following copyrighted material is intended for individual use of the
researcher, and may not be copied or distributed without written permission
from the copyright holder. You may use it for non-profit scholarly purposes.
The reference for this article is: Rivas, J. A. and G. M. Burghardt. 2002
Crotalomorphism: A metaphor to understand anthropomorphism by omission. pp: 9-
17. In The Cognitive Animal: Empirical and Theoretical Perspectives on Animal
Cognition. (Bekoff, M., Colin, A. and G. M. Burghardt, eds.) MIT Press,
Crotalomorphism: A Metaphor to Understand Anthropomorphism by Omission
Jesús A. Rivas and Gordon M. Burghardt
The Story of Country Blue
When foreign students come to study at the
People from a country called Blue normally wear blue clothes, blue hats, and blue sunglasses. Houses are blue and so are the cars and streets. Country Blue borders country Yellow where people wear yellow clothes, yellow hats, and yellow sunglasses. Houses as well as cars and street are yellow in country Yellow. These two countries are internally peaceful, but have conflicts with each other. They view the customs and policies of the other country as bizarre and evil. One day, a diplomat from Blue decided to visit Yellow, learn about their customs and traditions, and write an extensive article to his fellow Blue citizens explaining how people in Yellow view the world. He was convinced that they were not evil, they just saw the world in a different way. Therefore, the Blue diplomat put on yellow clothes, a yellow hat, and yellow sunglasses. After three month living in Yellow, the Blue diplomat returned to his country and reported that the citizens in Yellow were not bad, bizarre, or stupid. His article claimed that in country Yellow life was actually very nice and Green!
This tale characterizes somebody trying to understand another culture who neglected to consider a basic limitation: his own colored glasses. These he did not, or perhaps could not, remove. In an even more profound way, our human glasses are ingrained in us, and are very hard to remove (if possible at all). Nevertheless, if we are aware of having biased spectacles, we can attempt to address their effects upon us. In order to understand the cognitive accomplishments of a bee or beetle, squid or chimpanzee, we need to evaluate how they perceive their world. In doing so, technology can assist us, but we need to constantly remind ourselves that we are using our human senses and human-based technology, and are processing the information with a human brain.
Like The Poor, Anthropomorphism Will Always Be With Us
Anthropomorphism is defined as attributing human properties to nonhuman
entities. Such entities can be supernatural (gods) or animate or inanimate
nature. It is but an extension of the problems facing anyone trying to
understand another human culture, as in the Blue/Yellow example, or actually,
the experiences of any person other than yourself. The problems inherent in
inferring what other people or animals experience from their overt behavior was
recognized by Romanes and the early comparative psychologists (Burghardt 1985),
but they sought ways of surmounting the problem. After several decades,
however, psychologists and ethologists came to regard anthropomorphism as a
serious error that must be avoided no matter the cost. When
More recently, there have been attempts to look more closely at what anthropomorphism really is and how it operates. One of the first modern revivals was contributed by Lockwood (1985; 1989) who made it clear that not all anthropomorphic attributions were equal. For example, two kinds of anthropomorphism are restricted to non-scientific writing and therefore not a problem in science ("allegorical" and "personification"). Two others, that Lockwood called "superficial' and "explanatory" have potentially harmful consequences in science and these were the main culprits for Kennedy. Lockwood also identified a fifth kind of anthropomorphism, "projective" ("applied" anthropomorphism in the 1989 version), that he considered a legitimate strategy that had been used by authorities such as Darwin: the use of the personal perspective to convey what it is like to be another living being (see Crist this volume).
Burghardt (1985) tried to further demystify the concept of anthropomorphism by arguing that it is a legitimate, and perhaps particularly creative, way to do science if it is used to develop hypotheses that can be tested in a rigorous manner. Critical anthropomorphism (CA), was introduced as means of using various sources of information including "natural history, our perceptions, intuitions, feelings, careful behavior descriptions, identifying with the animal, optimization models, previous studies and so forth in order to generate ideas that may prove useful in gaining understanding and the ability to predict outcomes of planned (experimental) and unplanned interventions" (Burghardt 1991: 73).
Critical anthropomorphism was deliberately modeled after a proposed solution to a centuries old controversy in science and philosophy: the nature of the external world (Mandelbaum 1964). Here one important contrast is that between direct or "naïve" realists who accept that the world is just as it appears to us, and variants of subjectivists, idealists, and solipsists who basically argue that nothing exists but our own minds (or as Edgar Allen Poe wrote: "All that we see or seem is but a dream within a dream"). More germane is the position of skeptics who acknowledge the existence of an external world, but assert that our flawed senses show that we can never learn anything about it. Illusions serve the purpose of the latter nicely (c.f., Gregory 2001). Mandelbaum's (1964) solution was to advocate a radical critical realism based on both relevant sensory and neural data and predictive inferences. Furthermore, a process called 'transdiction' could be applied to ground "inferences to objects or events which not only have not yet been observed, but which in principle cannot be observed" (p: 63). Just as in the study of perception "critical realism" is the most scientifically congenial approach, so in the realm of animal minds, critical anthropomorphism is required.
Recently, an edited book (Mitchell, Thompson, and Miles 1997), a monograph (Crist 2000), and articles (e.g., Fisher 1990) have been devoted to reviewing and exploring the problems and perils of anthropomorphism and anecdotes in modern studies of animal behavior. Caporael and Heyes (1997) argue that the default condition of the human mind is anthropomorphic and this functions in understanding and dealing with other humans. Much earlier, C. Lloyd Morgan (1894) noted that anthropomorphism was very useful to farmers when dealing with domesticated animals; in a similar manner, anthropomorphism may inform many of the most creative scientific accomplishments (Burghardt 1985). Perhaps anthropomorphism is only harmful in science when it is unacknowledged, unrecognized, or used as the basis for accepting conclusions through circumventing the need to actually test them.
Anthropomorphism By Omission
Anthropomorphism by omission is the failure to consider that other animals have a different world than ours. We can, without realizing it, attribute human traits by failing to consider that many species perceive the world in a different manner than we do. Scientists may know this in theory, but if they do not deliberately acknowledge that different species have different perspectives and priorities than we do, they may draw anthropomorphic conclusions that are erroneous, or at least compromise their ability to adequately understand and predict, let alone influence, behavior. The only way to address this problem is to consider the internal and external world of the animal and try to evaluate what it is like to be the animal.
The idea of studying the private worlds of other animals was pioneered by Jacob von Uexküll (1909/1985) who attempted to bring the latest neural, physiological, and perceptual findings to bear in understanding the behavior of animals by considering both their inner world (Innenwelt) and how they perceived and responded to their environment (Umwelt). A major aspect of this approach was to evaluate differences among species in the salience of biologically relevant perceptual cues (Tinbergen 1951; Burghardt 1985). The cognitive ethology movement as pioneered by Don Griffin in the mid 1970s explicitly focused on the ways that animals perceive, interpret, and experience the world (Griffin 1978; Cheney & Seyfarth 1992; Glotzbach 1992; Burghardt 1997; Bekoff & Allen 1997 and chapters in this volume). An important component of this approach, though often understated, is to consider the animal being studied as an active participant, with the researcher trying to put him or herself in the animal's situation. This is especially true for those studying primate behavior (Herzog & Galvin 1997). Timberlake & Delamater (1991) proposed that to understand the behavior of an animal "Experimenters not only need to put themselves in the subject's shoes, they need to wear them walk, watch, hear, touch, and act like the subject" (p: 39). More recently, Bekoff (2000) has extended this view to advocate a biocentric anthropomorphism. Although it is true that we will never fully appreciate how another animal experiences the world, by doing our best to accomplish this through applying critical anthropomorphism, including the full range of available scientific data, we will get closer to understanding the life of the animal. Conceptually the task is no different from dealing with understanding another person that may differ from us in age, gender, sensory and motor abilities, personality, temperament, language, health, profession, wealth, status, or a host of other variables. Nevertheless, although we can never obtain access to the full inner life and private experiences of another human being, some people seem to be more successful at generalizing to situations other people get themselves in and are thus considered particularly insightful, empathic, or privy to human nature writ large (e.g., major novelists, playwrights, composers). Others can exploit their knowledge in political, social, and deceptive activities (Burghardt 1997). Partial knowledge is possible and useful even if full knowledge is both in practice and in principle unobtainable.
Our aim in this paper is to document the presence of anthropomorphism by omission and raise awareness about its presence and detrimental influences on science. If this argument is convincing, then when we analyze the behavior of animals, we not only can, but must, deliberately put ourselves in the animal's shoes - and not doing so is potentially and truly anthropomorphic. Thus we are extending the approach advocated by Timberlake and Delamater (1991) one step further: omitting to put oneself in the animal's shoes often leads to default anthropomorphism or anthropomorphism by omission.
Crotalomorphism - More Than A Metaphor?
Consider another story, but one involving the study of other species, species in which, unlike our tendencies with primates and domestic animals, anthropomorphism is not considered a serious threat to trained scientists.
A researcher is studying the behavior of a very colorful lizard. When this lizard sees a person it rapidly changes its color and matches the background, just as octopuses are well known to do. The researcher concludes that the change in color is a cryptic response to avoid predation. Just at this time, however, a large female timber rattlesnake (Crotalus horridus) quietly observing the researcher from some nearby brush is suddenly spotted by the researcher, who is both startled and scared. Rattlesnakes, being pit vipers, can detect patterns of infrared radiation from mammals through the loreal pits situated between the eyes and nostrils. Therefore, when she perceives the researcher she detects it as a very warm animal moving in a much cooler background (not unlike the way the human sees the colorful lizard). Now when the startled researcher saw the rattlesnake, adrenaline kicked in and the flow of blood to the arms and legs was reduced along with all other peripheral circulation; a normal response to stress. The researcher turned cooler and was therefore duller to the infrared detecting 'eyes' of the snake. Our clever rattlesnake concludes that the person is trying to escape by matching the cooler background: the drop in peripheral temperature was a cryptic response against predators with heat sensing organs.
This is an example of crotalomorphism by omission; for although there is evidence that predator stress can lower body temperature (Gabrielsen and Smith 1985), the snake's conclusion would be probably dismissed as erroneous by most human scientists. However, is her conclusion in any essential way different from the conclusion of the human researcher studying lizards? Crotalomorphism highlights the problem of interpreting the world solely by one species' standards. Together the snake and the scientist are playing out the same game as the people from country Blue and country Yellow!
We are convinced that unwitting anthropomorphism by omission is frequently present in several scientific fields related to animal cognition and we using examples from the literature, including our own work (Burghardt 1998), for we have not been immune ourselves.
A Few Case Histories
Foraging tactics in snakes:
When a northern water snake (Nerodia sipedon) catches a fish, it typically lifts it out of the water and takes it to the shore, where the snake swallows it. It is typically claimed that water snakes do not use any specialized technique other than grab and hold to subdue or kill their prey, since they lack venom and do not constrict. On the other hand, if an anaconda, a large, and largely aquatic tropical species, catches a deer and drags it to the water where the animal cannot breath, and its capacities to move are reduced (since deer are not aquatic), it is often concluded, reasonably we might add, that the anaconda is subduing the deer not only by constriction, but also by bringing it to an environment where the deer cannot breath or run. However, a deer forced into the water is no different than a fish forced to the land. While acknowledging the deer as being deliberately subdued, we fail to detect that the fish is being subdued as well. The former is much more evident to us because we would be subdued if we were in the position of the deer and not in the position of the fish. So, even without directly attributing human traits to the animals, it is clear that we may fail to consider the traits in which they are different. The fact that most scientists might readily agree when this possibility is pointed out, does not cancel our observation that removing fish from water is not readily viewed as a functional predatory tactic
Aposematic coloration has been considered a means to warn predators of potential danger. This has been the basis of much theoretical work, beginning with Fisher (1930). Yet there is often little consideration of how the presumed predators of aposematically patterned and colored animals actually identify prey, and if they can even see the presumed warning cues. Despite all the attention devoted to supposed warning coloration, we have not found much consideration of what warning coloration looks like to different predators. This is particularly evident when the predators are invertebrates rather than vertebrates. Even with vertebrates, many may be effectively color-blind. We came close to making this error with garter snakes responses to warning colors (Terrick, Mumme and Burghardt 1995) when in fact, in spite of having cones, the latest data find no wavelength discrimination ability in snakes (Burghardt, 1998). The discrimination we found may have been due to contrast, not wavelength. To other predators, color may be salient. To draw conclusions about the aposematic coloration of a species without asking first whether it is aposematic for its predators is another example of anthropomorphism by omission. In fact, if we really want to talk about aposematism and not be anthropomorphic we probably should invest less time studying aposematic coloration and more time researching aposematic scents, aposematic vocalizations, and aposematic textures.
Dominance in cockroaches:
Courtship in Drosophila:
Another example of the problems of anthropomorphism by omission is seen in the study of the role of sound during courtship of fruit flies (Drosophila spp) carried out by Boake and Poulsen (1997). Males flap the wings vigorously during courtship, producing sounds, and it was hypothesized that such sounds were an essential part of the courtship. An experiment was carried out with wing-clipped males that could not produce the sound. Such males were expected to have reduced mating success. To the authors' surprise, clipped males did better than the controls! The authors then mention that a reviewer (R. Hoy) pointed out that that the clipped males had lighter wings (stumps) and thus might vibrate them faster - hence vibration, more than sound, is the important part of the stimulus. In the world of humans, the shaking of the wings of an insect can only be detected by the sound they make. Due to the much larger mass of the researchers, it escaped their attention that the movement of the wings could produce substantial vibratory stimuli for an insect, though not for them.
The cat and the mouse:
In the field of cognitive ethology, anthropomorphism by omission has also taken its toll. Collin Beer (1997) states that ". . . the reach and complexity of connections attaching to ideas in the human case will usually far exceed [our italics] what is conceivable for any animal." This respectable and plausible claim is supported by Beer (1997: 203) by describing a cat crouching beside a hole down which it has just chased a mouse: "We should be inclined to say that the cat thought there was a mouse down the hole. But consider what thinking that would mean to us: it would mean that there was a furry mammal down the hole, a tetrapod vertebrate, a whiskered rodent, a warm-blooded cheese-eater, and a whole lot more that could not possibly occur to the cat. Only a small part (italics by us) of the network within which mouseness is nested for us extends into the cat's world." Beer had an uncharacteristic lapse in failing to consider that the world of the cat is different than ours and that only a small part of that world is obvious to us. The odors left by a chased and stressed mouse might allow the cat to obtain information as to whether the mouse is fat or thin, young or old, male or female, sick or injured. Perhaps it is aware of, and even enjoys, listening to the pitter-patter of the mouse running down the hole or smelling the odors the mouse and hole emit. We can't even begin to imagine the number of things that the cat may be aware of in that moment. Does it recall the taste of the last mouse it caught or the feeling of grabbing and biting it? Does it recall some former encounter when a mouse bit him at the time of the capture or remember prior feeding successes and failures at that location? Given the salience and importance of rodents to cats, their 'thoughts' and private experiences about mice in this situation might be far richer, and certainly quite different than ours.
Human and nonhuman language:
In the literature dealing with comparative communication and language the superiority of human language expression is contrasted with "less evolved", "simpler", or "less advanced" systems in other species (c.f., Brickerton, 1998; Allen and Saidel, 1998; Ujhelyi 1996). In addition to the various criticisms that can be made of such formulations (e.g., Allen and Saidel, 1998), such statements also are prone to committing the error of anthropomorphism by omission, for they often fail to recognize that other species have different worlds than ours. We often ignore the complex information contained in chemical cues and pheromones since we are so limited chemically. As another example, consider honey bees studying communication between humans: to them we would perform rather poorly since we do not, and perhaps cannot, dance out to our partners the location of the closest restaurant! To claim that our language is superior risks not being aware of its limitations compared to communication in other species because we are biased in our understanding of it. We are not denying that human language might be, and probably is, generally more complex than other forms of communication in all other animals. Nevertheless, it is uncritically anthropomorphic to begin comparative language research with this bias, which often goes unrecognized.
Zoos as Products of Anthropomorphism by Design:
Although not advertised by them, zoo exhibits, even the most modern, are often shaped much more by the needs of the human visitors or human caretakers than the animals supposedly shown living in 'natural' settings. Naturalistic exhibits reflect human concepts of nature, not necessarily those of the animals exhibited. Indeed, if the exhibit is too effective, the animals may not be readily spotted by visitors, and the exhibits are changed for people, the paying customers, not for the exhibited animals. This has happened in gorilla exhibits (Burghardt, 1996). Some zoo commentators have actually argued that the best modern zoos are those that treat the zoo as a theatrical experience for the public, not one oriented to the lifestyle of the captive species (Polakowski, 1989). This view seems to be pervasive in action, if not in rhetoric, in most zoos. A more honest and explicit recognition of the anthropomorphic nature of modern zoos would be helpful. It is not enough to have eliminated tea parties for chimpanzees.
A final example where anthropomorphism by omission has taken a toll is in the field of landscape ecology and conservation planning. Decision makers often develop wildlife management plans or enact legislation that considers various ecological benchmarks while often ignoring the perceptual capabilities and life histories of the animals that are to be protected. Consider the example of habitat managers designing corridors to connect patches of forest in a natural area. The use of corridors has been strongly debated as a management measure to allow species to move between separated natural areas, yet less emphasis has been made in the perceptual abilities and private experience of the animals that would use those corridors. Metapopulation analysis also evaluates gene flow across relatively isolated areas as a central issue in conservation biology (Wiens 1996). The spatial scale involved is a critical issue. If patches are too far apart, then the animals will not be able to find the patch by using their natural navigation method and gene flow will be interrupted (Wiens 1996).
Unfortunately, landscape modeling has often been addressed with little
knowledge of the perceptual world of the animals involved and little attention
to their private experiences. A well documented example is with the
white-footed mice, Peromyscus
leucopus (Lima and Zollner 1996). Agricultural lands
that allow some patches of natural vegetation are considered sufficient to
allow gene flow from one population to the other. Zollner and
Addressing the Pervasive
We have provided a few examples of anthropomorphism by omission, but the current scientific literature is literally plagued by examples of anthropomorphism by omission. We have shown that its presence can be detrimental to work in a variety of disciplines from high profile behavioral ecology, state of the art cognitive ethology, language studies, theoretical ecology (aposematism and mimicry), zoos, and decision making in conservation. It is not enough to avoid anthropomorphic vocabulary and claim to be strictly objective. Anthropomorphism is like Satan in the bible - it comes in many guises and can catch you unawares! Lockwood pointed out some of the guises; the most easily recognized are not the problem, but the conceit that one is immune is more often the problem. If anthropomorphism is a natural tendency of human beings, scientists are not immune; lurking unseen it can compromise efforts in many areas. By using critical anthropomorphism and trying to wear the animals 'shoes' we can overcome part of our natural bias and accomplish a more legitimate understanding of the life of other species and of nature (Rivas and Burghardt 2001). We encourage other researchers to put themselves in the position of their study animals not only as a novel, additional approach but as a required step in conducting good science.
As the articles in this volume attest, issues of animal cognitive abilities are prime areas where anthropomorphism by omission may occur, but it is also those working on animal cognitive behavior that are best positioned to discover what is necessary in order to avoid it.
Finally, the view of science embedded in our essay is one being urged in various quarters. For example, in a recent technical article on the use of statistical inference to interpret data in experimental psychology, the following quotation appeared that puts our message in a broader context:
The selection of hypotheses, their number, their location on the continuum of possible hypotheses, and their prior probabilities depend on the researchers' experience, their theoretical frame of mind, and the state of the field at the time of the study (Kreuger, 2001: 19).
and E. Saidel. 1998. The evolution of reference. In The
evolution of the mind, ed. D. Cummins, and C Allen, pp: 183-203.
1997. Expressions of mind in animal behavior. In Anthropomorphism,
anecdotes, and animals, ed R. W.
Mitchell, N. S. Thompson, and H. L. Miles, pp: 198-209.
Bekoff, M. 2000. Animal emotions: exploring passionate natures. Bioscience 50: 861-870.
Boake, C. R. and T. Poulsen. 1997. Correlates versus predictors of courtship success: courtship song in Drosophila silvestris and D. heteroneura. Animal Behaviour 54: 699-704.
D. 1998. The creation and re-creation of language. In Handbook
of evolutionary psychology: ideas, issues and applications. ed C.
Crawford and D. L. Krebs, pp: 613-634.
Burghardt, G. M. (1985). Animal awareness: current perceptions and historical perspective. American psychologist 40: 905-919.
G. M. 1985. Foundations of comparative ethology. Van Nostrand
G. M. 1991. Cognitive ethology and critical anthropomorphism: a snake with two
heads and hognose snakes that play dead. In Cognitive
ethology: the minds of other animals. Ed. C. A. Ristau, pp: 53-90.
G. M. (1996). Environmental enrichment or controlled deprivation? In G. M.
Burghardt & J. T. Bielitski & J. R. Boyce & D. O. Schaefer (Eds.), The
well-being of animals in zoo and aquarium sponsored research (pp.
G. M. 1997. Amending Tinbergen: a fifth aim for ethology. In Anthropomorphism,
anecdotes, and animals, ed. R. W. Mitchell, N. S. Thompson, H. L.
Miles, pp: 254-276.
G. M. (1998). Snake stories: from the additive model to ethology's fifth aim.
In L. Hart (Ed.), Responsible conduct of research in animal
behavior (pp. 77-95).
L. R. and C. M. Heyes 1997. Why anthropomorphize? Folk psychology and other
stories. In Anthropomorphism, anecdotes, and animals, ed. R.
W. Mitchell, N. S. Thompson, H. L. Miles, pp: 59-73.
Crist, E. 1999. Images of animals: anthropomorphism
and animal mind.
A. 1930. Genetical theory of natural selection.
Gabrielsen, G. W. and E. N. Smith. 1985. Physiological responses associated with feigned death in the American opossum. Acta Physiologica Scandinavian 123: 393-398.
A., Jr. and
J. S. 1992. The new anthropomorphism.
Kreuger, J. 2001. Null hypothesis significance testing: on the survival of a flawed method. American Psychologist 56: 16-26.
R. 1985. Anthropomorphism is not a four-letter word. In Advances
in animal welfare science, 1985/1986, ed. Fox,
M. W. and L. D. Mickley, pp: 185-199.
R. (1989). Anthropomorphism is not a four-letter word. In R. J. Hoage (Ed.), Perceptions
of animals in American culture (pp. 41-56).
M. (1964). Philosophy, science, and sense perception.
R. W., Thompson, N. S., and H. L. Miles (eds.). 1997. Anthropomorphism, anecdotes,
Moore, A. J. 1990. The inheritance of social dominance, mating behaviour and attractiveness to mates in male Nauphoeta cinerea. Animal Behaviour 39: 388-397.
Polakowski, K. J. 1989. A design approach to zoological exhibits: the zoo as theater. Zoo Biology Supplement 1: 127-139.
D. J. 1997. Panmorphism. In Anthropomorphism,
anecdotes, and animals, ed. R. W. Mitchell, N. S. Thompson, H. L.
Miles, pp: 92-103.
Terrick, T. D., Mumme, R. L. and G. M. Burghardt 1995. Aposematic coloration enhances chemosensory recognition of noxious prey in the garter snake, Thamnophis radix. Animal Behaviour 49: 857-866.
N. 1951. The study of instinct.
Tinberlake, W., and A. R. Delamater. 1991. Humility, science and ethological behaviorism. The Behavior Analyst 14: 37-41.
Ujhelyi, M. 1996. Is there any intermediate stage between animal communication and language? Journal of Theoretical Biology 180: 71-76.
J. von 1985. Environment (Umwelt) and the inner world of animals (C. J. Mellor
& D. Gove, Trans.). In: The Foundations of Comparative
Ethology, ed. G. M. Burghardt, pp: 222-245.
A. 1996. Wildlife in patchy environments: metapopulations, mosaics and
management. pp: 53- 84. In: Metapopulations and wildlife
conservation, ed. D. R. McCullough.