The Long Riders Guild Academic Foundation
The world’s first global hippological study


Origins of the LRGAF

Deadly Equines

Voices of Authority

Equestrian Wisdom & History Books

An Equestrian Writer's Guide

Academic Research

Historical Research

Military Research

Medical Research

Equine Slaughter & Hippophagy

Breeds & Equestrian Tribalism

Literary Research

Legends & Myths

Horsemanship & Training

Astonishing Rides, Rescues & Races

 Equestrian Inventions


Visit The Long Riders' Guild!

Website designed by Basha O'Reilly


Race, Ethnicity, Species, and Breed: Totemism and Horse-Breed Classification in America

John Borneman

Totem: the iconic representation of a specific ordering of plant and animal species. Clan: the representation of a group identity. Totemism: the relationship between totem and clan. From Emile Durkheim and his nineteenth-century antecedents to Claude Lévi-Strauss, the discussion of totemism has addressed the way in which people classify themselves with reference to the animal and plant world. This discussion began with the observation among different exotic peoples of the widespread practice of arranging certain animal and plant species into a pattern that, while differing from culture to culture in content, seemed to indicate a consistent formal relationship between totem and clan. The iconic representation of so-called nature – the totem – seemed invariably the model for the representation of intra- or inter-group identity for the clan.

This article examines the historical development of the American system of light-horse breed classification[1] – a striking instance in which the animal world is classified according to the categories used for persons, a case of "reverse totemism." The lay understanding of horse breeds is that they are particularly successful experiments in genetic engineering.  Whereas voluntaristic explanations of horse-breed categories, as based on the appreciation of objective and identifiable biological differentiation, are considered quite acceptable, the same kind of explanations for people groupings, such as "race" and "ethnicity," are now generally rejected.  When it comes to explaining their horse pets, most Americans – lay and scientific – will maintain that horse-breed distinctions are by and large objective issues of taste that bear little relation to concurrent developments in social structure, national identity, or state formation. In spite of the common knowledge that horse breeds differ from nation-state to nation-state, this fact of historical cultural contingency does not seem to enter into the consciousness of those who breed, own, and ride horses.

In explaining the American system of breed classification this essay will be organized around the following comparative observation: that the many light-horse breeds in the United States are organized into distinct caste-like species, with an exclusivist set of allegiances to each, whereas in continental Europe light-horse breeds are not seen as so separate and distinct. Furthermore, Europeans tend to be agreed upon a single performance standard for all such light-horse breeds, whereas Americans subscribe to a multitude of standards, some having nothing to do with, or having even an adverse relation to, performance.

The concepts of breed, species, and lineage, and the regulation of these differ markedly between America and the Continent, even though the animals involved in the classification are often of the same "stock" (consanguinity), or are put to the same use (function), or share similar morphological characteristics ("conformation"). While social scientists and historians in this century have been intensively engaged in formal structural analysis of category systems, only recently have they begun to deal with the origin, historical development, and transformation of the categories themselves (e.g., Todorov 1984; Sahlins 1981). I will examine the workings of these three primary (often-used) principles of classification in the structuring of particular breeds and their relationship to the evolution of a people, a national identity, and a state. This involves an explication of both the patterning of the underlying generative structure and the historically specific content. I will proceed via a discussion of the debate on totemism and myth within anthropology.

What Is Totemic Classification?

Let us be careful not to imagine that totemism has vanished like a cloud at the tap of the fairy wand – slight enough, in both senses of the word – of Malinowski.

 – Claude Lévi-Strauss, The Savage Mind

Totemism, after dominating much of the anthropological debate centered around "primitive thought" in the early part of the twentieth century, was defused as an intellectual issue by the early 1960s. Already in 1929 A. R. Radcliffe Brown suggested a shift in the debate, away from a question about the nature of primitive thought to one about universal thought patterns: He asked, "Can we show that totemism is a special form of a phenomenon which is universal in human society and is therefore present in different forms in all cultures?" (1952: 123). In the case of Australian systems, he explained, "The only thing that these totemic systems have in common is the general tendency to characterise the segments into which society is divided by an association between each segment and some natural species or some portion of nature. This association may take anyone of a number of forms" (1952: 122). In 1951 Radcliffe-Brown gave us a succinct working definition of totemism, the full import of which was not further developed until Lévi-Strauss wrote The Savage Mind in 1962. The Radcliffe-Brown definition reads: "The resemblances and differences of animal species are translated into terms of friendship and conflict, solidarity and opposition.  In other words the world of animal life is represented in terms of social relations similar to those of human society" (1952: 116). Working with and expanding upon this general definition, Lévi-Strauss articulated precisely how particularities and generalizations about systems of difference can be fashioned out of homologies made between different levels of categories, and how metaphorical and metonymical relations constitute the sociologic of different cultural classificatory schemes.[2] Moreover, Lévi-Strauss suggested redirecting study to "the ritualization of relations between man and animal" (emphasis added), which, he said, "supplies a wider and more general frame than totemism, and within which totemism must have developed" (1963: 61).

While mindful of the critical contribution of Lévi-Strauss to this debate, the simple tap of his wand – metaphorical and metonymical operations of the mind – does not complete the task of explicating totemism.  The formal ordering mechanisms of cognitive categories will reveal neither the origins nor the development of the structured and structuring practices, which are, after all, the raison d'être of both the categories and the totems. Lévi-Strauss also acknowledged the historicity of any specific application of totemism; he concluded that in "[so-called primitive societies] there is a constantly repeated battle between synchrony and diachrony from which it seems that diachrony must emerge victorious every time" (1966: 75, 155). Yet, in his own work he ignored the implications of this statement, namely, that any account of a system of classification that focuses solely on the semiological aspects will not be able to explain the significance of the conditions of its production. Without a historical account, the most significant aspects of any system of classification – the limits of its ability to generate structures – will remain unintelligible.

We can radicalize Lévi-Strauss's contribution to le totémisme aujourd'hui with three further observations. First, the postulation of homologies between so-called natural and cultural distinctions is not something reserved to the "primitive" mind. While Lévi-Strauss, like Radcliffe-Brown, admitted the theoretical universality of totemic devices, he drew most of his examples from "cold societies." Nor can the use of a totemic device be reduced to "pure thought," whose contemporary manifestation is a genetic inheritance from past times. Where it exists, it does so as a motivated action with mechanisms causing its present production. Second, all totemism is initially a kind of reverse totemism. The animal and plant world does not order itself in a way that is immediately recognizable from a pan-cultural or universal-human perspective. The ordering that humans perceive to exist in the animal world is not initially an inference, it is a projection (Foucault 1973). Third, through an examination of a specific "totemic mythology" we can complete the task that Lévi-Strauss set aside, of explaining "how and why it exists" (1975: 8). Lévi-Strauss stated that the "operative value" of totemic classifications derives from their formal character: "they are codes suitable for conveying messages which can be transposed into other codes, and for expressing messages received by means of different codes in terms of their own system" (1966: 75, 76). As for mythology, he argued that it "has no obvious practical function: . . . it is not directly linked with a different kind of reality" (1975: 10). Here he is absolutely wrong. What I will argue in this essay is that the semiotic codes that make up the patterning of totemism are brought into being by ideological factors, which subsequently also determine the content of the codes. The resulting myths are never innocent, never merely a product where "the mind is left to commune with itself" (1975: 10).

 The Wide World of Horses in America:

Criteria of Classification

 The horse has three gaits – walk, trot, and canter – that are generic to the horse as a species. These three gaits are not in every horse manifested as "pure gaits," meaning that the gait can lose its regularity. For example, the walk, a four-beat gait, becomes a pace (an impure gait) when the horse moves the legs diagonally instead of laterally. The two-beat trot, a diagonal gait, becomes impure when the horse simultaneously moves a front and hind leg laterally (that is, on the same side). At that point the trot loses its moment of suspension. The three-beat canter becomes impure when the horse lapses into a four-beat gait. International performance horses, which compete in the three Olympic sports of dressage,[3] show jumping, and combined training, are bred according to the conformational specifications that relate to the purity of the three gaits. This single performance criterion is also the basic generative structuring principle of the continental European horse world. The American horse world, however, is structured upon various breed-specific criteria, of which the performance criterion is but one among many principles.

Most northern European horse breeding is rigidly (by American standards) regulated by the state, which tests and designates approved stallions and mares considered qualified for breeding stock. This has resulted, over the course of the last two centuries, in the production of superior warm-blood performance horses (crosses between light breeds, such as Thoroughbreds, and local carriage horses and draft horses), some of which have been imported to America, where they are now also being bred. While the international sports performance horse is the focus of European horse categories, it still constitutes but a small percentage (less than 5 percent) of American horse breeding, pleasure riding, and competition. The categorization of the other 95 percent of the light-horse breeds in the United States, which serve purely national or local purposes, is the subject of this essay.

Unlike the Europeans, Americans have more than twenty-six kinds of breed recognized nationally by the American Horse Show Association (AHSA), yet either esoteric or unknown internationally, that are deemed functional and/or meaningful for ends other than international performance sports. These ends include breed criteria of, for example, color (Appaloosa, Pinto, Paint, Palomino), endurance and trail (Morgan, Peruvian Paso Fino), sprinting and cattle herding (Quarter Horse), elegant, high-stepping action (Saddlebred, Tennessee Walking Horse). While European breeds are evaluated on a single performance criterion, American breeds are evaluated not only on multiple scales of performance, but also on arbitrary breed standards other than performance, such as temperament, conformation, and coloring (Haynes 1976: 61).

Whereas the European breeds, focused upon a single performance criterion, will readily take superior animals from other breeds into their registries in order to improve the stock, Americans place great importance on breed "purity" and invest a great amount of energy in keeping the breeds pure and separate. At times the criteria will even be redefined in order to demarcate more clearly one breed from another. While the strategies used to maintain breed distinctions are multifarious, one strategy held in common among the various breeders involves the use of polythetic classificatory devices, such as versions of unilineal or cognatic descent categories, where no single significant feature is to be found in every member of the breed, but each member is assumed to share a common ancestor.

Thus mythologies are constructed around a particular prepotent foundation sire who is said to have originated the breed. Thereafter, all horses included within that breed registry, while perhaps not sharing any "substance" such as temperament, looks, or ability, are traced back to the same ancestor.

Within and across breed standards Americans have developed distinctive styles of riding, which have their correlates both in the ritually demarcated show arena and in everyday riding practices. At the most general level, Western and English styles are the major categories, but both are further subdivided in the horse-show world. For example, the "Western pleasure" (a stylized performance, aesthetically judged) and the "Western gaming" (timed competition) divisions of Western riding have their English riding counterparts in "dressage" and "hunt seat" (both aesthetically judged) and "show jumping" (timed competitions). The breed and performance standards are open to constant reformulation, depending upon changes in the state of the art for performance, and in the state of the science for breeding, and thus constitute a social field of overt struggle over classificatory schemes. Since the knowledge of the making and unmaking of the system of categories is in the public domain, it can be acquired, thus empowering individuals with various forms of ritual expertise. For example, certain performance classes for nearly all breeds are structured male/female and professional/amateur. These classes include, for example, "men's English pleasure" and "ladies English pleasure," or "Western riding, open" and "Western riding, amateurs only." Other riding domains, such as "dressage" or "combined training," require such an extensive investment in training, equipment, and time, that they become, in fact, limited to select classes of individuals. Many of these breed and performance standards conform to the major cleavages in American society and serve as indices of gender, color code, and class.

Yet, the democratic ideal in America has it that everyone has or should have equal opportunity to pursue pleasure, although this does not include a guarantee nor provide the mechanism whereby everyone will in fact have this equal opportunity. And the American myth correspondingly goes that anyone can own a horse. Anyone can also breed a horse, as there is no effective regulation of breeding by the state or by other regulatory agencies, as exists in Europe. Anyone can also show a horse. And, in fact, many people of disparate cultural groups and socioeconomic categories do own, breed, and show horses. Rather than make uniform the standards for owning, breeding, or showing horses, as in continental Europe, the American practice assumes that there is a horse to fit every pocketbook, to match every color preference, every temperament, every personal body-type, and so on and so forth. The horse in America is a democratic ideal.

To illustrate the development of specific American breeds and the distinctive strategies involved in their construction, I will offer an ethnography of four exemplary American breeds that are constituted according to the criteria of function and temperament (the Morgan), conformation (the Quarter Horse and the Arabian), and color (the Paint). These examples are intended to give the reader, on the one hand, a sense of the lack of functional necessity in breed categorization, and, on the other, a sense of the detailed elaboration of symbolic differences.

 Function and Temperament: The Morgan Horse

In the case of the first American breed, the Morgan, the origin is traced to Justin Morgan, a horse received by a Vermont schoolteacher in payment for a bad debt. Conceived in 1789 in Springfield, Massachusetts, this horse possessed excessive strength for his size and a temperament characterized as industrious, docile, kind, commonsensical, versatile, and independent (Horse Identifier 1980: 96).[4] A biography of this foundation sire reads, ". . . the small bay stallion entered a life of hard labor – and, at first, little recognition. Although standing only 14.0 hands and weighing scarcely 950 pounds at maturity, Figure (his original name) was put to any task which required horse power, from skidding heavy logs to racing the local talent at day's end. However the spunky little stud proved he not only had the mettle to attempt anything asked of him, but invariably left the competition eating his dust as well" (Mellin 1973: 5).

According to one widely held myth, the Morgan had been a necessity in the past: How could America have been settled without him? The founding of breed registries, of which the Morgan in 1894 was the first, is accompanied by an extensive list of rules, based on morphological criteria, to determine what the breed is and what it is not. In the words of one expert, "acknowledging that of course there are always likely to be found some variations on type within so versatile a breed, those people who are dedicated to the Morgan per se realize their obligation to perpetuate, in fact, his basic character and disposition" (Mellin 1973: 47). Once a breed is established, morphological criteria and a particular expression – the Morgan "look"– are used to distinguish it from other breeds: "It is a bright, proud expression, at once intelligent, mischievous, a bit defiant and totally irresistible! It is usually coupled with a snorty attitude, a tossing mane and an abundance of nervous animation" (Mellin 1973: 47). Can it be mere coincidence that the Morgan look matches the characteristics ascribed to the archetypical Jeffersonian yeoman farmer? This Jeffersonian ideal-typical construction of The American is the sum total of perceived American virtues.

It is appropriate that the Morgan, as the first official American breed to serve as a marker of distinction, simultaneously claims to incorporate the essence of Americanism. The development of a distinctively American way of life was accompanied by what the historian John Higham, in a definitive account of this phenomenon, calls racial nativism: "intense opposition to an internal minority on the ground of its foreign (i.e., 'un-American') connections" (1967: 4). Higham points to popular movements in the 1790s and 1850s in which nativistic agitation to define The American was particularly widespread. It was assumed, in both the popular and "scientific" literature of the nineteenth century, that there were at least three distinct European and many "non-Western" races. Yet, it wasn't until the end of the nineteenth century that political and literary speculation on "racial difference" received its "scientific" legitimation from biologists, anthropologists, and genealogists (Higham 1967: 149-57). In a parallel manner, the initial political and social origin of the Morgan horse of 1789 was only later, in 1894, legitimated by "science." Not until the end of the nineteenth century did the Morgan become a separate and distinct breed, based on scientifically verifiable "natural" biological distinctions.

While the notion of breed purity is common to most American breed registry associations, European warm-blood breeders seem unconcerned with purity, often using other breeds in order to improve their own, and switching identities freely. For example, a superior horse in the Hannoverian registry may be purchased by the Westphalian registry (or vice versa) and its offspring will be registered without qualification. It is worth noting, then, what happens when Americans import these European "unprotected" breeds. To keep the imported bloodlines pure, Americans have created separate registries for each European warm-blood, reproducing the breed-exclusive registry that exists within more established America breeds. In some cases, separate registries are even kept for those bred in Europe and for those bred in America, as if the geographical locus of conception made a difference. The justification given is that the change in climatic conditions results in a loss of breed purity. The cross-breeding of imported European warmbloods with American breeds (such as Thoroughbred and Quarter Horse) has resulted in the recent creation of man half-breed registries. There have also been attempts in the past ten years in America to create single performance horse registries that would incorporate qualified "performance" horses from different breeds. But, significantly, these registries do not intend to challenge the breed-exclusive categories, for they allow and even encourage listing in both the specific breed and the performance registry. In other words, dual identities are created.

Conformation: Arab Breeders and Quarter-Horse People

The construction of breed criteria based on morphological characteristic can be clearly illustrated by comparing two well-known and popular American breeds, the Quarter Horse and the Arabian. Although it has been an official breed only since 1940, the Quarter Horse embodies more than any other breed the American West, and its origin is often expressed in a folksy Western manner. L. N. Sikes writes, "The history of the Quarter Horse goes back a long way – back before the time when anybody kept good accounts of what stock horses in this country even looked like (1958: 13). Of course, Sikes is not unaware of the part he plays in myth building, for he appropriates other myths of the Quarter Horse as myth such as a virility myth told of Steel Dust, a foundation sire of the breed: "So much of a reputation did Steel Dust get that, pretty soon, he began siring colts in places he'd never been. In fact, up until recent years, lots of cowmen would refer to a Quarter-type horse as a 'steel dust'" (1958: 13). One chronicler of the Western horse, Robert Denhardt, comments that "every horse trader who has not recently joined a church will modestly admit that his horses are direct descendants of Steel Dust" (1967: 17). While the Morgan represents a particular east-coast-derived America archetype, the Quarter Horse appeals to a more general Western myth.  Says Denhardt, "The Quarter Horse is a scion of the oldest and most aristocratic of American equine families, a race which can trace one side of the family to the Mayflower and the other to the Conquistadors of Spain" (1967: 178).

Arab breeders, as they are called by other horse people, are among the most eccentric horse owners in America. This is attributed to many factors: They often fear riding their own spirited horses, they usually prize beauty and sentiment over function, and they commonly display an extreme fetishization of the parts of the horse. When asked to identify what makes an Arabian an Arabian to an Arab breeder, one man replied, "They have lots of mane. Lots of head with big eyes. And they are surrounded by fog." Arab breeders have made a fetish of the head, so that regardless of how the body of the horse is put together, or of how the horse moves (how it carries its rider), a proper Arab head – bulging eyes and nostrils, dished forehead, wild and flowing mane – is the most prized and determinative characteristic of the breed. Most pictorial representations of the Arab show only the head and at times part of the neck. Sometimes it is surrounded by mist.

This contrasts with representations of the Quarter Horse, which often focus on the rear. At times the horse's head is turned so that it is staring back at the camera, though many photos in Quarter Horse advertisements show only the rear end. The history of the Quarter Horse parallels that of the territorial expansion of the United States. It originated in the southern United States, and was bred to run a sprint, a quarter of a mile. The thrusting power of a horse is to be found in a powerful rear end: a long hip with a muscular loin and well-developed gaskin muscles. The Quarter Horse was bred for this powerful rear end. Because of its special stopping and starting ability, it is particularly suited for use in cattle herding on the open range. With the western expansion and the growth of the open range, Americans, not surprisingly, found Quarter Horses to be most suited to their needs. Thus, this American sport horse was developed for purely local ends – for short-distance racing and cattle herding. Yet, a third use of the Quarter Horse arose, that of a docile and tractable show horse. The number of horses devoted to this third use presently far exceeds that in the racing and cattle-herding oriented uses for which the horse was initially bred.

Paradoxically, as the initial functions of the Quarter Horse became less important, the prime symbol of those functions – the distinctive rear end – took on increased significance. The fetish surrounding this body part has resulted in the breeding of some horses that are uncomfortable to ride, for often the rear end is out of proportion to the front, creating not a more powerful engine but an unbalanced, downhill-moving horse. This kind of horse, known as the Bull-Dog type, "hits the ground hard" when it moves, jarring the teeth of its riders and decreasing the longevity of the horse, as its legs and feet have difficulty withstanding the pounding. One admirer of the leaner type of Quarter Horse refers to the Bull Dog as a "muscle-bound weight lifter trying to be a boxer" (Davis 1962: 9).

During the 1960s the popularity of showing horses greatly increased and show classes proliferated.[5] Halter classes, where the horses are led into the ring and judged purely on conformation with no concurrent attempt to link this to a performance aspect (like a swimsuit beauty competition, where the contestants need not swim), expanded greatly in number and took on increased status and significance. This was accompanied by increased specialization within each breed, so that a racing Quarter Horse was of a type different from a Quarter Horse used for show. And the show horse category was further subdivided into horses that could win "at halter" and those that could win "in performance." With the increased importance of the show-horse aspects to the majority of breeders, who prized (with money and status) competitive wins in the halter classes over those in performance classes, less attention was paid to those aspects that keep the horses sound, that is, physically healthy. Put simply, beauty became divorced from function. In order to accentuate the bulk (thus, the beauty) of the body, Quarter-Horse breeders during the 1960s and 1970s also selected horses with refined bones and feet, and they introduced some Thoroughbred blood – Thoroughbreds have slighter bodies and more refined limbs – into the registry. Many breeders today acknowledge that these particular conceptions of beauty, having an attenuated and even adverse relationship to function, have resulted in a physical weakening of the breed[6].

A particularly distinctive and popular class at Arabian horse shows is the "Arabian costume class," for which owners dress their horses and themselves in what they think of as typical Arabian costumes. The horse's head is appropriately highlighted, with plumes on the top. The horse's body is covered (who wants to see the body?) by sequined silk multicolored blankets with small white or blue pompons sewed around the edges. While most of the body is cloth covered, the head and neck are further accentuated with silver-plated bridles on which are inscribed the names of the horse, names of Arabic origin that none of the participants can pronounce. After all the performers are gathered in the arena, they circle the judge, each at a pace the individual prefers. They are then asked by the judge, usually a male, to perform individually. From a line-up in the middle of the arena, the performers individually circle the judge and the other riders, displaying their costumes to best advantage. Most often this is done at a hand gallop – a very fast canter – so that the rider's long silk robe and the pompons on the horse's blanket will billow in the wind, creating the feeling of speed and light movement. In several classes that I have watched, the female performers played upon Western ideas of the sexuality of the Arab world by going braless under their light silk robes. A Wisconsin woman, who had imported an Egyptian Arabian stallion, would also, after her gallop, race up to the male judge, stopping within inches of him, and lean forward so that he could smell her perfume and see her form under her Arab silks. She was always the talk of the show, and never lost a costume class during the four years (of the 1960s) in which I knew her.

Color: The Paint

The color breeds – for example, Paint, Appaloosa, Pinto, Albino, Palomino – while incorporating all three breed-classificatory criteria (function, conformation, and color) in both their mythology and their current registry standards, single out color as the necessary and sufficient condition of incorporation. A color breed such as the Paint, in order to constitute itself as a separate and distinct breed that is understood as a biological (that is, genetic) category, must assert that certain criteria, either singly or in combination, are uniquely Paint. In the effort to maintain and justify the distinctiveness of the color category, Paint owners and breeders engage in ceaseless semiotic and ideological maneuvering. In an article on the social uses of color codes, Marshall Sahlins provides an initial framework for explaining the semiotic aspect of the construction of the Paint breed: "Colors are in practice semiotic codes. Everywhere, both as terms and concrete properties, colors are engaged as signs in vast schemes of social relations, meaningful structures by which persons and groups, objects and occasions, are differentiated and combined in cultural orders" (1977: 166).

The Paint provides a specific example of the development of a color code, and can be used to extend Sahlins' insight. Two related questions should be kept in mind: First, can one isolate the various strategies used to produce this cultural order, and, second, what is the relationship between the ideological motivations and the semiotics of color coding?

Even though the founding of the Paint-Horse registry, with which its legitimation as a breed begins, did not occur until February 1962, much importance is placed upon its ancestry, that it has existed forever, or nearly so. Its beginnings are traced back to 20,000 B.C. in Spain, southern France, Arabia, and North Africa. There is some confusion in the early records (that is, prior to 1519, when the Paint is definitively dated as reaching America), as it is "impossible to trace the movements of each breed" because no distinction was usually made between the different kinds of spotted horses (Haynes 1976: 3).

After the fifteenth century the record supposedly becomes clearer, and we are told that, for instance, the Indians in America domesticated the Paint Horse "because they had an eye for anything bright or colorful," and that they apparently thought Paint Horses were better camouflaged than solid colored ones (Haynes 1976: 24, 25). "Paints" continue to be associated with the domesticated Other, which in this case is the American Indian. Popular forms of representation, from such diverse media as films and horse-show costume class performances, index the Paint as "colored." From the television serial "The Lone Ranger" in the 1950s, where the Ranger (white) rides a Palomino but Tonto (Indian) a Paint, to the film Silverado in the 1980s, where the hero rides a white horse and the buffoon a Paint, the mythical history of the breed is reconfirmed. Today the Paint Horse and the other color breeds can be found in all parts of the United States, although they tend to be more concentrated in the West.

Since 1962 the criteria of classification have changed several times because it has been difficult to decide, as one breeder put it, "How much color is a Paint Horse?" (Haynes 1976: 54). Initially, it was required that the Paint be recognizable as such from both sides, but this has been changed so that now only one side need be so recognizable. At present the other requirements for inclusion are height (fifty-six inches at age two years), gaitedness, and conformation approval (in the case of a stallion).  For those that fulfill these requirements but are not offspring of registered parents, patrilateral descent, that is, paternity from a stallion in another approved registry (for example, Quarter Horse or Thoroughbred), is an added requirement. Apparently, the dam's lineage can be considered inconsequential for the purposes of breed inclusion. This American oversight – denial of the genetic importance of the mother – is not peculiar to the Paint breed; it is a quite common practice in the origin of most breeds[7]. 

The requirements for participation in show classes for the Paint Horse are similar to those of the Quarter Horse, which the Paint Horse closely approximates in function and performance[8]. The one characteristic that all Paints are supposed to share – color – is also the only characteristic that is ruled out as a consideration when judging halter classes. The only class in which color is to be considered a factor is in a color class, a non-point (no cumulative year-end award) class judged solely on color. For show purposes, any Paint Horse, once accepted in the registry, has sufficient color.

What is most significant about the Paint Horse breed is that the criterion on which it bases its claim to a breed status is not and never can be genetically isolated, for "noncolor breeds" will at times produce horses that theoretically qualify as ideal Paints, and Paint Horses will often produce full-colored offspring. Many color patterns are carried by recessive genes[9].  Some horses unpredictably change color patterns several times during their lives. Nevertheless, given the social importance in America of horse breeds as a form of distinction making, each breed must necessarily be conceptualized as distinct, and invested with an ideal type. Horses not adequately measuring up to this ideal will be rejected as impure. Thus, the Quarter-Horse registry will refuse to register full-blood Quarter Horses with impeccable lineages but with too much white, or with color in the “wrong” place (that can, in other words, qualify as Paints), and will demand that the animals be withdrawn as breeding stock (by gelding the stallions and and spaying the mares). The rules for inclusion as a Paint are the the same as the rules for exclusion as a Quarter Horse, namely, that color markings as decided by "body contrasts" must or must not be of a particular type and in a specific location (Haynes 1976: 70).

The use of color as a "primordial classifier" illustrates the social embeddedness of semiotic codes and the consistency of ideological motivations  with any particular cultural order. The resort to mythical histories, to genealogies based on shared blood or patrilateral descent, and to functional claims (conduct codes) parallels the kind of strategies most often used to constitute the kinship systems of human groups: The consistent breakdown of the category-use does not result in a denial of their naturalness, but rather in a reformulation of the color combinations that signify the breed.  Color use never merely names objective differences in the visible spectrum of signified color patterns; it also, as Sahlins has argued, arbitrarily signifies and communicates culturally constituted social relationships.

 The Production of Distinction: Good Horse Flesh

As we have just seen, the practico-theoretical logics governing the life and thought of so-called primitive societies are shaped by the insistence on differentiation.

– Claude Lévi-Strauss, The Savage Mind


The practice of sports. . . owes part of its "interests," just as much nowadays as at the beginning, to its distinguishing function, and more precisely, to the gains in distinction which it brings.

–Pierre Bourdieu, "Sport and Social Class"

It should now be clear how American horse breeds, based on arbitrary distinctions of function, conformation, and color, are constantly manufactured. Furthermore, these distinctions and significations are not derived from some natural pan-cultural ordering of the animal kingdom, but are peculiarly American social distinctions that are in turn mapped onto differences found within a given species. The differences are indeed arbitrary: fetishes of parts (nostrils, eyes, and necks for the Arabian, rear ends for the Quarter Horse), priorities of color (for the Paint Horse or the Palomino), mythologies of origin and temperament (for the Morgan). By arbitrary I do not mean that these significations are unmotivated signs, but simply that the motivation chosen is one among other possibilities. There are, indeed, limits to the arbitrariness of a signification, but these limits are ideological constraints, not semiological ones. For the example of horse breeds, the meaning attached to function, conformation, and color is possible precisely because it is motivated by similar divisions in the social world of people[10].

In the remainder of this essay the question to be addressed, then, is how and why this particular set of ideological constraints motivates the semiotics of horse-breed distinctions. Why do Americans set up new rules for each breed rather than, like the Europeans, agree to one set of performance rules for all breeds? A passage from Lévi-Strauss, explicating the relation of totem to caste and the "conceptual transformations marking the passage from exogamy to endogamy," suggests an answer to this question:  

But if social groups are considered not so much from the point of view of their reciprocal relations in social life as each on their own account, in relation to something other than sociological reality, then the idea of diversity is likely to prevail over that of unity. Each social group will tend to form a system no longer with other social groups but with particular differentiating properties regarded as hereditary, and these characteristics exclusive to each group will weaken the framework of their solidarity within the society. The more each group tries to define itself by the image which it draws from a natural model, the more difficult will it become for it to maintain its links with other social groups and, in particular to exchange its sisters and daughters with them since it will tend to think of them as being of a particular "species" (1966: 177-78).

If Lévi-Strauss' conceptualization is correct, then the movement from exogamous to endogamous horse breeding would involve specific formative steps. How did the American horse breeds become species-like (or, one might ask, are endogamy and caste-status preconditions for constituting a breed)? This can be illustrated by examining the three processes involved in the creation of "good horse flesh": breeds are conceptualized, registries are created, and breed divisions are enforced.

While the definition of breed varies from writer to writer, certain themes are held in common. Wayne Dinsmore says a breed exists when a "substantial number of animals within a species . .  differentiat[e] them[selves] from others of the same species" (1978: 103). J. W. Evans relies on an argument of human or abstract needs to explain breeds: "Breeds were developed according to the needs of horse people in each locality, or they developed naturally" (1977: 101). Along with these two basic assumptions, many writers date the official origin (other than that offered in their mythologies) to the foundation of a registry. The importance placed on the official codification is often so great that, in his history of the Appaloosa (1975), for example, Jan Haddle dates the breed to the opening of its official registry in 1949, even though this is a year after the first full-Appaloosa horse show.

Haynes makes the above connections explicit: A breed is "a variety or stock of animals related by descent with certain inherited characteristics and capable of reproducing those characteristics," but has also come to mean "animals registered in some association or club" (1976: 61). This emphasis upon being recorded in the stud register or the registry of a particular breed is also regarded as the distinguishing characteristic of breed by the Manual of Horsemanship, which is published by the British Horse Society and The Pony Club, and is the most authoritative text on horse basics in all the Western English-speaking countries. The registry – that is, the concept consanguinity – has now taken on an institutional raison d'être, so that superior animals fitting the breed specifications will be rejected if their progenitors are not already members of the registry[11].  This needs to be so, says Haynes, because "it takes several generations of recorded and scientific breeding to establish a family or strain" (1976: 62).  Horse breeders themselves recognize the creative and constitutive nature of their endeavor. "There is no reason," says John Gorman, "why a group of horse breeders of a certain type and color cannot preserve the purity of the horses' breeding and eventually establish a breed" (1958: 229).

In order to establish "pure breeds," most breed associations at some point established extensive inspection programs through which individual animals were (or are) initially approved for a breed registry. Haynes points out that the desire to "ensure uniformity in the breed" led to "rigid stallion inspections" as one of many necessary "police actions" (1976: 52). The Quarter-Horse inspection program (no longer active because the breed registry is closed) can serve as an example of how these actions were carried out. If a horse's progenitors were not both registered, or if the progenitors were unknown, an owner could petition for inspection. After an initial inspection by an officially approved inspector, who was most often flown to the site where the petitioned animal was kept, the animal would be either rejected or accepted into the registry as an appendix-registered horse. The horse would then be required to compete and win in several different divisions (for example, racing, halter, performance) at official Quarter-Horse shows in order to accumulate the required number of points.

Thereafter, animals that fulfilled these requirements were automatically admitted into the full registry, with the possibility that they could be used as breeding stock[12].

The State, Social Structure, Cultural Categories

Just as there are many distinct, mutually exclusive – that is, caste-like – American horse breeds, there are also many styles of riding. On the Continent[13] not only is there a single standard – performance – for breeding, but also a uniform set of presuppositions concerning how training and riding the horse should progress[14]. These continental presuppositions are formalized and systematized in the art of "dressage." I will briefly sketch how this codification progressed, concentrating on the post-Medieval period, as it allows us to bring together the interactions among cultural categories, social structures, and the nascent states of Europe[15]. This comparative discussion of the sociopolitical context in which breed categorization developed is intended to illuminate the American situation, and not to provide an exhaustive account of European state and horse-breed development.

The first written texts on the different aspects of horsemanship in the West are attributed to Xenophon in 500 B.C. For the next twenty centuries, however, there was very little written on breeding type and riding style.  According to the current literature on horses and horsemanship, there were also no appreciable improvements in understanding, training, or breeding of the animals (at least in the West)[16] until the fifteenth century.  After the fall of the Roman Empire, the use of infantry declined and that of cavalry increased, although this does not appear to have had any immediate effect on improving the caliber of horsemanship. Yet, between 1500 and 500 B.C. the domesticated horse, arriving from the East, became common to much of Western Europe. This "noble beast . . . was associated with a new social distinction," comments Emmanuel Le Roy Ladurie, "[and] in its way it marked the appearance of a group of aristocrats, living off contributions levied from the peasants" (1979: 80). In addition to its function in changing agricultural techniques, the horse was simultaneously appropriated as an indexical symbol, associated with the growing nobility and with knighthood.

The formalization of riding and the development of haute école began again during the Renaissance in the fifteenth century. It reached a fountainhead at the riding School of Naples, which served as a source for subsequent developments among the Hapsburgs (and the Spanish Riding School) in Austria, the French, and the British. A student from that school, de Pluvinel, is thought to be the originator of the French school.  Antoine de Pluvinel's book, The Instruction of the King in the Art of Riding, printed in 1626, was written as a dialogue between the author and his pupil, Louis XIII. Following upon these beginnings, Louis XIV, from the splendor of his court, created a French riding school –  the School of Versailles –  in the Tuileries. In 1733 the Frenchman Robichon de La Guérnière, who conducted this school, wrote what is still considered the classic text on dressage. His teachings formed the basis for all subsequent cavalry schools, the most famous of which is the Cavalry School of Saumur, founded in 1771, a direct descendant of the School of Versailles.  Advances in horse breeding and training in France subsequently emanated from contacts with the Saumur school, which to this day retains its role in licensing instructors, trainers, and judges.

What one can conclude from this account is that horse riding in France is marked by the reliance of a particular cultural standard upon the court culture and upon the development of a military elite situated in and near Paris. The growth of the performance standard in France is inextricably linked both to the process of formalization of rules of etiquette among French nobility and to the utilization of the horse in the military-political centralization of France.[17] One is reminded of Alexis de Tocqueville's account of the demise of the ancien régime, his central thesis being that "in France, more than in any other European country, the provinces had come under the thrall of the metropolis, which attracted to itself all that was most vital in the nation" (1955: 72). Despite the fact that Louis XIV "tried to check the growth of Paris" six times, administrative centralization was the marked tendency of the French territory from the beginning of the seventeenth century, where the "true owner" of the landed estates in the kingdom "was the State and the State alone" (Tocqueville 1955: 189).  Tocqueville illustrates the tendency of the provinces to look toward Paris for leadership in all aspects of cultural and political life with a comment from a provincial: "We are only a provincial town; we must wait till Paris gives us a lead." He infers from this that the provinces seemed "not to dare to form an opinion until they knew what was being thought in Paris" (1955: 74) The German historian Otto Hintze is in accord that "the transition occurred only in the seventeenth and eighteenth centuries, the period of absolutism and administrative centralization; it was finally completed by Napoleon. As a unified state, moreover, France became the model of development for the entire continent" (1973: 168).[18

Whereas the development of "France" is marked by political centralization, that of "Germany" is by bureaucratization. The most common explanation for the bureaucratic centralization of Prussia in the eighteenth century is that external, chiefly military, threats from the new states in the West forced internal developments toward increasing centralization (Cf. Hintze 1973: 168-69; Craig 1978: 1-34; Wehler 1973). [19] Thus, in the Prussian case, military needs, initially tied to cavalry effectiveness, necessitated breeding of animals that correspond somewhat to our present light-horse category.[20]  This effort followed upon, and was modeled after, the successful school at Saumur in France, for monarchs and nobility commonly borrowed from each other. Since the purpose here is not to explain differences on the Continent,[21]but the differences between the American model and the continental one, it is important to emphasize that the Germans and French are much alike in holding a particular pan-national breed standard (although the manner in which and reasons for which these standards were propagated differ), whereas in America, with a weak state formed on pluralist (federalist) premises, there occurred a proliferation of breed standards.

For Americans, nation building – the creation of shared cultural-identity markers – took precedence over state building. In marked contrast to Europe, America had neither a court culture, nor large, threatening states on its borders. The classic formulation as to why political centralization never occurred to the same extent in America as in Europe was put forth by Louis Hartz, who argued that the absence of feudal social institutions, including an aristocratic cultural elite, made unnecessary the centralization of power required to dislodge it (1955: 43-46). Samuel Huntington concludes from this initial premise that, while the American state, in its subsequent development, "often helped to promote economic development,  . . . it played only a minor role in changing social customs and social structure" (1973: 193). Furthermore, without "external enemies" as a backdrop against which a national identity is imagined,[22] American national identity was (and is) imagined as an internal affair. This has meant both a nativist universalism, as Higham has articulated, and a pluralist melting-pot folk ideology concerning the nature and expression of cultural difference.

In France and Germany, centralization –  political in the French case and bureaucratic in the German one –  played the key role in establishing a universalistic cultural standard, which for horse breeding is expressed in a single performance criterion. In the United States, diverse cultural standards and social strata were never subject to strong political or bureaucratic structures capable of or in need of bringing about a uniform performance standard. Consequently, particularistic cultural standards, as in the domain of horse breeds, were generated as "separate but equal" social identity markers. Where state building precedes nation building on the Continent, centralized administration and hierarchical modeling leads to uniform breed standards. Where nation building precedes state building in the United States, the denial of social difference and of hierarchy leads to heterogeneous and separate but formally "equal" cultural breed standards.

Race, Ethnicity, Species, Breed

Totemism, as Lévi-Strauss argues, is the postulation of a homology between two systems of differences. It is a way of thinking that has no intrinsic content. Lévi-Strauss makes it clear that totemism is the establishment of homologies not between the terms themselves, but only between the differences "which manifest themselves on the level of groups on the one hand and that of species on the other" (1966: 115). He represents this as follows: 

Nature: species 1 # species 2 # species 3 # . . . . species n
Culture: group I #  group 2   # group 3 # . . . . group n

It is the same "pure totemic structure" that I am suggesting exists in the creation of breed categories in America. The relationships suggested are not between particular breed and particular ethnic or racial categories but between a system of human differentiation and a system of breed differentiation. It can be represented as follows:

Breeds: Quarter Horse # Morgan  # Paint # Arabian # breed n
Humans: Anglo-Saxon  # German  # Black  # Italian  # ethnic/racial group n.

The melting-pot ideology notwithstanding, American social groups increasingly tend to be in practice statistically endogamous, occupationally differentiated, and residentially separated from one another (Thurow 1980; 1975; Harrington 1984; 1980). The term ethnicity, which has traditionally been used to refer to forms of regional identity based upon customs and influences from outside the society in which the groups now live, is increasingly recognized as indicating relationships based upon differences used to demarcate indigenous groups from one another (Glazer and Moynihan 1975). Michael Hechter has postulated that a common form of ethnicity is a reactive group formation, where groups adopt historically established distinctions to demarcate themselves from other groups, rather than adopting these identities in an interactive, closed group (1978). Sociolinguistic work by John Gumperz (1982a; 1982b) indicating the persistence of linguistic and discourse differentiations has forced re-evaluation of the logic behind and reasons for the use of diverse discourse conventions.

There is no need to go into detail on the significance and persistence of collective descent and race as distinction-making categories in American history (Gossett 1963; on the relationship of race to ethnicity and class, see Altschuler 1982). Many writers have argued that race is the most significant category for Americans. A 1937 study of etiquette and race in the American South begins with the assertion that "the American people seem to exhibit a perennial interest in problems pertaining to contact and association of the many races which constitute the general population" (Doyle 193T viii). The logic of racial differentiation is often explained in a manner similar to Hechter's reactive-group-formation thesis. L. Copeland states, "Wherever the groups and classes are set in sharp juxtaposition, the values and mores of each are juxtaposed. Out of group opposition there arises an intense opposition of values, which comes to be projected though the social order and serves to solidify social stratification" (1959: 171). 


That is why myth is experienced as innocent speech: not because its intentions are hidden – if they were hidden, they could not be efficacious – but because they are naturalized.


[M]ythology harmonizes with the world, not as it is, but as it wants to create itself.


– Roland Barthes, "Myth Today"

Horse breeders and riders experience their breed classifications innocently.  The myths they spin about their cultural performances, while not maliciously motivated, are also not harmless. These classifications do not arise from virginal minds. Rather, at the point of their origin and in their reproduction they serve as perpetual alibis, by naturalizing and legitimating the social order about which they speak. Unlike "the primitive," the century-old creation of anthropological study, who is said to deny nature and to reify culture, Americans seek to reify certain parts of their culture which they falsely identify as nature. Americans forget that "nature," in as far as it is experienced and becomes part of a human life-world, is also a cultural construct.

It is no accident that American notions and usage of horse breeds  – where the concepts of race and ethnicity, blood and breeding, have all become reified into biological naturalisms – are first taken from human categories and then projected onto animal classifications. In the American case, Durkheim's basic insight about the relation of totem to clan can be confirmed. The concept of horse breeds is, in fact, a stolen language, stolen from our practice of social structure. This is a specific instance of a general phenomenon that can only be explained as a reverse totemism. Horse breeds are thus an example of what Sahlins has called scientific totemism (1976: 106).

The creation, in our image, of a world of differentiated animal species is also inextricably linked to the reciprocal influences of cultural categories, social-structural practices, and the formation of nation-state identities.

Although in everyday speech the terms breed and species, or race, ethnicity, and species may not be confused, in practice the different horse breeds are treated as if they were separate species. Since breed, like race, is confused with and often considered a matter of genetics and biology, and not culture, and since biology is considered the ultimate arbiter of phenomenological disputes, the naturalness of this social order is never questioned. In this case, the post-Mendelian scientific discourse on genetics enters the history of horse breeding after the forms of classification and their motivations have already been culturally cast. Today, the relationship of the science of genetics to popular representations and practices, when not serving a merely legitimating role for what is already there, remains tenuous.

Though this essay focuses on the analysis of the totemic nature of horse-breed categorization in America, the theoretical implications of the argument are broader. As Roland Barthes so perspicaciously argued, and as the specific examples here illustrate, the mythical systems produced through classificatory devices, while experienced as innocent speech, are in fact constructed, first, by a plagiarism of the social world, and second, by a harmonization of that plagiarism with its dominant discourse. This kind of myth is neither simply a charter for reality nor is it an invention of pure thought. It is both a language for analogically representing and reconstituting another reality – a hierarchical system of human differentiation –  and a means by which that reality can be validated.

John Borneman is a Professor of Anthropology at Princeton.  This article was first published in 1988 in "Comparative Studies in Society and History in 1988.  It is reproduced here with his permission.  It is also included in Modes of Comparison - Theory and Practice, edited by Aram A. Yengoyan, published by the University of Michigan Press in 2006.

 Notes and References     Page 2 - Postscript     Top of page          Back to Breeds page            Home       

© COPYRIGHT 2001 - 2014