Magic, Science and Religion


Malinowski’s classic attempt to separate magic, science, and religion has not worn well.  For some purposes, we now can find it very useful indeed, at least insofar as it separates pragmatic knowledge from unverifiable belief.  However, as Malinowski admitted, traditional societies often categorize knowledge is ways very different from this.  Knowledge of and ethics relating to the nonhuman environment are particularly difficult to describe this way.  A different way to talk about traditional environmental knowledge is as involvement in the world—more or less intense cognitive and emotional relations with nonhuman as well as human “others.”  Often, natural objects are, or contain, spirits that are part of one’s society.  Trying to sort out magic, science, and religion in such cases is useful, and has certainly been useful in the past (in the history of science as well as in anthropology), but perhaps more useful today is to try to understand local concepts of knowledge and interaction.  I present examples from the Maya and the Northwest Coast Native peoples.



E. N. Anderson

Dept. of Anthropology

University of California

Riverside, CA 92521-0418

“Yesterday’s ‘science’ is today’s ‘common sense’ and tomorrow’s ‘superstition.'”

–anonymous proverb, heard among scientists

“Spirituality is more than the awareness of one’s self.  It is the awareness of and responsibility for knowing your place and role in the world.  It is about being aware of the interrelatedness of all that was, is and will be.  It is about knowing your responsibilities for the past, present and future.”  Adrian Tucker (Australian Aboriginal), 1997; quoted in display, Australian Museum, Sydney.


Malinowski’s classic attempt to separate magic, science, and religion has not worn well.  For some purposes, we now can find it very useful indeed, at least insofar as it separates pragmatic knowledge from unverifiable belief.  However, as Malinowski admitted, traditional societies often categorize knowledge is ways very different from this.  Knowledge of and ethics relating to the nonhuman environment are particularly difficult to describe this way.  A different way to talk about traditional environmental knowledge is as involvement in the world—more or less intense cognitive and emotional relations with nonhuman as well as human “others.”  Often, natural objects are, or contain, spirits that are part of one’s society.  Trying to sort out magic, science, and religion in such cases is useful, and has certainly been useful in the past (in the history of science as well as in anthropology), but perhaps more useful today is to try to understand local concepts of knowledge and interaction.  I present examples from the Maya and the Northwest Coast Native peoples.


1.  Preface

In Nanjing, China, a university has offered a course in the ancient Chinese art of fengshui.  Fengshui, “wind and water,” involves planning houses, communities, and graves in accord with the ways of wind and water—both the real and tangible ways known to all the world, and the supernatural or imaginary ways that Chinese cosmology postulates.  This course has set off a storm of protest (Lee 2005).  One professor says “It’s a fake science… It only makes money for some swindlers” (Lee 2005:1).  Others denounce it as “superstition” (mixin in Chinese).  Still others laud it as a glorious achievement of Chinese culture.

The problem is that fengshui, as we know it today, is a single, unified field, but it has elements of what we of the western academic world would consider “magic,” “science,” and “religion.”  These distinctions have proved useful for anthropology, but they are confounded by traditional belief systems like fengshui, causing intractable problems for anthropologists as well as for Chinese university administrators.

Bronislaw Malinowski’s Magic, Science and Religion (1948) was actually the first anthropology book I read.  My father had a copy—I don’t think he ever read it.  As a high school student I discovered it and worked my way through it.  I must say it failed to engage my interest.  I got converted to anthropology a couple of years later, through visiting other countries and then discovering the great ethnographies of the Boas era.

Malinowski retailed the old Frazerian distinction:  magic is routine formulas that are supposed to work in the real world, but by supernatural mechanisms.  Science is the pragmatic knowledge based on observation and experience.  Religion is high-flown, systematized belief, involving supplication rather than routine instrumental manipulation.

Other definitions of magic had been offered.  Most notably, Hubert and Mauss in 1902 had concluded that magic is “any rite which does not play a part in organized cults—it is private, secret, mysterious…” (Mauss 2001:30, emphasis in original), but also emphasized it was social.  However, they too, in practice, regarded magic as the routine, formulaic stuff that is supposed to work automatically, and religion as more related to prayer, supplication, and abstract belief.

Since Malinowski, the tendency has been to sink “magic” into religion, or even to ignore it, and not much theorizing about it is being done today.  In spite of Mauss, I tend to agree with that tendency; I am not convinced of the existence of “magic” as a meaningfully isolable cross-cultural entity.  It seems to be merely the practical, working side of dealing with supernatural entities.

Malinowski, being no fool, questioned and qualified the distinction between magic, science, and religion—”problematized” it, as the postmodernists say.  However, he did not really reject it.  He was at some pains to show that the people he was still calling “savages” had just as solid, experiential, and pragmatic a knowledge of the world as any “civilized” folk.  He was attacking the view (he cited it to Lévy-Bruhl but pointed out that the view was widespread) that traditional people live in a “prelogical” world of mysticism, blind belief, and animal-like intuition.

Malinowski was guilty of many outrageous statements, such as “The road from the wilderness to the savage’s belly and consequently to his mind is very short, and for him the world is an indiscriminate background against which there stand out the useful, primarily the edible, species of animals or plants” (p. 44).  This was retrograde even at the time; no Boasian would have been caught dead writing such a line.  But, in general, Malinowski was on the side of the angels in this book.  He was hardly the first to say that traditional people had a wealth of solid empirical knowledge and high-flown spiritual experience—anthropologists from Morgan on had said that—but he was one of the more influential people saying it.

Malinowski knew that separate categories of “magic,” “science” and “religion” did not exist in Trobriand thought or any other traditional knowledge system.  He was quite aware that he was projecting a 20th-century European distinction onto systems that, themselves, did not really have anything like it.  However, he found the categories useful.  The time has come to question this.

2.  Religion

Malinowski saw religion as basically a way of utilizing belief in inferred, imaginary supernatural beings and forces to satisfy the emotional needs of the “savage.”

He said nothing about the civilized folk, but one assumes he was, as usual, sideswiping them via the Trobriands.  He never forgot his self-imposed mission to confront his elite European readers with an ironic reflection of themselves.  One can also assume that, like many early-20th-century social scientists, he expected religion to wither away in the near future.

Malinowski’s “religion” was strongly individualistic and psychological.  He rejected Durkheim’s  idea of religion (Durkheim 1995/1912) as the projection of society, and, by implication, Marx’ somewhat similar (though materialist) view.  He dismissed Durkheim’s theory as mere mysticism, which, along with much else, proves that he did not understand Durkheim very well.  (This is nothing against Malinowski—Durkheim did not go out of his way to make his case easy to follow.  A frequent mistake, not well avoided by Malinowski, is to claim that Durkheim himself thought society was a mystical reality!  No, he thought that Australian aboriginal rituals represented it so.)

I think most anthropologists today would agree with Durkheim that religion is an emergent phenomenon of society, and is, by definition, a system.  It has structure and social institutionalization.  The lone-individual side of religious sentiment is now called “spirituality” rather than “religion.”  The religious and cosmological belief system of a given society tends to reflect that society more or less closely.  Sometimes, to be sure, there are time lags; Chinese religion still sees a Heavenly Emperor with his magistrates, courtiers, and classic dancers, reflecting the reality of the Qing Dynasty (1644-1911).  Revolutionary change comes slowly in Heaven.  On the other hand, religion can and does change fast to accommodate new conditions; the meteoric rise of “fundamentalist” hate-based violence in the name of religion is a natural and, I think, inevitable reflex of the rise of hierarchies and giant government-business hookups in the late 20th century, with the consequent destruction of face-to-face communities, reduction of individual agency, and rise of mass violence.

Belief in supernaturals is probably a rather natural concomitant of ordinary human thought (Atran 2002).  Belief in supernaturals may stem from a human tendency to ascribe all actions to deliberate agency until proven otherwise, as Atran and many others suggest, and from an analytically separate tendency to extend humanity to everything—to anthropomorphize. People naturally assume human-like agency as a default, whenever explaining anything—a point made as early as Kant (1976/1796).  It makes evolutionary sense for people to look for someone causing things that happen in the world—rightly in cases of war and love, wrongly in cases of stars moving and glaciers forming.  This idea is already in Malinowski and famously in Evans-Pritchard, but is better discussed by Atran (2002), Bering (2005), and other recent writers.  Bering points out that children are more “religious” than modern western-educated adults, in that they are more prone to attribute ordinary events to unseen but thinking and wilful agents (supernaturals).   The “religious” emotions, moods, and feelings seem not a problem; they are not fundamental or homogeneous, and they occur in and probably are adapted to quite other situations.  But they are available for use in selling moral codes and ideological systems.

Certainly, religion acts in society to provide powerful sanctions for morality.  Supernatural beliefs are bent to use in promoting ethics.  This is one of the most basic functions of religion as we know it.  As Durkheim pointed out, religion thus represents ethics, and makes ethical ideas real and compelling and immediate.  Crudely, religion sells social morality.  It sells the social code and mythologizes the social contract.  This is probably its sharpest difference from science (though see below).  Ethics in religion is basic and inevitable, and is, to many, the whole reason to have religion at all.

Religion is now almost invariably defined as belief in supernatural beings (see Atran 2002).  However, Malinowski (and others of his time) differentiated religion from magic, which also depends on supernaturals.  And belief in supernatural beings is generally not considered adequate to make a religion.  For Scott Atran, author of one recent major book on the anthropology of religion, Mickey Mouse doesn’t count, and neither do devoutly held but allegedly “factual” or “scientific” belief systems like Marxism; religion must involve not only supernaturals but also counter-evidential beliefs and emotional sacrifices (Atran 2002:13).  Others disagree, finding Marxism and “capitalism” more like religion than like science or spirituality (Atran 2002:13; David Kronenfeld, personal communication over the years, most recently on drafts of this paper, 2004; Kronenfeld points out that ideological concepts, such as the Dialectic in Marxism, take on supernatural and counter-evidential qualities when made into mass slogans).

In general, most anthropologists have defined religion as belief in supernaturals.  Many other social scientists, on the other hand, seem to stay with Durkheim, and define religion as a social institution characterized by rituals and moral codes.  However, Malinowski, as well as many historians and theorists of religion, saw religion as basically defined by spiritual emotion—specifically, a sense of awe and reverence.  Often this is opposed to the coldly practical sense that is alleged to animate science.  Yet, many scientists feel awe, reverence, and veneration when contemplating the universe.  Conversely, many, perhaps most, religious people seem to view religion simply as routine social practice.  It is something they do without much feeling.  Recent events have reminded us that still other people, worldwide, have violent hatred as their sole religious emotion.  The emotional phenomenology of religion is too complex and diverse to be defining.

On the other hand, Malinowski was not wholly wrong in stressing personal emotion.  Some sort of intense emotional involvement appears to be always found somewhere in a community in their practice of religion, even if truly transcendent experience and “altered states” are not, and even if most communicants are dully bored with the rituals.  But Malinowski was wrong in arguing that religion stopped there.  Durkheim was surely right in arguing that a main job of religion is to get individual emotions (the ordinary, everyday ones as well as the wilder ones) involved in ritual, ceremony, and prayer.  Thus does a collective representation arise.  Thus, also, does it become more than an intellectual enterprise.  It is a real emotional force.  Whether or not religion is about awe or fear or love, it is certainly about some kind of production of psychological solidarity from synchronizing individuals’ emotionalities.

Recent anthropological accounts of religion tend to exaggerate the distinction from science by highlighting the aspects of religion that seem most exotic and irrational to the writers.  Thus, to pick only the most reasonable and sensible recent exemplars, Atran (2002) and Sosis and Alcorta (2003) emphasize supernaturals, counterintuitive beliefs, and the like; Sosis and Alcorta also emphasize visions, hallucinations, shamanism, sacrifice, and all the other favorite exotica of anthropology.  The problem here is that religion, everywhere in the world, is far more often a matter of going politely and sociably to church, temple, ch’a’chaak, or witchetty grub ceremony, there to sit patiently and be bored to death.  The ordinary humdrum side of religion is far more common, typical, and important to believers than the exotica.  At least in the United States, mystical experience is rare, and true religious ecstasy is very rare.  Religion for the vast majority of Americans is a much more ordinary affair, involving no altered mindstates.  Romantic anthropologists might argue that this is because capitalism and other political ideologies have taken over so much of the American mind—driving, often, the religious beliefs.  However, my observations in China, Mexico, and elsewhere convince me that the vast majority of humans have ordinary and practical religious lives, even in shamanistic and spirit-mediumistic religions.  Transcendent experience of every kind is rare and disruptive, not common, causal or constitutive.  Religion, usually, is mindless ritual conformity.

By contrast, spirituality is, by definition, emotional; it is the individual’s experience of awe, reverence, entrancement, enchantment, or similar emotions or transcendent feelings, inspired by natural or supernatural entities or forces.  Religion usually stimulates spirituality, and may be influenced by it.

The current claim that “secularism” or “secular humanism” is a “religion” does not make the grade by any standards.  First, secularism has no supernaturals—by definition.  Second, it has no communitas; nobody purports to be part of the secularist church or congregation or communion, nor does secularism have festivals, rituals, temples, or anything else to show.  (The French Revolutionaries briefly tried to impose such, but were laughed to shame.)  Third, it has no body of beliefs.  The few secular humanists out there do agree on some facts, but they have no litmus test, no professions that they must accept.  Indeed, skeptics differ enormously in worldviews—they are united only by skepticism (see journals like The Skeptic and The Zetetic, passim).

3.  Science

The division between magic, science, and religion was also important to Lévi-Strauss (1962) and others of the time.  All the thinkers of the structural and cognitive traditions of the 1960s emphasized the rational, systematic, empirical side of traditional knowledge, Lévi-Strauss’ “science of the concrete.”

In the 1950s and 1960s, interest in such systems climaxed in the development of the field of “ethnoscience.”  This field arose from the researches of several of George Murdock’s students, sent to work in Micronesia and the Philippines (Conklin 1957; Frake 1980).  The word was coined from the earlier term “ethnobiology,” introduced by John Harshberger in 1895.  Soon, terms like “ethnobiology,” “ethnozoology” and “ethnoornithology” followed.  The word “ethnoscience” seems to have disappeared somewhere in the intervening years, but the other terms persist, in spite of an attempt by Scott Atran to substitute “folkbiology” and other “folk-” words (Medin and Atran 1999) /1/.

Many of these systems are as purely empirical, self-correcting, developing, and truth-driven as any western science (Anderson 2000, 2003, 2005).  They also share with scientists a concern with insight, sensed experience, testing and probing, and the like (David Kronenfeld, personal communication, comment on draft of this paper, 2004). As science, they are limited more by lack of scope and equipment than by lack of some (mythical?) scientific mentality or method.  However, many, as we shall see, have supernatural entities built into them.  These problematize still more the basic distinction.  Insight and sensed experience are basic to both science and religion, narrowing somewhat the gap between them.

As ethnoscience was developing, the term “science” was being subjected to a great deal of critique.  For thousands of years—ever since the Greeks began to talk of scientia—it had had something like the straightforward, common-sense meaning that Malinowski knew.  It referred to systematized knowledge, as opposed to faith (belief without evidence) on the one hand and techne, mere craft, on the other.

In the early 20th century, Viennese logicians attempted to confine it to an exceedingly formal, even artificial, procedure, with very strict rules of verification or—more famously—”falsification” (Popper 1959).  Science was even supposed, in some positivist quarters, to be reducible to mathematical rules; anything not mathematically expressed was not science.  This was wildly out of line if one wanted to continue talking about Greek or Renaissance science, or even about the actual practice (as opposed to rigorous ideals) of 20th century science.  (Try expressing paleontology mathematically.)  Inevitably, a counterreaction set in, spearheaded by Thomas Kuhn’s classic The Structure of Scientific Revolutions (1962), but anticipated earlier by the brilliant work of Ludwig Fleck on medical history.  Kuhn concentrated on biases and limits within scientific practice, but the critiques of science rapidly expanded to include the lamentable tendency of scientists—all too human as they are—to believe pure nonsense and claim it to be scientific if it fit their biases.  This was especially noted in the cases of racism and sexism, but extended to economic theories, animal psychology, medicine, and indeed most fields.  However, one must note a very important difference.  Kuhn was attending to real problems with science itself, whereas the critiques of racism and sexism—however necessary—were simply rewarmings of Francis Bacon’s critiques of bias-driven pseudoscience.

In any case, we all came to realize that science as practice is a very human affair.  Pure science may be an ideal—or may not—but real-world science has its biases, mistakes, blind spots, and so forth.  Culture and economics intrude.  People try to justify their prejudices and errors.  Not only such socially convenient lies as racism, but even such neutral ideas as stable continents (Oreskes 1993, 2001), persist long after their time /2/.

From all this arose a great change in how “truth” is established.  Instead of going for pure unbiased observation, or for falsification of errors, we now go for “independent confirmation” /3/.  A result is not counted, a finding is not taking seriously, unless it is cross-confirmed, preferably by people working in a different lab or field and from a different theoretical framework.  I certainly don’t believe my own findings unless they are cross-confirmed!  (On all these matters, see Kitcher 1993 and Martin and MacIntyre 1994.)  We now see science as a social process.  Truth is established, but slowly, through debate and ongoing research.

The positivist agendas (including Malinowski’s), and all similar agendas in science, assume we can tell the supernatural from the natural.  This is where we begin to see real problems with these agendas, and the whole “modernist program” that they may be said to represent.  Telling the supernatural from the natural may have looked easy in 1948.  It seems less clear today.

We have many well-established facts that were once outrageous hypotheses:  the earth is an oblate spheroid (not flat), blood circulates, the sun is only a small star among billions of others.  We also have immediate hypotheses that directly account for or predict the facts.  However, we then move on to higher and higher levels of abstraction, inferring more and more remote and obscure intervening variables—up to the almost mystical cosmology that now postulates multidimensional strings, dark matter, dark energy, quark chromodynamics, and the rest.  Even the physicist Brian Greene has to admit that “[s]ome scientists argue vociferously that a theory so removed from direct empirical testing lies in the realm of philosophy or theology, but not physics” (Greene 2004:352).  To people like me, unable to understand the proofs, modern physics is an incomprehensible universe I take on faith—exactly like religion.  The difference between it and religion is not that physics is evidence-based; astrophysics theories, especially such things as string and brane theory, are not based on direct evidence, but on highly abstract modeling.  The only difference I can actually perceive is that science represents forward speculation by a small, highly trained group, while religion represents a wide sociocultural communitas. (Religion also has beautiful music and art, as a result of the communitas-emotion connection, but I suppose someone somewhere has made great art out of superstring theory.)

Modern science continues to test and probe and refine its evidential and theoretical bases, much more than a “traditional” society does.   However, traditional societies too made considerable progress over time, as we know from the few that left history and the many more that left archaeological records.  We live in a world of “rapid discovery science,” to use Randall Collins’ felicitous term for the self-conscious experimental science that emerged in Europe in the early modern period (Collins 1998).  Rapid discovery science is very different from traditional science, but the difference is one of degree at least as much as of  kind.

Rapid discovery science grew slowly in the late Middle Ages and took off spectacularly from the late 16th century.  It led to formalization and rational, formal discovery procedures, but only slowly; modern positivist discovery algorithms really date from the late 19th century, after the major groundwork of modern science had been long established.

The ethnoscientists proved that traditional ecological knowledge is not just random facts, but is systematized, elaborated, and often axiom-driven or theory-driven as much as any modern science.  (The theories are, necessarily, far less comprehensive and sophisticated, because of lack of equipment and so forth.)  Conversely, the historians of science have proved that contemporary laboratory science can be as culturally and socially negotiated, even “constructed,” as any traditional worldview.  Old-fashioned racist ideas that contrast the superior, rational, empirical Science of “the west” with the mystical nonsense of the rest are still sometimes aired (e.g. Wolpert 1993), but few scholars take them seriously.  The world has accumulated vast amounts of factual knowledge and reasonable theory in the past couple of hundred years—more than we had before.  We also have developed many formal algorithms for guaranteeing that we have the most accurate data possible.  However, even so, the process is not as different from what went on before as we used to pretend.

The really important question here is that of inferred explanatory mechanisms or variables.  Science advances by creating new theories and hypotheses.  Inevitably, facts underdetermine these (Kuhn 1962).  By definition, new ideas are leaps in the dark.  They are someone’s best guess about what explains, or at least ties together, observations.  They are guesses about what is inside a black box.

Traditional cultures, lacking modern laboratory equipment, could not test their inferences that gods, spirits, and the like were responsible for observed reality.  We can come up with explanations that fit the facts better (though that may merely push the gods to a more remote place in the causal chain).  But in the end we of the contemporary world must infer explanatory variables, many of which will surely turn out to be wrong.

Consider, again, modern physics.  The universe is approximately 96% composed of dark matter and energy—matter and energy we cannot measure, cannot observe, cannot comprehend, and, indeed, cannot conceptualize at all (Greene 2004).  We infer its presence from its rather massive effects on things we can see.  On a smaller and more human scale, we have the “invisible hand” (Smith 1776) of the market—a market which assumes perfect information, perfect rationality, and so on, among its dealers.  We have abstract and unverifiable black-box mechanisms in psychology (e.g. Freudian dynamic personality factors), anthropology (“culture,” at least in its Whitean form), and sociology (“class,” “discourse,” “network.”)  All these and other abstractions have become mystical entities in at least some theoretic discourses.  Modern schools in the social sciences often rest on beliefs, assumptions, and global abstractions more than on data.  Thus Atran’s specifications merely allow us to split an arbitrary realm, which we may please to call “religion,” from all the other realms of human knowledge.  All substantial scientific knowledge must rest on at least some inferences about unprovable, abstract, obscure entities.

When traditional peoples infer similar black-box mechanisms, we label those “supernatural.”  The traditional people themselves never seem to do this labeling; they treat spirit forces and spirit beings as part of their natural world.  If they do make a distinction, they may put things into boxes that seem wrong from the western scientists’ viewpoint.  Chinese see feng-shui, the art of site planning, as a practical art, not a part of religious teaching.  Indeed, feng-shui is based on empirical observation and does empirically work when confined to its traditional orbit (as opposed to being misapplied to a modern home, city, or office; see Anderson 1996a).  Yet it postulates dragons and tigers inside hills, and other matters foreign to laboratory science.  Westerners, at first, generally called it magic.  Later western scholars sometimes called it science, sometimes religion, sometimes a mix.  I was first impressed by this problem when Chinese friends ascribed a landslide to the shaking of a dragon in the hill.  At the time, western geologists could do no better; they spoke of “angle of repose,” a concept as mystical as the dragon.    Western science now understands such things more fully, but the comparison was sobering.

Some Chinese explained earthquakes the same way:  as dragons shaking in the earth.  Other Chinese explained earthquakes as waves caused by turbulent flow of qi (breath, or vital energy) in the earth.  The Chipewyans of north Canada explain earthquakes as the thrashing of a giant fish (Sharp 2001).  When I was an undergraduate, most American geologists did not yet accept the fact that earthquakes are usually caused by plate tectonics, and instead invoked scientific explanations just as mystical and factually dubious as the dragons and fish.  One should never be too proud about inferred variables inside a black box.

Similarly, I have argued at length elsewhere (Anderson 1996a and references therein) that Hippocratic humoral nutritional theory “sold” so well because it was about the best way of explaining empirical facts that could be found before modern laboratory chemistry arose /4/.   Traditional peoples around the world observed many effects of environment and food, and found many cures to obvious nutritional problems.  These they explained by humoral theory, which worked well enough to prove itself to everyone’s satisfaction.  The inferred variables—the Hippocratic-Galenic qualities and humors—did not, in fact, exist, as science proved only in the 20th century.  But they were inferred as black-box variables on the basis of extensive and excellent observation and deduction.  The messy reality of nutrition, disease, and environment interactions could not have been deduced before modern times.

The belief in hot and cold led to logically deducing several wrong conclusions in various parts of the world.  Diarrhea is usually considered due to cold, so liquids may be withheld, a very bad procedure.  Disease is usually explained by the Maya of Quintana Roo as being due to a cold shock hitting a warm body, so a great deal is done to prevent this and to warm the sick—harmless and sometimes beneficial, but not much use.  Fortunately, empirical observation of effective herbal remedies usually dominates logically-deduced but empirically wrong treatments.

Considering all folk explanations, and classifying the traditional ones as “religion,” Edward Tylor classically explained magic and religion as, basically, failed science (Tylor 1871).  He came up with a number of just-so stories explaining how religious beliefs could have been reasonably inferred by fully rational people who had no modern laboratory devices to make sense of their perceptions.  Malinowski’s portrayal of religion as emotion-driven was part of a general reaction against Tylor in the early 20th century.  Indeed, Tylor discounted emotion too much.  On the whole, however, there is still merit in Tylor’s work.  There is also merit in Malinowski’s and in Durkheim’s.  I think we all realize now that religion and magic, and science too, all do partake of the rational, the emotional, and the social.

Even so, the tendency today is to stand Tylor on his head.  Postmodernists dismiss science—and sometimes all truth-claims—as just another social or cultural construction, as solipsistic as religion and magic.  Some anthropologists still believe, or at least  maintain, that cultural constructions are all we have or can know.  This is a self-deconstructing position; if it’s true, it isn’t true, because it is only a cultural construction, and the statement that it’s only a cultural construction is only a cultural construction, and we are back with infinite regress and the Liars Paradox.

The truth is, rather, that culture is knowledge we need and use to adapt to our environments (see, e.g., the good discussion of this in McElroy and Townsend’s classic textbook of medical anthropology, 2004).  There are no “cultural constructions” hanging out there in spirit land; there is only cultural knowledge instantiated in individuals’ learned knowledge.  Most of that knowledge is factually correct, and is tested all the time against reality; a culture that taught that strychnine was good food would change or die out soon.

Thus, cultural constructions are usually right.  And people who use them are usually even more right.  Culture always includes several alternative plans or courses of action for any given situation; the individual actor must decide which one to use.  He or she then learns to fine-tune these plans in the light of reality.  To say that people all mindlessly follow one cultural construction is to say that learning and change cannot exist.

The extreme cultural-constructionist position is all too close to, and all too usable by, the religious fundamentalists who dismiss science as a “secular humanist religion.”  Both the postmodernists and the fundamentalists are, alas, all too often answered only by the self-styled apologists for science who maintain that science is True and is All Facts.  I imagine these people as having red-purple faces and using terms like “insolent puppy.”  Richard Dawkins has reportedly said (I hope this isn’t true) that there are no skeptics at 30,000 feet—as if the fancy technology of a plane is all there is to science!  One could as well establish the truth of religion by recourse to the famous line about atheists and foxholes.

Surely if they stopped and thought, the apologists for science would recognize that some unpredictable set of today’s inferred black-box variables is fairly sure to be a laughingstock 20 years from now—along with phlogiston, luminiferous ether (Greene 2004), and the angle of repose.  More:  they would have to admit that a science that is all correct and all factually proved out is a dead science!  Science is theories and hypotheses, wild ideas and crazy speculation, battles of verification and falsification.  Facts (whatever they are) make up part of science, but in a sense they are but the dead residue of science that has happened and gone on.  (This and the rest of my comments on science are mostly justified or extended from the writings of Ian Hacking, esp. Hacking 1999; Philip Kitcher 1993; Laudan 1996.  These writers have done a good job of dealing with the fact that science is about truth, but is ongoing practice rather than final truth.  On the other hand, see Anderson 2000 for a critique of Hacking.)

Thus, it is sometimes difficult to distinguish science from religion.  However, we can say that science is made up of facts that are empirically proved and ideas that can be theoretically proved, or at least tested, while religion is made up of things that must be taken on faith because they simply cannot be proved or disproved by any evidence.  Stephen Jay Gould (1999) has recently given us a strong argument for this position, and for the fundamental complementarity—and therefore the fundamental difference—of religion and science.  Yet Gould has to admit that his own science, paleontology, cannot be directly tested—and that not only traditional religions, but also modern fundamentalist religions, make a number of statements that can be tested.  At least, like paleontological ideas, they could be proved or disproved if we had a time machine.   The reality of six-day creation in 4004 BC, Noah’s flood, and Joshua’s musical destruction of Jericho have all been established as fact to fundamentalists’ satisfaction; geology, archaeology, and other sciences are bent to their ideas.  The same objective data serve to disprove the same events, of course, in the eyes of other observers.

Gould simply argues that this is a mistake—the religious are trying to do science, and shouldn’t mix the two.  The fundamentalists, of course, disagree.  As Gould points out, the “conflict between science and religion” is an 18th and 19th-century invention.  (Admittedly, it had an ancestry going back to the 17th century.)  Before then, people did not think the two were mutually exclusive or occupied different spheres.  Gould tries to eliminate the conflict, but only by making the difference even more profound.

Meanwhile, the postmodernists have shown, quite correctly, that all too much of modern “science” is really social bias dressed up in fancy language.  “Scientific” racism is the most obvious example of this.  Discredited “scientific” ideas about women, children, animals, and other vulnerable entities are only slightly less obvious.

Descartes, Newton, and others of their time were deeply and devoutly religious men.  Descartes imported Catholic dogma into his science; his separation of “mind and body” is actually l’âme et le corps.  Moreover, scientists of that day did much that we would call “magic.”  Christopher Marlowe’s Doctor Faustus, the prototypic early-modern seeker, saw magic as an alternative to science and philosophy.  Newton sought the philosopher’s stone.  Alchemy is now dismissed as magic; chemistry is science—but the two were one until the 18th century.  Astronomy is science, astrology is magic—but every competent stargazer did both until the late Renaissance, and saw them as part of one agenda.  It is this sort of thing that makes it difficult to contrast an essentialized, superior “Western science” with the mystical lore of the rest of the world.  We can count the Greeks and the Renaissance scientists as part of the modern positivist agenda only by the most outrageous back-projection and selective quoting.  In fact, the conceptualization of science as strictly factual-rational, religion as strictly otherwise, is quite recent, and seems to be evolving as we speak.  Concepts of “religion” and “science” not only differ from culture to culture; they differ within the same culture over time, often over quite short temporal intervals.

Religion is heavily involved with morals, while science is traditionally considered to be values-neutral.  The separation owes a great deal to David Hume’s argument that one cannot deduce an “ought” from an “is” (of course, Hume was more nuanced and subtle than this canned, though useful, summary of his philosophy; see Hume 1975).  Religion can prescribe morality, justifying it from divine law.  Science cannot.  Supposedly, it is about “is,” not about “ought.”  Ethical bias in science is often seen as a contamination, inevitable or not.  (The fundamentalist claim that science is really a religion of “secularism,” faked up to sell evil and cutthroat morals under the name of “evolution,” is based on complete lack of knowledge of science.)

However, once again, the real world is not so simple.  First, science has its own morality:  one is not supposed to lie, fake results, trash one’s fellows in anonymous reviews, or plagiarize.  Second, some sciences are specifically and openly moral.  Medical research makes no sense except as a healing art, and agricultural research is targeted at feeding more people, or at least producing more of some desired commodities.  Third, we can see from this that if one allows oneself even a single simple moral postulate, one may create an open-ended universe of scientific research.  If we ought to cure sickness, a great deal of “ises” follow from that one “ought.”  Medical science is the result.   If society is desirable, then there must be some way to hold society together and keep it from self-destructing by failure to adapt; this opens the door to otherwise relatively values-neutral applied research in anthropology, sociology, and political science.   This point, complete with the analogy to medicine, was already made almost 2500 years ago by Plato and Aristotle!  (See Lloyd 1996.)

In short, separating religion from science and both from magic is analytically important, valuable, and interesting, but it must always be a somewhat arbitrary separation.  It is constantly being problematized by the messiness of the real world and the messiness of real human thought.  No matter how defined, “religion” and “science” (to say nothing of “magic”) are ideal types—idealtypen—that do not describe the real world very neatly.  They are valuable ways of modeling thought and discourse, but not really descriptive, though they have considerable heuristic use in describing the western (and now the international) world since 1800 or 1850.

I suspect they are often invoked as a way of capturing turf.  Accounts of the “religion” of Native Americans often slide over into descriptions of hard-headed ecological wisdom that is better called “science.”  The huge and magistral series of volumes issuing from the Forum on Religion and Ecology, associated with Harvard Divinity School, has given us a powerful new look at religion and the environment—and, very often, the “religion” is hard to tell from “science.”  My own contribution focused on the science-friendly Chinese religion of Taoism (Anderson 2001; a general volume by the Forum organizers thoughtfully uses the title Worldviews and Ecology; Tucker and Grim 1994.)

Ethnoscience and folk science studies have been criticized for using the term “science” broadly enough to include a good deal of mystical and supernatural belief.  I am, obviously, very tolerant of such usages.  Perhaps more true is that folk sciences have the disadvantage of being local sciences, and thus to some extent collective cultural representations.  It is broadly true, I believe, that religion is the collective representation of a society (Durkheim 1995).  Modern science is not—it is internationally constructed and stated, and is proved by tests rather than by social usefulness.

Malinowski confined the “science” of the “savages” to mere technical matters. He admitted that they had a vast knowledge of seafaring, agriculture, crafts, and such.  He regarded their wider explanatory systems as “religion.”  (“Magic” was a low-level, esoteric, unsuccessful technology.)  The Trobrianders, like many other peoples (including American fundamentalists), tended to explain everything nonobvious by recourse to supernatural agency.  Many ethnoscientists seem to have implicitly made a similar distinction (see Frake 1980).

Perhaps this division works for the Trobrianders, but it does not work everywhere.  Roberto Gonzalez (2001) has demonstrated that Zapotec agricultural knowledge is a thoroughly pragmatic, accurate, well-systematized science.  Yet it explains some things by recourse to earth gods and other agencies that outsiders consider supernatural.  Among these are a very interesting and significant class of agents: the humors of Hippocratic medical fame—hot, cold, wet, and dry.  They were imported to Oaxaca from Spain in the 16th century, there to fuse with similar native concepts.  To the modern international bioscientist, the Hippocratic humors are supernatural or magical entities, but they were the best cutting-edge science from Hippocrates’ time up until the late 19th century!  They not only seemed reasonable; they were, arguably, the best explanations that people could come up with, given the observations and observational techniques they had.

This brings us back to Tylor, and to Atran (2002).  Religion can be seen as partly based on plausible but wrong inferences about ultimate cause.  Thus, in explaining the world, people naturally infer spirits and gods.  There appears to be a genuinely natural tendency for people to assume that trees, rocks, and animals are “people,” in some sense—having volition, consciousness, and humanlike will.  The Durkheimian observation that religion is a projection of the social order naturally follows from this (a point Atran rather misses).  There is obviously a great deal more than this to what we normally call “religion,” but inferences about the “people” out there clearly comprise one of the building blocks from which religion is made.

In short, inferred black-box causal arguments, once they are superseded, can get called either “science” or “religion,” depending largely on the whims and prejudices of the person doing the calling.  Magic, science, and religion as very useful terms, but terms that are limited in their application.

4.  Comparisons

On the other hand, we do need these terms.  It really is necessary to distinguish science from religion in the modern world, and to see the search for truth as different from the search for transrational belief—if only to free ourselves from the postmodernists with their claims that Newton’s Principia is a rape manual, or, as I once heard a postmodern anthropologist say, “sex has nothing to do with biology.”

Science is not a mess of facts; it’s a system that is meant to represent the world accurately and empirically.  It necessarily includes a lot of black-box variables that are hypothesized but are generally under examination, and that often turn out to be wrong.  Religion is not just a bunch of supernatural beliefs; it too is a system, in which emotionally compelling beliefs collectively represent social contracts and sell social ethics.  Magic partakes uncomfortably of both, and is thus a dicey term, very hard to use beyond a low level of abstraction /5/.

Thus, when we focus on religion as supernaturals, we are focusing on the wrong thing.  We should be focusing on religion as part of society, the part that stirs deep emotions to persuade its membership to follow the social codes.

Traditional belief systems of most interest to cultural and political ecologists are precisely those that most thoroughly mess up the distinction between science and religion.  Zapotec science prescribes certain ways of dealing with the world, and some of these are moral and ethical (Gonzalez 2001).  Similarly, Yucatec Maya worldviews are derived from a set of assumptions about the world that include emotional involvement and strong ethical teachings.  The world is based on non-neutral assumptions; its reality demands that people maintain the forest, keep trails clean, kill nothing unnecessarily, and so on.

The Northwest Coast and north woods Native peoples also have a cosmology that is not only “spiritual” but intensely moral.  They ask for permission before taking anything from nature; I have seen women ask permission of a tree to take a bit of its bark.  In regard to animals, respectful treatment is required:  killing only for necessity, not killing too many at a time, treating the remains in accord with ritual regulations that communicate respect, giving away one’s first kill and much of one’s subsequent meat take, and so on through countless rules.  These are taught through myths—typically stories that communicate the dreadful consequences of not following the rules (Anderson 1996b).

Other ways of breaking up the “knowledge” field exist, including the psychologists’ division of cognition, emotion, and conation (will), and Aristotle’s classic division into ethics, politics, physics, and metaphysics.

This brings us to the question of how non-European peoples talk about religion and knowledge.  Many simply use a single term for a very broad range of phenomena.  Widely in Australian aboriginal cultures, for instance, one word (tjukurrpa in the Central Desert) covers the mythical Dreamtime; traditional law, custom, ethics, and morals; ritual art (which means pretty much all art); and knowledge of how to manage the environment.  These are all one—not separated analytically or otherwise.  Somewhat similar is the Navaho concept of hozhoo, which means harmony, peace, beauty, health, and good resource management (see e.g. Witherspoon 1977).  To recover health, an ill patient sits (at key points in a healing ritual) on a ground painting of supernatural and natural beings to absorb the beauty and harmony of the painting and thus the blessings of the beings themselves.  For the Hopi, good ethical and ecological practice is the Hopi way, whereas bad actions—social, personal, and ecological—are ka-Hopi, “not Hopi” (Brandt 1954—a book which is still the best work on the ethics of a nonwestern, kin-based society).

Traditional Native American societies cut up knowledge in still different ways.  On the Northwest Coast, for instance, knowledge is divided sharply into esoteric vs public.  Esoteric is divided in turn into knowledge of shamans and knowledge of secret-society initiates.  Public knowledge is divided into special ritual and mythic knowledge of chiefs and ordinary knowledge available to everybody (see Anderson 1996 and references there).  Most “science”—that is, empirical knowledge not depending on supernatural belief—is in the public sphere, but not herbal medicine, which is mostly a shaman specialty.  These other ways of dividing up and classifying knowledge are valuable in their particular situations.  We need to be far more understanding and sensitive of such situational matters.

For one thing, the traditional Native American worldviews typically go with a particular educational strategy.  Sharp (2001), Goulet (1998), and Peter Gardner (2003) have been among several who have described the rather extreme form of this education that is found among the Dene (or Dine) peoples.  Children learn by doing, with almost no verbal instruction.  They copy what their elders do, without much correction, and learn by their mistakes as they try to imitate.  Verbal teaching, if any, is done by personal stories.  What little correcting of the young is done verbally is done by telling stories on oneself:  “When I was young, I…” is a typical formula starting off a story of obvious—but completely tacit—application to the younger hearers present.  This—with rather more verbal instruction—is the universal style of education among the older Maya and Northwest Coast people I know.  (See also Greenfield 2004 on such learning among the Tzotzil Maya.)  It was also the way my elders taught me field skills, from shooting to bird identification, in rural Anglo-America.  I suspect it is universal.  It is a very effective way to teach physical skills, habitual routines, and country knowledge in general.  It is also a very empowering way of teaching, within its own confines.  It makes the user really learn, and learn for him- or herself.  No superficial memorization here.  It is not necessarily a good way to teach the sort of analytic-generalizing skills we want in universities, but university teaching doesn’t help you ride a bike, let alone manage a slash-and-burn field.

The Yucatec Maya have used Spanish words cognate with the English ones for 500 years now, and have assimilated the European worldview.  However, I get a strong sense that the ancient Maya had a system much like the Northwest Coast one.  For one thing, the usual word for Malinowskian “magic” is not mágica (though that Spanish word is known) but secretos, “secrets.”  This indicates that what matters is a not a distinction between science and magic but a distinction between exoteric and esoteric.  This we know to have been a critical distinction for the ancient Maya.  Also, the Yucatec Maya divide the “knowledge” domain in interesting ways.  To know is ojel.  To recognize or remember is k’ajal or k’ajal ool (generally shortened to k’ajool)Ool may be a shortening of ojel here, but it also means “heart” or “soul.”  The separation between “know” and “recognize” is so similar to the Spanish distinction of saber and conocer that one must suspect some influence, but the basic distinctions are ancient.

To understand is na’a(t).  The cognate word in Tzotzil Maya, na’, has been the subject of an important study by Zambrano and Greenfield (2004).  They find that it is used as the equivalent of “know” as well as “understand,” but focally it means that one knows how to do something.  It is a word so closely related to practice and action that, when asking if someone wants a soda, one says “Mi xana’ yuch’el rasqu?” -literally, “do you know the drinking of soda?” (Zambrano and Greenfield 2004:256).  A better translation would probably be “Do you wish to experience a soda?” but the point is made:  na’ means, focally, to do something on the basis of knowledge of it.   This keys us into the difference between Tzotzil knowing and Spanish or English knowing:  Tzotzil know by watching and then doing (as do many other Native Americans; see Goulet 1998, Sharp 2001), while Spanish and English children and adults know by hearing lectures or by book-learning.  It seems fairly likely that a culture that sees knowledge as practice would not make a fundamental or basic distinction between magic, science, and religion.  The distinction would far more likely be between embodied knowledge or habitus and knowledge known only from others’ stories.  Such distinctions are made in some Native American languages.

The Chinese break up knowledge in a totally different way.

There is no traditional equivalent in Chinese for any of our three terms.  The nearest equivalent to “magic” would be the various terms for charms and spells.  The nearest equivalent to “religion” in old-time Chinese (there is a modern Chinese term glossing the western word) is a verb-object construction, bai xian, meaning “worshiping supernaturals.”  It covers the same phenomena as the English word.  Another relevant term is jiao, “body of organized teaching,” which includes organized religion as opposed to folk cults, and also includes organized bodies of philosophic tradition, such as Confucianism.  A jiao was defined by its having a body of canonical text material and an organized priesthood or philosophical teaching tradition.  Folk cults and the like were dismissed by elite persons as mixin, “superstition” (a fairly modern coinage).  The folk referred to them simply as bai xian, or bai fo (“worshiping Buddhas”).  Magic was not recognized as a separate realm; what Malinowski called “magic” was lumped with either bai xian or with ordinary knowledge, depending on the amount of worship activity involved.  Much worldly work was done by spells and charms, but they were part of religious practice.

Sciences, and knowledge in general, was described in terms of xue, “studies.”  This term included literary studies, historical studies, and even religious studies, as well as sciences.  Some disciplines that are close to western equivalents differ in blurring the magic-science-religion lines; thus “medicine” has a close Chinese equivalent, yao, but yao in China includes religious and magical practices to drive away sickness demons and placate plague gods.  Modern writers may note a “curious combination of medical and magical thinking” (Wilms 2002:33) in the writings of Sun Simiao in the 7th century, but Sun would not have found anything curious; he was merely writing up the medicine he knew.  I dare say a state-of-the-art American medical textbook of 2004 will look like “a curious combination of medical and magical thinking” in a thousand years, if there is anyone around then to read it.

Li xue, “study of basic principles,” covered both mystical and philosophical foundations of the universe and humanity.  The modern term for “science” is ke xue, “study of categories,” but it is a Chinese reading of a Japanese translation of the western word.  Neither it nor its referent (western-style science, as a concept) had any traditional currency.

In short, for traditional literate Chinese, knowledge was classified as various kinds of xue.  Organized religious or philosophical tradition,  jiao, was opposed to unorganized cults.  These and other categorizations of knowledge cut right across the western category lines we are investigating.  All this makes sense in terms of traditional Chinese society, as I hope to show in a future paper /6/.

5.  The Longer View

There is some truth to the old cliché that “the Indians” live in a religious or spiritual world, but that cliché misses the point, which is that the Indians live in a world about which they know an incredible amount of factual information and toward which they have an intensely ethical stance.  The detailed factual knowledge and the free-standing moral philosophy are not separable from each other, or from the supernatural beliefs.  The problem of deducing an “is” from an “ought” is resolved by seeing the “oughts” as natural laws, part of the “ises.”  The need to represent factual knowledge and ethical treatment of resources in one system means that Native American systems of thought cannot be separated into “religion” and “science.” There is thus a very basic, very fundamental way in which we cannot see these systems in terms of “religion” and “science.” This has often been pointed out in anthropology, but it needs continual reassertion.

What has not been pointed out is that this different way of organizing knowledge follows directly from the need to know almost everything about the local environment.  If one must know, and take seriously, every significant plant, tree, creek, patch of ground, weather indicator, and star, and must be constantly open to more and more and more learning, one must inevitably develop an intense emotional involvement with all the above—both as individual entities and as a totality.  This confrontation—total self totally open to total fact—is the reality of survival for hunting-gathering and traditional agricultural peoples.  They may not feel “awe,” but they do not have the modern urbanite’s luxury of feeling nothing.  An intense spiritual involvement with the whole landscape necessarily follows.

Contra the myth of the noble savage, this does not mean that Native Americans always live by their highest ethical principles, any more than anyone else does.  The Maya I know are deeply committed to the morality of not overhunting—not killing more than they need—but they have shot out the game in the Yucatan Peninsula anyway.  This is no more surprising than the disconnect between American political ethics and the recent election.  Ethics are about what one should do, and are always deliberately set at impossibly high levels, because everyone knows that real people fall short of their aspirations.  To get people to do X, law, ethics, and religion have to demand more-than-X.  If the safe speed is 40 mph, we must post it at 25.  Christianity demands universal love in hopes that people will at least act caring toward a few more people than they would otherwise.  Native American cosmologies demand a much wider respect—nothing less than kinship, sociability, responsibility, and care toward all beings.  It should not surprise us if reality is often short of this.  (On the failure of traditional people to live by their ecological morals, see esp. Kay and Simmons 2002, but also Evans 1998, McNeill 1992, Ponting 1991 on environmental damage in Mediterranean cultures from ancient Greece on down, and Kirch 1994, 1997 on Polynesia, where early waste was often followed by learning better and recovering ecologically).

The point, then, is not to promote the “noble savage” nonsense, but to say that Native American systems of environmental thought, at least the ones known to me, are based on an experiential or phenomenological substrate that is fundamentally different from “religion” or “science” as the modern world knows them.  It is different not because it has a “prerational worldview,” or any such nonsense, but because it has fundamentally different problems to handle.
Kay Milton (2002) stresses the importance of “loving nature.”  She studied contemporary British and Irish ecological activists, and found them motivated by love, not by economic or scientific motives.  She thoughtfully expanded on this perception, in light of the fact that humans integrate emotion and cognition and cannot separate these in practice (she cites Damasio 1994).  Human phenomenological reality is a harmonious, or at least thorough, integration of the two.  Loving nature, in so far as it applies to religious feelings or actions, would clearly be in the “spiritual” rather than the “religious” realm.  For saving nature, love is essential, and so is structured ethical practice—in short, we must use spirituality and religion, and of course science too, to save the world.  (I am indebted to Michael Winkelman’s commentary on this paper, Dec. 3, 2005, for clarifying this point.)
For my Maya and Northwest Coast friends, “loving nature” is not quite the perfect description.  Their emotional response is both less and more than that.

A Maya not exposed to contemporary environmentalism would not normally say in yakuntik le baalche’oobo (“I love the things of the trees”—”things of the trees” is about as close as one can get in Maya to “nature”).  Maya do love many wild plants and animals, and do say so; my coworker Felix Medina Tzuc recently said to me “me encanta la selva, me encantan las arboles” (“the forest enchants me, the trees enchant me”; Medina Tzuc, personal communication, 2005).  The Maya understand a holistic world, including humans and nonhumans.  They represent it religiously, via the various gods of forests, fields, animals, and so on.  Their emotional response to it is intense, quiet, often loving, and impossible to verbalize.  (I know it from sharing thousands of hours, over many years, in the forests and fields with many close friends.)  Basically, there is a deep, intense, pervasive involvement.  However, for many Maya, “love” is quite an inappropriate word; they do not say they “love nature” as they do say they love their homes, spouses, children.

On the whole, the same is true of many Northwest Coast people, though here my experience is that many of them do have so intense an emotional feeling that “love” is the only word.  Many Northwest Coast Native people today feel that the trees, animals, and rocks of their areas are home and family—living spirit persons who are intensely loved as one loves one’s human neighbors, friends, and house.

This is the result of thousands of years of having to take forests and animals seriously. If one has to interact with plants and animals over time, one cannot help developing emotional and moral feelings toward them.  Humans simply do not remain neutral about things they have to take constantly into account (Anderson 1996a).  Interactions construct our world; our very selves are born of interacting (Dilthey 1988; Levinas 1985).  Interactions with beings we take seriously are powerfully emotional events, and, indeed, more than that; phenomenologically, they construct our world (Dilthey 1988; Levinas 1985).  Our selves are the products of our interactions; some Diltheyans would say our selves are our interactions.  In the cases noted here, these Native American peoples must depend on the forests and animals, and must also be responsible for caring for them and—to at least some appreciable extent—conserving them.  This is constructed religiously, but is based on sound empirical principles.  In such cases, we are dealing with a phenomenological universe orthogonal to science and religion.

The emotions raised by the forest, among the Maya and the Northwest Coast peoples, are much deeper and more complex than those raised by a forest among most Americans.  Yet, even the modern American will usually look at a forest with a mix of awe, concern, curiosity, and other emotions.  She may desire to save the forest, yet fear the mountain lions in it.  She may wonder what there is to eat, or what birds nest there.  She may know enough of the English and Dutch landscape art tradition to have her reaction to the forest strongly colored by Constable’s or the Ruisdaels’ views of woodland.  Countless other associations may occur—cognitive yet powerfully informed by emotion.

The whole brain—senses, associations, feelings, motions, gestures, work habits, rational analyses, and much more—deals with whole nature.  Total self meets total fact.  Malinowski’s and Durkheim’s views of emotion and religion stand rehabilitated, but in very different terms from those they used.

A worldview grounded in this sort of involvement does not lead to cutting the world into magic, science, and religion.  It leads, rather, to cutting the world into ethical vs nonethical behavior, into local vs nonlocal place (cf. Feld and Basso 1996), into belief vs clearly counterfactual claims, into effective vs ineffective ways of living and working, into prosocial vs antisocial behavior (remembering that animals and plants are part of society), and into one’s immediate social world—including animals, plants, and often even rocks—vs other distant social worlds.

I do not reject “science” and “religion” as ways of cutting up experience.  (However, I would restrict the term “magic” to parlor shows for night clubs and children’s parties.)  They are appropriate and useful ways of cutting up knowledge.  I am emphasizing that they are adapted to a particular situation.

6.  Conclusions

“Science,” then, can be seen as the ideal word to use if one is looking at the empirical, evidence-driven side of a knowledge system.  Science is empirical, and rests on open-minded inquiry.  It is done by individuals who are concerned with verifiable, empirical results, and often with probing further into new realms.

“Religion” is used if one is looking at the belief-driven or tradition-driven side of human knowledge.  It is based on widely-shared social beliefs, usually of very long standing and of very high levels of perceived antiquity and legitimacy.

Science and religion, however, overlap broadly when one looks at inferred explanatory variables, and also when one looks at complex and emotional attitudes toward the world.

Magic—basically a sciency agenda done in a religious way—disappears analytically, though the concept still may have uses for someone somewhere.

These terms, when applied to societies other than modern western or western-influenced ones, do not refer to identifiable and institutionalized sectors of activity (as they do in the modern US).  Instead (as Malinowski knew) they provide a rather arbitrary classification system for knowledge, imposed in a thoroughly etic way.  Other societies have their own, quite different, systems for classifying knowledge.  Usually, their cultural models do not include a natural/supernatural split.

Thus, “magic, science, and religion” are not very adequate terms to represent traditional knowledge systems.

Classifications are about being useful, not about being cast in stone.  We of the modern international scholarly community have found a particular way of classifying knowledge to be rather useful.  Perhaps it is not so useful now.  Meanwhile, other cultures have classified knowledge in radically different ways.  These often cross-cut our classification system.  This does not mean they are wrong; it means they serve different purposes.  Traditional people usually do not want to separate environmental knowledge from ethics, or either one from ritual.

“Worldview” (see Kearney 1996; Tucker and Grim 1994) is always available, as are “cosmology” and “belief system.”  “Science” has the merit of conveying respect.  However, it has too limited and misleading a set of connotations these days.  And we have the turf-capturing problem noted above.  Even “traditional” has its problems.  It does not deal adequately with new but culturally embedded systems, like the emergent ecological knowledge systems of creole and metis populations, or the folk systems of western scientists, or the contemporary worldviews of fundamentalist religious persons.

One problem is that the domain of “religion” has been shrinking steadily for years—Max Weber’s “disenchantment” (Gerth and Mills 1958).  Religion or spirituality or at least a spiritualized cosmology was knowledge, or included all knowledge, for the most ancient or traditional societies.  Weber traces out the shrinkage, which he expected (not without some wistfulness) to become total in the 20th century, with religion disappearing.  Instead, it has come roaring back, but still as a residual category—often more an excuse for hatred and murder than a serious spiritual force, let alone a source of knowledge.

We are forced back on “cultural models,” “knowledge systems,” and “ecological knowledge” as labels for the things we study.  Different cultures categorize knowledge in different ways, according to their needs.  They also develop different ideas of what “knowledge” is, how one acquires it, and how one teaches it.  The knowledge that results is inevitably some mix of truth and error.  The fact that knowledge is socially constructed does not mean that it is wrong; in fact, culture is a way of adapting and learning, and thus cultural knowledge of practical matters is usually very accurate (Anderson 2000) /7/.

Phenomenologically, knowledge is a single vast realm.  It can be broken up in various ways, but these divisions are all somewhat arbitrary.  They may capture some very useful distinctions, but they are not God-given.

We simply have to remember that there is nothing special or privileged about our anthropological ways of classifying knowledge.  Other groups have their own classifications—interesting, valuable, and demanding more study.  And cultural knowledge systems typically merge all the components that Malinowski and many subsequent thinkers need to separate out, for analytic purposes, as magic, science, and religion.


/1/  Of course, there is an issue of pejorative connotations here.  “Folk” is clearly problematic.  Even “ethno-” and “traditional” are criticized by some.  This seems to me to be paranoia, but perhaps I am not an adequate judge.  More to the point, the term “traditional ecological knowledge” has recently been criticized for implying that said knowledge is “backward and static….  Much development based on TEK thus continues to implement homogenous Western objectives by coopting and decontextualizing selected aspects of knowledges specific to unique places, eliminate their dynamis, and focus more than anything else on negotiating the terms for their commodification” (Sluyter 2003; he cites to Arturo Escobar’s excellent essay of 1998, and I wish to second the importance of Escobar’s work).  Most of us that study TEK do not commit these sins.  But many people do, especially non-anthropologists working for bureaucracies.  International bureaucratic “spin” has indeed corrupted the use of the term, making it problematic.  Many scholars have pointed this out (see e.g. Bicker et al. 2004, passim).  The implication of stasis is particularly unfortunate.  The appropriation of useful knowledge from traditional societies is absolutely necessary if humanity is to survive, but we may have to find a better way to talk about it.  We certainly have to work out better ways to compensate the bearers of this knowledge, as Sluyter implies in the second part of the quote.  The problem of decontextualizing is particularly complex.  To understand these systems, we need to study them in their entirety, but to use their key discoveries for humanity, we need to do some decontextualizing.  If we had waited to understand everything about local uses of quinine bark or artemisia leaves, malaria would kill many times as many people as it does today.  There are knotty ethical problems here.

/2/  This leads to the extreme idea that science is all just politics—that there is no truth, and that “truths” are merely claims selectively deployed for political reasons.  Though this idea is widely circulated in self-styled “progressive” circles, it is an intensely right-wing idea, stemming from Nazi and proto-Nazi thought and ideology (including the thought of the hysterically anti-Semitic Nietzsche, and later of Martin Heidegger and Paul de Man, both committed and militant Nazis, and influenced also by the right-wing philosopher and former Nazi-sympathizer Paul Ricoeur), and deployed today not only by academic elites but also by the fundamentalist extremists who denounce Darwinian evolution as just another origin myth.  The basically fascist nature of the claim is made clear by such modern right-wing extremists as Gregg Easterbrook (2004), who has recently attacked the Union of Concerned Scientists for protesting against the politicization of science under the Bush administration.  Easterbrook makes the quite correct point that the Union of Concerned Scientists is itself a politically activist group, but then goes on to maintain that, since scientific claims are used for political reasons, the claims are themselves purely political.  He thus confuses, for example, the fact that global warming due to greenhouse gases is now a major world problem with the political claims based on this fact; he defends the Bush administration’s attempt to deny or hide the fact.  There are at least three separate types of statement to consider here:  Proven scientific facts; objective assessments of how those facts impact the general welfare; and outright political choices about what to do about these impacts.  The Union of Concerned Scientists is aware that the first two are different from each other and very different from the third.  Easterbrook, and his pseudo-progressive anthropological kin, are not—or perhaps they simply have a vested interest in pretending they are not.
Somewhat related are the various critiques of the concept of “nature” (Cronon 1983; Escobar 1999; Hvalkof and Escobar 1998; for a vast compendium of “nature” views, see Torrance 1998).  It is clear that “nature” is a highly contested culture-bound category that cannot be essentialized, and that, as Escobar points out, an essentializing political ecology would be hopeless nonsense.  We are looking at processes, and usually at ill-defined ones.  On the other hand, there is a difference between Jasper National Park and Times Square, and a difference between the most remote Amazon forests and downtown Sao Paulo.  The difference is highly relevant to human survival.  One may also note that “tradition” has been quite properly and actively critiqued (Hobsbawm and Ranger 1985), but that does not change the fact that speaking English is more “traditional” in London than in Singapore.  Of course traditions change and adapt.  This does not prove that they do not exist, any more than the fact that I have changed greatly in the last 64 years proves that I do not exist.  De-essentializing concepts does not consign them to the garbage bin; it makes them more useful.

/3/  David Kronenfeld has commented (writing on a draft of this paper, 2005) here:  “Science itself is also an attitude—probing, trying to ‘give nature a chance to say no,’ and so forth….science is not a thing of individuals but is a system of concepts and of people.”

/4/  David Kronenfeld has commented at this point:  “It is empirical in providing an approximate summarization that works pretty well; it is not really what we know as science because it lacks the interlinked roles of formal representation, formal logical organization, attention to detail, and role of testing in theory refinement.”

However, Zapotec and other traditional sciences have the last three of these.  They lack only a rigorous formal structure.  But, then, so does modern nutritional science, except for the extreme laboratory-based studies relying on rats and guinea pigs.  So does much of natural history and field geology.  Only the most “physical” of modern sciences have a fully developed, rigorous, quantifiable, deductive theory framework.

/5/  An issue that should concern anthropologists is the ways that, in modern American society, some mistaken beliefs are classified as “pseudoscience,” some as “religion,” and some merely as “controversial/inaccurate/disproved science.”   In psychology, parapsychology is firmly in the “pseudoscience” category, but racism remains “scientific,” though equally disproved and ridiculous.  Freudian theory, now devastated by critiques, is “pseudoscience” to many but is superseded-“science” to many others.  It is obvious that such labels are negotiable, and are negotiated.

The respectability and institutional home of the propounder of a theory is clearly a major determinant; a mistake made at Harvard is science, a mistake outside of academia is pseudoscience.

All this should not be taken to mean that I do not believe there is a valid distinction here!  Good science is real, and is about real things.  Pseudoscience is usually a quite obvious shuck.  There is a difference between fact and nonsense, whatever the postmodernists may say.  The problem here is that nonsense propounded by a Harvard or Stanford professor is all too apt to be taken seriously—especially if it fits with popular prejudices, like the thoroughly bogus and transparently ridiculous race psychology of Harvard professor Richard Herrnstein and Stanford professor Thomas Jukes.

The tendency to use “science” to describe truth-claims and “religion” to describe untestable beliefs has been attacked above.  This usage is normative, not descriptive.  It is a rather underhanded attempt to confine religion to the realm of the untestable and therefore irrelevant.  (This is one objection made by almost every reviewer of Gould 1999.)

Words for “science” and scientists go back to medieval English, and ultimately to Greek via Latin.  The claim that “scientist” was not invented till W. Whewell coined it in 1840 is true, but it merely replaced an earlier word, “scient,” used as a noun or adjective; this word had been around since the 1400s, but had become obsolete.  In short, none of these are new concepts in the western world, and, consequently, restricting the meanings to modern phenomena is illegitimate.

/6/  Comparison between Chinese and Greek science are apparently irresistable to historians.  Recently, this has been brought to a fine art by the collaboration of Geoffrey Lloyd and Nathan Sivin (Lloyd 1996, 2002; Lloyd and Sivin 2002).  They find less difference  between the traditions than do many others.  However, Greek science was more prone to define itself through controversy, and thus to be more self-conscious about basic principles—not only the famous “axioms,” but also ideas of cause, logic, and proof.  Chinese invested more effort in technical applications and in large-scale systems thinking.  Further work is neceessary, but we now know that Greek and Chinese scientific knowledge systems were not as radically distinct as some have alleged, and both differed from modern science in similar ways.

/7/  I want to distance myself explicitly from the view that “cultural models,” “cultural constructions,” or “the superorganic” somehow have a life of their own.  Contrary to the old Leslie White view, there is no “culture” that somehow mystically instills itself into our minds and determines our thought.  Culture consists of useful knowledge—data and rules—that we learn and then use in adapting to daily challenges and opportunities.  It includes countless alternatives that we can invoke at will.  If I want to pluralize “sheep” as “sheeps,” or even “sheepen,” I can do it, in spite of cultural rules to the contrary, and I will be understood by standard-English-speaking hearers (who will correctly assume I am playing language games—and will, if they are young enough, be amused).  Creative use of knowledge and rules is what life is all about, and any culture that imposed a rigid crust of “constructions” on its bearers would immediately die out.


David Kronenfeld’s suggestions and comments and Leslie Sponsel’s intellectual support have been enormously helpful to me.


Anderson, E. N.  1996a.  Ecologies of the Heart.  New York:  Oxford University Press.

—  1996b.  “An Introduction to Wilson Duff.”  In Bird of Paradox by Wilson Duff.  Surrey, BC:  Hancock House.  Pp. 16-119.

—  2000.  “Maya Ornithology and ‘Science Wars.'”  Journal of Ethnobiology 20:129-158.

—  2001.  “Flowering Apricot:  Environmental Practice, Folk Religion, and Daoism.”  In Daoism and Ecology, ed. N. J. Girardot, James Miller, and Liu Xaiogan.  Cambridge:  Harvard University Press.  Pp. 157-184.

—  (with José Cauich Canul, Aurora Dzib, Salvador Flores Guido, Gerald Islebe, Felix Medina Tzuc, Odilón Sánchez Sánchez, and Pastor Valdez Chale).  2003.  Those Who Bring the Flowers.  Chetumal, QR, Mexico:  ECOSUR.

—  2004.  “‘Loving Nature’ among the Maya.”  Paper, Society for Ethnobiology, annual conference, Davis, CA.

—   (with Aurora Dzib Xihum de Cen, Felix Medina Tzuc, and Pastor Valdez).  2005.  Political Ecology in a Yucatecan Community.  Tucson:  University of Arizona Press.

Atran, Scott.  2002.  In Gods We Trust.  New York:  Oxford University Press.

Bering, Jesse.  2005.  “The Evolutionary History of an Illusion:  Religious Causal Beliefs in Children and Adults.”  In Ellis, Bruce J., and David F. Bjorklund (eds.): Origins of the Social Mind:  Evolutionary Psychology and Child Development.  NY:  Guilford.  Pp. 411-137.

Brandt, Richard.  1954.  Hopi Ethics.  Chicago:  University of Chicago Press.

Collins, Randall.  1998.  The Sociology of Philosophies.  Cambridge, MA:  Harvard University Press.

Conklin, Harold.  1957.  Hanunoo Agriculture.  Rome: FAO.

Cronon, William.  1983.  Changes in the Land.  New York:  Hill and Wang.

Damasio, Antonio.  1994.  Descartes’ Error.  New York:  G. P. Putnam’s Sons.

Dilthey, Wilhelm.  1988.  Introduction to the Human Sciences:  An Attempt to Lay a Foundation for the Study of Society and History.  Tr. Ramon A. Betanzos.  Detroit:  Wayne State University Press.

Durkheim, Emile.  1995.  Elementary Forms of the Religious Life.  Tr. Karen Field (Fr. orig. 1912).   New York: Free Press.

Easterbrook, Gregg.  2004.  “Politics and Science Do Mix.”  Los Angeles Times, April 6, 2004, p. B13.

Escobar, Arturo.  1998.  “Whose Knowledge, Whose Nature?  Biodiversity, Conservation, and the Political Ecology of Social Movements.”  Journal of Political Ecology 5:53-82.

— 1999.  “After Nature:  Steps to an Antiessentialist Political Ecology.”  Current Anthropology 40:1-30.

Evans, L.  1998.  Feeding the Ten Billion.  Cambridge:  Cambridge University Press.

Feld, Steven, and Keith Basso (eds.).  1996.  Senses of Place.  Santa Fe, NM:  School of American Research.

Frake, Charles.  1980.  Language and Cultural Description.  Ed. Anwar S. Dil.  Stanford:  Stanford University Press.

Gardner, Peter.  2003.  “Rethinking Foragers’ Handling of Environmental and Subsistence Knowledge.”  Ms. circulated at American Anthropological Association meeting, Chicago.

Gonzalez, Roberto.  2001.  Zapotec Science.  Austin: University of Texas Press.

Gould, Stephen Jay.  1999.   Rocks of Ages: Science and Religion in the Fullness of Life.  New York:  Ballantine.

Goulet, Jean-Guy.  1998.  Ways of Knowing.  Lincoln: University of Nebraska Press.

Greene, Brian.  2004.  The Fabric of the Cosmos.  New York:  Knopf.

Hacking, Ian.  1999.  The Social Construction of What?  Cambridge, MA:  Harvard University Press.

Greenfield, Patricia.  2004.  Weaving Generations Together:  Evolving Creativity in the Maya of Chiapas.  Santa Fe:  School of American Research Press.

Hobsbawm, Eric, and Terence Ranger (eds.).  1985.  The Invention of Tradition.  Cambridge:  Cambridge University Press.

Hume, David.  1975.  Enquiries Concerning Human Undestanding and Concerning the Principles of Morals.  Ed. L. A. Selby-Bigge (orig. 1777).  Oxford:  Oxford University Press.

Hvalkof, Sven, and Arturo Escobar.  1998.  “Nature, Political Ecology, and Social Practice:  Toward an Academic and Political Agenda.”  In:  Building a New Biocultural Synthesis, ed. by Alan H. Goodman and Thomas L. Leatherman.  Ann Arbor;  University of Michigan Press.  Pp. 425-450.

Kant, Immanuel.  1976.  Anthropology from a Pragmatic Point of View.  Tr. Victor Lyle Dodwell (German orig. 1796).  Carbondale and Edwardsville, IL:  Southern Illinois University Press.

Kay, Charles, and Randy Simmons (eds.).  2002.  Wilderness and Political Ecology.  Salt Lake City:  University of Utah Press.

Kearney, Michael.  1996.  Worldview.  Novato, CA:  Chandler and Sharp.

Kirch, Patrick.  1994.  The Wet and the Dry.  Chicago:  University of Chicago Press.

—  1997.  “Microcosmic Histories:  Island Perspectives on Global Change.”  American Anthropologist 99:31-42.

Kitcher, Philip.  1993.  The Advancement of Science.  New York:  Oxford University Press.

Kuhn, Thomas.  1962.  The Structure of Scientific Revolutions.  Chicago:  University of Chicago Press.

Laudan, Larry.  1996.  Beyond Positivism and Relativism.  Boulder, CO:  Westview.

Lee, Don.  2005.  “Lessons in Harmony Lead to Discord.”  Los Angeles Times, Sept. 27, pp. A1, A20.

Levinas, Emmanuel.  1985.  Ethics and Infinity.  Interviews w Philippe Nemo.  Tr. Richard A. Cohen.  Pittsburgh, PA: Duquesne Univ. Press.

Lévi-Strauss, Claude.  1962.  La pensée sauvage.  Paris:  Plon.

Lloyd, Geoffrey.  1996.  Adversaries and Authorities.  Cambridge:  Cambridge University Press.

— 2002.  The Ambitions of Curiosity:  Understanding the World in Ancient Greece and China.  Cambridge:  Cambridge University Press.

Lloyd, Geoffrey, and Nathan Sivin.  2002.  The Way and the Word:  Science and Medicine in Early China and Greece.  New Haven:  Yale University Press.

Malinowski, Bronislaw.  1948.  Magic, Science and Religion.  Glencoe, IL:  Free Press.

Martin, Michael, and Lee MacIntyre (eds.).  1994.  Readings in the Philosophy of the Social Science.  Cambridge, MA:  MIT Press.

Mauss, Marcel.  2001.  Magic.  Tr. Robert Brain, 1972.  (Fr. orig., Esquisse d’une théorie de la magie, M. Mauss and H. Hubert, appeared in Année sociologique for 1902, actually published 1904.)  London:  Routledge.

McNeill, J. R.  1992.  The Mountains of the Mediterranean World.  Cambridge:  Cambridge University Press.

Medin, Douglas, and Scott Atran (eds.).  1999.  Folkbiology.  Cambridge, MA:  MIT Press.

Milton, Kay.  2002.  Loving Nature.  London:  Routledge.

Oreskes, Naomi.  1993.  The Rejection of Continental Drift.  New York:  Oxford University Press.

— 2001 (ed.).  Plate Tectonics:  An Insider’s History of the Modern Theory of the Earth.  Boulder:  Westview.

Poniting, Clive.  1991.  A Green History of the World.  New York:  Penguin.

Popper, Karl.   1959.  The Logic of Scientific Discovery.  London:  Hutchinson.

Sharp, Henry.  2001.  Loon.  Lincoln:  University of Nebraska Press.

Sluyter, Andrew.  2003.  “Material-Conceptual Landscape Transformation and the Emergence of the Pristine Myth in Early Colonial Mexico.”  In Political Ecology: An Integrative Approach to Geography and Environment-Development Studies, ed. By Karl Zimmerer and Thomas Bassett.  New York:  Guilford.  Pp. 221-239.

Smith, Adam.  1910 (orig. 1776).  The Wealth of Nations.  New York:  Dutton.

Sosi, Richard, and Candace Alcorta.  2003.  “Signaling, Solidarity, and the Sacred:  The Evolution of Religious Behavior.”  Evolutionary Anthropology 12:264-274.

Torrance, Robert (ed.).  1998.  Encompassing Nature.  Washington, DC:  Counterpoint.

Tucker, Mary Evelyn, and John Grim (eds.).  1994.  Worldviews and Ecology.  Maryknoll, NY:  Orbis.

Tylor, Edward.  1871.  Primitive Culture.  London: John Murray.

Wilms, Sabine.  2002.  The Female Body in Medieval China.  Ph.D. dissertation, Dept. of East Asian Studies, University of Arizona, Tucson, AZ.

Wolpert, Lewis.  1993.  The Unnatural Nature of Science.  Cambridge, MA: Harvard University Press.

Zambrano, Isabel, and Patricia Greenfield.  2004.  “Ethnoepistemologies at Home and at School.”  In Culture and Competence: Contexts of Life Success, ed. By Robert J. Sternberg and Elena L. Grigorenko.  Washington:  American Psychological Association.  Pp. 251-272.

Leave a Reply