(Scott Atran, Douglas Medin) The Native Mind
(Scott Atran, Douglas Medin) The Native Mind
(Scott Atran, Douglas Medin) The Native Mind
A Bradford Book
The MIT Press
Cambridge, Massachusetts
London, England
6 5 4 3 2
2007013988
1
Contents
Preface
vii
Introduction
Cultural Epidemiology
10
277
References
Index
321
291
17
49
63
121
143
209
255
225
Preface
The research described in this book began when we found ourselves sitting next to each other at a workshop dinner at the University of Michigan more than two decades ago. We talked. The patterning of shared and
distinctive perspectives made the prospect of research collaboration exciting. When graduate students, postdoctoral fellows, and faculty colleagues
were added to the mix, a community of scholars evolved to address what
we think are fundamental theoretical, empirical, and methodological
issues at the interface of anthropology and psychology.
The research was and is intense. In the eld, hot days of interviewing
were followed by heavy nights of analysis, spiked and lightened occasionally with rotgut rum (not charged to our grants). We worked through discussions that ranged from competing denitions of culture and nuances of
cultural modeling to speculating about how soon the rst Jet Skis would
be seen on Lake Peten Itza in the Guatemalan rain forest. We vividly
recall one night of analyses when the power went out (a not infrequent
occurrence). We continued in the dark, going from three to two to one
computer as battery power played out. Finally, we called it quits for the
night when we could no longer distinguish between the cursor and the
insects landing on the computer screen.
Not everything was fun. This sort of research is expensive, and we
spent an inordinate amount of time writing grant proposals. In a classroom, if you ub an experiment, you can go to the next class. In the eld,
you may have to wait for the next seasonif youre luckyto get people
from several countries, universities, academic disciplines, and dierent
forest activities to coordinate schedules again. It also was not a thrill to
discover that the rst set of soil samples we collected was useless because
the critical identifying information was put down on paper in pencil, only
to become some unidentiable fungus after being enclosed in the humid
sample containers. Finally, we collected so much data that sometimes we
spent a lot of time tracking down where the original data were and in
what stage of analysis.
viii
Preface
This research was supported by grants from NSF (SBR 931978, SBR
9422587, SBR 9707761, SBR 9983260, BCS 0216762, SBR 0424629, BCS0132469, REC-0529650), NSF/EPA (SES-9981762), NIH (MH55079), the
Russell Sage Foundation (87-99-02), and CNRS (contract no. 92 C 0758,
MRT). This work is the summary of a combined eort by a number of
long-term research collaborators: Norbert Ross, Sandra Waxman, Elizabeth Lynch, Edilberto Ucan Ek, Valentina Vapnarsky, John Coley, and
Ximena Lois. Other important contributors and collaborators include:
Jeremy Bailenson, Michael Baran, Serge Blok, Russell Burnett, Doug
Cox, Paul Estin, Brett Hayes, Alejandro Lpez, Brad Love, Bobbi Low,
Richard Nisbett, Connie Pankratz, Elizabeth Prott, A. Kimball
Romney, Hillarie Schwartz, Michael Shum, Brian Smith, Edward Smith,
Paolo Sousa, Gert Storms, Christopher Timura, Sara Unsworth, Dan
Bartels, and Phil Wol. Dan Sperber, Lawrence Hirschfeld, Michael
Cole, Lance Rips, Susan Gelman, Ed Hutchins and William Batchelder
provided constructive comments on earlier drafts of the manuscript. Jennie Woodring was instrumental in the copyediting and detail tracking
part of the production process. We are deeply grateful to our research
partners and participants in Guatemala, Mexico, Brazil, Wisconsin, Illinois, and Michigan, and especially to the Maya and Menominee communities that took us in. Last but foremost, we are thankful for our
loved ones who support our scholarly passions.
Introduction
It is the best of times; it is the worst of times. Recent years have seen stunning advances in understanding the basis of life, ranging from unraveling
the human genome to discovering extraterrestrial sources of generation
and extinction. Although some would argue that macrobiology has been
neglected in favor of microbiology and biotechnology, there has been undeniable progress in understanding complex systems, including ecosystems. From remote sensing to environmental chemistry, scientists are
delivering insights into how to protect the environment. And awareness and concern about life on our planet is widespread. A recent survey
found overwhelming endorsement of the statement that we have a moral
duty to leave the Earth in as good or better shape than we found it
(Kempton, Boster, and Hartley 1995, 257). No part of the earth is
untouched by advances in both the science of nature and commitments
to support it.
But it is also the worst of times. There is an increasing sense of diminished human contact with nature, a phenomenon some refer to as the extinction of experience (Nabhan and St. Antoine 1993) and others as a
Nature-decit disorder (Louv, 2006). Respondents in the same survey
above agreed that the majority of people are completely cut o from nature. They spend their time indoors and when theyre outdoors, nature is
just an inconvenience to them.
Under such circumstances, commonsense knowledge of nature is poor,
sometimes surprisingly so. As we write this, buckeyes and horse chestnuts
are owering in northern Illinois, but fewer than half of Northwestern
University students surveyed say they have even heard of horse chestnut
or buckeye. Below is part of an interview with a Northwestern Honors
student who expressed surprise that 3- and 4-year olds were asked to
give examples of plants. She was then asked to generate examples herself:
Chapter 1
It does not take a genius to gure out that we live in a fragile world. We
are told not to eat sh more than once a week and pregnant women are
advised to avoid sh altogether because of mercury contamination.
Recent droughts have left Lake Powell at half its former size and the
Western region of the United States faces serious water shortages. Global
warming seems almost minor by comparison. And it is not just what we
are doing to the earth. We live under the shadow of terrorism that
threatens to escalate into nuclear, chemical, or biological warfare.
Even business as usual involves frequent cultural clashes over natural
resources, such as access to salmon, preserving spotted owls, or rights to
land. This book is about both of these dimensions: rst, the relationship
between how people think about the natural world and how they act on
it, and second, cultural dierences in these understandings and how
they contribute to intergroup conict or cooperation. Our enterprise is
grounded in the cognitive sciences, and relevant ascriptions include resource dilemmas, mental models, culture and cognition, folkbiology, categorization and reasoning, protected or sacred values, and environmental
decision making. Each of these topics is central to our eorts. An important overarching theme is that they can best be addressed by bringing psychology and anthropology together.
Claims about the nature of human nature, with their associated policy
implications, require the very best analyses that an interdisciplinary cog-
Introduction
nitive science has to oer. Peoples behavior toward the natural world is
surely conditioned in part by their ways of knowing and modeling it.
What are these modes of knowledge and understanding? How are they affected by goals, theories, and intimacy of contact with the biological
world? What is universal, what is not, and what are the implications of
such observations for insights into the development of biological cognition? How does cognitive and cultural development lead some people to
protect their environment and others to destroy it? These questions shape
the present book.
1.2
Much of human history has been spent (and is being spent) in intimate contact with plants and animals, and it is dicult to imagine that
human cognition would not be molded by that fact. In subsistence
cultures, survival depends on a detailed appreciation of the habits, aordances, and interactions linked to the biological world. In technologically
advanced cultures, which are increasingly faced with environmental degradation and nonsustainable use of natural resources, no less may be at
stake.
There are a series of mutually reinforcing reasons for studying peoples
understanding of the biological world. First, biology represents a natural
unit of analysis and appears to be a core component of human cognition.
To get along in the world, people need to be able to understand and predict the general properties and behaviors of physical objects and substances (physics), the more specic properties of plants and animals
(biology), and the particular properties of their fellow human beings (psychology). We will argue that biology represents a distinct module of mind
that is associated with universal patterns of categorization and reasoning.
Some of these principles are robustly universal and some depend on having more than modest contact with nature. Yet others are highly dependent on particulars of cultural models and associated values. Each of
these three classes of ndings has important theoretical and practical
implications.
A second reason for a focus on biology is that biological kinds provide
a natural metric for cultural comparisons. Although the specic animals
and plants vary considerably across the world, scientic taxonomy constitutes a structure for comparing kinds at corresponding levels of specicity
(e.g., bird and sh versus duck and trout). Ethnobiologists often begin
their research in some area by conducting a survey of local plants and
Chapter 1
In many respects anthropology and psychology are perfect foils for each
other because of their complementary strengths (and weaknesses). Initially, we focus on limitations of cognitive psychology in order to set up
a contrast with anthropology.
Introduction
Universality
One of the strengths of cognitive psychology is its focus on systematic, controlled comparisons. Historically, this
concern was so strong that psychologists studying learning made use of
nonsense syllables to limit any inuence of prior experience or knowledge. Although this particular habit has been discarded in favor of using
meaningful materials, experimentalists have concentrated on nding
materials with particular desirable properties (in terms of controlling for
Chapter 1
extraneous factors), with little concern for the relation between such
materials and the range of stimuli over which one might wish to generalize. The idea of systematic sampling is somewhat alien, perhaps because it is not always clear how to answer the question, systematic with
respect to what? For example, if a psychologist wants to compare reasoning involving living kinds versus human artifacts as stimuli, he or she
typically would generate examples, subject only to the constraint that
undergraduates be familiar with them. Rarely, if ever, would it occur to
the psychologist to ask what kinds of artifacts or what kinds of living
kinds there are and how one might go about selecting a representative
sample.
A related limitation of research in cognitive psychology is
that conceptual behavior is often studied with little concern about reference. For studies involving adults, the stimuli are often words and seldom
does the researcher establish any relation between these words and what
their referents are. For living kinds other than mammals, college students
often have little idea about reference beyond a very general level (e.g.,
such-and-such is a tree). For developmental studies concerned with
living kinds, the stimuli are typically toys, which are at best representations of living kinds. Again reference is rarely established. Of course,
for some questions of interest, reference may not matter, but our impression is that, as in the case of populations and stimuli, convenience
and control tend to dominate a systematic analysis of the domain of
interest.
Reference
Anthropology as a Contrast
Introduction
could be driven by relevance to humans (utility), perceptual discontinuities, or even size (Hunn 1999).
Scientic taxonomies are, of course, hierarchical taxonomies and, as
such, provide both a set of standards and a heuristic for asking other
questions about universal aspects of folktaxonomies. There are two important analytic points involved here. One is that although the particular
kinds of plants and animals to be found may vary across cultures, the abstract structure in terms of species, genus, family, order, class, division,
and kingdom will be represented. Consequently, scientic taxonomy provides something of a conceptual grid for cross-cultural comparisons. The
second, related point is that using a scientic taxonomy allows one to establish corresponding ranks such that it becomes meaningful to state that
oak is at the same level or rank as is trout. This does not mean that they
are psychologically at the same rank, but it does provide a basis for asking questions such as whether some culture dierentiates mammals more
than sh. As it turns out, ethnobiologists have found that folk ranks and
folktaxonomies only loosely approximate scientic taxonomies but formal taxonomy has served as an eective standard for cross-cultural comparisons (Hunn 1975).
Note that the practices that are most natural for an ethnobiologist
address each of the limitations that we have attributed to cognitive psychology. Folktaxonomic analyses provide a framework in which one can
propose and evaluate hypotheses about cognitive universals (Berlin, Breedlove, and Raven 1973). The main criticism we can oer for the issues in
question is that ethnobiologists have tended to focus rst on small-scale
subsistence cultures to the neglect of larger, more industrialized cultures,
and second on culturally competent adults rather than children (Hirschfeld 2003). In sum, so far pretty good for anthropology.
Critique of Anthropology
Chapter 1
Introduction
warms to the sun. (See Strauss and Quinn 1992 for a critique of this view
from within anthropology.)
If anything we may be guilty of downplaying this issue. At times within
anthropology, the methodological point that anthropological observations
are socially constructed has been elevated to a form of self-immolation
that threatens to destroy the science part of anthropology as a social
science and move it squarely into literature. As one of our anthropological colleagues is fond of saying, eldwork should focus on research
that is liable to awe our own, often complacent culture with the diversity of collective human imagination and action. We readily grant the
importance of demonstrating the rich variation in human thought and experience, but we think that more rigorous science could better help to
make the case.
One tricky thing about knowledge is that
there are no free peeks at mental representations. This is true both for
the scientist and the informant. At one point researchers interested in
developing computers as expert systems hoped that knowledge could be
transferred from human expert to machine simply by asking the expert
to report what he or she knew. This eort was largely unsuccessful because experts cannot, by an act of will, simply make their knowledge accessible. Articial intelligence knowledge engineers and psychologists
have learned to use indirect measures of knowledge and to draw inferences from patterns of behavior. This is an important operating procedure in cognitive sciencethat is, developing and testing methods and
models that foster inferences about knowledge representation and use.
Quantitatively based models and theories are not complete strangers to
anthropology, but neither are they intimate friends.
10
Chapter 1
The challenge of understanding biological cognition is daunting. Consider the presumably simpler task of understanding temperature regulation, a problem that has its own evolutionary history. Here it has been
found that temperature regulation in human beings involves the integration of multiple parallel systems (e.g., shivering, sweating, putting on
clothes) that vary in their renement and redundancy (e.g., Satino
1983). We should not expect anything less for something as intricate as
peoples understanding of the natural world.
1.4 Theoretical Issues in the Cognitive Science of Folkbiology
Introduction
11
the one hand, our perceptual system is surely an adaptation to the natural
world, and if similarity-based models are going to succeed anywhere, it
should be here. On the other hand, the biological world is apparently a
world of fairly stable clusters of complex features whose remarkable endurance in the face of constant change can presumably be explained in
terms of naturally occurring causal patterns. Understanding causal patterns in the world is a primary goal of theory-driven knowledge in
science, and the history of science is coterminous with trying to understand biological causality in particular. If theory-based knowledge were
to develop anywhere outside of sciencein other cultures or in everyday
thinkingit should be here.
From the perspective of similarity, there are evident patterns of covariation for biologically related attributes: toothless two-legged beings generally have wings, feathers, and y; leaves, owers, and fruits generally go
together with stems and roots; and so on. Perhaps most people in the
world are aware of these covariations without necessarily understanding
their causal origins or interrelations, such as the role of feathers in ight
or of leaves in stem development. In other words, there could be quite
a bit of biologically relevant data that is stored but not theoretically
assimilated.
Nevertheless, people in dierent cultures acknowledge, and often try to
better understand, at least some of the causal interrelations among covariant biological attributes. These include irreversible patterns of biological
growth (maturation); the apparent constancy of covariant morphological,
anatomical, and behavioral patterns across generations (reproduction and
inheritance); the success of mutually constraining actions of interrelated
attributes in maintaining life (bodily functioning); and the breakdown of
interrelated bodily functions (illness and death). Moreover, these naive
attempts at causal explanation are themselves interrelated, often with the
sort of resultant explanatory bootstrapping and integration of the database that could help to kick o the development of science.
Suppose, as ethnobiologists generally agree, people everywhere witness
certain covariant biological patterns (roughly corresponding to perceptually salient species or genera), but interpret the causal relationships
underlying these patterns in dierent ways. This might suggest that
similarity-based reasoning is prior to theoretically based reasoning, at
least in the biological domain. This was a message of developmental
studies in the 1980s (Carey 1985; Inagaki and Sugiyama 1988; Keil 1989).
More recent studies have lowered the age at which children are thought
12
Chapter 1
to reason causally about biological kinds. But the origins of causal reasoning in folkbiology remain a matter of controversy.
A closely related question concerns which factors shape the acquisition
of biological knowledge and the extent to which their inuence extends to
adult (more or less steady-state) knowledge. Researchers in the area of
cognitive development have been actively studying the role of language
in conceptual development (see Waxman 1999, 2004) and are increasingly
turning to an analysis of the role of input conditions (Hatano and Inagaki
2003; Gelman et al. 1998), at least at intermediate stages of development.
3. Is folk biology a naive form of scientic biology? To some extent,
the fact that most psychologists prefer the label naive biology or intuitive biology over the ethnobiologists folkbiology implies somewhat
dierent understandings and uses of scientic biology as a standard of
comparison. For those interested in the structure and development of biological causality in our own culture, folkbiological concepts often appear to contain rudimentary or inchoate elements and clusters of
more sophisticated scientic concepts. Although there has been little systematic study of the input conditions and processes by which scientic
concepts are assimilated into lay thinking, there is hardly any doubt that
science is pervasively involved in how people in our culture come to think
about the biological world. The inuence of science may be especially
pronounced among the university subpopulations psychologists prefer to
study, but most of the general population is heavily exposed to scientic
concepts in one form another through schooling, nature programs on
television, popular books, the press, and so forth.
The elaborate folkbiological inventories that ethnobiologists have
shown time and again for many small-scale subsistence societies often
match and occasionally even surpass in intricacy and accuracy the knowledge of eld biologists working in the same locales as those societies (e.g.,
Bartlett 1936; Simpson 1961; Bulmer and Tyler 1968). Moreover, few ethnobiologists would consider it enlighteningbut rather misleadingto
characterize the signicant dierences between folk knowledge in other
cultures versus science in terms of relative degrees of intuition or navete.
Admittedly, ethnobiologists might well agree with psychologists about referring to lay biology in our culture as naive in comparison to the relative sophistication of science as well as folkbiological knowledge in other
cultures.
A key issue is whether basic folk concepts, such as folk species or
generics, are dierent in kind from contemporary scientic concepts,
Introduction
13
such as the idea of a species as a logical individual (i.e., a lineage of connected parts) rather than a logical class (i.e., a meaningful collection of
individuals) (Ghiselin 1999). If they are not really dierent in kind, but
only in degree of sophistication, then there may be no reason for holding
on to the lay concept at all, except perhaps as an optional psychological
convenience for navigating the everyday world (see also Kripke 1972;
Putnam 1975). If, however, folk and scientic concepts are dierent in
kind, then perhaps they have separate but equalor at least dierent
roles to play in the attainment of knowledge (Dupre 1999; see also
Braisby et al. 1996). Folk concepts would be useful for accommodating
to the everyday world and scientic concepts for exploring the cosmos at
large (including extended thoughts about evolutionary dimensions of
space and time that would be largely irrelevant to ordinary understanding
and action).
Finally, one might accept that folk and scientic concepts may be different in kind, or that folk concepts are in some sense psychologically
more convenient in a given culture or at a given stage of history or development, but argue that folk concepts ought to be replaced by scientic
concepts (Hull 1999; cf. Russell 1948). For example, if it is true that people ordinarily believe that living kinds (including humans) have underlying essences (see Hirschfeld and Gelman 1994), then it is also likely that
people will treat natural variation as deviance. If so, then the essentialist
folk concept should be discarded along with other outworn commonsense myths, such as belief in witches or races, no matter how hard it is
to unlearn them. Even if this should be case, however, understanding how
people do in fact think about biological kinds (and other biologically related phenomena discussed in this book, such as diseases) may help us all
to better cope with them.
What is at stake in the interdisciplinary study of peoples understanding of biology? A lot. Can human beings make the transition
from locally sustainable adaptation to (technologically driven) global
economies without irreparably damaging our environment or destroying
local cultures? To address such issues researchers may need to integrate
questions about the structure of biological cognition with systematic analyses of how knowledge is linked to action in diverse ecological and cultural contexts (Atran and Medin 1997; Atran, Medin, and Ross 2005).
We hope that this book provides new intellectual tools for understanding
how humans come to know nature.
Summary
14
Chapter 1
1.5 Themes
In this book, we describe historical, cross-cultural, and developmental research on how people conceptualize nature (naive or folkbiology) and
how they act on it (folkecology). This represents the results to date of an
ongoing multidisciplinary, multinational project begun in 1991. Here we
concentrate on cognitive, cultural, and historical processes in the devolution of knowledge and the consequences of devolution for environmental
management. Our approach integrates three disciplinary perspectives:
For cognitive psychology, we examine how results gathered from standard populations in industrialized societies often fail to generalize to humanity at large. This leads us to an account of several fundamental
human processes of categorization and reasoning that dier substantially
from current accounts. An important factor motivating our experiments,
and our interpretation of them, is how plausible the results appear in light
of evolutionary biology and psychology.
For developmental research, we nd that usual study populations represent instances of impoverished experience with nature. This has serious
implications for science education in our own society. Perhaps even
more vital, this may help to reverse todays dismal prospects of integrating science and folk knowledge in other societies in ways that do not denigrate or destroy valuable and often irreplaceable local understandings of
nature.
For cultural and environmental studies, we show that even groups living
in the same habitat can manifest strikingly distinct behaviors, cognitions,
and social relations relative to it. Understanding why some people work
in a way that degrades the environment while others manage to preserve
and even enhance ecological diversity and resilience has critical implications for environmental and political decision making. It bears directly
on how our species might deal with increasingly dire problems of sustaining our common environment as globalization advances. This line of research suggests a novel way of studying culture and culture processes and
it points to a perspective on decision making that emphasizes values
and meanings over probabilities and utilities.
We argue that cultural transmission and formation does not consist only,
or even primarily, in the inheritance of shared codes of thought and behavior, but in complex distributions of causally connected representations
across minds. Instead of viewing culture as a top-down structure that
imposes itself on individual minds, we focus on modeling microprocesses
Introduction
15
Book Summary
16
Chapter 1
For more than a decade we have been investigating the cognitive consequences of reduced contact with naturewhat Nabhan and St. Antoine
(1993) refer to as extinction of experience and what we call devolution. This chapter provides background for our research program in the
domain of naive or folkbiology. It bears not only on knowledge loss or
devolution but also on universal aspects of biological cognition. We begin
with a brief summary of candidate principles for evolved universals and
associated historical developments in scientic taxonomy. Then we turn
to evidence demonstrating that cultural support for attention to nature
has been diminishing since the onset of the industrial revolution.
2.1
18
Chapter 2
19
informants are likely to infer that other oaks may also be susceptible to
this disease. As we will see, the detailed character of the induction varies
with experience and cultural background.
There is also growing cross-cultural evidence of a commonsense assumption that each folkspecieswhat we will refer to as generic species for reasons discussed belowhas an underlying causal nature, or
essence, uniquely responsible for the typical appearance, behavior, and
ecological preferences of the kind. On evolutionary grounds one would
expect that innate potential is vested at the generic-species level: for the
most part, generic species are genetically, geographically, and reproductively isolated (Mayr 1982 calls these nondimensional species). Hence,
we would expect presumptions of essence to be at the generic-species
level, where innate potential is.
There are thus strong constraintsplausibly naturally selectedon
how people organize local knowledge of biological kinds. Universal appreciation of generic species may be one such functional adaptation.
Pigeonholing generic species into a hierarchy of mutually exclusive taxa
allows incorporation of new species and biological properties into an
inductively coherent system that can be extended to any habitat, facilitating adaptation to many habitats (a hallmark of Homo sapiens). In the
chapters that follow, we will provide additional evidence that folkbiology
is a constrained domain of development and that its core aspects are either innate or universally acquired under some minimal, adequate input
conditions.
2.2
In every society people think about plants and animals in the same special
ways (Berlin 1992). The science of biology also treats plants and animals
as special kinds of objects, but applies this treatment to humans as well.
Folkbiology, which is present in all cultures, and the science of biology,
whose origins are particular to the Western cultural tradition, have corresponding notions of living kinds. Consider four corresponding ways in
which ordinary folk and biologists think of plants and animals as special
(Atran 1998).
Four Points of General Correspondence between Folkbiological Taxonomy and
Scientific Systematics
20
Chapter 2
21
22
Chapter 2
example, beating of the heart and circulation of blood give prima facie
mechanical evidence for causal activity. In addition, loss of blood and
stopping of the heart are often signs of loss of life. Thus, heart and blood
may be privileged candidates for the locus of essence, as they have been
throughout the history of European societies (Atran 1990). Even contemporary Americans who undergo heart transplants show evidence of
believing that at least some aspects of essence have been transmitted
from the donor to the recipient (Sylvia and Novak 1997). In dierent cultural settings, other plausible candidates (e.g., milk as conveyer of essence
through nursing) may have priority (Stoler 1995). Willingness to allow
transformations of essential kindhood (e.g., through blood transfusions,
organ transplants) also may depend on cultural context (Walker 1992;
Mahalingam 1998; Waxman, Medin, and Ross 2007). Even in cultures
where the adult discourse is antiessentialist, it appears that both children
and adults essentialize animals (Astuti 2002; Astuti, Carey, and Solomon
2004). Indeed, essence-related notions of nonintentional and nonmechanical causal processes, continue to agitate science.
Vitalism is the folk belief that biological kindsand their maintaining
parts, properties, and processesare teleological, and hence not reducible
to the contingent relations that govern inert matter. Its cultural expression varies (cf. Hatano and Inagaki 1994). Within any given culture people may have varying interpretations and degrees of attachment to this
belief: some who are religiously inclined may think that a spiritual essence determines biological causality; others of a more scientic orientation might hold that systems of laws that suce for physics and chemistry
do not necessarily suce for biology. Many, if not most, working biologists (and cognitive scientists) implicitly retain at least a minimal commitment to vitalism: they acknowledge that physicochemical laws should
suce for biology, but suppose that such laws are not adequate in their
current form, and must be enriched by further laws whose predicates are
dierent from those of inert physics and chemistry.
It is not evident how complete elimination of teleological expressions
(concepts dened functionally) from biological theory can be pursued
without forsaking a powerful and fruitful conceptual scheme for physiology, morphology, disease, and evolution. In cognitive science, a belief
that biological systems, such as the mind/brain, are not wholly reducible
to electronic circuitry, like computers, is a pervasive attitude that implicitly drives considerable polemic, but also much creative theorizing. Even
if this sort of vitalism represents a lingering folk belief that science may
23
ultimately seek to discard, it remains an important and perhaps indispensable cognitive heuristic for regulating scientic inquiry.
3. In addition to the spontaneous division of local ora and fauna into
essence-based species, such groups, as Darwin (1859, 431) noted, have
from the remotest period in . . . history . . . been classed in groups under
groups. The structure of these hierarchically included groups, such as
white oak/oak/tree or mountain robin/robin/bird, is referred to as
folkbiological taxonomy. Especially in the case of animals, these nonoverlapping taxonomic structures can often be scientically interpreted in
terms of speciation (related species descended from a common ancestor
by splitting o from a lineage).
In all societies that have been studied in depth, folkbiological groups,
or taxa, are organized into hierarchically organized ranks. Most folkbiological systems have between three and six ranks (Berlin 1992). Taxa of
the same rank are mutually exclusive and tend to display similar linguistic, biological, and psychological characteristics. Ranks and taxa, whether
in folkbiological or scientic classication, are of dierent logical orders,
and confounding them is a category mistake. Biological ranks are secondorder classes of groups (e.g., species, family, kingdom) whose elements
are rst-order groups (e.g., lion, feline, animal). Folkbiological ranks
vary little across cultures as a function of theories or belief systems
(Malt 1995). Ranks are intended to represent fundamentally dierent
levels of reality, not convenience.
Generalizations across taxa of the same rank thus dier in logical type
from generalizations that apply to this or that taxon. Termite, pig, and
lemon tree are not related to one another by a simple class inclusion
under a common hierarchical node, but by dint of their common rank
in this case the level of generic species. A system of rank is not simply a
hierarchy. Hierarchiesthat is, structures of inclusive classesare common to many cognitive domains, including the domain of artifacts. For
example, chair often falls under furniture but not vehicle, and car
falls under vehicle but not furniture. But there is no ranked system of
artifacts: no inferential link, or inductive framework, spans both chair
and car, or furniture and vehicle, by dint of a common rank, such as
the artifact species or the artifact family (see Coley et al. 2004 for experimental evidence that artifacts are treated dierently from natural kinds).
Modern systematicsthe branch of biology that concerns scientic
taxonomyis currently in the process of divesting itself of ranks in favor
of unranked phylogenetic lineages (clades), although the process is far
24
Chapter 2
from complete. Ever since Darwin, biology no longer recognizes a principled ontological distinction between species and variety, genus and
species, family and genus, and so forth. Nevertheless, contemporary systematists and other biologists continue to make use of ranked taxonomic
hierarchies as a research heuristic. By tabulating the ranges of extant and
extinct genera, families, classes, and so on, systematists can provide a
usable compendium of changing diversity throughout the history of life.
For example, by looking at just numbers of families, it is possible to ascertain that insects form a more diverse group than tetrapods (i.e., terrestrial vertebrates, including amphibians, birds, mammals, and reptiles). By
calculating whether the taxonomic diversity in one group varies over time
as a function of the taxonomic diversity in another group, evidence can
be garnered for or against the evolutionary interdependence of the two
groups. Some comparisons of the relative numbers of families of insects
and owering plants, reveal the surprising fact that insects were just as
taxonomically diverse before the emergence of owering plants as after.
Consequently, evolutionary eects of plant evolution on the adaptive
radiation of insects are probably less profound than previously thought
(Labandeira and Sepkoski 1993). The heuristic value of (scientically
elaborated) folk-based strategies for inquiry is compelling, despite evolutionary theorists being well aware that no true distinctions exist between various taxonomic levels.
4. Biological taxonomies not only organize and summarize biological information, they also provide a powerful inductive framework for making
systematic inferences about the likely distribution of organic and ecological properties among organisms. In modern systematics, this strategy
receives its strongest expression in the fundamental principle of systematic induction (Warburton 1967; Bock 1973). On this principle, given a
property found among members of any two species, the best initial hypothesis is that the property is also present among all species that are
included in the smallest higher-order taxon containing the original pair
of species. For example, nding that the bacteria Escherichia coli share a
hitherto unknown property with robins, a biologist would be justied in
testing the hypothesis that all organisms share the property. This is because E. coli link up with robins only at the highest level of taxonomy,
which includes all organisms. This or any general-purpose system of taxonomic inference for biological kinds is grounded in a universal belief
that the world naturally divides into the limited causal varieties we commonly know as (generic) species.
25
These four principles provide the backbone and background for studying the role of culture and experience in cognizing nature. That is, they
suggest candidates for universals as well as variations that may derive
from limited contact with plants and animals or from dierent cultural
lenses for perceiving biological kinds. In the next section we will see how
these principles have inuenced the development of scientic taxonomy.
2.3
Historical Developments
26
Chapter 2
6,000 known species to 600 genera). The place of a new species in the natural order of genera would be initially determined in either of two ways:
(1) by empirical intuition, that is, readily visible morphological agreement
with a European representative or some other preferred type species of
the genus, or (2) by intellectual intuition, that is, analytic agreement with
the generic fructication (fruit and ower) according to the number, topological disposition, geometrical conguration, and magnitude of its constituent elements. Within this Cartesian framework, the one criterion
would be ultimately commensurate with the other, allowing a mathematical reduction of the new species to its associated type by reason of their
common fructication. In this way, the customary native knowledge of
the folk naturalist would be rationally extended to a worldwide scale.
Such was the aim of Carolus Linnaeuss (1735) natural system.
Under John Lockes inuence, the English naturalist John Ray (1703)
questioned whether fructication characters encoded the essential order
of plant life. Analytic convenience might justify reliance on readily visible
parts of the fruit and ower as a classicatory strategy, but there was no
guarantee such analytic characters could be arranged into a preset combinatory system. In the case of animals, reduction of visible parts to computable characters proved unwarranted.
The geometrical rate of exploration and discovery further undermined
the taxonomic priority of the genus. As awareness of new forms increased
another order of magnitude, the family concept became the new basis for
taxonomy. The family was itself rooted in local groupings that native folk
implicitly recognize but seldom name, such as felines, equids, legumes,
and umbellifers. The ancients called these eide anonyma or genera innominata. The local series of such groupings does not fully partition a local
environment, but is riddled with gaps. A strategy emerged for closing the
gaps: looking to other environments to complete local gaps, naturalists
sought to discern a worldwide series that would cover the lacunae in any
and all environments. This would reduce the ever-increasing number of
species and genera to a mnemonically manageable set of basic, family
plans that were still perceptually distinguishable. Linnaeus (1751) dubbed
this strategy the natural method for completing family fragments.
French Enlightenment naturalists elaborated the natural method,
favoring empiricism over rationalism. Michel Adanson (1763) introduced
the idea of classication by family resemblances (air de famille) for
completing a worldwide family series. Antoine-Laurent Jussieu (1789)
reduced the thousands of genera proposed since Tournefort to exactly
100 families, but acknowledged this number reected convenience rather
27
28
Chapter 2
plants of [ecological] associations. . . . Confronted with such a situation, the botanist will nd that his diculties vanish as if by magic if he undertakes to learn the
ora as the natives know it, using their plant names, their criteria for identication (which frequently neglect the fruiting parts entirely), and their terms for habitats and types of land.
As Linnaeus needed the life-form tree and its commons species to actually do his work, so did Darwin ([1872] 1883, 353354) need the life-form
bird and its common species:
[In the Galapagos Islands] There are twenty-six land birds; of these twenty-one
or perhaps twenty-three are ranked a distinct species, and would commonly be
assumed to have been here created; yet the close [family] anity of most of these
birds to American species is manifest in every character, in their habits, gestures,
and tones of voice. So it is with other animals, and with a large proportion of
plants. . . . Facts such as these, admit of no sort of explanation on the ordinary
view of creation.
From a strictly cosmic viewpoint, the title of his great work, On the
Origins of Species, is ironic and misleadingmuch as if Copernicus had
titled his attack on the geocentric universe, On the Origins of Sunrise. Of
course, in order to attain that cosmic understanding, Darwin could no
more dispense with thinking about common species than Copernicus
could avoid thinking about the sunrise (Wallace [1889] 1901, 12). In
fact, not just species, but all levels of universal folktaxonomy served as
indispensable landmarks for Darwins awareness of the evolving pathways of diversity: from the folkspecics and varietals whose variation
humans had learned to manipulate, to intermediate-level families, and
life-form classes, such as bird, within which the godlier processes of natural selection might be discerned.
So far we have been discussing folktaxonomy at a fairly abstract level.
We now turn to a specic example that we examine in considerable detail,
taxonomy among the Itza 0 Maya of Guatemala.
2.4 ItzaO Maya Folktaxonomy
29
The most general rank in any folkbiological taxonomy is the folk kingdom,3 that is, plant or animal. Such taxa are not always explicitly named,
and represent the most fundamental divisions of the (nonhuman) biological world. These divisions correspond to the notion of ontological category in philosophy (Donnellan 1971) and psychology (Keil 1979). From
an early age, it appears, humans cannot help but conceive of any object
they see in the world as either being or not being an animal, and there is
evidence for an early distinction between plants and nonliving things
(Gelman and Wellman 1991; Keil 1995; Hickling and Gelman 1995;
Hatano and Inagaki 1996). Conceiving of an object as a plant or animal
seems to carry with it certain presumptions that are not applied to objects
thought of as belonging to other ontological categories, like the categories
of person, substance, or artifact.4
The next rank down is that of life form.5 The majority of taxa of lesser
rank fall under one or another life form. Most life-form taxa are named
by lexically unanalyzable names (primary lexemes), and have further
named subdivisions, such as tree and bird. Biologically, members of a single life form are diverse. Psychologically, members of a life form share a
30
Chapter 2
31
This contrasts with the other plant and life-form categories, which seem
to have mutually dened ecological roles (see Atran 1990; Berlin 1992):
birds and trees in the air (ik 0 ) and upper forest tier; mammals and herbs
on the ground (lu 0 um) in the forest understory; vines in the connecting
middle (tan-chumuk) tiers; grasses in the open lands (chak 0 an); sh
in the water ( ja 0 ). To be sure, the boundaries between these adaptive
zones are permeable by members of each life form; however, each life
forms has its respective habitat, or home (otoch). Accordingly, because
the chicken (aj-kax) has its home exclusively on the ground, and cannot
live in the air like other birds, it is not a bird, nor is it included under
any of the other life forms (although for Tzeltal Maya the chicken is the
prototypical bird; Hunn 1977).
For the mejenb 0 a 0 al@che 0 , whose morphologies and ecological proclivities are very distant from humans and other vertebrates, correspondence of folk to modern systematics blurs as one descends the ranks of
the scientic ladder, and violations of scientic taxonomy tend to be
more pronounced. Still, in this respect as in others, Itza 0 taxonomy diers
little from that of any other folkbiological system, such as that which
initially gave rise to systematics, including evolutionary systematics. For
Linnaeus (1751, sec. 153), a natural system is rooted in a natural instinct
[that] teaches us to know rst objects closest to us, and at length the
smallest ones: for example, Man, Quadrupeds, Birds, Fish, Insects, Mites,
or rst the large Plants, last the smallest mosses.
Generic Species
As we noted before, the core of any folk taxonomy is rank of generic species, which contains by far the most numerous taxa in any folkbiological
system. Most cultures have a set of life forms, but all cultures have a set
of generic species. Ethnobiologists who otherwise dier in their views of
folktaxonomy tend to agree that this level best captures discontinuities in
nature and provides the fundamental constituents in all systems of folkbiological categorization, reasoning, and use (Bulmer 1974; Hunn 1982;
Morris 1996; Descola 1996; Ellen 1999).
People in all societies studiedand thus likely in all cultures
spontaneously partition the ontological categories animal and plant into
generic species in a virtually exhaustive manner. Virtually exhaustive
means that when an organism is encountered that is not readily identiable as belonging to a named generic species, it is still expected to belong
to one. The organism is assimilated to one of the named taxa it resembles
(Berlin 1999). This partitioning of ontological categories seems to be part
32
Chapter 2
33
34
Chapter 2
primary lexemes, like winesap (a kind of apple tree) and tabby (a kind of
cat). Foreign organisms suddenly introduced into a local environment are
often initially assimilated to generic species as folk specics. For example,
the Lowland Maya originally labeled the Spanish pig village peccary,
just as they termed wheat Castillian maize. Similarly, the Spanish referred to the indigenous pacas and agoutis as bastard hares, just as
they denoted the Maya breadnut tree Indian g (Beltran [1742] 1859).
Over time, as introduced species acquire their own distinctive role in the
local environment, they tend to assume generic-species status and, as with
most other generic species, are labeled by a single lexeme (e.g., corn in
American English now refers exclusively to maize). Thus, the original
Lowland Maya word for the peccary, k 0 ek 0 en, now refers exclusivey to
the introduced pig, whereas the native peccary is obligatorily marked in
the composite expression k 0 ek 0 en()che 0 forest k 0 ek 0 en.
The subordinate ranks of folkspecic and varietal correspond to ranges
of perceptible natural variation that humans are most apt to appropriate
and manipulate as a function of their cultural interests. Partitioning
into subordinate taxa usually occurs as a set of two or more taxa that
lexically contrast along some readily perceptible dimension (color, size,
and so on); however, such contrast sets often involve cultural distinctions that language and perception alone do not suce to explain (Hunn
1982). An example is the Itza 0 Maya contrast between red mahogany
(chak[]chak-al@te 0 ) and white mahogany (sak[]chak-al@te 0 ). Red
mahogany actually appears to be no redder than white mahogany. Rather,
red mahogany is preferred for its beauty because it has a deeper, darker
woodgrain than white mahogany. But why red as opposed to white,
rather than simply dark as opposed to light?
A majority of Itza 0 folkspecics reect color contrasts, and the most
habitual contrast is between chak and sak (Atran et al. 2004), despite the
fact that distinctions involving green, yellow, or black may be no
less obvious to the naked eye. One interpretation is that use of contrasting
color specics, which almost invariably involve just the ve primary
colors, is related to the overriding importance of these colors in Maya
cosmology (for Lacandon [Lakantun] see Bruce 1968; for Yukatek see
Barrera Marn, Barrera Vasquez, and Lopez Franco 1976). In this ancient cosmology, the red east is the true direction of rain and the good
life, whereas the white north is the false direction of cold and deception.
This is not to deny that color contrasts generally signal perceptible distinctions among folkspecics. It merely suggests that color perception
35
36
Chapter 2
intermediates are usually restricted to locally occurring fragments of biological orders, families, or genera. Examples include Araneida (tarantulas
and other spiders), Anura (frogs and toads), Psittacidae (parrots and
macaws), Dasypractidae (agoutis and pacas), Meliaceae (mahogany and
tropical cedars), and Annona (custard apples).
Insofar as they reect a cognitively biased, phenomenal appreciation of
the surrounding environment, taxonomies help to set the constraints on
life that make a culture possible. It is little wonder, then, that folkbiological taxonomies tend to be among the most stable, widely distributed, and
conservative cognitive structures in any culture. Once set into place, such
a structure would likely survive even catastrophic historical upheaval to a
clearly recognizable degree. Ancient and contemporary Maya societies
would be no exception. Even with the social order and cosmological
system sundered, the folkbiological structure would persist as a cognitive
basis for cultural survival under three conditions. First, there must be signicant biological continuity in the ecological distribution of species. Second, there must be signicant linguistic continuity with the dialect that
rst encoded the knowledge. Third, there must be a sustained interaction
between people and living kinds where knowledge of the various species
matters.
As we will see, Itza 0 folkbiology (and that of many other indigenous
small-scale societies) fullls all three conditions, whereas the folkbiology
of majority-culture Americans (and that of many other urbanized societies) fails to meet the third condition. The consequences of this failure
for knowledge of the biological world and action toward it (environmental
management) are dramatic. Lets take a look at some historical evidence
on loss of knowledge or at least loss of cultural support for learning about
nature.
2.5 Loss of Knowledge and Familiarity with Nature
37
that the nonnative eucalyptus is not conducive to maintaining biodiversity in the face of competition for scarce water. Likewise, few residents
of Chicago are able to identify a buckthorn, much less comprehend that
a re can selectively weed out invasive buckthorns without aecting bur
oaks and other native prairie tree species.
It is hard to escape the impression that, on an individual and cultural
level, knowledge about living kinds is diminishing. As we mentioned earlier, anthropologists studying traditional societies often note with concern
the loss of indigenous language and a lessening of knowledge about the
natural world (e.g., Diamond and Bishop 1999; Nabhan and St. Antoine
1993; Wester and Yongvanit 1995). In technologically oriented cultures,
contact with biological kinds may be so minimal that researchers can
demonstrate signicant dierences in childrens biological reasoning as
a function of whether they do or do not have goldsh as pets (Inagaki
1990; Hatano and Inagaki 1987).
A survey we conducted at Northwestern University provides some index of what undergraduates know about one domain of biology, namely
trees. We provided the names of eighty trees and asked the students to
circle the trees they had heard of before, regardless of whether they knew
anything about them. More than 90 percent said they had heard of birch,
cedar, chestnut, g, hickory, maple, oak, pine, and spruce. But fewer than
half indicated any familiarity with alder, buckeye, catalpa, hackberry,
hawthorn, honey locust, horse chestnut, larch, linden, mountain ash,
sweet gum, and tulip treeall of which are common to the Evanston
area where Northwestern University is located. Of course, these observations by themselves do not implicate a loss of knowledge. It may be that
Northwestern undergraduates from a hundred years ago would have
proved equally unfamiliar with biological kinds. Nevertheless, such low
levels of knowledge are consistent with the possibility that knowledge
about trees is declining.
The Devolution Hypothesis
38
Chapter 2
and values. We will refer to this kind of exposure as cultural support. The
idea of cultural support has to do with the degree to which a society promotes a particular area of knowledge. It does not, then, have to do with
whether there are specialists who know or care about particular kinds.
Rather, it has to do with the extent to which people focus on a domain
of knowledge in their everyday interactions. For example, to what extent
do parents call childrens attention to plants and animals, and when they
do so, is their reference to robins, trout, and maples or to birds, sh, and
trees? Declines in cultural support, like declines in exposure to the natural
world, could lead to devolution.
In the remainder of this section, we summarize our work on the devolution hypothesis with respect to the life-form trees (Wol, Medin, and
Pankratz 1999). Trees are of special interest because they could represent
a particularly strong test of the devolution hypothesis. In terms of contact
with the natural world, we may not expect devolution with respect to
trees at all. While people in urban environments may have only limited
exposure to all but a few mammals (e.g., cats, dogs, squirrels), they are
likely to have seen many dierent kinds of trees. And trees, because of
their size, are not likely to be ignored. As argued by Hunn (1999), size is
a key factor in determining which natural kinds in a culture attract attention and get named. If the prerequisites for conceptual organization consist solely of an inherent curiosity about living kinds and a perceptual
system tuned to discontinuities in nature (Berlin 1992), then even urbanized cultures should show an appreciation for dierent kinds of trees. On
the other hand, it is possible, despite continued direct exposure to trees,
that knowledge about them has devolved because cultural support for
trees has declined.
Measuring Cultural Support
39
that this measure of cultural support is likely to be a conservative measure of what people may know. An author might write about noticing
cottonwoods along a riverbank without being able to pick a cottonwood
out of a biological lineup. The use of writing as a measure of what people
know is therefore likely to overestimate the knowledge of an average
citizen, hence underestimate devolution. By the same token, if changes
are found, they are most likely to be historically signicant.
Because our interest is in a longer time
span than U.S. written history aords (in terms of databases we might access), we selected a database from England for study: the OED, a historical dictionary. We chose the OED for a variety of reasons. The OED
seeks to capture the evolution of all words in the English language except
those that became obsolete before 1150 or are intelligible to only the specialist. The rst edition was published in 1933 after nearly seven decades
of work. The second edition, the OED2, was published in 1989. It combines the original edition, four supplemental volumes published after
1933, and results from a fourth major reading program.
The dictionary contains approximately 616,500 word forms (Berg 1993;
Murray 1989). Denitions for these words are illustrated with quotations
from each century of use, with extra quotations provided for signicant
changes in meaning. The quotations were drawn from a wide range of
books, with special emphasis on great literary and scientic works, but
also among other things, books of foreign travel, letters of foreign correspondents, magazines, and diaries. The total number of quotations in the
OED2, roughly 2.5 million, was drawn from a sample of between 5 and 6
million quotations. Given the breadth of the inquiry, we have no reason
to expect that the quotations represent a biased sample with respect to the
questions we aim to address. The sample may well be biased in terms of
reecting interests, values, and accessibility, but these sorts of biases are
more or less orthogonal to our focus.
Recently, the entire twelve-volume set was retyped into a special computer database format allowing for online searching of all denitions and
quotations. The OED online corpus may be searched for any key words
(e.g., tree, maple tree, maple, and so on) and search codes may be written
such that the date, source, and full quotation context will be returned.
General Predictions
40
Chapter 2
with a given historical period (we used 100-year blocks for our analyses),
and (2) the number of sources (kinds of publications from which the
quotes are drawn) relative to the total number of sources associated with
a given period. Our rst analysis examines the general prediction that if
knowledge of trees is devolving, there should be an overall drop in the
number of quotes and number of sources across time. A second major
analysis examines more specic hypotheses concerning the relative usage
of tree terms at dierent levels of taxonomic organization.
Of course, there may be historical periods of time where cultural support for biological knowledge is increasing (evolution rather than devolution). The predictions here would be more or less reversed. As we will see,
our analyses suggest both periods of evolution and devolution. Before
turning to specic procedures, we rst state our assumptions about levels
of specicity and identify potential problems that may arise with analyses
such as ours.
Methodological Issues
There are ve general concerns associated with using texts to assess change across time. One problem involves changes in
spelling and in naming. For example, our search revealed twenty dierent
spellings of oak and twenty-ve dierent spellings of tree. Spelling consistency only became fairly uniform in the nineteenth century. Obviously,
one needs to search the corpus for each of the alternative spellings.
Likewise, some trees have multiple common namesfor instance, in
England another name for linden is whitewood. The same prescription
holds here.
A second concern is that the results may be aected by the particular
meaning of the term being invoked in a quotation. For example, the
term pine can be used to refer not only to a particular kind of tree, but
also a particular kind of wood (e.g., pine oor), location (e.g., pine
grove), activity (e.g., pine away), or proper name (e.g., the cleaning product, Pine Sol). In the following analyses, only direct references to particular kinds of trees (the rst use) were included because it is for these uses
that the devolution hypothesis makes the clearest predictions.
A third concern is that the sources for quotes may change across time
in a systematically biased manner. For instance, during the age of exploration and colonization, new publications appeared (e.g., the Australian
Journal ) devoted not to life in England, but rather to life in the British
colonies. These often include descriptions of the (novel) ora and fauna.
Threats to Validity
41
Other Issues
The purpose of this rst analysis was to test the main prediction of the
devolution hypothesis: If knowledge about trees is declining, there should
be an overall drop in the use of tree terms.
42
Chapter 2
The process of preparing the quotes for analysis had three main
phases: (1) abstracting the entries containing quotations, (2) coding the
entries, and (3) correcting for uneven sampling in the OED. These three
phases are discussed in turn.
Method
Abstracting Entries
Results Our search for tree terms generated a total of 22,319 quotations.
An automatic coding of each quotations source was performed using a
43
Figure 2.1
Proportion of quotations and sources in the OED referring to trees along with
associated 95 percent condence intervals.
program that checked lists of 134 foreign and 45 technical sources. The
resulting 15,146 quotations with sources not present on these lists were
roughly equivalent to 900 pages of text and were further analyzed by
hand according to the criteria described in the method section. The resulting 6,548 quotations that both made direct reference to trees and
came from folk sources were roughly 29 percent of the original set of
quotations.
The ndings provided strong support for the main prediction of the devolution hypothesis: Cultural support for trees, as measured by the relative number of quotations and sources in the OED, declined markedly in
the last century. As described above, tree counts were analyzed relative to
the estimated number of folk quotations and sources in the OED in order
to eliminate dierences due to sampling. Figure 2.1 shows the resulting
proportions for each period of time. The condence intervals in gure
2.1 represent ranges having a 95 percent probability of covering the true
population values, assuming a binomial distribution.
An examination of gure 2.1 shows that the proportions for quotations
and sources were fairly constant through the sixteenth, seventeenth, and
eighteenth centuries. In the nineteenth century, the relative number of
quotations and sources increased, suggesting that knowledge of tree terms
evolved during this period. However, the gains of the nineteenth century
were completely lost in the twentieth century, which witnessed a striking
44
Chapter 2
decline in both quotations and sources using tree terms. Note that the
start of the decline corresponds closely with the start of the industrial revolution. The condence intervals indicate that the evolution occurring
in the nineteenth century and the devolution occurring in the twentieth
century are signicant. The condence intervals also indicate that the
twentieth-century decline was so great that writing about trees is less extensive now than in any other time in the history of the English language.
The only dierence between the two measures seems to occur between
the sixteenth and seventeenth centuries: sources indicate evolution while
quotations do not. This dierence does not change the important conclusion that we can be condent that the observed changes in quotations are
not due to an overrepresentation from a particular kind or set of sources.
In sum, the ndings are perfectly consistent with the idea that there have
been periods of evolution and, more recently, devolution in knowledge
about trees.
Analysis 2: Examining Tree Terms at Different Levels of Specificity
The same set of quotations used in Analysis 1 was used for an analysis of
the specicity or level of quotes. One of the main goals in this analysis
was to better understand the observed decline in tree terms in the twentieth century. However, a closer examination of the quotations could also
be used to provide further insight into the apparent lack of change existing between the sixteenth and eighteenth centuries and the observed evolution of tree terms in the nineteenth century.
Three levels of organization were coded (gure 2.2). The lifeform level (Life form) was indicated by use of the word tree, or one of
its twenty-four other spellings. The folk-generic level (Generic) was
indicated by quotations containing one of the twenty-two prechosen tree
terms listed in Analysis 1. Quotations demonstrating the folkspecic level
contained one or another of the twenty-two prechosen folk-generic tree
terms.
Method
45
Figure 2.2
Proportion of quotations in the OED for dierent levels of specicity along with
associated 95 percent condence intervals (after Wol et al. 1999). Note that before about 1700 folk generic terms (e.g. oak, bear) referred mostly to monogeneric European species, whereas after about 1700 generic terms often referred
to polytypic species built around a European type.
46
Chapter 2
we found that the pattern shown in gure 2.2 holds whether the tree term
was the topic of the quotation or was incidental to it.
The pattern of frequency counts during the twentieth century is most
consistent with devolution. Crucially, the twentieth century is the only
century where frequency counts for all levels of organization declined.
Thus, in contrast to the shift-in-knowledge hypothesis, an overall drop in
tree terms cannot be explained as a drop in the life-form level alone,
which masked increases at more specic levels of organization.
One potential challenge to the devolution
hypothesis is that the observed decline in tree terms in the twentieth century may be a statistical artifact of the OED. Assuming the twentieth century experienced an enormous explosion in new categories, it is certainly
possible that talk about any one category may have been diluted. Thus,
the apparent decline in the twentieth century may not be due to devolution, but rather to decreased talk about any one thing because there were
more things to talk about. This possibility is relatively easy to check. If
the twentieth-century decline is due to dilution, similar rates of decline
should be observed for categories other than tree categories. If, however,
the decline is due to changes in knowledge, rates of decline are likely to
vary widely between the categories. To test this possibility, life-form-level
terms (or their equivalent) from three other domains were analyzed using
the same criteria as used in Analyses 1 and 2. The specic categories analyzed were sh, weapon, and bird. The ndings provide further support
for the devolution hypothesis. In contrast to the dilution hypothesis, not
all the categories declined during the twentieth century. Specically, quotations referring to the category sh steadily increased from the sixteenth
century until the present. This may partially be a function of the fact that
sh also appears in food contexts. Quotations containing the category
weapon slowly declined during the sixteenth to nineteenth centuries and
then asymptote during the twentieth century. Changes in the category
bird mirrored those of the category tree, but not as dramatically. In sum,
because declines in the twentieth century are not inevitable, we can be
more condent that the observed declines in tree terms are due to changes
in knowledge and not dilution.
The results from this research support the claim that knowledge about trees evolved during the sixteenth to nineteenth centuries and
devolved during the twentieth century. We showed that the twentieth century was marked not only by a major decline in frequency in tree terms
overall (Analysis 1), but at all levels of specicity (Analysis 2). These
Implications
47
twentieth-century declines cannot be explained as simply due to an explosion of categories diluting talk about any particular kind of category.
Diluting would predict that all categories should decline, but as indicated
by the categories sh and weapon, decline is not inevitable.
When a domain devolves, does it reverse the order of its evolution? The answer to this question appears to be
a cautious no. When a domain evolves, knowledge of the domain motivates the creation of ever more precise category labels. When a domain
dies, it may be that the knowledge of the associated concepts declines
faster than knowledge of specic terms. Thus, the language may preserve
certain distinctions beyond the time these distinctions are still understood.
It is as if knowledge builds up a terminological structure in the language,
but when knowledge declines, the structure, like an abandoned building,
may remain for a while. In chapter 4 we will describe evidence consistent
with this suggestion.
2.6
Conclusions
We have given a great deal of background and perhaps tested your patience. This chapter has provided an introduction to folk and scientic
taxonomies, elaborating in detail on the Itza 0 Maya taxonomic system.
The historical review reveals parallel developmentsfolk conceptions
nd their way into science because there are certain ways to think and
organize nature that are natural. Finally, cultural support for interest in
biological kinds appears to be eroding in technologically oriented cultures. This observation sets the stage for many of our ndings in chapter
4. But before turning to these ndings, we outline our methodology in the
next chapter.
Mesoamerican Populations
A key focus of our work concerns three cultural groups in the same municipality in Guatemalas Department of El Peten: native Itza 0 Maya,
Spanish-speaking immigrant Ladinos, and immigrant Q 0 eqchi 0 Maya.
Itza 0 and Q 0 eqchi 0 were each circumscribed by entirely overlapping and
perfectly redundant criteria of proximity of residence, ethnic selfidentication, and a multigenerational history of pervasive family interconnections. The Q 0 eqchi 0 were also identied by their mother tongue.
The Ladino population was initially circumscribed by proximity of
residence, language (Spanish), ethnic self-identication, and lack of a
50
Chapter 3
The Itza 0 , who ruled the last independent Maya polity, were reduced to
corvee labor after their conquest in 1697 (Atran, Lois, and Ucan Ek 0
2004). San Jose was founded as one of a handful of reductions for concentrating remnants of the native Itza 0 population (and fragments of related groups). In 1960, the military government opened Peten (which
includes 35,000 km 2 , about one-third of Guatemalas territory) to immigration and colonization. In the following years, about half the forest
cover of Peten was cleared. Supported by a debt-for-nature swap, Guatemalas government set aside remaining forests north of 17 10 0 latitude as
a Maya Biosphere Reserve in 1990. The San Jose municipality now lies
within the Reserves ocial buer zone between that latitude and
Lake Peten Itza 0 to the south. Today San Jose has some 1,800 habitants,
about half of whom identify themselves as Itza 0 , although only older
adults speak the native tongue (a Lowland Mayan language related to
Yukatek, Mopan, and Lakantun).
Immigrant Ladinos
51
The hamlet of Corozal, also within the Municipality of San Jose, was settled at the same time by Q 0 eqchi 0 speakers, a Highland Maya group from
the Department of Alta Verapaz, just south of Peten. Q 0 eqchi 0 ltered in
as nuclear families, migrating in two waves that transplanted partial
Highland communities to Corozal: (1) directly from towns in the vicinity
of Coban (the capital of Alta Verapaz), and (2) indirectly from Alta Verapaz via the southern Peten town of San Luis (home to a mixed community of Q 0 eqchi 0 and Mopan Maya). Q 0 eqchi 0 immigration into Peten
began as early as the eighteenth century, though massive population displacement into Peten is recent. Although many of the nearly 400 Q 0 eqchi 0
of Corozal understand Spanish, few willingly converse in it. Q 0 eqchi 0 is
not mutually intelligible with Itza 0 . To help understand results with Lowland Q 0 eqchi 0 immigrants, we also studied a native Highland Q 0 eqchi 0
group.
Native Highland QOeqchiO
Studies with Highland Q 0 eqchi 0 from Aldea Paapa in the Coban region
of Alta Verapaz, Guatemala, focused on the issue of whether Q 0 eqchi 0
immigrants arrive in Peten with a cognitive model that is already impoverished with respect to knowledge of species relationships, or whether
they are simply unable to use richer Highland models because these are
inappropriate to Lowland ecology.
Native Lacandon (Lakantun) Maya
Studies with the Lacandon Maya, whose agroforestry practices closely resemble those of the Itza 0 , were mainly concerned with intergenerational
change among adult generations living in the community of Mensabak
in Chiapas, Mexico.
Yukatek Maya Children
52
Chapter 3
It has also been helpful to collect data from a number of U.S. populations. When we began to study folkbiology with the standard undergraduate populations it soon became clear that the typical college student
knows very little about plants and animals. Consequently we sought out
a variety of other U.S. populations. There is also evidence that urban and
suburban children may have relatively impoverished experience with nature (compare Stross 1973 on Maya childrens knowledge and naming
of plants with Dougherty 1978 on Berkeley children) and, therefore, our
developmental studies also involved several dierent groups.
Undergraduates
This category includes diverse groups with distinct kinds of expertise: bird
watchers, shing experts, parks maintenance workers, landscape architects, and professional taxonomists. They typically had at least twenty
years experience in their occupation or avocation.
Native American Menominee
Adults The Menominee (Wild Rice People) are the oldest continuous
residents of Wisconsin. There are 4,000 to 5,000 Menominee living on
tribal lands in and around three small communities. As in the past, the
reservation is heavily forested. Hunting and shing are important activities for most adult males and for many females. Education is an important value and there are two tribal colleges on the reservation.
Children
53
tribal school in Neopit, Wisconsin. Keshena and Neopit are the two
largest towns on the Menominee reservation; Menominee Tribal
Enterprisesresponsible for managing the forestis in Neopit and the
main tribal oces are in Keshena. Although Menominee children tend
to know some Menominee words, especially those for clan animals, they
are basically monolingual English speakers (though there are vigorous
eorts underway to restore the language, and the tribal school includes
classes in the Menominee language).
Rural Majority Culture
Adults
Children
Urban Children
The urban children came either from Boston or Chicago. The Boston,
Massachusetts, school is located in East Boston and serves a middle-class
community. The Chicago school is a magnet school and serves a diverse
population, most of whom are middle-class. For some comparisons we
interviewed parents of these children.
3.3
54
Chapter 3
There is no theoretically neutral way to dene culture. We have just suggested that the idea that culture is whatever is left when all potentially
55
Cultural values, beliefs, and behaviors are not static but relentlessly develop, dissolve, merge, and mutate. Nonetheless, it seems sensible to
look for sharp contrasts by means of selecting subpopulations that have
retained more traditional knowledge. These considerations lead one to
employ sampling techniques most likely to reveal cultural dierences
rather than focusing on estimating population parameters. Consider
again the Lopez et al. studies with the Itza 0 Maya. Younger Itza 0 might
have notions of biology that dier from those of Itza 0 elders, dierences
that reect assimilation to Western culture. Thus a random sample
56
Chapter 3
may tend to hide rather than reveal cultural dierences. Instead of randomly selecting participants, Lopez et al. restricted their sample to Itza 0 speaking Maya as the best representatives of Itza 0 culture. It is not that
there was some pure Itza 0 culture in the past that nowadays is being
degradedcultural change is a constant. Itza 0 cultural life is a rich blend
of ideas and habits stemming from dierent inputs, including a great deal
of Spanish inuence. A random sample is only appropriate when one
wants to make claims about population parameters, something that we
believe is rarely relevant in cultural comparisons.
Cross-Cultural but Culturally Sensitive Methodogy
57
cultures the world over (Lopez et al. 1997). By contrast, there was no signicant dierence in the performance of American students asked to sort
items that go together by nature or as being most similar.
Similar sorts of analyses and pretesting accompanied preparation of
all of our instructions. One advantage of tailoring instructions to a variety of nonstandard populations is that they can be further applied to
other populations with greater ease and condence than if they had been
simply translated from instructions given to undergraduates or other
groups aliated with large research universities and urban environments
in the United States. Moreover, we have found that the instructions so
pretested usually can be successfully reapplied to standard populations.
We turn now to an important methodological tool, the cultural consensus
model.
3.5
58
Chapter 3
59
60
Chapter 3
between each pair of subjects is generated (as the product of their respective consensus parameters). Next, the expected agreement matrix is subtracted from the raw agreement matrix to yield a matrix of deviations
from expected agreement (see Hubert and Golledge 1981). If raw and residual agreement are signicantly associated, then a signicant portion
of residual agreement consists of deviations from the consensus. One can
then explore other factors (e.g., cultural subgroups, social network distance), which might predict or explain the residual agreement.
Another marker for reliable residual agreement is when an analysis
over two or more groups reveals systematic dierences in factors beyond
the rst. If two groups dier in their second factor scores, then withingroup agreement extends beyond the overall consensus. For example,
Medin et al. (1997) asked tree experts to sort local species of trees and
found a clear overall consensus, coupled with second factor scores correlating strongly with occupation (e.g., parks maintenance, taxonomist,
landscaper). Subsequent comparisons revealed systematic dierences in
the basis for sorting across groups.
Our method of modeling cultural consensus allows us to avoid synthetic interpretations of peoples thoughts and actions (as part of this or
that culture) and to describe emergent cultural patterns derived statistically from measurements of individual cognitions and behaviors. Rather
than merely assuming cultural consensus from statistical reliability, we
impose additional conditions that allow us to identify and demonstrate
patterns of consensus more precisely (e.g., aggregated folktaxonomies) so
as to better make independent predictions about a population (e.g., use of
taxonomy to generate biological inferences).
Once cultural dierences are found, we can proceed to ask a series of
more analytic questions about things like the following: (1) Are these
ideas spread by means of abstract models and inference strategies or is
the information conveyed in quite literal, concrete form? (2) Do factors
like income or occupation or density of social networks or a variety of
other input conditions moderate cultural dierences (either within or between groups)? Within the present framework the goal in studying variation is to have a theory about the distribution of ideas and ow of
information, not to isolate some (magical, reied) entity, culture.
3.6 Summary
61
64
Chapter 4
Our present knowledge of evolutionary mechanisms and history is generally too poor to generate causal explanations of cognition. Often, evolutionary accounts are mere consistency argumentsjust-so stories
that lack evidentiary standards for ruling out indenitely many contrary
evolutionary scenarios (Atran 2002). There have been more constrained
evolutionary accounts of higher-order cognitive functions specic enough
to motivate competing and informative research (e.g., Pinker and Bloom
1990; Cosmides and Tooby 1992). So far, however, these accounts may
do little more than retrodict ndings generated independently of any
evolutionary considerations (e.g., Hauser, Chomsky, and Fitch 2002;
Sperber, Cara, and Girotto 1995).1 At the same time, we hope to illustrate how evolutionary argument can be usefuleven if not necessary
to progress in the eld. A factor motivating our experiments, and our
interpretation of them, is evolutionary plausibility. We do not claim that
evolutionary arguments have explanatory value here, but they may have
important heuristic value.
Humans and their ancestors undoubtedly depended for their survival
on intimate interaction with plants and animals, which likely required
anticipatory knowledge of at least some plant and animal species. This
makes it likely (but not necessary) that adaptations for special dealings
with plants and animals evolved, and, further, that they evolved in a
manner somewhat independent of adaptations for dealings with other
people. For example, identication and categorization is dierent for
humans, on the one hand, and for animals and plants, on the other. There
are cognitive mechanisms primarily dedicated to tracking humans as
individuals, such as facial recognition (e.g., Carey and Diamond 1977;
Diamond and Carey 1986), syntactic and semantic structures of pronominalization and proper naming (Balogh, Swinney, and Tigue 1998;
Arnold et al. 2000), social game strategies (Axelrod 1985; Nowak and
Sigmund 1998), and so forth. For animals and plants, the default recognition strategies are focused at the collective, species level (individualization
of pets involves anthropomorphic extensions of person-identication
strategies). From an evolutionary vantage point, it hardly would matter
which member of a plant or animal species a person could eat or be eaten
65
by, but it would matter greatly who in particular a person could mate,
ght, or cooperate with (Eldredge 1986). One implication of this analysis
for folkbiological cognition is that results focused at the level of individuals do not necessarily carry over to studies at the level of species,
and vice versa.
4.2
Dierent cognitive scientists have oered alternative and sometimes conicting notions of modules, so we will take a few paragraphs to say what
66
Chapter 4
67
Table 4.1
Empirical and theoretical claims and the status of evidence bearing on them
Claim
Status of evidence
Basis for typicality ratings and typicality eects in reasoning knowledgedependent and undergraduates are
often the odd group out.
4.3
Categorization tasks are of independent theoretical interest and selfcontained, but they are also designed to provide the inferential framework for category-based reasoning. In this section we focus on models
for use of categories in inductive reasoning in general, and biological inference in particular. The empirical phenomena of interest are referred to
as typicality and diversity eects in reasoning. To set the stage for our discussion, we briey review one of the most inuential models of induction,
the similarity-coverage model (SCM) of Osherson et al. 1990.
An important function of taxonomic classication is enabling generalizations between categories. Osherson et al. (1990), building on previous
work by Rips (1975), identied a set of phenomena that characterize
category-based inferences in undergraduates, and formalized a model
that predicts the strength of those inferences. The model relies on the notion of similarity and similarity relations as a guide to induction.
68
Chapter 4
69
Of course to test the SCM model, one needs some measure of similarity. There are two problems with the idea of collecting pairwise similarity
judgments by asking informants to rate the degree of similarity on some
scale. The rst is that, for many informants, rating scales may not be
meaningful. The second is that the number of judgments required grows
geometrically with the number of stimuli. For example, with 44 items,
946 judgments would be needed.
A convenient alternative strategy is to use a sorting technique to derive
a metric of similarity. In the Lopez et al. study, which we will take up
shortly, participants were asked to sort local mammals into groups, to
put the animals that go together by nature into as many groups as you
want. Subsequent sorting into sub- and superordinate categories created
a hierarchical taxonomy for each participant, which were then combined
to create a group taxonomic hierarchy. The rationale for eliciting such
taxonomic hierarchies was to be able to indirectly, but automatically,
compute measures of similarity, typicality, and category coverage from a
single cognitive structure. This method also directly links categorization
(sorting) with reasoning. If the sorting reects idiosyncratic or taskspecic strategies, then the reasoning data should not be orderly. Thus,
in a sense, the sorting and reasoning data are mutually reinforcing.
To justify combining individual sorts into an aggregate cultural taxonomy, Lopez et al. rst applied the Romney, Batchelder, and Weller 1986
cultural consensus model to the informant-by-informant agreement matrix for both the Itza 0 and undergraduate sample. Both groups showed a
strong consensus. For undergraduates the similarity relationships were
largely organized along the dimensions of size and ferocity, as other
researchers have observed (e.g., Henley 1969). Indeed size alone accounted for more than 70 percent of the variance. A multidimensional
scaling of Itza 0 sorting results is shown in gure 4.1. The two-dimensional
solution is shown to be considerably less satisfactory and size accounted
for only 16 percent of the variance. In addition to morphological similarity, ecological considerations come into playboth the otter, a water
mammal, and the bat, an aerial mammal (considered to be a bird by the
Itza 0 ) are isolated from the other mammals. In short, Itza 0 sortings were
based on many factors and not just size.
With these results in hand, distance in the consensual group taxonomy
provides a key measure of similarity. This measure was then used to study
category-based inferencing.
Similarity predicts that the stronger inference should be the one where
the premise is closest to the conclusion, with closeness measured as the
70
Chapter 4
Figure 4.1
Multidimensional scaling of the Itza 0 taxonomy for local mammals. From Lopez
et al. 1997.
71
Table 4.2
(%)
M*
SD
American condition
Similarity
Typicality
Diversity
(92)
(94)
(96)
3.67
3.75
3.83
0.65
0.45
0.39
8.86**
13.40**
16.32**
Itza 0 condition
Similarity
Typicality
Diversity
(85)
(79)
(38)
3.42
3.17
1.50
0.79
1.03
0.91
6.19**
3.93**
1.92
Why do many experts and Itza 0 not show diversity? Consider, rst, the
Itza 0 . Their justications revealed that diseases did not function as blank
predicates for the diversity items but instead served as triggers for ecologically based explanations. In many cases, ecological considerations led
participants to conclude that the argument with more diverse premises
was actually weaker. For example, one Itza 0 woman favored the argument rat, pocket mouse / mammal over tapir, squirrel / mammal. She
argued that tapirs and squirrels are less likely to pass on the disease because they require an ecological agent (a bat biting them) to get the disease in the rst place, whereas rats and pocket mice are close enough
companions that they do not need an ecological agent (a bat biting
them) to get the disease.
Ecological considerations also led to diversity-based inductions in a few
cases. Thus, a dierent Itza 0 informant reasoned, to the contrary, that
rats and pocket mice live only where there is corn, sleep above ground,
and do not travel in parts of the forest where other animals may catch
their disease.
We did a follow-up study with the Itza 0 using palms as stimuli. Again
we did not observe diversity eects. For palms, although similar premises
were chosen more frequently than diverse premises, the dierence failed
to reach signicance. Nevertheless, Itza 0 preference for causally based
ecological reasoning was evident here as well. For example, one person
favored argument (v) over argument (vi), arguing that because the coconut and the royal palm are tall and treelike, their disease is more able to
spread to other palms:
72
Chapter 4
(v) coconut (kookoj ) & royal palm ( palmareaal ) / all palms (tulakal
uy-et 0 ok xa 0 an)
(vi) coconut (kookoj ) & basket whist (b 0 ayal ) / all palms
In this case, as in many others, size is indicative of the broader ecological coverage of the forests canopy. In other instances, ecological considerations again led to diversity-based inductions. For example, one
informant accepted (vi) as being stronger than (v) by saying: Dont you
see that the coconut is a big tree and the basket whist clings to it worse
than a vine, isnt that so? It can encounter the coconut, climb it and catch
the same disease the other has [and give it to the other palms]. In other
words, vinelike basket whists can help spread the disease of treelike coconuts to all other palms, whereas the treelike royal palm would presumably contribute little more to the spread of the disease than would the
coconut alone. In this example, as in others, the focus seems to be on
broader ecological coverage in terms of the vertical, or storied, relationships between forest species rather than in terms of horizontal relationships of broad spatial coverage.
In the absence of a theoryor at least presumption of a theoryof
causal unity underlying disparate species, there is no compelling reason
to consider a property discovered in two distant species as biologically intrinsic or essential to both. It may make as much or more sense to consider the counterintuitive presence of a property in dissimilar species as
the likely result of an extrinsic or ecologically accidental cause. For
Itza 0 , taxonomic distance can provide one indication of the extent to
which ecological agents are likely to be involved in predicting biological
properties that do not conform to surface relationships. This may account
for negative diversity on some tasks (Lopez et al. 1997). This does not
mean that Itza 0 fail to grasp or use a diversity principle. In justications,
Itza 0 clearly reject a context-free use of the diversity principle in favor of
context-sensitive reasoning about likely causal connections. In tasks
designed to assess risk-diversication strategies (e.g., sampling productivity from one forest plot or several), Itza 0 consistently showed an appreciation of the diversity principle in these other settings (Atran 1995; Lopez
et al. 1997).
More generally, what counts as a biological cause or property may
be somewhat dierent for folk, like the Itza 0 , who necessarily live in intimate awareness of their surroundings, and those, like American folk,
whose awareness is less intimate and necessary. For Itza 0 , awareness of
biological causes and properties may directly relate to ecology, whereas
73
74
Chapter 4
Interestingly, the tree experts did not show typicality eects. Their justications for typicality probes often appealed to family size, where
family refers not to scientic families but to generic species. This result
may indicate the psychological salience of the genus level.
To further test the generality of these ndings on typicality and diversity, we tested Itza 0 on yet other kinds and properties (e.g., has little
things inside), and we also tested other U.S. expert groups. Lets look
at one of these lines of research in further detail.
Triangulating with Birds
Bailenson et al. (2002) studied three populations categorizing and reasoning about birds. The populations were (1) Itza 0 Maya elders of Guatemala, (2) U.S. bird experts (bird watchers), and (3) U.S. novices recruited
through ads placed on campus at Northwestern University.
The experts were ten men and ten women (mean age 51 years) having either occupations or extensive experience related to birds. The average number of years spent watching and studying birds (termed birding
hereafter) was 22 years. On a 7-point continuous scale with 1 indicating
very little knowledge about birds and 7 indicating total expertise,
the mean self-reported rating for experts was 5.1. Most of them viewed
birding as an extremely involving hobby, often dedicating their vacation
time to traveling to places where they could nd birds that they had never
seen before. The novices were eight men and eight women (mean
age 21 years, range: 1840) who were recruited through the university
and paid for their participation. On the 7-point rating scale mentioned
above, the mean self-reported rating for our novices was 2.33. The Itza 0
informants were eight men and two women (mean age 66). All were
bilingual in Itza 0 and Spanish, although experimental instructions and
responses were in the Itza 0 language. It was assumed that all Maya elders
would be experts given their continuous and extensive experience with
forest plants and animals. All Itza 0 were well acquainted with the experimenters, and at relative ease in the session.
The stimulus materials were pictures of Chicago-area U.S. birds as well
as pictures of birds of lowland Guatemala. The idea was to see if the
experts responded dierently to local versus exotic species. Itza 0 can be
thought of as novices with respect to U.S. birds, but they have extensive
experience with birds that they may bring to bear with novel bird species.
Each set consisted of full-color illustrations of 104 bird species laminated onto index cards. For the U.S. set, illustrations were taken from
the Golden and National Geographic eld guides, books designed to aid
75
bird identication. The other set (Tikal Birds) was taken from the book
The Birds of Tikal (Smithe 1966). The specic selection of birds was based
on the inventory list carried out by the University of San Carlos (Guatemala) for the UN-sponsored Maya Biosphere Reserve.
The structure of the scientic taxonomy representing the U.S. bird set
was designed to correspond maximally with that representing the Tikal
bird set. The Tikal bird set consisted of thirty families and seventeen
orders, while the U.S. bird set consisted of thirty-three families and seventeen orders. One notable dierence was in the number of passerines
(songbirds) in the two sets. Although passerines are the numerically dominant group both in Chicagoland and Mayaland, they are somewhat
more prevalent in Chicagoland. There were some birds and taxonomic
groups that were common to both sets (eighteen shared orders, twelve
shared families, twelve shared genera, and ve shared species).
The rst study used the Lopez et al. sorting procedure. Again, we asked
participants to sort pictures of local and exotic birds into groups that go
together by nature. The main goal of this study was to compare withinand across-group patterns of sorting. In that regard there are two important questions: (1) Do people within a group agree suciently in their
sorting that it is sensible to claim that there is a consensual cultural or
group model? (2) Are the patterns of sorting reliably dierent across
groups? To address these questions we use the cultural consensus model
(CCM), and looked at patterns of residual agreement. If the groups dier,
then individuals within a group should agree with each other to a greater
extent than is predicted by the overall consensus analysis.
What should we predict concerning agreement across groups and
agreement with scientic taxonomy? Based on the work of Boster and his
associates, we might expect that novice sortings would correlate highly
with scientic taxonomy. Whether the two experts groups show a strong
correlation with science should hinge on whether they have specialized
goals. Given that the primary goal of birders is to identify birds, we see
no reason for expecting that their goals violate the structure of bird taxonomy. Both bird-identication and traditional taxonomic systems are
based on morphological similarities and dierences. If this analysis is correct, then both U.S. experts and U.S. novices should base their categorization of both U.S. and Mesoamerican birds on the natural or default
taxonomy and show good agreement with scientic taxonomy.
It was less clear what to predict the Itza 0 experts would do. On the one
hand, they do have specialized goals with respect to some birds (e.g.,
hunting them for their meat) but, on the other, our prior work suggested
76
Chapter 4
that they have rich ecological knowledge concerning relationships between particular kinds of birds and both plants and other animals. The
latter observation suggests a general-purpose (default) representation. If
the Itza 0 have both special-purpose and general-purpose representations,
then we would expect that the special-purpose representation should be
much more evident in their sorting of familiar Mesoamerican birds than
the unfamiliar U.S. birds.
All participants were tested individually. They were told that we were
interested in how they organized their knowledge about birds. First, we
showed them all 104 bird cards one at a time and asked them to name
them as specically as possible. Next, all 104 cards were placed in front
of the participant, for the sorting task. As in the Lopez et al. studies, initial sorts were followed by lumping and splitting to produce a hierarchical
taxonomy for each informant.
The naming data are useful in providing an independent index of expertise and relative familiarity with the two picture
sets. We scored each naming response on a 3-point scale, with a 3 representing an exact species match, a 2 representing a correct genus match,
and a 1 representing a match at order or higher (i.e., a bird response
was scored a 1). For the Itza 0 this measure is somewhat conservative in
that, unlike novices, they rarely said bird and instead often used intermediate categories such as esh-eating bird. The three groups named
all the birds from both stimulus sets except the Itza 0 , who only named
birds from the Tikal set. Experts were more accurate at naming U.S.
birds (M 2.55) than Tikal birds (M 1.66), but novices showed little
dierence (M 1.25 for U.S. birds versus 1.14 for Tikal birds). These
results established that the U.S. experts were more familiar with the U.S.
birds than the Tikal birds and that their naming skills were superior to
those of novices for both sets of birds. The Itza 0 averaged 1.92 for Tikal
birds and were less accurate at naming passerines (M 1.39) than other
birds (M 2.11). Notably, U.S. experts were equally good on passerines
and nonpasserines.
Each informants hierarchical sorting was used to derive a birdby-bird similarity (distance) matrix. The lowest level at which two given
birds go together in a folk taxonomy represents the distance between
them. In each condition, the bird-distance matrices produced by each
informant were correlated with each other, yielding a single pairwise
Sorting
77
Combined Consensus
78
Chapter 4
Cluster Analysis
79
average link method (Sneath and Sokal 1973). The results for nonexperts,
experts, and Itza 0 for U.S. and Tikal birds are summarized in gures 4.2
to 4.7.
The cluster analysis produces a hierarchical taxonomy where examples
within a cluster are more similar than examples across clusters. For instance, at the bottom of gure 4.2 one can see a tight cluster (wood
duck, common loon, mallard, common merganser, green winged teal,
and goldeneye), mostly members of the duck family. The corresponding
cluster for the U.S. experts in gure 4.3 includes the black duck, wood
duck, green winged teal, and mallard. Note that this cluster adds a duck
that belongs to the same genus (Anas) and does not include the common
loon, which actually belongs to a dierent order.
For both sets of birds, the three groups showed overall similarity,
coupled with systematic group dierences. As the gures indicate, for
each of the taxonomies there were groups of predators, game birds, water
birds, hummingbirds, and woodpeckers, to name a few of the groups.
Some notable dierences in the taxonomies are as follows. Whereas U.S.
novices and U.S. experts generally kept passerines (small songbirds) in a
large single group, the Itza 0 experts had them spread out across the taxonomy in several dierent clusters.
We also found a dierence in subjects sorting of water birds. On the
U.S. bird set, U.S. experts had a large water birds cluster, featuring
ducks, grebes, geese, shore birds, and herons/egrets. This cluster was
fairly isolated from the rest of the taxonomy. Although novices also had
a water-bird category, it was more spread out, was not as isolated from
other birds, and was interrupted by nonwater birds, such as gamebirds,
nightjars (birds that eat insects while they are ying), the pigeon, and the
turkey vulture. This also reduced the correspondence of novice sorts to
scientic taxonomy.
As in Lopez et al., we used the data from the
sorting study to develop typicality and diversity probes to see how participants used bird categories and salient examples of birds in reasoning.
Again the focus was on typicality and diversity. Recall that in the
category-based induction model of Osherson et al. 1990, both of these
phenomena hinge on coverage.
Given the results from Lopez et al. 1997 as well as Prott, Coley, and
Medin 2000, we expected that U.S. novices would exhibit more diversity
responding than either of the other two groups. It would not be surprising
if the U.S. bird experts showed some modest amount of diversity
Category-Based Induction
80
Chapter 4
Figure 4.2
Cluster analysis of nonexperts sorting U.S. birds. The numbers next to the bird
names correspond to an alphabetic sort of the names.
Figure 4.3
81
82
Figure 4.4
Chapter 4
Figure 4.5
83
84
Figure 4.6
Chapter 4
Figure 4.7
85
86
Chapter 4
responding given that they are quite familiar with the scientic taxonomy.
Overall, however, our hypothesis was that domain knowledge makes it
less likely that a person will employ abstract reasoning strategies. Instead
we expected to observe more concrete justications such as the ecological/
causal reasoning.
Based on previous work we decided against using
identical properties for the Itza 0 and U.S. induction probes. Half of the
probes involved disease and this was constant across groups. For the
other half we used enzyme for North American subjects and little
things inside for Mesoamerican subjects. We piloted both terms with
both groups. We found that North American adult participants are confused by little things inside but not enzyme, protein, or disease
X, whereas Maya subjects were confused by enzyme and protein
but not by little things inside or disease X. Earlier studies show that
the patterns of results on dierent kinds of biological induction tasks for
American undergraduates were statistically the same for enzyme and
disease, whereas the Itza 0 showed the same patterns of results for little
things inside and disease (Atran et al. 1997; Coley et al. 1999). As in
the sorting study we used probes involving both U.S. birds and birds of
Tikal. There were no dierences as a function of property so we collapsed
across this variable.
For both kinds of probes we presented two pairs of birds and then
asked about the property in question (disease, enzyme, or little things inside). For example, for the typicality trials, we displayed both birds in
each pair and said: Lets assume that we discovered two new diseases.
All we know about these diseases is that Disease A is found in these types
of birds and Disease B is found in these. Which disease do you think is
more likely to be found in all birds? Similarly, for the diversity trials,
we placed one pair of birds on the left-hand side and one pair of birds
on the right-hand side, and asked the same question.
Only the undergraduates (novices) showed any indication of a typicality eect. A look at the justications for choices conrms
this pattern. The most striking dierence is that novices use typicality as a
reason for the choice more than half of the time, while experts and Itza 0
never indicate typicality. Both Itza 0 and U.S. experts tended to use range
or ecological factors as justications. These justications are summarized
in table 4.3.
Typicality Results
87
Table 4.3
TYP
BEH
ECO
GEO
NUM
EVO
U.S. expert
Tikal
U.S.
0
0
0
0
18
12
39
37
2
4
21
23
Itza 0
Tikal
U.S.
0
0
12
7
60
58
18
26
1
9
0
0
47
56
2
4
7
3
4
5
18
18
2
0
Nonexpert
Tikal
U.S.
Note: TYP Typicality, BEH Behavior, MOR Morphology, ECO Ecology, GEO Geographic range, NUM Number, and EVO Evolutionary age.
Diversity
88
Chapter 4
89
90
Chapter 4
experts and Itza 0 ), who have substantial knowledge, goals, and activities
involving the items they classify and reason with, information other than
that derived from perceptual clustering and similarity judgment is relevant to understanding natural biodiversity. Behavior and ecology, for example, appear to be crucial to the deeper and broader understanding of
nature that bird-watchers seek. Such concerns also may be critical to the
way the Maya and perhaps other peoples in small-scale societies manage
to live and survive with nature. If so, then it is practically impossible to isolate folkecological orientation from other aspects of cultural knowledge.
We think that our pattern of results can best be
understood with a theory that has not been applied previously to problems of category-based induction. One of our test sessions with a tree expert provided the impetus for this shift of view. The expert was being
given typicality probes such as the following: Suppose we know that
river birches get disease X and that white oaks get disease Y, which disease do you think is more likely to aect all trees? In this case, the expert
said disease X, noting that river birches are very susceptible to disease, so
if one gets it they all get it. The very next probe involved the gingko
tree and the expert chose the disease associated with it as more likely to
aect all trees on the grounds that gingkos are so resistant to disease
that if they get it, it must be a very powerful disease. He then said that
he felt as if he had just contradicted himself, but that nonetheless these
seemed like the right answers.
Normatively, this experts answers do not represent a contradiction. Instead, he appeared to be using the information that was most salient and
accessible to guide his reasoning (birches are notoriously susceptible to,
and gingkos notoriously resistant to, diseases). Simply put, the expert
was using the knowledge that he considered most relevant. We believe
that Sperber and Wilsons (1986) relevance theory provides a good framework for understanding the patterns of responding in all our populations.
Furthermore, it leads to a number of novel predictions that contrast with
those of other models of induction.
In relevance theory, relevance is seen as a property of inputs to cognitive processes:
A Relevance Framework
91
vidual had in mind (or in an experimental situation, was presented with). (Van der
Henst, Politzer, and Sperber 2002)
Consider again the Itza 0 pattern of sorting and reasoning about birds. Recall that their daily-life circumstances lead them to attend to the larger,
more ecologically important forest birds. These are ecologically important both to perceivable eects on the forest and to Itza 0 needs. For example, raptors compete with the Itza 0 for large gamebirds (e.g., the wild
turkey), and so Itza 0 hunters clearly must pay attention to both groups of
birds. Consequently, their choices of nonpasserines on reasoning probes
are driven by these omnipresent background concerns. Specically, their
extensive knowledge of large gamebirds and raptors has consequences
for both eect and eort. All else equal, it is easier for them to retrieve knowledge about nonpasserines and when they do so, this retrieved
knowledge has greater consequences.
Undergraduates, in contrast, have little background knowledge to
bring to bear on the sorts of reasoning tasks we have used and consequently it is not surprising that they rely heavily on more abstract reasoning strategies. At the same time we do have evidence that their responses
are sensitive to both eect and eort. This line of work was motivated by
a follow-up study involving reasoning about mammals. Here we tested
undergraduates individually and asked them to justify their responses.
The one-on-one context implicitly asks for more eort, which should
lead to more eect. Under these circumstances diversity eects were
much reduced and we started to see justications in terms of the range
and population size of dierent mammals.
The above pilot study has led us to examine relevance eects in undergraduate populations more systematically (Medin et al. 2003). The probes
rely on identifying accessible background knowledge to bring out the effect side of relevance and manipulating the premise and conclusion categories to show consequences on the eort side. As an example of the
former, we nd that the argument that bananas have enzyme X, therefore
92
Chapter 4
93
ition is that robins are better examples of birds than are chickens. Furthermore, the consensus has been that the basis of typicality eects is
similarity relationshipsrobins are better birds because they are more
similar to other birds than are chickens (for empirical and theoretical
treatments of typicality, see Smith, Shoben, and Rips 1974; Rosch and
Mervis 1975). Once again, however, these observations rest on a narrow
empirical base with respect to study populations.
Work on typicality judgments among Itza 0 shows that inductively useful notions of typicality may be driven more by considerations of idealness than central tendency (Atran 1999a). In each case for which we
have direct Itza 0 ratings, the truest or most representative livingkind categories are large, perceptually striking, culturally important, and
ecologically prominent. For example, the three most highly rated mammals are the jaguar (also called The Lord of the Forest), the mountain
lion (the jaguars principal rival), and the tapir (also called The Beast of
All Seven Edible Kinds of Flesh). The three most highly rated snakes
are the large and deadly fer-de-lance (Bothrops asper, also called The
True Snake) and its companions, the large and venomous tropical rattlesnake (Crotalus durissus) and the smaller but deadly coral snake
(Micrurus sp.). The three most representative birds are all large, morphologically striking, and highly edible galliformes (wild fowl): ocellated turkey, crested guan, and great curassow.
One might speculate that the instructions were not comparable or that
typicality has a dierent meaning in the Itza 0 language. Further observations undermine this possibility. Lynch, Coley, and Medin (2000) found
that U.S. tree experts based their typicality judgments on ideals (e.g.,
height, absence of undesirable characteristics) and that central tendency
was uncorrelated with judgments. Lynch et al. used instructions that followed verbatim those by Rosch and Mervis (1975) in their original studies
showing central-tendency-based typicality eects.5 The best predictor of
undergraduate typicality ratings was word frequency. This ts with the
idea that many undergraduates do not know the referents of these terms
and, therefore, rely on the only information they have.
Bailenson et al. 2002 also collected typicality ratings. Central tendency
was correlated with judgments only for the novices. We did not try to
measure ideals or other factors that might determine typicality ratings
among bird experts, but in another domain of biologyfreshwater
shwe did.
To demonstrate the generality of these ndings and to get a better understanding of ideals, we ran a study of culture
94
Chapter 4
95
Procedure
96
Table 4.4
Desirability
Fish
Me
MC
Me
MC
Sci CT
Me
MC
Fam
Size
(cm)
Hab
2.4
3.7
5.3
4.1
3.5
6.2
4.0
6.8
6.8
2.8
3.9
4.5
4.0
2.3
3.5
4.0
4.5
2.5
3.9
4.7
1.6
4.1
5.3
3.1
2.8
6.1
2.6
6.3
5.9
3.2
4.8
4.0
3.2
1.9
3.5
3.2
4.8
1.9
3.8
5.2
1.32
0.08
0.42
0.08
0.64
0.42
0.86
0.69
0.67
0.43
0.43
0.91
0.76
0.39
0.69
0.61
0.32
0.33
0.52
0.20
0.89
0.20
0.43
0.41
0.58
0.45
0.14
0.88
0.87
0.07
0.10
0.22
0.44
0.15
0.71
0.27
0.18
0.33
0.25
0.49
0.48
0.02
0.52
0.23
0.77
0.65
0.77
0.23
0.23
0.65
0.10
0.65
0.10
0.48
0.90
0.77
0.10
0.48
0.65
0.65
.50
.17
.43
.17
.17
.43
.17
.14
.14
.33
.00
.17
.17
1.00
.17
.17
.17
1.00
.17
.57
.50
.17
.33
.58
.00
.33
.00
.08
.08
.67
.08
.00
.08
.83
.00
.00
.00
.83
.00
.25
.64
.92
.92
.67
.70
1.00
.50
.98
.97
.94
.91
.32
.30
.95
.59
.88
.80
.89
.94
.85
152
62
49
61
26
41
11
70
103
122
127
12
10
109
13
10
155
183
30
31
1
0
1
1
1
1
1
1
1
0
1
1
1
1
1
0
1
0
0
1
Chapter 4
Typicality
2.0
6.7
3.7
6.3
6.0
5.3
6.9
3.6
3.7
3.9
4.2
5.1
2.7
6.1
4.2
3.8
3.4
6.2
6.8
4.9
5.3
3.9
3.7
6.0
1.3
5.9
2.9
6.4
6.5
5.8
5.9
2.5
3.3
3.1
3.4
5.6
2.4
5.9
3.8
3.3
1.9
5.1
6.6
4.4
5.3
3.3
4.0
6.1
1.34
0.31
0.91
0.26
0.09
0.42
0.69
0.16
0.56
0.76
0.15
0.16
0.51
0.24
0.42
0.73
0.30
0.75
0.05
0.22
0.33
0.00
0.08
0.12
1.00
0.21
0.23
0.25
0.25
0.49
0.84
0.23
0.54
0.13
0.30
0.15
0.14
0.22
0.49
0.83
0.19
1.04
0.02
0.33
0.46
0.38
0.12
0.05
5.73
0.52
0.48
0.23
0.23
0.65
0.23
0.23
0.77
0.90
0.40
0.02
0.48
0.52
0.48
0.90
0.48
0.48
0.02
0.48
0.52
0.23
0.02
0.10
.67
.86
.17
.14
.43
.57
.14
.17
.17
.17
.40
.14
.33
.71
.14
.02
.00
.00
.43
.57
.29
.17
.19
.43
.42
.50
.00
.67
.67
.33
.00
.58
.00
.00
.42
.67
.67
.50
.00
.08
.00
.08
.75
.50
.33
.50
.25
.42
.85
1.00
.65
1.00
.98
.92
.92
.89
.71
.89
.98
.67
.68
1.00
.92
.61
.38
1.00
1.00
.88
.91
.82
.86
.97
64
97
8
183
133
40
114
74
23
13
43
76
89
69
33
15
7
274
91
45
53
64
47
40
97
Note: CT central tendency, Sci scientic, Me Menominee, MC majority culture, Fam familiarity, Hab habitat.
1
1
1
0
0
1
1
1
1
1
0
0
1
0
0
1
1
1
0
1
1
0
0
1
Lamprey eel
Largemouth bass
Mudminnow
Musky
Northern pike
Pumpkinseed
Rainbow trout
Redhorse
Redtail chub
River shiner
Rock bass
Sauger
Sheephead (drum)
Smallmouth bass
Smelt
Spottail shiner
Stickleback
Sturgeon
Walleye
White bass
White crappie
White sucker
Yellow bullhead
Yellow perch
98
Chapter 4
standardized and multiplied by 1. Desirability was derived from justications associated with the sorting task. Participants often formed categories of sh that they described as undesirable (rough or garbage sh)
or desirable (prestigious game sh and sh that are good eating).
Each shs desirability was computed as the proportion of times it
was assigned to desirable groups minus the proportion of times it was
assigned to undesirable groups. This was done for the cultural groups separately and combined. Each shs characteristic adult size was included as
a possible predictor because it may be related to ideals and because size
has been found relevant in related contexts (Hunn 1999; Lynch, Coley,
and Medin 2000). Familiarity was computed as the proportion of participants who knew a sh during the initial familiarity task. Finally, a shs
habitat was coded as 1 if the sh is found mainly in rivers and streams,
1 if the sh is found mainly in lakes, and 0 if the sh is commonly found
in both types of water.
Results Typicality ratings and the values of the predictor variables are
presented in table 4.4. The rst thing to note is that high ratings were
given to desirable game sh like musky, northern, walleye, and largemouth and smallmouth bass. High ratings were also given to other desirable sh like the bluegill, the trouts, and the yellow perch. Low ratings
were given to rough (undesirable) sh like the gar and the dogsh. Minnows and other baitsh received intermediate ratings. To get a broad perspective, typicality, folk CT, and desirability were computed over all
participants (rather than for each cultural group, as shown in table 4.4),
and correlations among these and the other candidate predictors were
computed. The correlations are shown in table 4.5. Typicality is very
highly related to desirability (r .80) and fairly well related to familiarity
(r .50). Both of these correlations are reliable ( ps < :01), and no other
predictor variable is reliably correlated with typicality.
Figure 4.8 collates much of the data in table 4.4 into subcategories corresponding to the predictions we described earlier. Gamesh comprise the
musky, northern pike, sauger, largemouth bass, and smallmouth bass.
Pansh include the black crappie, bluegill, green sunsh, pumpkinseed,
rock pass, white crappie, and yellow perch and rough sh, the American
eel, black sucker, dogsh, gar, lamprey eel, redhorse, sheephead, and
white sucker. Baitsh comprise the minnows, shiners, and chubs.
As predicted, Menominee participants gave higher ratings overall,6 and
there was a signicant interaction of cultural group and sh group.7 This
interaction took the form predicted in most respects. Menominee partici-
99
Table 4.5
Folk
CT
Sci
CT
Desirability
Familiarity
.06
.28
.64**
.80**
.24
.30*
.50**
.38*
.05
.28
Habitat
Size
.24
.03
.12
.26
.51**
.22
.70**
.23
.14
.43**
.02
Note: Because Menominee participants gave higher typicality ratings overall, ratings were standardized for each participant before being averaged and submitted
to correlational analysis.
CT central tendency, Sci scientic.
* p < :05, ** p < :01.
Figure 4.8
100
Chapter 4
Implications
101
102
Chapter 4
103
104
Chapter 4
Figure 4.9
Inductive inferences for Itza 0 Maya and U.S. students compared (after Coley et al.
1997).
105
106
Figure 4.10a
Chapter 4
Figure 4.10b
107
108
Chapter 4
Discussion of Results
These results indicate that both the inexperienced Americans and the
Itza 0 elders prefer taxa of the generic-species rank in making biological
inferences. If inferential potential were a simple function of perceptual
similarity, then American nonexperts should prefer life forms for induction (as with Rosch et al. 1976). The ndings suggest that root categorization and reasoning processes in folkbiology owe something to conceptual
assumptions (about the causal locus of biologically essential attributes at
the generic-species level) and not exclusively to general, similarity-based
(e.g., perceptual) heuristics. To be sure, language may signal expectation
that little or poorly known generic species are more biologically informative than better-known life forms for Americans (e.g., via common use of
binomials, such as oak / red oak). But why presume that an appropriately
tagged item is the locus of a deep causal nexus of biological properties
and relationships? Why suppose at all that there is such a nexus that
spontaneously justies and motivates expectations, inferences, and explorations relating little-known or nonobvious aspects of a presumably fundamental biological reality? Indeed, our experiments still show reliable
results in the absence of clear linguistic cues (e.g., oak / white oak / swamp
white oak vs. dog / poodle / toy poodle).
In related work with U.S. botanical experts and undergraduates,
Schwartz and Medin (2000) used a converging technique to get at inductive privilege. In their studies, premises and conclusions always involved
the varietal level, but the two kinds varied as to whether they belonged
to the same species, the same genus, the same family, or the same life
form. For example, a premise involving fastigiata black alders and a conclusion involving heritage river birch trees would be a probe of the family
level, since the most specic taxonomic rank including birches and alders
is the family level. The dependent variable was rated inductive condence
on a 9-point scale. The advantage of this technique is that it allows one to
probe for unnamed ranks.
The results are shown in gure 4.11. Both groups provide clear evidence of privilege at the generic-species level, again indicating that relative expertise does not aect where the inection point in inductive
condence lies. On a ner level of detail there are expertise eects. The
undergraduates show no dierence between the life-form and family
levels, where the experts do.11 Shared genus (e.g., red oak, white oak) is
marked in language whereas shared family usually is not. Apparently the
undergraduates had no idea of taxonomic relationships other than those
109
Figure 4.11
Interaction of level of expertise and taxonomic level for the kingdom plants
(Experts, n 39; Novices, n 24).
marked in language. In contrast, the botanical experts did make this distinction with respect to secondary privilege.
More recently Coley et al. (2004) have both added a developmental
component and provided evidence that the knowledge versus expectation
may be specic to biological kinds. They tested U.S. 5-year-olds, 8-yearolds, and adults on feature-listing and induction tasks. For all age groups
the greatest increase in features listed tended to occur at the life-form
level, whereas the greatest increase in inductive condence tended to
occur at the generic-species level. Knowledge and induction were more
concordant for artifact stimuli.
Undergraduates lack of close contact with biological kinds may be
precisely what allows us to tease apart the contributions of perceptual
processes and abstract expectations to the privileged level in induction.
There is now considerable evidence for perceptual learning (e.g., see
Goldstone 1994; Schyns, Goldstone, and Thibaut 1998) in general as well
as evidence that the basic level on perceptual tasks becomes more specic
with expertise (e.g., Tanaka and Taylor 1991; Johnson and Mervis 1997).
Expertise is almost always a relative term and one equally could cast
these results into a dierent frame: so-called expert performance on perceptual tests could be the default outcome of normal development, and
110
Chapter 4
undergraduate performance on perceptual tests (favoring the more abstract life-form level) could be the result of a failure to undergo normal
perceptual development with respect to biological kinds. If this were true,
then we would expect Itza 0 to perform like experts on perceptual tests,
and only for cases of impoverished input would we expect a discrepancy
between abstract expectations and perceptual processes. Arguably, there
is an evolutionary design to a cognitive division of labor between
domain-general perceptual heuristics and (domain-specic) learning and
inference mechanisms, the one enabling exible adaptation to variable
conditions of experience, and the other invariably steering us to those
enduring aspects of biological reality that are both causally recurrent
and relevant to the emergence of human life and cognition.
We consistently found a decisive break in inductive strength
just above the rank of generic species. Nevertheless, we also found secondary evidence that supports the downgrading of American folkbiological knowledge versus the upgrading of Maya knowledge, relative to
the generic-species level. Specically, we nd Americans have more faith
in inductions to superordinate life-form taxa than the Itza 0 , and Itza 0 differentiate among subordinate taxa more than students. This observation,
coupled with some suggestive data on the decreasing salience of biological kinds in Western societies, raises further issues concerning the relativity of expertise.
Summary
This has been a long chapter and it is time to sum up. The results
described in the chapter inform our understanding of folkbiology as a
module and the cognitive consequences of diminished contact with nature. They also have important implications for the psychology of categorization and reasoning.
Categorization
Two of the most robust and signicant ndings in the psychology of concepts are basic-level and typicality eects. Our work suggests important
modications in each of these.
Typicality
The standard assumption has been that goodness of example, or typicality, is driven by similarity relations. A good example of a category is one
111
that looks like its fellow category members and unlike members of contrasting categories (e.g., Rosch and Mervis 1975; Smith, Shoben, and
Rips 1974; Smith and Medin 1981). As we noted, the similarity-coverage
model assumes that goodness-of-example eects extend to category-based
induction.
Once again, however, results based on the standard undergraduate
population proved atypical in the case of biological kinds. First, when
the stimuli being judged are names of trees, undergraduates even fail to
show similarity-based typicality. Instead, word frequency or familiarity
is the best predictor (Lynch, Coley, and Medin 2000). Apparently, undergraduates know too little about trees to have a basis for computing similarities. More to the point, populations with domain familiarity, whether
professional taxonomists or Itza 0 agroforesters, consistently organize categories in terms of ideals, such as the taxonomists American elm or the
Mayas wild turkey.
We believe that people who have serious commerce in a domain rarely
approach it in a content-neutral manner, passively recording the regularities associated with the category. We saw that the Itza 0 , for example, bias
their observations of biological kinds toward those that are most perceptually and ecologically salient (e.g., large gamebirds, predators, and poisonous snakes). Parks workers worry about susceptibility to disease and
other maintenance problems with local trees, and their typicality ratings
reect this concern. Majority-culture shermen attend to gamesh and
Menominee shermen expand that focus to include sacred, culturally important sh. In brief, the ways people deal with the world aect the ways
they cognize it.
Consequently, models of categorization need to be sensitive to the likelihood that the most relevant and best examples of a category will tend to
be learned rst, and that later learning will be aected by and build on
earlier learning (again see Berlin 1992; Love, Medin, and Gureckis 2004;
Steyvers and Tenenbaum 2005).
Basic Level and Essentialism
112
Chapter 4
Our data are consistent with the idea that biological essentialism may
be universal and linked to an evolutionarily adaptive appreciation of generic species. For contemporary peoples in small-scale societies who continue to live intimately with nature, the level of generic species is the most
relevant, as it likely was for our hominid ancestors. When we used an induction task where performance can be based on knowledge or expectation or both, we found convergence across cultures and expertise on the
generic-species level as privileged for biological inference. The fact that
biological experts also privilege the generic-species level on perceptual
tests suggests that the divergence in question has little to do with how
psychologists versus ethnobiologists measure the basic level. Rather, the
apparent salience of the life-form level for undergraduates on feature listing and perceptual tests appears to be a peculiarity of the devolved state
of undergraduate biological knowledge in particular, and that of industrialized populations in general (for a example involving the German language, see Zubin and Kopcke 1986).
Why should the generic-species level be privileged for biological inference in the face of uncertainty? Because that is where the action was, and
often still is, in human dealings with biological kinds. It would also be
sensible for the perceptual system to be tuned to this same level of biological reality, and we suspect that this is the default condition for human
beings who depend directly on nature for survival (i.e., without the intermediary of shops and supermarkets). Some perceptual learning may be
necessary to achieve this consonance (e.g., Goldstone 1998; Schyns and
Rodet 1997; Johnson and Mervis 1997), experience that undergraduates
may lack. More generally, people may have a perceptual-familiarity heuristic that allows them to rapidly and economically navigate their everyday world. This heuristic may be importantly inuenced by cultural
support (Wol, Medin, and Pankratz 1999). There is increasing evidence
from studies with infants that words act as invitations to form basic-level
concepts (Waxman and Markow 1995; Waxman 1999), which in our society tend to focus on the life-form level (except for familiar pets and domestic animals; hence, bird, sh, and dog are basic).
Category-Based Inference
113
indirect, rather than direct. That is, idealness per se plays no role in the
rationale for responses. Instead, it is the implicit organization of knowledge organized around goals that both creates category ideals and drives
category-based inference. For example, the Itza 0 Maya nd passerines
less relevant than gamebirds and raptors for understanding the forest
(the forest being the primary focus of their understanding of the biological world). Consequently, they have much more knowledge about
the large birds, knowledge that is recruited on reasoning tasks.
Although previous induction models have implicitly assumed that
diversity-based responding is universal, it clearly is not. When we probed
Itza 0 , bird-watchers, tree experts, and shermen in areas where they had
knowledge we hardly ever observed diversity responses (and sometimes
found below-chance diversity). Obviously, observations such as these require a reformulation of inference theories, perhaps along the lines of relevance theory.
Itza 0 noncompliance with diversity-based reasoning apparently results
neither from a failure to understand the principle of diversity nor from
any problems of computational load. As with the most evident divergences between American and Itza 0 performance on similarity and typicality tasks, divergence from diversity apparently results from real-world
concerns. In the absence of a theoryor at least the presumption of a
theoryof causal unity underlying disparate species, there is no compelling reason to consider a property discovered in two distant species as
biologically intrinsic or essential to both (see also Prott, Coley, and
Medin 2000). This does not mean that Itza 0 do not understand a diversity
principle. In fact, in a series of tasks designed to assess risk-diversication
strategies (e.g., sampling productivity from one forest plot or several)
Itza 0 consistently showed an appreciation of the diversity principle in
these other settings (Lopez et al. 1997). This suggests that although diversity may be a universal reasoning heuristic, it is not a universally relevant
aspect of folkbiological taxonomy, as we also found in U.S. populations
having more direct interest in the natural world.
Autonomy and Universality
114
Chapter 4
All organisms must function to procure energy to survive, and they also
must procure (genetic) information for recombination and reproduction
(Eldredge 1986). The rst requirement is primarily satised by other species, and an indiscriminate use of any individual of the other species (e.g.,
energywise, it does not generally matter which individual lion eats you or
which individual apple you eat). The second requirement is usually only
satised by genetic information unique to individual conspecics (e.g., ge-
115
116
Chapter 4
Universality
Ever since the pioneering work of Berlin and his colleagues, evidence
from ethnobiology and experimental psychology has been accumulating
that all human societies have similar folkbiological structures (Berlin,
Breedlove, and Raven 1973; Berlin 1992; Hunn 1977; Hays 1983; Brown
1984; Atran 1990, 1999a). These striking cross-cultural similarities suggest that a small number of organizing principles universally dene
folkbiological systems. Basic aspects of folkbiological structure (e.g., taxonomic ranking, primacy of generic species) seem to vary little across cultures as a function of theories or belief systems.
Precocity of Acquisition
Acquisition studies indicate a precocious emergence of essentialist folkbiological principles in early childhood that are not applied to other
domains (Keil 1995; Hatano and Inagaki 1999; Atran et al. 2001). We
will provide further evidence on ease of acquisition in chapter 5.
Independence from Perceptual Experience
Cerebral impairments (Williams syndrome, brain lesions caused by certain types of herpes virus, and so on) suggest selective retention or loss
of folkbiological taxonomies or of particular taxonomic ranks. Neuropsychological studies have reported a pathological performance in recognition at the life-form and generic-species levels (e.g., recognizing an item
as an animal but not as a bird or robin), and dissociation at the life-form
level (e.g., not recognizing items as trees). Existing studies, however, do
not say anything about the generic-species rank as the preferred level of
representation for reasoning, perhaps because of methodology (linked to
averaging over items and failure to include sets of generic species) (Warrington and Shallice 1984; Sartori and Job 1988; Job and Surian 1998;
Caramazza 2002).
Resistance to Inhibition
117
essences. Such beliefs greatly help people explore the world by prodding
them to look for regularities and to seek explanations of variation in
terms of underlying patterns. This strategy may help bring order to ordinary circumstances, including those relevant to human survival. But in
other circumstances, such as wanting to know what is correct or true for
the cosmos at large, such intuitively ingrained concepts and beliefs may
hinder more than help. For example, the essentialist bias to understand
variation in terms of deviance is undoubtedly a hindrance to evolutionary
thinking. In some everyday matters, the tendency to essentialize or explain variation in terms of deviation from some essential ideal or norm
(e.g., people as mental or biological deviants) can be an eortlessly
natural but wrong way to think.
Because intuitive notions come to us so naturally they may be dicult
to unlearn and transcend. Even students and philosophers of biology
often nd it dicult to abandon commonsense notions of species as
classes, essences, or natural kinds in favor of the concept of species as a
logical individuala genealogical branch whose end points are somewhat arbitrarily dened in the phyletic tree and whose status does not
dier in principle for that of other smaller (variety) and larger (genus)
branches. Similarly, racismthe projection of biological essences onto
social groupsseems to be a cognitively facile and culturally universal
tendency (Hirschfeld 1996). Although science teaches that race is biologically incoherent, racial thinking is as notoriously dicult to suppress as
it is easy to incite.
Ease of Cultural Transmission
Human cultures favor the rapid selection and stable distribution of ideas
that (1) readily help to solve relevant and recurrent environmental problems, (2) are easily memorized and processed by the human brain, and (3)
facilitate the retention and understanding of ideas that are more variable
(e.g., religion) or dicult to learn (e.g., science) but contingently useful or
important. Folkbiological taxonomies readily aid humans everywhere in
orienting themselves and surviving in the natural world. The content of
these taxonomies tends to be stable within cultures (high interinformant
agreement, substantial historical continuity) and their structure isomorphic across cultures (see Boster 1991; Lopez et al. 1997). Folkbiological
taxonomy also serves as a principled basis for transmission and acquisition of more variable and extended forms of cultural knowledge.
Consider the spontaneous emergence of totemismthe correspondence
of social groups with generic speciesat dierent times and in dierent
118
Chapter 4
parts of the world. Why, as Levi-Strauss (1963) aptly noted, are totems so
good to think? In part, totemism uses representations of generic species
to represent groups of people; however, this pervasive metarepresentational inclination arguably owes its recurrence to its ability to ride piggyback on folkbiological taxonomy. Generic species and groups of generic
species are inherently well structured, attention-arresting, memorable,
and readily transmissible across minds. As a result, they readily provide
eective pegs on which to attach knowledge and behavior of less intrinsically well-determined social groups. Totemic groups thereby also become
memorable, attention-arresting, and transmissible across minds.
These are the conditions for any idea to become culturally viable (see
Sperber 1996 for a general view of culture along the lines of an epidemiology of representations). A signicant feature of totemism that
enhances both memorability and its capacity to grab attention is that it
violates the general behavior of biological species: members of a totem,
unlike members of a generic species, generally do not interbreed, but
only mate with members of other totems so as to create a system of social
exchange. Notice that this violation of core knowledge is far from arbitrary. In fact, it is such a pointed violation of human beings intuitive
ontology that it readily mobilizes most of the assumptions people ordinarily make about biology (Atran and Sperber 1991).
4.8 Conclusions
119
In her inuential 1985 book Susan Carey proposed that young childrens
understanding of living things is initially embedded in a folkpsychological, rather than folkbiological, explanatory framework and that human
beings act as the prototype. Her data suggested that children did not develop an independent model of biology where humans were seen as one
animal among many until they were 10 to 12 years old. In short, on this
view, children have to undergo a fundamental conceptual change to
achieve an autonomous biology.
A strong form of evidence for this theory comes from an inductive inference task where children are told that some novel property is true of
one biological kind (e.g., Humans have a little green thing inside them
called an omentum), then are asked whether that property is true of
other biological kinds (e.g., Do you think that dogs have an omentum?). Three major ndings bolster the claim that young childrens conceptions of the biological world are anthropocentric. First, children more
readily project properties from humans onto other living kinds than
from other living kinds onto one another. The other two ndings are
consequences of this dierence in induction potential. The second result
122
Chapter 5
To evaluate the role of cultural milieu and conditions of learning in childrens inductive reasoning, we initially studied four populations: urban
Boston children, rural Wisconsin majority-culture children, Menominee
children, and Yukatek Maya children of varying ages (4 to 11) and adults
(Ross et al. 2003; Atran et al. 2001). All testing in the United States
was in English; Yukatek Maya was used for the Maya children and
adults.
123
124
Table 5.1
Adults
males
average
females
males
average
females
males
average
Human
human
mammal
bird
reptile
invertebrate
tree
stu
artifact
sun
1.00
0.63
0.63
0.75
0.67
0.50
0.42
0.50
0.58
1.00
0.57
0.43
0.36
0.43
0.36
0.43
0.29
0.50
1.00
0.60
0.53
0.55
0.55
0.43
0.42
0.39
0.54
0.96
0.46
0.50
0.38
0.62
0.19
0.23
0.23
0.42
1.00
0.57
0.48
0.19
0.33
0.10
0.05
0.00
0.38
0.98
0.52
0.49
0.29
0.47
0.14
0.14
0.12
0.40
1.00
0.58
0.25
0.17
0.13
0.00
0.04
0.08
0.13
1.00
0.71
0.58
0.46
0.25
0.04
0.04
0.00
0.08
1.00
0.65
0.42
0.31
0.19
0.02
0.04
0.04
0.10
Dog
human
mammal
bird
reptile
invertebrate
tree
stu
artifact
sun
0.64
0.89
0.46
0.64
0.32
0.25
0.29
0.25
0.36
0.54
0.89
0.68
0.46
0.39
0.25
0.14
0.11
0.32
0.59
0.89
0.57
0.55
0.36
0.25
0.21
0.18
0.34
0.83
1.00
0.33
0.17
0.17
0.08
0.08
0.25
0.17
0.17
1.00
0.58
0.25
0.25
0.08
0.00
0.00
0.25
0.50
1.00
0.46
0.21
0.21
0.08
0.04
0.13
0.21
0.25
1.00
0.17
0.04
0.21
0.04
0.08
0.13
0.08
0.71
0.96
0.42
0.38
0.17
0.00
0.08
0.00
0.08
0.48
0.98
0.29
0.21
0.19
0.02
0.08
0.06
0.08
Chapter 5
females
0.08
0.50
0.42
0.50
0.42
0.33
0.33
0.42
0.33
0.50
0.71
0.36
0.50
0.29
0.14
0.36
0.14
0.50
0.29
0.61
0.39
0.50
0.35
0.24
0.35
0.28
0.42
0.58
0.81
0.58
0.46
0.58
0.19
0.23
0.12
0.31
0.33
0.81
0.62
0.33
0.33
0.05
0.05
0.00
0.19
0.46
0.81
0.60
0.40
0.46
0.12
0.14
0.06
0.25
0.79
0.79
0.25
0.17
0.17
0.25
0.25
0.04
0.13
0.67
0.83
0.54
0.46
0.42
0.21
0.38
0.00
0.08
0.73
0.81
0.40
0.31
0.29
0.23
0.31
0.02
0.10
Bee
human
mammal
bird
reptile
invertebrate
tree
stu
artifact
sun
0.33
0.25
0.46
0.50
0.92
0.25
0.17
0.42
0.58
0.29
0.29
0.29
0.29
0.39
0.14
0.21
0.07
0.36
0.31
0.27
0.37
0.39
0.65
0.20
0.19
0.24
0.47
0.50
0.54
0.38
0.42
0.85
0.19
0.31
0.12
0.27
0.33
0.52
0.52
0.43
0.76
0.10
0.05
0.00
0.19
0.42
0.53
0.45
0.43
0.80
0.14
0.18
0.06
0.23
0.58
0.21
0.08
0.00
0.75
0.38
0.17
0.08
0.13
0.75
0.33
0.29
0.29
0.63
0.63
0.13
0.08
0.13
0.67
0.27
0.19
0.15
0.69
0.50
0.15
0.08
0.13
Pec
human
mammal
bird
reptile
invertebrate
tree
stu
artifact
sun
125
126
Chapter 5
Figure 5.1
Figure 5.2
127
Figure 5.3
wild animals and invertebrates may favor them less as sources of induction. The fact that dogs are a better base for induction than are peccaries
is consistent with this observation. One candidate explanatory principle is
that the more properties a child knows about some kind, the more likely
they are to generalize some new property to other living kinds. We will
evaluate this and other explanations later in this chapter.
Young children (especially the girls) generalized in a fairly undierentiated way from humans (gure 5.3). It is not clear how to interpret this
pattern of results. One possibility is that these children lack a clear grasp
of how humans t into the tree of life (the girls show the same pattern
with the peccary, an animal with which they are unfamiliar). Another
possibility is that humans, being the primary focus of ecological interactions, provide a plausible inductive base for thematic relationships that
may have little correlation with taxonomic distance.
On the whole, Yukatek Maya children show some intriguing gender
dierentiation. These gender dierences may reect the strong sexual division of activity that is institutionalized early in the rst year of life. In
the jeetz@meek 0 ceremony, Maya girls are introduced by the women to
household utensils, whereas Maya boys are introduced by the men to agricultural and hunting tools. Later in life, Maya women will spend their
time almost wholly in the vicinity of the house and house garden, in close
interaction with domestic animals. By contrast, Maya men spend days,
128
Chapter 5
weeks, and even months in the forest away from home. For Maya
females, dogs are household animals, whereas men value dogs as hunting
animals. Maya boys also venture out into the forest with their fathers at
an early age, and so become familiar with wild animals, such as the peccary, before girls do. These ndings suggest that induction patterns may
be inuenced by relative familiarity with animals and by the culturally
specic character of the functional and ecological relationships between
humans and other natural categories of elements.
In this study (Ross et al. 2003) we compared
urban majority culture, rural majority, and rural Menominee childrens
inductive generalizations. The pattern of responding varied substantially
across groups. The young urban U.S. children (56-year-olds) generalized
in a broad, almost completely undierentiated manner. The only clear
trend was greater generalization from a human base to a human target
than to other targets. Older urban children (78 and 910-year-olds) generalized in terms of biological anity but showed a strong asymmetry in
reasoning between humans and other animals. Although these data do
not replicate Careys precise ndings, they agree in the outcome that
young urban children did not generalize based on biological anity.
The young, rural majority-culture children revealed a dierent pattern;
they showed the mature pattern of generalizing in terms of biological
anity. Interestingly, both they and older rural children showed asymmetries in reasoning between humans and animals and often justied a failure to extend a property from an animal to humans on the grounds that
people are not animals. This observation strongly suggests that the
asymmetry does not derive from humans being conceptualized as the
prototypic animal. Instead, seeing humans as animals may be something of a developmental achievement, as suggested by Johnson, Mervis,
and Boster (1992; see also the sorting task in Carey 1985). Finally, older
rural children gave some evidence of reasoning in terms of ecological
relations, as when they justied generalizing from bees to bears because
a bee might sting a bear or a bear might acquire the property by eating
the bees honey.
Menominee children demonstrated yet a third pattern. First, even the
youngest Menominee often reasoned in terms of ecological relations. In
addition, children of all ages generalized in terms of taxonomic relatedness and showed no reliable human-animal asymmetries. The Menominee
origin myth has people coming from the bear, and even the youngest
children are familiar with the animal-based clan system. In short, there is
129
130
Chapter 5
Table 5.2
Rural
Menominee
Population
79
911
57
79
911
57
79
911
Human-mammal
Mammal-human
Mammal-mammal
Human-insect
Insect-human
Human-plant
Plant-human
Mammal-insect
Insect-mammal
Mammal-plant
Plant-mammal
Insect-plant
Plant-insect
0.70
0.33
0.96
0.47
0.31
0.50
0.13
0.50
0.50
0.32
0.32
0.40
0.31
0.73
0.35
0.90
0.35
0.08
0.27
0.72
0.38
0.56
0.12
0.38
0.44
0.62
0.52
0.24
0.76
0.07
0.28
0.16
0.10
0.38
0.38
0.14
0.24
0.26
0.24
0.58
0.16
0.78
0.34
0.20
0.17
0.12
0.28
0.25
0.14
0.25
0.30
0.34
0.80
0.47
0.92
0.63
0.53
0.30
0.30
0.47
0.56
0.22
0.42
0.58
0.33
0.58
0.42
0.75
0.46
0.63
0.30
0.42
0.57
0.56
0.18
0.50
0.46
0.29
0.70
0.47
0.82
0.55
0.35
0.22
0.41
0.56
0.56
0.33
0.42
0.40
0.41
0.72
0.52
0.88
0.48
0.38
0.34
0.38
0.52
0.56
0.24
0.46
0.42
0.43
We think that dierent patterns of distinctive feature and category label activation play a major role in asymmetries. First consider category labels. In U.S. culture, humans are both
considered to be animals (and not plants) and to be contrasted with animals (footnote: animal may also have a default meaning as beast or
mammal or quadruped). Fleshing out this idea (no pun intended),
one needs to add that distinctive features (categories) of the target diminish inductive condence. The idea is that both animal and humans as
a contrast with animal are associated with the category people, but
only animal is associated with the animals typically used. When some
animal other than humans is a target, animal may be activated and
tend to prime the sense of animal associated with humans. In contrast,
when some nonhuman animal is the base and human the target, the category human may get activated in the target and the induction is called
into question.
The argument would carry through in the same way for distinctive
features. Mammal-to-human generalization is limited by the fact that
humans have many distinctive features; the claim would be that other
mammals have fewer distinctive features. (Note that this claim represents
a means of incorporating familiarity eects, but in the form of limiting
131
132
Chapter 5
These observations seriously complicate the interpretation of the induction task. Consequently, the induction task may have limited utility,
unless it is supplemented by additional converging evidence, such as justications. The justication data we do have imply that distinctive features
and categories of the target category are crucial to asymmetries. In addition, ecological reasoning is more likely to be used when the ecologically
active agent is in the target position. Despite these complications, our
data indicate that anthropocentrism in reasoning is the exception, not
the rule.
Before moving on, lets return to Carey and her procedure. It is the
case that 4- and 6-year-olds were trained and tested on separate days,
though this apparently was not done for older children and adults. The
context was that children were being taught about spleen or omentum along with a review/probe of other, presumably more entrenched
properties like has a lung or has a heart. The teaching was fairly
elaborate, including showing kids a diagram of just where the omentum
is. The test probes asked if something was true of some kind without any
explicit appeal to other kinds that might or might not have that property.
The data analyses were restricted to children who did correctly attribute
the practiced property to the practiced base. In any event, inferences
invited by the experimenter (as in our studies) may show a dierent pattern than those that arise spontaneously where the task is not transparently about inference. We have some preliminary evidence that this is
the case and that the pattern of task dierences varies between urban
and rural children. Even more intriguing, we have suggestive evidence
that the anthropocentric pattern of reasoning in young urban children is
an acquired cultural model. That is, 3-year-old urban children do not
show the anthropocentrism that we observe in 4- to 5-year-old urban children. In short, we are not at the end of the story (Waxman and Medin
2007).
5.3
133
134
Chapter 5
was born to a cow. Still, the child may not know that having certain
kinds of parents causes a cow to be a cow (Carey 1995).
We have been studying several culturally distinct populations to test
the extent to which childrens assumptions about innate species potential
govern projection of both known and unknown properties. In one study
(for details see Atran et al. 2001), Yukatek Maya children and adults
were presented with a forced-choice task involving an adoption scenario.
They were asked whether an adult animal adopted at birth would resemble its adoptive parent (e.g., cow) or birth parent (e.g., pig) on four dierent individual traits: known behaviors (e.g., moo / oink), known physical
features (e.g., straight / curly tail), unknown behaviors (e.g., looks for
chachalacas / looks for pigeons), and unknown physical features (e.g.,
heart gets atter / rounder when it is sleeping). Known traits were context-free, category-typical features that the children readily associate
with species, whereas unknown traits were chosen to minimize any possibility of factual or prelearned associations of traits with categories. Each
unknown trait within a set was attributed to the birth parent for half the
participants and to the adoptive parent for the other half. This ensured
that projection patterns of the unknown traits were not based on prior
associations.
Stories were accompanied by sketches of each parent. Sketches were
designed to unambiguously represent a particular species of animal with
minimum detail. In addition, sketches of known physical features (e.g., a
sketch of a curly or straight tail), unknown physical features (e.g., at versus round heart), and relevant aspects of unknown behavioral contexts
(e.g., closed versus open eyes when afraid, stops in front of mahogany
versus cedar trees) were shown to participants. These sketches in no way
indicated the species to which the traits belonged.
The story was followed by two comprehension questions: (1) Who
gave birth to the baby? and (2) Who did the baby grow up with? Children then were presented with the experimental probes. For example they
might be told: The cow mooed and the pig oinked. When the baby is all
grown up will it moo like a cow or oink like a pig? The probes were followed by a bias control in which the participant was asked, When the
baby was growing up did it eat with animals that looked like X or animals that looked like Y? (Notice that this last probe involves an inference and is not simply a memory check.)
Overall, results showed systematic and robust preference for attributions from the birth parent (see table 5.3). This preference was observed
for all Yukatek age groups and for known and unknown behavior and
0.74**
0.96***
1.0***
0.90
45-year-olds
67-year-olds
Adults
Mean
0.87
0.68*
0.97***
0.96***
Phys. feat.
Behavior
Group
Known
0.88
0.71
0.97
0.98
Mean
0.81
0.69**
0.82***
0.90***
Behavior
Unknown
Percent birth-parent choice for each probe type for each group
Table 5.3
0.81
0.68*
0.83***
0.93***
Phys. feat.
0.81
0.69
0.83
0.92
Mean
0.87
0.65
0.99***
0.97***
Kind
0.74
0.56
0.79**
0.88***
Blood
0.02
0.06***
0.01***
0***
Bias control
(Food)
136
Chapter 5
Table 5.4
Percent birth-parent choice for Brazilian children (after Sousa et al. 2002)
4-year-olds
5-year-olds
6-year-olds
7-year-olds
Adults
Known
behavior
Known
trait
Unknown Unknown
behavior trait
Blood
Control
0.87***
0.92***
0.71*
0.83**
1.00***
0.87***
0.96***
0.87***
0.83**
1.00***
0.78**
0.78**
0.71*
0.79**
0.83**
0.13***
0.00***
0.04***
0.00***
0.00***
0.83**
0.87***
0.75*
0.83**
0.87***
0.33
0.25*
0.26*
0.35
0.96***
137
were not counterbalanced and tended to be much more negative for the
adoptive parent. Informants may have the belief that negative properties
are more powerful and dominate positive qualities (e.g., as in the historical one-drop rule in Southern states; see also Stoler 1995). Second, the
children in the Bloch et al. study were older than they were in our studies.
Thus, Hirschfeld (1996) shows that, for racial categories, fth and sixth
graders show strong social eects not apparent in second graders. We
nd greatest agreement (and a birth bias) in the youngest children in our
various populations. The ideal test case for our hypothesis is a culture
where the adults are not essentialists about ethnicity (see Astuti 1995,
but also Gil-White 2001 for cautions concerning claims about adult conceptions). Here we would still expect that young children would be essentialists (certainly for animals and perhaps for humans as well) even if
adults were not (though adults may be essentialists about animals other
than humans).
Finally, there is reason to expect that reasoning about animal and plant
species may be dierent from reasoning about people. Indeed, in followup studies with the Vezo of Madagascar, Astuti, Solomon, and Carey
(2004) found a reliable birth bias for the youngest children they tested (6
years old) when animals rather than humans were used in the adoption
scenario.
Another objection to our data is that we may be guilty of overinterpreting the results in the sense that projection on the basis of species membership should not be equated with projection on the basis of some essence
(see Rips 2001 for an amplication of this criticism). An alternative view
is that children are employing ideas about causal relations but that they
may have no notion of essence whatsoever (Strevens 2000). Although
this distinction may be subtle, we have discussed it at length elsewhere
(see the Ahn et al. 2001 commentary) and will conne ourselves to a few
remarks in the context of summarizing this section.
5.4
Summary
138
Chapter 5
claim is that from a quite early age children have intuitions that the mechanisms underlying essential causes are biological. The essential causal relations are those involving, for example, birth, biological relatedness, and
internal structure. Just how detailed these notions are and how they are
modied by experience and cultural milieu await further comparative
study.
These same sorts of comparative studies reveal components of biological cognition that vary systematically as a function of cultural milieu and
input conditions (intimacy of contact with nature). The fact that young
Native American children often reason in terms of ecological relations
poses a challenge for interpreting patterns of projection on the induction
task. On the other hand, the prominence of ecological reasoning points
to a component of childrens biology that has scarcely been studied, in
part because this pattern is scarcely evident in developmental studies
with standard populations.
5.5 Implications for Science Education
139
say that rocks and water are alive. Later on, we did another interview
where we constructed more detailed probes where we asked whether
things were alive, died, needed food, needed air, had babies, and so on.
Under these conditions the percentage of Menominee children who said
that plants are alive dropped dramatically (from about 75 percent to
about 30 percent). We think that this reects the conict between cultural
notions about what is alive and the formal notion of alive taught in
science (the fact that young rural majority-culture children were somewhat more likely to say that plants are alive under this detailed probing
suggests that it was not just that the task was confusing).
On a more abstract level, lack of transfer may result from a mismatch
between cultural ways of organizing knowledge and the organization provided in science instruction. As we have seen and will see again later,
Menominee children and adults often reason ecologically and if biology
textbooks use ecology as an organizing principle, compatibility would be
maximized. We examined some of the textbooks being used and found
that when ecology was included it tended to be one of the last chapters
in the book.
In our own research we have seen the consequences of this sort of mismatch. Specically, in the follow-up study alluded to earlier we included
both native or exotic species as bases (on separate daysone or the other
on a given day) and the targets were always a mixture of native (squirrel,
bee, pine tree) and exotic (lion, tarantula, cactus) species. Although our
original objective was to compare reasoning with familiar and unfamiliar
bases to familiar and unfamiliar targets (including the possibility that our
exotic bases might be more familiar to urban than to rural children), the
key result for our Wisconsin populations seems to have been that we dramatically reduced or eliminated ecological reasoning (e.g., an inference
from bees to bears on grounds that bears eat honey or that bees sting
bears). When we looked at the subset of data where the base and target
matched a base and target used in the Ross et al. study, we found no difference for pairs where we had not seen ecological reasoning before and
much reduced generalization for pairs where we had observed ecological
reasoning before. This was true for all Menominee children and for the
older rural majority-culture children. In short, the data suggest that we
failed to access childrens ecological knowledge because the mixed structure of our probes (both native and exotic) only made sense from a taxonomic sense (in other words, it was an ecological jumble). These results
show that knowledge acquired outside the classroom will not necessarily
nd its way into the classroom.
140
Chapter 5
The most abstract level of cultural models, values, and practices may
turn out to be the most important. A number of educators have made
the point that science (instruction) is not acultural or culturally neutral
(e.g., Allen 1998). For her dissertation, Megan Bang (2005), a learning
sciences graduate student at Northwestern University, systematically analyzed practices associated with science curricula and instruction and
compared them with the practices of rural Menominee, rural EuropeanAmerican, and urban Indian parents and children. Her observations and
interviews are consistent with the view that a major challenge to Indian
childrens learning science in school is the mismatch between school and
community practices. Specically, her interviews suggest that there is
a better match between majority-culture parents views of nature and
science instruction than between Menominee parents views and instruction. For example, both the texts and the majority-culture parents tend
to imply that nature is an externality to be exploited, cared for, learned
about, and so on (see Kellert 1993 for orientations toward the natural
world). Menominee parents and urban Indian parents (only some of
whom are Menominees) tend to emphasize that we are a part of nature
and that nature is not an externality (Bang et al. 2005). This a part versus apart orientation may render some aspects of science education
alien to Native Americans in the same way that an economic approach
to a family (e.g., How many dollars is a daughter worth?) is repugnant
to most people.
This brief description does not do justice to Bangs dissertation. She
analyzed practices such as how diagrams presented in traditional science
texts diered from diagrams in texts written by Native American
educators. She also investigated dierences in discourse style between
European-American parents (get to the point) and Indian parents
(provide the context rst and the setting may be as important as the socalled topic or point of the story). She measured the latter by asking
parents to describe the last time they went shing and counting the number of words before sh were mentioned (median of 27th versus median
of 83rd word). Cultural compatibility between community and school
practices was consistently higher for European Americans than for Native Americans (Bang, Medin, and Atran 2007).
The obvious way to test this analysis is to make science instruction
more culturally consonant for Indian children and to see whether their
learning improves. For example, one might structure biology units using
ecosystems as the primary organizing principle and provide contextual
grounding of new information. As a member of our research team, Bang
141
144
Chapter 6
At the end of the chapter we will return to the parallels between species
in biology and intuitive notions of culture in cognitive science. Bearing in
mind these issues concerning stability and change, we turn now to current
stances on how culture and cultural processes should be studied. Each of
them is useful for some purposes and all of them have limitations.
6.1 Culture as Norms and Rules
6.2
145
Cultural Psychology
The recent upsurge of interest in cultural psychology (for one review and
critique, see Oyserman, Coon, and Kemmelmeier 2002 and associated
commentaries) has produced a variety of intriguing ndings and has done
psychology a service by calling attention to cultural variation. Many of
these studies show that knowledge systems previously thought to be universal actually vary widely across the world (for a review, see Cohen
2001). The lesson drawn is that psychologists who choose not to do
cross-cultural psychology may have chosen to be ethnographers instead
(Nisbett et al. 2001, 307). In brief, cultural psychology is succeeding
in divesting academic psychology of implicit and ingrained ethnocentric
biases.
What denes or constitutes cultural psychology? The area draws much
of its inspiration from researchers such as Hofstede (1980) and Triandis
(1995), who sought to characterize cultural dierences in terms of a small
number of relevant dimensions. The project is successful if multiple
sources of evidence converge on the same small set of dimensions. Examples of such dimensions that have received a lot of attention are individualism versus collectivism and egalitarian versus hierarchical social
structure. Other researchers such as Nisbett (2003) have used sociohistorical analysis to derive dimensions of cultural dierences in worldviews or
preferred modes of thought. Examples of these dimensions are analytic
and logical (categorical, axiomatic, and noncontradictory) versus holistic
and dialectical (thematic, no rst principles or excluded middle). In short,
Nisbett and his associates are suggesting that cultural studies must include not only contents per se, but also thinking processes that themselves
may be dierentially distributed across cultures.
Cultural psychologists import the rigor and controls of standard experimental procedure into anthropological concerns, providing clear
identication of the participants, thoughts, and behaviors tested. Cultural psychologists are thus able to systematically exploit anthropological
insights to demonstrate that mainstream psychologys long-held assumptions about cognitive processes can be quite mistaken. In our opinion,
cultural psychology has several limitations. First of all, the leap from statistical regularity in some sample population to the culture may suer
from precisely the sort of reasoning criticized in mainstream psychologys
leap from Americans or Europeans (or, more typically, psychology undergraduates) to the world at large. The same inchoate conception of culture
146
Chapter 6
once used by many anthropologists and still used by most ordinary folk
remains customary in much cultural psychology. In this view, culture
becomes a stable and shared set of beliefs, practices, or strategies to be
studied as yet another population parameter / personal attribute.
This ahistorical, consensual view of culture limits the ability to explain
and understand cultural dierences once they are encountered. In other
words, it is not clear how explanation or interpretation can be extended
beyond simple description. In some cases researchers have been able to
exert some experimental control by priming tendencies to act individualistically versus collectively (e.g., Gardner, Gabriel, and Lee 1999; Briley,
Morris, and Simonson 2000). These sorts of studies reinforce the dimensional analysis and potentially extend its scope. There is always the
risk, however, of circularity in analysis. If priming does not aect some
candidate task measuring individualism versus collectivism, then maybe
the prime was ineective or the task does not entail individualism and
collectivism.
Perhaps we are guilty of prejudging the initial phase of a two-stage
project. In stage 1, cultural psychologists tend to characterize culture as
an external, historically determined system that becomes internalized in
the individual through acculturation (or some other causally opaque
process), either diusely or as some specialized part of the psyche responsible for cultural (or social) cognition. A stage 2 focus on within-culture
variations in modes of thought might illuminate how dierent cultural
institutions shape ways of thinking and vice versa.
For cultural psychologists trained as anthropologists, the focus is on
the extrasomatic or extragenetic nature of culture as an integrated
corpus of external control mechanisms that program individual minds
and bodies, molding them in patterned ways recognizable across individuals (Geertz 1973). We agree that expressions of the human psyche are
profoundly embedded and structured within social and historical contexts, but we dissent from the invited implication of a one-way inuence,
with individual minds being passive recipients of culture.
So far we have followed current practice in using the term cultural psychology to describe the recent upsurge of cross-cultural comparisons by
cognitive and social psychologists. This may be a bit misleading in that
one of the pioneers of the use of the term, Richard Shweder (1990, 87),
uses it to refer to a set of ideas that entail rejecting psychic unity as well
as rejecting the idea of characterizing cultural dierences as variation
along a small number of dimensions:
147
To avoid confusion in nomenclature, we will categorize Shweders approach to cultural psychology under the next framework, context and situated cognition.
6.3
There are alternative views of cultural psychology that call into question the use of standard forms of experimental procedure (methodological behaviorism) as fundamentally awed on grounds that they are
ethnocentrically biased in their focus on the individual mind/brain. Instead of considering cognitions to be embedded exclusively in individual
mindswith culture as just one component of individual cognition
these theorists maintain that human cognitions should be properly situated in cultural-historical context and practical activity (Cole 1996; cf.
Vygotsky 1978). A related concern is that cultural cognitions may be better understood as distributed cognitions that cannot be described exclusively in terms of individual thought processes, but only as emergent
structures that arise from irreducible levels of interactional complexity
involving dierential linking of individual minds in a given population
(Hutchins 1995).
Researchers such as Michael Cole believe that culture cannot be entirely conceptualized in terms of cognitions, belief systems, and the like,
but must instead consider a cultures artifacts (construed broadly enough
to include language). Cole (1996) argues that subjects and objects are not
only directly connected but also indirectly connected through a medium
constituted of artifacts. These artifacts are simultaneously material and
conceptual. One consequence of this view is an emphasis on studying
cognition in context, where cognitive labor may be distributed across
individuals as well as artifacts (such as plumb lines or computers).
Since context includes peoples conceptions of artifacts, it is inherently
relational.
We share some of these concerns raised by the situated view, such as
(1) diculties with standard experimental procedures, including 2 2
148
Chapter 6
Our only disagreement with this is that although cognition has been extensively studied, cognition in context has not; hence we see a continuing
need to attend to mental representations.
6.4
149
Culture as a Superorganism
150
Chapter 6
We believe, however, that understanding cultural formation and evolution depends profoundly on understanding the proximate cognitive
mechanisms involved. Perhaps we can best summarize with an analogy:
macroeconomics is a legitimate eld of study and generates important
insights into economic activity on the basis of assumptions, for example,
of an ecient market (and optimal individual behavior). But these insights
do not in the least undermine microeconomics; and, more to the point,
observations from microeconomics, such as loss aversion (e.g., Kahneman and Tversky 1979) and mental accounting (e.g., Thaler 1985), have
had a signicant impact on macroeconomics.
6.5 The Grammar of Culture
151
The fundamental properties attributed to the human mind, such as binary contrast, were few and simple-minded (or so general and vague as
to be applicable willy-nilly to any phenomena at all).4 This is not to deny
the insights that structural anthropologists garnered into the relationships
between dierent aspects of cultural life within and across populations
(e.g., linking myth, kinship, folkbiology, hunting, and cooking practices,
residential architecture, and so on). Instead, it is only to deny that structuralist theories provide any principled causal explanation concerning
how these relationships might have come about.
More current anthropological views of the grammar of culture are less
committed to a specic theory of the cognitive architecture responsible
for cultural productions than to the belief that culture consists of a
bounded set of rule-bound systems, each with its own grammarlike
structure. A more recent work in linguistic anthropology describes the
culture-as-grammar view as follows:
To be part of a culture means to share the propositional knowledge and the rules
of inference necessary to understand whether certain propositions are true (given
certain premises). To the propositional knowledge, one might add the procedural
knowledge to carry out tasks such as cooking, weaving, farming, shing, giving a
formal speech, answering the phone, asking for a favor, writing a letter for a job
application. (Duranti 1997, 2829)
I-Culture
152
Chapter 6
world over the last thousand years or so). Just as the English language
was shapedand is still being shapedby broad historical events that
did not take place inside a single head (including the Norman invasion
of England and the global Internet), so too has Western European or Chinese or Navajo culture been shaped by complex processes (Pinker 2002).5
If the analogy holds, then psychologys contribution to understanding
culture might best focus on how children grow an I-Culture through
the combination of an innate, biologically specied culture acquisition
device and the exposure to stimuli in the world (or, equivalently, how
individuals are capable at all of participating in E-Culture). As Gary
Marcus (2004, 27) proposes, The very ability to acquire culture is, I
would suggest, one of the minds most powerful learning mechanisms.
This suggests a line of inquiry for culture studies parallel to that taken
by generative linguistics over the past fty years, in which the fundamental guiding questions include: What do people know when they know
culture? and How do people come to acquire culture?
Unlike the structuralist version of culture as grammar, this version
does not prejudge the complexity or variety of cognitive mechanisms
that may be involved in cultural acquisition. Like more current anthropological versions, however, it seems to assume that I-Culture is a bounded
system, or an integrated collection of systems, generated (under appropriate experience) by some articulated set of cognitive principles.
But we contend that there is no systematically bounded or integrated
culture as such. There is nothing at all grammatical or generatively
rule-bound about the relations that connect, say, language, religion,
the nation-state, and science (or that connect the capacities to acquire
knowledge of, and participate in, languages, religions, nation-states, and
sciences). There are only family resemblances to what is commonsensically referred to as culture (or religion or science), but no overarching or integrated structure.
6.7 Generativist (Agent-Based) Models of Culture
Recent advocates of agent-based computational models of cultural phenomena also sometimes borrow self-consciously from the framework of
generative linguistics, where few and nite rules generate rich and complex structures. For most current agent-based models, however, the focus
is not on the generative power of mental mechanisms as such (as it is for
advocates of cultural grammar or I-Culture) but on connectionist and
153
constructivist modeling of how (micro)processes at the level of individual decisions and actions yield macrostructural cultural norms and other
social regularities, such as spatial settlements (Dean et al. 1999), economic classes (Axtell, Epstein, and Young 1999), political alliances (Axelrod and Bennett 1993), voting patterns (Kollman, Miller, and Page 1992),
ecological management networks of religious water temples (Lansing and
Kremer 1993), and so on: To the generativist, explaining the emergence
of macroscopic societal regularities, such as norms or price equilibria,
requires that one answer the following question: How could the decentralized local interactions of heterogeneous autonomous agents generate the
given regularity? (Epstein 1999, 41). In agent-based models of cultural
phenomena there is no central, top-down control over individuals.
Rather an initial population of autonomous heterogeneous agents, situated in a specied spatial environment, begins to interact according to
rather simple local rules (e.g., if agent X manifests behavior A in the immediate, spatially proximate neighborhood of agent Y at time T, then X
and Y will both manifest A at time T1; never attack an immediate neighbor; trade with a neighbor only if that neighbor is red; and so on). Over
time, these concatenated individual interactions generateor grow
macrostructural regularities from the bottom up:
Of course, there will generally be feedback from macrostructures to microstructures, as where newborn agents are conditioned by social norms or institutions
that have taken shape endogenously through earlier agent interactions. In this
sense, micro and macro will typically co-evolve. But as a matter of specication,
no central controllers or higher authorities are posited ab initio. (Epstein 1999, 42)
154
Chapter 6
Summary
155
The framework theory that we endorse draws on insights from a number of the theories we have just reviewed. In particular our focus is on
cultural processes, and consequently our approach is rst cousins of both
the situation and agent-based modeling approaches. We now turn to our
approach and lay out its methodological and conceptual implications.
6.8
Cultural Epidemiology
156
Table 6.1
What is culture?
Cultural change
Within-culture
variability
Cognitive processes
and their relevance
1. Intuitive
Viewed as loss
Viewed as noise
Learning and
memory
Not addressed
2. Cultural
psychology
Not addressed
Not addressed
Inference, reasoning,
perception (cognitive
toolbox)
Not addressed
3. Situated
cognition
Cognitions, belief
systems, and artifacts
Cultures are
dynamic
Variability associated
with dierent practices and artifacts
Distributed, often
context-specic
Not addressed
4. Culture as
superorganism
Emergent system
aecting individuals
Adaptive
Acknowledged, but
not relevant
Depends on
domain-specic
functionality
5. Culture as
grammar
Shared knowledge,
procedures, rules
Not addressed
Not addressed
Mental structures
revealed by crosscultural comparisons
Not addressed
6. I-culture
Bounded rulelike
system organized by
cognitive processes
Not addressed
Driven by E-culture
Universal cultural
acquisition device
Important
Chapter 6
May be emergent
outcome from
perturbation of
steady state
Treated as signal,
key to analyzing
cultural transmission
processes
Imitation, rule
following
Not usually
addressed
8. Cultural
epidemiology
Distribution of ideas,
beliefs, and behavior
in ecological contexts
Cultures are
dynamic
Treated as signal,
key to analyzing
cultural processes
Inference, reasoning,
perception, and
notions of relevance
Important
7. Agent-based
modeling of
culture
157
158
Chapter 6
159
There is little or no detail available in typical normative accounts of social structure in the anthropological literature that would allow evaluation of patterns of individual variation, agreement, and disagreement
within and between groups (but see Aunger 2002 for a counterexample).
Without such detail, normative claims are dicult to verify or falsify. The
overarching reason is simple: anthropologists are typically instructed to
go out into the eld alone for some months orin exceptional cases
some few years and bring back a description of the society studied. The
popular image of the anthropologist with a pith helmet and notebook is
not very far o the mark, only now the pith helmet is a baseball cap or
canvas fedora, and the notebook is a PC. In this situation, there is little
alternative to normative description (except the narratives of antipositivist postmodernism, which do little to foster dialogue with the larger scientic community).
Detailed analyses of the relations between ecology, technology, social
networks, and so forth require large interdisciplinary eorts, over many
eld seasons, at a cost that usually exceeds typical ethnographic eldwork
by one or several orders of magnitude. The pertinent academic and government funding institutions are not set up for this kind of project, and so
the eort is rarely made (for a notable exception, see Henrich et al. 2001).
We have been fortunate to be involved in two such eorts: one in Mesoamerica and another in North America.
A critical case for the importance of cultural selection versus environmental determination comes from a variation on the common garden
experiment in biology. When members of a species have dierent phenotypes in dierent environments, samples are taken from both environments
and replanted in only one. If the dierences still exist, they are probably
genetic (two genotypes); if not, then they are probably environmental
(one genotype producing two phenotypes). Here we use a variation on
162
Chapter 7
In earlier studies (chapter 4), we found that Itza 0 Maya informants consistently appealed to ecological relations on category-based induction tasks.
That observation, coupled with the Itza 0 Maya record of sustainable
agroforestry, suggested to us that there may be a connection between
folkecological models and behavior. In preliminary studies we also found
that Spanish-speaking Ladino and Q 0 eqchi 0 Maya immigrant populations
in the area practice agroforestry in a much less sustainable manner (Atran
and Medin 1997). This situation provided the opportunity to see if understandings of the forest are correlated with action on it. These conjectures
led us to a series of systematic cross-cultural and within-cultural comparisons that are pertinent to a variety of conceptual issues in cognition, decision making, and culture theory (Atran 1999a, 2002; Atran, Medin, and
Ross 2005; Ross 2002).
The Common Setting
163
Figure 7.1
Map of the Maya Biosphere Reserve, El Peten, Guatemala (not drawn entirely to
scale).
Our three groups lie within the Maya Biosphere Reserves ocial
buer zone between 17 10 0 north latitude and Lake Peten Itza to the
south (gure 7.1). Here, vegetation is quasi-rainforest; mean annual temperature is 25 C; mean annual precipitation is 1,6001,800 mm. In the reserve and adjacent areas, Itza 0 comprise a majority of the population in
only 1 settlement, Q 0 eqchi 0 are a majority in 25 settlements, Ladino
immigrants are a majority in 134 settlements, and Ladino Peteneros
(in the area for at least three generations) are a majority in six settlements
(Grunberg and Ramos 1998).
In 1998, San Jose had 1,789 habitants. Most identied themselves as
Itza 0 , although only a minority spoke their native Mayan tongue. Itza 0
represent the last Lowland Maya with demonstrable ties of genealogy
and practice to pre-Columbian civilization in Petens northern forests
(Atran 1993; Atran and Ucan Ek 0 1999; Atran, Lois, and Ucan Ek 0
164
Chapter 7
2004), where the population once exceeded the regions current level by
about an order of magnitude (Culbert and Rice 1990). Nearly all 625 people in neighboring La Nueva are Ladinos (mixed European and Amerindian descent).
As noted in chapter 3, Corozal was settled at the same time by Q 0 eqchi 0
speakers, a Highland Maya group. The Q 0 eqchi 0 now constitute the
largest identiable ethnic group in Peten, while maintaining the smallest
number of dialects and largest percentage of monolinguals (Wilson 1995,
38; cf. Stewart 1980). This set of observations reects the suddenness,
magnitude, and relative isolation of the Q 0 eqchi 0 migration.1
For all groups, people pay rent to the municipality for a farm plot.2
Itza 0 and Ladinos interact often, because their villages are 1 kilometer
apart. All groups practice agriculture and horticulture, hunt game, sh,
and extract timber and nontimber forest products for sale. Each household (about 5 persons) has usufruct rights on 30 manzanas (21.4 hectares)
of ejido land (municipal commons). Farmers pay yearly rent of less than
a dollar for each manzana cleared for swidden plots, known as milpa,
whose primary crop is maize. Yearly crop patterns can vary widely,
owing in part to microclimate and rainfall uctuation. People can hold
plots in scattered areas and can change plots. Plots from all groups may
abut. Hunting is tolerated on neighbors plots, but not access to anothers
crops or trees. Q 0 eqchi 0 live 18 kilometer from both groups; however,
daily buses connect the Q 0 eqchi 0 to the other two groups (who also farm
regularly around Corozal).
To ensure maximum social coverage from our sample, initial informants could not be immediate blood relatives (children, grandchildren,
parents, grandparents, siblings, rst cousins, nieces, nephews, uncles,
aunts), immediate anes (spouses, in-laws), or godparents (compadres).
The distribution view of culture that we adopt leads one to employ sampling techniques most likely to reveal cultural dierences rather than focusing on estimating population parameters.
In the present study we assumed that younger Itza 0 Maya might have
notions of biology that diered from those of Itza 0 elders and that these
dierences might reect an assimilation to Western culture. In addition, Itza 0 is a dying language and few younger Itza 0 speak it. Thus a random sample would tend to hide rather than emphasize the dierences we
were interested in. Instead of randomly sampling, we restricted our initial
sample to Itza 0 -speaking Maya as the best representatives of Itza 0 Maya
culture. Cultural change is a constant and we assume that across time
165
and outside inuences (of varying nature), the knowledge base diers between individuals and across generations. Ultimately, our goal is to trace
the distributions of ideas and beliefs both within and across generations.
In chapter 8 we describe results from a younger, Spanish-speaking Itza 0
sample.
7.2
Historical Geography
166
Chapter 7
The name Itza 0 itself is likely a compound of itz (resin, sap, life
essence, vital substance, hidden power) or the derivative itz 0 in
(brotherhood, lineage, kinship) and (j)a 0 (water) (see Barrera
Vasquez and Rendon 1963, 29). The Classic Maya itz-am (am agentive
marker, or he who is) may denote a shaman who brings itz to the
world (Freidel et al. 1995, 51). Itz-am@na 0 , then, would be the Shamans
House (na 0 or naj house) that enveloped the Classic Maya world.4
By the time Classic Peten Maya civilization collapsed (Terminal Classic
8001,000 AD), the Itza 0 , who may have originated as migratory
Chontal lineages from western Peten, had replaced the Classic CholYukatek lineages in parts of Peten (Seibal) and Yucatan (Chichen Itza 0 )
(Fox 1987; Tourtellot et al. 1992). Arguably, the Itz@a 0 thought of themselves as the Water Wizards, ultimately extending the Classic Peten
cult of Itz-am@na 0 into a Postclassic water cult centered at the Great
Cenote of Chichen Itza in Yucatan (chi 0 mouth ch 0 e 0 en water
hole) (see Barrera Vasquez and Rendon 1963, 2529; Pina Chan 1980;
Porter 1988).
During the Postclassic period, the Itza 0 vied with the Mexicanized Xiw
Maya for control of Yucatan (xiw means grasses, herbs, or reeds
in Yukatek and is a translation of the Nahuatl totellin or Toltec
Reed People). For a time, they kept the peace. In the league
of Mayapan (may calendrical cycle apan Nahuatl for watering
place), the Itza 0 and Xiw agreed to disagree about who had the right calendar for understanding and controlling the course of events in the world.
But the peace was periodically broken when important calendrical events
conicted. This happened, for example, when the ruler of Mayapan sacriced the rain priest of Chichen, Xib 0@chaak (male [strong man] thunder-
167
storm), to demonstrate that the once all-powerful Maya rain god Chaak
was now subordinate to the Mexicanized god Kukul@kan (feathered serpent, a Maya translation of the name of the principal Toltec deity, Quetzalcoatl) (Tozzer 1941, 3234; Barrera Vasquez and Rendon 1963,
147149; Edmonson 1982, 1520).
The Conquest
On March 22, 1697, nine days after the destruction at Lake Peten Itza 0
of the last independent Maya confederacy, Ursua wrote to the King of
168
Chapter 7
169
In fact, the battle apparently was not the relatively bloodless coup that
Ursua suggests. It was a carefully planned campaign by a relatively welltrained European troop armed with galliot, cannons, muskets, and crossbows against an Indian confederacy demoralized by civil war and armed
only with stone and wood. Indeed, as the Mercedarian friar, Diego de
Rivas, later testied, Of the entry made by Don Martn Urzua [I] state
that so great was the number of those who opposed it, so innumerably
many were those killed by the bullets we shot, it seemed like an island in
the lake was formed by the bodies of the dead Indians (AGI Guatemala
345, Parecer de Fray Diego de Rivas, 15 Noviembre 1698, folio 389
recto). Numbers of Itza 0 did ee into the forest, and some likely committed suicide by drowning themselves in the lakenot out of cowardice, barbarous stupidity, or fatalism, but more likely out of a desire not
to be Spanish prisoners or slaves.
Cultural Survival
For both the Spanish and Itza 0 , the immediate post-Conquest years were
marked by a scorched-earth approach to dealing with the other side, virulent epidemics, and starvation. Deaths and desertions among the soldiers and settlers decimated and severly weakened the edgling colony at
Nuestra Senora de los Remedios y San Pablo, Laguna del Itza 0 (formerly
Noj@peten and subsequently Flores). Nevertheless, the Spanish eventually managed to corral about 8,000 Itza 0 and Mopan into eighteen towns,
or reducciones, under the control of the secular clergy (AGI Mexico 3159,
Reporte de 1707 del Gobernador de El Peten). Within a decade, nearly
half of the surviving Indians ed or died from smallpox (viruela) and
other European diseases (AGI Mexico 702, Informe de Luis Coello Gaytan al Rey ledo en el Concejo de Indias, 5 Febrero 1716; cf. Gerhard
1991, 60). This is far from the tens of thousands that Ajaw Kan@ek 0 was
said to govern on the eve of the Conquest (Avendano y Loyola [1696]
1987, 4748; Villagutierre 1701, bk. 5, chap. 11, 332; Cano [1697] 1984,
8; Ximenez 19291931, vol. 2, bk. 4, chap. 68, 210222).
The original mission of San Joseph, founded in 1702, was located between present-day Santa Elena and San Benito, opposite the former Itza 0
island capital. In the wake of the chaos and rebellion of the early postConquest years, a new reduccion of San Joseph was established at the
present site of San Jose sometime before 1750. The earliest marriage record from San Jose dates from that year (5 mayo 1750, Joseph Cante de
San Joseph con Mara Tun de San Andres, Libro de Casamientos de la
170
Chapter 7
Parroquia de los Pueblos de San Andres, San Joseph y San Geronimo, ano
de 1751).
The extant (but partial) record of marriages and baptisms between
1751 and 1788 reveals the most frequently mentioned surnames to be
those of Itza 0 (and allied Kowoj and Mopan) patronyms present in the
area before the Conquest. In descending order of frequency they are (in
their original spelling): Tun (12), Chayax (11), Canek (9), Tz 0 in (7), Chabin (7), Kinyocte (7), Cuouh (6), Chata (6), Tut (6), Quixoban (5),
Xiquen (5), Citcan (5), Cante (4), Chan (4), Puc (4), Kanchan (4), Tzuntecun (4), Tesucun (4) (see Ximenez 19711977, bk. 5, ch. 65; AGI Guatemala 345, Razon Indibidual y General de los Pueblos Poblaciones y
Rancheras de Esta Provincia del Zuiuha Peten Itza Por Declaracion que
han hecho El Rey Canek y El Kimcanek y El Capitan Don Martn Cham
y El Capitan Kulut Coboh, 910 Octubre 1698, 302 verso311 recto).6
A social-network analysis of Itza 0 speakers and their close intimates in
present-day San Jose includes thirty-two surnames (for methodology, see
Atran et al. 1999). Of these, eight are Spanish (Ramos, Lopez, Daz, Cortez, Lines, Garca, Morente, Cinturon), eight are Yukatek families that
came to San Jose between about 1750 and 1900 (Huex, Colli, Vitzil,
Mex, Panti, Tz 0 ul, Mis, Yej), and sixteen are pre-Conquest names from
Itza 0 -ruled territory (Chayax, Cohouj, Chan, Suntecun, Zacal, Tesucun,
Zac, Cauich, Ek, Tut, Xiken, Batab, Cante, Chata, Quixchan, Chuc).
The three social tiers of Peten society lived within a somewhat less violent web of parasitic relationships than elsewhere in Guatemala (Schwartz
1990): Criollo (European origin), Ladino (Spanish speakers of mixed or
Indian ancestry), and Indio (Mayan-speaking natives). Native Creoles
and Ladinos continue to play an important, and sometimes dominant,
role in Petens social and economic life. By contrast, Lowland Maya communities now comprise an increasingly marginalized minority.
In the 1930s, Guatemala dictator General Jorge Ubico instituted a virulent anti-Maya language policy that led the Lacandon (Lakantun) Maya
to ee Peten and resulted in loss of Itza 0 as a rst language in San Jose.
Today in Peten, there are no Lowland Mayanspeaking groups between
the Mopan-Q 0 eqchi 0 town of San Luis in southeast Peten (a few thousand
speakers) and the Itza 0 settlement of San Jose on the northwest shore of
Lake Peten Itza (a few dozen speakers). After some two millennia of recognizable continuity, Itza 0 Maya language and forest culture verge on
extinction.
The intense scholarly attention that pre-Conquest Itza 0 continue to attract contrasts markedly with a lack of interest in the post-Conquest Itza 0 .
171
Forest plants with nutritive value comparable to maize and wheat, like
breadnut (ramon) and palmnut (corozo), were supposedly used more to
sustain pack animals and other stock (Villagutierre [1701] 1985, bk. 7,
chap. 7) than people (cf. Hellmuth 1977, 434). Itza 0 were forced to overextend maize cropping to sustain Spanish overreliance on cereal. Production often fell short of demand. Spaniards cried famine, bewailing idle
barbarian customsuch as relying on root crops, and seeking escape
from hunger and exploitation in the fruits, game, and cover of the forest:
172
Chapter 7
At times the Spanish and the rest of the population were forced (as the documents put it) to eat such foods as ramon . . . camote (sweet potato), yuca (manioc),
name and macal (yams), green plantains, and mamey and sapote fruit. Although
this list may indicate that the Indians persisted in producing a diversity of crops
other than grains despite the Spanish attempt to get them to concentrate on maize
and beans . . . the Spanish had culturally dened nutritional standards that hardly
made these considerations good news. (Schwartz 1990, 55)
In fact, Itza 0 survived conquest and colonization with many preColumbian dietary and medicinal strategies, many of which arguably
date to Classic times (Atran, Lois, and Ucan Ek 0 2004).
7.3 Folkecology and the Tragedy of the Commons
The tragedy of the commons and other similar social and ecological
dilemmas are basically variants of a deep problem in decision and game
theories known as the prisoners dilemma (Hardin 1968; Bromley
1992). Consider a group of n persons who share a common territory of
xed size on which they hunt (or graze) animals. Each hunter (or herdsman) has one of two choices: he can cooperate with the others by not
overhunting (grazing) on the commons; or he can hunt (graze) in a way
that is advantageous to him, but that ultimately results in the overuse
and destruction of the common resource. The second option appears
more rational in the short term: the short-term advantage to one who
overhunts (grazes) (e.g., 1) always outweighs the short-term disadvantage
to him when that disadvantage is equally distributed among the other
hunters (herdsmen) (1=n). If all people cooperate, the common resource
is preserved. But if the rationale of the prisoners dilemma pervades the
camp, no one will have an incentive to cooperate and all will defect.
Field and laboratory studies by anthropologists (Atran 1986; Berkes
et al. 1989), psychologists (Thompson and Gonzalez 1997), and political
scientists (Ostrom et al. 1994) indicate that individual calculations of rational self-interest collectively lead to a breakdown of a societys common
resource base unless institutional or other normative mechanisms are
established to restrict access to cooperators: it is irrational to continue to
act to sustain a diminishing resource that others increasingly deplete. This
so even when peoples basic needs are satised (Boniecki 1977), no
matter how small the group or how informed of the looming tragedy
(White 1994). Earlier observations by our research team, however, suggested that exclusive concern with economic rationality and institutional
norms might not suciently account for behavioral dierences among
groups in Lowland Mesoamerica (Atran and Medin 1997).
173
Although the three groups share a reliance on land and awareness of local
species for survival, analyses of a 3-year period of milpa practice among
twelve to sixteen informants in each group showed striking dierences in
the groups utilization of land and knowledge of species. Data in this section are chiey self-reports elicited from informants, but long-term spot
checks and subsequent measurements (reported below) conrm a correspondence to actual behavior. Reports exhibit no evident bias (e.g., elicited maps of milpa plots depict land cleared in amounts systematically
greater than municipal tax records show; elicited maps also tend to be
more accurate and up to date).
ANOVAs were used to reveal group dierences between Itza 0 , Ladinos,
and Q 0 eqchi 0 , with the Schee statistic ( p < :05) used for post hoc comparisons. The following abbreviations are used with the comparative
statistics: I Itza 0 Maya, L Ladino, Q Q 0 eqchi 0 Maya; M milpa
(swidden plot), G guamil (fallow milpa), R reserve (secondary forest). Analyses revealed no dierences among groups in age, family size,
land available to cultivate, or per capita income from all traceable
sources.8 Q 0 eqchi 0 produce one set of crops per year; Itza 0 and Ladinos
usually produce two (table 7.1). Q 0 eqchi 0 cut and burn forest for new
Table 7.1
N
Itza0 (R)
I (O)
Ladino (R)
L (O)
Q 0 eqchi0 (R)
Q (O)
Other Q 0 eqchi0 *
16
10
16
10
12
10
Crops/ Years of
year
land use
Hectares Years
cleared
fallow
Species/
year
cultivated
2.3
1.8
1.6
2
2.6
2.4
4.1
3.6
3.7
7.8
9.7
3.3
6.4
3.6
6.2
2.5
1.6
4.7
3.6
3.3
3.3
174
Chapter 7
plots every year, compared with an average of 2.3 years for Itza 0 and 1.8
for Ladinos [I; L < Q].9 Dierence in burn frequency produces dierences in destructiveness, independently of need for income.
Itza 0 dier from Ladinos and Q 0 eqchi 0 in reporting: amount of land
cleared for cultivation [F(2; 41) 5:45, p < :01, I < L; Q], fallow length
[F(2; 41) 6:982, p < :002; I > L; Q], and number of species cultivated
[F(2; 34) 13:94, p < :001; I > L; Q] (table 7.1). How can one map these
dierent patterns of use onto an overall measure of destructiveness? For
present purposes we make the strong simplifying assumption that destructiveness (D) is an increasing function of land used per cycle through a
plot (L) and rate of cycling through a plot (R). That is, D L R. To
determine L and R we use A (amount of land a farmer clears), Yc (number of years a cleared plot is used continuously), and Yf (number of years
the land is left fallow). From this it is straighforward to determine that
L A (Yc Yf )=Yc and that R 1=(Yc Yf ). Multiplying these
two terms yields the result that D A=Yc, which is simply land cleared
per year. By this measure, Q 0 eqchi 0 destroy more than ve times as much
forest, but Ladinos less than twice as much, as Itza 0 .10
Q 0 eqchi 0 clear plots in a contiguous S-pattern that rolls through the
forest leaving few trees within or between plots, including hill crowns.
Ladinos intermittently leave trees between and within plots. Itza 0 regularly ring plots with trees, clear rebreaks around valuable trees inside
plots, and change plots in a noncontiguous pattern. This is a strategy apparently shared with some groups of Lowland Lacandon and Yukatek
Maya in Mexico, and there is evidence of a pre-Columbian origin
(Gomez-Pompa et al. 1987; Remmers and De Koeijer 1992). Itza 0 explain
it in terms of forest regeneration: birds, such as the chachalaca (ix
b 0 aach Ortalis vetula), roost in the milpas outer ring (t 0 ool che 0 ) but
y to inner stands (watal che 0 ) to feed on crops and excrete undigested
seeds of outlying trees. Left to fallow, areas around inner stands begin to
emulate and bridge with the outer ring. Birds take undigested seeds of
valuable inner-stand trees, such as the ramon (oox Brosimum alicastrum), to the outer ring, thereby increasing its value for people.
Remote sensing conrms extensive deforestation along Q 0 eqchi 0 migration routes into Peten (Grunberg and Ramos 1998; Sader 1999; Grunberg
2000). Reported patterns of crop diversity, coupled with awareness of
greater ecological complexity and reciprocity between animals, plants,
and people, also should favor regeneration of forest used by Itza 0
versus Ladinos. Despite mutual imagined similarities between Itza 0 and
Q 0 eqchi 0 , on nearly all reported measures, Ladinos are closer to Itza 0
175
176
Chapter 7
Table 7.2
Spanish name
Scientic name
achiote**
aguacate**
almendra**
arroz
ayote
banano
(platano)
cacahuate**
camote
cania
Bixa orellana
Persea americana
Terminalia catappa
Oryza sativa
Cucumis moschata
Musa acuminata
1
2
1
1
8
1
0
0
0
0
6
0
0
2
0
1
8
0
0.033
0.133
0.033
0.067
0.733
0.033
Arachis hypogaea
Ipomoea batatas
Saccharum
ocinarum
Allium cepa
Cedrela mexicana
Cnidoscolus
chaymansa
Sicana odorifera
Capsicum annum
Coriandum sativum
Cordia dodecandra
Cocos nucifera
Canavalia ensiformis
Phaseolus vulgaris
Psidium guava
Phaseolus lunatus
Vigna unguiculaa
Pachirrhyzus erosus
Spondias purpurea
Citrus limonia
Zea mays
Xanthosoma
yucatense
Citrus reticulata
Mangifera indica
Byrsonia bucidaefolia
Byrsonia crassifolia
Citrus aurantium
1
4
1
1
0
0
1
0
1
0.100
0.133
0.067
1
1
1
0
0
0
0
1
2
0.033
0.067
0.100
2
0
1
2
2
2
6
1
0
3
3
2
3
10
5
0
2
0
0
1
6
8
0
4
2
2
1
1
10
1
0
1
1
0
0
3
6
0
0
1
2
1
1
10
4
0.067
0.100
0.067
0.067
0.100
0.367
0.667
0.033
0.133
0.200
0.233
0.133
0.167
1.000
0.333
2
2
2
3
1
0
0
0
0
0
0
0
1
1
0
0.067
0.067
0.100
0.133
0.033
0.067
cebolln
cedro**
chaya
chilacayote
chile**
cilantro**
ciricote**
coco**
frijol abono
frijol
guayaba**
ib**
ix pelon
jcama
jocote**
limon**
maz
makal
mandarina**
mango**
nance agria**
nance dulce**
naranja
agria**
naranja
dulce**
Citrus sinensis
177
Table 7.2
(continued)
Itza0
Spanish name
Scientic name
ocoro
papaya**
payak
pepitoria**
pepino**
pimienta**
platanos
(varios)
sandia
tecomate
tuki**
tsol**
yuca
yuquilla
Hibiscus esculentas
Carica papaya
Discorea alata
Cucumis sativa
Cucurbita mixta
Pimenta dioica
Musa sp.
1
0
4
1
0
1
5
3
1
0
0
1
1
6
3
0
1
0
5
1
4
0.233
0.033
0.167
0.033
0.200
0.100
0.500
Citrullus anatus
Lagenaria siceraria
Annona purpurea
Cucurbita pepo
Manihot esculenta
Maranta
arundinaceae
Pouteria mammosa
1
2
1
1
1
2
2
0
0
0
1
1
0
0
0
0
2
0
0.100
0.067
0.033
0.033
0.133
0.100
0.033
9.7
6.4
6.2
zapote
mamey**
Mean
tree cover shows the same pattern (gure 7.2), with eects of Group
[F(2; 81) 6.17, p :003; I > Q; L(marginal)], Location [F(2; 81)
75:08, p < :0001; R > M; G], Group Location [F(4; 81) 3.43,
p :01; M: I(marginal) > Q; G: I > Q; L(marginal); R: I > Q]. For total
land cleared (M G), Itza 0 dier reliably from Q 0 eqchi 0 and Ladinos.
Group dierences cannot owe to base-rate dierences in species frequency, given the adjacency of parcels across groups.
Soil analysis also suggests that Itza 0 agroforestry is least harmful and
most productive. All soils are moderately alkaline with no signicant
group dierences in pH or organic matter (Atran et al. 1999). Dierences
are most apparent for (normalized) measurements of phosphorus and
nitrates. Neither is abundant in geological materials of limestone regions
and their availability represents limiting factors on life-support systems
(Rice 1993).
Each soil sample was rated on a scale of 1 to 22 as a joint function of
texture (sandy clay loam < clay loam < silty clay loam < sandy clay <
clay < silty clay) and structure (small grain < medium grain < large
178
Chapter 7
Figure 7.2
Biodiversity (number of tree species) and tree cover (square meters per hectare) as
a function of ethnic group and location type (error bars: 95 percent condence
interval).
179
grain < small block < medium block < large block). The best soil ( 1)
is sandy clay loam composed of small granular structures that become
neither too hard when dry nor too compact when wet to prevent water
and root penetration. The worst soil ( 22) is silty clay structured in large
blocks, which become rock hard when dry and extremely compact when
wet. This scale reects the fact that not all possible combinations of texture and structure were present. Physical character of soils was analyzed
by ANOVA: Group (I; L; Q), Location (M; G; R) Level (1 10 cm,
2 20 cm). Only Level proved signicant [F(1; 162) 11.37, p .001;
1 < 2]. There were no reliable between-group dierences for any location.
Averages for each group across all locations fell within the range of clays
with block structures (I 14.1, L 16.9, Q 14.0). These are able to
hold water and x phosphorus, but become unworkable and impede root
growth during very dry and wet spells (frequent in Peten). Erosion and
lack of tree cover magnify the eect.
Phosphorus and nitrate levels were analyzed using Group
Location Level ANOVAs. Phosphorus showed eects for Location
[F(2; 162) 25.67, p < :0001; M > G; R], Level [F(1; 162) 18.86,
p < :0001; 10 cm > 20 cm] and Group Location [F(4; 162) 3.79,
p .006; M: I; L > Q; R: L > I]. Itza 0 dier from Q 0 eqchi 0 in the upper
milpa level (p < :05), where phosphorus is most abundant and useful to
new plant growth. Nitrate levels show eects of Group [F(2; 162)
11.42, p < .0001; I(marginally) > L; Q], Location [F(2; 162) 6.44,
p .002; M > G) and Group Location [F(4; 162) 2.87, p .02;
M: I; L > Q; G: I > L; Q]. For total land cleared (M G), Itza 0 dier
marginally from Ladinos, signicantly from Q 0 eqchi 0 .
Overall, Itza 0 have the highest milpa and lowest reserve scores for
phosphorus, indicating greater phosphorus storage by plants in reserve
with more available for release in milpa (gure 7.3).13 High levels of
phosphorus in milpa arise from burning; however, intense heat volatilizes
nitrates essential to leaf formation. Thus, higher phosphorus levels should
be correlated with lower nitrate levels (and perhaps less foliage cover in
the long run). But for Itza 0 , and to a lesser extent Ladinos, the reverse is
true (gure 7.4). Interrelated factors allow Itza 0 to enjoy relatively high
phosphorus and nitrate levels. Itza 0 cultivate more varieties of nitrogenxing pole beans that climb maize stalks than do Q 0 eqchi 0 or Ladinos.
Q 0 eqchi 0 and Ladinos weed only once shortly after planting. Itza 0 weed a
second time before maize has owered and leave the weeds as mulch. Intense rainfall at this time favors bacterial decomposition of mulch, which
releases nitrogen (also phosphorus, potassium, and magnesium). Finally,
180
Chapter 7
Figure 7.3
Itza 0 tend to light smaller and more dispersed res to clear land, and to
protect valuable trees with rebreaks 2 m around in width. (A side eect
is that the less intense heat causes less volatilization of nitrogen.)
In sum, physical measurements generally corroborate reported behaviors and track their consequences, indicating that Itza 0 practices encourage a better balance between human productivity and forest
maintenance than do immigrant practices. However, signicant dierences in immigrant practices reveal that immigrant Ladinos are measurably closer in behavior to native Maya than are immigrant Maya.
Studies of milpa practices among other immigrant Q 0 eqchi 0 communities
in Peten conrm the patterns in our study (Fagan 2000). In this context,
Itza 0 appear to behave irrationally insofar as their restraint subsidizes
another groups proigacy: the more cooperators produce for free-riders,
the more the free-riding population is able to expand and lay waste.14
Interestingly, Itza 0 tend to believe that Ladinos are more destructive in
their practices than Q 0 eqchi 0 (eleven of fourteen Itza 0 indicated this in
an informal survey we conducted), perhaps because the Q 0 eqchi 0 have
retained corporate rituals that the Itza 0 are now adopting.
181
Figure 7.4
Because analyses revealed no between-population dierences in age, family size, land available to cultivate, or per capita income from all traceable sources, we sought to determine if group dierences in behavior are
reected in distinct cognitive patterns we elicited in folkecological models.
In preliminary studies, we asked informants which kinds of plants and
animals are most necessary for the forest to live? From these lists we
compiled a set of twenty-eight plants and twenty-nine animals most frequently cited across informants (plant kinds were all generic species, except for two life forms, grass and bush).15 The twenty-eight plant kinds in
the study include twenty kinds of trees and 1 ligneous vine counted
among the species in the preceding study. Although these twenty-one species represent only 17 percent of the total number of species enumerated,
they account for 44 percent of all trees in Itza 0 parcels, 50 percent in
Ladino parcels, and 54 percent in Q 0 eqchi 0 parcels. This conrms the salience of the species selected for the folkecology study (table 7.3).
182
Chapter 7
Table 7.3
Plant name
P1*
P2*
P3*
P4*
P5*
FRUIT TREES
ramon
chicozapote
ciricote
allspice
strangler g
P6*
P7*
P8*
PALMS
guano
broom palm
corozo
P9
xate
P10
P11
pacaya
chapay
Sabal mauritiiforme
Crysophilia stauracantha
Orbignya cohune
Scheelea lundellii
Chamaedorea elegans
C. erumpens
C. oblongata
Chamaedorea tepejilote
Astrocaryum mexicanum
P12
P13
GRASSES/HERBS
herb/underbrush
grasses
(various families)
Cyperaceae/Poaceae
P14*
P15*
P16*
P17*
P18*
P19*
P20*
P21*
P22*
OTHER PLANTS
mahogany
cedar
ceiba
madrial
chaltekok
manchich
jabin
santamaria
amapola
P23*
P24*
P25*
P26*
P27
P28
yaxnik
kanlol
pukte
water vine
cordage vine
killer vines
Scientic name
Brosimum alicastrum
Manilkara achras
Cordia dodecandra
Pimenta diocia
Ficus obtusifolia
F. aurea
Swietenia macrophylla
Cedrela mexicana
Ceiba pentandra
Gliricidia sepium
Caesalpinia velutina
Lonchocarpus castilloi
Piscidia piscipula
Calophyllum brasilense
Pseudobombax ellipticum
Bernoullia ammea
Vitex gaumeri
Senna racemosa
Bucida buceras
Vitis tilaefolia
Cnestidium rufescens
(various epiphytes)
183
Table 7.3
(continued)
Ref.
Animal name
Scientic name
A1
A2
A3
ARBOREAL ANIMALS
bat
spider monkey
howler monkey
A4
A5
A6
kinkajou
coatimundi
squirrel
A7
A8
A9
A10
BIRDS
crested guan
great curassow
ocellated turkey
tinamou
A11
A12
A13
A14
A15
toucan
parrot
scarlet macaw
chachalaca
pigeon/dove
Penelope purpurascens
Crax rubra
Meleagris ocellata
Tinamou major
Crypturellus sp.
Ramphastos sulfuratus
Psittacidae in part
Ara macao
Ortalis vetula
Columbidae
A16
A17
A18
A19
A20
A21
A22
A23
RUMMAGERS
collared peccary
white-lipped peccary
paca
agouti
red-brocket deer
white-tailed deer
tapir
armadillo
Tayassu tacaju
Tayassu pecari
Cuniculus paca
Dasyprocta punctata
Mazama americana
Odocoileus virginianus
Tapirus bairdii
Dasypus novemcintus
A24
A25
A26
A27
A28
A29
PREDATORS
jaguar
margay
mountain lion
boa
fer-de-lance
laughing falcon
Felis onca
Felis wiedii
Felis concolor
Boa constrictor
Bothrops asper
Herpetotheres cachinnans
Chiroptera
Ateles georoyi
Allouatta pigra
A. palliata
Potus avus
Nasua narica
Sciurius deppei
S. aureogaster
184
Chapter 7
Figure 7.5
Reported positive plant impact on animals for Itza 0 , Ladinos, and Q 0 eqchi 0 . Animal and plant numbers refer to the ordering of species in table 7.3. The height of
each point refelects the proportion of informants reporting each interaction.
185
Figure 7.5
(continued)
How Plants Affect Animals
The plant and animal kinds are organized into categories used later in the
analysis. Instructions and responses were given in Itza 0 , Spanish, or Q 0 eqchi 0 .
Equal numbers of informants were asked to explain how each plant helped
or hurt each animal, and how each animal helped or hurt each plant.
The procedure had two parts. We asked participants how each plant
aected each animal. The task consisted of twenty-eight probes, one for
each plant. On each trial, all animal picture cards were laid out and the
informant was asked if any of the animals search for, go with, or
are companions of the target plant, and whether the plant helped
or hurt the animal. Questions were pretested for simplicity and easy applicability across cultures. Unaliated animals were set aside. For each
animal, informants were asked to explain how the plant helped or hurt
the animal. Next, they were asked how each animal helped or hurt each
plant. To explore interactions among people and plants, we asked
each informant to explain whether people in their community actually
help or hurt each item on the plant list, and vice versa.
For each task, we used the CCM to determine if a single underlying
model of ecological relations held for all informants in a population. We
186
Chapter 7
collapsed over the dierent ways one kind might help or hurt another,
and the dependent variable for each pair was whether the plant or animal
in question helped, hurt, or had no eect on the other kind. Agreement
across informants was determined by whether their answers match or
mismatched for each pair. To establish consensus, all tasks involved a
minimum of twelve participants from each group, with equal numbers of
males and females. Data were adjusted for guessing (Romney et al. 1986).
Finding consensus justies further study of groupwide patterns. Analyses
of residual agreement were used to reveal group dierences.
Results on plants helping animals are summarized in gure 7.5. Each
of the three groups produced a distinct model on the forest ecology task.
Two results are apparent on how participants see plants aecting animals:
(1) Itza 0 and Ladinos show a highly similar pattern of relations, and (2)
Q 0 eqchi 0 perceive many fewer relations, which tend to be a subset of those
seen for the other two groups. The overwhelming majority of interactions
within each group involved plants helping animals by providing them
food. Plants providing shelter to animals was also a common response.
An ANOVA for plants helping animals showed Q 0 eqchi 0 reporting on
average many fewer relations (46.8) than either Ladinos (163.2) or Itza 0
(187.5), who did not dier from each other.16 Itza 0 and Ladinos showed
a large overlap for which plants help which animals: r (I; L) .82 versus
r (I; Q) .42 and r (L; Q) .54.
A large cross-group consensus emerged. Often all Q 0 eqchi 0 reported no
eect, making the modal answer no eect. Thus, Q 0 eqchi 0 responses
drive the overall consensus. Given this situation, residual analyses are
more eective than simple measures of interinformant agreement in
revealing cultural models. We analyzed a 3 36 residual agreement matrix. For each of thirty-six informants (twelve in each group) there were
three measures: average residual agreement of that informant with members of the same group and that informants average residual agreement
with members of each of the other two groups. Within-group agreement proved reliably greater than across-group agreement.17
Itza 0 and Q 0 eqchi 0 have greater within- than between-group residual agreement. Ladinos show higher within- than between-group residual agreement vis-a`-vis Q 0 eqchi 0 , but do not share more residual
agreement with one another than with Itza 0 . This nding is consistent
with the idea that the Ladino model for plant-animal relations is a subset
of the Itza 0 model. One distinction between Itza 0 and Ladinos was the
latters tendency to generalize the benecial eect on animals of economically and culturally important plants, such as mahogany (the prime
187
wood export) and ceiba (Guatemalas national tree) without apparent justication (Atran et al. 2002). Relations noted by Q 0 eqchi 0 were basically
subsets of those reported by other groups. Overall, Ladino and Itza 0
models converge on how plants help animals. The Q 0 eqchi 0 model is a
severely limited subset of the Itza 0 and Ladino models.
Animals Affecting Plants
Reports of how animals aect plants yielded even larger dierences (gure 7.6). Q 0 eqchi 0 report too few interactions (only 10 out of 812 possible
relations) for consensus analysis. Itza 0 and Ladinos show strong crossgroup consensus, but also greater residual agreement within than between groups. Negative reports of animals hurting plants occur with
equal frequency (8.0 percent of cases by Itza 0 , 8.2 percent by Ladinos).
Ladinos report few relations of animals helping plants. For example, Itza 0
are 4 times more likely to report positive interactions and 3.4 times more
likely to report reciprocal relations (a plant and animal helping each
other).
With respect to positive relations, Itza 0 report that classes of animals
dierentially aected classes of plants, whereas Ladinos do not. To illustrate, plant kinds were collapsed into four categories (fruit, grass/herb,
palm, and other), as were animal categories (arboreal, bird, rummager,
and predator). An ANOVA reveals a plant-by-animal interaction for
Itza 0 but not for Ladinos,18 (1) arboreals were much more likely to interact with fruit trees than with other plant groups, (2) birds were also most
likely to interact with fruit trees, but had moderate levels of interactions
with palms as well, (3) rummagers interacted primarily with grasses/
herbs, and to a lesser extent with fruit trees, and (4) predators showed
few if any interactions with plants.19
On a qualitative level, although both groups acknowledge that animals
have a large impact on fruit trees, Itza 0 dier from Ladinos in understanding these relations. In their justications of plant-animal relationships, Ladinos almost always see animals as harming plants by eating
fruit. Itza 0 justications reveal a more nuanced appreciation of the relationship between seed properties and processing: if the seed is soft and
the animal cracks the fruit casing, the animal is likely to destroy the seed
and thus harm the plant; but if the seed is hard and passes through the
animals body rapidly, the animal is apt to help the plant by dispersing
and fertilizing the seed (Atran and Medin 1997).
A more detailed examination of positive and negative relations reinforces the view that Itza 0 and Ladinos are attending to the same relations
188
Chapter 7
Figure 7.6
Reported positive and negative animal impact on plants for Itza 0 and Ladinos.
Animal and plant numbers refer to the ordering of species in table 7.3. The height
of each point reects the proportion of informants reporting each interaction.
Figure 7.6
(continued)
189
190
Chapter 7
but interpreting them dierently. The two groups have essentially the
same model of negative relationsfor no pair was there greater than a
.40 mean dierence in endorsing a negative relation. For a given animalplant pair both positive and negative relations could be reported. For
both groups there was a reliable, positive correlation between reporting a
positive relation for a pair and reporting a negative relation for that same
pair.20 The group dierence is mediated by the fact that Itza 0 are much
more likely to report a positive relation, even when a negative relation is
also present. We will examine this dierence in more detail when we consider mechanisms of cultural transmission.
These ndings suggest a complex Itza 0 folkecological model of the forest, wherein dierent animals aect dierent plants, and relations among
plants and animals are reciprocal. As Bartlett (1936) and Lundell (1937)
noted when carrying out the rst systematic ecological surveys of Peten,
native Maya (Itza 0 ) awareness of local ecological associations served as
remarkably detailed and accurate guides to subsequent scientic identication and analysis. On a qualitative level, the Ladinos appear to be operating under a dierent cultural model. In a preliminary interview where
we asked Ladinos how animals help plants (thus presupposing that they
do) the typical response was, Animals dont help plants; plants help
animals. Ladinos also possess a relatively elaborate model, but relations
are more unidirectional and less specic. Q 0 eqchi 0 acknowledge a much
reduced role for plants, and almost no role for animals in the folkecology
of the forest.
Human-Animal Relations
191
and practice are intermediate. Itza 0 agroforestry thought and practice encourage a potentially sustainable balance between human productivity
and forest maintenance.
Human Impact
Ladino
Q 0 eqchi0
artisanry
housing
food
medicine
rewood
cash
ornament
other
27
25
21
7.5
5.7
5.0
1.6
7.3
housing
food
rewood
artisanry
medicine
cash
ornament
other
24
22
17
15
8.5
5.0
0.9
8.2
food
housing
cash
artisanry
rewood
medicine
ornament
other
30
25
17
15
10
2.6
0
0
192
Table 7.5
centrality
plant
impact
centrality
plant
impact
centrality
plant
0.64
0.62
0.48
0.11
0.11
0.20
0.40
0.36
0.09
0.05
0.30
0.25
0.20
0.13
0.09
0.34
0.17
0.07
0.06
0.16
0.37
ramon
chicle
mahogany
cedar
ciricote
xate
madrial
ceiba
allspice
guano
chapay
amapola
pacaya
corozo
broom palm
grasses
jabin
chaltekok
manchich
santamaria
herbs
0.75
0.58
0.58
0.55
0.50
0.42
0.33
0.33
0.17
0.17
0.17
0.08
0.00
0.00
0.00
0.00
0.13
0.14
0.18
0.25
0.25
0.20
0.15
0.10
0.16
0.47
0.61
0.64
0.12
0.36
0.25
0.14
0.30
0.29
0.22
0.17
0.00
0.14
0.01
0.11
0.25
0.06
guano
pacaya
xate
ceiba
mahogany
chicle
ramon
allspice
madrial
grasses
cedar
ciricote
amapola
broom palm
cordage vine
yaxnik
pukte
chaltekok
santamaria
herbs
water vine
0.33
0.25
0.08
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.08
0.08
0.08
0.04
0.09
0.07
0.21
0.17
0.08
0.07
0.05
0.04
0.02
0.02
0.02
0.01
0.01
0.01
0.00
0.00
0.00
0.15
0.03
0.01
madrial
broom palm
grasses
ramon
amapola
chapay
allspice
herbs
mahogany
pacaya
santamaria
water vine
jabin
pukte
yaxnik
ceiba
corozo
kanlol
strangler g
cordage vine
killer vines
Chapter 7
1.00
1.00
0.83
0.83
0.83
0.75
0.67
0.67
0.67
0.67
0.58
0.58
0.58
0.58
0.58
0.50
0.42
0.42
0.42
0.25
0.17
Q 0 eqchi0
Ladino
0.47
0.28
0.16
0.07
0.01
0.09
0.03
strangler g
yaxnik
pukte
cordage vine
water vine
killer vines
kanlol
0.33
0.44
0.50
0.60
0.67
0.67
0.75
0.09
0.13
0.20
0.00
0.60
0.24
0.06
chapay
jabin
corozo
manchich
strangler g
killer vines
kanlol
0.08
0.25
0.25
0.25
0.58
0.67
0.75
0.00
0.05
0.03
0.01
0.13
0.12
0.09
chaltekok
guano
xate
manchich
chicle
cedar
ciricote
0.08
0.08
0.25
0.33
0.33
0.58
0.58
193
194
Chapter 7
195
Further Ground-Truthing
Itza 0 folkecological models also relate directly to observed behavior. Regression analysis revealed that for Itza 0 , ratings of human impact (the extent to which people report their actions as helping or hurting particular
species) and weed status (factoring out plants considered weeds) predicted
frequencies of trees counted in informant parcels (r 2 .46, p .004, with
both predictors reliable). No comparable relation emerged for Ladinos or
Q 0 eqchi 0 . Regressions also revealed dierent predictors of human impact
on plants for each group. For Itza 0 , ecological centrality (number of associations in a groups consensual ecological model for a given plant) and
combined utility (value of a plant for wood, shelter, and cash combined)
predicted reported human impact (r 2 .44, p < .001, with both predictors reliable). In short, ecological importance and overall utility predicted
which plants the Itza 0 seek to protect, which in turn predicts the plants
encountered in sample plots.24
For Ladinos, cash value was the only reliable predictor of impact, indicating that Ladinos protect plants having cash value. For Q 0 eqchi 0 , none
of these variables predicted impact signature and the (nonsignicant)
correlations were consistently negative, indicating the Q 0 eqchi 0 tend to
destroy valuable plants. In sum, the three groups have very dierent mental models of the forest, and correspondingly distinct patterns of use. Only
Itza 0 seem to have a positive vision of the role of plants, animals, and
humans in helping the forest survive that is based on species reciprocity.
For neither of the other two groups is there a reliable association between
mental models of the forest and patterns of use.
Our tentative line of reasoning is that Itza 0 , and perhaps other native
peoples with a long history of ecological maintenance,25 might not treat
resources as traditional decision and game theory suggeststhat is, as
objects of a payo matrix (extensional items substitutable along some
metric, such as one that assigns monetary value to every object). Instead,
196
Chapter 7
7.7
197
To better undertstand how Itza 0 think about spirits and God we conducted an experiment on their theory of mind. Cognitive and developmental psychologists use the notion of a theory of mind to refer to a
core aspect of folkpsychology, namely, the ability to correctly anticipate
the mental states of other intelligent agents. We showed seven female
and seven male Itza 0 Maya adults a tortilla container and told them,
Usually tortillas are inside this box, but I ate them and put these shorts
inside. We asked each informant in random order what a person, God,
and the forest spirits (arux) would think was in the box. Each informant
(save one man) responded that God had a true belief because, as several
respondents stated, He can see through the basket as if it were transparent. All (except the same man) thought a person coming upon the basket
would have a false belief about its contents. Six men and four women
thought the forest spirits (arux) would know the baskets true contents.
Overall (for men as well as women), mental states of humans also were
perceived as dierent from those of God27 and forest spirits,28 but God
and forest spirits were not signicantly dierent from one another
(Atran and Norenzayan 2004).
To further explore the developmental trajectory of these sorts of beliefs,
we posed a similar but somewhat more comprehensive set of questions to
forty-eight Yukatek-speaking children (twenty-six boys, twenty-two girls)
(Knight et al. 2004). We asked each child in random order what a person,
God, the sun (k 0 in), principal forest spirits ( yumil k 0 ax 0 ob 0 , Masters of
the Forest), and other minor spirits (chiichi 0 ) would think was in the
box. As with American children (Barrett et al. 2001), the youngest Yukatek (4 years) overwhelmingly attribute true beliefs to both God and people in equal measure. After age 5, the children attribute mostly false
beliefs to people but attribute mostly true beliefs to God.29 Thus, 33 percent of the 4-year-olds said that people would think tortillas were in the
container versus 77 percent of the 7-year-olds. In contrast, no signicant
correlation was detected between answers for God and age [r(46) .06].
Collapsing over ages, Yukatek children attribute true beliefs according to
a hierarchy of human and divine minds, one in which humans and minor
spirits are seen as easier to deceive (gure 7.7). Mental states of humans
were perceived as dierent from those of God (Z 3.357, p .001), and
from those of Masters of the Forest and the Sun Deity (Z 1.89, p .06
for both). God is seen as all-knowing, and local religious entities fall
somewhere in between (gure 7.7). In brief, from an early age Lowland
198
Chapter 7
Figure 7.7
Maya seem to reliably attribute to supernaturals cognitions that they believe are dierent and truer than those attributed to humans.
7.8 Implications for Conceptualizing the Commons
199
200
Chapter 7
actions on common-pool resources, but they may not suce. There also
appears to be an important cognitive dimension to behavioral research
on how people learn to manage environmental resources. Valuation
studies suggest that cognition of supernatural agents may serve not only
to guarantee trust and foster cooperation between nonkin, as standard
commitment theories assume (Frank 1988; Irons 1996), but also foster
human interaction with nonhuman resources in relations of indirect reciprocity (Alexander 1987).
It is no surprise that native Maya with centuries-old dependence on a particular habitat have a richer model of forest ecology than
immigrants. But longevity in a given context does not guarantee sustainable agroforestry practices. Our observations suggest that Itza 0 have a
complex of knowledge, practices, and beliefs associated with sustainability. One should be very cautious in moving from correlations to cause
we do not know whether any one component of the Itza 0 belief system is
either necessary or sucient to support Itza 0 practices. Given that important proviso, the extent to which knowledge, values, and beliefs about the
forest spirits reinforce each other is remarkable.
It is also surprising that Ladino immigrants, who share no evident tradition with native Maya, come to measurably resemble them in thought
and action. As we will see in the next chapter, network analyses reveal reliable but noninstitutionalized linkages that allow socially well-connected
Ladinos access to Itza 0 forest expertise.
Summary
201
202
Chapter 7
Table 7.6
Scientic name
Reference number
jeketzol
kiche0 ak0 ach
saqb0 in
kuk
kej
pich0
jalaaw
kaqkoj
q0 uq0
uut0
pu0
che0 jej
selepan
sosol
mukuuy
k0 uch
iq0 b0 olay
imul
kaqik0 anti
ch0 ojix
sotz0
xalaw
tz0 unon
rax k0 aj
aaqam
sis
yuq
Ortalis vetula
Tyassu sp.
Mustela frenata
Sciurus sp.
Odocoileus virginianus
Picidae sp.
Agouti paca
Felis concolor
Pharomachrus mocinno
Columbidae sp.
Penelope purpurascens
Picidae sp.
Ramphastos sulfuratus
Coragyps atratus
Columbidae sp.
Buteo sp.
Bothrops sp.
Lagomorpha sp.
Micrurus sp.
Felis pardalis
Chiroptera sp.
unidentied bird
Trochilidae sp.
unidentied green snake
Dasyprocta punctata
Nasua narica
Mazama americana
A01
A02
A03
A04
A05
A06
A07
A08
A09
A10
A11
A12
A13
A14
A15
A16
A17
A18
A19
A20
A21
A22
A23
A24
A25
A26
A27
203
Table 7.7
Scientic name
Reference
number
Human
impact
Ecological
centrality
pach0 aya
turans
saqi tul
tzurmuy
ike
peenz
sepres
ch0 lin
tem che0
mes che0
k0 iib0
chaj
tza0 aj
sub0
oqob0
tz0 inte0
j0 k 0 l
ch0 ut
ji
aam che0
Poaceae/Cyperaceae
Prunus persica
unidentied vine
Annona sp.
Capsicum annum
Pimenta dioica
Thuja orientalis
Citrus sinensis
unidentied tree
unidentied tree
Chamaedorea tepejilote
Pinus sp.
Venonia leicarpa
unidentied tree
unidentied tree
unidentied tree
unidentied epiphyte
unidentied tree
Quercus sp.
unidentied tree
P13
P15
P18
P16
P14
P10
P05
P07
P11
P12
P20
P01
P06
P19
P03
P08
P09
P17
P02
P04
0.92
0.92
0.92
0.83
0.75
0.58
0.5
0.5
0.5
0.5
0.5
0.42
0.42
0.42
0.33
0.33
0.33
0.33
0.25
0
0.1
0.25
0.16
0.2
0.12
0.19
0.41
0.24
0.34
0.36
0.29
0.51
0.33
0.36
0.55
0.3
0.17
0.15
0.54
0.44
204
Chapter 7
Figure 7.8
Reported positive plant impact on animals, and positive and negative animal impact on plants, for Highland Q 0 eqchi 0 . Plant and animal numbers refer to species
listed in tables 7.6 and 7.7.
205
Figure 7.8
(continued)
migrations into the Lowlands show little concern with maintaining forest
biodiversity. Under protection from Dominican clergy for centuries,
Highland Q 0 eqchi 0 institutionally managed their own highly commensalist and intense forms of cultivation. When land was scarce, Q 0 eqchi 0
migrated into the Peten Lowlands, often for the short term.
Other Q 0 eqchi 0 communities that immigrated into Peten and adjacent
areas of Belize both before and after the civil war behave similarly to our
study group (Carter 1969; Fagan 2000). When environmentally related
economic diculties arise (e.g., banana blight, hurricanes, and so on), immigrant leaders may send delegations to sacred places in the Q 0 eqchi 0
Highlands to seek aid and redress from Highland spirits (see Schackt
1984). But our immigrant Q 0 eqchi 0 do not concern themselves with Lowland spirits or consult Itza 0 . When we asked why they fail to consult Itza 0
about the forest, the Q 0 eqchi 0 often remark that they do not feel the need
to seek out or placate Lowland spirits as long as the Q 0 eqchi 0 remain true
to their ancestral deities.31
One conclusion from these ndings is that sacred values, per se, are not
enough for sustainability. At the very least, a combination of rich ecological models and sacred values may be required. We do not know how this
combination plays out under rising population densities, continuous
environmental degradation, or even loss of traditional language and
knowledge. We are currently exploring these issues in cross-generational
206
Chapter 7
Lacandon Maya
Our studies with the Lacandon Maya were mainly concerned with intergenerational change among the men of the two adult generations living in
the community of Mensabak. The rationale for this focus was twofold.
First, given the distributional view of culture, we might explore withinculture dierences that go beyond expertise eects. Second, within-culture
dierences among Lacandon Maya hold particular interest. Members of
the second generation of married adults were born or grew up in villagelike communities, whereas fathers and grandfathers originated from dispersed households and settlements.
To elicit mental models of folkecology, a freelisting task was used
again to generate a list of species most important for the forest to live.
The CCM produced a single factor solution (ratio eigenvalue 1:2 16.4,
variance 87 percent) indicating the existence of one underlying model
shared by all informants. Nevertheless, members of the two adult generations separate on both their rst and second factor loadings, further suggesting two submodels for the members of the two generations.
One dierence is that members of the rst generation report signicantly more interactions than members of the second generation. The rst
generations consensual model exhibits a clear structure that separates the
animals and plants along lines of taxonomy and habitat (Ross 2002).
This separation is based on specic plant-animal relations that involve
certain physiological characteristics, such as having a hard shell (as we
found with Itza 0 ).
We used the expert networks to explore possible links between relation
to an expert and levels of agreement. Second-generation adults clearly
regard rst-generation adults as experts; however, we could nd no evidence for a relation between proximity to an expert and ecological knowledge. In addition we failed to nd reliable residual agreement between
fathers and sons. As a whole, these data only describe expertise dierences among the members of the Lacandon community. The dierences
appear to reect a marked shift in recent history, namely, a dramatic
change in settlement patterns that distanced the younger generation from
forest life. The expertise dierences observed cannot be easily explained
as dierences in amount of factual knowledge. Rather, dierences in per-
207
ceived goals (the need to tend the forest) and learning landscape (like
Itza 0 , Lacandon elders say that one learns by walking alone in the
forest) lead individuals to draw dierent conclusions from the same
observations.
Other research suggests that overall patterns of knowledge and behavior
among native Lacandon Maya versus Tzeltal and Tzotzil Maya (born to
immigrant families from the Highlands that had settled into the area) resemble that of Itza 0 versus Q 0 eqchi 0 immigrants (Nigh 2002). The fact
that these descendants of immigrants have lived all their lives in the forest
indicates that mere personal exposure to the local ecology is not a deciding factor in sustainability of practices.
7.11
The puzzle for decision theory is: How do people manage limited resources in a sustainable manner without apparent institutional or other
obvious normative constraints to encourage and monitor cooperation?
Multiple factors are involved in explaining the stability of representations
within and across our study populations.
In the area of decision making and the commons, the prevailing view
at least in economics and political sciencehas been that human behavior
in society is driven by self-interest, mitigated by institutional constraints.
Like models of induction that rely on universal similarity, abstract decision models employ a homogeneous notion of utility, where content
biases and protected values simply are annoying. For example, protected
values are annoying because their utility may be hard to measure
(Baron and Spranca 1997; Ritov and Kahneman 1997), and content
biases only serve to distort rational calculations of utility (but see Tanner
and Medin 2004 for a contrasting view).
Thus, analyses of the commons problem may appear to be trapped
somewhere between isolated individual interests that lead inevitably to
commons destruction and a focus on institutions that has little need for
cognitive science. To be sure, there is a good body of social science research that identies certain conditions for cooperation in articial experimental situations (e.g., Messick and Brewer 1983; Ostrom 1998), but it is
hard to see how to transfer these ndings to complex, real-world situations such as we nd in Peten and Wisconsin. Furthermore, this body of
research provides no role for content or values other than in terms of fungible (transparently interchangeable) gains and losses. There is no place
208
Chapter 7
Cultural Epidemiology
In chapter 7 we observed large dierences across the three groups of agroforesters. Now we want to probe more deeply into cultural processes
within and between groups. We begin by describing social and expert network data. The goal was to trace variations in knowledge and beliefs with
variations in social distance from others, including experts. This is what
we mean by cultural epidemiology.
8.1
Social network analysis bears out the close relationship in mental models
and forest behaviors between Itza 0 and Ladinos. For each community we
began with six men and six women not immediately related by kinship or
marriage.1 Each informant was asked to name, in order of priority, the
seven people outside of the household most important for your life.
Informants were asked in what ways the people named in this social
network were important for their lives. Some days later each informant
was also asked to name, in order of priority, the seven people to whom
you would turn if there were something that you did not understand and
wanted to nd out about the forest/shing/hunting. Informants were
asked about the kind of information they would seek in these expert networks. After performing these tasks with our initial group of informants,
we used a snowball method to extend these ego-centered networks to
wider patterns of social relations in which they operatethat is, social
networks and expert networks were then elicited from the rst and last
persons named in the social network. When either the rst or last person
named was not available, we interviewed either the second or sixth person
named. The decision to establish network closure after a single iteration
(one roll) was based on previous studies suggesting that in practice it is
rarely necessary to seek direct ties involving more than one intermediary.
210
Chapter 8
The three populations markedly dier in their social and expert network structures, with dierent consequences for the ow of information
about the forest.2 To fully appreciate the dierences, one needs to consider the social and expert networks in conjunction. But rst consider social networks, which are summarized in the graphs shown in gure 8.1.
As we will see, the circle graph of the Ladino network shows a clear
gender division of the community. The Itza 0 social network is the most
diuse, and the two clusters correspond, not to gender but to the two major moieties (subgroups, often organized in terms of practices such as intermarriage rules). The MDS that accompanies each circle graph in gure
8.1 uses the pattern of connections as a similarity metric and scales the
similarity relationsin this casein a two-dimensional space (while preserving the connections shown in the circle graph): for the Itza 0 , the rst
dimension dierentiates the two major moieties and the second reects
the presence of clans and families that are connected in dierent ways
to the major moieties but not each other. The Q 0 eqchi 0 form the most
socially interconnected community. They show a dense, highly interconnected network, with no dominant individual or subgroup. The following
analysis supports this impression.
To measure social connectedness we used the group l-level. Lambda
sets describe the line connectivity of nodes (Borgatti et al. 1990). The
line connectivity for a pair of nodes is equal to the minimum number of
lines that must be removed from the graph in order to leave no path between them. For our purposes, we use the group as the Lambda set and
calculate the minimum connectivity (rather than the average) as an index
of overall group cohesiveness. In the case of the Itza 0 , for example, if one
link is removed, then at least one member of the group becomes separated
from the group. For Ladinos, two links must be removed to separate at
least one member from the group. For Q 0 eqchi 0 , four links must be
removed to separate at least one member from the group. Level 5 (l 5)
includes 90 percent of Q 0 eqchi 0 , 21 percent of Ladinos, and only 10 percent of Itza 0 . In short, the Q 0 eqchi 0 are the most and the Itza 0 the least
interconnected.
Representations of the Q 0 eqchi 0 show a dense, highly interconnected
social network, with no dominant individual or subgroup. This redundant
social structure favors communal and ceremonial institutions that organize accountability, and that are richer among Q 0 eqchi 0 than among Itza 0
or Ladinos. Only Q 0 eqchi 0 practice agroforestry in corporate groups:
neighbors and kin clear and burn each households plot, kin groups seed
together, and the community sanctions unwarranted access to family
Cultural Epidemiology
211
Figure 8.1
Social networks for Itza 0 , Ladinos, and immigrant Q 0 eqchi 0 . Circle graphs (top)
and multidimensional scaling (bottom) are alternative representations of the
same data sets.
212
Chapter 8
Cultural Epidemiology
213
that the Itza 0 community is currently divided into two social factions: one
dominated by Y, the other by VW and TN. Person V is Ws father and
person T is Ns father. V and T are also cited as two of the top three Itza 0
forest experts. Y and V head two families that have continuous genealogical links to pre-Conquest Itza 0 clans of the same name, Chayax and
Tesucun.
8.2
Information Transmission
One possibility compatible with the Itza 0 social structure is that ecological
knowledge is directly transmitted from socially well-connected forest
experts, such as Y, to other Itza 0 . To evaluate this possibility, we analyzed patterns of residual agreement in relation to social and expert network structure. We wanted to see how other individuals and subgroups
compared to our most cited expert Itza 0 informant, Y. We focused exclusively on the nonempty cells, because knowledge transmission should primarily take the form of noting an existing relation, not the absence of
relations. Analyses within the Itza 0 sample revealed little residual agreement and this agreement was inconsistent across dierent tasks. In no
case could we discern relationships between residual agreement and social
or expert network proximity. In other words, Itza 0 social structure does
not show evidence of specic pathways for learning about the forest, at
least for our sample.
While this lack of evidence may be an eect of high repetition of information among all informants and the generally high level of expertise,
there is an alternative scenario to learning about the forest that is more
consistent with independent discovery than direct social transmission of
ecological knowledge. When asked how they learn about the forest, Itza 0
mostly claim to acquire knowledge elicited in our tasks by walking
alone in the forest they call the Maya House. For Itza 0 , diusely
interconnected social and expert networks suggest multiple social pathways for individuals to gain, and for the community to assimilate and
store, information about the forest. Cultural stories, values, and the like
may bias the interpretation of experience in dierent ways: for example,
a bird or monkey eating fruit may be seen to be transparently harmful
by Q 0 eqchi 0 and Ladinos, but interpreted by Itza 0 to be helpful.
Our analysis of cultural models and social transmission is frankly speculative, but it does have some testable consequences. The general idea is
that a persons cultural upbringing primes that person to (1) pay attention
214
Chapter 8
Cultural Epidemiology
8.3
215
In line with evolutionary models of social learning, one may assume that,
when in doubt or ignorance about a certain domain of activity vital to everyday life, people will look to those with knowledge in order to emulate
them (Boyd and Richerson 1985; Henrich and Boyd 1998). One promising strategy would be to rst look for knowledge from those to whom
deference (respect) is shown by others (Henrich and Gil-White 2001). At
least in many small-scale societies, knowledge bearers tend to be elders,
political leaders, economically well o, and so on. In the Itza 0 case, forest
experts are experts in a variety of relevant domains (e.g., soils, trees, hunting, collecting plants), elder males, and former political town leaders.
Informally we have noted that Ladinos today continue to express doubt
about their forest knowledge and express a desire to acquire knowledge from the Itza 0 . Apparently, the most respected and socially wellconnected Ladinos attend to those Itza 0 to whom other Itza 0 defer; and
these Ladinos, in turn, become subjects of emulation and sources of
knowledge for other Ladinos.
But how do Ladinos go about obtaining the relevant knowledge without initially knowing how it is relevant? Observers do not have direct access to the deep knowledge they wish to emulate, only to surface signs
or markers of that knowledge.
First of all, there may be some transmission of norms or ruleswe
have witnessed Itza 0 showing Ladinos how to control burns when clearing
land for milpa and discussing where to plant dierent species of fruit
trees. Other learning factors may be involved in transmitting knowledge,
including normative prototypes and narratives, but in fairly indirect ways.
Thus, Ladino prototypes and stories of Itza 0 experts as forest wizards
may share little actual content with the normative pronouncements and
narratives of the Itza 0 themselves. Moreover, Itza 0 disavow teaching the
Ladinos anything about the forest. The line of reasoning that follows is
frankly anecdotal, but one that should motivate further research.
For present purposes, of greatest relevance is evidence suggesting Ladinos may be acquiring knowledge through dierent isolated examples that
trigger inferential structures to support generalizations. Our data suggest
that two distinct forms of inference may aect mental models of the
forest: (1) inferences from general knowledge of ecological relationships,
such as whether relations are positive or negative and where in the forest
they are likely to occur, and (2) category-based induction over ecological
and taxonomic groups. Consider the rst form of inference. A Ladino
216
Chapter 8
Cultural Epidemiology
217
218
Chapter 8
Cultural Epidemiology
219
8.4
220
Chapter 8
statistically reliable distributions of cognitions and behaviors. Social network analyses further reveal that members of each of these communities
almost never include people identied with other communities among
their intimate social relations. These commonsense cultural constructs
allowed us initially, if roughly, to distinguish populations that subsequently revealed themselves to consist of reliably distinct cognitive, behavioral, and social-relational patterns. In addition, we showed where
patterns cross over populations.
Our use of the commonsense notion of culture to initially distinguish
populations is not a case of circular reasoning, because patterns of similarities and dierences within and between populations could not be predicted in advance. As with Darwins use of the commonsense notion of
species, which rst focused his attention, subsequent discoveries revealed
only rough correspondence between the commonsense construct (species)
and historically contingent patterns of evolution (more or less geographically isolated and interbreeding populations). Darwin continued to use
the commonsense idea of species (Wallace [1889] 1901, 1) only as a
heuristic notion that could ground attention as diverse and often inconclusive scientic analyses advanced, while denying it any special ontological status or reality (Atran 1999b). Likewise, intuitions about what
constitutes a culture may continue to help orient research, but should
not be mistaken for a nal or correct framework of explanation.
Thus, although our ndings reinforce separating the Q 0 eqchi 0 from the
other groups, our ndings also strongly suggest that Itza 0 and Ladino
populations are beginning to merge on a number of dimensions (male expertise, residential proximity, converging use of Spanish as the principal
language, and so on). In fact, researchers in the area from several disciplines now refer to a more generalized Petenero culture that joins Itza 0
and Ladinos but still generally excludes the Q 0 eqchi 0 (Schwartz 1990).
This merging of cultures probably owes in large measure to Ladinos
assimilating certain Itza 0 values, or at least learning and adapting some
of the important ways that Itza 0 imbue forest life with meaning, whereas
perhaps Itza 0 are shedding some of the olderand in a commercial age,
outwornvalues that rendered the forest their sacred Maya House.
The distributional view of culture implies a methodology that departs
in distinct ways from traditional anthropology, where the intrepid explorer becomes immersed in culture X and returns to report how Xers
think and behave: We are not interested in what A or B may feel
qua individuals . . . we are interested only in what they feel and think qua
members of a given community [where] their mental states receive a cer-
Cultural Epidemiology
221
222
Chapter 8
We have argued that the study of culture is the study of variation within
and across populations. From the theoretical perspective of decision
theory, our work extends the tragedy of the commons to situations involving multiple groups transmitting knowledge and belief systems in distinct
patterns that can be traced to historically conditioned conceptions of nature and social and expert network distance. This same perspective is also
relevant to application: cultural cognitions aect environmental values,
decision making, and prospects for human survival under conditions of
global change (Atran, Lois, and Ucan Ek 0 2004).
Our previous work on category-based induction enabled us to identify
inferential patterns in acquisition and transfer of folkbiological knowledge. We saw that these patterns reect both universal constraints on
biological inductions and culturally specic biases in construal and organization of information. The view of culture as a patterned distribution of
cognitions and behavior set the stage for addressing issues of learning, inference, and transmission of information, within and between cultural
groups.
To explain cultural consensus and stabilization of folkecology we focused on the likely causal roles of (historically conditioned) mind-internal
mental models for representing and processing cultural cognitions, and
on mind-external ecological factors (including social arrangements) for
transmitting cultural cognitions. We found that statistically consensual
cultural cognitions and practicesor cultures for shortinvolve complex causal chains that go both inside and outside the mind. These chains
irreducibly link individual minds and their internal representations with
psychophysical interactions between individuals and their external environment (including interactions with other individuals).
By targeting the microprocesses (including evolved cognitive aptitudes
like the folkbiology module) by which these cultural chains form, we have
sought to account for regularities and recurrences in sociocultural macrophenomena. This contrasts with standard explanations in social psychology that seek to account for individual cognitions in terms of the
inuences of sociocultural macrophenomenawhere the causal character of inuence is left opaque. Our approach also runs counter to
customary accounts in anthropology, sociology, economics, and political
Cultural Epidemiology
223
226
Chapter 9
227
1. The fur trade era Soon after their initial arrival, French traders
established the fur trade. The French fur trade era was perhaps the most
benign threat to the Menominee, at least at rst. The area of the upper
Great Lakes was extremely suitable for such an endeavor, given the existence of large tracts of forests intersected by rivers that would allow easy
transportation and inhabited by an indigenous population that was very
skilled in both hunting and trading (Beck 2002, 27). The trade in furs
introduced sweeping changes to the local economy.
The French had no designs on Menominee lands, preferring to prot
from trade. The French had congenial relations with Indians and intermarriage between French traders and Menominee women was not
unusual. But the fur trade also changed the social structure of the
Menominee. As the closest sources of beaver pelts began to run out,
Menominee men had to travel longer distances in search of furs. As a
consequence, the village structure gave way to smaller hunting groups or
bands for signicant portions of the year. It is dicult to estimate the impact of this shift on Menominee culture, but it no doubt disrupted the
clan system to some extent. This may be one factor causing uncertainty
about the precise structure of the Menominee clan system in the precontact period (e.g., Homan [1896] 1970). The French and Indian war
marked the end of French inuence in this area.
2. Disease Contact with whites meant contact with diseases for which
native peoples had no natural immunities. The eects were devastating
throughout the New World and the Menominee were not spared. A
1736 survey put the warrior strength of the Menominee at 160 (Keesing
[1939] 1987). It is hard to get precise estimates, but it is probably safe to
assume that at least half the Menominee population fell victim to disease.
3. Religion The Menominee were fairly receptive to early Catholic
(French) missionaries, in part because Menominees did not conceive of
religious beliefs as mutually exclusive. Later in the nineteenth century,
under American inuence, there were pressures to give up traditional
beliefs. At the point that the Menominee people settled into permanent villages on the Menominee reservation, around midcentury, the Menominee
who were Christians tended to settle in and near what is today Keshena
and South Branch; the so-called pagans (following more traditional Indian religion) settled further away, north and west of Keshena. The present village of Zoar, established in 1881, continues to be a center for
traditional spiritual practices.
Although most Menominees today are Catholics, many Menominees
integrate this faith with beliefs and practices associated with their
228
Chapter 9
traditional religion. In addition, one can see that the church has, in some
cases, incorporated Menominee symbolism. For example, the church in
Keshena depicts corn on its exterior.
4. Land and treaties The French and the British came to Menominee
country for trade. After the War of 1812, British inuence all but disappeared, to be replaced by Americans who wanted morethe land itself.
Settlers poured into the Midwest in general and Wisconsin in particular.
The treaty era had begun.
The idea that land is something that one could own was alien to Indian
sensibilities. Land was not an object or commodity, but a relational entity, like a grandmother. That is, you wouldnt, couldnt, and shouldnt
sell land any more than you would sell your grandmother. Consequently,
Indian conceptions, at least for initial treaties, were that they were receiving gifts for usufruct or use privileges, not giving up land ownership
(again, because land is not the sort of thing that one owns).
As is commonly known, treaties meant displacement to less hospitable
areas, often several thousand miles away from home. When the implications of treaties became clear to Indian tribes, they began to resist them,
but it soon became clear that the U.S. government was going to take the
land, treaty or no treaty. Although the treaties usually included payment
for the land, the amounts were at best symbolic and usually only applied
in order to provide a facade of legality. (In a treaty of 1831 Menominee
received $285,000 for 3 million acresabout 8 cents per acre.)
Americans made their rst appearance at Green Bay in 1815, and as
they had done previously when the British replaced the French in 1761,
the Menominee did their best to adapt to the new situation. In March
1817 at St. Louis, they signed their rst treaty with the United States.
A second treaty took place in 1831, and a third in 1836. In the latter
Menominee surrendered another 4.2 million acres.
As Wisconsin statehood approached in 1848, Chief Oshkosh and the
Menominee were pressured into ceding their remaining Wisconsin land
in exchange for a 600,000-acre reservation on the Crow Wing River in
Minnesota. Chief Oshkosh sent scouts to the Crow Wing area and they
reported back that the land was desolate and inhospitable. This served
to strengthen the Menominee resolve to resist displacement from their
(remaining) land. Although the Menominee signed the treaty in 1848,
they nevertheless refused to move to Minnesota. Instead they pled for a
reservation in Wisconsintheir homeland. This was nally granted after
many negotiations and interventions from outsiders, missionaries (includ-
229
ing Father Bonduel, for whom the town of Bonduel is named), and settlers (Beck 2002, 179).
The nal settlement, the Wolf River treaty, was signed in May 1854
and established a reservation for the Menominee in northern Wisconsin. The terms of the treaty of 1854 assigned twelve townships to the
Menominee reservation. The agreement was changed in 1856, when the
Menominee (again, under pressure) ceded two townships for the purpose
of creating a separate reservation for the Stockbridge Indians (who had
been displaced from the East Coast). The newly established Menominee
reservation contained 235,000 acres of their original homeland, a tiny
fraction of the lands they had originally occupied. Figure 9.1 shows the
size and location of contemporary Menominee lands.
5. Timber interests Before, during, and after the treaty era, businessmen
and settlers, aided by government ocials, were eager to exploit Wisconsins timber resources, including those on tribal lands. Todays
Menominee forest is not the result of a selection process that left the
Figure 9.1
230
Chapter 9
231
As we write this, the Bureau of Indian Aairs is being sued for literally
billions of dollars for misuse and misappropriation of tribal trust funds. A
major problem hampering the investigation is the BIAs inability to nd
relevant records in its oces.
8. Boarding schools In the 1920s and 1930s there was an attempt to
assimilate Indians to white culture by taking the Indian out of them.
Many Indian children were forced to attend boarding schools at considerable distances from their homes. Children from dierent tribes and
language groups were thrown together, and students were punished for
speaking their native language. Among other things, one product of this
experience for many Native Americans was distrust of and disidentication with the educational system. Many parents avoided speaking the
Menominee language in front of their children, so that their children
would not be punished when they went to school. Menominee became a
dying language. Over the past few decades, however, the tribe has made
extensive eorts to keep the language alive.
9. Termination The most recent threat to the Menominee as a people
was the Termination Act of 1954. This ended federal recognition of Indian rights and privileges guaranteed by treaties. No new names were to
be added to the Menominee tribal roll after 1954. The eort was, in effect, to legislate the Menominees out of existence. Concretely, when the
plan went into eect in 1961, it meant that the Menominee were subject
to a crushing nancial burden. The tribal clinic and hospital soon closed.
The tribal court system was ended, and the sawmillthe primary business on the reservationhad to focus on eciency rather than maximizing Menominee employment. Menominee County became a pocket of
poverty almost overnight and the tribe struggled to provide services.
Termination also led to a compromise on sovereignty. In 1959 a development group, Menominee Enterprises, was created by a general council
meeting. About ten years later, in an attempt to raise funds and establish
a tax base, Menominee Enterprises agreed to a project that involved creating a human-made lake, Legend Lake, and selling lakeshore lots to
nontribal members.
But the Menominee people again proved resilient. The Legend Lake
development triggered a storm of protest over the loss of land and the energy released by it fueled a broad, sustained eort to restore tribal recognition. One key development was a 1968 U.S. Court of Claims ruling that
the Menominees did not relinquish their hunting and shing rights when
the tribe was terminated. This ruling set the stage for a Supreme Court
ruling that Termination did not abrogate treaty rights. Land sales ended
232
Chapter 9
in 1972, though nontribal members continue to own land and live around
Legend Lake.
The story of the restoration movement would make a book in itself
(see Pero 1982; Davis 2000; Beck 2002). After more than a decade of
struggle, the tribe succeeded. The Menominee Restoration Act was
signed by President Richard M. Nixon in 1973. Under the terms of
the Restoration Act, the Menominee lands would be protected under
the treaty trust relationship with the U.S. government. Furthermore, the
Menominee achieved their greatest degree of autonomy since before
the treaty era.
10. Casinos and continuing threats to sovereignty The relatively recent
development of gaming compacts and tribally operated casinos has constituted an economic boon for many tribes, including the Menominee. It
has allowed the tribe, among other things, to create and support the College of the Menominee Nation.
But there is also a strong downside to casinos. One is the misperception
that Indians have all been made wealthy from casino prots. In the case
of the Menominees, the highest per capita payment over the past ve
years has been $100 and for the past two years, when the tribe has had
some budgetary problems, the per capita has been zero (per capita is the
amount that each enrolled member receives). Compare this with the per
capita of well over $1,000 that every Alaskan citizen receives each year
(the gure was $1,540 in 2002) from the State of Alaska from oil and oil
pipeline revenues.
A more signicant downside is the pressure from members of the state
legislature and the preceding governor (Tommy Thompson) to the eect
that Wisconsins tribes should give up certain treaty rights in exchange
for the renewal of gaming compacts. In the case of the Ojibwe or Chippewa, the aim has been to force them to relinquish their hunting and shing rights in the ceded territories. (The Ojibwe did not give up their
hunting and shing rights when they ceded the land that makes up much
of northern Wisconsin; consequently, they are entitled to hunt and sh on
o-reservation lands.) Of most relevance to the Menominee tribe is the attempt to remove any legal standing Menominees may have for ghting
o-reservation threats to the Wolf River, such as the proposed Crandon
mine some 3040 miles north of the reservation. Even without the mine,
it is a reality that the absence of industrial production on the reservation
provides no protection from acid rain and mercury contamination brought
in from the outside.
233
Summary
9.2
Start with the rising sun and work toward the setting sun, but take only the mature trees, the sick trees, and the trees that have fallen. When you reach the end of
the reservation, turn and cut from the setting sun to the rising sun, and the trees
will last forever.
Menominee leader, usually identied as Chief Osh Kosh
Looking at satellite images, one can readily locate the Menominee reservation because of the salience of the forest. This is true even on the northern border, where the reservation borders the Nicolet National Forest.
The Menominee forest is richer in larger trees, has a richer mix of species,
and is denser than the Nicolet forest. It also has a higher per acre production of timber and maintains a higher number of board-feet of commercial species (Davis 2000, 15).
The forest not only provides timber for the Menominee, but it is also
a place for hunting deer, bear, and other game, as well as for gathering
nontimber products such as berries, ginseng, and other medicinal plants.
As in the past, hunting also provides the Menominee with an important
food source. Menominee are avid hunters and both bow hunting and rie
hunting are practiced on the reservation. Bear are used for their meat,
their fur, and also for medicinal purposes. The bear population appears
to be healthy on the reservation. Given the isolation of the Menominee
forest islands, no moose are found and the wolf population is small.
Hunting is regulated by the tribe. Individuals who want to hunt have
to apply for a deer tag for each deer they want to kill. Limits are based
on population numbers (actual and ideal), just as is the case o the
reservation.
The reservations other major natural resource is water. Over 300 miles
of trout streams, over forty lakes, and several rivers provide plenty of
opportunities for shing, a major activity for Menominees. Menominee
sh during all seasons and in all dierent kinds of waters and styles.
234
Chapter 9
Trout are very important target sh, as are the dierent species of pansh. Just as for hunting, the tribe sets its own shing regulations.
The average annual income on the reservation is about two-thirds of
the income o the reservation (median family income in 2000 on the reservation was about $26,000, compared to $38,000 in Shawano County).
Even these gures are somewhat misleading, because the average household size on the Menominee reservation is considerably larger than in
Shawano County (3.74 versus 2.51). About 36 percent of the people on
the reservation live below the poverty line compared with about 8 percent
in Shawano County. The age distributions are also substantially dierent
on and o the reservation. Slightly more than half of the Menominee are
under 18 compared with a corresponding gure of about 25 percent for
Shawano County.
Shortly after restoration, the Menominee developed their own constitution and a tribal government was established in 1977 (Ricciuti 1997, 30).
Today this government consists of eight tribal legislators and a chairperson. Members are elected in three-year intervals. The tribe has its own
court system and police department. Besides Menominee Tribal Enterprises, which manages the forestry business, the tribe receives revenues
from its casino operation. Still, the revenues from gambling are barely
enough to keep up with the costs of running an autonomous government.
In many respects Menominee today appear to adopt a rural majorityculture lifestyle. Quite a few Menominee seek employment outside the
reservation, and some even send their children o the reservation to
Wisconsin public schools in neighboring Shawano County. Similarities
to majority-culture individuals can also be found with respect to outdoor activities such as shing, hunting, snowmobiling, and so on. These
activities are very popular in the general area and also attract many
tourists each year. Still, these supercial similarities conceal large underlying dierences reecting history, values, and distinct perspectives on
nature.
The visible resultsthe maintenance of the forest in light of large-scale
deforestation in the surroundings as well as constant economic incentives to the contraryare quite impressive and provide a literal groundtruthing of Menominee respect for nature. Perhaps the critical element
is not values per se but rather what happens when one set of values
come into conict with another (Kempton, Boster, and Hartley 1995).
The Menominee have demonstrated time after time that they are unwilling to trade their forest, lakes, and rivers for money or higher employment. Many Menominee believe that the Menominee would not be the
235
Menominee without the forest, in the same way that a farmer cannot be a
farmer without land.
9.3
236
Chapter 9
Intercommunity relations are multifaceted and almost all generalizations would be misleading. Here are just a few of these facets. First,
there is some back-and-forth when it comes to employment. Menominee
Tribal Enterprises awards logging contracts on a competitive basis that
includes a point system rewarding contractors who either are themselves
Menominee or hire Menominee loggers. Many majority-culture adults
from the surrounding community attend the College of the Menominee
Nation, which itself employs an ethnically diverse faculty. The Menominee
casino has provided an economic stimulus not only to Menominee County
but also Shawano County. In summary, there is a fair amount of interchange between the peoples of the two counties that is mutually benecial.
Not everything is positive. As we noted earlier, there is resentment on
the part of majority-culture sportsmen of tribes setting their own hunting
and shing regulations, especially when those regulations are more liberal
than those of the state of Wisconsin. Typically these dierences are seen
as resource depleting. There is also a fair amount of real and perceived
prejudice and stereotyping. Menominee complain about being followed
about when they shop in certain stores, a form of ethnic proling. Some
Menominee who fail to be recognized as Indians by majority-culture people tell stories about hearing more overt expressions of prejudice.
In short, there is a mixed picture of majority culture and Menominee
relations. On the one hand, for issues like supporting U.S. troops in Iraq
or ghting a proposed mining operation that threatens the Wolf River,
the communities become one. On the other, there is always the potential
for sharp divisiveness on issues where the interests of the communities are
not perceived as shared. As we will see, even where the two groups have
common superordinate goals, diering mental models and associated
values and practices may lead to misperception and intergroup conict.
9.5 Mental Models
237
238
Chapter 9
Standard sorting techniques and other probes were used to explore each
groups categorization of local sh species. On a spontaneous sorting
task involving forty-four local species of sh, fteen Menominee and fteen majority-culture experts showed overall consensus,1 but also reliable
group dierences. An analysis of variance on residual agreement (Nakao
and Romney 1984) revealed greater within- than between-group agreement and a signicant population by within- versus between-group interaction. The form of this interaction is that only the Menominee informants
displayed reliably greater within- than between-group residual agreement.
In short, it appears that the Menominee and majority-culture informants
share a common cultural model of sh but that the Menominee, in addition, share a somewhat distinct conceptual organization of sh.
Additional analyses indicate that the Menominee consensus contains
an ecological component absent in the sorting of majority-culture experts.
Multidimensional scaling (MDS) yielded a dimension for Menominee
experts that correlates with sh habitat. In addition, Menominee experts
were reliably more likely than majority-culture experts to mention habitat
in their explanations for the sorts they created. This dierence reects
preferences for organizing categories rather than knowledge dierences
239
240
Chapter 9
relatively more ecological justications (40 percent), and fewer goalrelated (29 percent) and taxonomic-morphological (31 percent) justications. The majority-culture nonexperts, by contrast, gave fewer ecological
justications (16 percent) and more goal-related (43 percent) and
taxonomic-morphological (41 percent) justications. Whereas the pattern
of Menominee justications is robust across the two levels of expertise,
the majority-culture pattern changes, such that, with expertise, majorityculture informants come to give more taxonomic-morphological and
fewer ecological and goal-related justications. Some majority-culture
experts explicitly mentioned how their orientation toward shing had
changed over the years, moving away from the stereotypic sportsmans
model that targets shing contests or going for the trophy sh.
Menominee shermen tend to take an ecological orientation to
conceptualizing sh. They also commonly express the attitude that every
sh has a role to play and are less likely than majority-culture shermen
to think of sh in terms of positive (gamesh) or negative (garbage sh)
utility. Although both groups report wanting to save sh as a resource,
the goal of conservation is supported by dierent strategies in the two
groups. Menominees have a strong do not waste ethic and tribal
regulations prohibit the wanton destruction of any sh, even the sh that
Wisconsin DNR regulations (WDNR 2002) refer to as rough sh.
Majority-culture experts, in contrast, tend to focus on catch-and-release
as a conservation strategy.
These group dierences might be best described as dierent orientations, with majority-culture experts being more goal-oriented and
Menominee more ecologically oriented. Although both groups presumably share the goal of preserving sh as a resource, we wondered how
the dierences in orientation might be reected in values and attitudes
toward dierent shing practices.
Summary
241
242
Chapter 9
Table 9.2
Menominee
Majority culture
Black sucker
Bluegill
Bluntnose minnow
Brook trout
Brown trout
Gar
Largemouth bass
Muskellunge
Northern pike
Perch
River shiner
Smallmouth bass
Sturgeon
Walleye
Yellow bullhead
12.0
5.4
13.0
2.2
2.6
14.5
4.4
8.7
6.5
5.9
11.9
7.3
9.1
4.6
11.1
12.1
4.8
12.0
6.8
7.9
13.7
6.4
5.4
5.5
5.9
11.0
6.5
8.6
2.9
10.0
The consensus rankings of the fteen species of sh are summarized in table 9.2. Lower numbers correspond to higher rankings. Somewhat to our
surprise we found a strong cross-group consensus3 between Menominee
and majority-culture shermen. Nonetheless the overall consensus was
coupled with reliable group dierences.4 These dierences indicate the existence of clear cultural submodels. Looking at the actual rank ordering,
we nd the biggest dierences with respect to brook and brown trout (average ranking for Menominee: 2.1 and 2.6; average ranking for majority
culture: 6.2 and 7.2; the dierences for both sh are highly signicant5) as
well as with respect to the muskie and walleye (average ranking for majority culture: 5.3 and 3.0; average ranking for Menominee: 8.7 and 4.5;
only the dierence for muskie is signicant6). Although both groups value
all of these sh, Menominee assign a higher value to the two trout species,
and majority-culture shermen preferentially value muskie and walleye.
These modest dierences should not distract us from the wider consensus
that exists between the two groups. For example, the six sh ranked lowest are not only the same for both groups but are even placed in exactly
the same order. The overall cross-group correlation of rankings was .81.
243
Table 9.3
Goal rankings
Majority culture
Menominee
Goal rankings
Self
Group Ogroup
Self
Group Ogroup
2.4
3.4
3.2
3.7
4.6
3.3
3.0
4.4
3.0
4.6
3.2
3.7
4.6
4.6
2.6
2.9
3.9
3.9
2.4
2.9
3.7
1.9
2.6
4.2
2.7
5.0
2.4
3.2
1.7
5.4
3.1
3.3
2.5
4.5
3.5
3.5
Note: The column labeled self gives the average of the individual ratings in Experiment 1 broken down by group (rst and fourth column). Group refers to
predictions for ones own group and Ogroup to predictions by the other group
in Experiment 2. Lower numbers correspond to greater importance.
Goal Rankings
The average rankings for the six goals are summarized in the rst and
fourth columns of table 9.3. Given that each goal has been endorsed by
several experts in earlier interviews, we had no strong reason to expect consensus either within or across groups or even a clear ranking
of goals. Indeed, we did not nd consensus across groups or for either of
the two groups individually. Despite the lack of consensus, we were nonetheless able to detect group dierences on specic goals. Menominee
experts give signicantly higher importance to shing for food,7 while
majority-culture experts tend to place higher value on shing as a challenge to outsmart the sh (the latter dierence was marginally signicant8). These data are in line with the observation that majority-culture
shermen tend to see shing as a contest or sport.
Ratings of Practices
244
Chapter 9
Table 9.4
Catch-and-release only
Spearsh suckers/carp
Spear walleyes/northern
Trophy mounted
Bluegill/sunsh food
Northern/muskie food
LM/SM bass food
Setpoles for trout
Selling sh
Keep undersized sh
Fishing contests
Fishing spawning beds
Suckers for sturgeon
Cull for biggest limit
Using sh nders
Exceed limit for family
Giving all sh away
Menominee
Self
Group Ogroup
4.5
5.9
1.0
4.4
6.0
2.9
3.4
2.5
2.0
1.6
4.9
3.0
3.6
2.9
5.9
2.9
3.9
3.1
5.9
1.1
4.8
6.4
3.2
3.8
2.6
1.7
2.6
4.9
4.2
3.8
3.2
5.9
4.1
3.3
3.3
5.0
5.4
3.9
7.0
6.8
6.8
3.3
1.5
2.4
5.0
4.3
2.9
3.5
4.8
5.3
5.1
4.3
3.4
1.2
6.7
6.2
5.6
5.5
1.1
1.7
2.2
6.7
4.0
3.0
3.4
6.5
2.8
3.2
4.7
5.1
4.0
4.9
6.9
6.1
6.5
3.3
1.3
2.3
4.3
2.8
2.1
1.9
3.6
5.2
4.8
3.1
6.2
5.9
4.6
6.5
5.7
5.7
6.1
4.4
4.8
4.4
5.9
5.7
4.9
6.1
6.0
4.8
of table 9.4. We found modest overall consensus9 and reliable group differences. The main group dierences are as follows. Menominee experts
gave higher ratings to catching bass, northerns, and muskie for food
(strong approval versus neutral), higher ratings to someone taking more
than the limit to feed their family (modest approval versus modest disapproval), and higher ratings to spearshing walleyes (neutral versus strong
disapproval). The Menominee shermen were sharply divided on spearshing walleyes and the average reects an equal mixture of strongly
positive and strongly negative ratings. Menominee opposed to it say that
females are being speared and their spawn wasted. Menominee in favor
of spearing say that they only spear the males. (Data to be reviewed later
on Ojibwe spearshing suggests that about ten males are speared for
every female walleye speared.)
245
Summary
9.8
246
Chapter 9
247
Table 9.5
Menominee
Majority
Rated
Menominee
Majority
Menominee
Majority
Black sucker
Bluegill
Bluntnose minnow
Brook trout
Brown trout
Bullhead
Gar
Largemouth bass
Muskie
Northern pike
Perch
River shiner
Smallmouth bass
Sturgeon
Walleye
11.9
4.7
13.4
1.6
2.3
10.9
14.4
4.2
8.6
6.0
7.5
12.4
7.5
9.6
4.3
11.7
6.7
13.4
5.5
5.6
10.3
13.9
2.8
4.6
5.5
7.8
12.9
7.3
8.7
1.4
11.5
6.1
12.2
2.4
3.1
10.6
13.9
6.8
7.9
6.2
6.2
12.6
7.5
5.4
4.8
11.9
4.0
13.2
5.1
7.1
10.0
14.1
4.3
5.9
5.4
6.2
12.5
7.2
10.9
2.1
culture shermen would rank walleye and muskie very highly (and more
highly than the Menominee), and that also was the case.
Overall, each group very accurately anticipates the rankings that members of the other group would give. The two groups generally agree on
their rankings (the cross-group correlation was .87), but this accuracy
extends well beyond general agreement. Menominee estimates of majority
values correlate .95 with majority values, and majority estimates of
Menominee values correlate .93 with Menominee values.
Goal Rankings
The predicted rankings for the two groups are summarized in table 9.3
(p. 243). Again, smaller numbers refer to higher priorities. Columns 2 and
5 give predictions for ones own group, and columns 3 and 6 give the predictions by the other group. Each group was fairly accurate at anticipating the goals of members of their own group. Menominee experts thought
that majority-culture experts would be much more focused on catching a
trophy-size sh than they actually are. Menominee shermen also underestimated the importance for majority-culture experts of shing as an activity to pass down to future generations and for being close to nature. The
248
Chapter 9
The results on values and attitudes came as a shock to us. Predictions for
own group and other group are summarized in table 9.4 (p. 244). Columns
2 and 5 give predictions for ones own group and columns 3 and 6 give
the predictions by the other group for a given group. Table 9.4 indicates
that Menominee experts think that majority-culture shermen would be
more approving of shing contests and getting a trophy sh mounted
than majority shermen report. Table 9.4 also shows that majorityculture experts think Menominee experts would approve selling sh,
keeping undersized sh, shing on spawning beds, culling smaller sh to
get the biggest-sized limit, and using setpoles to catch trout. As we noted
before, such practices are disapproved of by both groups. Majority shermen even believe that Menominee would approve shing for suckers hoping to get a sturgeon on the line (sturgeon are sacred for the Menominee).
A cross-group consensus analysis was conducted to see how well the
two groups agree in their perceptions. That is, do majority-culture and
Menominee experts (1) have the same beliefs about majority-culture
values and attitudes and (2) have the same beliefs about Menominee
values and attitudes? This cross-group analysis reveals consensus for
both majority-culture and Menominee experts only with respect to the
majority-culture responses.10 In light of the individual group consensus
this suggests that the Menominee model of majority-culture experts is in
basic agreement with majority-culture experts perceptions of their own
values and behaviors. This basic cross-group consensus is coupled with
signicant residual group dierences, because members of both groups
dier signicantly on their second factor scores.
Corresponding cross-group analyses with respect to the Menominee response pattern fail to show consensus. This underscores an asymmetry
with respect to cross-group perception. While Menominee and majorityculture experts concur on a model of majority-culture expert values and
behavior, both groups dier widely in their perceptions of Menominee
values. In short, majority-culture models of Menominee are strikingly different from Menominee individual responses and Menominee predictions
for the group consensus. Overall, these data indicate that majority-culture
shermen hold strong, incorrect expectations concerning Menominee attitudes and values.
9.9
249
Sources of Misperception
Where do these misperceptions come from? We believe that these misperceptions come from dierences in specic goals and knowledge organization, reinforced by patterns of media coverage. Dierences in specic
goals can lead to rejection of another groups values and practices.
In an exploratory analysis we have examined relationships between the
sh-ranking task and stereotyping. There is one nal discrepancy between
perception and actuality that we have deferred talking about until now.
The majority-culture sh experts thought that the Menominee experts
would rank the sturgeon more highly than they actually do. The Shawano
dams on the Wolf River prevent sturgeon from being able to reach the
reservation itself. Many majority-culture shermen may be aware of
Menominee eorts to get ladders installed on these dams so that sturgeon
could return to the reservation for spawning. Some may know that sturgeon are considered sacred. So it is not surprising that majority-culture
experts thought that Menominee experts would value sturgeon highly. Indeed, we ourselves were initially surprised that our Menominee experts
did not rank sturgeon more highly. The responses of the Menominee
experts tend to be more pragmatic. A typical comment was we dont
have them on the reservation any more. One expert who is an elder did
not rank sturgeon high because he thinks the meat is too rich.
The overall mean of 5.4 for the majority-culture expectations about
Menominee ranking conceals a great deal of variability, and we decided
to investigate further. Specically, we looked how answers to the values
probe pretending to sh for suckers hoping to get a sturgeon on the
line correlated with beliefs about Menominee valuing sturgeon more
than white shermen. Recall that Menominee disapprove of this practice
but that expert majority-culture shermen as a group thought that
Menominee would approve of it more than their own group does. For
each majority-culture expert we computed two scores: (1) rating for
Menominee approval of pretending to sh for suckers minus the same
anticipated rating for majority-culture shermen and (2) anticipated
Menominee ranking of sturgeon versus expected ranking for majorityculture experts. We then correlated these two scores across our majorityculture sh experts.
One hypothesis is that experts who knew enough about Menominee
culture to know that they value sturgeon would be less likely to think
that Menominee would approve of trying to get sturgeon on their lines
250
Chapter 9
for entertainment. If that were the case then we should observe a negative
correlation between the two scores. The observed correlation was 0.70,
highly signicant and in the opposite direction! Those that thought
Menominee experts would value sturgeon also thought that they would
approve of getting sturgeon on the line for entertainment.
Another way of describing the results is that the majority-culture experts who know enough about Menominee shing values to anticipate
that they would not rank sturgeon highly were also those experts who
judge that Menominee would not approve of pretending to sh for
suckers hoping to have a chance to wrestle with a sturgeon.
We also looked at the correlation between thinking that Menominee
would rank sturgeon high and a combined measure of values and practices associated with stereotyping: (1) selling sh, (2) keeping undersized
sh, (3) culling smaller sh to get the largest bag limit, (4) shing on
spawning beds, (5) using setpoles to catch trout, and (6) pretending to
sh for suckers hoping to get a sturgeon on the line. Again for each
majority-culture expert we took the dierence between anticipated approval by Menominee versus majority experts as our index of stereotyping. The correlation between this index and thinking that Menominee
experts would rank sturgeon comparatively higher was 0.65, which is
statistically signicant. So the correlation holds not only for the item concerning suckers and sturgeon but also for stereotyping as a whole.
Overall, these observations suggest that knowing a bit about
Menominee values in the abstract was not enough to undermine stereotyping, but knowing Menominee shermens specic values was. Of
course, it could be that the judgment that Menominee shing experts
would not value the sturgeon more highly than majority-culture shermen was based on lack of knowledge rather than a specic belief. To address this question, we did a nal correlation analysis.
Recall that majority-culture experts as a group knew that Menominee
place greater relative value on trout. In another analysis we looked at the
correlation between predicting that Menominee would value sturgeon relatively more and knowing that Menominee value trout relatively more.
The correlation was signicant and negative (0.62). In other words, the
majority-culture experts who correctly thought that Menominee value
trout tended to think correctly that Menominee would not preferentially
value sturgeon. Using the six items mentioned previously to get an overall
measure of stereotyping, we nd a reliable negative correlation (0.49)
between knowing that Menominee preferentially value trout and stereotyping. The better the majority experts knew Menominee rankings, the
251
less stereotyping they displayed. We are currently gathering social network data as a converging source of evidence, and so far our data are
consistent with the idea that knowing specic Menominee who sh is negatively correlated with stereotyping. Finally, we examined relationships
between sh ranking and stereotyping. Specically, we look at ranking
of the big ve sportssh (walleye, northern, muskie, smallmouth bass,
largemouth bass). The higher this ranking, the greater the stereotyping
(.50).
9.10
The most striking nding is that the very modest actual dierences in
goals, values, and attitudes are accompanied by massive perceived dierences. Furthermore, the eect is strongly asymmetrical. Menominee shexpert judgment modestly exaggerated the sportsmans model of shing,
but majority-culture judgments of Menominee values are wildly discrepant from stated Menominee values. One explanation that can readily be
rejected is that the Menominee stated values do not correspond to actual
behaviors. Recent surveys of sh populations in lakes and rivers on the
Menominee reservation show that sh populations are healthy and abundant (Schmidt 1995). In short, the Menominee tribe has done a good job
of managing sh as a reservation resource.
We suggest that these misperceptions are mediated by dierences in
specic goals and associated knowledge organization, reinforced by patterns of media coverage (for related analysis of eects of media coverage
see Gilens 1996; Gilliam and Iyenger 2000). The sportsmans model of
searching for trophy-sized sh is common in the media. It is easy to get
the idea that getting a trophy sh is the be-all-and-end-all of shing. Fishing contests on cable television only reinforce this impression. Sporting
magazines are full of photographs of particularly large gamesh that
anglers have caught. Rarely does an article mention someone catching
two 16-inch walleyes and making a nice meal from them (and there certainly would be no photo). It is also important to note that although the
Menominee as a whole tended to have stereotypes about majority-culture
shermen, there were a number of exceptionsthese are group trends
that do not hold for every individual.
If the gap between prediction and reality is large for Menominee predicting majority-culture values, then it is enormous for majority-culture
shermen predicting Menominee values. The fact that they thought that
Menominee would be more approving of spearshing walleyes than they
252
Chapter 9
are is not so surprising. But they also thought, contrary to fact, that
Menominee shermen would strongly approve of virtually every practice
that both groups condemn.
9.11
Cultural Support
253
for every one taken by Ojibwe spearers. The Ojibwe also maintain sh
hatcheries, strip the spawn from any females they spear, and restock in
the same waters where they spear (e.g., in 1998 Ojibwe stocked ceded territory waters with over 26 million walleye fry and more than 700,000
walleye ngerlings; WDNR 2002). Despite these numbers, many sportshermen may balk at the image of spearers taking large female walleyes.
But WDNR monitoring also undermines this image. Sexing of harvested
sh during the 19851999 period shows a breakdown of 83 percent males,
10 percent females, and 7 percent of unknown sex. The average length of
walleyes taken has been 15.5 inches. So the image that best ts is of a 15inch male walleye, not a 25-inch female.
Overall, the most striking nding is the contrast between perception
and reality: despite the strong overall consensus in knowledge, goals, and
values, majority-culture shermen see Menominee as vastly dierent.
These results show that dierences in how groups conceptualize nature
are critical to understanding intergroup conict over resources.
9.12
Conclusions
Our data show that expertise cannot be separated from cultural milieu,
even when people engage in more or less the same activities. The parallels
between the Itza 0 and the Menominee are striking, especially when one
notes that both groups also have sustainable forestry practices. As with
Itza 0 and Lacandon, some Menominee men express the belief that if a
person treats nature in a greedy or wasteful manner then spirits will punish them, and oer tobacco as a prayer of thanks. Cultural paths (in the
sense of reliable distributions of conceptual representations in a population of minds) appear to provide something of a framework theory for
organizing experience. This is seen, for example, in the Itza 0 Maya tendency to see reciprocal relations (animals helping plants as well as being
helped by them) and in Menominee shermens ecological orientation.
These studies reinforce the distributional view of culture. Residual
analysis indicated that expert Menominee and majority-culture shermen
have a shared model, but that, in addition, Menominee shermen have a
distinct model based on salience of ecological relations. These dierences,
coupled with dierences in underlying subordinate goals, can give rise
to dramatic intergroup misperception, even when both groups share the
same superordinate goal of resource conservation. The interaction of culture by expertise in the basis for sorting also suggests dierent developmental trajectories in the two groups. Indeed, our developmental data
254
Chapter 9
10
We have covered a lot of territory but our journey is far from complete.
In this chapter we review our main ndings in the context of discussing
the implications of our results for theory, methodology, and application,
including policy. This summary will be coupled with comments concerning current and future directions.
A Few Clarifications
Before turning to implications, we rst try to avoid some possible misunderstandings by adding a few points of clarication.
Modules
One reviewer of an earlier draft of this book generally liked it, but took
us to task for adopting an extremist position on modularity. Although
we made some changes here and there to reduce the likelihood that we
would be misunderstood, we will say a bit more here. First, we do not
claim that there is a distinct area of the brain dedicated to all and only
biology. Second, we do not claim that the innate propensities for learning
about the biological world necessarily have propositional content. All we
claim is that there is enough there, so that with some minimal interaction
with the world, a number of candidates for universal principles emerge,
including the privilege of the generic-species rank and a presumption of
underlying essence.
In chapter 4 we oered a set of converging criteria (operational guidelines if you will) for a biology module and reviewed a wide range of evidence bearing on these criteria. Others may disagree either with these
standards or with the state of the evidence bearing on them. This is ne
and conforms to our notions about how good science operates.
256
Chapter 10
One of our most trusted colleagues advised us against employing the suggestion that cultures are like species. He oered two reasons: (1) the basis
for an analogy should be grounded in stable, shared knowledge and readers might well have highly varied and incorrect notions about species, and
(2) many invited implications of the analogy would also be incorrect. Perhaps our analogy would have been a bit more precise if we had instead
suggested that peoples misconceptions about culture are like peoples
misconceptions about speciesfor example, that they have universally
shared essential properties. In any event, we will trust the reader to
avoid the rst problem and oer a few examples to address the second.
First, cultures can blend and mix in ways that species almost always do
not (sure, botanists will tell you that oak species are very promiscuous,
but that is relative to other plant species where mixing is much rarer).
Second, the ontogeny of an individual plant or animal is probably much
more stable than the ontogeny of an individual member of some cultural
group.
Units of Analysis
257
Taking Sides
The reader could easily come away with the impression that we have
failed to remain objective or neutral, instead transparently favoring the
Itza 0 Maya and the Menominee. We reject both the framing of this accusation and its conclusion. First, being neutral is not the same as being objective. Generally our goal is to take the perspective (not the side) of
whatever group we are working with and it is not clear what neutral
means in this context. We think objectivity is more likely to emerge from
taking multiple perspectives than from trying to take no perspective.
In short, we would like to be accused of taking the perspective of all
the groups we work with. Even in the case of interviewing candidate suicide bombers and their sponsors (which we do in our new work), we believe there is value in trying to understand the underpinnings for such an
extreme action.
Second, we do not see cultural research as a one-way enterprise where
we administer some interview or test and the participant gets paid with
dollars (or some local currency). The research enterprise is more collaborative and some of our studies are based on suggestions made by participants. It is also not one-way in that the groups we work with may ask us
to help in other ways. So we when are asked to speak at a Rotary Club,
meet with a parents group, teach a course at a tribal college, or help to
establish a rainforest reserve, we generally try to be helpful. And when we
need our horse or car serviced and one of our options is a stable or garage
owned by one of our informants, guess where we take it.
Third, we think that there are some formal obligations when one is
working with groups that historically have been underserved. For example, our work in Wisconsin has been based on an understanding of appropriate research methods for working with American Indian communities.
There is a long history of research in American Indian communities that
has often not been in their best interest (as they see it), a legacy that has
made many native communities suspicious toward research. Over the
years indigenous researchers themselves have worked to develop appropriate methods and criteria for conducting research (Hermes 1999; Smith
1999; Mihesuah 1998; Guyette 1983). There are some general lessons that
have driven the approach to this work.
To begin, all of the literature generally agrees that the participatory
action research (PAR) is the best framework of inquiry. PAR has generally been dened as an integrated approach that relies on the participation of community members to investigate the issues at hand, while
258
Chapter 10
building local skills for the purpose of increasing autonomy. PAR includes the following criteria: elder input; use of traditional language;
community participation in research agenda, sta selection, and budget;
community payo; respect for cultural values; and informed consent
(Hudson and Taylor-Henley 2001). Additionally, when conducting research with reservation communities, investigators must go through the
tribal research approval process; a research board approval from a mainstream institution is not sucient (Lomawaima 2000).
In Guatemala, there is less of a history of research exploitation, at least
in Peten, but it has been no less important to work closely with community leaders in conducting research. Lois, Vapnarsky, and Atran also
helped to organize a language program aimed at keeping Itza 0 Maya
alivefor years government functionaries discouraged use of the language in public and banned children from speaking it in schools through
nes and beatings. The language-program eort was based on community
goals, and not an imposition of researcher values.
Finally, attention to and respect for cultural groups that have been
neglected or oppressed may have some intrinsically benecial aspects. To
the extent that the research fosters cultural identication and dignity, it
may support cultural survival and help people maintain their bearings
even in the worst of circumstances. Count us in on this.
Shallow Ethnography
Although we have made protestations about the perils of parachute researchand solemnly sworn that we have conducted appropriate historical, linguistic, and ethnographic researchit would be misleading to
claim that we have done everything that we could have. Ethnographic research is not like putting on a suit for a wedding where you have only one
chance to get it right. We have recorded many Itza 0 stories and ceremonies (some already reported elsewhere, some to come), but it was not until
we found qualitative dierences in folkecological models that we started
to think about looking for themes related to reciprocity. Similarly, it was
only after discovering Menominee childrens precociousness in ecological
reasoning that it made sense to look for its roots in everyday Menominee
cultural practices.
We are trying to make two related points here. One is that ethnography
is never theoretically neutral and the other is that it is an iterative process.
So our ethnographic research may be shallow with respect to some puzzles we would like to address but not with respect to others.
259
One might claim that the primary value in studying folkbiology in college
students at major research universities or among children in these same
communities is for the purpose of examining the cognitive consequences
of diminished contact with the natural world. These consequences are far
from minor. The Oxford English Dictionary studies showed a striking
quantitative loss of cultural support for learning about biological kinds
260
Chapter 10
in the twentieth century, and our studies with undergraduates yield little
familiarity with kinds other than mammals.
It is likely the case that the list of folkbiological universals could
be considerably lengthened and strengthened if one added a stipulation
requiring some minimal experience (which would eliminate most college
students at major research universities), just as language generalizations
are amplied by excluding wholly articial languages, such as Signed
English (a language constructed by a committee that turned out to be
unlearnable). The fact of devolution allows us to see which aspects of
folkbiology are most resistant to loss.
But qualitative changes in folkbiological understanding may be even
more important. One shift appears to be from an ecological orientation
where people see themselves as an integral part of nature to a conception
where nature is an externality and people are apart from it. In the Bang
et al. interviews with majority culture and Menominee parents, one group
dierence was in what one might call distancing discourse. For example, a Menominee parent might say We use milkweed for soup and a
majority-culture person parent might say Hard maple is used in ooring. The latter is more distancing than the former.
It is a challenge to identify factors that might lead to a qualitative shift
in folkbiological orientation, but one can make some guesses. Just turn
on the Discovery Channel to see programs devoted to wildlife. These programs almost always concern distant rather than local places. If you go to
a bookstore that sells childrens books, you will nd lots of material devoted to animals. But you will struggle to nd anything on local animals
(a side bet: youll nd vastly more on dinosaurs), not to mention plants or
ecosystems. In short, common sources of information about nature tend
to focus on exotic species in distant places where ecological interactions
typically do not go beyond predator-prey relations and ora are rarely
more than stage props.
Finally, quantitative and qualitative changes almost surely interact. If
all you know is oak, then you will not notice that some species of oaks
grow near water and others do well on the edges of prairies. If you cannot
distinguish the actors, then their interactions will not be meaningful to
you.
Folkbiological Variability
261
the same activities. These dierences represent a challenge for future research. Should they be conceptualized in terms of how focusing on certain
kinds of goals leads to devolution? Are these dierences mediated by differences in worldviews and mental models? What cognitive and cultural
processes are responsible for these large dierences in mental models?
In rural Wisconsin we see cultural dierences in ecological orientation in
the youngest children we have tested. So, whatever the source of the differences, it seems to be transmitted across generations at an early age. Although we have made some progress, we need to know much more about
the role of cultural and experiential factors in the development of folkbiological understandings.
Categorization and Reasoning
Theory and data on decision making may be the biggest success story that
the cognitive sciences have to oer (Kahneman and Tversky 1983). Regardless of whether it derives from success or boredom, however, the eld
262
Chapter 10
263
The twin facts that Itza 0 Maya devotionally value the ramon tree and see
the forest spirits or Arux as protectors of the forest who will punish transgressors has led us to the view that morally motivated decision making
plays a critical role in both environmental decision making and may also
be important in intergroup conict. Moreover, guring strictly from a
standpoint of maximizing current self-interest in a competitive environment, where other groups chop down ramon trees and otherwise degrade
the forest, protecting ramon appears to make little rational sense. Preserving what others destroy seems to be a waste of time and energy and
hence, in the long run, a waste of life. So why do Itza 0 continue to protect
the forest much as they believe the forest spirits do? There is, of course, a
body of work on morally motivated decision making, but we think our
work provides a somewhat dierent perspective on it.
But rst lets see what others say. Phil Tetlock and colleagues (2000)
denes sacred values as any value that a moral community implicitly
or explicitly treats as possessing innite or transcendental signicance
that precludes comparisons, tradeos, or indeed any other mingling with
bounded or secular values. There has been more than a decade of research on sacred or protected values (PVs) and decision making, and
some of their key properties have been identied. First of all, PVs are
linked to moral outrage and other emotions (Baron and Spranca 1997),
especially when a person holding a PV is oered a secular value in exchange for a PV (e.g., selling ones child, auctioning of body parts, or selling futures that bet on the likelihood of acts of terrorism; Medin et al.
1999; Tetlock 2002, 2003).
A second important generalization grows out of work by Jon Baron
and collaborators. They have amassed considerable evidence that PVs
are associated with a large omission bias (Baron and Greene 1996; Baron
and Ritov 1994; Ritov and Baron 1992, 1995, 1999) and that this bias
grows out of the use of deontological (e.g., do no harm) rather than
consequentialist decision rules. (A deontological decision rule is one based
on oughts and obligations or prohibitions regardless of outcomes, in direct contrast to a consequentialist decision rule which mandates choice of
the action that leads to the best overall outcome.) For example, in one scenario, people might be told about a threat to fteen species of sh and then
oered actions that will save those fteen species but threaten a number
of other species. In this sort of trade-o situation, some participants
typically those with PVssay that they would not want to cause the loss
of a single species.
264
Chapter 10
The general nding is that participants with PVs are less willing to
trade o than participants who do not express a PV, suggesting that these
people are less sensitive to a trade-o s consequences than people without
PVs. However, whether one appears to be focused on acts versus omissions also may to vary with (social) context. For example, when asked to
imagine people in a position of responsibility, participants show an act
rather than an omission bias (Haidt and Baron 1996), which may again
represent the application of a deontological rule. In some contexts, people
with PVs may feel a moral obligation to act, independent of the likelihood of success because it is the right thing to do (as Kant might have
said, Moral virtue is its own reward).
The characterization of devotional values as biases to be overcome
has tended toward the pejorative, and researchers have bemoaned the
fact that protected values get in the way of conict resolution and tradeos. For example, how can a court decide on monetary damages for
cases like the oil spill associated with the Exxon-Valdez when Alaskans
say that a pristine shoreline is of innite value? As both Baron and Tetlock have noted, people with PVs seem to treat them as having innite
value (i.e., in refusing to consider trade-os), but this premise entails a
logical impossibilityif PVs have innite value for people who endorse
them, they should spend literally all their time protecting and promoting
that value. For this reason, some have suggested that these values are
only pseudo-sacred (Baron and Leshner 2000; Thompson and Gonzalez
1997). Others have noted that people with PVs may nonetheless engage
in indirect trade-os (McGraw and Tetlock 2005; Tetlock 2000b). One
may be tempted to think of protected values as self-serving posturing
but acts such as suicide bombings by well-educated and well-adjusted
middle-class young adults (Atran 2003) or a monks self-immolation
(Gambetta 2005) undermine this stance (cf. Skitka and Mullen 2002).
Our work indicates that sacred or protected values cannot be dismissed
as a form of posturing. Not only do Itza 0 say that the ramon tree is
protected by spirits, but also when we do tree counts on their forest plots,
we nd a greater number of ramon trees than in adjacent plots managed
by Q 0 eqchi 0 Maya and Ladinos.1 These intriguing issues, coupled with
world events where people with PVs engage in heroism in some cases
and suicide terrorism in others (cf. Atran 2004), underline the importance of understanding morally motivated decisions, and suggest that
there are signicant empirical and theoretical challenges that demand further attention (Atran, Axelrod, and Davis 2007).
265
We do not need to dwell on the fact that, for our purposes, it is just a
nonstarter to treat or dene cultures and groups in terms of shared properties. In our initial studies with tree experts (Medin et al. 1997), we were
prepared to nd an overall consensus but our analysis suggested instead
that there were three distinct subgroups that corresponded almost perfectly with type of occupation. In the same vein, ordinarily one would
not expect a hodgepodge of immigrants coming from scattered areas to
constitute a culture, but as we saw in chapters 7 and 8, the Ladinos are
indeed forming a culture of sorts.
We are currently collecting data on folkecological models and social
and expert networks for three generations of Itza 0 and Ladinos. This will
allow us to trace knowledge change across generations and across groups.
Almost any pattern of results will be interesting (Do the youngest Itza 0
retain the notion of reciprocity? Does the notion of reciprocity jump
from Itza 0 to Ladinos? Do changes across generations show a dierent
266
Chapter 10
pattern as a function of gender?) It would be hardly conceivable to undertake this work without the cultural consensus model and a notion of cultural processes as dynamic. Finally, we note that the distributional view
of cultural processes is compatible with agent-based modeling. The modeling, coupled with the network and mental models data, can be used to
test ideas about information transmission as a function (for example) of
compatibility of mental models.
Methodological Implications
For those willing to swim upstream, the sights can be very rewarding. At
the risk of belaboring the point, the generalizations concerning three of
the most central phenomena in the cognitive science of categorization
(typicality, basic level, and inductive reasoning) have been fundamentally
changed by the sorts of cultural and expertise comparisons that this book
267
Cognitive and developmental psychology bring to cross-cultural comparisons some obvious strengths, such as concern for universal processes, statistical techniques for reliably sampling populations and distinguishing
groupwide patterns (messages) in the midst of considerable (often noisy)
individual variation, and controlled experiments that precisely delimit
the object of study. But as we noted at the beginning of the book, these
strengths can become weaknesses if not tempered by anthropological concerns, such as awareness of the need to avoid (1) essentializing cultural
groups, (2) interpreting variation as deviation, or equating the use of articial stimuli or stimuli in contrived contexts with natural referents and
situations.
Anthropologys forte is the ability to extract cultural knowledge from
groups of people and make it understandable to others, ideally without
losing the information required to recontextualize it and make it relevant
to those same peoples. Like a naturalist who would never imagine trying
to understand pandas apart from the bamboo forest that sustains them,
or water lilies without their ponds, so many anthropologists nd it unacceptable to consider decontextualized categories, concepts, populations,
and problems as representing the world, or what is psychologically most
interesting about it. But anthropologys strengths, too, turn to weaknesses
if not tempered by psychologys concerns, namely, wariness about counting the plural of anecdote as data or acknowledging that some informants special knowledge bespeaks a unique or dierent sort of mind
(however thickly embedded in narrative, history and the environment).
So, for us, the methodological union of psychology and anthropology
is not merely one of practical convenience, but of intellectual necessity.
Cultural Consensus Modeling and Cultural Epidemiology
Perhaps the best illustration of the mutually reinforcing benets that arise
from this union of psychological and anthropological methods is the t of
cultural consensus modeling to cultural epidemiology, which we consider
to be a major synthesis in our work. Rather than simply taking standard
measures of statistical reliability in groupwide patterns as evidence for
268
Chapter 10
cultural consensus, we used cultural consensus modeling to impose additional constraints that enabled us to identify and demonstrate patterns of
consensus more precisely (e.g., aggregated folktaxonomies, regularities in
ecological reasoning), so as to better make independent predictions about
a population (e.g., use of taxonomy to generate biological inferences,
knowledge of dierences in ecological reasoning to anticipate dierences
in environmental management). From the standpoint of cultural epidemiology, cultures are not bounded entities with component parts, but variable distributions of thoughts and behaviors through environments that
are often uid. In line with this stance, our use of cultural consensus
modeling allowed us to avoid having to blend interpretations of peoples
thoughts and actions into component parts of a culture. Instead, we were
able to describe emergent cultural patterns derived statistically from measurements of individual cognitions and behaviors, without losing any of
the information and insight attending individual variation.
Again motivated by cultural epidemiologys view of cultures as intersecting distributions of thoughts, behaviors, and environments, we used
cultural consensus modeling to examine the relations between agreement
patterns both within and across dierent populations. This generated
metacultural models that permitted us to trace possible pathways of
learning and information exchange within and between cultural groups:
for example, the likely transfer of Itza 0 ecological knowledge to socially
well-connected Ladino male experts, and then on to Ladino women. The
result illuminated more general processes of cultural formation, transformation, and evolution.
The merging of cultural consensus modeling with cultural epidemiology also sets the stage for potentially more informative agent-based and
causal modeling. Thus far we have only revealed groupwide patterns of
thoughts and behaviors, shown reliable correlations between dierent
sorts of patterns, and identied likely interactive pathways between them,
including key nodes (e.g., forest experts) in those pathways. But we have
not yet demonstrated causal connections, much less how these connections
are built up over time through exchanges between interacting agents.
What we have done is to establish a general framework for informed
agent-based and causal cultural modeling to take place.
Decision Making and Transcending Utility
There are two fairly novel aspects to our methods for dealing with decision
making. First, we sought to ground-truth our own, and our informants,
inferences about how their mental models of the environment translate
269
into ecological behavior and how that behavior aects the environment,
including measures of soil composition, canopy cover, crop diversity and
species counts, as well as time-sensitive satellite imagery of deforestation
patterns, and archival work on historical changes. Not only does such
ground-truthing validate ndings about the relationship between environmental thought and action, but without it we nd that policymakers (and
many scientists outside our main elds of interest) just will not take such
ndings seriously (nor should they).
Second, we extended our methods for eliciting mental models of ecological relationships among animals, plants and humans to the realm of
spirits. In one of our most intriguing studies, we found that Itza 0 believed
they would be punished if they violated spirit preferences, and that these
fears translated into behavioral patterns whose consequences on the composition of the forest could be traced back more than a millennium. We
also found that Menominee, like Itza 0 , appear to conceive of relations between humans and animal species in reciprocal terms. The implication of
these ndings is that standard rendering of decision making in terms of
utility maximization must be modied or rethought if these kinds of behavior are to make senseas they do to members of the populations we
study and to us.
This work has motivated us to branch out again, generalizing our concerns for resource conicts to cultural (including political) conicts more
generally, and involving ourselves in new eld sites in the Middle East,
Southeast Asia, and elsewhere. These concerns, in turn, have compelled
us to elaborate new methodological tools aimed at exploring the ways
cognition and emotional judgments might interact in sacred values to affect decision making and risk among individuals and groups. These methods include tests of (1) psychological barriers to trade-os (e.g., it is taboo
to sell o cultural heirlooms or sell out ones country), (2) immunity from
free-rider eects (e.g., conservationists make costly eorts not to deforest
even if most people continue deforesting, and suicide bombers willingly
die knowing their comrades will live), (3) disregard for material costbenet analysis (e.g., soldiers will rescue a buddy even if the rescue
greatly endangers many additional lives), (4) action bias and resistance
to certain framing eects (e.g., conservation eorts and risky military
action for cherished goals may be just as likely under frames of loss and
gain), (5) privileged ties with emotions (e.g., people respond with outrage
to immoral oers, such as money or sex to become a traitor; and people
morally outragede.g., humiliatedmay pursue the agent of that outrage with a vengeance even if it kills them), and (6) homogenization and
270
Chapter 10
271
Figure 10.1
Predictions of the percentage of the population who would use violence to oppose:
a peace deal perceived to violate a sacred value (taboo condition), the taboo
deal plus an added instrumental incentive (taboo), or the taboo deal plus a
sacred value concession without instrumental value, from the adversary (tragic)
for (A) Israeli settlers (linear trend F [1; 195] 5.698, P :018), and (B) Palestinian refugees (F [1; 384] 7.201, P :008). Parallel results obtained for emotional reactions by: (C ) settlers reporting anger or disgust at an Israeli
leader who would agree to the trade-o being evaluated (F [1; 260] 4.436,
P :036), and (D) refugees reporting joy at hearing of a suicide bombing
according to the type of trade-o being evaluated (F [1; 418] 7.48, P :007).
The trend of emotional intensity and support for violence in each case,
taboo > taboo > tragic, could not be predicted by an instrumental rationality
account of human behavior.
272
Chapter 10
one month before Hamas was elected to power. In this experiment, hypothetical peace deals (see supporting online materials) all violated the Palestinian right of return, a key issue in the conict (Shamir and Shikaki
2005). For the 80 percent of participants who believed this was an essential value, we once more observed that for violent opposition the order
between conditions was taboo > taboo > tragic, where those evaluating
a tragic deal showed lowest support for violent opposition (see gure
10.1B). Further, the same order was found for two measures ostensibly
unrelated to the experiment: (1) the belief that Islam condones suicide
attacks; and (2) reports of joy at hearing of a suicide attack (see de Quervain et al. 2004 for evidence of joy as a neurophysiological correlate of
revenge). Compared to refugees evaluating a taboo or taboo deal, those
evaluating a tragic deal believed less that Islam condoned suicide attacks,
and were less likely to report feeling of joy at hearing of a suicide attack
(see gure 10.1D). In neither the settler nor the refugee study did participants responding to the tragic deals regard these deals as more implementable than participants evaluating taboo or taboo deals.
These experiments, as well as recent follow-up studies with Palestinian and Israeli leaders (Atran, Axelrod, and Davis 2007), reveal that in
political disputes where sources of conict are cultural, such as the IsraeliPalestinian conict or emerging clashes between the Muslim and JudeoChristian world, violent opposition to compromise solutions may be
exacerbated rather than decreased by insisting on instrumentally driven
trade-os, while noninstrumental symbolic compromises may reduce support for violence.
Here we have only hinted at this new direction in our research (Tanner
and Medin 2004) and its potentially important implications (Atran 2006),
especially in regard to the complex and poorly understood relations
between consequentialist (utility-oriented) and deontological (moral and
sacred) reasoning. But it is worthwhile noting that this research would
have never gotten o the ground if not for our rudimentary attempts at
eliciting and exploring the behavioral eects of sacred values among the
Itza 0 Maya and Menominee.
Applications and Policy Implications
273
Conceptions of Culture
Our research and the work of others such as Susan Gelman and Larry
Hirschfeld show that some biological concepts, such as essentialism, are
very easy to think and very hard to set aside. Quite apart from any religious conicts associated with the concept of evolution, one must discard
well-entrenched notions about the nature of species in order to understand evolutionary theory (Hull 1999). In contrast, young children appear
to have rudimentary notions about inheritance that might provide an
eective foundation for formal instruction on biological inheritance.
Our work has additional educational implications for Native American
children in particular and perhaps children of color more generally. It is
very easy for majority-culture parents, teachers, and children to think of
culture as something that other people (those dierent from themselves)
have. Consequently, they nd it natural to think that instruction, including science instruction, is acultural, when in fact much of formal education in the United States (or Canada or France) is infused with practices
that represent cultural barriers to minority children. An analogy may
convey something of the avor of what we think often takes place. Imagine tourists from a country where people drive on the right side of the
road (e.g., United States) are transported to another culture where people
drive on the left (e.g., England). Now lets look at the poor tourists in the
role as pedestrians trying to cross the streetall of their attentional habits for looking before crossing will only get in the way. And, in fact, signicant numbers of U.S. tourists get into car-pedestrian accidents when
274
Chapter 10
walking in British cities. The analogy is straightforward: the set of practices that Native American children bring to the classroom are ones that
are natural outside of school, but create a clash in the classroom, because
the rules are dierent there.
As we noted earlier, we are currently analyzing in school and out of
school practices in Native American urban and Native American and
majority rural contexts. Our goal is to test the ideal that as cultural compatibility between in and out of school settings increases, classroom learning will increase.
The Environment and Environmental Decision Making
275
and benets, will usually not get cultural minorities or ethnic groups to let
others exploit territory, resources, symbols or names considered to be
holy or integral to their identity. Although sacred values often block conicting groups from reaching or even considering a negotiated settlement,
such values may also provide opportunities breakthrough and accommodation. If one side can nd unprotected values in the face of the other sides
protected values, then even symbolic token concessions of trivial value for
one side (the unprotected-values side) may yield huge benets for both
sides.3 For example, a relatively low-cost gesture from majority-culture
shermen to the Menominee would be to support the installation of sh
ladders on the three dams that currently block sturgeon (which the
Menominee consider sacred) from migrating up to spawning grounds on
the reservation in the spring. This might do wonders for improving intercommunity relations.
Border Crossings
Studying how people think about and act on nature is intimately involved
with a wide range of fundamental theory and policy concerns: from how
minds, societies, and environments build each other, to bettering science
education, environmental decision making, and the prospects for resolving resource conicts. And the lessons may go beyond that to even more
general issues of cultural conict and survival. One consequence of globalization is that many aspects of life centrally involve cross-culture
contact, communication, and negotiation. This is true not only across nations but also within them, as we have seen in Peten and in Wisconsin.
For research in culture and cognition to be eective in addressing these
issues requires a willingness and costly commitment to cross academias
cultural borders and perhaps to break some down.
Notes
Chapter Two
1. There are broadly three dierent kinds of competing scientic classication
schemes for biological organisms: cladistics, phenetics, and classical evolutionary
taxonomy. Cladists tend to focus only on those characters that reveal strict
branching sequences in phylogeny (Hennig 1966). Pheneticists ideally attempt to
base classications on as many observable characters as possible without prior
weighting in terms of their presumed relative importance in evolution (Sneath
and Sokal 1973). Evolutionary taxonomists ideally use as many observable characters as possible, but weight them according to their likely evolutionary role in
the process of natural selection (Mayr 1969). Because evolutionary taxonomy
deals with the joint eects of phylogenetic descent and adaptive radiation, it uses
both cladistic and phenetic perspectives to reconstruct limited patterns of evolutionary relationships among many morphological, behavioral, and ecological
characteristics. For example, from a cladistic (phylogenetic) standpoint crocodiles
are closely related to birds but not to turtles, whereas from a phenetic vantage
point crocodiles are closer to turtles. From the perspective of evolutionary taxonomy, however, the novel and wide-ranging adaptation of birds to life in the air
renders them a class apart (Aves) from both the crocodiles and turtles (Reptilia).
Field naturalists, behavioral ecologists, and biogeographers tend to prefer evolutionary taxonomies for mapping natural diversity, whereas most recent comparative work in biogenesis, microbiology, and genetics uses cladistic analysis as a
more reliable basis for making historical and causal inferences (but see the discussion below of Labandeira and Sepkoski 1993 for some surprising discoveries in
paleobiology via evolutionary taxonomy). There are no absolute ranks in cladistic
classications, unlike in evolutionary taxonomies (e.g., species, genus, family, order, class, and so on), and one could argue that continued preference for ranking
in evolutionary taxonomy owes to a lingering commonsense (folkbiological) bias
that articially boosts the correlation between evolutionary and folkbiological
taxa.
2. Botanists and ethnobotanists see privileged folkbiological groups as more
akin to scientic genera (Bartlett 1940; Berlin 1972; Greene 1983). Plant genera
especially are often readily recognized morphologically without technical aids
278
Notes
(Linnaeus 1751). Zoologists and ethnozoologists tend to view them as more like
scientic species, where reproductive and geographic isolation are more readily
identied in terms of behavior (Simpson 1961; Diamond 1966; Bulmer 1970).
3. English speakers ambiguously use animal to refer to at least three distinct
classes of living things: nonhuman animals, animals including humans, and mammals (prototypical animals). Beast seems to pick out nonhuman animals in English, but is seldom used today. Plant is ambiguously used to refer to the plant
kingdom, or to members of that kingdom that are not trees.
4. Only animals and plants are always exclusively individuated in terms of their
unique generic-species essence, whereas humans are variously individuated as
both individual agents and as social actors in accordance with inferred intentions
rather than expected clusters of body parts. Itza 0 , like folk everywhere, always
identify an individual animal or plant, rst and foremost, as a member of the generic species that presumably causes that individual to be. But Itza 0 , like most
people in the world, individuate humans, or winik, without exclusive recourse to
a single superordinate level of superordinate existence, such as the level of species.
Depending on context, a person may be Itza 0 or Yukatek, Maya or Ladino, man
or woman, mother or godmother, neighbor or stranger, hunter and/or farmer, or
some combination that presumably determines that persons intentional self.
5. Life forms vary across culture. Ancient Hebrew or modern Rangi (Tanzania)
include herpetofauna (reptiles and amphibians) with insects, worms, and other
creeping crawlers (Kesby 1979), whereas Itza 0 Maya and (until recently) most
Western cultures include herpetofauna with mammals as quadrupeds. Itza 0
place phenomenally isolated mammals like the bat with birds, just as Rofaifo
(New Guinea) place phenomenally isolated birds like cassowaries with mammals
(Dwyer 1976). Whatever the content of life-form taxa, the life-form level, or rank,
universally partitions the living world into broadly equivalent divisions.
6. According to Brown (1982, 102), Itza 0 see mammals as part of an unnamed
residual category that includes invertebrates except for worms. For Mayan languages generally, he claims mammal is a residual life form encompassing creatures
left over after encoding bird, fish, and snake. The evidence for the former claim
comes from Otto Schumanns (1971) supercial dictionary and the unpublished
notes of Pierre Ventur (Brown 1979, 382). Evidence for the latter claim comes
secondhand, via dictionaries. Overall, our experiments show that patterns of induction among mammals are the same as those for bird, fish, tree, or vine (see
chapter 4). In sorting tasks, mammals are always isolated from the other animals
as an exclusive group, with two exceptions: the bat (sotz 0 ) is always classied with
the birds, and the otter ( pek 0 -il ja 0 ) is always classied with other mammals but
occasionally cross-classied with some water-dwelling reptiles (crocodiles and turtles, but not water snakes). Brown also relies on linguistic evidence to claim that
kan (snakes) is an Itza 0 life form. But sorting and inference tasks (Atran 1999a)
clearly indicate that snakes and lizards (uy-et 0@ok juj ) are taxonomically closer
to one another than either of these intermediates is to other intermediates of the
herpetofauna life form (b 0 a 0 al@che 0 k-u-jil-t-ik-u-b 0 aj ), such as turtles (aak) or
amphibians (b 0 a 0 al@che 0 k-u-siit 0 ).
Notes
279
280
Notes
and is therefore encapsulated. Virtually any game (e.g., chess) or routine activity (e.g., car driving) relies on a restricted database that gives it privileged access
to a certain range of input. This would seem to trivialize the notion of modularity
and rob it of any descriptive or explanatory force.
Indeed, according to Fodor (2000, 23), the best case that can be made for the
computational theory of mind (i.e., the view that all conceptual processes are
Turing-like computations over syntactic-like representational structures) is in
terms of conceptual modularity. However, because conceptual modularity is
pretty clearly mistaken, the claim is also likely to be mistaken that the computational theory of mind has very much to tell us about how the mind congures the
world. For Sperber (2001), Fodors pessimism is unwarranted because it ignores
the fact that privileged access to an input set depends on the competition for mental
resources. Evolutionary task demands generally favor certain naturally selected
modular structures for processing certain types of naturally recurrent and statistically relevant input (all other things being equal). In principle, then, an explanatory account of modularity in terms of evolutionary task demands and related
developmental considerations of modularity is preferable to a purely descriptive
account in terms of encapsulation, mandatoriness, and the like.
4. At the time this study was conducted we thought that we were observing
central-tendency-based typicality eects, but we realized later that typicality in
this sense was confounded with typicality based on ideals. Subsequent studies (to
be described shortly) suggest that idealness is the key factor.
5. Barsalou (1985) argued that idealness rather than central tendency predicts
typicality in goal-derived categories (e.g., foods not to eat on a diet, things to
take from ones home during a re, camping equipment), although central tendency still supposedly predicts typicality in taxonomic categories (furniture,
vehicles), including folkbiological categories (birds).
6. F(1; 63) 7.32, Mse 3.5, p < .01.
7. F(5; 315) 3.14, Mse 0.88, p < .01.
8. Means 6.8 and 6.1, t(63) 4.33, p < .01.
9. Means 6.2 and 5.1, t(64) 3.06, p < .01.
10. Means 2.9 and 2.2, t(64) 2.23, p < .05.
11. The two-way interaction of level of expertise by taxonomic level was signicant, F(3; 183) 2.78, Mse 1.92, p < .05.
12. Paul Griths (2002) argues that because the items on any such symptomatic
list do not necessarily co-occur in any given case, and cannot unequivocally demonstrate innateness, then notions of innateness are inherently confused and should
be discarded. The same could be said against modularity. But the list represents
only an evidential claim, not a causal claim about innateness or modularity. It
provides a family of heuristics rather than a causal diagnosis.
13. Although the adaptive relationship of structure to function is often manifest,
as with the giraes neck or the rhinoceross horns, often it is not. In such cases,
evolutionary theorists adopt a strategy of reverse engineering. Reverse engineering is what military analysts do when a weapon from an enemy or competitor
Notes
281
in the arms market falls into their hands and they try to gure out exactly how it
was put together and what it can do. Reverse engineering is easiest, of course, if
the structure contains some signature of its function, like trying to gure out what
a toaster does given the telltale sign of toasted bread crumbs left inside. But in
many cases recognizing the appropriate signs already requires some prior notion
of what function the structure may have served. Thus, after a century and a half
of debate, it is only now that scientists clearly favor the hypothesis that bipedality
was primarily selected to enhance eld of view. Comparative studies of humans
with bipedal birds and dinosaurs, as well as experiments comparing energy expenditure and running speed in two-footed versus four-footed running and walking,
appear to exclude the competing hypotheses that bipedality evolved for running
or energy conservation. Paleontologists still do not know why triceratops had
neck frills and stegosaurs had back plates (defense, thermal regulation, sexual or
species signaling, and so on), except that lugging around all that cumbersome
baggage for millions of years must have had some adaptive function(s). For most
higher-order human cognitive faculties, however, there may be little useful comparative evidence from elsewhere in the animal kingdom. This is because of their
apparent structural novelty, poor representation in the fossil record (e.g., stone
tools tell little of language or theory of mind), and lack of surviving intermediate
forms. The moral is that reverse engineering can be helpful, and occasionally successful, but success is by no means guaranteed even in the richest of evidentiary
contexts.
Chapter Five
1. Still other characteristics may be explained in terms of individual, random
variation; however, our use of paired category-typical characteristics minimizes
this eventuality.
2. In another study, however, Gelman and Wellman (1991) asked children to reason about plants without identifying the species membership. For example, they
described a seed that came from an apple and was planted in a eld of corn, without identifying the seed as an apple seed. The results were largely the same as
with the animals and supported a nature-over-nurture bias (see Hickling and Gelman 1995; Gelman 2003).
3. For example, in Brazil, several of the 6- to 7-year-old children based their
responding on an explicit analogy with the Disney movie, Tarzan, which was
widely shown at the time of the study. They evinced a signicant but weaker birth
bias than 4- to 5-year-olds, consistent with Tarzans mixed human/ape behavioral
characteristics.
Chapter Six
1. Functionalism, which is alive and well in biology, should not be confused with
functionalism in anthropology, which has been in decline for at least half a century. Functionalism in anthropology, a dying metaphor, was initially derived
282
Notes
Notes
283
284
Notes
Notes
285
on public works projects for supplementary wages and greater reliance on sales of
surplus maize.
9. F(2; 41) 12.92, p < .001.
10. F 2; 41 25.04, p < .0001, D(I) .753, D(L) 1.39, D(Q) 3.92. Our
equation oversimplies the consequences of dierent patterns of use, which involve a trade-o between costs of farming a plot longer versus benets of
fallowing longer. In theory, the costs and benets could be quantied to assess
sustainability, but we have already seen that these groups farm dierently. One
potential limitation of our formula is that a shorter growing period for Q 0 eqchi 0
could leave the land in better shape for recovery. But soil tests (reported below)
reveal that nitrogen, a limiting factor in these calcied soils, is much more abundant in Itza 0 fallow land (guamil) than Q 0 eqchi 0 fallow.
11. We normalized highly variable distributions of raw scores with a natural log
transformation.
12. F(2; 27) 3.339, p < .05.
13. An ANOVA was performed on a composite of standardized scores for basic
nutrient elements: P (K Mg Ca). Calcium is antagonistic to xing of
potassium and magnesium, so the composite represents a balance of the available
nutrient elements: phosphorus for root growth, potassium for stem strength, magnesium for photosynthesis, calcium for cell formation. Results paralleled those
of phosphorus for Location (F(2; 162) 15.15, p < .0001; M > G; R), Level
(F(1; 162) 34.10, p < .0001; 1 > 2), and Group Location (F(4; 162) 4.02,
p .004; M: I(marginally) > Q; R: L > I).
14. As more immigrants have moved into the municipality of San Jose, milpa use
has shifted toward privatization. New immigrants are no longer automatically
granted use rights to 30 manzanas of land and all farmers retain their current
rights. In principle, farmers could benet from personal investments aimed at preserving sustainability in their own plot. But whether 30 manzanas suces for sustainability is not independent of the status of the surrounding land and the forest
practices of ones neighbors. The same plot that would be self-sustaining when
surrounded by healthy forest would not survive if surrounded by depleted land
(Schwartz 1995).
15. Plant vouchers were deposited at the University of Michigan Herbarium.
Vouchers numbers and photographs of plants appear in Atran, Lois, and Ucan
Ek 0 2004.
16. F(2; 33) 23.10, p < .001.
17. For each group, F(2; 22) > 23, p < .001.
18. F(9; 99) 26.04, p < .0001.
19. Participants were given two scores for each pairing of animal and plant
groups, reecting the proportion of positive and negative interactions acknowledged. A score of .25 for negative arboreal-fruit interactions indicates that the
participant identied negative interactions between one-quarter of all possible
pairings of arboreal animals and fruiting plants. Scores were entered into 2 (type
of interaction: positive, negative) 4 (animal group: bird, rummage, arboreal,
286
Notes
Notes
287
ally all their time and eort protecting and promoting that value. Moreover,
innite utility is incompatible with any sort of preference schedule: expected
utilities are weighted averages, which makes little sense when one of the terms is
innite. Thus, some have suggested these values are only pseudo-sacred (Baron
and Leshner 2000; Thompson and Gonzalez 1997); others have noted that people
with sacred values may nonetheless engage in indirect trade-os (McGraw and
Tetlock 2005; Tetlock 2000b). One may be tempted to think of sacred values as
self-serving posturing, but the reality of acts such as suicide bombings undermines this stance (Atran 2003a). Moreover, sacred values necessary to an individuals identity may take on truly absolute value only when value-related identity
seems gravely threatened, just as food may take on absolute value only when sustenance for life is threatened. A deeper point is that notions of maximization of
anticipated benets perhaps cannot best account for such spiritually driven behaviors, and ad hoc moves to maintain standard (probabilities and utilities) rationality at all costs result in a concept of rationality doing little explanatory work.
31. In April 2001, we presented our ndings on folkecology directly to the
Q 0 eqchi 0 , Ladino, and Itza 0 communities. The Itza 0 and Ladino experts in our
studies acknowledged the plausibility of the analysis showing the latters knowledge to be a proper subset of the formers. Q 0 eqchi 0 leaders also conrmed the
general reliability of our results and analyses.
Q 0 eqchi 0 representatives asked for help from Itza 0 on two counts. First, in light
of our ndings that smaller res allow Itza 0 to maximize natural fertilizers, including both phosphorus and nitrates in upper-level soils, the Q 0 eqchi 0 asked for instruction on Itza 0 burning techniques. Second, following presentations by Itza 0
women on the advantages of biodiversity for maintaining a living pharmacopoeia,
Q 0 eqchi 0 women asked Itza 0 for instruction on which plants to preserve for medicinal uses. The meeting began and ended with Itza 0 and Q 0 eqchi 0 prayers.
Q 0 eqchi 0 did not express interest in conservation as such, but were eager to produce more crops with less land while maintaining a richer stock of medicinally
useful plants. They asked that other meetings be organized on similar lines, that
contacts between the communities be more regular, and that specimens or photographs and descriptions of Itza 0 medicinal plants be made available.
32. A further observation is that the Itza 0 consider the ecologically central ramon
tree to be always worthy of protection and unlike the other two groups would
never use ramon as rewood. Although research in the psychology of decision
making sometimes views sacred or protected values as a hindrance to proper decision making and a source of cognitive biases (e.g., Baron and Spranca 1997),
there is other evidence suggesting that protected values may be associated with
the absence of framing eects and related biases (Fetherstonhaugh et al. 1997;
Friedrich et al. 1999; Tanner and Medin 2004).
Chapter Eight
1. To ensure maximum social coverage from our sample, initial informants could
not be immediate blood relatives (children, grandchildren, parents, grandparents,
288
Notes
siblings, rst cousins, nieces, nephews, uncles, aunts), anes (spouse, in-law), or
godparents (compadres).
2. The greatest overlap in the two networks occurs among Itza 0 and the least
among Q 0 eqchi 0 . For Itza 0 , fourteen of the most cited social partners are among
the twenty-two most cited forest experts. Although the Itza 0 social network is not
highly centralized, the most cited social partner is also the second most cited forest
expert, whereas the top forest expert is also the third most cited social partner.
For Ladinos, eleven of the most cited social partners are among the twenty-ve
most cited forest experts. Of these eleven, all are Ladino men. Ladino women
tend to mention Ladino men as experts; however, the top Ladino experts most
often cite the same Itza 0 experts as the Itza 0 themselves do, suggesting diusion
of information from Itza 0 experts to a select group of socially well-connected
Ladino men. For Q 0 eqchi 0 , who have by far the most densely connected and centralized social networks, only six of the most cited social partners are among the
eighteen most cited forest experts (these are cited much less often as experts than
outside institutions are).
3. Note that I is not directly connected to D1 (who belongs to the Itza 0 faction
opposed to Itza 0 -Y and to which I is allied).
4. For Ladinos, three of the four most cited Itza 0 experts are also the three named
most by Itza 0 . We combined Itza 0 and Ladino responses about plant-animal relations and found a metacultural consensus (rst factor scores all positive, ratio
eigenvalue 1:2 10.4, variance accounted for 52%). Then we regressed gender
and frequency of being cited as an expert against Ladino rst factor scores in the
combined consensus model. The r-square on Ladino scores was .63 (F(2; 10)
6.97, p .02) with gender ( p .02) and expertise ( p .008) reliable. One subgroup of men (with one woman) averaged 5.8 expert citations, 6.0 social network
citations, and an average culture competence (i.e., mean of rst factor scores) of
.73 (versus .75 for Itza 0 ). Averages for the other subgroup (with one man) were
respectively 0, 1.3, and .59.
5. We have independent evidence that people in these communities form and use
taxonomic hierarchies that correspond fairly well with classical scientic taxonomy (and especially so at the generic-species level). For example, using standard
sorting experiments (see Medin and Atran 2004), we elicited highly consensual
mammal taxonomies (see Lopez et al. 1997). For each population there was a single factor solution (I 7.2:1, 61%; L 5.9:1, 50%; Q 5.8:1, 48%). First factor
loadings were uniformly positive, and mean rst factor scores reected highly
shared competence for each population (I 77, L .71, Q .68). The aggregated Ladino taxonomy correlated equally with Itza 0 and Q 0 eqchi 0 taxonomies
(r .85), indicating very similar structures and contents. All three populations
grouped taxa according to general-purpose similarity rather than special-purpose
concerns (e.g., wild peccary with domestic pig, house cat with margay, and so on).
Special-purpose clusters, such as domestic versus wild, or edible versus nonedible,
can also be elicited (Lois 1998). But they do not belong to the general consensus
of kinds that go together by nature (see the idiosyncratic version of Itza 0 folk
taxonomy in Hoing and Tesucun 1997).
Notes
289
6. One possibility is that these data may simply indicate a social learning system
in which women are inuenced by women and men by men. To evaluate this idea
we conducted a residual analysis of the Ladino plant-animal relationships to see
if residual agreement was higher among men and among women than across genders. It was not. Instead, we found that women agreed with women reliably more
than men agreed with men [F 1; 10 9.64, Mse 3.16, p .01]. We think this
result reects the stability of inference processes at the family and genus level on
the part of women, relative to the diversity of concrete experience among Ladino
men.
7. One might interpret this cultural bias toward reciprocity as a shared abstract
expectation, but it requires additional supportive observations.
Chapter Nine
1. Ratio of rst eigenvalue 7.6 to 1, 57% of variance accounted for, average
rst factor score .75.
2. Ratio of rst to second eigenvalue 4.2:1, 30% of variance accounted for, average rst factor score .52.
3. Ratio of rst to second factor eigenvalue: 7.2; rst factor explains 63% of the
variance, all rst factor scores positive and high, average: 0.76.
4. Expressed in signicant dierences with respect to the second factor loadings,
F 22.9; Mse 1.2; p 0.000.
5. F 21.6; Mse 164; p 0.000.
6. F 8.2; Mse 86; p 0.007.
7. F 11.8; Mse 26; p 0.002.
8. F 3.6; Mse 7.5; p 0.06.
9. Ratio of rst to second eigenvalue 2.6, rst factor 44% of variance, mean
rst factor .63.
10. Eigenvalue 9.3 to 1, 66% of variance accounted for by the rst factor, and
average rst factor score .80.
Chapter Ten
1. There is also more recent research suggesting that moral values may be associated with less rather than more bias. Connolly and Reb (2003) criticized the paradigm used to assess omission bias and under their improved procedure, omission
bias all but disappeared. Earlier we mentioned the Tanner and Medin (2004) ndings that protected values led to, if anything, an act bias. In addition, they found
that protected values eliminated framing eects (i.e., the shift in choices when the
same objective situation is described in terms of gains versus lossesnegative
framing leads to more risk seeking, i.e., loss aversion). Dan Bartels in our laboratory has also found that protected values are associated with the reduction
290
Notes
References
292
References
References
293
Atran, S. 2006. The moral logic and growth of suicide terrorism. Washington
Quarterly 29:127147.
Atran, S., R. Axelrod, and R. Davis. 2007. Sacred barriers to conict resolution.
Science 317:10391040.
Atran, S., P. Estin, J. Coley, and D. Medin. 1997. Generic species and basic
levels: Essence and appearance in folk biology. Journal of Ethnobiology 17:2245.
Atran, S., X. Lois, and E. Ucan Ek 0 . 2004. Plants of the Peten Itza 0 Maya. Museum of Anthropology, University of Michigan Memoirs, No. 37. Ann Arbor:
Museum of Anthropology Publications.
Atran, S., and D. Medin. 1997. Knowledge and action: Cultural models of nature
and resource management in Mesoamerica. In M. Bazerman, D. Messick, A. Tinbrusel, and K. Wayde-Benzoni, eds., Environment, Ethics, and Behavior: The Psychology of Environmental Valuation and Degradation. San Francisco: Jossey-Bass.
Atran, S., D. Medin, E. Lynch, V. Vapnarsky, E. Ucan Ek 0 , and P. Sousa. 2001.
Folkbiology doesnt come from folkpsychology: Evidence from Yukatek Maya in
cross-cultural perspective. Journal of Cognition and Culture 1:342.
Atran, S., D. Medin, and N. Ross. 2005. The cultural mind: Ecological decision
making and cultural modeling within and across populations. Psychological Review 112:744776.
Atran, S., D. Medin, N. Ross, B. Lynch, J. Coley, E. Ucan Ek 0 , and V. Vapnarsky. 1999. Folkecology and commons management in the Maya Lowlands.
Proceedings of the National Academy of Sciences USA 96:75987603.
Atran, S., D. Medin, N. Ross, E. Lynch, V. Vapnarsky, E. Ucan Ek 0 , J. Coley,
C. Timura, and M. Baran. 2002. Folkecology, cultural epidemiology, and the
spirit of the commons: A garden experiment in the Maya Lowlands. Current Anthropology 41:123.
Atran, S., and A. Norenzayan. 2004. Religions evolutionary landscape: Counterintuition, commitment, compassion, communion. Behavioral and Brain Sciences
27:713770.
Atran, S., and D. Sperber. 1991. Learning without teaching: Its place in culture.
In L. Tolchinsky-Landsmann, ed., Culture, Schooling and Psychological Development. Norwood, N.J.: Ablex.
Atran, S., and J. Stern. 2005. Small groups nd fatal purpose through the web.
Nature 436:620.
Atran, S., and E. Ucan Ek 0 . 1999. Classication of useful plants among Northern
Peten Maya. In C. White, ed., Reconstructing Ancient Maya Diet. Salt Lake City:
University of Utah Press.
Au, T., and L. Romo. 1999. Mechanical causality in childrens folkbiology. In
D. Medin and S. Atran, eds., Folkbiology. Cambridge, Mass.: MIT Press.
Aunger, R. 2002. Exposure versus susceptibility in the epidemiology of everyday beliefs. Journal of Cognition and Culture 2002:113157.
Avendano y Loyola, Fray Andres de. 1987 [1696]. Relation of Two Trips to Peten
Made for the Conversion of the Heathen Ytzaex and Cehaches. F. Comparato, ed.,
C. Bowditch, and G. Rivera, trans. Culver City, Calif.: Labyrinthos.
294
References
References
295
296
References
References
297
Boyer, P. 1994. The Naturalness of Religious Ideas: A Cognitive Theory of Religion. Berkeley, Calif.: University of California Press.
Braisby, M., B. Franks, and J. Hampton. 1996. Essentialism, word use, and concepts. Cognition 59:247274.
Briley, D., M. Morris, and I. Simonson. 2000. Reasons as carriers of culture:
Dynamic versus dispositional models of cultural inuence on decision making.
Journal of Consumer Research 27:157178.
Bromley, D., ed. 1992. Making the Commons Work. San Francisco: Institute for
Contemporary Studies Press.
Brown, C. 1979. Growth and development of folk botanical life-forms in the
Mayan language family. American Ethnologist 6:366385.
Brown, C. 1982. Growth and development of folk zoological life-forms in the
Mayan language family. American Ethnologist 9:97111.
Brown, C. 1984. Language and Living Things: Uniformities in Folk Classication
and Naming. New Brunswick, N.J.: Rutgers University Press.
Brown, C. 1995. Lexical acculturation and ethnobiology: Utilitarianism and intellectualism. Journal of Linguistic Anthropology 5:5164.
Brown, C., J. Kolar, B. Torrey, T. Truong-Quang, and P. Volkman. 1976. Some
general principles of biological and non-biological classication. American Ethnologist 3:7385.
Brown, D., and S. Boysen. 2000. Spontaneous discrimination of natural stimuli
by chimpanzees (Pan troglodytes). Journal of Comparative Psychology 114:392
400.
Bruce, R. 1968. Gramatica Lacandona. Mexico, DF: Instituto Nacional de Antropologa e Historia.
Bruman, C. 1999. Why a successful concept should not be discarded. Current
Anthropology (Special Issue) 40:S1S14.
Buon, G. 17741789. Histoire generale et particulie`re. Suppl. Paris: Imprimerie
Royale.
Buon, G.-L. 17491767. Histoire Naturelle Generale et Particulie`re, 15 vols.
Paris: Impimerie Royale.
Bulmer, R. 1970. Which came rst, the chicken or the egg-head? In J. Pouillon
and P. Maranda, eds., Echanges et communications: Me langes oerts a Claude
Levi-Strauss. The Hague: Mouton.
Bulmer, R. 1974. Folk biology in the New Guinea Highlands. Social Science Information 13:928.
Bulmer, R., and M. Tylor. 1968. Karam classication of frogs. Journal of the
Polynesian Society 77:335385.
Burnett, R., D. Medin, N. Ross, and S. Blok. 2005. Ideal is typical. Canadian
Journal of Psychology 159:510.
Cain, A. 1956. The genus in evolutionary taxonomy. Systematic Zoology 5:97
109.
298
References
Campbell, D. 1958. Common fate, similarity, and other measures of the status of
aggregates of persons as social entities. Behavioral Science 3:1425.
Cano, Fray Augustn. [1695] 1984. Manche and Peten: The Hazards of Itza Deceit
and Barbarity. F. Comparato, ed., C. Bowditch and G. Rivera, trans. Culver City,
Calif.: Labyrinthos.
Caramazza, A. 2002. The organization of conceptual knowledge: The view from
neuropsychology. Paper presented at the British Academy Symposium on Conceptual Knowledge, June.
Carey, S. 1985. Conceptual Change in Childhood. Cambridge, Mass.: Bradford
Books.
Carey, S. 1995. On the origin of causal understanding. In D. Sperber, D. Premack, and A. Premack, eds., Causal Cognition. Oxford: Clarendon Press.
Carey, S. 1999a. The Origin of Concepts. Cambridge, Mass.: MIT Press.
Carey, S. 1999b. Sources of conceptual change. In E. Scholnick, K. Nelson, S.
Gelman, and P. Miller, eds., Conceptual Development: Piagets Legacy. Mahwah,
N.J.: Erlbaum.
Carey, S. 2003. The origin of concepts. Paper presented at the Joint Program
Lecture Series in Culture and Cognition and Evolution and Human Adaptation,
University of Michigan, Ann Arbor, February.
Carey, S., and R. Diamond. 1977. From piecemeal to congurational representation of faces. Science 195:312313.
Carter, W. 1969. New Lands and Old Traditions: Kekchi Cultivation in the Guatemalan Lowland. Gainesville: University of Florida Press.
Cavalli-Sforza, L., and M. Feldman. 1981. Cultural Transmission and Evolution.
Princeton, NJ: Princeton University Press.
Cerella, J. 1979. Visual classes and natural categories in the pigeon. Journal of
Experimental Psychology: Human Perception and Performance 5:6877.
Chase, A., and D. Chase. 1989. The Investigation of Classic Period Maya Warfare
at Caracol, Belize. Mayab 5:518.
Chomsky, N. 2000. Minimalist inquiries: The framework. In R. Martin,
D. Michaels, and J. Uriagereka, eds., Step by Step. Cambridge, Mass.: MIT
Press.
Cohen, D. 2001. Cultural variation: Considerations and implications. Psychological Bulletin 127:451471.
Cole, M. 1996. Cultural Psychology. Cambridge, Mass.: Harvard University
Press.
Coley, J. 1995. Emerging dierentiation of folkbiology and folkpsychology: Attributions of biological and psychological properties to living things. Child Development 66:18561874.
Coley, J., B. Hayes, C. Lawson, and M. Moloney. 2004. Knowledge, expectations, and inductive reasoning within conceptual hierarchies. Cognition 90:217
253.
References
299
Coley, J., D. Medin, and S. Atran. 1997. Does rank have its privilege? Inductive
inferences in folkbiological taxonomies. Cognition 63:73112.
Coley, J., D. Medin, J. Prott, E. Lynch, and S. Atran. 1999. Inductive reasoning
in folkbiological thought. In D. Medin and S. Atran, eds., Folkbiology. Cambridge, Mass.: MIT Press.
Conklin, H. 1962. Lexicographical treatment of folk taxonomies. In F. Householder and S. Saporta, eds., Problems in Lexicography. Report of the Conference
on Lexicography, November 1112, 1960. Bloomington: Indiana University
Press.
Connolly, T., and J. Reb. 2003. Omission bias in vaccination decision: Wheres
the omission? Wheres the bias? Organizational Behavior and Human Decision Processes 91:186202.
Cosmides, L., and J. Tooby. 1992. Cognitive adaptations for social exchange. In
J. Barkow, L. Cosmides, and J. Tooby, eds., The Adapted Mind: Evolutionary
Psychology and the Generation of Culture. New York: Oxford University Press.
Cowgill, U. 1962. An agricultural study of the southern Maya lowlands. American
Anthropologist 64:273286.
Csibra, G., G. Gergely, S. Bro, O. Koos, and M. Brockbank. 1999. Goal attribution without agency cues: The perception of pure reason in infancy. Cognition
72:237267.
Culbert, T., and D. Rice. 1990. Precolumbian Population History in the Maya
Lowlands. Albuquerque: University of New Mexico Press.
DAndrade, R. 1989. Cultural cognition. In M. Posner, ed., Foundations of
Cognitive Science. Cambridge, Mass.: MIT Press.
DAndrade, R., and C. Strauss. 1992. Human Motives and Cultural Models. Cambridge: Cambridge University Press.
Darwin, C. 1859. On the Origins of Species by Means of Natural Selection. London: Murray.
Darwin, C. [1872] 1883. On the Origins of Species by Means of Natural Selection.
6th ed. New York: Appleton.
Davis, T. 2000. Sustaining the Forest, the People, and the Spirit. New York: State
University of New York Press.
Dawkins, R. 1976. The Selsh Gene. New York: Oxford University Press.
Dean, J., G. Gumerman, J. Epstein, R. Axtell, A. Swedlund, M. Parker, and S.
McCarroll. 1999. Understanding Anasazi culture change through agent-based
modeling. In G. Gumerman and T. Kohler, eds., Modeling Small-Scale Societies.
New York: Oxford University Press.
Demarest, A. 1993. The violent saga of a Maya kingdom. National Geographic
183:95111.
Dennett, D. 1995. Darwins Dangerous Idea. New York: Simon & Schuster.
Descola, P. 1996. Constructing natures. In P. Descola and G. Palson, eds., Nature
and Society. London: Routledge.
300
References
References
301
302
References
Gelman, S., and H. Wellman. 1991. Insides and essences. Cognition 38:213
244.
Gerhard, P. 1991. La Frontera Sureste de la Nueva Espana. Mexico. DF: UNAM.
Ghiselin, M. 1981. Categories, life, and thinking. Behavioral and Brain Sciences
4:269313.
Ghiselin, M. 1999. Natural kinds and supraorganismal individuals. In D. Medin
and S. Atran, eds., Folkbiology. Cambridge, Mass.: MIT Press.
Gilens, M. 1996. Race and poverty in America: Public misperceptions and the
American news media. Public Opinion Quarterly 60:515541.
Gilliam, F. D., Jr., and S. Iyenger. 2000. Prime suspects: The inuence of local
television news on the viewing public. American Journal of Political Science
44(3):560573.
Gil-White, F. 2001. Are ethnic groups biological species to the brain? Current
Anthropology 42:515554.
Ginges, J., S. Atran, D. Medin, and K. Shikaki. 2007. Sacred bounds on rational
resolution of violent political conict. Proceedings of the National Academy of
Sciences 104:73577360.
Goldstein, W., and E. Weber. 1995. Content and discontent: Indications and
implications of domain specicity in preferential decision making. In J. R. Busemeyer, R. Hastie, and D. L. Medin, eds., The Psychology of Learning and Motivation, Vol. 32, Decision Making from a Cognitive Perspective 83136. San Diego:
Academic Press.
Goldstone, R. 1994. Inuences of categorization on perceptual discrimination.
Journal of Experimental Psychology: General 123:178200.
Goldstone, R. 1998. Perceptual learning. Annual Review of Psychology 49:585
612.
Gomez-Pompa, A., J. Flores, and V. Sosa. 1987. The pet kot: A man-made
tropical forest of the Maya. Interciencia 12:1015.
Graunke, G. 2003. American rights guardian update. Protecting American Rights
and Resources Newsletter 5, 7 (2003).
Greene, E. 1983. Landmarks in Botany. 2 vols. Stanford, Calif.: Stanford University Press.
Grice, H. 1975. Logic and conversation. In P. Cole and J. L. Morgan, eds., Syntax and Semantics, Vol. 3: Speech Acts. New York: Academic Press.
Griths, P. 2002. What is innateness? The Monist 85:7085.
Grignon, D., R. Alegria, C. Dodge, G. Lyons, C. Waukechon, C. Warrington, C.
Caldwell, C. La Chapelle, and K. Waupoose. 1998. Menominee Tribal History
Guide: Commemorating Wisconsin Sesquicentennial 18481998. Keshena, Wis.:
Menominee Indian Tribe of Wisconsin.
Grunberg, G., and V. Ramos. 1998. Base de Datos sobre poblacion, tierras y
medio ambiente en la Reserva de la Biosfera Maya. Peten, Guatemala: CARE
Guatemala and CONAP.
References
303
Grunberg, W. 2000. Modeling Deforestation: Risks for the Maya Biosphere Reserve, Guatemala. University of Arizona: School of Renewable Natural Resources.
Guntheil, G., A. Vera, and F. Keil. 1998. Do houseies think? Patterns of induction and biological beliefs in development. Cognition 66:3339.
Guyette, S. 1983. Community-Based Research: A Handbook for Native Americans.
Los Angeles: University of California Press.
Haidt, J., and J. Baron. 1996. Social roles and the moral judgement of acts and
omissions. European Journal of Social Psychology 26:201218.
Hall, P., and M. Pecore. 1995. Case Study: Menominee Tribal Enterprises. Madison: Institute for Environmental Studies and the Land Tenure Center, University
of Wisconsin at Madison.
Hardin, G. 1968. The tragedy of the commons. Science 162:12431248.
Harris, M. 1974. Cows, Pigs, Wars, and Witches. New York: Random House.
Harrison, P., and B. Turner. 1978. Pre-Hispanic Maya Agriculture. Albuquerque:
University of New Mexico Press.
Hatano, G., and K. Inagaki. 1987. Everyday biology and school biology: How
do they interact? Quarterly Newsletter of the Laboratory of Comparative Human
Cognition 9:120128.
Hatano, G., and K. Inagaki. 1994. Young childrens naive theory of biology.
Cognition 50:171188.
Hatano, G., and K. Inagaki. 1996. Cultural contexts of schooling revisited: A review of the learning gap from a cultural psychology perspective. Paper presented
at the Conference on Global Prospects for Education: Development, Culture and
Schooling, University of Michigan.
Hatano, G., and K. Inagaki. 1999. A developmental perspective on informal
biology. In D. L. Medin and S. Atran, eds., Folkbiology. Cambridge, Mass.:
MIT Press.
Hatano, G., and K. Inagaki. 2003. When is conceptual change intended? A
cognitive-sociocultural view. In G. Sinatra and P. Pintrich, eds., Intentional Conceptual Change. Mahwah: Erlbaum.
Hatano, G., and K. Inagaki. Forthcoming. The formation of culture in mind: A
sociocultural approach to cognitive development. In J. Mehler, S. Carey, and L.
Bonatti, eds., Cognitive Development and Conceptual Change. Cambridge, MA:
MIT Press.
Hauser, M. 2000. Wild Minds: What Animals Really Think. New York: Henry
Holt and Company.
Hauser, M., N. Chomsky, and W. T. Fitch. 2002. The faculty of language.
Science 298:15691578.
Hays, T. 1983. Ndumba folkbiology and general principles of ethnobotanical
classication and nomenclature. American Anthropologist 85:592611.
Heider, F., and S. Simmel. 1944. An experimental study of apparent behavior.
American Journal of Psychology 57:243259.
304
References
Hellmuth, N. 1977. Cholti-Lacandon (Chiapas) and Peten-Ytza agriculture, settlement pattern and population. In N. Hammond, ed., Social Process in Maya
Prehistory: Studies in Honor of Sir J. Eric S. Thompson. London: Academic
Press.
Hempel, C., and P. Oppenheim. 1948. Studies in the logic of explanation. Philosophy of Science 15:567579.
Hennig, W. 1966. Phylogenetic Systematics. Urbana: University of Illinois Press.
Henley, N. M. 1969. A Psychological Study of the Semantics of Animal Terms.
Journal of Verbal Learning and Learning Behavior 8:176184.
Henrich, J., and R. Boyd. 1998. The evolution of conformist transmission and the
emergence of between-group dierences. Evolution and Human Behavior 19:215
241.
Henrich, J., and R. Boyd. 2002. On modeling cognition and culture. Journal of
Cognition and Culture 2:87112.
Henrich, J., R. Boyd, S. Bowles, C. Camerer, H. Gintis, R. McElreath, and E.
Fehr. 2001. In search of Homo economicus: Experiments in 15 small-scale societies. American Economic Review 91:7379.
Henrich, J., and F. J. Gil-White. 2001. The evolution of prestige. Evolution and
Human Behavior 22:165196.
Hermes, M. 1999. Research methods as a situated response: Toward a First Nations methodology. In L. Parker, D. Deyhle, S. Villenas, eds., Race Is . . . Race
Isnt: Critical Race Theory and Qualitative Studies in Education. Boulder, Colo.:
Westview Press.
Herrnstein, R. J. 1984. Objects, categories, and discriminative stimuli. In H. L.
Roitblat, T. G. Bever, and H. S. Terrace, eds., Animal Cognition. Hillsdale, NJ:
Lawrence Erlbaum Associates.
Hickling, A., and S. Gelman. 1995. How does your garden grow? Evidence of an
early conception of plants as biological kinds. Child Development 66:856876.
Hirschfeld, L. 1995. Do children have a theory of race? Cognition 54:209252.
Hirschfeld, L. 1996. Race in the Making. Cambridge, Mass.: MIT Press.
Hirschfeld, L. 2002. Why dont anthropologists like children? American Anthropologist 104(2):611627.
Hirschfeld, L., and S. Gelman, eds. 1994. Mapping the Mind. New York: Cambridge University.
Homan, W. J. [1896] 1970. The Menomini Indians. In Fourteenth Annual Report
of the US Bureau of Ethnology, 189293. New York: Johnson Reprints.
Hoing, C. 1996. Indigenous linguistic revitalization and outsider interaction: The
Itzaj Maya case. Human Organization 55:108116.
Hoing, C., and F. Tesucun. 1997. Itzaj-Maya-Spanish-English Dictionary. Salt
Lake City: University of Utah Press.
Hofstede, G. 1980. Cultures Consequences: International Dierences in WorkRelated Values. Beverley Hills, Calif.: Sage.
References
305
306
References
Jeyifous, S. W. 1992. Developmental changes in the representation of word meaning: Cross-cultural ndings. British Journal of Developmental Psychology 10:285
299.
Job, R., and L. Surian. 1998. A neurocognitive mechanism for folk biology?
Behavioral and Brain Sciences 21:577578.
Johnson, K., and C. Mervis. 1997. Eects of varying levels of expertise on the
basic level of categorization. Journal of Experimental Psychology: General
126:248277.
Johnson, K. E., and C. B. Mervis. 1998. Impact of intuitive theories on feature
recruitment throughout the continuum of expertise. Memory and Cognition
26:382401.
Johnson, K., C. Mervis, and J. Boster. 1992. Developmental changes within the
structure of the mammal domain. Developmental Psychology 28:7483.
Johnson, S., and G. Solomon. 1997. Why dogs have puppies and cats have kittens: The role of birth in young childrens understanding of biological origins.
Child Development 68:404419.
Jones, G. 1998. The conquest of the last Maya kingdom. Stanford: Stanford University Press.
Jussieu, A. L. 1789. Genera plantarum. Paris: Herissant.
Kahneman, D., and A. Tversky. 1979. Prospect theory: An analysis of decision
under risk. Econometrika 47:263291.
Kahneman, D., and A. Tversky. 1983. Choices, values, and frames. American
Psychologist 39, no. 4 (April): 341350.
Kannengeiser, C. 1995. Enseigner les origins du Christianisme dans le cadre actuel
des sciences religieuses au Quebec. Religiologiques 11:111119.
Karabatsos, G., and W. Batchelder. 2003. Markov chain estimation for test
theory without the answer key. Psychometrika 68:373389.
Keesing, F. M. [1939] 1987. The Menomini Indians of Wisconsin: A study of Three
Centuries of Contact. Madison: University of Wisconsin Press.
Keesing, R. 1972. Paradigms lost: The new anthropology and the new linguistics.
Southwest Journal of Anthropology 28:299332.
Keil, F. 1979. Semantic and Conceptual Development. Cambridge, Mass.: Harvard University Press.
Keil, F. 1989. Concepts, Kinds, and Cognitive Development. Cambridge, Mass.:
MIT Press.
Keil, F. 1995. The growth of causal understanding of natural kinds: Modes of
construal and the emergence of biological thought. In A. Premack and D.
Sperber, eds., Causal Cognition. Oxford: Oxford University Press.
Keil, F., D. Levin, B. Richman, and G. Gutheil. 1999. Mechanism and explanation in the development of biological thought. In D. L. Medin and S. Atran, eds.,
Folkbiology. Cambridge, Mass.: MIT Press.
References
307
Kellert, S. R. 1993. The Biological Basis for Human Values of Nature. In S. Kellert and E. Wilson, eds., The Biophilia Hypothesis. Washington, D.C.: Island
Press.
Kempton, W., J. Boster, and J. Hartley. 1995. Environmental Values in American
Culture. Cambridge, Mass.: MIT Press.
Kesby, J. 1979. The Rangi classication of animals and plants. In R. Reason and
D. Ellen, eds., Classications in Their Social Contexts. New York: Academic
Press.
Klein, G. 1999. Sources of Power: How People Make Decisions. Cambridge,
Mass.: MIT Press.
Knight, N., P. Sousa, J. Barrett, and S. Atran. 2004. Childrens attributions of
beliefs to humans and God: Cross-cultural evidence. Cognitive Science 28:117
126.
Kollman, K., J. Miller, and S. Page. 1992. Adaptive parties in spatial elections.
American Political Science Review 86:929937.
Kripke, S. 1972. Naming and necessity. In D. Davidson and G. Harman, eds.,
Semantics of Natural Language. Dordrecht: Reidel.
Kroeber, A. [1923] 1963. Anthropology. New York: Harcourt, Brace & World.
Labandeira, C., and J. Sepkoski. 1993. Insect diversity in the fossil record. Science
261:310315.
Laland, K., and G. Brown. 2002. The golden meme: memes oer a way to analyse human culture with scientic rigour. Why are social scientists so averse to the
idea. New Scientist 175:4043.
Laland, K., and F. J. Olding-Smee. 2001. The evolution of memes. In Robert
Aunger, ed., Darwinizing Culture: The Status of Memetics as a Science. Oxford:
Oxford University Press.
Laland, K., F. J. Olding-Smee, and M. Feldman. 2000. Niche construction, biological evolution and cultural change. Behavioral and Brain Sciences 23:131146.
Landa, Diego de. [1566] 1985. Relacion de la cosas de Yucatan. Ed. M. Rivera
Dorado. Cronicas de America, no. 7. Madrid: Historia 16.
Lansing, S., and J. Kremer. 1993. Emergent properties of Balinese water temple
networks: Coadaptation on a rugged tness landscape. American Anthropologist
95:97114.
Lave, J., and E. Wenger. 1991. Situated Learning. New York: Cambridge University Press.
Lawson, E. T., and R. McCauley. 1990. Rethinking Religion. Cambridge: Cambridge University Press.
Leach, E. 1976. Culture and Communication. Cambridge: Cambridge University
Press.
Levi-Strauss, C. 1963a. The bear and the barber. The Journal of the Royal Anthropological Institute 93:111.
Levi-Strauss, C. 1963b. Structural Anthropology. New York: Anchor Books.
308
References
References
309
310
References
References
311
312
References
References
313
Ritov, I., and J. Baron. 1992. Status-quo and omission bias. Journal of Risk and
Uncertainty 5:4961.
Ritov, I., and J. Baron. 1995. Outcome knowledge, regret, and omission bias.
Organizational Behavior and Human Decision Processes 64:119127.
Ritov, I., and J. Baron. 1999. Protected values and omission bias. Organizational
Behavior and Human Decision Processes 79:7994.
Ritov, I., and D. Kahneman. 1997. How people value information: Attitudes
versus economic values. In M. Bazerman, D. Messick, A. Tenbrunsel, and K.
Wade-Benzoni, eds., Environment, Ethics, and Behavior: The Psychology of Environmental Valuation and Degradation. San Francisco: New Lexington Press.
Roberts, J. 1964. The self-management of cultures. In W. Goodenough, ed.,
Explorations in Cultural Anthropology: Essays in Honor of George Peter Murdock.
New York: McGraw-Hill.
Romney, A. K. 1998. Toward a theory of culture as shared cognitive structures.
Ethos 26:314337.
Romney, A. K., and W. Batchelder. 1999. Cultural consensus theory. In R. Wilson and F. Keil, eds., The MIT Encyclopedia of the Cognitive Sciences. Cambridge, Mass.: MIT Press.
Romney, A. K., W. Batchelder, and S. Weller. 1986. Culture as consensus: A
theory of culture and informant accuracy. American Anthropologist 88:313338.
Romney, A. K., J. Boyd, C. Moore, W. Batchelder, and T. J. Brazill. 1996. Culture as shared cognitive representations. Proceedings of the National Academy of
Sciences USA 93:46994705.
Rosch, E. 1975. Universals and cultural specics in categorization. In R. Brislin,
S. Bochner, and W. Lonner, Eds., Cross-Cultural Perspectives on Learning. New
York: Halstead.
Rosch, E., and C. Mervis. 1975. Family resemblances: Studies in the internal
structure of categories. Cognitive Psychology 7:573605.
Rosch, E., C. Mervis, W. Grey, D. Johnson, and P. Boyes-Braem. 1976. Basic
objects in natural categories. Cognitive Psychology 8:382439.
Ross, B. 1997. The use of categories aects classication. Journal of Memory and
Language 37:240267.
Ross, L. 1990. Recognizing the role of construal processes. In I. Rock, ed., The
Legacy of Solomon Asch: Essays in Cognition and Social Psychology, 7796. Hillsdale, N.J.: Erlbaum.
Ross, L., and C. Stillinger. 1991. Barriers to conict resolution. Negotiation Journal 8:389404.
Ross, N. 2000. Lacandon Maya intergenerational change and the erosion of
folkbiological knowledge. Proceedings of the 7th International Congress of Ethnobiology, Athens, Georgia.
Ross, N. 2001. Bilder vom Regenwald: Mentale Modelle, Kulturwandel und
Umweltverhalten bei den Lakandonen in Mexiko. Muenster: Lit Verlag.
314
References
References
315
Schwartz, N. 1990. Forest Society: A Social History of Peten, Guatemala. Philadelphia: University of Pennsylvania Press.
Schwartz, N. 1995. Colonization, development, and deforestation in Peten,
Northern Guatemala. In M. Painter and W. Durham, eds., The Social Causes of
Environmental Destruction in Latin America. Ann Arbor: University of Michigan
Press.
Schyns, P., R. Goldstone, and J. Thibaut. 1998. Development of features in object
concepts. Behavioral and Brain Sciences 21:154.
Schyns, P., and L. Rodet. 1997. Categorization creates functional features.
Journal of Experimental Psychology: Learning, Memory, and Cognition 23:681
696.
Shafto, P., and J. D. Coley. 2003. Development of categorization and reasoning
in the natural world: Novices to experts, nave similarity to ecological knowledge.
Journal of Experimental Psychology: Learning, Memory, and Cognition 29:641
649.
Shamir, J., and K. Shikaki. 2005. Public opinion in the Israeli-Palestinian twolevel game. Journal of Peace Research 42:311328.
Shweder, R. 1990. Cultural psychology: What is it? In J. Stigler, R. Shweder, and
G. Herdt, eds., Cultural Psychology: Essays on Comparative Human Development.
New York: Cambridge University Press.
Simons, D. J., and F. C. Keil. 1995. An abstract to concrete shift in the development of biological thought: The insides story. Cognition 56:129163.
Simpson, G. 1961. Principles of Animal Taxonomy. New York: Columbia
University.
Skitka, L., and D. Houston. 2001. The dark side of moral conviction. Analyses of
Social Issues and Public Policy 2:35241.
Sloman, S. 1993. Feature-based induction. Cognitive Psychology 25:231280.
Sloman, S. 1994. When explanations compete: The role of explanatory coherence
on judgments of likelihood. Cognition 52:121.
Sloman, S., and B. Malt. 2003. Artifacts are not ascribed essences, nor are they
treated as belonging to kinds. Language and Cognitive Processes 18:563582.
Smith, E., and D. Medin. 1981. Categories and Concepts. Cambridge, Mass.: Harvard University Press.
Smith, E., E. Shoben, and L. Rips. 1974. Structure and process in semantic memory. Psychological Review 81:214241.
Smith, L. T. 1999. Decolonizing Methodologies: Research and Indigenous Peoples.
NewYork: Zed Books.
Smithe, F. 1966. The Birds of Tikal. Cambridge: Harvard University Press.
Smyth, M., and D. Christopher. 1992. Large-site archaeological methods at Sayil,
Yucatan, Mexico: Investigating community organization at a prehispanic Maya
center. Latin American Antiquity 3:321.
Sneath, P., and R. Sokal. 1973. Numerical Taxonomy. San Francisco: Freeman.
316
References
Sober, E. 1996. Learning from functionalism: Prospects for strong articial life. In
M. Boden, ed., The Philosophy of Articial Life. New York: Oxford University
Press.
Sober, E., and D. S. Wilson. 1998. Unto Others: The Evolution and Psychology of
Unselsh Behavior. Cambridge, Mass.: Harvard University Press.
Solomon, G., S. Johnson, D. Zaitchik, and S. Carey. 1996. Like father, like son:
Young childrens understanding of how and why ospring resemble their parents.
Child Development 67:151171.
S.O.S.: se muere biosfera Maya. 1998. Prensa Libre, Guatemala daily, April 22.
Sousa, P., S. Atran, and D. Medin. 2002. Essentialism and folkbiology: Further
evidence from Brazil. Journal of Cognition and Culture 2:195203.
Soza, J. M. 1970. Monografa del Departamento de El Peten, 2 vols. Guatemala:
Editorial Jose de Pineda Ibarra.
Sperber, D. 1985. Anthropology and psychology: Towards an epidemiology of
representations. Man 20:7389.
Sperber, D. 1996. Explaining Culture. Oxford: Blackwell.
Sperber, D. 2001. In defense of massive modularity. Paper presented to the
Innateness and Structure of the Mind Workshop, University of Sheeld,
November.
Sperber, D., F. Cara, and V. Girotto. 1995. Relevance theory explains the selection task. Cognition 57:3195.
Sperber, D., and L. Hirschfeld. 2004. The cognitive foundations of cultural stability and diversity. Trends in Cognitive Sciences 8:4046.
Sperber, D., P. Premack, and A. Premack, eds. 1995. Causal Cognition. Oxford:
Clarendon Press.
Sperber, D., and D. Wilson. 1986. Relevance: Communication and Cognition. Oxford: Blackwell.
Spindler, G., and L. S. Spindler. 1971. Dreamers with Power: The Menomini Indians. New York: Holt, Rinehart & Winston.
Spindler, G., and L. S. Spindler. 1991. Reactions and worries. Anthropology &
Education Quarterly 22(3):274278.
Stewart, S. 1980. Grammatica Kekchi. Guatemala: Editorial Academica Centro
Americana.
Steyvers, M., and J. B. Tenenbaum. 2005. The large-scale structure of semantic
networks: Statistical analyses and a model of semantic growth. Cognitive Science
29(1):4178.
Stierle, A., G. Strobel, and D. Stierle. 1993. Taxol and taxane production by Taxomyces andreanae, an endophytic fungus of Pacic Yew. Science 260:214216.
Stoler, A. 1995. Race and the Education of Desire. Durham, NC: Duke University
Press.
Strauss, C., and N. Quinn. 1992. Preliminaries to a theory of culture acquisition.
In H. L. Pick, P. Van Den Broek, and D. C. Knill, eds., Cognition: Conceptual
References
317
318
References
Classic period. In D. Chase and A. Chase, eds., Mesoamerican Elites: An Archaeological Assessment. Norman. Okla.: University of Oklahoma Press.
Tozzer, A., ed. 1941. Landas relacion de las cosas de Yucatan. Papers of the Peabody Museum, 18. Cambridge, Mass.: The Peabody Museum.
Tracy, L. 1998. State trends. Council of State Governments Newsletter, summer
1998.
Trager, G. 1939. Cottonwood tree: A Southwestern linguistic trait. International Journal of American Linguistics 9:117118.
Triandis, H. 1995. Individualism and Collectivism. Boulder, Colo.: Westview Press.
Triandis, H. 2001. Individualism-collectivism and personality. Journal of Personality 69:907924.
Tribal claims were too much. 1999. Milwaukee Journal Sentinel, April 25. Available at www.jsonline.com/news/editorials/apr99/0425tribal.asp.
Tversky, A., and D. Kahneman. 1981. The framing of decisions and the psychology of choice. Science 211:453458.
Tversky, A., and D. Kahneman. 1986. Rational choice and framing decisions.
Journal of Business 59: S251278.
Van der Henst, J.-B., G. Politzer, and D. Sperber. 2002. When is a conclusion
worth deriving? A relevance-based analysis of indeterminate relational problems.
Thinking & Reasoning 8(1):120.
Villagutierre Soto-Mayor, Juan de. 1701. Historia de la conquista de la provincia
de el Itza, reduccion, y progressos de la de el Lacandon y otras naciones de indios
barbaros, de la mediacion de el reyno de Guatimala, a las provincias de Yucatan, en
la America septentrional. Madrid: Lucas Antonio Bedmar, y Narvaez.
Villagutierre Soto-Mayor, Juan de. [1701] 1983. In F. Comparto, ed., R. Wood,
trans History of the Conquest of the Province of the Itza. Culver City, Calif.:
Labyrinthos.
Vygotsky, L. 1978. Mind in Society. Cambridge, Mass.: Harvard University Press.
Waddington, C. 1959. Canalisation of development and the inheritance of
acquired characteristics. Nature 183:16541655.
Walker [Jeyifous], S. 1992. Supernatural beliefs, natural kinds, and conceptual
structure. Memory & Cognition 20:655662.
Walker, S. J. 1999. Culture, domain specicity and conceptual change: Natural
kind and artifact concepts. British Journal of Developmental Psychology 17:203
219.
Wallace, A. F. 1961. Introduction. In A. F. Wallace, Culture and Personality.
New York: Random House.
Wallace, A. R. [1889] 1901. Darwinism. New York: Macmillan.
Warburton, F. 1967. The purposes of classication. Systematic Zoology 16:241
245.
Warrington, E., and T. Shallice. 1984. Category specic semantic impairments.
Brain 107:829854.
References
319
Waxman, S. 1999. The dubbing ceremony revisited: Object naming and categorization in infancy and early childhood. In D. Medin and S. Atran, eds., Folkbiology. Cambridge, Mass.: MIT Press.
Waxman, S. 2004. Everything had a name, and each name gave birth to a new
thought: Links between early word-learning and conceptual organization. In
D. G. Hall and S. R. Waxman, eds., From Many Strands: Weaving a Lexicon.
Cambridge, Mass.: MIT Press.
Waxman, S., and D. Markow. 1995. Words as invitations to form categories:
Evidence from 12-month-old infants. Cognitive Psychology 29:257302.
Waxman, S., and D. Medin. 2007. Experience and cultural models matter:
Placing rm limits on anthropocentrism. Human Development 50:2330.
Waxman, S. R., D. L. Medin, and N. Ross. 2007. Folkbiological reasoning from
a cross-cultural developmental perspective: Early essentialist notions are shaped
by cultural beliefs. Developmental Psychology 43:294308.
Weller, S. 1987. Shared knowledge, intracultural variation, and knowledge aggregation. American Behavioral Scientist 31:178193.
Wester, L., and S. Yongvanit. 1995. Biological diversity and community lore in
Northwestern Thailand. Journal of Ethnobiology 15:7187.
White, L. 1949. The Science of Culture. New York: Farrar, Straus.
White, S. 1994. Testing an economic approach to resource dilemmas. Organization and Human Decision Processes 58:428456.
Williams, G. 1992. Natural Selection. New York: Oxford University Press.
Wilson, D. S. 2002. Darwins Cathedral. Chicago: University of Chicago Press.
Wilson, E. O. 1978. On Human Nature. Cambridge, Mass.: Harvard University
Press.
Wilson, E. O. 1992. The Diversity of Life. New York: Norton.
Wilson, R. 1995. Maya Resurgence in Guatemala. Norman: University of Oklahoma Press.
Wisconsin Department of Natural Resources. 2000. Wisconsins Lake Sturgeon
Management Plan. Milwaukee, Wis.: Bureau of Fisheries Management and Habitat Protection.
Wisconsin Department of Natural Resources. 2002. Open Water Spearing and
Netting in Northern Wisconsin by Chippewa Indians during 2001. Great Lakes
Indian Fish and Wildlife Commission Administrative Report, Wisc2002-01. Odanah, Wis.: Great Lakes Indian Fish and Wildlife Commission.
Wiseman, F. 1978. Agricultural and historical archaeology of the Maya Lowlands. In P. Harrison and B. Turner, eds., Pre-Hispanic Maya Agriculture. Albuquerque: University of New Mexico Press.
Wisniewski, E., and D. Medin. 1994. On the interaction of theory and data in
concept learning. Cognitive Science 18:221281.
Wol, P., D. Medin, and C. Pankratz. 1999. Evolution and devolution of folkbiological knowledge. Cognition 73:177204.
320
References
Index
322
Index
subgroups, 2324
systematic inferences, use in, 24
theories of categorization, implications of research ndings for, 261
universality of, 1719
Causal reasoning, 1012. See also
Reasoning in folkbiology
Cavalli-Sforza, L., 283n7
CCM (cultural consensus modeling),
5760, 69, 75, 77, 185, 206, 222,
267268, 279n1
Central tendency, 89, 93, 94, 95, 97,
99, 100, 112
Child development. See Development
of folkbiological cognition in
children
Children, education of. See Education
China, symbolic worth in, 290n3
Chippewa (Ojibwe), 232
Chomsky, Noam, 151
Cladistics, 277n1
Classical biological classication, 25
Classical evolutionary taxonomy,
277n1. See also Scientic vs. folkbiological taxonomies
Cluster analysis in birds study, 7985,
8085
Cognition, situated, 147148
Cognitive function and folkbiology,
63118. See also Reasoning in
folkbiology
autonomy and universality, 113114
development in children (see Development of folkbiological cognition
in children)
evolutionary context, 6465
FBS (folkbiological system), 6364
generic species privileged status for
biological inference, 101110, 104,
106, 107, 109, 111112
ideals, 88, 92101, 280n5
implications of study results for
categorization and reasoning, 110
114
inductive inference, 92, 103, 104
typicality eects, 92101, 96, 97, 99,
110111
Index
323
324
Index
Index
325
Education
development of folkbiological
cognition in children and science
education, 138141
Menominee children, boarding
schools for, 231
policy implications of research
ndings for, 273274
Emergence, 147, 153, 283n6
Encapsulation, 279280n3
English language usage
ambiguous use of kingdoms in,
278n3
OED devolution study, 3947, 259
Enlightenment, 2627
Environmental decision making, policy
implications of research ndings
for, 274
Epidemiology, cultural. See Cultural
epidemiology
Epstein, J., 153
Essentialism, 2122, 219. See also
Generic species
Ethnography. See Anthropology
Ethology and biology as module of
mind, 115
Evolutionary context of cognitive
function and folkbiology, 6465
Evolutionary taxonomy, classical,
277n1. See also Scientic vs. folkbiological taxonomies
Experience
development of inductive reasoning
and (see under Development of
folkbiological cognition in children)
extinction of, 1, 17 (see also
Devolution)
independence of biological induction
from, 116
Experts in biology
birds study, expert knowledge
advantages shown by, 8889
information transmission and, 213
214
networks, cultural epidemiology of,
209212, 211
perceptual experience, independence
of biological induction from, 116
326
Experience (cont.)
social learning and, 215219
as study population, 52
Extinction of experience, 1, 17. See
also Devolution
Exxon-Valdez oil spill, 264
Familiarity hypothesis, 129, 130
FBS (folkbiological system), 6364
Feldman, M., 283n7
Fish studies
dierent cultural models of sh and
shing (see under Intergroup
conict and mental models in
North America)
typicality, and ideals, 92101
Fodor, J., 279280n3
Folkbiological system (FBS), 6364
Folkbiology and folkecology, 116
child development of understanding
of (see Development of folkbiological cognition in children)
cognitive function of (see Cognitive
function and folkbiology)
common resources, cultural dierences in use of (see Common
resources, Mesoamerican cultures
and use of )
culture and individual, relationship
between, 89, 1415
dened, 14
devolution in knowledge of (see
Devolution)
fragility of environment and, 23,
14
interdisciplinary approach to, 410,
1415, 267, 275
as naive form of scientic biology,
1213
reasons for studying, 34
scientic taxonomy vs. (see Scientic
vs. folkbiological taxonomies)
theoretical issues in, 1013
universals of, 1925, 259
variability of, 260261
Folkpsychological and folkmechanical
systems, 6364
Folkspecics, 3335
Index
Index
327
328
Index
Index
329
330
Index
Index
331
332
Index
Index
333