IMPROVING THE SUCCESS OF BREEDING PROGRAMS FOR LIVESTOCK GUARDING DOGS, WITH REFERENCE TO SOUTH AFRICA Matthew Berry, Owner, Boavida Kennels, Durban, South Africa Daniel Stewart, Primary Animal Health Care Coordinator, U.N.W.H.O; Rabies Elimination Project, KwaZulu-Natal, South Africa Jean-Marc Landry, Institute for the Promotion and Research on Guarding Animals, Martigny, Switzerland Raymond Coppinger, Professor Emeritus of Biology, Hampshire College, Amherst, MA USA Director, Livestock Guarding Dog Project, Hampshire College Abstract Livestock guarding dogs (LGDs) are ubiquitous throughout pastoral communities. They possibly are the most common working dogs in the world. In North America or Europe, however, except in transhumance cultures, they have been adopted mostly as purebred pet, and show dogs and sometimes as household guardians. During the last 40 years, in response to changing environmental values, programs have been created to introduce ranchers and farmers to LGDs as one method of non-lethal predator control. One of the major questions is where do we get the thousands of dogs needed to populate North America, Europe, and now South Africa. With LGD programs in America and Europe showing promise, other countries have created similar ventures. Researchers from established programs have assisted setting up new projects, or helping to solve problems within existing programs. For the four of us, South Africa is our latest endeavor. This report focuses on creating and assisting the introduction and education of South African stockmen to LGDs and compares their acquisition, breeding, training, socialization and costs with other projects in which we have been involved. Wildlife Protection with LGDs Starting in the 1970s, in the USA and Europe, wildlife management laws and attitudes about public policy rapidly changed. Large carnivore populations were locally extinct or endangered. In the USA, conservationists became alarmed at the effect of lethal control of predators on other wildlife species. Traps, poisons, bombs and guns on the western ranges killed not only cougars, coyotes, bears and eagles but other wildlife. In more populated areas lethal control was dangerous to people as well. Thus in the 1970s, legislation against lethal methods of predator control increased. Because a large segment of the public disapproved of the extermination of predators and endangered species (Reiter et al., 1999; Breitenmoser et al., 2005), programs were created to protect these animals and to restore species in danger of eradication. Endangered species were now encouraged to repopulate regions. Either the remaining individuals were protected (e.g., wolves in the Alps, Portugal, Slovakia, and Italy), or a species was reintroduced into a previously occupied area (e.g., lynx into Switzerland or wolves into Yellowstone National Park). Thus an immediate need arose for alternative nonlethal methodologies. LGDs were not common in either the USA or Europe. In the USA LGDs were used by the Navajo Indians and some Hispanic cultures (Black, 1981; Black and Green, 1985). English-speaking farmers and ranchers, however, had no memorable cultural history with LGDs. Most European countries had landrace “breeds” which were often thought of as a “national dog” However, the wide use of LGDs was lost during and after World War II, with the few exceptions of landrace dogs associated with Mediterranean transhumance cultures. In order to demonstrate to US farmers and ranchers the value of Old World LGDs, two projects were created in the late 1970s. The first was at Hampshire College in Massachusetts, funded by private foundations and the U.S. Department of Agriculture (Coppinger et al., 1988). This was a three-part program which 1) studied the behavioral differences between breeds of dogs, 2) ran a cooperator program with 1500 landrace dogs collected and bred from working stock of transhumance pastoralists in Italy, Yugoslavia, and Turkey, and 3) continuously reported research results, cooperator success and failure stories, and the impact of LGDs on all the farms and ranches in the program. The second project was at the U.S. Sheep Experiment Station in Dubois, Idaho (USDA–APHIS) (Green & Woodruff, 1990b). This study had virtually the same goals and the same sample size of dogs as the Hampshire College study. The main difference between the two projects was where they got their dogs. The Dubois study used mainly LGD breeds (Great Pyrenees, Komondor, and Kuvasz) recognized by the American Kennel Club (AKC) and purchased from American breeders, while the Hampshire study used only working dogs collected from pastoralists. The results of both projects were similar and encouraging. Over 70% of the dogs were successful in reducing predation (Coppinger et al., 1988; Green & Woodruff, 1990a), and no significant differences could be shown between the breeds and the landrace dogs. The most important finding was that both attentive and trustworthy behaviors are a consequence of early puppy socialization with livestock. Protectiveness was simply a consequence of a dog or dogs present among the stock — and this is what deterred predators. European projects were numerous and similar. University scientists and their students started LGD projects in a dozen countries. LGDs were acquired from breeders (non-working dogs) or from pastoral landrace dogs. Some projects restored ancient landrace dogs, e.g., in Bulgaria (Sedefchev, 2005) or in Portugal (Ribeiro & Petrucci-Fonseca, 2005). Other projects focused on the weakness of some traditional prevention methods and proposed recommendations to increase the efficiency of LGDs (Mertens & Schneider, 2005; Rigg, 2005), while others introduced the use of LGDs in a non-traditional way (e.g., patrolling with a LGD in areas of unattended flocks; Hansen, 2005). Other projects adapted the method of introducing a pup to tourist areas (Landry et al., 2005) and studied the interaction between LGDs and hikers (Landry, 2010), in order to be able to give recommendations for decreasing biting incidents. Only few studies were dedicated to test the efficiency of LGDs against predators (Petrucci-Fonseca et al., 2000; Landry & Raydelet, 2010). Rigg (2001) gave an overview of LGDs. Landry and his colleagues created a research-oriented newsletter (Carnivore Damage Prevention News), dedicated to sharing the experiences among the different European projects. Now there are several thousand LGDs working throughout Europe (about 1500 dogs in the Alps alone). Researchers on both continents attempted to quantify the benefits of LGDs. The studies were done by scientists whose motivations were to test the system and report the results in the scientific and agricultural journals. Producers in the USA and Europe were convinced by the results and many now claim that LGDs are an integral and valuable part of their business. In the USA, ranchers selecting pups from good working dogs have quickly developed and increased the numbers of second and third generation animals. Regionally — perhaps inadvertently — landrace dogs have begun to evolve. In certain pastoral sections of the USA one finds 90% of the producers using dogs — where there were none in the 1970s. Significantly, dogs are being born and raised on the farms where they are expected to work. Now in many areas one can purchase a locally-adapted dog for a small cost, or sometimes a friend will give you one. This is fairly standard among many pastoral communities around the world. The introduction of LGDs into South Africa has been quite a different story. Introduction of American and Eurasian LGDs to Africa began in 1994, when the Cheetah Conservation Fund of Namibia imported several Anatolian shepherd dogs and pups and set up a project modeled after the Hampshire College Livestock Dog Project. Since the turn of this century, other conservation agencies have initiated similar projects, including Cheetah Outreach, The Landmark Foundation and the Endangered Wildlife Trust of South Africa, and Cheetah Conservation of Botswana. These organizations actively promote LGDs as a practical and effective way of helping to solve predation conflicts. Until recently, South African conservation groups have tended to focus on one breed, the Anatolian Shepherd. A breakthrough in South Africa’s entry into the LGD world was Johan Gallant’s article in Farmers Weekly (4 June 2004) advocating the use of endemic landrace dogs. He wrote about the native guarding dogs of the Zulus, and about the pastoral peoples in the kingdom of Lesotho who had beautiful working LGDs. In 2004, Gallant and Matthew Berry arranged to import Maluti Mountain dogs into South Africa and test their effectiveness on livestock/wildlife conflicts. There are perhaps 100 being tested. Cheetah Conservation Botswana uses only landrace LGDs and have the most exciting program in the region. They give an annual award for “the best LGD” of the year. LGDs as landrace dogs exist in huge numbers around the world. One estimate, extrapolating from flocks of sheep on migration between Portugal and the Himalayas, counts perhaps 1,600,000 LGDs (Coppinger & Coppinger, 2001). Any dog that has been socialized as puppy and bonded with livestock (sheep, goats, cows, horses, water buffalo, ostrich, etc.) can be considered a LGD. Because they are bonded with the livestock, they are trustworthy with that stock (dogs tend not to show predatory routines to animals they socially bond with) and they are attentive, socializing with animals which they are bonded to. Dogs that are always present with the livestock (distinguishing them from farm or village dogs) provide protection from predators. Predators are cautious and avoid dog(s), or are distracted by them from their predatory routine. The LGD is therefore defined by its function and behavior, not by its morphology. Like household guarding dogs, LGDs behave in a manner appropriate to the socialization they receive as very young pups. Does that mean that any dog properly bonded will do? The first answer is yes, any type or kind of dog can be bonded with livestock. The second answer is — well, not quite. Chihuahuas won’t do. Sick dogs won’t do. The dog has to be appropriate for the environment and the culture it serves. LGDs exist in a variety of geographies, climates and pastoral cultures. Big dogs do better in cold regions while desert LGDS tend to be small. In many places, livestock pastures are shared with hikers and campers and it is important not to have big aggressive dogs that bite strangers. There are many considerations that go into the meaning of an “appropriate” LGD. The best way to acquire an appropriate dog is to get a working dog or pups from good working parents from a nearby farm. At this stage of the use of LGDs in South Africa, the culture of using landrace dogs is not widespread. The reason for introduction programs however is that no local LGDs are available. The agency supporting a demonstration project finds the “appropriate” dogs, but quite often the agency is naïve and falls prey to the mythologies of what a LGD does. There are two major sources of LGDs for these programs: 1) landrace dogs acquired from some distant pastoral community, and 2) purebred LGDs purchased from local or distant breeders. Both have advantages and disadvantages. 1) Landrace LGDs: Most pastoralist societies have LGDs. The size and shape of these dogs are primarily a response to local environmental factors such as climate, with its seasonal extremes, its own parasites, and a culture that includes seasonal nomadic migrations. In the vast majority of pastoral societies, shepherds have little to no control over LGD reproduction. The phenotype of the landrace LGD is the product of natural selection within the specific pastoral culture. Landrace LGDs are a product of post-zygotic selection (Coppinger & Coppinger, 2005). Pups are already on the ground when selection by humans begins. Pastoralists select from among these pups, based on their management style and often for capricious reasons. Females are often culled for a variety of reasons, and litters are often reduced to one or two pups. Surviving pups may be of a favored color, which becomes popular in the region. Any dog that is disruptive to the livestock husbandry is destroyed at any time during its life cycle. A good-working, favored guardian will be cared for, increasing the likelihood of its reproducing. As a result, regional phenotypic traits appear among those LGDs which have appropriate behavior. These LGDs are the survivors of local parasites, local diseases, and local predators. They adapt or evolve to the region and their job. 2) Purebred LGDs: Breeds are the product of breed clubs. A breed club is usually integrated with a national and/or international kennel club. Breeds are typically created by hobbyists capturing a few specimens of some regional landrace dog working in a contained pastoralist community. For example, the Kangal dog was created from a few pairs of dogs imported into England from Turkey in the 1960s (Derbent & Yilmaz, 2008). A few animals are extracted from a regional landrace population and transported to a country that has kennel clubs. Kennel clubs have rules about how a few landrace dogs are transformed into registered breed. Breed clubs usually require a written breed standard — a description of the ideal phenotype. The phenotype of individual animals is compared at dog shows and individuals that have the “ideal” phenotype are rewarded. These individuals become preferred as sires or dams, resulting in what is known as pre-zygotic or artificial selection. The difference between “landrace” and “breeds” is enormous. Landrace dogs exist in large populations randomly breeding over some defined geographical area, e.g., the Pyrenees Mountains, the Anatolian Plateau, Kangal Province, or Turkey, or the Hungarian Pusta, or the Yugoslavian transhumance where the LGDs migrate from Thessaly in Greece to the Sharplanina Mountains of Albania, Macedonia, and Kosovo. A breed is a sexually isolated population not allowed to breed randomly, derived from a few individuals adopted and removed from their place of origin. Landrace dogs adapt over many generations to a particular region and culture. Breeds are small, sexually-isolated inbred, populations of individually cared for, wormed, vaccinated , sheltered animals that are, now, historical representations of the original landrace population. The problem with purebreds placed back into pastoral cultures after years of being pet or show dogs is that, no matter how perfectly a dog meets the “breed standard,” it is not necessarily suitable for the conditions of an adopted livestock setting. Selection by breeders for size, shape or color has little to do with whether it is bonded to livestock and displaying an appropriate behavior for the adopted pastoral culture. Many different kinds of problems arise when setting up demonstration LGD programs . For the Americans the problem was that the nearest landrace dogs were thousands of miles away in Europe and Turkey and breeders charged show dog prices. Why didn’t they use Navajo or Hispanic landrace LGDs? In the early days of the projects “everybody” thought that LGDs were distinct “breeds.” The Navajo dogs looked like mongrels. None of the project leaders came to the conclusion that Navajo dogs were a smallish landrace population perfectly adapted to a semi-arid scrub environment. Everybody thought LGDs were supposed to be big. Nobody realized that five 17kg LGDs were much more effective than one 85kg dog bred for the show ring. In addition the ranching community was angry at having lost their traditional lethal techniques, nor did they want wolves restored, nor did they want some college professor telling them that some Mexican cur was the answer to all predation problems. In the early days of the project, the dog had to be special; it had to be, for example, a Turkish shepherd dog bred for 7000 years to kill wolves. The problems are the same in South Africa where it is easier for a farmer to try a LGD which is an imported Anatolian shepherd dog than to try a regional Kaffir dog (Swarrt, 2003) The big problem with bringing landrace dogs from Europe and Asia Minor to the USA was that dog fanciers made them into a breed. The original few imported transhumance dogs are now a sexually isolated population and prone to all the genetic and inbreeding problems that purebreds have. The Hampshire program crossbred the imported landrace dogs and kept records on these as well as those that were bred to their own breed. The American agricultural community understands the concepts of hybrid vigor so that wasn’t a problem. But it was a problem for the breeders who thought their breed was being corrupted. Most of the dogs sponsored by the Landmark Foundation, the Endangered Wildlife Trust, and Cheetah Outreach in South Africa have been Anatolian shepherds, obtained locally, with a few imported individuals. Anatolians are the only LGD breed which is available locally. They were deemed very successful in the Namibian project. Interestingly, the Namibian stock came from landrace dogs that Coppinger collected in Turkey in the 1980s. He selected dogs that were not too big and that did not show aggressive personalities. Selection was for working LGDs that would be appropriate to the America farm. Also, interestingly, many of the dogs he collected became founding stock for the newly emerging (1996) Anatolian Shepherd Dog Club of America. In gratitude members of that new club donated dogs to the Hampshire College project, which funded the first Anatolians for Namibia. By the time those dogs got to South Africa, the myth that came with them was that they were big, fierce, and specially bred for 7000 years to kill predators. Mythology is fun until it begins to interfere with reality and policy. Many breeders and conservationists exaggerate the assumed inherited qualities of “breed”, and that diverts the attention from the essential puppy development processes. You cannot raise a litter of nine pups in a garage or pen with three sheep and expect to get nine LGDs. It does not matter what breed they are! Mythology also exaggerates the overall image in the mind of the general public of big, ferocious dogs, bred to kill. Those visions have alarmed people in the USA and Europe and motivated policy-makers to discuss whether there should be laws forbidding big, vicious dogs from the range and holding shepherds criminally liable for being reckless. Purchasing purebred LGDs is often prohibitive for both farmers and conservation projects. Paying for a papered purebred, to stand out in some wilderness area protecting sheep, borders on the absurd. What are the chances it has been raised correctly? It is really hard to educate people that success of the system is not in the dog’s genetics, but in its first few months of life. In South Africa, working LGDs are very often neutered. Purchasing neutered animals from breeders, who in some cases are trying to retain control of production and sales, is counter-productive to a developing regional program. The economics are straightforward. The average ranch dog lives 18 months and the average fenced-ranch dog lives 33 months (Lorenz et al., 1986). Only 70% of the surviving “started” dogs become working LGDs (Coppinger et al., 1988). There are 50,000 livestock producers in South Africa, meaning there would have to be 65,000 LGDs on about a three-year rotation. Those of us interested in conservation of wildlife have to ask — where are those dogs going to come from and how much are they going to cost? The point is that there are sources of landrace dogs in and around South Africa. The producers need a supply of a working stock that can reproduce itself. Under those conditions, nature and natural selection will result in new and local landraces. We know that is not what the dog breeding community wants to hear, but it is what the stock producers practicing non-lethal predator control needs to hear. Bibliography Black, H.L., 1981. Navajo sheep and goat guarding dogs: A New World solution to the coyote problem. Rangelands 3: 235–237. Black, H.L., Green, J.S., 1985. Navajo use of mixed-breed dogs for management of predators. Journal of Range Management 38: 11–15. Breitenmoser, U., Angst, C., Landry, J.M., Breitenmoser-Würsten, C., Linnell, J.D.C., Weber, J.M., 2005. Non-lethal techniques for reducing depredation. In: Woodroffe, R., Thirgood, S., Rabinowitz, A. (Eds.), People and Wildlife: Conflict or Coexistence? Cambridge University Press, pp. 49-71. Coppinger, R., Coppinger, L., Langloh, G., Gettler, L., 1988. A decade of use of livestock guarding dogs. In: Crabb, A.C., Marsh, R.E. (Eds.), Proc. 13th Vertebrate Pest Conference, pp. 209-214. Coppinger, R., Coppinger, L., 2001. Dogs: A New Understanding of Canine Origin, Behavior and Evolution. Scribner, N.Y. Coppinger, R., Coppinger, L., 2005. Livestock guarding dogs: from the transhumance to pre-zygotic selection. Carnivore Damage Prevention News 9: 2-9. Derbent, S., Yilmaz, O., 2008. Le Karabash. Chiens des bergers nomandes. Son histoire, son travail, son avenir. Propre éditeur. 202 pages. www.karabash.eu Green, J.S., Woodruff, R.A., 1990. ADC livestock dog program update: a focus on managing dogs. In: Davis, L.R., Marsh, R.E. (Eds.), Proc. 14th Vertebrate Pest Conference, pp. 233-236. Green, J.S., Woodruff, R.A., 1990b. Livestock guarding dogs: protecting sheep from predators. U. S. Department of Agriculture, Agricultural Information Bulletin 588. Hansen, I., 2005. Use of livestock guarding dogs in Norway – a review of the effectiveness of different methods. Carnivore Damage Prevention News 8: 2-7. Landry, J.M., 2010. Chiens de protection et randonneurs. Quoi de Neuf 23: 17-18. (http://www.oncfs.gouv.fr/Quoi-de-neuf-Le-Bulletin-d-information-du-reseau-loup-download130) Landry, J.M., Burri, A., Torriani, D., Angst, C., 2005. Livestock guarding dogs: a new experience for Switzerland. Carnivore Damage Prevention News 9: 40-47. Landry, J.M., Raydelet, P., 2010. Efficacité des chiens de protection contre la prédation du lynx dans le Massif jurassien. Report for the French Ministry of Environment. Pôle Grand Prédateur, Juin, 2010.

Excellent. For lots of reasons. But most of all I love how it's a scientific study unambiguously noting as a matter of fact that purebreeds are fundamentally a garbage concept. And also it does a great job of dispelling the mythology surrounding LGDs. I wish other natural working dog cultures were studied with such professionalism. The whole world of dogs should be up to this standard, but only LGDs are thanks largely to coppinger. The rest is just reduced to a childish "breed encyclopedia", "dogs 101" etc bs level. Purebreeds should just be ignored by everyone except for dog show loonies, and working dogs should be a serious category of scientific research and understanding. I'm no scientist myself so all I can really do is complain. I want scientists to hop to it and start studying the world of working dogs so I as an enthusiast can learn from them. LGDs are a great start and there is some great work being done here and I wish them all the best in establishing reproducing LGD populations in these sheep producing cultures that have lost LGDs somewhere along the way. Hopefully doing away with such ugly evils as poison baits and etc, and ultimately helping human beings live in harmony with wildlife the way we SHOULD be able to (we're just animals after all). Australia should be the next on their agenda but unfortunately no one cares to conserve the feral dogs of australia but me. To me they belong here just like aborigines but I'm fighting a losing battle trying to argue that point. Anyway, great thread and thank you!

The thing is dingoes are just dogs, pure and simple. They were the first wave of dogs, then a new wave of dogs came in. With this new wave of dogs, came many other changes to australia, not least of which a wider variety of prey animals in certain areas. And so the wild dogs of australia, which by now I think should be considered part of the furniture, have changed in some areas from normal dingoes to different dog designs in response to the new prey animals and with help from the genes of domestic dog types which can contribute an aptitude for hunting these animals. What we get from the australian people is this big resistance to all this natural stuff taking place, they're trying to kill off the feral dogs and kill off their prey and ofcourse failing, and if anything just prolonging this awkward destructive transitionary period. A totally natural one really which has happened on most other landmasses at different times in history. Eurasian animals invaded north america, caused a lot of extinctions and destruction but it all sorted itself out and created the north american wilderness we see today. North american animals invaded south america, again caused a lot of destruction and there was an awkward transition with extinctions and etc but it became the magnificent wilderness of south america. The destruction in australia has already been done, it was done when the aborigines and later dingoes came to australia and completely decimated the native flora and fauna. Australia pre-europeans was a decimated biosphere, eerily devoid of large animals, an incomplete eco-system. The animals that have come with the europens noticed this vacancy and ofcourse are running amok, having a field day filling all the vacant niches. And people are scrambling to inhibit this process by any means necessary. Poisoning and trapping and aerial shooting and whatever you can be done to stop the animals from forming a new ecosystem. My view is- let them form a new ecosystem. Let it sort itself out and build up into a diverse and vibrant full eco-system as seen in south america. Dogs are shaping up to be the main predators, and heading in some interesting directions with some (unfortunately shot) feral dogs verified to tip the scales at around 75 kgs. Mind you I'm not saying ferals in australia shouldn't be hunted, by no means, but they should be hunted by natural methods that test the quality of the prey species and naturally spares the top specimens. In other words, hunt them with dogs. Anyway, getting a bit of topic I suppose.