Futures 41 (2009) 6–12

Contents lists available at ScienceDirect

Futures
journal homepage: www.elsevier.com/locate/futures

Ecological complexity, fuzzy logic, and holism in indigenous knowledge

A R T I C L E I N F O A B S T R A C T

Article history: Some indigenous knowledge is said to be holistic in the way it deals with the environment.
Available online 19 July 2008 Given the difficulties of Western science with complex environmental problems, any
insights from the holism of indigenous knowledge are of major interest. Based on
examples from Inuit and other northern peoples, it appears that indigenous knowledge
approaches complex systems by using simple prescriptions consistent with fuzzy logic.
Specifically, indigenous knowledge pursues holism through the continued reading of the
environment, collection of large amounts of information, and the construction of
collective mental models that can adjust to new information. Such an approach serves the
assessment of a large number of variables qualitatively, as opposed to focusing on a small
number of variables quantitatively.
ß 2008 Published by Elsevier Ltd.

1. Introduction

It is said that indigenous knowledge is holistic, but not many studies have documented how this holistic approach works
in practice. Exceptions include studies of Polynesian navigation by Turnbull [26] and Balinese irrigation by Lansing [13]. In
dealing with ecosystems, the holism of indigenous knowledge is in the context of complexity. Ecosystems are nested systems
(e.g., a small watershed inside a larger watershed and so on) and typically show variation in spatial and temporal scales,
making them extremely difficult to predict and control [14]. Ecosystem management is a postnormal science, as Funtowicz
and Ravetz might put it, a poor fit with conventional positivistic science that works best when systems are bounded and
control is possible [8]. Western science-based societies have tended to simplify ecosystems to be able to manage them (e.g.,
agriculture) and dampen natural variability and renewal cycles – but at the cost of impairing the long-term health and
resilience of ecosystems [11].
The knowledge and practices of indigenous societies are of significance in this context. Even though indigenous
knowledge does not have the techniques and quantitative tools at the disposal of Western science, some systems of
indigenous knowledge seem to have developed ways to deal with complexity. Any insights from indigenous wisdom in
regard to ecosystems are of huge potential interest, given that modern society has not been particularly successful in
managing ecosystems sustainably.
Diamond [7] and others have shown that many ancient societies did not live in harmony with their environment, and not
all indigenous groups necessarily do so in the contemporary world. However, where traditional societies have co-existed
with their ecosystems for a long time, there is strong evidence that this is not explainable solely in terms of low numbers of
people and low-impact technology. The accumulation of evidence, especially in the last two decades or so, indicates that
some indigenous groups have resource-use practices that suggest a sophisticated understanding of ecological relationships
and dynamics [3]. Such practices tend to be backed up by worldviews that see human as part of the ecosystem, based on
relationships of respect for the land and for living beings [2, Chapter 5].

* Corresponding author. Tel.: +1 204 474 6731; fax: +1 204 261 0038.
E-mail address: [email protected] (F. Berkes).

0016-3287/$ – see front matter ß 2008 Published by Elsevier Ltd.

Further, there is evidence that indigenous resource-use practices can evolve over time by adaptive learning, based on the
elaboration of a gradual understanding of the environment as well as on lessons learned from mistakes [4]. Geographic
variations of such sophisticated management approaches are found, for example, in reef and lagoon tenure systems in
Oceania [12]. Even though our understanding of indigenous ways of knowing and knowledge systems is still rudimentary,
the picture that is beginning to come into focus indicates that some traditional societies have experience in reading
environmental variables to deal with ecological complexity. But it is difficult to see how these indigenous approaches might
work. How do they do it?
Gadgil and his colleagues suggested that the key might be the use of ‘‘rules of thumb’’, simple prescriptions based on a
historical and ever-expanding cultural understanding of the environment. These rules are often backed up by religious belief,
ritual, taboos and social conventions. In the area of what science would call biodiversity conservation, for example, there
seems to be four ‘‘rules of thumb’’ used in various indigenous societies. These are (a) the total protection of certain selected
habitats (e.g., sacred forests and other customary sanctuaries), (b) the total protection of certain species of animals or plants
(e.g., taboo species), (c) prohibitions concerning vulnerable stages in the life history of certain species (e.g., hunting taboos in
south India for fruit bats at daytime roosts), and (d) practices of monitoring of populations and their habitat [9].
Such prescriptions in the form of rules of thumb have the advantage of turning complex decisions into rules that can be
remembered easily and enforced locally through social means. Traditional reef and lagoon tenure systems in Asia-Pacific,
with their taboo areas, taboo species and ritually announced opening and closing dates for permissible harvests, operate
under the same logic [12]. The First Salmon ceremony of many indigenous groups of the Pacific Northwest of North America
may have functioned in a similar way. A ritual leader and his helpers would watch the salmon swimming up the river and
presumably make a qualitative assessment of a sufficient number of mature fish (spawners) escaping upstream to
perpetuate the population. Then the fishery would be declared open and the event marked by a ceremony [23]. It may be
argued that such a system, led by an experienced leader, can produce results similar to one achieved by a biological
management system with population models, counting fences, daily data management and harvest quota enforcement – but
without the whole research infrastructure, quantitative data needs, and associated costs.
Rules of thumb that cut across complexity, as in reef and lagoon taboos, is one way to grasp how indigenous knowledge
systems might deal with ecosystem complexity. Another is to think of local knowledge as fuzzy cognitive maps that are
qualitative models of a system consisting of variables and the causal relationships between those variables [20]. A third and
related approach is to use fuzzy expert systems. Mackinson developed a model that utilizes fuzzy logic to capture and
integrate scientific and local knowledge in the form of heuristic rules to fish herring [15]. Similarly, looking at local and
traditional knowledge systems in the Caribbean island state of Grenada, Grant and Berkes identified ten categories of
knowledge that are important for finding and catching tunas and other large pelagic fish. Conceptualized as a decision-
making rule structure, these 10 categories of information, along with the feedbacks among them, can be thought to
constitute an expert system [10].
Hence fuzzy logic appears to be a good fit with indigenous knowledge, and an approach that may help understand, or
provide insights, on the question of how local and indigenous knowledge systems may be dealing with complexity. Our
emphasis in this article is not fuzzy cognitive maps or fuzzy expert systems but fuzzy logic as initially conceived by Zadeh
[27,28]. The paper pursues the idea that indigenous knowledge is able to deal with ecosystems as complex adaptive systems
by using simple prescriptions, consistent with fuzzy logic thinking.
First we explore the nature of indigenous knowledge, and the similarities and differences between such knowledge and
science. Language-based qualitative nature of indigenous knowledge establishes parallels with fuzzy logic models. After
covering some fuzzy logic basics, we discuss the idea that indigenous knowledge pursues holism by considering a large
number of variables qualitatively, while Western science tends to concentrate on a small number of variables quantitatively.
The examples and illustrations are from northern Canada, involving four northern indigenous groups, the Inuit, the Inuvialuit
(western Canadian Arctic Inuit), the Cree, and the Dene. Northern Canada serves as a suitable ‘laboratory’ for this discussion
because there is a great deal of material available, and because of our own interest in the area over three decades.

2. Nature of indigenous knowledge

Indigenous knowledge is a body of knowledge built up by a group of people through generations of living in close contact
with nature. It is local knowledge held by indigenous peoples or local knowledge unique to a given society, including some
non-indigenous ones. When the knowledge is of ecological nature (and not all traditional knowledge is) one may use the
term, traditional ecological knowledge. The working definition we have used for this term is ‘‘a cumulative body of
knowledge, practice and belief, evolving by adaptive processes and handed down through generations by cultural
transmission’’ [2, p. 7].
Every society has its own understanding of how the natural world works, a repertoire of habits, skills, and styles from
which members of a society construct their livelihoods. As a knowledge–practice–belief complex, indigenous knowledge
includes an intimacy with local land, animals, and plants. It also includes institutions (rules and norms) about interacting
with the environment, and it includes worldview, as the worldview shapes the way people make observations, make sense of
their observations and learn. These levels of ecological knowledge may overlap, change over time, and interact with each
other. For example, the worldview informs what new knowledge is legitimate and can be incorporated into resource-use
practice, what social sanctions and taboos on hunting and farming practices may be used [3,4].
8 F. Berkes, M.K. Berkes / Futures 41 (2009) 6–12

For the insiders, indigenous knowledge is often a way of life and a way of knowing. For Western analytical purposes, there
is a distinction to be made between knowledge as process, as opposed to knowledge as the thing known, consistent with
Gregory Bateson [1]. When Northern indigenous participants were asked to describe traditional knowledge, there was
consensus on the following meanings that seem to include both the process of knowing and the knowledge itself: practical
common sense; teachings and experience passed through generations; knowing the country; rooted in spiritual health; a
way of life; an authority system of rules for resource use; respect; obligation to share; wisdom in using knowledge; using
heart and head together [2].
There are both similarities and differences between indigenous knowledge and Western science. Both kinds of knowledge
are ultimately based on observations of the environment, both provide a way of knowing based on these observations, and
both emerge from the same intellectual process of creating order out of disorder. But there are also differences. Indigenous
knowledge, as an integral part of culture, tends to have an explicit social context; science does not. Indigenous knowledge
traditions have their own rules about processes of knowing, and these tend to be different from the rules of science regarding
evidence, repeatability and quantification.
Highlighting the last point, not all kinds of Western scholarship use numbers, but quantification is certainly one of the
hallmarks of modern science. In many indigenous knowledge systems, by contrast, quantification is not emphasized; often it
is not even considered important or relevant. And we know this from personal experience, indigenous knowledge holders are
often baffled by the preoccupation of scientists to count and measure everything. What they value is the understanding of
the environment, how to read and interpret signals from the environment, and the relationships within it, including those
involving humans [2].
With regard to quantification, one can take the position that indigenous knowledge and Western science are different; at
the same time, one can also take the position that they are complementary. Arguably, quantitative vs. qualitative
understanding of phenomena is a false dichotomy. Quantification complements qualitative understanding, and both kinds of
understanding together are more powerful than each alone.
Holism is the key characteristic of indigenous knowledge for the purposes of the present article. The following example
may serve as an illustration of holistic thinking. The Arctic region has been undergoing rapid environmental change in the
past few decades. Northern indigenous observations of climate change and abnormalities in animals (which may be related
to Arctic ecosystem contamination), once dismissed by scientists, are being taken seriously. Indigenous observers do not
carry out chemical tests for pollutants or gather quantitative data. But their ways of observing and assessing environmental
changes provides insights regarding indigenous holism.
O’Neil et al. worked with the Inuit of Hudson Bay region in northern Canada and documented how they made sense of the
contaminants issue. Their major concern was the observation of abnormalities in many animals, focusing on seals. The
diagnosis of a sick animal relied on many locally developed indicators. The Inuit knew which animals were sick or abnormal.
They had a sense of what normal animals should look like, based on their collective experience over many years. They made
reference to specific signs that indicated that an animal was not well and should not be eaten: animals with manimiq (lumps),
discoloured bones, abnormal liver, bumps and blueish spots in the intestines, and skinny animals. The diagnosis did not rely
only on anatomical observations and the quality of the meat; the Inuit also observed the behaviour of the animal, its feeding,
swimming and response to predators, reading signs of wellness continuously and cumulatively. Interpretations of elders
were given weight, presumably because the Inuit expected that experienced hunters would have a more finely developed
sense about the state of health of an animal [19].
Indigenous knowledge holders accumulate such information as a result of many years of observations (analogous to
extensive sampling), the sharing of knowledge with other hunters and fishers (data pooling), and forming a collective mental
model of what healthy animals would look like. Their ‘‘data’’ on animal health and abnormalities are language-based, rather
than numbers-based, and comparisons are performed on perceived ranks (e.g., fat, thin, very thin). How these data are
characterized and used strongly depend on the terminology used.
The mental processes of ‘‘data collection’’, concept formation and retention, and mental model formation among
indigenous people follow patterns consistent with the language used, as language shapes terms and concepts. For example, it
is well known that the Inuit do not attach much value to numerical precision. They also do not often appear to make simple
linear, cause-and-effect type connections, as often done in the Western science. Rather, they see environmental change and
related observations as empirically connected. Systematic generalizations regarding cause–effect relationships are in
general regarded negatively. According to the Inuit worldview, making simplifications and generalizations of complex
phenomena is ‘‘childish’’ and without sense (without ihuma) [18].
We argue that holistic thinking among the Inuit is possible because precise categorizations and generalizations are
avoided. If all the relationships embedded in a holistic term were to be specified, the whole idea would become
unmanageably complex. There seems to be an inverse relationship between the complexity of a system and the degree of
precision that can be used meaningfully to describe it. Zadeh [28, p. 28] calls this idea, the principle of incompatibility: ‘‘as
the complexity of a system increases, our ability to make precise and yet significant statements about its behaviour
diminishes until a threshold is reached beyond which precision and significance (or relevance) become almost mutually
exclusive characteristics.’’
It is in this sense that precise quantitative analyses of the behaviour of complex systems are not likely to have much
relevance to the real world. Some kinds of indigenous knowledge work because (a) there are large amounts of information;
(b) it is collected continuously; and (c) changes are incorporated into the collective mental model as new information flows
 F. Berkes, M.K. Berkes / Futures 41 (2009) 6–12 9

in. Herein lies the essential similarity of indigenous knowledge and fuzzy logic. The three points above are also the backbone
of fuzzy logic models. In both cases, the analysis of the complex system behaviour is carried out, not by using numerically
precise data, but by using language-based data that are qualitative and rich.

3. Fuzzy logic and indigenous knowledge

Fuzzy sets, a generalization of set theory, were developed as a way to represent the imprecise nature of information in
everyday life. To use an illustration offered by Bezdec [5], suppose you are approaching a red light and you must advise a
driving student when to apply the brakes. Would you say, ‘‘apply the brakes 25 m before the red light’’, or would you say,
‘‘Brake pretty soon’’? The latter of course, because the former instruction is too precise to be implemented. In most everyday
situations, precision may be quite useless while vague directions, consistent with the imprecise nature of data, can be
interpreted by the human brain and acted upon.
Practical applications of fuzzy logic (also referred to as fuzzy thinking, fuzzy models or fuzzy sets) in electrical and
computer systems include self-monitoring and adjusting ‘‘smart’’ systems such as intelligent cars, subway trains (e.g.,
Sendai, Japan) that can detect and adjust to changing conditions, and your humble rice cooker with fuzzy logic technology.
All ‘‘soft computing’’, including decision-support systems, Geographic Information Systems (GIS) and Geographic
Positioning Systems (GPS) use fuzzy logic [29]. Applications are found in computer science and artificial intelligence, in fact,
in almost all engineering fields [5].
Fuzzy logic is a mathematical approach for dealing with complex systems where only approximate information on
components and connections are available. It is a way to deal with uncertainty and uses rules of thumb. It is suitable for
concepts and systems that do not have sharply defined boundaries, or where the information is incomplete or
unreliable. It is an unusual approach because it breaks with the yes–no binary tradition used in most science and
computer applications.
Classical logic of precision, which forms the basis of Western science, was developed by the Greek philosophers, notably
Aristotle, whose works gave rise to the so-called ‘‘laws of thought’’. One of these is the ‘‘law of the excluded middle’’ which
forms the basis of later Cartesian logic. This law states that every proposition has to be either true or false. It is thought that
the tendency of Western science to regard matters as either black or white stems from this bivalent yes–no logic and the
Cartesian tradition based on it. Even during Aristotle’s time, there were some philosophers, Plato among them, who did not
agree with the law of the excluded middle. Plato postulated that beyond true and false, there was a middle ground. One could
say that if Cartesian science is based on Aristotle, the idea of fuzzy logic is based on Plato.
In fuzzy logic, things need not be precisely defined or quantified before they can be considered mathematically; fuzzy
logic models do not need precise information inputs. Like the human mind, fuzzy logic puts together related objects into
categories in such a way as to reduce the complexity of the processing task. Fuzzy logic provides the tools to classify
information into broad categorizations or groupings, simulating the workings of the human mind. According to the founder
of fuzzy logic, Zadeh [28, p. 28], ‘‘the premise is that the key elements of human thinking are not numbers, but labels of fuzzy
sets, that is, classes of objects in which the transition from membership to non-membership is gradual rather than abrupt.’’
Zadeh’s approach has three main distinguishing features: use of linguistic variables in place of (or in addition to)
numerical variables; characterization of simple relations between variables by fuzzy conditional statements; and
characterization of complex relations by fuzzy algorithms. A linguistic variable is defined as a variable whose values are
sentences. For example, if ‘‘fat, thin, very thin’’ are values for fatness, then fatness is a linguistic variable. Fuzzy conditional
statements are expressions in the form ‘‘IF a THEN b.’’ A fuzzy algorithm is an ordered sequence of instructions, as in a recipe
for chocolate fudge [28].
Fuzzy models rely on language-based information that is converted to simple mathematical expressions that can then be
manipulated to make mathematical inferences. For the purposes of indigenous knowledge, we need not be concerned about
the mathematics of fuzzy sets. The important consideration is language. To take the example of Inuit observations of ‘‘skinny
seals’’ or ‘‘fat fish’’, fatness is the linguistic variable and the adjectives are linguistic values. Qualifying terms such as ‘‘and’’,
‘‘or’’, ‘‘not’’ are called linguistic connectives.
Using the example of Inuit observations of unhealthy seals, let us say that experienced hunters among the Inuit of Hudson
Bay have been hunting large numbers of seals (a large sample size) and finding that many of them have abnormalities. Over
several hunting seasons (continuous set of observations), hunters have noticed that some seals are thin, some have
discoloured bones, and some have abnormal livers. After a while, the experienced hunters would begin to formulate an
opinion about the general goodness of the seals to eat.
To put it in fuzzy model terms, the hunters observe seal fatness (variable 1) during the sampling. There is an existing
mental model of the various values (different degrees of fatness/thinness) of this variable from experience and collective
memory of experienced hunters and elders. Each seal is evaluated instantaneously against this model. The seals may be
assessed to be generally thinner, and variable 1 is assigned a fatness/thinness value. In fuzzy models, it is assigned a certain
weight between 1 and 10. Other variables such as discoloured bones (variable 2), condition of livers (variable 3), and so on,
are assigned different weights, based on the existing mental model of a healthy seal that is good to eat.
All the relevant variables according to the model are weighted. IF variable x has a degree of thinness a, THEN the seals are
assigned a fuzzy conditional statement. This type reasoning is used for all the variables specified, and helps evaluate the
suitability of the seal for eating. In fuzzy models, several ‘‘IF a THEN b’’ type statements are used simultaneously and in a
10 F. Berkes, M.K. Berkes / Futures 41 (2009) 6–12

flexible way, changing the weightings as observations accumulate. Although we do not attempt this here, fuzzy models
are able to quantify (by assigning numerical values or weights) the qualitative judgments of hunters based on their expertise.
Note that Inuit observations on the fatness/thinness of seals (and other variables) are not quantitative. For purposes of the
kind of science that requires quantitative data to make conclusions, Inuit observations are difficult to use. Critics may dismiss
them as anecdotal, or worse, irrelevant. Rather than requiring quantitative data, fuzzy logic is able to work with the
approximate values assigned to the categorizations of fatness/thinness that can be inferred from the language used by the
hunters. Using weightings and mental models of Inuit experts, and then assigning numerical values to these, fuzzy logic
simulates human judgment in making sense of a large number of variables.

4. Making sense of indigenous knowledge as fuzzy logic

The illustration of fuzzy logic using the case of Inuit indigenous knowledge of seal wellness, stops short of addressing one
key question: what is the advantage of valuing Inuit knowledge in such a case, if quantitative scientific data can be obtained
instead? The short answer is that both kinds of knowledge are desirable because they extend the range of information
available. The long answer is, the two kinds of knowledge have different relative strengths and a potential for
complementarity. Indigenous knowledge appears to bring some unique advantages in dealing with multiple variables and
complexity.
In this section, we illustrate this point, first by further pursuing the example of Arctic contaminants, and extending the
seal wellness case. Second, we use the example of Inuvialuit observations of climate change in the western Canadian Arctic to
make the point that the ability of indigenous knowledge to deal with many variables is valid for different kinds of
environmental knowledge. The example also illustrates that indigenous knowledge complements science with scale-specific
information.
Based on the available literature, Donald Cobb and colleagues compiled a comprehensive table of environmental quality
indicators used in the study of contaminant-related effects in fish and marine mammals, one column for indicators used by
scientists (toxicologists), another column for indicators used by northern indigenous knowledge holders. The two sets of
indicators, as used by the two knowledge systems, assess environmental conditions and wellness in their own way [6].
The science of toxicology uses many indicators at the chemical, biochemical and cellular levels. At these levels,
indigenous knowledge is generally not very useful, except that some northern indigenous people are apparently able to
taste and smell some contaminants, or effects of contaminants, in animal tissues. However, at the levels of the whole
organism, population and community, local observations and indigenous knowledge seem to provide a range of indicators
comparable to science. Some toxicological effects, such as physiological changes, are not observable to hunters. However,
such physiological effects may express themselves as behavioural effects, and the Inuit are experts on reading animal
behaviours. The review found some indicators that may be noted by indigenous observers and not normally studied by
science. These include the detailed observations of different kinds of fat on the body of an animal that are used as indicators
of health [6].
A major difference between science and indigenous knowledge is that toxicological studies tend to work with a single
analytical tool at a time, focusing on one or a small number of indicators. By contrast, indigenous knowledge focuses on a
large number of less specific (and probably multicausal) indicators used simultaneously as a suite. Whereas scientific
approaches seek indicator specificity and result in quantitative studies of a small number of indicators, those based on
indigenous knowledge do not produce formalized generalizations. But the use of a broad suite of simpler indicators (instead
of a few detailed and costly ones) gives the hunting community feedback on many aspects of environmental health, a holistic
picture of the environment. The use of a broad suite of simple indicators provides built-in adaptability in a fuzzy logic sense.
That is, such community-based indicators could be readily modified with changing conditions, providing flexibility in the
collective mental models used and interpretations made.
The Inuit Observations of Climate Change study further illustrates the ability of indigenous knowledge to deal with
multiple variables and complexity, and shows that local observations can provide information at the appropriate spatial
scale to complement science [22].
Most scientific studies of climate change use models to generate projections at the global and regional levels. Scientific
field studies on climate change typically provide discipline-specific results, based for example on climatic variables
(temperature, precipitation, etc.) or based on plants (distributions, changes in treeline) or based on different groups of
animals (changes in food and physiology, migration patterns, etc.). By contrast, the Inuvialuit offer a full range of
observations on all aspects of climate change, similar to the broad suite of simple indicators in the Arctic contaminants case.
As summarized by Mark Nuttall and colleagues, the Inuvialuit of one small Arctic community, Sachs Harbour, are able to
provide some 25 kinds of evidence of climate change, spanning the areas of physical environmental change; predictability of
the environment; travel safety; access to resources; and changes in animal distributions and condition. In the area of
physical environmental change, for example, Sachs Harbour Inuvialuit noted the reduction of sea-ice cover and lack of
multiyear ice around the community in summer, and the thawing of permafrost. All of these changes made environmental
predictability, travel safety and resource access more difficult. Hunters rely on their ability to predict the weather and the
timing of animal movements, and to read snow and ice conditions (‘‘is the ice firm and safe to travel?’’). Increased ice
movements in winter and spring, changes in the distribution of leads, cracks and pressure ridges, as well as overall thinning
of the ice, have all affected access to resources [17].
 F. Berkes, M.K. Berkes / Futures 41 (2009) 6–12 11

The collective mental models of the Inuvialuit are very exacting in combining the variables they consider relevant. For
example, sea-ice observations are evaluated jointly with changes in wind frequency and strength. Weather instrumentation,
along with remote sensing, provides precise information on each of these variables but not on the combination of these
variables [16]. Whereas global models produce results based on average change, the Inuvialuit consider numerically precise
‘‘average change’’ as more or less irrelevant, and focus instead on the magnitude and frequency of extreme weather events,
variability, and predictability [17].
Complex systems phenomena, such as climate change, occur at multiple levels, and there is no one correct level of
analysis. The system must be analyzed simultaneously across geographic scale, from the global to the local. But the relative
emphasis of science has been at the global level. The fact that indigenous knowledge provides local-level understanding is
particularly important because it complements science precisely at the level where information is poor.

5. Prospects for indigenous knowledge

‘‘Not everything that counts can be counted, and not everything that can be counted, counts’’ (Albert Einstein)
We started by asking how indigenous knowledge develops holistic approaches, and focused on fuzzy logic as a way to
explain how rules of thumb and other simple prescriptions can be used to deal with complexity. It is well known in the
theory of complex adaptive systems that complexity can emerge from simple rules [14]; conversely, it seems, simple rules
are appropriate for dealing with complex adaptive systems. How does it work? Indigenous knowledge seems to build
holistic pictures of the environment by considering a large number of variables qualitatively, while science tends to
concentrate on a small number of variables quantitatively. It is a tradeoff captured in Zadeh’s [28] principle which says
that an inverse relationship exists between the complexity of a system and the degree of precision that can be used
meaningfully to describe it.
It should be qualified that not all kinds of indigenous knowledge are holistic, and obviously not all Western science is
reductionistic. Some (but not all) kinds of ecology, systems science, gestalt psychology, quantum physics and fuzzy logic are
examples of holistic areas within the predominantly reductionistic tradition of Western science. The dominant tradition,
captured in the ‘‘normal science’’ terminology of Funtowicz and Ravetz, is appropriate for relatively simple, controllable
systems [8]. By contrast, there is a large degree of uncertainty in complex adaptive systems in which ‘‘not everything that
counts can be counted’’. This limits the ability to make precise (and yet significant) statements about system behaviour.
Complex adaptive systems cannot be readily comprehended by the use of conventional science – simply because we would
be overwhelmed by data and not even be sure that the data are meaningful. Zadeh’s insight was to recognize the nature of
this problem and depart from precision, ironic for an engineer. In the process, his theories were apparently not well received
by some major US corporations but embraced by some Japanese companies that ran with them. The rest is soft computing
history.
Our analysis suggests that Zadeh’s solution is a good fit with indigenous knowledge and its holistic treatment of
ecosystem complexity, with the use of rules of thumb and broad suites of simple indicators, as in the Arctic contaminants
case. The science of environmental toxicology is well developed, but there is an increasing recognition that the use of a few
indicators, no matter how well chosen, may be inadequate in capturing complexity [6]. The indigenous solution is brilliant in
building a holistic understanding by monitoring a large number of indicators over a long period of time, accumulating and
accessing a large amount of qualitative data, and building a collective mental model of what healthy animals and
environment ought to look like. Even in a highly analytical science such as toxicology, one can see the potential contribution
of qualitative understanding provided by indigenous knowledge.
Similarly in the area of climate change, Inuvialuit observations cannot replace scientific measurements and models. But
they can contribute to the overall understanding of the system, complementing science by filling in the otherwise missing
local scale, providing baseline information, helping formulate research questions and hypotheses, providing insights
regarding impacts and adaptation, and supplying community-based monitoring [22]. The crucial part of developing
indigenous understanding was the ability to assess the veracity of change, based on their mental model of the predictability
of the environment, the range of expected variation, and the expected frequency and severity of extreme weather events. On
all three counts, the expected environmental variability in the collective mental model was exceeded in the 1990s and
onwards [16,17,22].
Based on our analysis, indigenous knowledge can be analyzed as fuzzy logic and quantified with fuzzy sets, if it were
desirable to do so. We do not attempt to do that here, in part because we are concerned that to do so will take indigenous
knowledge out of its cultural context and add a level of analysis that is not appropriate. In any case, even without the
quantification into fuzzy sets, holistic insights of indigenous knowledge are being recognized from northern Canada to
southern India in environmental assessment and other fields [21]. The challenge is to find appropriate ways of bridging
Western science and indigenous knowledge without absorbing the diversity of knowledge traditions into one dominant
science [2,21,25].
Indigenous knowledge is a challenge to the positivist–reductionist paradigm [2] and to the essential question of what
constitutes knowledge [25]. David Turnbull has argued that when local knowledge is probed deeply, ‘‘in no case does it come
out looking like the standard Western notion of information;’’ rather, it tends to be a ‘‘blend of knowledge, practice, trusted
authority, spiritual values, and local social and cultural organization: a knowledge space’’ [25, p. 560]. It is a lesson for futures
12 F. Berkes, M.K. Berkes / Futures 41 (2009) 6–12

studies that the ancient wisdom of indigenous peoples is such a good fit with fuzzy logic, a very new science – sometimes you
have to go back to go forward [3]. The holistic approach is the key.
The basic idea is captured by Bateson who has criticized Cartesian science for creating false dualities such as the split
between mind and nature (or nature and culture). Bateson observed that ‘‘The continuum of nature is constantly broken
down into a discontinuum of variables in the act of description’’ [1, p. 165]. The conventional scientific solution has been to
quantify a few of the variables, whereas the solution in indigenous knowledge has been to find ways of perceiving that
continuum of nature and working with it. For Westerners, holistic Western sciences such as fuzzy logic help comprehend the
concept. Indigenous knowledge holders do not need fuzzy logic to understand holism; they already practice it.

Acknowledgements

We thank the hunters and elders, too numerous to mention, of several communities of Inuit, Inuvialuit, Cree and Dene
peoples for sharing their understandings of the environment. We thank David Turnbull for providing comments on the
paper. This work was supported by the Social Sciences and Humanities Research Council (SSHRC) and the Canada Research
Chairs program (http://www.chairs.gc.ca).

References

[1] G. Bateson, M.C. Bateson, Angels Fear: Towards an Epistemology of the Sacred, Bantam Books, New York, 1987.
[2] F. Berkes, Sacred Ecology, Second Edition, Routledge, New York and London, 2008.
[3] F. Berkes, C. Folke, Back to the future: ecosystem dynamics and local knowledge, in: L.H. Gunderson, C.S. Holling (Eds.), Panarchy: Understanding
Transformations in Human and Natural Systems, Island Press, Washington, DC, 2002, pp. 121–146.
[4] F. Berkes, N.J. Turner, Knowledge, learning and the evolution of conservation practice for social-ecological system resilience, Human Ecology 34 (2006) 479–
494.
[5] J. Bezdec, Fuzzy models – what are they and why? Transactions on Fuzzy Systems 1 (1992) 1–5.
[6] D. Cobb, M. Kislalioglu Berkes, F. Berkes, Ecosystem-based management and marine environmental quality indicators in northern Canada, in: F. Berkes, R.
Huebert, H. Fast, M. Manseau, A. Diduck (Eds.), Breaking Ice: Renewable Resource and Ocean Management in the Canadian North, University of Calgary Press,
Calgary, 2005, pp. 71–93.
[7] J. Diamond, Collapse. How Societies Choose to Fail or Succeed, Penguin Books, New York, 2005.
[8] S. Funtowicz, R. Ravetz, Science for the post-normal age, Futures 25 (1993) 739–755.
[9] M. Gadgil, F. Berkes, C. Folke, Indigenous knowledge for biodiversity conservation, Ambio 22 (1993) 151–156.
[10] S. Grant, F. Berkes, Fisher knowledge as expert system: a case from the longline fishery of Grenada, the Eastern Caribbean, Fisheries Research 84 (2007) 162–
170.
[11] C.S. Holling, G.K. Meffe, Command and control and the pathology of natural resource management, Conservation Biology 10 (1996) 328–337.
[12] R.E. Johannes, The renaissance of community-based marine resource management in Oceania, Annual Review of Ecology and Systematics 33 (2002) 317–
340.
[13] J.S. Lansing, Priests and Programmers, Princeton University Press, Princeton, NJ, 1991.
[14] S.A. Levin, Fragile Dominion: Complexity and the Commons, Perseus Books, Reading, MA, 1999.
[15] S. Mackinson, Integrating local and scientific knowledge: an example in fisheries Science, Environmental Management 27 (2001) 533–545.
[16] T. Nichols, F. Berkes, D. Jolly, N.B. Snow, the Community of Sachs Harbour, Climate change and sea ice: local observations from the Canadian western Arctic,
Arctic 57 (2004) 68–79.
[17] M. Nuttall, F. Berkes, B. Forbes, G. Kofinas, T. Vlassova, G. Wenzel, Hunting, herding, fishing and gathering: indigenous peoples and renewable resource use in
the Arctic, in: Arctic Climate Impact Assessment, Cambridge University Press, Cambridge, 2005, pp. 649–690 (Chapter 12).
[18] K. Omura, Science against modern science: the socio-political construction of otherness in Inuit TEK (traditional ecological knowledge), Senri Ethnological
Studies 67 (2005) 323–344.
[19] J. O’Neil, B. Elias, A. Yassi, Poisoned food: cultural resistance to the contaminants discourse, Arctic Anthropology 34 (1997) 29–40.
[20] U. Ozesmi, S.L. Ozesmi, Ecological models based on people’s knowledge: a multi-step fuzzy cognitive mapping approach, Ecological Modelling 176 (2004)
43–64.
[21] W.V. Reid, F. Berkes, T. Wilbanks, D. Capistrano (Eds.), Bridging Scales and Knowledge Systems, Island Press, Washington, DC, 2006.
[22] D. Riedlinger, F. Berkes, Contributions of traditional knowledge to understanding climate change in the Canadian Arctic, Polar Record 37 (2001) 315–328.
[23] S.L. Swezey, R.F. Heizer, Ritual management of salmonid fish resources in California, Journal of California Anthropology 4 (1977) 6–29.
[25] D. Turnbull, Reframing science and other local knowledge traditions, Futures 29 (1997) 551–562.
[26] D. Turnbull, Masons, Tricksters and Cartographers: Comparative Studies in the Sociology of Scientific and Indigenous Knowledge, Harwood Academic
Publishers, Reading, 2000.
[27] L.A. Zadeh, Fuzzy sets, Information and Control 8 (1965) 338–353.
[28] L.A. Zadeh, Outline of a new approach to the analysis of complex systems and decision process, Transactions on Systems, Man and Cybernetics SMC-3 (1973)
28–44.
[29] A.X. Zhu, B. Hudson, J. Burt, K. Lubich, D. Simonson, Soil mapping using GIS, expert knowledge and fuzzy logic, Soil Science Society of America Journal 65
(2001) 1463–1472.