Tuesday, May 31, 2011

Excessive Force Claims Throw Harsh Light on Deficiencies in Training, Deployment, and Supervision of Police Canine Unit

Samuel Campbell and Lisa Parker had a date one night in October 2007. After dinner they went to the Springboro Eagles Club for drinks. At about 12:30 Parker decided to walk home because she lived nearby and was intoxicated, but Campbell stayed until after 1 a.m. He realized he had Parker’s car keys and decided to go to her apartment to give them back. He saw her lying on the couch through the window in the front door, pounded loudly for five to ten minutes, but could not rouse her. He went to the back door and pounded there, but again with no result. Another tenant, hearing the pounding, called the Springboro police department, which dispatched Officers Clark and Anderkin, who arrived at Parker’s residence at 2:26 a.m.

Campbell began to head home through Parker’s back yard when he heard the approaching sirens. He lay down on the ground near an outbuilding of an adjoining property in an attempt to avoid a confrontation with the police. The police also saw Parker sleeping in her front room but were also unable to wake her. There was some damage on the doors and a third officer, Larry Bush, called the landlord to request a key to Parker’s residence because the officers believed a potential intruder might still be in the area. Clark and Anderkin decided to use a police dog, Spike, to track the intruder, believing that they were dealing with an attempted burglary.

Spike was a Belgian Malinois trained in narcotics detection and patrol work, and Clark was his handler. Clark deployed Spike near the side of the house. Spike pulled Clark towards the back porch and then led Clark through Parker’s back yard. Anderkin followed to provide cover. Spike led the officers to an adjoining yard where there was a fence leading up to the outbuilding near which Campbell was lying on the ground.

Spike tried twice to go over the fence, and Campbell testified that he was trying to determine how to get over the fence when Spike found Campbell lying on the ground and bit him. Campbell, to the contrary, testified that he had made eye contact with Clark some moments before and that Clark was leading the dog directly to him before the bite. Both parties agree that Officer Clark issued no warning and said nothing to Campbell before the dog bit him. Spike bit Campbell on the left leg and continued to bite him for 30 to 45 seconds, a long enough time to indicate either noncompliance on the part of Campbell or an inability on the part of Clark to control the dog.

Clark testified that he shouted at Campbell multiple times to put his hands up but Campbell did not comply and that Campbell kicked the dog in the head repeatedly. Only after Campbell raised both hands did he call off the dog. Once Campbell stood up he was handcuffed. He was taken by ambulance to a local hospital where his wounds were cleaned and stitched. The wounds later became infected and it took five or six months for him to heal completely.

That was one incident described in a civil rights suit against the city of Springboro, Ohio. There was another.

One evening in October 2008, a year after the first incident, Chelsie Gemperline, an 18 year old, was arrested for underage drinking. Gemperline and a friend, Sara Osborne, had invited people to a party at the home of neighbors who were out of town for whom Osborne was house-sitting. At about 1:30 a.m., Clark was dispatched to a loud party. While trying to assess what was going on, some of the partygoers left the house. Clark determined he was dealing with underage drinkers and called for backup. Several units responded.

Clark was uncomfortable with deploying Spike to find someone who may have fled because he feared the dog might bite the individual. Gemperline was found hiding in a closet on the second floor. She was questioned about her brother, who had left, but refused to help the officers locate him. Gemperline was handcuffed and became belligerent. She admitted, when being deposed, that she had said “fuck you, bitches” to them during the scuffle.

Once placed in a police car, Gemperline slid her right hand out of the handcuffs, lowered the window of the car, climbed out and ran down the street. She hid in a plastic playhouse in the back yard of a house six or seven houses away from where the party had been. Sergeant Aaron Zimmaro told Clark to get Spike and look for Gemperline. Zimmaro was primarily concerned with Gemperline’s safety. Clark put Spike on a 20-foot lead and began looking for Gemperline. Clark regarded the situation as a search for a fugitive, not a missing person, as indicated by the following description of his deposition:

“According to Officer Clark, there are two different types of tracks, a 'missing persons' track, which is done to locate a non-criminal missing person, and a 'tactical' or 'fugitive' track, which is done to locate a person who has committed a criminal offense…. Spike was not trained to respond differently based on the type of track…. However, Officer Clark used different equipment for a missing persons track than he did for a fugitive track. When doing a fugitive track, Spike wore a harness and a twenty-foot lead…. During a missing persons track, Spike wore a much shorter four-foot leash, which gave Officer Clark more control over Spike's movement and decreased the likelihood of Spike biting the subject of the search….”

Both Sergeant Zimmaro and Clark testified that they did not intend for Spike to bite Gemperline, but this might be negated as to Clark by a statement he made on the scene:

“Officer Clark also made the following comments, captured in part by his vehicle recording device: ‘Jeez Louise ... [unintelligible] this bitch, ... I've had it,’ and ‘... she's gonna get a nice rude awakening here in one second or two, ... it's not gonna feel very good.’ … Officer Clark denied making the first statement, but admitted to the second statement.”

Clark believed Spike was smelling something on the deck of a house down the street, but when they didn’t find anything he thought the dog might be “goofing off.” On heading back to his car, the dog went head-first at the window of the child’s playhouse where Gemperline was hiding. Spike reached his head far enough through the window that he nipped Gemperline’s chin and bit her right thigh. She tried to pry him off her leg but he kept biting. She continued to struggle until she either passed out or went into shock. Rather than giving a verbal release command, Clark choked off the dog, forcing it to release its bite. He believed it would be more difficult to give a verbal release command if Gemperline was hitting or actively resisting Spike.

A dog should be able to be recalled even if the dog is still struggling and it seems likely that Clark would have been taught this initially. The website of Lynnwoods Kennels, where Clark received some of his initial training with Spike, and which sold Spike to Springboro, contains the following statement (scroll down to 6) :

“Some schools of thought believe that the dog should release when commanded even if the suspect is not submitting and is fighting with the dog. This is a judgment call, but I agree. We must teach the dog that when I say it is over it is just that, over! There are times that the suspect is in so much fear of the dog that he or she is in survival while the dog is on them and they will not or can not stop fighting. If the suspect seems to be in this state we have to get the dog off him, and I sure do not want to stand in the middle of this action to do it. If the suspect does not settle down after the dog gets off we can always allow the dog to re-engage.”

Gemperline did not recall hearing Clark say anything to her. Clark testified that he did not shout any warnings when he entered the back yard but later said that he had told her to stay still and put her hands up as he performed the choke-off. After Clark pulled Spike off Gemperline, another officer flipped over the play house, putting Gemperline on her stomach, when he handcuffed her. The cuffs were removed when the extent of Gemperline’s injuries were seen. The deputy carried Gemperline to the street and called for an ambulance.

Gemperline spent a few days at a hospital and then a week and a half at a medical center receiving additional treatment. She was unable to walk for a time and was given crutches that she relied upon for six months. She has a permanent deformity in her right thigh.

The day after the Gemperline incident, the Springboro police department suspended its canine program, and Spike was retired. Officer Clark was terminated in 2009.

Before looking at the legal issues in the suit, it is necessary to describe the background of Spike and Clark.

Springboro is “a somewhat affluent city with a rather low crime rate,” according to the federal district court for the Southern District of Ohio. A canine unit was established in 2005, consisting of Officer Clark and Spike. Clark and Spike completed a 300-hour canine handling course in May 2005. The state of Ohio requires that police canine units be certified by the Ohio Peace Officer Training Commission. Clark and Spike received two certificates from the Commission, one for criminal apprehension, canine control, and canine searches, and the second for tracking, article search, and the detection of marijuana, cocaine, heroin, methamphetamines, and their derivatives. Both certificates were good for two years. On the same day they received their initial state certification, Clark and Spike also received accreditation from the North American Police Work Dog Association (NAPWDA) for narcotics detection and other skills, referred to as “Utility Phases,” including obedience, article search, area search, tracking, building search, and aggression control.

Clark believed that he and Spike were supposed to complete eight hours of maintenance training every other week. Clark and Spike attended three training programs in 2006 and 2007. One was a 5-day NAPWDA convention, while another was a 40-hour training program provided by the Miami Valley Police Canine Association. Also, there was a 40-hour training workshop at Lynnwoods Kennels, the facility from which Spike was originally obtained. They did not receive maintenance training on a regular basis. Clark testified that his supervisors did not allot sufficient time for Spike’s training. Clark complained about this, but how much was not clear. Spike’s certifications lapsed in the summer of 2007, but he was recertified in September 2007 by the Ohio Training Commission. The team also received a third certificate of accreditation from NAPWDA that month.

As a canine unit policy, the Springboro police department used, apparently rather informally, the Law Enforcement Canine Model Policy of the International Association of Chiefs of Police (IACP). Clark had obtained the policy from the website of an expert witness on police canine matters, but Clark had no contact with the IACP. No one was specifically assigned the responsibility of supervising the canine unit to ensure that Spike was suitable for duty. Every time Spike was deployed, Clark had to complete a canine deployment form, but it appeared to the court that Clark “was essentially supervising himself” as to his canine responsibilities. None of Clark’s supervisors showed virtually any familiarity with training and certification requirements or understood what IACP policies might have required.

The district court found that Spike had been trained to use the bark-and-hold approach, rather than the bite-and-hold approach to suspect apprehension, but despite this Spike bit 13 people during his police career. His biting during tracking assignments seemed to increase over time. In 2005, he went on ten tracks in which he bit no one. In 2006, he completed 28 tracks, apprehending 14 suspects, five of whom he bit. In 2007, he apprehended six individuals and bit five of them.

The suit was based on claims under 42 U.S.C. 1983, alleging excessive force, failure to train and supervise the canine unit, as well as Ohio law claims for assault and battery and malicious prosecution (the latter only as to Campbell). The court then analyzed whether qualified immunity applied, which required that it determine whether the force used to subdue the plaintiffs “was reasonable from the perspective of a prudent officer on the scene.” This determination is to be made on the facts and circumstances of the situations involved. The court found that several factors in the incidents involving Campbell and Gemperline indicated that the force used was excessive and objectively unreasonable under the circumstances. As to Campbell, the court noted:

“Officer Clark did testify that he was concerned that Campbell may still be in the area because the officers knew that he had only recently left. But there is no evidence that he believed Campbell was lying in wait to possibly ambush them. Instead, the fact that the officers did not express any concern for their own safety while they were investigating the exterior of Parker's house suggests the opposite.” Three factors were deemed of particular significance: (1) Spike was not undergoing proper maintenance training and may not have been performing as expected in the field, (2) no warnings or orders were given to Campbell, which may have been a violation of proper procedures, and (3) the extended period of time during which Spike continued to bite Campbell (up to 45 seconds) was extreme.

As to Gemperline, the lack of a consistent training regimen was also deemed significant, as was the failure to shout out a warning and the length of time that Spike was able to bite Gemperline. One witness for the plaintiffs provided an affidavit saying that the real reason that Clark used the choke-off may have been that Spike did not always respond to Clark’s verbal commands as he should have, which makes this part of the training issue. The court summarized:

“Essentially, when the facts are viewed in a light most favorable to Gemperline, the evidence in this case demonstrates the following: Defendant Clark deployed a police dog known for biting subjects during tracks to locate a 110 pound intoxicated eighteen year old who had evaded arrest for underage consumption of alcohol, despite knowing that the dog had not regularly undergone required maintenance training and was not always responding properly to verbal commands. Furthermore, when the dog began air scenting, Officer Clark failed to issue any verbal warning to give Gemperline an opportunity to surrender, although he either knew or should have known that she was nearby and possibly hiding in a playhouse in the area near where the dog air scented. Finally, when the dog engaged Gemperline, Officer Clark chose to physically pull Spike off of Gemperline rather than issuing a verbal command, thereby unnecessarily extending the duration and increasing the force of the bite. Under that scenario, a reasonable jury could find that the amount of force used in apprehending Gemperline was excessive.”

The court summarized the excessive force case law as follows:

“Sixth Circuit decisions demonstrate that there is a continuum of permissible versus impermissible use when it comes to police dogs. On the permissible end of that spectrum are cases wherein officers deploy properly trained police dogs to locate individuals who were believed to be involved in nefarious criminal activity, who may have been armed and dangerous, and who failed to surrender or respond in any manner after officers gave several warnings…. On the other end of that spectrum lies a case in which a canine officer allowed a little-trained police dog to get close enough to a subject of a track to bite the subject despite the fact that the subject had already been subdued and placed in handcuffs.”

The court then clearly states its opinion of the law as applied to the facts before it:

“Based on the precedent discussed above, Officer Clark had reason to know that the conduct he engaged in while apprehending and arresting Plaintiffs was unlawful. Officer Clark deployed Spike with knowledge that Spike had not undergone necessary maintenance training on a regular basis and had a high propensity for biting subjects during tracks. Further, in both cases, Officer Clark never issued any verbal warnings or commands though there is evidence showing that he knew or should have known that the plaintiffs were hiding close by and in fact knew or should have known the precise location in which each was hiding. Finally, in both cases, there are questions of fact as to whether Officer Clark allowed Spike to bite Plaintiffs for an unreasonable and unnecessary amount of time, particularly where neither Plaintiff showed signs of active resistance other than attempting to prevent Spike from further biting them. Under the circumstances, Officer Clark should have known that his conduct violated the clearly established rights of the Plaintiffs to be free from the unreasonable and improper deployment of a police dog in the course of their apprehensions and arrests.”

If I were defense counsel and read these statements in an order on pre-trial motions, I might well say to myself: If I can’t get a jury to think different, I’m screwed. Time to send out settlement feelers.

The court denied Officer Clark’s motion for summary judgment as to the excessive force claims. As to the city’s liability, the court noted that a “systematic failure to train or supervise police officers adequately can amount to deliberate indifference on behalf of a city towards its inhabitants.” On this issue the court extended its disapproval beyond Clark:

“Chief Kruithoff never assigned anyone with specific knowledge or training regarding the operation of the SPD's canine unit to supervise the unit. Instead, when it came to his duties as a canine handler, Officer Clark was essentially left to monitor himself. He was under the general supervision of Lieutenants Wheeler and Parker and a number of rotating sergeants. However, none of those supervisors received adequate training in the operation of a canine unit such that they would be capable of effectively supervising the unit. Accordingly, although they allegedly were in charge of monitoring the canine unit's certification and training, they expressed a lack of knowledge as to what type of certification and training was required. Perhaps as a result of their lack of knowledge, Officer Clark's supervisors allowed him and Spike to continue working for several months after their state certification had lapsed and failed to take any formal action to address Officer Clark's repeated complaints that he was unable to find time to perform necessary maintenance training with Spike.”

With such an across-the-board failure, the court could not “at this point conclude as a matter of law that the City of Springboro could not be held liable for violations of Plaintiffs’ constitutional rights.”

“[T]here is evidence in this case that in establishing the SPD's canine unit, Chief Kruithoff took few if any steps to ensure that the unit functioned in accordance with the law. He chose to essentially abdicate any duty he may have had to set policies governing the operation of the unit and to provide training for the officers who were supposedly charged with supervising the unit. Plaintiffs set forth sufficient evidence from which a jury could conclude that the supervision and training in this case were so lacking that the resultant violations of Plaintiffs' Fourth Amendment rights was almost a foregone conclusion.”

The nail goes in deeper:

“[T]he Court finds that a reasonable jury could conclude that Officer Clark acted with a malicious purpose, or in a wanton or reckless manner.”

The case is a warning to law enforcement canine handlers to keep up their training regimens, and renew certifications in a timely manner. Clark may have had to take the fall, but there is plenty of blame to go around. A handler must have the support of supervisors, and that support should not be blind. The case is also a warning to police chiefs, who must realize that implementing a canine unit requires accepting responsibility for the operation of the unit, including the willingness and ability to devote the necessary funds, time, equipment, training, management, and the adoption of appropriate policies and protocols. The community has a right to expect that the police department will be accountable to the community for the safe operation of such a unit.

Courts have not easily found excessive force or stripped away qualified immunity, and the benefit of the doubt has often been given to handlers, particularly when they can show good faith in deploying trained and certified police dogs, even when the deployment has led to serious injury and sometimes death. Nevertheless, there is a line, and Springboro, Ohio, and members of its police department appear to have crossed it here.

Campbell v. The City of Springboro, Ohio, 2011 WL 1575525 (April 26, 2011).

This blog was written by John Ensminger and L.E. Papet. For a broad discussion of excessive force suits regarding police dogs, see Police and Military Dogs, Chapter 20: Suspect Apprehension and Bite Issues (forthcoming Taylor & Francis/CRC Press).

Tuesday, May 24, 2011

Small Dogs Appeared Early in the History of Canine Domestication, but Did This Happen in the Middle East?

Research published in 2007 indicated that small dogs share a variant of the IGF1 gene on dog chromosome 15. Variants in the IGF1 (insulin-like growth factor 1) gene are “fixed in the majority of small distantly related dog breeds which suggests that small size evolved early in the history of domestication.” A few giant breeds (mastiffs, bullmastiffs, and rottweilers) share the “small” allele, but no wild canid so far tested has been found to have this variant.

Surveying a great many grey wolves and dogs, members of the same research team (Melissa Gray, Nathan Sutter, Elaine Ostrander, and Robert Wayne) have concluded in research published in 2010 that “grey wolves of Middle East origin were slightly closer to domestic dogs” as to the specific size-relevant allele.

Included in the wolves with haplotypes closest to small dogs are Spanish wolves. The authors conclude that their analysis was consistent with “a Middle Eastern origin for the small domestic dog haplotypes.” The authors elaborate:

“Many of the domestic dogs from Middle Eastern sites are small whereas those from Belgium, Germany and Western Russia are larger in size, which supports our hypothesis that small body size evolved early in the history of domestic dogs and probably in the Middle East…. Therefore, in concordance with past archeological studies, our molecular analysis provides strong evidence for the early evolution of small size in dogs in the Middle East, more than 12,000 years ago.”

The researchers observe that Akitas “were positioned between grey wolves and the main cluster of domestic dogs,” a finding providing some fuel for an alternative interpretation of the data discussed in the next section of this blog.

The researchers distinguish their results from those obtained by Savolainen et al. in 2002, which argued for an East Asian locus of canine domestication. The researchers note that Savolainen’s team looked at mitochondrial DNA, whereas the current study looked at nuclear DNA. They acknowledge that there could be a number of explanations for the difference in the results from the two approaches, but they argue that their findings are more consistent with the archeological record. “Small size could have been more desirable in more densely packed agrarian societies where dogs may have lived partly indoors or in confined outdoor spaces.”

A Contrary View. Two researchers in Savolainen’s camp responded quickly. In a comment published in the same journal as the original paper, these critics, Klutsch and Crapon de Caprona, note that only two wolves from China were sampled in the study by Gray et al. “Obviously, this is not enough to evaluate if Chinese wolves have a higher or lower genetic diversity from Middle Eastern wolves and they could by pure change be unrepresentative.”

The critics find the evidence from the Gray paper concerning Akitas important, and note:

“The clearest extractable result is that Akitas (as an ancient East Asian dog breed) are most closely related to wolves. The Akita included in this study exists in two sizes. The American Akita is derived from the ancient Japanese Akita, but has acquired a larger size, and is thus now considered to be another breed by the Japanese Kennel Club within the Fédération Cynologique Internationale. Since both breeds exist in North America, it would have been important to study the Akita samples according to the size of the dogs sampled… [I]t would have been important to know if it is the smaller Japanese breed (clearly bred for hundreds of years outside the Middle East) that is closest to the Middle Eastern wolves but also to other Asian wolf samples. The most parsimonious conclusion here would have been that the Japanese Akita dog is related to Asian wolves and Middle Eastern wolves represent a fraction of Asian wolves.”

The critics also argue that DNA studies must take into account the effects of hybridization between dogs and wolves, which has been argued by Pang et al. as occurring in the Middle East. The critics conclude that the data presented “do not lend strong support to the conclusion of Gray et al. that the SDH [small dog haplotype] is derived from Middle Eastern wolves in ancient times.”

It does not appear that the criticism of the research on small dog genetics disputes that this uniform characteristic of small dogs (found oddly in a few big dogs) began early in the period of domestication. The issue in dispute is where the small dog phenomenon began.

Response to the Response. Gray and Wayne responded to the criticisms of Klutsch and Crapon de Caprona. They argue that “if South-East Asia or any other region were the origin for the small haplotype we would expect at least one of these haplotypes to cluster with the small dog haplotype, and we find none.” The authors cite the paper by vonHoldt et al., also finding a Middle Eastern origin for domestication. Eight authors of the original 2007 or the 2010 small gene paper were also co-authors on the multi-authored (37) paper published by Nature online in March 2010.

Dogs as Creators of Wealth. Two scientists, one from Oxford and one from the Laboratory of Genomic Diversity in Maryland, joined the fray by noting that “there are almost as many different imputed locations for the dog’s origins as there are studies…. Moreover, wolves are highly mobile, resulting in scant geographic patterning in their genetics and, worse, dogs can interbreed with regional native wolf populations.” Finding that the research of Gray et al. may have avoided some pitfalls in prior genetic research, however, they accept that the Small Dog Haplotype probably descends from Middle Eastern wolves, provided “Middle Eastern wolves today most closely resemble Middle Eastern wolves of yesteryear.”

These two scientists, Driscoll and Macdonald, describe the Natufians, whose burials are the earliest firm evidence of canine domestication, as hunter-gathers who lived in permanent or semi-permanent settlements, and argue that this culture reflects a “narrow transition between Paleolithic nomadic hunter-gathering and the comparatively advanced settled agro-economics of the Neolothic.” This, they say, may be “the moment in which wolves crossed the canine Rubicon.”

Citing the work of the Coppingers, Driscoll and Macdonald say that the “initial association of wolves and humans was doubtless instigated by the wolves.” They speculate that “proto-domestic wolves were resident scavengers at the rubbish dumps of permanent settlements rather than nomadic camp followers,” and that this led to a separate lifestyle and reproductive isolation from larger wolves.

“To speculate yet more wildly, perhaps intraguild competition then led to small garbage-wolves that became vigilant barkers at the approach of larger and hostile wolves and, in so doing, further divided the population genetically. This series of developments would set up the assortative mating theoretically required for the sympatric divergence of the wolf population.”

The Natufians, argue Driscoll and Macdonald, “having originally selected runts on a whim, thereafter not only perpetuated their whimsy—selecting for cuddliness—but also began selecting for pint-sized functionality (such as ratting and entering burrows).” Since “diminution” would be maladaptive in the wild, it was probably selected for under the influence of people. The Natufians must have “had some loose cultural concept of tolerance for dogs, if not of caring and ownership of them.” Dogs may have even been the first “living capital,” a kind of wealth that could be accumulated, leading to economic hierarchies.

A flowchart in the paper presents an evolutionary timeline for two stages of shrinking: first when wolves at the dumps became smaller; second by artificial selection after some level of domestication, “resulting in the ancestor of the small breeds found today worldwide.” The flowchart presented here has been adapted for blog space and to emphasize the timeline of this interesting idea. Dogs, under the argument—and I think there is something to it—are a critical underpinning of certain basic premises to civilization—sedentary agriculture, wealth, private property, social hierarchies, even capitalism.

Waiting for Dogot. Klutsch and Savolainen have recently argued that future studies should look at Y-chromosome evidence, as well as ancient DNA samples of both wolves and dogs, as these “may give new insights into early domestication history and the dog’s migration routes.” Canids are an extremely mobile family. A recent paper finding, among other things, that Ethiopian wolves are closely related to coyotes, is curious in part because the ranges of the two groups (eastern Africa and the Americas) nowhere overlap.

The dispute between the Middle Easterners and the Far Easterners on the locus of domestication, and now on the locus of the origin of smallness in dogs, is the best theater in canine research I know. I left science long ago after the 19th expedition of the Te Vega and this sort of warfare is the part of scientific research I miss most. I wish both camps well, and hope they will continue to develop evidence and propose theories that will entertain us for years to come. The conflict itself advances our knowledge.

  1. Coppinger, R. and L. Coppinger (2001). Dogs: A Startling New Understanding of Canine Origin, Behavior, and Evolution (Scribner New York).
  2. Driscoll, C.A., and D.W. Macdonald (2010). Top Dogs: Wolf Domestication and Wealth. Journal of Biology, 9(10).
  3. Gray, M., N.B. Sutter, E. A. Ostrander, and R.K. Wayne (2010). The IGF1 Small Dog Haplotype Is Derived From Middle Eastern Grey Wolves. BMC Biology, 8, 16.
  4. Klutsch, C.F.C., and M.D. Crapon de Caprona (2010). The IGF1 Small Dog Haplotype Is Derived from Middle Eastern Grey Wolves: A Closer Look at Statistics, Sampling, and the Alleged Middle Eastern Origin of Small Dogs. BMC Biology, 8, 119. 
  5. Pang., J.F., C. Kluetsch, X.J. Zou, A. Zhang, L.Y. Luo, Luo, H. Angleby, A. Ardalan, C. Ekström, A. Sköllermo, J. Lundeberg, S. Matsumura, T. Leitner, Y.P. Zhang, and P. Savolainen (2009). mtDNA Data Indicate a Single Origin for Dogs South of the Yangtze River, Less Than 16,300 Years Ago, from Numerous Wolves. Biological Evolution, 26, 2849-2864.
  6. Rimbault, Maud, Holly C. Beale, Jeffrey J. Schoenebeck, Barbara C. Hoopes, Jeremy J. Allen, Paul Kilroy-Glynn, Robert K. Wayne, Nathan B. Sutter, and Elaine A. Ostrander.(2013). Derived Variants at Six Genes Explain Nearly Half of Size Reduction in Dog Breeds. Genome Research, 23, 1985-1995. This article was published two years after this blog was written but is important to read in connection with the subsequent history of this issue.
  7. Savolainen, Peter, Y.P. Shang, L. Luo, J. Lundeberg, and T. Leitner (2002). Genetic Evidence for an East Asian Origin of Domestic Dogs. Science, 298, 1610-1613.
  8. Sutter, N.B., C.D. Bustamante, K. Chase, M.M. Gray, K.Y. Zhao-Li, Zhu , B. Padhukasahasram, E. Karlins, S. Davis, P.G. Jones, P. Quignon, G.S. Johnson, H.G. Parker, N. Fretwell, D.S. Mosher, D.F. Lawler, E. Satyarij, M. Nordborg, K.G. Lark, R.K. Wayne, and E.A. Ostrander (2007). A Single IGF1 Allele Is Major Determinant of Small Size in Dogs. Science, 316, 112-115.

Sunday, May 15, 2011

The Killers of Actaeon

According to the most common version of the myth, Actaeon, son of Aristaeus and Autonoë, was hunting when he surprised Artemis bathing in a wood. Angered, she turned him into a stag and he was torn apart by his own hunting dogs. After their master's death, the dogs searched for him in vain, howling in their grief, until the centaur Chiron made a lifelike image of Actaeon to soothe them.

There are many variations of the story, but the most common depiction in art, particularly in vase painting, was of the dogs attacking Actaeon, the goddess watching and sometimes contributing to his death with her bow and arrow. The savage and sacrificial nature of the attack is emphasized in the first picture, a vase from Badisches Landensmueum (Karlsuhe, Germany, posted by Aaron J. Atsma on Theoi Greek Mythology, dated between 400 and 350 BC). The dogs are greyhound-like, but probably Laconians. A statue excavated at Herculaneum shows dogs attacking a stag, probably Actaeon fully transformed, with equal savagery.

P.M.C. Forbes Irving, in a study of metamorphosis in Greek myths, describes the essence of the story as "a hunting nightmare in which the hunter identifies with his victim." Nevertheless, he says that hunting, in such stories, has "a primarily imaginative and metaphorical significance," since hunting "had for a very long time played a fairly marginal role in Greek life." Actaeon's crime, if there was one, may have been rape, and there may be a revenge element in the actions of the goddess.

The deer, according to Forbes Irving, has metaphorical significance. "If the hunter was the supreme example of masculinity and daring, the deer is the traditional symbol of cowardice and weakness; it is with this psychological change that Ovid completes his description of the transformation: ‘and last of all she planted fear within his heart.’"

Walter Burkert, the great German scholar of Greek religion, describes the myth as having "a particularly ancient quality." The behavior of the dogs "goes beyond anything that could be observed in nature" since "real dogs cannot be comforted by an image." Burkert says the search of the dogs is a human ritual where the search for a torn-up victims ends in a symbolic restoration. Actaeon's "death is a sacrificial ritual of the hunt, consecrated by the Mistress of the Beasts and performed in the form that had been standard since Paleolithic times. The actors are dogs struck mad by 'wolf’s frenzy,' werewolves whose shrine is in a mountain cave."

Vase paintings appear from the 6th century BC. In earlier pictures, Actaeon sometimes wears a deerskin, but by the middle of the 5th century he is often shown with antlers. The second vase, attributed to the Choephoroi Painter (Harvard 1960.37, dated 350-340BC), shows Actaeon just as the attackers are first making contact with their victim. The dogs are not easily identified with any ancient breed, but the one flying through the air towards Actaeon's sword (enlarged in the third plate) may be a Molossian, unafraid to attack even when about to be met with a deadly weapon. I had thought at first that the artist might be thinking more of a battle scene than a hunt since a sword seemed unlikely for a hunter. D.B. Hull, however, in his book on hounds and hunting, says that hunters sometimes carried the xiphos, which he describes as a "double-edged straight sword with a blade that usually swelled in width near the middle so that it was leaf-shaped," exactly what we see here. The sword was for protection in the event of an unexpected attack by a lion or wild boar.

Actaeon is usually shown dying without friends, but in the Louvre vase below (Louvre CA 3482, dated 460-440 BC), a comrade is also threatened by the mad dogs but may be escaping. The dog underneath Actaeon as he falls is likely a Laconian, while the one above him could be a Molossian. The Greeks often went on hunts with various types of dogs bred and trained for separate functions in the hunt, which may be something the painter, known as the Painter of Woolly Satires, was told about rather than having seen since the dogs are in positions they would hold while eating or standing, not running or attacking. The wolf-like color of the dog on top, which almost looks like a drawing of a lying dog that has been cut out and pasted in an inappropriate place, bothers me. Molossians were sometimes thought to have lion blood, not wolf blood, and are not shown with unusual coloration on any other vase I am aware of. Wolves were known to be able to cross with dogs (an evolutionary connection would not have been understood) and a cross might have been seen as a vicious animal (as is believed by many now--just consider the laws on wolf and coyote hybrids). Burkert's speculation about the dogs really being werewolves might find almost Jungian support here, but we must always beware making unconscious projections on the screen of antiquity, as E.R. Dodds long ago warned. A passage is Saara Lilja's study, Dogs in Greek Poetry, may help:

"Rabies may have been the real factor behind the well-known myth about Actaeon who, turned into a stag, was chased and mangled by his own hounds. But it should be remembered as well that the dog in old myths and in the Homeric epics may have sometimes been identified with the wolf. O. Keller's suggestion (Thiere des classischen Altertums 168) that λύσσα is etymologically akin to λύκος would suit a rabid dog's wolf-like behaviour."

There was a ritual reflecting the part of the myth concerning the dogs' visit to the cave of Chiron, who built the statue of their master to calm them. It took place on Mount Pelion in Thessaly, at the time of the rising of Sirius, the dog star, at the hottest point of summer. Young men of the upper classes of the area each sacrificed a sheep or a ram, then each put on the skin of the victim and climbed the mountain to the cave of Chiron. Burkert says this ritual corresponds to the journey of the dogs to the mysterious mountain cave. No dogs or dog skins appear to be involved in the ritual, however, and the association with Actaeon may be a later layer to the initially local tradition surrounding the ritual.

Burkert, W. (1972). Homo Necans: The Anthropology of Ancient Greek Sacrificial Ritual and Myth, 112-113. University of California Press, Berkeley (1983 translation of 1972 German edition)
Dodds, E.R.( 1951). The Greeks and the Irrational, 148 (discussing Orphism). University of California Press, Berkeley.
Hull, D.B. (1964). Hounds and Hunting in Ancient Greece. University of Chicago Press, Chicago.
Irving, P.M.C.F. (1990). Metamorphosis in Greek Myths. Oxford University Press, Oxford.
Lilja, S. (1976). Dogs in Ancient Greek Poetry. Commentationes Humanarum Litterarum 36. Societas Scientiarum Fennica. Helsinki.
Oxford Classical Dictionary (1996). Entry for Actaeon by Herbert Jennings Rose and Jennifer R. March. Oxford University Press, 3d edition.

Thanks to Richard Hawkins of Fern Hill Scottish Deerhounds for suggesting this blog topic.

Saturday, May 14, 2011

Wolves May Lose Federal Protection in 29 States

The U.S. Fish and Wildlife Service has proposed to substantially reduce North American wolf protection by declaring all or parts of 29 eastern states as not being in the historical range of the gray wolf, thereby removing those areas from the application of the Endangered Species Act as to gray wolves.

The 29 states (or parts of states) where gray wolves will no longer protected, if the regulatory change is made final, are Maine, Massachusetts, Connecticut, New Hampshire, Rhode Island, Vermont, New York, New Jersey, Pennsylvania, Delaware, Maryland, Virginia, North Carolina, South Carolina, Georgia, Florida, West Virginia, Kentucky, Tennessee, Alabama, Arkansas, Mississippi, Louisiana, and parts of Ohio, Texas, Oklahoma, Missouri, Indiana, and Illinois.

The Fish and Wildlife Service is not saying that there are no wolves in those 29 states, or that they are no longer endangered, but is rather determining that the wolves that belong in those areas are not, and never were, gray wolves (except where gray wolf populations have spread beyond their historical range). The wolves in those areas are now to be denominated eastern wolves or eastern timber wolves.

Rather than automatically allowing eastern wolves to pick up the protection previously given to the mistakenly denominated gray wolves in the areas, the agency has decided to spend some time studying the issue before giving any protection to eastern wolves. While it must be hoped that study will lead to action, it is by no means certain that the federal government really has the well-being of wolves, gray, eastern, or otherwise, as its primary goal. There are indications that the Fish and Wildlife Service has been compromising protection of wolves in an effort to placate politicians who would delist gray wolves almost everywhere (but certainly in their own districts).

This rather facile way of disenfranchising so many wolf populations is made possible through the Fish and Wildlife Service's acceptance of the conclusion of some scientists that Canis lupus lycaon, the eastern wolf, should no longer be a subspecies of the gray wolf, but should now be treated as a separate species, Canis lycaon. The two categories interbreed, resulting of mixed populations that include hybrids. Though not all scientists agree, there is genomic evidence for the distinction the Fish and Wildlife Service is now making. See Wilson et al.(2003). The Service says that it will evaluate this species for possible protection under the Endangered Species Act “in the near future.”

States and tribes may also protect wolves, but the mood in most states is not to protect wolves but to hunt them, though some tribes are much more charitable to our ancient shadow. The wolves might well ask why we don’t stay in our historical range and stop pushing them out of theirs. Road density has been found to be inversely correlated with the density of wolf populations. See Mech et al. (1988). If roads become dense enough, wolves disappear. Human population growth is a great enemy of wildlife outside of zoos.

There is a long history to this, the early stages of which the Fish and Wildlife Service summarizes as follows:

“European settlers in the Midwest attempted to eliminate the wolf entirely in earlier times, and the U.S. Congress passed a wolf bounty that covered the Northwest Territories in 1817. Bounties on wolves subsequently became the norm for States across the species’ range. In Michigan, an 1838 wolf bounty became the ninth law passed by the First Michigan Legislature; this bounty remained in place until 1960. A Wisconsin bounty was instituted in 1865 and was repealed about the time wolves were extirpated from the State in 1957. Minnesota maintained a wolf bounty until 1965.”

The plate above, from the early 15th century book on hunting by Gaston Phoebus, shows a way of trapping wolves used in the middle ages. Variations of such horrors are used even now.

In 1967, the Department of the Interior listed as endangered the eastern timber wolf (Canis lupus lycaon) in the Great Lakes region (32 Fed. Reg. 4001, March 11, 1967), followed by listing the subspecies in the northern Rocky Mountains (Canis lupus irremotus) in 1973 (38 Fed. Reg. 14678, June 4, 1973). In 1976, the Mexican wolf (Canis lupus baileyi) was listed (41 Fed. Reg. 1740, April 28, 1976), as was the Texas gray wolf (Canis lupus monstrabilis) (41 Fed. Reg. 24064, June 14, 1976). In 1978, the Department realized that the entire species was endangered throughout the lower 48 states (43 Fed. Reg. 9607, March 9, 1978), except for Minnesota.

The eastern wolf, now sometimes recognized as a separate species (Canis lycaon), occupied the northeastern United States, while the red wolf (Canis rufus) was found in the southeastern United States. The extent of the range of the gray wolf, and the overlap of ranges of wolf species and subspecies, has been debated for at least 65 years, going back to the early studies of Young and Goldman. Genomic studies have placed more precision on range boundaries, and have indicated that the gray wolf did not occur in the eastern United States.

To be fair, the wolf populations in some states, such as Minnesota, Wisconsin, and Michigan have grown under the Endangered Species Act. From 1,000 to 1,200 wolves in Minnesota in 1976, there are now around 3,000. Wisconsin has gone from a negligible population of about 34 in 1990 to nearly 700, and Michigan has gone from 10 in 1990 to nearly 600. The animals are also more widely distributed than they were a few decades ago, but the populations of Wisconsin and Michigan remain small.

Illegally shooting and trapping wolves continues, but the Fish and Wildlife Service notes that because such killings “generally occur in remote locations and the evidence is easily concealed, we lack reliable estimates of annual rates of intentional killings.” In the Agassiz National Wildlife Refuge in Minnesota, there were around 20 wolves before managed and illegal shootings reduced the number to six in 1999. This has grown to between 10 and 12 in recent years. In Wisconsin, the percentage of wolf mortality attributable to illegal killing varies considerably from year to year. In 2006, illegal killing accounted for 67% of all mortality, only 19% in 2007, 23% in 2008, 62% in 2009, and 38% in 2010.

The Fish and Wildlife Service has gathered federal, state, and tribal representatives to review overall wolf strategy with a view to (in the Service’s own words):

1. Promote and sustain wolf recovery.
2. Comply with requirements of the Endangered Species Act (16 U.S.C. 1531 et seq.).
3. Minimize the regulatory burden on the states, tribes, and the general public.
4. Facilitate state and tribal management of wolves.
5. Minimize wolf-human conflicts.
6. Promote public acceptance of wolf listing and recovery actions.

The first of these policy objectives is good for the wolves but, one way or another, the remaining ones amount to bureaucratic fudging, which no wolf should trust. An exception involves tribal management. The Service describes protective efforts and policies of tribes, nations, and regional subgroups, including the Ojibwe (Chippewa), Odawa, Ho-Chunk, Menominee, the Keweenaw Bay Community (a band of the Chippewa), and others.

Just because the Fish and Wildlife Service wants to cut loose the eastern wolf (even if only temporarily), this does not mean that it will succeed. Certain groups, such as Defenders of Wildlife, the National Wildlife Federation, and the Humane Society of the United States, have taken on the Department of Interior’s attempt to delist wolves before, and have sometimes won. In Humane Society of the U.S. v. Kempthorne, for instance, the federal district court of the District of Columbia asked how the Service could delist a population within a larger listed group. On this questioning of its authority, the Service became rather defensive and got the Solicitor General of the Department of the Interior to issue a memo in support its authority to revise lists to “reflect recent determinations.” The Service cites this memo as the basis for other actions it has taken towards delisting and will undoubtedly continue to cite it if a lawsuit is brought on its delisting of a large part of the wolf population of the U.S.

Sources: Department of the Interior, Fish and Wildlife Service, Endangered and Threatened Wildlife and Plants; Proposed Rule to Revise the List of Endangered and Threatened Wildlife for the Gray Wolf (Canis lupus) in the Eastern United States, Initiation of Status Reviews for the Gray Wolf and for the Eastern Wolf (Canis lycaon), 76 Fed. Reg. 26086 (May 5, 2011)(revising 50 CFR 17.11, the present version of which (p. 25 of the pdf) lists the gray wolf's habitat as the coterminous (lower 48) states with certain exceptions); Young, S.PL. and Goldman, E.A., Wolves of North America (Dover Publications 1944); Wilson, P.J., Grewal, S., McFadden, T., Chambers, R.C., and White, B.N. (2003). Mitochondrial DNA Extracted from Eastern North American Wolves Killed in the 1800s Is Not of Gray Wolf Origin. Canadian Journal of Zoology, 81(5), 936-940; Humane Society of the United States v. Kempthorne, 579 F.Supp.2d 7 (D.D.C. 2008) ; See Mech et al. (1988) Wolf Distribution and Road Density in Minnesota, Wildlife Society Bulletin, 16, 85-87; Defenders of Wildlife v. Secretary of the Interior, 354 F.Supp.2d 1156 (D. Or. 2005); National Wildlife Federation v. Norton, 386 F.Supp.2d 553 (D.C. Vt. 2005); Humane Society v. Kempthorne, 481 F.Supp.2d 53 (D. D.C. 2006); Defenders of Wildlife v. Tuggle, 607 F.Supp.2d 1095 (D. Az. 2009) (Mexican wolf); Defenders of Wildlife v. Salazar, 729 F.Supp.2d 1207 (D. Mt. 2010); New Mexico Cattle Growers Ass’n v. U.S. Fish & Wildlife Service and Defenders of Wildlife, 1999 WL 34797509 (D. N.M. 1999).

One website that continues to follow wolf issues closely is Wolfwatcher.com.

Monday, May 9, 2011

Egyptian Jackal Reclassified as the African Wolf by Genetics Researchers

The golden jackal, Canis aureus, classified by Linnaeus in 1758, is found throughout north and east Africa, the Middle East, southeastern Europe, and parts of Asia. In 1833, Hemprich and Ehrenberg identified the Egyptian jackal, Canis aureus lupaster, as a subspecies of golden jackal, though Thomas Huxley noted as early as 1880 that the Egyptian jackal seemed suspiciously similar to the Indian wolf. A graduate student at Leeds, Magda Nassef, had investigated one gene segment in 2003 and concluded that the Egyptian jackal was more similar to a grey wolf than to any jackel. A larger study now confirms this.

The grey wolf's range extends to the Sinai Peninsula but not to mainland Africa, though there is an Ethiopian wolf (Canis simensis) whose range overlaps that of golden jackals found in the Ethiopian highlands. A team from the Universities of Oslo, Oxford, and Addis Ababa, determined that the Ethiopian highland jackals are in fact Egyptian jackals. More importantly, the team concluded that the Canis aureus lupaster, whether found in Egypt or Ethiopia, is a type of grey wolf. The paper's phylogenetic tree diagrams show the Ethiopian wolf as most closely related to the coyote (Canis latrans).

The researchers state that the Egyptian jackal “likely represents an ancient wolf lineage that colonized Africa prior to the radiation of the Holarctic wolf and as such should be reclassified as the African wolf.” This colonization probably happened during the Pleistocene (a broad time range from 2.6 million years ago to 12,100 years ago). They state that Canis aureus lupaster should no longer be considered a subspecies of jackal, and in fact represents the only grey wolf to inhabit the African continent.

The authors further shake up current phylogenetic classifications by concluding that the golden jackal is not in a monophyletic group with the two other named jackal species, the side-striped jackal (Canis adustus) and the black-backed jackal (Canis mesomelas), meaning that there is “reason to question whether the colloquial name jackal has any taxonomic integrity.”

The map reproduced here is taken from the paper posted by this research team, and shows the distribution of the golden jackal in dark grey. The range of the Indian wolf, Canis lupus pallipes, to which the now described African wolf is closely related, is shown in the center of the map, while the range of the Himalayan wolf, Canis lupus chanco, another relative, extends beyond the right edge. The black squares represent Canis lupaster animals that were sampled for DNA, while the stars represent Canis aureus animals sampled. The numbers show the number of animals sampled at each site.

In Ethiopia, jackals are, according to the authors, “systematically persecuted because of their threat to livestock.” The Egyptian jackal is also not protected.

Sources: Rueness, E.K., Asmyhr, M.G., Sillero-Zuibiri, C., Macdonald, D.W., Bekele, A., Atickem, A., and Stenseth, N.C. The Cryptic African Wolf: Canis aureus lupaster Is Not a Golden Jackal and Is Not Endemic to Egypt. PLoS ONE, 6(1), e16385 (January 2011); Nassef, M. (2003). The Ecology and Evolution of the Golden Jackel (Canis aureus), Master’s thesis, University of Leeds.

Sunday, May 1, 2011

Anthropomorphism in Antiquity

Consider the following account:

"I myself reared a hound with the greyest of grey eyes, and she was fast and a hard worker and spirited and agile, so that when she was young she once dealt with four hares in a day. And apart from that she is most gentle (I still had her when I was writing this) and most fond of humans, and never previously did any other dog long to be with me and my fellow huntsman Megillus as she does. For since she was retired from the chase, she never leaves us, or at least one of us. If I am indoors she stays with me, and accompanies me if I go out anywhere; she escorts me to the gymnasium, and sits by while I am exercising, and goes in front as I return, frequently turning round as if to check that I have not left the road somewhere; when she sees I am there she smiles and goes on again in front. But if I go off to some public business, she stays with my friend, and behaves in the same way to him. If one of us is ill, she does not leave him. If she sees us even after a short period of time, she jumps up in the air gently, as if welcoming him, and she gives a bark with the welcome, showing her affection. When she is with one of us at dinner she touches him with her paws alternately, reminding him that she too should be given some of the food. And indeed she makes many different noises, more than any other dog that I think I have seen; and she shows audibly what she wants. And because when she was being trained as a puppy she was punished with a whip, if anyone even to this day should mention a whip, she goes up to the one who has said it and crouches down like one beseeching, and fits her mouth to his mouth as if she is kissing, and jumps up and hangs from his neck, and does not let him go until the angry one gives up the threat. And so I think that I should not hesitate to write down the name of this dog, for it to survive her even in the future, viz. that Xenophon the Athenian had a dog called Hormé, very fast and very clever and quite out of this world."

This passage was written in the second century AD in a book on hunting with hounds, Cynegeticus, by Arrian of Nicomedia, who refers to himself as Xenophon (from either part of his name or a nickname taken from his idol, Xenophon son of Gryllus, who lived six centuries earlier). In his book on Arrian, Philip Stadter describes this passage as revealing more about Arrian’s life than any other passage in his works.

"We are permitted a brief glimpse into his life at Athens with his friend Megillus and his daily routine visiting the gymnasium or doing public business. More evident, however, is his feeling for his hound, his pride in its hunting skill, its little tricks and habits. The passage implies a reciprocity of friendship between man and dog, seen in the walk from the gymnasium or the awareness of the dog at dinner. Such faithfulness and devotion prompt Arrian finally to immortalize his dog by including its name in his work."

Arrian’s writing is remarkably modern, as is often true of works from the Silver Age of Roman literature. (Just read Lucian of Samosata’s account of traveling to the moon in A True Story.) The description of a favorite dog could be written today, though the use of a whip (μάστιξ) would now be considered barbaric. Phillips, in his commentary on Arrian, notes that the primary function of the whip may have been as a sound to get a dog's attention, but the fact that Hormé supplicated anyone who mentioned it suggests that some aversive conditioning was involved.

The phrase “dealt with” in the first sentence, ανταρκέω (perfect, αντήρκεσεν), is translated as “hold out against” or “persist against” in the Liddell & Scott Greek-English Lexicon, and presumably means that the dog could either catch four hares a day or be close to the hare when it was caught by another dog or driven into a net. In other words, it could keep up with a very fast animal and was certainly a gazehound (sighthound). The dog was probably what Arrian referred to in an earlier passage as a Celtic hound, vertragus in the Latin spelling (ούετραγος in Greek). D.B. Hull, in his book on ancient Greek hounds and hunting, states that the word reappears in Old French as veltre, meaning greyhound.

A word that is made perhaps somewhat too modern in the translation of the passage is the adjective that M.M. Willcock renders as “out of this world.” The adjective, ίερωτάτη, a superlative of ίερός, could be translated as supernatural, but also as divine, sacred, mystical, under divine protection. It seems to me that Arrian finds something of the divine in his dog’s powers, and in their ability to understand each other.

Arrian had a distinguished career, including holding the office of proconsul of Hispania Baetica in southwest Spain, around 126 AD, where Richard Hawkins suggests he may have first encountered the vertragus. (See the statue from the Rijksmuseum on the Fern Hill website.) A genome study published in Nature in 2010 found two distinct groups of sighthounds. The larger group includes greyhounds, Italian greyhounds, whippets, Irish wolfhounds, Scottish deerhounds, and Borzois, while two breeds fit within the ancient breeds, the Saluki and the Afghan hound. It would appear that the Celtic running hound may have been a progenitor of one or more of the breeds in the first group.

The first picture above of a dog in a supplicating posture was taken by Barbara McManus, Professor of Classics emerita at the College of New Rochelle, showing the lower portion of a statue of Diana with a hunting dog in the Vatican Museum (provided courtesy of the VRoma Project). The dog is probably a Laconian, but could be a vertragus. I've changed my opinion on this since I first posted this blog because Richard Hawkins, who has a better eye for this than I do, is inclined to think the dog a Laconian. Also, though I haven't found anything on the dating or provenance of the statue, it's condition suggests a Roman imperial date. Statues of that vintage are often copies of Greek originals, and if that applies here, the statue was probably of Artemis with a Laconian since Celtic running hounds were not known in the Attic period. Of course, an argument could be made that a Roman copyist might modernize the representation of a dog to the most popular contemporary breed.

I fully understand the desire to immortalize a favorite dog, as Arrian did for Hormé (΄Ορμη, Greek for dasher, impulse, onrush). When I suggested that a picture of my dog appear on the cover of Service and Therapy Dogs in American Society, the art director of my publisher discouraged the idea and offered several other alternatives. Only when my wife found a picture of Chloe that captured much of the purpose of the book was the idea accepted. (Don’t tell me I was wrong if you want to keep me in a good mood.)

The second picture is from a funeral stele found in Boeotia, executed by Alxenor of Naxos, who left his name on an inscription. Percy Gardner describes the monument as follows:

"This delightful monument represents a worthy Greek citizen in one of his lighter moods.  Standing in a position of ease, he rests his weight on a staff which supports his shoulder, and holds out in sport a grasshopper to a favourite dog, who leaps up in an attitude somewhat constrained, and clearly resulting from the narrow limits of the monument."

Grasshoppers were apparently treats in antiquity. The third plate is also from Gardner's book and shows a funeral stele of a man holding a wine cup and a pomegranate. Professor Adolf Fürtwangler of the University of Berlin had argued that the horse symbolized Hades and the dog Hecate. 

Gardner believed the horse and dog were "really a survival of an ancient custom, whereof we find traces in the graves of Greece and Italy, by which the horse and dog of a deceased warrior were slain and buried in the same place with him. Whether their bones were mingled with their master's, or whether they are merely figured on his gravestone, the meaning is much the same, that wherever the lord is, there are his faithful attendants: 'Admitted to that equal sky, his faithful dog shall bear him company,' as Pope says.  In any case, horse and dog on a tomb are certainly a mark of knightly rank."

The quote from Pope is actually about North American Indians, who had a similar burial practice. The supplicating posture of the dog indicates the closeness of the relationship that the deceased had with it, or at least that the sculpture of the tomb wanted it to appear the two had. 

A wonderful understanding of affection between dogs is found in a marble statue of two hounds now in the Vatican Museum, the fourth picture here, possibly from the 2d century AD. A similar pair is displayed in the British Museum. These statues were found near Civita Lavinia (modern Lanuvio south of Rome, a few miles below Lake Nemi), not far from where the second century Roman emperor, Antoninus Pius (138-161), had a palace, though a connection to the emperor or his estate is not assured. The dogs are most likely Celtic running hounds as well, but could arguably be Laconians, another sleek hunting dog from antiquity, which takes its name from the region of Laconia around Sparta.

Sources: A.A. Phillips & M.M. Willcock, Xenophon and Arrian on Hunting with Hounds (Aris & Phillips, Ltd. 1999); Philip A. Stadter, Arrian of Nicomedia (University of North Carolina Press 1980); Richard Hawkins, An Jansen & Waidman, Arrianus, De Lange Jacht en Lurecoursing (Eburon 2003); D.B. Hull, Hounds and Hunting in Ancient Greece (U. Chicago Press, 1964); B. vonHoldt et al., Genome-wide SNP and haplotype analyses reveal a rich history underlying dog domestication. Nature 464, doi:10.1038 (April 2010) (See Figure 1.); W. Dansey, Arrian on Coursing (J. Bohn, London 1831); P. Gardner (1896). Sculptured Tombs of Hellas.  Macmillan & Co., London. For a small dog in a Pompeian fresco, see the blog of July 19, 2010. Many interesting images of Roman dogs can be found by trolling through the VRoma Image Archive.

Thanks to Richard Hawkins who told me that Arrian's Cynegeticus may be the best dog book ever written. I'm beginning to agree. Thanks to James Grout for permission to use the picture from the SPQR website.