 percent of the

re-EuroAmerican uplands. Estimates for 1990 are a stunning 2.6 percent for “natu-
ral” longleaf (2.1 percent for longleaf-dominated uplands) and 0.4 percent for
“natural” slash pine uplands, with successional mixed hardwood-pine forests (44
percent), croplands (20.8 percent), pine plantations (15.2 percent), developed lands
(10.2 percent), and pasture (6.4 percent) having replaced the old-growth native
pine forests.

According to a 1995 inventory, longleaf pine acreage continues to decline in
the greater Southeast and in Florida (Outcalt and Sheffield 1996:2, 20). Most losses
have occurred on privately owned lands. Because remaining stands on private
lands are continuing to reach saw-timber size, losses will most probably continue
at a high rate. Based on a study of North Carolina longleaf, Frost (1993:21) figures
that few existing stands are being fire-maintained and as a result the majority of
stands are heavily invaded by hardwood paces. If this pattern is typical of the
t less than 0.7 percent of the pre-EuroAmerican

Southeast region, F a a
longleaf forests remains under’ “natural” conditions : t efforts on county,
state, federal, and try and p , Boyette
1996). However, of the longleaf states, only Texas shows small i increases on both
public and industry lands (Outcalt and Sheffield 1996:20). No increases are shown

for private lands.
Restoration efforts aimed at both longleaf and slash pine forests include new

294 WALKER Vol. 20, No. 2

management plans that emphasize periodic burning of the forest ground layers.
Efforts by conservation | groups large and small, such as The Nature Conservancy,
Tall Timbers R (north of Tallahassee, FL), Longleaf Alliance Johnson
1996; Longleaf Alliance n.d.) of Auburn University’s School of Forestry, and the
Longleaf Partners Funds/Longneedle Press (Moore and Goodwin 1995, 1996) and
Longleaf Ecology and Forestry Society (LEAFS), both of Gainesville, Florida, are
educating the public and landowners of the values of restoring native longleaf
ecosystems. For example, a recently published management guide for landown-
ers (Franklin 1997) provides guidelines for burning practices and for compatible
timber and cattle production, once again following the centuries-old tradition of
pineland grazing. The developing trend in landowner education is the promotion
of compatibility between longleaf reforestation and economic viability (e.g.,
Franklin 1997; Landers et al. 1995). Modern studies show that with appropriate
management, overall longleaf growth rates are comparable to the other pines on
most lands (Franklin 1997:5).

Southwest Florida’s Lee County together with the Florida Game and Fresh
Water Fish Commission have initiated reforestation in the new HCMP, planting a
mix of south Florida slash and longleaf pines. A restored, fire-managed pine forest
would be good habitat for a gopher-tortoise preserve, fulfilling one conservation
goal of the HCMP. In addition, an archaeological National Register nomination
(for the multiple historic logging sites) and a public education program including
on-site ecological and historical interpretation and trails are being considered for
the near future.

ACKNOWLEDGMENTS

This paper is in part based on the 1996 final report for the Hickey’s Creek Archaeological
Survey project, co-authored with Robin L. Denson and Gary D. Ellis of the Gulf Archaeology
Research Institute (GARI) and submitted to Lee County’s Division of Public Parks &
Recreation Services, Fort Myers, Florida. It is an expansion and revision of the author’s
earlier hypotheses concerning the presence and extent of longleaf pine at the HCMP.
Numerous agencies, groups, and individuals contributed to the Hickey’s Creek project.
Listed alphabetically, some of the most prominent are Gloria Andrews, Mark Bateman, Jan
and Robin Brown, Roger Clark (Lee County Parks & Recreation), Raymond Crawford,
Russell Dorsey (GARI), Charles Foster (Alva Museum), Dan Garner, Bill Gifford (GARI),
Carol Goodwin (Goodwin Heart Pine), Larry Harris (UF), Bud and Shirley House (Randell
Research Center [RRC]), Beverly and Bobby Little, Kevin Lollar (Fort Myers News Press),
William Marquardt (Florida Museum of Natural History [FLMNH]), Kent Perkins (FLAS/
FLMNH), James Pickens, Donna Ruhl (FLMNH), Gloria Sajgo and Annette Snapp (Lee
County Planning), Anna and Ray Stober (RRC), Jim Radz (GARI), Mark Renz, Self & Rost,
Inc. Engineers, Southwest Florida Archaeological Society, Southwest Florida Historical
Society, Don Taggart, Fred Tyers, Ken Wishlow, and Howard Yamataki (Soil Conservation
Service). In addition, Sue Ellen Hunter produced Figure 1, Myrna Sulsona translated the
abstract, and Donna Ruhl, William Marquardt, Doria Gordon, Roger Clark, Gail Wagner,
and anonymous reviewers suggested improvements to the article.

Winter 2000

JOURNAL OF ETHNOBIOLOGY 295

LITERATURE CITED

ABRAHAMSON, WARREN G., and
DAVID C. HARTNETT. 1990. Pine
flatwoods and dry prairies. Pp. 103-149

University of Central Florida Press,

Orlando.
gp eat Tk; ICE A: 1976. Florida

Part Sten sae Cattleman’s
Association, Kissimm
ANONYMOUS. 1988. “The Village:

Commemorating 75 Years of Christian
Commitment and Compassion 1913-
1988. Advent Christian Village, Dowling

BALEE, WILLIAM (editor). 1998. Advances
in Historical Ecology. Columbia
University Press, New York.

BARTRAM, WILLIAM. 1791. The Travels
of William Bartram (M. oS Doren,
editor, 1955). Dover, bigs

BENGTSON, GEORG , JOHN DUPRE,
WILLIAM TWOMEY, and ROBERT
HOOPER. 1993. Longleaf ecosystem
restoration in the wake of Hurricane
Hugo. Pp. 339-347 in Proceedings, Tall
Timbers Fire Ecology Conference No.

Timbers Research Station, Tallahassee.
BOARD, PRUDY TAYLOR and PATRICIA
POPE BARTLETT. 1985. Lee County: A
Saad History. The Donning
mpany, Norfolk.
BOYETTE, WARREN G. 1996. Results of
leaf Pine

I, II, 11). North Carolina Division of
Forest Resources, Raleigh.

RINTON, DANIEL G. 1869. Guide-Book
of Florida and the South, for Tourists,
Invalids and Emigrants, with a Map o
the St. John River (1978 edition). The
University Presses of Florida,
Gainesville.

BUTLER, CARROLL B. 1998. Treasures of
the Longleaf Pines Naval Stores. Tarkel
Publishing, Shalimar.

CRUMLEY, CAROLE L. 1994a. Preface. Pp.
xiii-xiv in Historical Ecology: Cultural
Knowledge and Changing Landscapes,

arole L. Crumley (editor). bees of
American Research Press, Sa

. 1994b. Historical Ecology: a

multidimensional ecological

orientation. Pp. 1-16 in Historical

Ecology: Cultural Knowledge and

Changing Landscapes, Carole L.

Crumley (editor). School of American

Research Press, Santa Fe.

8. Foreword. Pp. ix-xiv in
Advances in Historical Recta William
Balée (editor). Columbia University
Press, New Yor

DELCOURT, HAZEL R. and PAUL A.
DELCOURT. 1997. Pre- columbian
Native fi

Appalachian PADS. Conservation
:1010-1014.

DELCOURT, PAUL A. and HAZEL R.
DELCOURT. 1998. The influence of
prehistoric human-set fires on oak-
chestnut forests in the southern
Appalachians. Castanea 63(3):337-345.

eee = ISON,
WILLIAM E. SHARP, and KRISTEN J.
GREMILLION. 1998. Prehistoric human
use of fire, the eastern agricultural
complex, and Appalachian oak-
chestnutforests: Fhavisam ats of Cliff
Palace Pond, ucky. American
Antiouity 63(2): 768-278,

DODD, JR., C. cane dah 1995. Reptiles
and amphibians n the endangered
longleaf pine cca eten Pp. 129-131 in

and Health of U.S. Plants, Animals, and
Ecosystems, E. LaRoe, G. Farris, C.
Puckett, P. Doran, and M. Mac (editors).
U.S. Department of the Interior,
National Biological Service,
Washington, D.C.

DODRILL, DAVID E. 1993. Selling the
Dream: The Gulf American Corporation
and the Building of Cape Coral, Florida.
The University of Alabama Press,
Tuscaloosa

296 WALKER

ENGSTROM, . -B.:.... TODD..:., 1993.
Characteristic mammals and birds of
longleaf pine forests. Pp. 127-138 in
Proceedings, Tall Timbers Fire Ecolog
Conference No. 18, The Longleaf Pine
Ecosystem: Ecology, Restoration, and
Management, Sharon M. Hermann
(editor). iia Timbers Research Station,
Tallahas

FLORIDA COASTAL MANAGEMENT
PROGRAM (FCMP). 1995. Protecting
Florida’s Coastal Ecosystems:
Conservation Through Local Land
Acquisition and Management. Florida
i fossa of Community Affairs,
Tallahas

FOLKERTS, "GEORGE W., MARK A
DEYRUP, and D. CLAY SISSON. 1993.
Arthropods associated with xeric
longleaf pine habitats in the
southeastern United States: a brief
overview. Pp. 159-192 in Proceedings,
Tall Timbers pain Ecology Conference

No. 18, The Longleaf Pine Ecosystem:
Ecology, yg hii and Management,
Sharon M. Hermann (editor). Tall
Timbers Research Seton. Tallahassee

FRANKLIN, ROBERT M._ 1997.
Stewardship of Longleaf Pine Forests:

Ecosystem: Ecology, Restoration, ~
Management, Sharon M. Herman
(editor). ~ Timbers Sankt Station,
Tallahasse

GODOWN, MARIAN B. and ALBERTA C.
RAWCHUCK. 1975. [serpy a s Fort

Myers. Press Printing, Fort Mye

GUYER, CRAIG and MARK A. BAILEY.
1993. Amphibians and reptiles of
longleaf pine communities. Pp. 139-158
in Proceedings, Tall Timbers Fire
Ecology Conference No. 18, The
Longleaf Pine Ecosystem: Ecology,
Restoration, and Management, Sharon

. Hermann (editor). Tall Timbers
Research Station, Tallahassee.

Vol. 20, No. 2

GRISMER, KARL H. 1982. The Story of Fort
Myers: The History of the Land of the
Cc 1 1 ro | ic ral <T1 pe

A facsimile reproduction of the 1949
edition. The Island Press Publishers,
Fort Myers Beac

HARCOMBE, P. A., J. S. GLITZENSTEIN,
R. G. KNOX, S. L. ORZELL, and E. L.
BRIDGES. 1993. Vegetation of the
longleaf pine region of the west Gulf
Coastal Plain. Pp. 83-104 in Proceedings,
Tall Timbers Fire Ecology Conference

Timbers oye Station, Tallahassee.
HARRAR, ELLWOOD S. and J. GEORGE

yea 1962. Guide to Southern

Trees. Dover Publications, Inc., New

Yor

HARRIS, LARRY D. 1999. peak tuarked
Florida’s ancient forests: old Flori
words and pictures. Florida Phteaders
of the Environment Special Bulletin,

Summer.

HARSHBERGER, JOHN W. 1914. The
vegetation of south Florida south of 27°
30' north, exclusive of the Florida Keys.
Transactions of the Wagner Free
Institute of Science of Philadelphia Vol.

VII(3).
HEADLAND, — THOMAS N. 1997.

Current Anthropology : “ie 605-609.
HOFSTETTER, RON 1974. The

effect of fire on he Le td ps and

Sawgrass communities of southern

Society Memoir
Environments of South Florida: Present
and sts Patrick J. Gleason (editor and
compiler).

JOHNSON, RHETT. 1995. Game
management in the longleaf pine forest
type. Alabama’s Treasured Forests
(summer):28-30.

. 1996. Longleaf pine: a southern
legacy. ee s Treasured Forests
Su -9

KALISZ, PAULJ., ALAN W. DORIAN, and

EARLL. STONE. 1986. Prehistoric land-
use and the distribution of longleaf pine
on the Ocala National Forest, Florida:
an interdisciplinary synthesis. The

Florida Anthropologist 39(3):183-193.

Winter 2000

LANDERS, J. LARRY, DAVID H. VAN
LEAR, and WILLIAM D. BOYER. 1995.
The longleaf pine forests of the
southeast: requiem or renaissance?
Journal of Forestry 93(11):39-44.

LITTLE, ELBERT L., JR. 1971. Atlas of
United States Trees. Volume 1: Conifers

1146. U.S. Government Printing Office,

Washington, D.C.

. 1978. Atlas of United seen Trees:

Volume 5: Florida. U.S. Forest Service

“parrgmosoranae Publication ae 1361.
2. nment Printing Office,

Washington, D

_______ , 1980. The Audubon Society Field
Guide to North American Trees: Eastern
Region. Alfred A. Knopf, Inc., New

York.

LONGLEAF ALLIANCE. n.d. The Longleaf
Alliance: Longleaf Pine...A Southern
re Ha Papbe Foci Auburn University,

MARQUARDT, WILLIAM H. and
CAROLE. -L. ‘CRUMLEY. © 1987.
Theoretical issues in the analysis of
spatial patterning. Pp. 1-18 in Regional

MACLAREN, PATRICIA A. and JAMES A.
STEVENSON. 1993. Florida
Department of Natural Resources
management of longleaf pine
ecosystems. P. 407 in Piecetings: Tall
Timbers reste Ecology Conference Ae
18, The Longleaf Pine Ecosyste
Ecology, Restoration, and Mariageisiont,
Sharon M. Hermann (editor). Tall
Timbers Research Station, Tallahassee.

MILANICH, JERALD T., and CHARLES

. 1993. H to and
Indians of —— University Pei of
Florida, Gaines

MORR, CHARLES. 1896. The Timber Pines
of the oe United States. U.S.
Departm f Agricu — Division of
Forestry, Bulletin’ No.

MOORE, JULIE. esas ees of the
longleaf forest. P. 18 in Longleaf
Legacies: pele Pine...We Can Save
it for Profit and Pleasure, Julie Moo
and Carol Goodwin (editors). Longleaf
Partners Funds, Gainesville.

JOURNAL OF ETHNOBIOLOGY 297

hiconiiem S99G6D, The groundcover: where
the diversity i is found. P. 19 in Longleaf
Legacies: Longleaf Pine...We Can Save
it for Profit and Pleasure, Julie Moore
and Carol Goodwin —. Longleaf
neg Funds, Gainesville

pith role of fire in
the eases e. P. 22 in Longleaf
Legacies: Longleaf Pine...We Can Save
it for Profit and Pleasure, Julie Moore
and Carol Goodwin (editors). Longleaf
Partners Funds, Gainesville.

and CAROL GOODWIN. 1996.
~~ Longleaf Legacies: Longleaf Pine...We
Can Save it for Profit and Pleasure, A
Calendar for 1996. Long Needle Press,
Gainesville, FL.

MOYROUD, RICHARD. 1996-1997. South
Florida slash pine: Pinus elliottii variety
densa. The Palmetto, winter:11-12.

MYERS, RONALD L. 1990. Scrub and high
pine. Pp. 150-193 in Ecosystems of
Florida, Ronald L. Myers and John J.
Ewel (editors). University of Central

Florida Press, Orlando.
OUTCALT, KENNETH W. and RAYMOND
FFIELD. 1996. The Longleaf Pine
Forest: Trends and Current Conditions.
Resource Bulletin SRS-9. U.S.
Department of Agriculture, Forest
Service, Southern Research Station,

sheville.
PALIK, BRIAN. 1995. aba ait of
n Longlea

it for Profit and Pleasure, A Calendar for
1996, Julie Moore and Carol Goodwin
Long Needle Press,

ville.

PANSHIN, A. J. and C. DEZEEUW. 1980.
Textbook Wood Technology:
Structure, Identification, Properties, and
Uses of the Commercial Woods of the
United States and Canada. McGraw-
Hill Book Co on Hem New York.

PEET, ROBERT K. and DOROTHY J.
ALLARD. 1993. Longleaf pine
vegetation of the southern Atlantic and
eastern Coast regions: a
preliminary classification. Pp. 45-81 in

Proceedings, Tall Timbers Fire Ecology
Conference No. 18, The Longleaf Pine
Ecosystem: —— Restoration, and
Management, Sharon M. Herm
eos — aber oss Station,
Tallaha

298 WALKER

PYNE, STEPHEN J. 1998. Forged in fire:

Ecology, William Balée henseen
Columbia University Press, New York
REBERTUS, ALAN J., G. BRUCE
WILLIAMSON, and WILLIAM J.

PLATT. 1993. Impact of temporal

eavanna

and pines. Pp. 215-225 in Proceedings,
Tall Timbers ste Ecology Conference
No. 18, The Longleaf Pine Ecosystem
Ecology, Restoration, and Management,
Sharon M. Hermann (editor). Tall
Timbers Research Station, Tallahassee.
RECORD, SAMUEL and ROBERT HESS.
1943. Timbers of the New World. Yale
University Press, London.
RILEY, L., R. CLARK, K. FORSYTH, K.
DRYDEN, and M. ALLEN. 1993. Florida
Communities Trust Preservation 2000
Grant Application for the Lee County
Hickey Creek Mitigation Park.
ROMANS, BERNARD. — A Concise
Natural History of East and West
Florida (a facsimile speaacane 1962).
University of Florida Press, Gainesville.
ROUSE, PARKE, JR. 1988. The Timber
Tycoons: The Camp Families of Virginia
and Florida and Their Empire 1887-
1987. The William Byrd Press,

ichmond.

SCHWARZ, G. FREDERICK. 1907. The
Longleaf Pine in be rk Forest: A Silvical
Study. John Wiley & Sons, London.

SARGENT, CHARLES S. 1884. Beoart on
the Forests of North America (Exclusive
of Mexico). Department of the Interior,
Census Office, Washington, D.C.

SCARRY, C. MARGARET, and VINCAS P.
STEPONAITIS. 1997. Between
farmstead and center: the natural and
social landscape of Moundville. In
People, Plants, and Landscapes: Studies
in Paleoethnobotany, Kristen J
Gremillion (editor). University of
Alabama Press, Tuscaloosa.

SMALL, JOHN K. 1930. Vegetation and
erosion on the ste Keys. The
Scientific Monthly 30:33-49.

SMITH, BUCKINGHAM (trans.). 1968.
Narratives of De Soto in the Conquest
of Florida as Told by A Gentleman of
Elvas and in a Relation by Luys
Hernandez de Biedma (orig. trans.
1866). Palmetto Books, Gainesville.

Vol. 20, No. 2

SNYDER, JAMES R., ALAN HERNDON,
and WILLIAM B. ROBERTSON, JR.
1990. South Florida ‘nockland. Pp. 230-

Florida, Ronald L.

Myers and John : Ewel (editors).

University of Central Florida Press,

Keo xzvetame

Orlando.

SUDWORTH, GEORGE B. 1913. Forest
Atlas: Geographic Distribution of North
American Trees, Part 1. Pines. U.S.D.A.
Forest Service, Washington, D.C.

TEBEAU, CHARLETON W. 1957. Florida’s
Last Frontier: The History of Collier
County. University of Miami Press,

Miami.

TURNER, GREGG M. 1999. Images of
America: Railroads of Southwest
Florida. Arcadia Publishing, Charleston.

Food and Agricultural ‘Sciences,
Agricultur al
Soil Science Department, and Florida
Department of Agriculture and
Consumer Services, U.S. Government
Printing Office, Washington, D.C.

VIGNOLES, CHARLES. 1823. Observations
Upon the Floridas (a facsimile
reproduction 1977). The University
Presses of Florida, Gainesville.

WAGNER, GAIL E. In press. Eastern
woodlands anthropogenic ecology. In
People and Plants in Ancient North

merica, Paul Minnis (editor).

Smithsonian Instititution Press.
Washington, D.

WAHLENBERG, W. &. 1946. Longleaf Pine:

Lathrop Pack Forestry
Foundation and U.S. a of
Agriculture, Washington, D.C.

WALKER, LAWRENCE C. 1991. The
Southern Forest: A Chronicle.
University of Texas Press, Austin.

WALKER, KAREN J. 1997. Historical

p south Florida perspec
Palmetto 17(2):16-19.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 299

ROBIN L. DENSON, and GARY WINTERHALDER, BRUCE. 1994. Concepts

DE ELLIS. 1996. Archaeological Survey in historical ecology: the view from
of the Hickey Creek Mitigation Park evolutionary ecol 7-41 in
Report submitted to the Division of Historical Ecology: Cultura al Knowledge
Recreation and Parks Services, Lee and Changing Landscapes (Carole L.
County, FL. Crumley, editor). School of American
WATTS, W. A. 1993. The history of longleaf Research Press, Santa Fe.
pine in Florida. P. 15, Abstract, in ©WUNDERLIN, RICHARD. 1998. Guide to
Proceedings, Tall Timbers Fire Ecology the Vascular Plants of Florida.
Conference No. 18, The Longleaf Pine University Press of Florida, Gainesville.
Ecosystem: Ecology, Restoration, se ZEISS, BETSY. 1983. The Other Side of the
Management, Sharon M. Herman River: Historical Cape Coral. Cape
on ay Timbers Research Station, Coral.
Tallaha
WINN, JOHN. 1996. A longleaf pine
restoration project. The Palmetto,

Winter 1996/1997: 15-15.

300 BOOK REVIEWS Vol. 20, No. 2

Biodiversity and Native America. Minnis, Paul E. and Wayne J. Elisens (editors).
University of Oklahoma Press, Norman. Pp 310, x, maps, figures, and illustra-
tions, index. $34.95 cloth. ISBN 0-8061-3232-9.

Biodiveristy and Native America is a significant and. essential read for students
and researchers interested in past and present ips among
Aboriginal peoples in North America. The book spans most of the geography of
the United States, with representative articles covering portions of Mexico and
Canada. This book is a beginning of what will hopefully be many books on the
subject, the editorial product of Paul Minnis and Wayne Elisens, University of
Oklahoma. The papers largely came together as a result of a 1997 symposium
workshop organized around the topic of “Biodiversity and Native North America.”
The book showcases 10 articles reflecting a variety of geographical and cultural
perspectives, set in three sections.

Section one: Issues and Overviews contains three articles that examine First
Nations’ management and conservation (Sonora Desert), ethnobotony (Mexico)
and ethnopharmacology (broad U.S. overview with implications for Peru). The
first article by Gary Paul Nabhan bridges issues from li gy with
conservation biology. Robert Bye and Edelmira Linares offer a fine paper summa-
rizing the complex and evolving ethnobotanical relations among the over 54
Indigneous language groups in Mexico. These groups have extensive knowledge
for the Holarctic and Neotropic plant kingdoms. Walter Lewis presents an infor-
mative, but narrow, discussion of ethnopharmacology and a possible future built
on collaborative agreements. Lewis does not discuss how the influence of tremen-
dous amounts of capital on traditional systems, including issues of ownership (is
knowledge of medicinal plants individual or collective property), will be worked
out.

Section two: Ethnographic Case Studies, gives the greatest representation of
“biodiversity and Native America” in this volume. The three papers represent the
sampled knowledge of Northern Paiute, Owens Valley Paiute, Southern Paiute,
Timbisha (Panamint), Shoshone, Ute, and Washoe; Stl’atl’imx (Lillooet / Lil’ wat);
and Raradmuri. Catherine Fowler recounts in her paper how fieldwork with Numic
speaking Great Basin Indian people in the 1960’s and 1970’s brought forward the
concept “we live by them.” This “Native knowledge” may be gathered and fo-
cussed as ethnobiological constructs, but it is embedded and expressed in the
cultural practices and daily lives of untold numbers of Aboriginal peoples from
whom she learned/studied. The article by Sandra Peacock and Nancy Turner is
an excellent example of collaborative ethnobotanical research. The authors bal-
ance academic/scholarly knowledge with traditional plant knowledge to
understand traditional resource management and biodiversity conservation for a
portion of the traditional territory of the Secwepeme, Stl’atl’imx, Nlaka’pamux,
and ae people. In the third paper in this section, Enrique Salm6én exam-
ines the Raramuri t of iwigara. The paper is a well researched and important
paper that made me questions the role of traditional environmental knowledge.

Many of the papers in this edition advocate changing western land manage-
ment practices and thinking (and I support such efforts). However, if the role of

Winter 2000 JOURNAL OF ETHNOBIOLOGY 301

ancient knowledge really is to reform science and western constructs we need to
know how they will and can function within our commercialized modern world.
What the potential and real impacts on how the Rardmuri are able to practice
iwigara in the northern Sierra Madre Occidental is left unstated. A dominant view
among most Indigenous groups is that ancient knowledge has intrinsic and cul-
tural value, which scientists have a hard time understanding and accepting.

Section three, Prehistory and Biodiversity, contains three interesting papers.
This section will disappoint those readers who may have grown somewhat accus-
tomed to the culturally specific, situated, and interdisciplinary tone of the volume
to this point. The three papers in this section are more general with an over-riding
anthropological perspective. Furthermore, the contributions in this section reflect
a much narrower geographic focus, which does not fit with the rest of the book.
With this said the three papers should not disappoint readers, as they offer signifi-
cant contributions. Focussing his discussions on Northern New Mexico, Richard
Ford examines the significance of human disturbance on biodiversity, arguing that
“it is a mistake to regard Native Americans as insignificant managers of biotic
resources or as passive a in the shaping of the landscape and the diver-
sification of habitats” (219). G tzd through
archaeological research for es pe cover a broad area of the eastern United States
and a portion of southern Ontario. Fritz’s paper suggests the difficulties that bi-
ologists and anthropologists run into when assessing the scale of environmental
impacts and changes to anthropogenic influences. Unfortunately, this paper ap-
pears to reflect a classical positivi st resea earch approach BY failing to involve or even
consult local knowledge in ing prehistoric changes. While this may be
the least informative paper eae Traditional knowledge, Fritz is able to dis-
pel myths that Eastern Amerindians were non-agrarian nomadic peoples, and so
it is nonetheless an important paper for readers to consider. The final paper in this
edition by Julia Hammett, an already published journal article (was originally her
master’s thesis), examines the ethnohistory of the southeastern United States.
Hammett cautions that historical accounts are “loaded with distortions, biases,
and contradictions” and suggests that it is important to understand the cultural
context of the original authors, an ambitious project to say the least (p. 253). This
paper is both informative and provocative and completes the book in a way that
urges readers on to further research.

Chris Hannibal-Paci
First Nations Studies
University of Northern British Columbia

Journal of Ethnobiology 20(2): 303-325 Winter 2000

A CLADISTIC APPROACH TO Dr ny
ETHNOBOTAN

T
DYE PLANTS OF THE SOUTHWESTERN UNITED STATES

KIMBERLY HAMBLIN HART
Department of Botany and Range Science
Brigham Young University
Provo, Utah 84602 USA

PAUL ALAN COX
National Tropical Botanical Garden
3530 Papalina Road
Kalaheo, Hawati 96741 USA

ABSTRACT.—An intensive review of the ethnobotanical literature on dye plants
used by 11 indigenous tribes in the Southwestern region of the United States
revealed that 108 plants have been used to manufacture dyes for coloring wool,
leather, cotton and other plant fibers. seme pee ee are also used to obtain
pigments for pottery and body paint used to color food. Of the 11
different plant dye traditions evaluated in this ay the Navajos use the greatest
number of plants (n=69) for dye purposes. Considering innovations in dye plant
traditions shared among tribes to be analogous to shared derived characters in
phylogenetic analyses (termed “synapomorphies”), a cladistic analysis shows that
traditions of dye plants are most derived among the Navajo and Hopi tribes. The
traditions of dye plants of these two tribes are also more closely related to each
other than either tradition is to dye plant traditions from other tribes. The cladistic
approach of analyzing shared derived technologies appears to be a useful way of
generating hypotheses concerning cultural diffusion of plant uses in other
ethnobotanical studies.

Key words: cladistics, dye plants, ethnobotany, Southwestern Native Americans.

las plantas

RESUMEN.—Una revision intensiva de la literatura ent

del tinte usados por 11 tribus indigenas en la region al sudoeste de los Estados
Unidos revelé que 108 plantas se han utilizado para fabricar los tintes para las
lanas del colorante, el cuero, el algod6n, y otras fibras de la planta. Un ciertas
especies de la planta también se utilizan para obtener los pigmentos para la
cerdmica y la pintura de cuerpo en ane otras se utilizan para colorear el
alimento. De las 11 trib io, la tribu de Navajo utiliza el
numero mas grande de las plantas (n=69) para los propésitos del tinte.
Considerando innovaciones en las plantas del tinte compartidas entre las tribus
para. ser el equivalente si fermip pepe se dice “synamorphies,” un analisis
el tribus de Navajo y de Hopi son seradhe

mas de las plantas del tinte. Estas do

uno al otro en sus Lao de la planta del tinte que estan a pas ei otra
tribu. El acercamiento cladistic de analizar tecnologias derivadas compartidas
aparece ser una manera ttil de generar hipétesis referentes a la difusi6n cultural
de las aplicaciones de la planta en otros estudios ethnobotanical.

304 HART and COX Vol. 20, No. 2

RESUME. Une | revue approfondie de la literature ethnobotanique est presenté
sur 108 pl tribus indigénes a la region suroeste
des Etats Unis. Elles sont utilisées pour teindre de la laine, le cuir, le coton, et
quelques autres fibres vegetales. La tribu Navajo utilis le plus grand nombre des
plantes ¢ comme seiniunes hams = 69). Une analyse cladistique mnchigie ae i lined

comme teintures et aussi elles sont plus similares entre se.

INTRODUCTION

The Southwestern region of the United States is considered ethnobotanically
to be “the best studied area in the world” (Ford 1985:401). In this region, compre-
hensive studies have been made of the plants used by indigenous people for
medicine, food, clothing, and art (Bell and Castetter 1937; Castetter, Bell and Grove
1938; Dennis 1939; Dunmire and Tierney 1995; Fewkes 1896; Kent 1957; Palmer
1878; Sauer 1950; Standley 1911; Winter 1974). Other studies have focused on the
ethnobotany of particular tribes (Castetter and Underhill 1953; Cook 1930; Elmore
1943; Ford 1968; Hough 1897; Jones 1931, 1948; Mathews 1886; Reagan 1929;
Robbins, Harrington and Freire-Marreco 1916; Stevenson 1915; Swank 1932; Ves-
tal 1952; White 1945; Whiting 1939; Wyman and Harris 1941, 1951). However,
comparative ethnobotanical studies are rare. In the early 1960’s, Whiting identi-
fied an urgent need for “summary reports, comparative historical studies, and
broadly based reviews of comparable data throughout the area” (Whiting 1966:318).
Doebley (1984) responded to this call with comparative studies of wild grasses,
yet few other similar studies have been done. Twenty years after Whiting made
his statement, Richard Ford (1985) and Robert Bye (1985) both noted that there
remains a void in the area of comparative work.

We have compared use of plants for dyes and paint among different south-
western indigenous tribes based on historical and contemporary accounts. For
this purpose we considered all plants used to color wool, cotton, and leather, for
food coloring, as well as for pigments for body and pottery paint. The purpose of
our study is two-fold: Q) to Provide a comprehensive review and comparison of
dye plants used by dians, and (2) to show how cladistic analy-
ses may be used to generate hypotheses concerning cultural diffusion of plant
uses between tribes.

Linguists, systematists, and biogeographers have previously used cladistic
techniques to study common origins of languages, biological species, and biogeo-
graphical regions mpspertively, Unlike comparative methods that rely on overall
similarity, s such diagrams (also
know as cladograms) based on shared derived features or characters
(synapomorphies). Thus, although ph
lizards are more closely related t to ‘each other than either are to birds because of
overall similarity, cladistic analyses group birds with crocodiles because of shared
derived features of skull anatomy (Ridley 1993). In biology, characters used for
cladistic analyses can be different features of anatomy, molecular sequence, be-
havior, physiology and so forth.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 305

We believe that cladistic analysis might be a useful method for cross-cultural
ethnobotanical comparisons. A unique technological innovation that is subse-
quently shared by different cultures could be considered a shared derived feature,
called in cladistic terminology a “synapomorphy.” For example, if use of a par-
ticular plant as a medicine originated with a single individual, but subsequently
spread to different cultures through time, that use could be considered to be a
synapomorphy for those cultures. Synapomorphies are used in cladistic analyses
to indicate possible branching patterns in cladistic trees. Such shared derived in-
novations can then be used to generate relationship trees for the technology of
interest (such as dye plants, medicinal plants, crop varieties, etc.). Diagrams of
these relationships, presented as trees, are termed “cladograms.”

Technological features in common to different cultures that do not share a
common unique derivation could be termed “symplesiomorphies.” For example,
the use of conifers as firewood is probably common to all cultures where conifers
occur, but likely cannot be traced to a single unique innovation, and hence is an
example of a symplesiomorphy. Symplesiomorphies unfortunately, are of little or
no value in determining relationship trees or cladograms.

Some cultures may produce technological innovations that do not spread to
other cultures. Such unique unshared innovations are termed “autapomorphies.”
For example, use of an endemic species of algae by the Hawaiian people cannot
possibly have spread to other islands, and hence could be considered an
autapomorphy. Autapomorphies, while interesting for a particular culture, do not
shed light on relationships to other cultures.

Characters used for cladistic analyses in cross-cultural ethnobotanical studies
could include technological, medicinal, artistic, architectural, ritual innovations.
It is not necessary to compare biological entities; we here study plant uses because
as ethnobotanists our interests are focused on the interactions between plants and
people. Cladistic studies require that observable information is translated into dis-
crete characters (Kitching et al. 1998). In cross-cultural ethnobotanical studies one
can easily identify plants as used or not used, making such characters prime can-
didates for cladistic analyses. Thus, we are proposing to evaluate relationships
between uses of plants by different tribes based on shared technological innova-
tions of dye plant use rather than grouping these uses on the basis of overall
similarity. It is important to note that we are not, however, attempting to consider
genetic or cultural relationships of the tribes themselves. It is only the plant uses,
and not the people themselves, which are the objects of our analysis. Thus, while
our diagrams of plant use relationships are not intended to suggest genetic or
cultural relationships between different tribes, they can be used to generate hy-
potheses of how different discoveries of new dye plants might h P through
various cultures. :

While cladistic techniques are simple, and for a limited number of different
traditions of plant use (three or four) can easily be done by hand, the number of
possible alternative relationships trees (and hence the number of calculations) in-
creases exponentially with the number of tribes. As a result, we have had to use a
computer program to evaluate the number of trees. As will be described shortly,
the program basically determines which, of all possible relationship trees, is the

306 HART and COX Vol. 20, No. 2

most parsimonious- the one requiring the least number of steps of culture trans-
mission, parallel innovation, and culture loss. This most parsimonious tree is then
proposed as a candidate for evaluation by other researchers. Often, with a large
number of taxa (here considered to be different tribal traditions of plant dye use),
different trees of equal number of steps are discovered during the computer algo-
rithm. We have here chosen to present a summary of the features in which all of
these most simple trees agree: such a diagram is called a strict consensus tree.
Further information on cladistic techinques can be obtained from a variety of text-
books in systematic biology.

METHODS

General Comparison.—As a means of understanding native American traditions of
dye plants use, we conducted interviews with Navajo weavers on the Navajo res-
ervation in Southern Utah and Northern Arizona and observed some collections
of dye species and dying techniques. We then expanded our study to a regional
basis by conducting an intensive literature review, compiling ethnobotanical in-
formation on 11 different tribes: Eastern Keres, Hopi, Jemez (Towa), Navajo, Papago,
Pima, Southern Tiwa, Tewa, Western Apache, Western Keres (Acoma and Laguna),
and Zuni. We chose to study the dye plant traditions of these tribes because of the
geographical proximity of the tribes to each other, their pattern of cross-cultural
interactions, and the availability of previous ethnobotanical studies.

Some of the dye plants used in the past are no longer used today, yet for our
analysis we include both historical and contemporary uses with no effort to dis-
tinguish between the two. In our study we selected from literature accounts only
those plants identified to the level of both genus and species, since records from
different tribes of a plant identified only by a generic epithet might conflate differ-
ent species, skewing our analysis. For consistency, plants identified beyond the
species level to the varietal level were truncated to species. Appendix 1 lists each
plant and the tribes that used it. Figure 1 illustrates approximate tribal boundaries
and the number of dye plants use by each tribe. Ow definition of tribal bound-
aries is somewhat arbitrary since these ver been static but vary
in time with changes in culture, modes of transportation, and the colonization/
reservation boundaries forced upon different indigenous groups. For this reason,
we used a slightly modified version of regional boundaries defined in The Hand-
book of North American Indians (Ortiz 1983).

Cladistic Analysis.—A cladistic analysis based on shared derived characters
(synapomorphies), in this case, shared cultural innovations in use of dye plants,
was performed by coding each of the 108 dye plant species as either used or not
used for each of the 11 tribes. Our data matrix is provided in Appendix 2. No effort
was made to differentiate between plants used to dye wool or other materials for
two reasons: (1) we are presuming that one plant used for one particular material
would most likely be tried on other materials as well, therefore not be exclusive to
wool, cotton, leather, or other materials, and (2) literature accounts tend to focus
on the plants used rather than on the types of materials dyed. Our coded data

Winter 2000 JOURNAL OF ETHNOBIOLOGY 307

FIGURE 1.—Map showing the geographic proximity of tribal regions. Number of dye
plant species used by each tribe are indicated (adapted from Ortiz 1983).

matrix was analyzed with the computer program HENNIG86 (Farris 1988). In the
analysis we gave each dye plant species an initial weight of one and selected the
non-additive option. The complete search algorithm, implicit enumeration (ie),
was used to generate relationship trees of minimal length. A strict consensus tree
was obtained for the trees obtained from implicit enumeration of the unweighted
characters (Figure 2). We then found equally parsimonious trees by using the xsteps
command with the w option utilized, thus applying species weights according to
their fit to the trees. Weights applied were calculated by the program as the prod-

308 HART and COX Vol. 20, No. 2

uct of character consistency index, ci (Kluge & Farris 1969) and the retention in-
dex, ri (Farris 1989a, 1989b). Weightings were applied in successive rounds of
implicit enumeration until no changes in tree length, consistency index, or reten-
tion index could be obtained from successive rounds. We then obtained a strict
consensus tree for the weighted sample.

In a strict cladistic sense, we make no claim about the monophyletic nature of
the traditions we have here analyzed; in fact the uses we analyze may be
paraphyletic because 1) we do not know if all of these plant dye uses can some-
how be traced back to a singular innovation in the uses of plants as dyes, and
hence share the same ancestral tradition, and 2) it is doubtful if we have here in-
cluded all possible traditions derived from an ancestral tradition; little is known,
for example, about Anasazi use of dye plants.

In cladistic analyses, often an outgroup possessing the “primitive” state is
chosen in order to determine character polarities. Not wishing to make any state-
ment about relative age and technological status of any of the 11 tribes we studied
by claiming that one tribe’s use of a plant somehow preceded or was ancestral to
another tribe’s use of the same plant, we rooted our analysis in the uses of plants
by a hypothetical tribe that has never used any dye plants: hence all character

Primitive

Papago

Becker

ih Sie W. Apache
2 Zuni

Jemez

W. Keres

avajo

Hopi

FIGURE 2.— Cladistic relationships of dye plant traditions of southwestern Amerindian
tribes; strict consensus tree of unweighted characters, length= 131, consistency index =
0.82; retention index = 0.51

Winter 2000 JOURNAL OF ETHNOBIOLOGY 309

States begin at zero. We note that alternative methods of cladistic analysis that
include unrooted networks are available, but such analyses do not change the
topology of our resultant trees, and we believe our postulation of a zero-use cul-
tural antecedent to current Amerindian dye uses is, by reductio ad absurdum true: if
we were to go back far enough in time (at the extreme, the first aboriginal immi-
grants to North America), we would eventually find a group of people who did
not use any North American plants for dyes. This group would be the most “primi-
tive” group as far as dye technology is considered. Our subsequent analysis
assumes that knowledge of how to use dye plants is passed from generation to
generation rather than being independently recreated de novo each generation—
the cultural equivalent of recurrent homoplasy in the cladistic sense.

RESULTS

Enumeration of Dye Plant Species—A total of 108 species, including 103 vascular
plants, two fungi and three lichens, have been recorded as sources of dye pig-
ments for wool, cotton, leather, body and pottery paints, and the coloring of food
by the 11 tribes (Appendix 1). The 103 vascular plants represent 38 different fami-
lies. The majority of the dye plant species are used to dye wool. Of these 108 species,
the Navajo use 69, the Hopi use 24, the Western Keres use 14, the Tewa and Zuni
use 10, the Jemez use eight, the Western Apache use seven, the Papago use six, the
Southern Tiwa use four, and the Eastern Keres and Pima both use three species
(Figure 1).

Cladistic Analysis.—Our cladistic analysis based on shared cultural innovation
(synapomorphies) in use of dye plant species initially resulted in nine trees, each
requiring 131 steps, a consistency index of 0.82 and a retention index of 0.51. A
strict consensus tree, which presents all features on which these nine trees agree,
showed a basal unresolved trichotomy, but we sought to improve the consistency
and retention indices by successive character weightings. We then performed two
rounds of successive approximations weighting which was analyzed by implicit
enumeration, and seven trees of different topologies (Figure 2) were obtained with

TABLE 1.—The 11 most commonly used dye plants and the associated tribes
that use those plants.

Plant species Tribes that use them

Alnus tenuifolia Jemez, Navajo, S. Tiwa, Tewa, W. Apache, W. Keres, Zuni
Cercocarpus montanus E. Keres, Jemez, Navajo, S. Tiwa, W. Keres
Chrysothamnus nauseous Navajo, Tewa, W. Apache, W. Keres, Zuni
Cleome serrulata E. Keres, Navajo, S. Tiwa, Tewa, Zuni
Pinus edulis Hopi, Jemez, Navajo, W. Keres

Atriplex canescens Hopi, Navajo, Tewa

Betula occidentalis Hopi, Jemez, Tewa

Castilleja integra Navajo, W. Apache, Zuni

Descurainia pinnata Hopi, Jemez, Tewa

Psilotrophe tagetina W. Apache, W. Keres, Zuni

Rhus aromatica Hopi, Navajo, W. Keres

310 HART and COX Vol. 20, No. 2

a higher consistency index of 0.98 and a retention index of 0.90, all with 940 steps.
Although the strict consensus tree of the weighted samples reduced resolution
from the unweighted analysis by collapsing the original basal trichotomy into a
basal hexatomy, all other topological features of the tree remained the same as the
unweighted tree (in fact, the tree produced from character weighting, is, topologi-
cally, still a subset of the unweighted tree.). We note that the consistency index
(but not the retention index) may be an overestimate of the robustness of our analy-
sis since this statistic is sensitive to autapomorphies (characters restricted to one
tribe, in our analysis, uniquely derived plant uses not shared with other tribes), in
which some traditions we studied, particularly that of the Navajo, abound.

Of the 108 plant species used as characters, only 27 are synapomorphies. This
level of autapomorphy which, as mentioned above, does affect the consistency

Primitive

Papago

aaa Pima

E. Keres

S. Tiwa

Tewa

23 W. Apache
wi

Zuni

Jemez

Vi W. Keres

87 avajo
15,52

Hopi

FIGURE 3.—Cladistic relationships of dye plant traditions of southwestern Amerindian
tribes; strict consensus tree of weighted characters, with two successive rounds of
implict enumeration; length= 940, consistency index = 0.98; retention index = 0.90. The
synapomorphies are indicated by a number which refers to specific plant species as
identified in Appendix 1.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 311

index. Yet, including all of the data does not affect global parsimony or successive
weighting methods, and, of course, leaves the tree morphology unaltered. How-
ever, many of the nodes of the consensus tree are supported by a relatively few
number of synapomorphies, so it is conceivable that the topology of the consen-
sus tree could change as plant uses from other additional tribes outside of our
study area are added to the sample.

A strict consensus tree (Figure 3), which combines the features on which that
all of the seven most parsimonious trees agree, shows dye plant use of the Hopi
and the Navajo to be more closely related to each other than the use of plants by
either tribe is to their sister group, plant used among the Western Keres.
Synapomorphies (shared characters or innovation in use of a plant species for
dye) linking the Hopi and Navajo include Carthamus tinctorius (an introduced spe-
cies), Juniperus osteosperma, Rumex hymenosepalus, Thelesperma megapotamicum, and
Thelesperma subnudum. Dye use among members of the larger clade consisting of
the Navajo, Hopi, and Western Keres was more closely related to dye use among
the Jemez than to any of the other tribes considered. The synapomorphy (shared
innovation) linking the clade composed of the Navajo, Hopi, and Western Keres is
Rhus aromatica. Dye use among the clade consisting of the Navajo, Hopi, Western
Keres, and Jemez was more closely related to each other than to all other tribes on
the basis of Pinus edulis as a synapomorphy. The other clades consistently grouped
in the strict consensus tree were the Western Apache and Zuni linked by the
synapomorphy of Coreopsis cardaminefolia. Use of Chysothamnus nauseosus link the
Western Apache, Zuni, and Tewa, although our analysis indicates an independent
origin for the use of this species among the Navajo and Western Keres. A less
parsimonious solution is, of course, that the other tribes lost this knowledge. Such
homoplasy may disappear from the cladogram as plant used for dyes from more
tribes are added to the data set, and, in an adapation of cladistic biogeography, as
comparative cladograms for plants used for different purposes (i.e. medicinal,
ritual, etc.) are overlaid with dye plant use. The Papago and Pima share two
synapomorphies—Krameria parviflora and Prosopis velutina. However, the pattern
of branching cannot be resolved in the strict consensus tree for the Eastern Keres
and Southern Tiwa. However, in the unweighted tree they form a sister group to
the Papago and Pima.

DISCUSSION

Enumeration of Dye Plant Species.—Certain questions are raised from our study in
both the enumeration of plant uses and in the subsequent cladistic analysis. Why
do the Navajo use so many unique plants (cultural E phies), especially in
comparison to the other tribes? We believe that the importance of dye plants in the
Navajo economy, specifically in weaving, creates an incentive for Navajos to use
more dye plants. For the Navajo, weaving has been, and continues to be, an im-
portant source of income (Hedlund 1992; Roessel 1983). Weaving as a source of
commercial income for the Navajo was established by 1900 (Wheat 1979). Indeed,
at that time the Navajo rug was the only handwoven good from natives of the
Southwest that still had significant trade value (Minge 1979). The Navajo have

312 HART and COX Vol. 20, No. 2

been praised for the highest quality of weaving observed among regional indig-
enous groups. Some have estimated that the Navajos emerged as premier weavers
by the 1800’s (Mathews 1891; Wheat 1979). The Navajo adopted weaving about
300 years ago and yet they didn’t use a great number of dye plants until the begin-
ning of the 20th century (Hedlund 1992). Indeed, one of the earliest recordings of
dye plant use among the Navajo only mentions seven dye plants (Mathews 1891).
Aniline dyes were also employed during the early part of this century, but by the
1930’s there was a resurgence of interest in natural dyes (Reichard 1936), and in
today’s market a weaver can get a better price for a rug made with vegetal dyes
than one that is made with aniline dyes. For these reasons we believe the Navajo
have a stronger incentive to use vegetal dyes and to continue experimenting and
finding more plants that produce good dyes, even looking outside their own cul-
tural knowledge of dye plant use. This incentive may have also contributed to the
Navajo looking to Anglo/ Western sources for plant dye information, as found in
Amsden’s Navajo Weaving (1940). Our interviews with different weavers and trad-
ers show a general consensus that experimentation with new plants to find new
dyes is common today. This is also supported in the literature (Hedlund 1992;
Jones 1948). Hence there is an economic motivation for use of plant dyes. This
motivation may be a determining factor in the continued use of natural dyes and
may contribute to the fact that many weavers today are continually experiment-
ing with new plants and combinations of plants for unique dyes. This economic
incentive may be significant in the large difference of dye plants used between the
Navajo and other tribes.

Tribal population size could also influence the variation in dye flora sizes
among tribes assuming that larger tribes, having more people, had greater collec-
tive knowledge about what plants make good dye plants. If this were the case, we
would expect larger tribes to use more plants. Today the Navajo tribe is the largest
of the tribes studied, but it is difficult to assess and correlate fluctuations in tribal
size with fluctuations in dye plant use.

Cladistic Analysis.—Of interest in the cladistic analysis is the absence of
symplesiomorphies common to all tribes, i.e., dye plants that all 11 tribes use and
were derived from some earlier tradition of use or people not included in our
analyses. The most commonly used plant is Alnus tenuifolia. Seven of the 11 tribes
use this plant. The four tribes that do not use it are the Hopi, Eastern Keres, Papago,
and Pima. Table 1 shows the 11 most commonly used plants and which tribes use
them.

Does the absence of symplesiomorphies mean that different clades (tribes)
independently invented the use of dye plants, or some tribes lost the use of a
particular plant, or that each tribe merely utilized those plants that were most
common and therefore readily available? Obviously the latter hypothesis cannot
be true for every tribe, especially when the use of non-native species is consid-
ered. But for those tribes only using a few dye plants, independent development
of plant dyes is possible. The absence of symplesiomorphies could indicate that
different tribes lost the use of a particular plant as acculturation through the influ-
ence of Western culture increased with the movement of more European-Americans
into their regions. In the case of Alnus tenuifolia we can assume that each of the

Winter 2000 JOURNAL OF ETHNOBIOLOGY 313

nevert tribes independently invented the use of this plant, but a more parsimoni-
ous h thesis would be that Alnus tenuifolia is actually a symplesiomorphy which
was “lost” four separate times by the Eastern Keres, Hopi, Papago and Pima.

As we consider other commonly used plants like Cercocarpus montanus or Cleome
serrulata, the question becomes more problematic. These two plants are used by
five of the 11 tribes. Were they each once used by all tribes, thus being a
symplesiomorphy? If so, the knowledge would have been lost six times. Or is it
more likely that the five tribes independently came to use these two plants? Use
and diffusion of plant knowledge of such plants may be difficult to assess. Yet
some plants lend th Chrysothamnus nauseosus could
easily be placed on the cladogram below the Tewa, and use of it could have been
lost by both the Jemez and Hopi.

The Navajo have 51 autapomorphies (plants used by only that tribe—a
uniquely derived, but unshared, innovation). The rest of the tribes have notice-
ably fewer autapomorphies and are as follows: Hopi-12, Western Keres-six,
Tewa-four, Papago-three, Western Apache and Jemez-two, Pima-one, Eastern
Keres, Southern Tiwa and Zuni-zero. The presence of unique cultural uses of dye
plants suggests that some indigenous groups are putting more energy into find-
ing dye plants, while others are content to use fewer plants and have less variety
in their range of color for dyed materials. The large number of sence ame
that the Navajo have correlates well with their cultural and p
woven rugs as discussed above.

This cladistic analysis provides some hypotheses on the cross-cultural diffu-
sion of dye plant use/knowledge. It seems plausible that the Navajo and Hopi
would be closely related in dye plant use because of their geographical proximity
to each other and the similarity of the environment in which they live. It is feasible
that as the Nava tern region they learned about
plant use from theti nearest t neighbors—the Hopi. Clearly, some knowledge was
being shared between tribes—the Navajo learned to weave from the Pueblo people.
And through their contact with other southwestern tribes, like the Hopi, it is likely
that the Navajo learned about plant use, in this case dye plant use.

The relationship between the Western Apache and Zuni plant dye use is sur-
prising at first, given their distinct language differences. But as one closely examines
their environments, both live within the White Mountain range which contains a
distinctly different flora from the high plateau deserts where tribes that are cultur-
ally more similar live. Thus their shared relationship in dye plant use appears to
be a function of their shared environment, rather than a closely shared culture.

The Papago and Pima relationship of plant dye use is no surprise—their tribal
regions are much further west and south than the Pueblo tribes and the Navajo/
Western Apache. It would be expected that their flora is the most different of all
tribes studied based on the ecology of their homeland. Indeed, of all the tribes
studied, they have the smallest potential dye flora within their ecological bound-
aries. Also, the Papago and Pima come from the Uto-Aztecan language stock, as
do the Hopi, but the Hopi live in much closer proximity to the Puebloan tribes and
share many cultural traditions with them. The Papago and Pima are more unique
in their cultural background and it would be expected that they would emerge as
more closely related to each other in dye plant use than to other tribes.

314 HART and COX Vol. 20, No. 2

CONCLUSIONS

This comparative study shows a wide range of plants used by Native Ameri-
cans for dye purposes. Such variation suggests several scenarios in the evolution
of dye plant use: (1) those tribes that place a greater emphasis on dyeing, due to
factors such as the economics of dyed materials or cultural significance, may have
actively sought to find plants that yield pigments and thus increased their overall
dye flora, (2) some tribes may have lost dye plant knowledge through accultura-
tion and assimilation into the Western culture, (3) some tribes could have
independently invented the use of certain plants for dyes, and (4) larger tribes
may have retained more inf tion about their tribal dye flora whereas dye plant
use may decrease as tribal size decreases over time. Most likely, a combination of
these factors account for the variation seen among the eleven tribes considered in
this study.

The fact that some tribes use very few plants is as telling as those tribes that
use many dye plants. The cultural importance of weaving, dyeing and painting
varies between tribes. We might assume that those tribes that place a higher sig-
nificance on such activities will have a larger dye flora. And conversely, those
tribes who use few plants may place a lesser value on weaving and dyeing. By
comparing plant use in other areas, we could piece together potential cultural
values for each tribe, based on size of flora used for different means (medicinal,
agricultural, ceremonial/ritual, building, etc.). Dye use is merely one piece of a
larger picture that helps us understand not only cultural uses of plants, but those
things that are important in different cultures as well.

Cladistic analyses can generate hypotheses of cross-cultural diffusion of dye
plant use hat might not be readily apparent if one were to limit cultural compari-
sons to overall similarities. Again, we reiterate that this analysis does not suggest
overall cultural relationships between the eleven tribes studied since we consid-
ered only one small aspect of material culture: dye plant use. Our analysis does,
however, suggest hypotheses on how dye plant knowledge may have spread be-
tween the different tribes and which tribes haring ethnobotanical knowledge.
We find a strong ethnobotanical link between the Hopi and Navajo, the Zuni and
Western Apache, and the Papago and Pima. These different indigenous groups
could have been sharing information about dye plants with each other, both po-
tentially enlarging their own dye flora from the others’ ethnobotanical knowledge.
The exact history of use and knowledge will not be known, but hypothetical situ-
ations can be generated by cladistic studies which are amenable to falsification by
archaeological or ethnohistorical data.

Further cladistic analyses on different ethnobotanical uses—such as medici-
nal and agricultural plants, plants used for clothing, shelter and tools, and plants
with ritual significance—could be overlaid in the same way that vicariate bio-
geographers overlay different plant and animal phylogenies to discover
relationships between diverse geographical areas. Such iterative cladistic analy-
ses (towards which our study is only a small step) could provide fascinating clues
and trends into ethnobotanical cross-cultural interactions. By overlaying such analy-
ses we might generate hypotheses of cultural interactions that may not be readily
apparent otherwise.

Winter 2000 JOURNAL OF ETHNOBIOLOGY 315

As indigenous knowledge systems throughout the world continue to disap-
pear, it is important to understand how ethnobotanical knowledge diffuses across
cultural boundaries. Using plants as shared innovations and comparing tribal use
of plant species using cladistic analyses may provide one key to understanding
such knowledge transfer. It is a simple technique that can clarify relationships
between indigenous cultures and elucidate the exchange of knowledge and tech-
nologies. Cladistic analyses may also render insights on plant technologies that
were independently invented versus those that were exchanged across cultural
boundaries.

LITERATURE CITED

AMSDEN, CHARLES AVERY. 1949. Naxajo
Weaving, its Technique and History.
oe New Mexico Srcae

Albuquerque

BELL, WILLIS H. ., and EDWARD F.
CASTETTER. 1937. The utilization of
mesquite and screwbean by the
aborigines i in the American sei
Southwest V. University of Nes vena
Bulletin on Biological Series 5:3-55.

BYE, ROBERT A., JR. 1985. Botanical
a of ph tae of the
greater Southwest. Economic Botany
39:375-386.

CARTER, GEORGE F. 1945. Plant
Geography and Culture History in the
American Southwest. The Viking Fund,

c., New York.
CASTETTER, EDWARD F., WILLIS H.
BELL, and ALVIN R. GROVE. 1938. The
early utilization and the distribution of
Agave i in the American | oaurey ert:

Southwest VI. University 0 of N ew
Mexico, Bulletin 335, Biological Series
5:3-92.

RUTH M. UNDERHILL.

New eevee Bulletin 275, Biological
Series

COLTON, “IARY-RUSSELL FERRELL.
1965. Hopi Dyes. Museum of Northern
Arizona Press, Flagstaff.

COOK, SARAH LOUISE. 1930. The
ethnobotany of the Jemez Indians
Unpublished MA thesis. University oi
New Mexico, Albuquerque.

DENNIS, MARY E. 1939. Plants and
animals used as sources of dye, paint,
and skin dressing in the Southwest.
— MS thesis. University of

New Mexico, Albuquerque

DOEBLEY, JOHN F. 1984. “Seeds” of wild
grasses: A major food of Southwestern
Indians. Economic Botany 38:52-64.

DUNMIRE, WILLIAM W. and GAIL D.
TIERNEY. 1995. Wild Plants of the
Pueblo Province. Museum of New
Mexico Press, Santa Fe.

ELMORE, FRANCIS H. 1943. ey
of the Navajo. The University of N
Mexico Bulletin, Monograph Series Vol.

1 no-7.

FARRIS, JAMES S. 1988. HENNIG86
Reference. Version 1.5. Software and
cane publnee by author.

989a. The retention index and
index.

the caled consistency
Cladistics 5:417-419.
89b. The retention index and
eo ei excess. Systematic Zoology
38:406-407.
FEWKES, J. WALTER. 1896. en
ethnobotany. e American
Anthropologist IX:14-21.
FORD, RICHARD I. 1968. An ecological
] eae: foo? al et _ —

Juan Pueblo, New Mexico. Ph.D.
Dissertation, University of Michigan,
Ann Arbor.

Anthea olen

perspective of ethnobota any in the
greater eae Economic
Botany 39:400-415.

FRANCISCAN FATHERS. 1910. An
Ethnologic Dictionary of the Navaho
Language. St. Michael’s, Arizona.

316 HART and COX

HEDLUND, ANN LANE. 1992. Reflections

of the weaver’s world: the Gloria F. Ross

peony of paayreiy a ae
eaving. Denver Art Muse nv

HOUGH, WALTER. 1897. The Hopi in
relation to their plant environment.
American oat st 10:33-44.

Environmental
ei in Arizona. American
arora 11: 133-155.

JEANCON, JEAN ALLARD and F. H.
DOUGLAS. aden aoe spinning,
dyeing and weaving. Denver Art
Museum, Dapertinent of Indian Art,
leaflet no. 3, Denver.

JONES, VOLNEY H. 1931. The ethnobotany
of the Isleta Indians. Unpublished
master’s thesis, univer of New
Mexico, Sate e

unusual Navajo
dye (Endothia aust ian 21:17-

KENT, KATE PECK. 1957. The cultivation

Southweiterh United —
Transactions of the American
Philosophical rag 47:459-733.
KITCHING, IAN J., PETER L. FOREY,
CHRISTOPHER J. HUMPHRIES, and
DAVID M. WILLIAMS. 1998. Cladistics:
The Theory and Practice of Parsimony
Analysis. Second Edition. Oxford
University Press, Oxford.
KLUCKHOHN, CLYDE. 1968. Navajo
Material Culture. Belknap Press,

ambridge.
KLUGE, ARNOLD G. and JAMES S.
FARRIS. 1969. Sra phyletics
and the evolu of anurans.

Systematic Jockey 18: 1-32.

MATHEWS, WASHINGTON. 1886. Navajo
names for plants. American Naturalist

776-777.

. 1891. Navajo dye stuffs.
Smithsonian Institution Annual Report
1890-1891: 613-615.

MAYES, VERNON O. and BARBARA
BAYL i Navajo
Herbal. Navajo Community College

Press, Tsaile.

MINGE, WARD ALAN. 1979. Effectos del
pais: A history of tha along the Rio
Grande. Pp. 8-28 in Spanish Textile
Tradition o Nes Mexics and Colorado,
tee bb nes or haut of New
Mexico Press, Sant

Vol. 20, No. 2

ORTIZ, ALFONSO ap as eta rae

Handbook of North Am

oo Vol. 10. pare mare
Institution, Washington.

PALMER, EDWARD. 1878. Plants used by
the Indians of the United States.
American Naturalist 12:593-606, 646-
655.

PEPPER, GEORGE H. 1903. Native Navajo
dyes. The Papoose 1:1-12

REAGAN, ALBERT B. 1929. Plants used by
the White Mountain Apache Indians of
Arizona. Wisconsin Archeologist 8:143-

REICHARD, GLADYS A. 1936. Navajo
Shepherd and Weaver. J. J. Augustin
Publisher, New York.

RIDLEY, MARK. 1993. Evolution. Blackwell
Scientific Publications, Boston.

ROBBINS, WILFRED WILLIAM, JOHN
PEABODY HARRINGTON
BARBARA FREIRE-MARRECO. 1916.
Ethnobotany of the Tewa Indians.
peony Institution, Bureau of
American Ethnology, Bulletin

ROESSEL, RUTH. 1983. Navajo arts and
crafts. Pp. 592-604 in Handbook of
North American Indians, Vol. 10, A
Ortiz, (editor). Smithsonian Institution,

ashington.
RUSSEL, FRANK. 1908. The Pima Indians.
a of barr ares Ethnology, 26th

nnual Report:
SAUER, JONATHON D. 1950. Amaranths

Southwestern Journal of Anthropology

6:412-415.

STANDLEY, PAUL C. 1911. Some useful

snes plants of New Mexico.
mithsonian Institution Report, 1911.

STEVENSON. MATILDA COXE. 1915.
Ethnobotany of the Zuni Indians.
Bureau of American Ethnology Annual
Report 30:31-102.

SWANK, GEORGE R.
ethnobotany of the Acoma and Laguna
Indians. Unpublished “MA thesis.
University of New Mexico,

ise The

Albuquerque.

VESTAL, PAUL A. 1940. Notes on a
collection of plants from the Hopi
Indian region of Arizona made by J. G.
Owens in 1891. Harvard Museum
Leaflets 8:153-168.

Winter 2000

1952. Ethnobotany of the Ramah
Tae Reports of the Ramah Project,
Papers of the Peabody Museum of
American Archaeology ee Ethnology,
Harvard University 11:1

WHEAT, JOE BEN. 1979. Rio Grande,
Pueblo, and Navajo weavers: Cross-
-36 in Spanish

Colorado, Nora Fisher, (editor).
Museum of New Mexico Press, Santa Fe.

WHITE, LESLIE A. 1945. Notes on the
ethnobotany of the Keres. ina of the
Michigan Academy rts, and
Letters 30:557-568.

WHITING, ALFRED F. 1939. Ethnobotany
of the Hopi. Museum of Northern
Arizona gps No. 15.

. The present status of
canitotas ny in the Southwest.
Economic Botany 20:316-325.

WINTER, JOSEPH CHARLES. 1974.
Gare ge 1 * el i eee. | tL ‘

and Great Basin. Unpublished PhD
dissertation. University of Utah, Salt
ty.

JOURNAL OF ETHNOBIOLOGY 317

WYMAN, LELAND C. and STUART K.
ARRIS. 1941. Navajo Indian medical
ethnobotany. University of New Mexico
Bulletin No. 366.
. 1951. The ethnobotany of the
Kayenta Navaho. Univ. of New Mexico
Bulletin No. 366.
na aera Saag (5): 1-76.
YOUNG, STELLA. 1940. Navajo Native
Dyes, Their net and Use
Chilocco Agricultural rt Printing,
Dept., Chilocco, Oklahom
'To whom ogee fa cence ‘should be
addres

318 HART and COX Vol. 20, No. 2

APPENDIX 1.—Plants used for dye purposes and corresponding tribes that use(d) them.
Plants are used to dye wool, unless otherwise indicated. Numbers in each box refer to
reference(s) that indicated tribal use of each plant.

7) wae oO 7)
4 | = 2 § © me ols s| x —
— © Se] Ee = = las Ole ©
SST eT etre Se he lea sR
= dea (eke Seka e-toc
© (3 ras ®
uJ = fa) =
ASTERACEAE 37
1. Actinea gaillardia
2. Actinea leptociada 37
SCROPHULARIACEAE ee
3. Adenostegia wrightii paint
1,7,9,12, 30
BETULACEAE 6 |15,19,20 16 FAT: 7.23\\ sth 7,23
4. Alnus tenuifolia leather oy a5.) a ©| leather
2,3
AMARANTHACEAE aon 29
5. Amaranthus ‘aula wafer
cruentus . bread
13,14 7
6. Amaranthus palmeri wafer wafer
bread bread
ASTERACEAE 37
7. Artemisia tridentata
GRAMINEAE a7
8. Arundo donax :
YOR
acetal ree 14 18,32,
Atriplex canescens wate 37 725
bread
|10.Atriplex rosea 18
|BERBERIDACEAE 20
11.Berberis fremontii 729
| 12.Berberis repens 20,37 8
BETULACEAE 7,14 6 14
13.Betula occidentalis leather | leather leather
14.Carduus 1
occidentalis
i — 7,10,13,
14,32,34 7
afer
soe bread
SCROPHULARIACEAE 37 7,29
16.Castilleja integra 23 inate
17.Castilleja 34
linariaefolia

Winter 2000

JOURNAL OF ETHNOBIOLOGY

319

Eastern Keres

Hopi
Jemez (Towa)
Navajo

Papago

Pima

Southern Tiwa

Tewa

Western
Apache

Western Keres

Zuni

LINACEAE
18.Cathartolinum
puberulum

MN
Ww
o

paint

ULMACEAE
19.Celtis reticulata

breviflorus

21.Cercocarpus
tanus

leather | 20,22,24 |
28,3

16
leather

30
leathe

ASTERACEAE
22.Chrysothamnus
latisquamena

23.Chrysothamnus
nauseosus

20,32,37

7,23

729

24.Chrysothamnus
parryi

body
paint

25.Chrysothamnus
viscidiflorus

9
plant
materia

CAPPARACEAE
26.Cleome serrulata

Ww
Ww

20,37

16,33

7ikks
25,35
paint

27 .Coreopsis
cardaminefolia

ROSACEAE
28.Cowania mexicana

CUCURBITACEAE
129.Cucumis melo
(introduced)

body
paint

30.Cucurbita
foetidissima

CHENOPODIACEAE
31.Cycloloma
atriplicifolium

RANUNCULACEAE
|32.Delphinium
scaposum

24,37

|CRUCIFERAE
133.Descurainia pinnata

8
paint

720
genus
only

320 HART and COX Vol. 20, No.
8 g S é
o to) ° ro) be © eo
ol Se KR — te) o Sot eo) oe nee
vee
. e| =| a = =<) 2
© ® ro) o
LJ = ” =
IASTERACEAE 47
34.Encelia farinosa ;
35.Endothia singularis 7
(fungus)
GNETACEAE 912
36.Ephedra trifurca ‘
37.Ephedra viridis 37
38.Erysimum 8,30
OLEACEAE 6
39.Forestiera body | 37
neomexicana paint
ERICACEAE
40 eria 72
humifusa
ASTERACEAE
41.Gutierrezia 18
sarothrae
42.Helenium hoopesii =
43.Helianthus annuus 9
44 Helianthus faa,
petiolaris 14
SAXIFRAGACEAE i058
45.Heuchera bracteata ee
46.Hymenoxys 7,37
metcalfei
IRIDACEAE
ee AAO eg 20
47. Iris missouriensis
| JUGLANDACEAE
48.Juglans major 1237
49.Juglans regia "
(introduced)
|CUPRESSACEAE 18
50.Juniperus deppeana |

Winter 2000

JOURNAL OF ETHNOBIOLOGY

Eastern Keres

Hopi

Jemez (Towa)

Papago
Pima

Southern Tiwa

Tewa

Western
Apache

Zuni

51 Juniperus
monosperma

. é SIWestern Keres

52 —
osteosperm

body
paint

—

53.Juniperus
scopulorum

FABACEAE
54.Krameria parvifolia

47
cotton

7,26
a Pris leather

ZYGOPHYLLACEAE
55.Larrea tridentata

47
tattoo

56.Letharia vulpina
(lichen)

FABACEAE
57.Lupinus kingii

58. Medicago sativa
(cultivated)

NYCTAGINACEAE
59.Mirabilis multiflora

LILIACEAE
60.Nolina microcarpa

CACTACEAE
61.Opuntia

paint

anni
62.Opuntia
phaecantha

63.Opuntia polycantha

64.Parmelia
molliuscula (lichen)

1937

FABACEAE
65.Parryella filifolia

AMPELIDACEAE
.Parthenocissus
vitacea

body
paint

ASTERACEAE
67.Pectis angustifolia

68.Petradoria pumila

S22 HART and COX Vol. 20, No. 2
8 g E é
© oO fo) oO | ood S 2 @®
seal ae ee ee ow 0 e c S igs =< a
Se; 3S N = a £ o eine & =
o| = o S oO aX < — |©o a o Ni
D = a s =< 2
Ar} s 4 =
FABACEAE
69.Phaseolus vulgaris 3,5,34
(cultivated)
LORANTHACEAE
70.Phoradendron 9,23
juniperinum
PINACEAE 9
71.Picea pungens
TSI5,
; : 19,20,
72.Pinus edulis 34 6 22°32 30
CAPPARACEAE
73.Polanisia 8,25
FABACEAE 7,10 one
74.Prosopis velutina paint hair dye
ROSACEAE 16
75.Prunus americana — leather
76.Prunus emarginata 9
77.Prunus 37
melanocarpa
78.Prunus persica .
(introduced)
79.Prunus virginiana 20
UMBELLIFERAE
0. Pseudocymopterus 37
montanus
81.Psilotrophe 23 | 8,30 | 7,8,29
a
MONOTROPACEAE
82.Pterospora 37
andromedea
PYROLACEAE 36
83.Pyrola chlorantha paint
ROSACEAE 9
84.Pyrus malus
FAGACEAE
85.Quercus gambelii 20,37

Winter 2000 JOURNAL OF ETHNOBIOLOGY 323
8 g § 8
“alt oe] ele le ies =
c| 9 > a} £ ® o |na £ =
o| <x ® bod 6 a r= ie oe |
< | 21 & 5 = &
0 ® o ©
uJ Bey ” =
86.Quercus pungens 7,36
1,7,9,15
ANACARDIACEAE 5.33 18,19,22 7
87.Rhus aromatica : 24,27,
31,35,36
POLYGONACEAE 1,7,9,12,
: 5,8,34 15,18,19
hymenosepalus e452,57
725
SALICACEAE body
89 Salix irrorata paint
CHENOPODIACEAE 37
90.Salsola kali
Sacrobatus 20
vermiculatus
{ASTERACEAE 24
92.Senecio douglasii
7
bod
paint
CRUCIFERAE 7,25
94 Stanleyella wrightii
ASTERACEAE 9.37
95.Tagetes micrantha :
96. Taraxacum A vei
officinale
97.Tetradymia 18
canescens
98.Thelesperma = re 18,32,
megapotamicum aa 37
99. Thelesperma 9
subnudum a4
TYPHACEAE 7
100.7ypha angustifolia
1101.Ustilago zeae bee
y
(fungus) paint
ERICACEA
{102.Vaccinium 9
humifusu

324 HART and COX Vol. 20, No. 2

Eastern Keres
Hopi
Navajo
Papago
Pima
Southern Tiwa
Tewa
Western
Apache
Western Keres
Zuni

AMPELIDACEAE
103. Vitis arizonica

2 fon)
58 | Jemez (Towa)
we

'104.Xanthium body
paint

commune
105.Xanthoparmelia
en 18,32
lichen)
LILIACEAE
106. Yucca glauca 18,32
GRAMINEAE
107.Zea mays 5,34
(cultivated)
ASTERACEAE i
|108.Zinnia grandiflora

* Ref f ific citati f each plant d diff t tribes. See literature cited for full
citation.

1. Amsden 31. Vestal 1940
2. Bell rie icp 1937 32. Vestal 1952
3. Carter 1945 33. White 1945
4. rete and Underhill 1953 iti
5. Colton 1965
6

74
; Cook "1930 36. Wyman and Harris 1951
7. Dennis 1939 37. Young 1940
8. Dunmire < Tierney 1995
9. Elmore 194.
10. Fewkes 186
11. Ford 1968

12. Franciscan Fathers 1910
sa

16. Jones 1931

17. J

20. pet and Lacy 1989

25. Robbins, Harrington and Freire-Marreco 1916
26. Russel 1908
27. Sauer 1950
28. Standley 1911
. Stevenson 1915
30. Swank 1932

Winter 2000 JOURNAL OF ETHNOBIOLOGY

sist: 2. ¥ a|$ = =} bab 512 ial f-~
lalalzielg lala eal = a sla als
o 5 0}/° < FS re) a\=
ol Fs ro) o
=) —= 2
~ 3 ef me] a maa} | fein} | | 64
~ } ok ab 5
= ) ial pa 6
= are 7
_ ae 8
= } ‘ale ard (gag pre 9
_ al 10
= =| ad 11
= = —_ 12
ab ( = ab] ad 13
> 68 ad) 14
~ 69 ol tal ;
a ) a lati oul 4
wade
= ey ee 72 = 3
= 73 ah
=| 74 = )
oath ae 7: aa oh ablabl lad)
‘a r —_=! 4
ie s =| =a} =a} oa) = }
a 3 rs |
_ ] a
Be ) = =| =! ally ab)
2 a Be ar eg are
n 4 alt)
a 3 —_ ]
ee q call )
ee ) art
= a ai =k =| }
oath 3 Ss
= ) «hl
—_ ) baad
a alt
al J Ss }
ali } =| }
an q _ 40
p< coe 4
= os 4
ab) F a 43
a ee 3 = é
a i 4 é
_ 100 - y.
= 102 - 48
aii 103 ~ 49
cai 104 50
lk 105 al é
4 Ge 22
107 53
ale 108 u

APPENDIX 2.—Data matrix for cladistic analysis. Characters are represented by
individual plant species as numbered in Appendix 1. A blank cell denotes plant not
used by tribe, a “1” denotes plant used.

325

NOTICE TO woe 53 00298 2038
The Journal of Ethnobiology’s revised “Guidelines for Authors” appears in volume 19, 9, number two |
(winter 1999). If you need a copy of these Guidelines you may request a copy from the Editor.
Careful scrutiny of recent issues of the Journal should provide appropriate stylistic models for any
manuscript you may wish to submit.

Th bmitti f ideration for publication in the Journal should send three
hard copies and o one copy on a te density 3.5" pe arse in IBM-compatible or Apple format with
original camera-ready figures. Manuscripts submitted in inappropriate style and format will be
returned.

Manuscripts should be sent to:
WILLIAM BALEE
Editor, Journal of Ethnobiology
Department of Anthropology
Tulane University
New Orleans, LA 70118 USA
(wbalee@tulane.edu)

BOOK REVIEWS
We welcome suggestions on books to review or actual reviews from readers of the Journal. If you
submit a book review for consideration please send two hard copies and one on diskette. Send
suggestions, comments, or reviews to:
MICHAEL K. STEINBERG
Book Review Editor, Journal of Us
Department of Geography- opology
University of Southern ihe
Gorham, ME 04038
(mstein@usm.maine.edu)

SUBSCRIPTIONS
Subscriptions to the vet i Ethnobiology should be addressed to Virginia Popper, UCLA Institute
of Archae eology, Box 951510, Fowler A-210, Los Angeles, CA 90095-1510 USA (Email:

ada, and Mexico, $35
HACE Heo Shas eal eC pas t for Mexico, for whic

vpopper@ucla.edu). Subscription sia are win v4 institutional nape nit in the U.S., Canada
ana

and Mexico, $90 for instituti
or students el }
subscriptions are $25; individual best from Kicica, Asia, Avaivalis, and Europe are $45. For
postage outside of the USA, Canada, or Mexico, add $10. Make checks payable to the Jour nal of
Ethnobiology. Defective or lost copies will be replaced if a written request is received within one year
of issue. For information on back issues contact Cecil Brown, Department of Anthropology,
Northern Illinois University, DeKalb, IL 60115 USA, (815) 753-0246. For further details on subscrip-
tions and the Society of Ethnobiology in general, consult http:/ /www.ethnobiology.org.

CONTENTS

ETHNOBIOTICA iv

MAYA KNOWLEDGE AND “SCIENCE WARS”
Anderson 129

EASTERN SUMBANESE BIRD CLASSIFICATION
Forth 161

KNOWING, GATHERING AND EATING: KNOWLEDGE AND ATTITUDES
ABOUT WILD FOOD IN AN ISAN VILLAGE IN NORTHEASTERN
THAILAND

Somnasang & Moreno-Black in7

THE JAGUARS OF ALTAR Q, COPAN, HONDURAS: FAUNAL ANALYSIS,
ARCHAEOLOGY, AND ECOLOGY
Ballinger & Stomper 223

PERCEPTION AND MANAGEMENT OF CASSAVA (MANIHOT ESCULENTA
CRANTZ) DIVERSITY AMONG MAKUSHI AMERINDIANS OF GUYANA
Elias et al. 239

THE HISTORICAL ECOLOGY OF SOUTHEASTERN LONGLEAF PINE AND
ITS SOUTHERNMOST EXPRESSION
Walker 269

A CLADISTIC APPROACH TO COMPARATIVE ETHNOBOTANY
Hart & Cox 303

BOOK REVIEWS 159, 193, 217, 237, 266, 300

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Journal of ethnobiology."

See other formats