Monday, June 8, 2015

New Definition of "Service Dog" May Get SDs for PTSD into VA Buildings, but Old Defnition May Keep Them off VA Transport Vehicles

Regulations proposed in the Federal Register on May 27 would restrict transport of service dogs to those that have been funded, or at best could have been funded, by the Department of Veterans Affairs.  Such dogs must have been trained by a member organization of Assistance Dogs International or the International Guide Dog Federation, so would not include service dogs trained by other organizations or by the veterans themselves. (Dogs trained by organizations that are candidates for membership in ADI or IGDF would also not satisfy the definition, as stated by the VA at 77 Fed. Reg. 54372, September 5, 2012.) Such dogs would have to be used for visual, hearing, or substantial mobility impairments, so would not include service dogs for PTSD and other psychological conditions.

The problem with this proposal is that once the service dog access regulations proposed by the VA in the Federal Register on November 21, 2014, are made final, as may happen soon, veterans will be able to enter VA facilities with service dogs for PTSD, but unless the transport rules are revised to conform, they will not be able to bring their dogs on vehicles that will take them to the appointments.  While it might be expected that logic will not let this happen, the administrative paralysis that has afflicted operation of the VA in recent years could mean that logic will have little to say about the matter.  We can pray, but we can also write letters, and I submitted the following comment letter concerning the regulatory confusion to the Department of Veterans Affairs.

The letter has now been posted on the website. Thanks to Veronica and Brad Morris of Psychiatric Service Dog Partners for reading an earlier draft of this letter. 

June 4, 2015

William F. Russo, Director
Regulation Policy and Management
Department of Veterans Affairs
810 Vermont Ave. NW, Room 1068
Washington, DC 20420

Re: RIN 2900-AO92-Veterans Transportation Service

Dear Director Russo:

This comment is submitted regarding two provisions in the proposed rules concerning the Veterans Transportation Service as published in the Federal Register on May 27, 2015, 80 Fed. Reg. 30190.  Both provisions refer to service dogs.  I am an attorney in private practice, licensed in the State of New York. I have written a book, Service and Therapy Dogs in American Society (CC Thomas 2010), which describes service animal rules in various contexts, and also maintain a blog that regularly covers legal issues regarding dogs where I have discussed issuances of various agencies, including those of the Department of Veterans Affairs.

Proposed 38 CFR 70.71(b) states that “[r]egardless of a veteran’s eligibility for beneficiary travel, VA may provide VTS [Veterans Transportation Service] to veterans enrolled in VA’s health care system who need transportation authorized under § 70.72 [Types of transportation] for: … (2) Retrieval of, adjustment of, or training concerning medications, prosthetic appliances, or a service dog (as defined in 38 CFR 17.148)…” (emphasis added). The other reference to a service dog in the proposed rules occurs at 38 CFR 70.73(a), which provides that in requesting VTS, the requester “must provide the facility director or designee with information necessary to arrange these services, including … any special needs that must be accommodated to allow for transportation (e.g., wheelchair, oxygen tank, service or guide dog)…” (emphasis added). The second provision does not separately reference a definition for “service or guide dog” so it must be presumed that the definition would there also be taken from 38 CFR 17.148.[1]

The preamble to the proposal, at 80 Fed. Reg. 30192-3, elaborates:

Enrolled veterans would be eligible under paragraph (b) [Enrolled veterans] if they are traveling for a scheduled visit or urgent care; for retrieval, adjustment, or training concerning medications or prosthetic appliances; to acquire and become adjusted to a service dog provided pursuant to 38 CFR 17.148; for an unscheduled visit; or to participate and attend other events or functions for the purposes of examination, treatment, or care. (emphasis added)

This paragraph is apparently intended to indicate that if a veteran needs VTS in order to acquire or become adjusted to a service dog, such a dog would have to be one that satisfies the requirements of 38 CFR 17.148, which has a narrower definition of service dog than that contained in proposed 38 CFR 1.218(a), 79 Fed. Reg. 69379, November 21, 2014. If providing transportation services to acquire a service dog is regarded as one of the expenses that the VA will undertake to cover under the general service dog funding regulation, 38 CFR 17.148, then there is some logic to such a restriction.[2]  The phrase “becoming adjusted to a service dog” would seem to be somewhat less easily categorized as an activity that should be restricted to a service dog for which the VA would provide funding.  Although it is not clear how often veterans are provided VTS solely for the purpose of becoming adjusted to traveling with service dogs, all veterans with all service dogs, including service dogs under the broad definition of proposed 38 CFR 1.218(a), must be able to habituate their dogs to transportation vehicles.

The preamble also states that when an eligible person is contacting a VA facility regarding an examination, treatment, or care at the facility, a request may be made for transportation services.  The request may be made to a Mobility Manager in “many cases,” or to someone designated by the facility director.  The request is to contain necessary information for the transportation of the veteran, which can include “special needs that must be accommodated to allow for transportation (e.g., wheelchair, oxygen tank, service or guide dog) and other relevant information.” 

This clearly indicates that transportation services can be provided to any eligible veteran who is coming to a VA facility for an examination, treatment, or care, and who may need to be accompanied by a service dog.  Such a dog should not be presumed to be restricted to a dog eligible for VA funding under 38 CFR 17.148, but rather to any service dog that can obtain access to VA property under proposed 38 CFR 1.218(a), a much broader definition and one largely if not completely consistent with regulations regarding service animals issued by the Department of Justice (28 CFR 36.104, etc.).  If service animals that could be brought with a veteran on a VTS transport vehicle were restricted to the definition provided in 38 CFR 17.148, the anomalous situation would inevitably arise, upon finalization of 38 CFR 1.218(a), that a veteran could be entitled to bring a service dog for PTSD into a VA facility but would not be able to obtain transportation services in order to get the dog to the entrance of the facility. 

It might be argued that proposed regulations on transportation services should not be required to take into account proposed regulations on another issue, and this argument may as a policy matter have merit.  I note, however, that at a May 19 meeting of the VA Advisory Committee on Prosthetics and Special-Disabilities Programs, slides presented by Joyce Edmondson, Co-Chair of the VHA Animals in Health Care Committee, indicate that proposed 38 CFR 1.218(a) “is currently in the concurrence process with expedited review being completed at request of the Secretary.”  It would, therefore, seem appropriate to anticipate the finalization of those regulations with language to the effect that transportation services should be provided to veterans with service dogs as defined in regulations relevant to the access of animals to VA facilities.

Distinction between Service Dogs under 38 CFR 17.148 and Proposed 38 CFR 1.218(a)

Under 38 CFR 17.148, service dogs are defined as “guide or service dogs prescribed for a disabled veteran….”  Clinical requirements must be met, including that the “veteran is diagnosed as having a visual, hearing, or substantial mobility impairment….”  Even if a veteran is diagnosed as having such a condition, the regulation provides that “[i]f other means (such as technological devices or rehabilitative therapy) will provide the same level of independence, then VA will not authorize benefits under this section.”  I have been informally advised by one group working with veterans that this sentence has on occasion been used to justify medications, often substantial amounts of medications, as a preferable alternative to the use of a service dog. 

In addition to the one sentence definition of service dogs, 38 CFR 17.148(c) defines “[r]ecognized service dogs … for the purpose of paying benefits…” as having to satisfy the following requirement:

The dog and veteran must have successfully completed a training program offered by an organization accredited by Assistance Dogs International or the International Guide Dog Federation, or both (for dogs that perform both service- and guide-dog assistance). The veteran must provide to VA a certificate showing successful completion issued by the accredited organization that provided such program.[3]

Thus, the veteran is to have a document that indicates completion of a training program with one of two organizations, or both.  Such a document might become a means by which employees of transport services would be able to verify that a service dog satisfies the requirements of 38 CFR 17.148 and thereby preclude access to vehicles by service dogs without such documentation.  Indeed, to accept dogs that did not qualify under 38 CFR 17.148 might arguably be a violation of the mandates of the proposed revisions to 38 CFR Part 70.

In contrast, proposed 38 CFR 1.218(a)(11)(viii) defines a “service animal” as follows:

A service animal means any dog that is individually trained to do work and perform tasks for the benefit of an individual with a disability, including a physical, sensory, psychiatric, intellectual, or other mental disability. Other species of animals, whether wild or domestic, trained or untrained, are not service animals for the purposes of this definition. The work or tasks performed by a service animal must be directly related to the individual’s disability. The crime deterrent effects of an animal’s presence and the provision of emotional support, well-being, comfort, or companionship do not constitute work or tasks for the purposes of this definition. Service dogs in training are not considered service animals. This definition applies regardless of whether VA is providing benefits to support a service dog under § 17.148 of this chapter. (emphasis added)

There is no specific organizational connection required for a dog to be a service animal under this provision, and there is no reason that a dog could not be trained by a veteran himself.  Indeed, there are programs under which veterans are presently being taught to train dogs, including service dogs.[4]  Also, there is no requirement that the service animal’s function be solely related to a “visual, hearing, or substantial mobility impairment,” as specified in 38 CFR 17.148, and functions related to a “sensory, psychiatric, intellectual, or other mental disability” are specifically allowed for service animals under the proposed provision.  If there were any doubt, the final sentence specifically distinguishes this section from the funding provision. 

Suggested Modifications to 38 CFR Part 70 Proposal

It is, of course, possible that proposed 38 CFR 1.218(a) will be finalized before the proposal under discussion here is finalized.  In that event I respectfully suggest that the parenthetical cross-reference in 38 CFR 70.71(b)(2) to 38 CFR 17.148 be changed to 38 CFR 1.218(a).  If finalization of 38 CFR 1.218(a) does not occur prior to the finalization of the current proposal, I suggest that the parenthetical reference in 38 CFR 70.71(b)(2) be altered to read:

(as defined in 38 CFR 17.148 or in such regulations as apply to the access of animals to VA property) 

Although I do not believe an exception on transportation services should be made based on the VA’s service dog funding policy, if it were deemed necessary to separate general transportation services from transportation specifically to acquire a service dog, the cross-reference to a broader definition could be restricted to proposed 38 CFR 70.73(a), where a sentence could be added after the sentence with the parenthetical reference to “service or guide dog” stating: “For purposes of this provision, a service dog is one defined under such regulations as apply to the access of animals to VA property.”

If an alteration to the effect of one of these suggestions is not accepted, the preamble to the final regulation under 38 CFR Part 70 should advise veterans as to the reason for a policy decision that transportation services will only be provided to service dogs satisfying the restricted requirements of 38 CFR 17.148. 

Please contact me with any questions regarding this comment.  I can be reached at or at 917-613-4960.

Respectfully submitted,

John J. Ensminger 

[1] No provision other than §17.148 in Title 38 of the Code of Federal Regulations defines or even refers to service dogs.  Present 38 CFR 1.218(a)(11) refers to seeing-eye dogs, a provision that would be significantly expanded under the proposed service and therapy dog access regulations referred to at length in this letter.  38 CFR 17.37(h)(i) refers to certain kinds of care which, under special circumstances, a veteran may receive, mentioning “seeing-eye or guide dogs.”  38 CFR 18.444 refers to recipients of funds to provide educational programs for veterans and states that such recipients cannot impose certain kinds of rules on “handicapped students,” an example of which would be prohibiting students from bringing guide dogs into campus buildings. 
[2] Under 38 CFR 70.1, Part 70 “provides a mechanism” under which the VHA is “to make payments for travel expenses incurred in the United States to help veterans and other persons obtain care or services from VHA.”  Although this makes 38 CFR Part 70 itself something of a funding regulation, there is no logical reason to subsume its funding limits under the funding provisions of 38 CFR 17.148.  Obviously, somebody with crutches or a walker that needed transportation, but who did not obtain such items with VA assistance or approval, would not be told to leave their prosthetics at home because they were not or could not be obtained with VA funding. 
[3] A second provision regarding “recognized service dogs” concerns dogs obtained prior to the issuance of the regulation. 
[4] See, e.g., Rick A. Yount, Meg D. Olmert, and Mary R. Lee (2012), “Service Dog Training Program for Treatment of Posttraumatic Stress in Service Members,” pp. 63-69 in The United States Army Medical Department Journal: Canine-Assisted Therapy in Military Medicine (April-June 2012), describing a service-dog training program as “a safe, effective, nonpharmaceutical intervention to treat the symptoms of posttraumatic stress disorder (PTSD) and traumatic brain injury in Veterans and service members undergoing treatment at a large Veterans Administration residential treatment facility.” (available for download at

Monday, June 1, 2015

VA Secretary Puts Service Dog Rules Revision on Fast Track

Slides presented at the May 19 meeting of the VA Advisory Committee on Prosthetics and Special-Disabilities Programs indicate that the proposed access regulations on service and therapy dogs, described here last November, have been given an expedited review status by Robert A. McDonald, Secretary of Veterans Affairs. 

Although no date for issuance of the final rules has been announced, presumably the Secretary’s designation will mean that this should occur in the next few months.  One of the slides in the presentation, reproduced to the left, indicates there will be training programs for police service employees of the VA since, often working at the entrances to VA facilities, “enforcement falls largely to them.”  Other affected staff have already participated in conference calls regarding the proposed rules according the last bullet point in the slide.

Leased Facilities Subject to ADA Service Dog Rules 

Another slide from the May 19 presentation, reproduced below, makes the interesting observation that leased facilities used by the VA, which may include outpatient clinics and Vet Centers, "may be covered under the Americans with Disabilities Act for access depending on how lease agreement and what access law applies."  A note under the main bullet point adds: "Care should be taken when negotiating these agreements."  This suggests to me that there has been a problem somewhere, perhaps in a facility where the VA does not have absolute control over the space, but shares it with some other entity or service provider.  

If the lease agreement did not provide that VA policy on service dogs should apply to the facility, it could mean that patients or employees with dogs defined as service dogs by the Department of Justice were bringing in psychiatric service dogs, so veterans began demanding the same privilege, or merely started bringing PTSD dogs with them to appointments or to work. (Many veterans are both VA employees and recipients of medical services at VA facilities.) Someone at the VA who had negotiated such a lease agreement is probably in hot water, though the flame should be lowered when the access rules are finalized.   

Transport Vehicles to Handle Service Dogs

Also, on May 27, the Department of Veterans Affairs issued proposed rules regarding transportation of eligible persons to or from a VA facility for examination, treatment  or care.  The proposal states that, for veterans enrolled in the VA’s health care system who need transportation, adjustments are to be made if the veteran requires transportation with a service dog (proposed 38 CFR 70.71(b)(2)).  In arranging transportation services, the veteran or other individual making the request on behalf of the veteran is to indicate “any special needs that must be accommodated to allow for transportation (e.g., wheelchair, oxygen tank, service or guide dog), and other relevant information.”  Proposed 38 CFR 70.73(a), 80 Fed. Reg. 30190, May 27, 2015.

A service dog, for the purpose of such an accommodation in transportation, is defined under the VA's service dog funding provision, 38 CFR 17.148, so will not include psychiatric service dogs used, for instance, by veterans with PTSD.  The cross reference for the definition of "service dog" should change to 38 CFR 1.218(a), the proposed access regulation, once that is made final. Otherwise, veterans with PTSD dogs, and other service dogs for which the VA does not provide funding, may be able to bring the dogs into a VA facility after the access rule is finalized, but will not be able to take the dogs with them onto VA transport vehicles in order to get to a facility. Ideally, of course, the definition of "service dog" in the funding rule should be changed as well.


One of the slides used at the Advisory Committee meeting is about "pseudo-service dogs," and provides that employees can ask users if a dog is a service dog and what tasks the dog performs for the veteran.  It states that handlers who invite others to pet or play with their dogs probably do not have service dogs.  Also any unruly, unclean, un-house-broken, or generally misbehaving dog is deemed likely to be a pseudo-service dog. It can be expected that such generalizations will be part of the training that security officers and others will be receiving in the future.

The fact that the phrase "concurrence process" is used to describe the completion of the regulatory review (first slide above, second bullet point) may indicate that there have been disputes regarding the proposals within and between some sections of the VA, and that the individuals responsible for finalizing the rules are trying to get some level of cooperation across the agency. If nothing will be released until there is "concurrence," the optimism I expressed in the second paragraph above for a quick finalization may be misplaced. I hope I shall be proven wrong as to this speculation.    

Thanks to Larry N. Long, who works with the Advisory Committee, for providing a copy of the slide presentation of Joyce Edmondson, VHA Animals in Health Care Committee Co-Chair. 

Saturday, May 2, 2015

VA Advisory Committee to Discuss Service Dogs on May 19

On May 19 and 20, the Federal Advisory Committee on Prosthetics and Special-Disabilities Programs will hold meetings at the VA Central Office, 810 Vermont Ave. NW, Washington, DC 20420.  Meetings will begin at 8:30 a.m. and adjourn at 4:30 p.m. on May 19 and at 12 noon on May 20.  The meeting is open to the public but members of the public will not be given time to comment. 

On May 19, the Committee will receive briefings on, among other things, service dogs.  It is not indicated whether this will include any update on the status of the VA's proposed revisions to its service animal policies, discussed here in a blog last November. Under 38 U.S.C. 543, the Committee is to advise the Secretary of the Department of Veterans Affairs on prosthetics and special-disabilities programs administered by the VA, including providing advice on the adequacy of funding for special-disabilities programs. The Charter of the Committee, which is posted on the VA's website, indicates that the Committee's advice may concern research on programs that deal with "spinal cord injury, blindness or vision impairment, loss of or loss of use of extremities, deafness or hearing impairment, or other serious incapacities in terms of daily life functions."  The latter phrase would cover mental health disabilities, including PTSD, so any discussion of service dogs should not be restricted to service dogs for physical disabilities. 

Any member of the public wishing to attend should contact Larry N. Long, Designated Federal Officer, at 202-461-7354.  He can also be emailed at Photo ID will have to be presented at the Guard's Desk of the building. 

Veterans and others concerned with the VA's service animal policies--at least those living in the DC area--should consider attending.  Although they presumably won't be allowed to speak, if Mr. Long receives enough calls from veterans asking about what service dog issues will be discussed, the Committee might allocate more time to those issues.

The VA's announcement is included in the Federal Register for May 4. Department of Veterans Affairs: Advisory Committee on Prosthetics and Special-Disabilities Programs; Notice of Meeting, 80 Fed. Reg. 25362 (May 4, 2015)

Monday, April 6, 2015

Children of Anubis

The following article was published in two parts in the January/February and March/April 2015 issues of PetsNews.  I had wanted to update the original blog, posted in May 2012, so Eytan Hendel's request to publish a Hebrew version gave me an opportunity to add new material that had come to my attention in the intervening years. 

In Oklahoma City there is a memorial to those who died on April 19, 1995.  On the south side of the park are the remains of the entrance plaza to the Alfred P. Murrah Federal Building destroyed by the explosion.  In the center of the park is a long rectangular pool, beside which is a lawn with 168 glass and bronze sculptured chairs representing the people who died there that day.             

Figure 1. Anubis embalming Sennedjem
Visiting the city in April 2012, I took Chloe for a walk to the memorial as dawn was breaking one morning.  We had been told at the hotel that there was always a ranger on duty, though we did not see one. We saw no one once we entered the area.  We walked beside the pool and read some names, found the five chairs of those who had died outside the building.  We looked at the walls, the square arches that give the area the feel of a temple, an ancient temple, almost Egyptian, and I felt tears coming for people I had never known, felt the weight of the loss that was not mine, but was.  It was near the entrance arch that I felt something else, something cold. It was as if the earth had begun to shift from the spirits beneath, as if the chairs would speak.  I was suddenly afraid, yet ashamed of this senseless fear. 

It was then that Chloe began running, thrashing in desperate, frantic circles as some fear gripped her entire being.  A bee sting, the shadow of a bird overhead?  Perhaps, but I did not think of such possibilities until I began to tell others about what happened and needed to assure them of my sanity.  I knew then, though it makes no sense now—the irrational is always lost to any real description after the brief moment of its existence—that Chloe felt the same air I had, the same wisp of death, the same imminence of something not to be encountered above the earth, as if she had seen Anubis himself standing at the end of the pool, arms aloft, summoning a modern acolyte to her ancient duties. 

Graveyards and Battlefields

Almost all cultures—ancient, modern, eastern, western, Native American—make associations of dogs and other canids with death.  As observed by Roscher (1896), the presence of howling dogs where corpses are placed is particularly frightful in the night, as if the animals are powers from the underworld who can pull us across the boundary between life and death.  As translated by Kershaw, Roscher says that “wolf and dog have the same significance precisely in the cult and myth of the underworld demons but are otherwise strictly distinguished from one another.”

That this is the one realm where myth, and perhaps our unconscious, collective or otherwise, does not distinguish the domesticated from the wild canid is an important observation.  Kershaw elaborates:

“The connections of Canis with Death are manifold.  Dogs, or certain behaviors of dogs, presage death.  Dogs, especially large black dogs, are incarnations of death-spirits, or death-demons appear as dogs or are dog-headed; the same demons, e.g. Hekate, can appear as wolves.  Dogs are psychopompoi.  Fearsome dogs, such as Garmr and Kerberos, guard the realms of death.  All this is true of Indo-Europa, but identical conceptions are found worldwide.” 

Figure 2. Two human-fox burials of the Epipaleolithic Period
The association of the Canidae with death in “the cult and myth of underworld demons,” again borrowing from Roscher, may extend further even than dogs, wolves, and jackals.  Maher et al. (2011) report on fox-human burials at upper Paleolithic sites in Jordan.  Arguing that if “the earliest domestic dogs in the Near East are small, it is not much of a stretch to think that similarly-sized foxes could have been considered as potential domesticates to prehistoric people.”  Maher et al. speculate that “the burial of a fox with a human might have had the same social, ideological or symbolic significance as that of a human with a dog.”  There was, they argue, a “special relationship (i.e., companion) to the humans in these graves.” 

(The subject of abandoned domestication efforts could be an entire treatise in itself.  Dinell (1992; see also Jarman, 1972) cites several examples, including that red deer were “husbanded and perhaps herded in the Mesolithic,” only to be “discarded in favor of sheep.”  South American canids that had been at least partially domesticated lost favor among native groups to European dogs after the conquistadors arrived.) 

The difficulty of differentiating the canids in the graveyards extends to images, such as the early depictions of Anubis which, as Osborn and Osbornova (1998) observe, could have the head of a dog or jackal, but ears more fox-like, the tail bushy and short like a jackal.  “The black color is typical of Anubis statues and paintings, although black canines were rarely seen in Egypt.”

Even at the beginning of historical times, dogs were coupled with death by Homer, where many references describe them feeding on the slain.  “Nay of a surety many a one of the Trojans shall glut the dogs and birds with his fat and flesh, when he is fallen at the ships of the Achaeans  (ή τις Τρωων κορεει κυνας ήδ’ οιωνους δημω και σαρκεσσι, πεσων επι νηυσιν Αχαιων).” (Iliad, VIII.379-80, Murray translation, 1928)  Lilja (1976) cites strands of Homeric scholarship suggesting that this would have been expected from wild or half-wild pariah dogs supplementing the food they took from dunghills.

Figure 3. Wall painting in tomb of Inerkhau
Funerary Rituals

Herodotus wrote that a Persian’s corpse is not buried until it has been mauled by birds or dogs (ὡς οὐ πρότερον θάπτεται ἀνδρὸς Πέρσεω ὁ νέκυς πρὶν ἂν ὑπ᾽ ὄρνιθος ἢ κυνὸς ἑλκυσθῇ, Histories 1.140).  Writing of the Bactrians east of Persia, Strabo (Geography 11.11.3) describes dogs known as “undertakers,” perhaps better translated as “entombers,” as suggested by Mair (2007):

“Now in early times the Sogdians and Bactrians did not differ much from the nomads in their modes of life and customs, although the Bactrians were a little more civilised; however, of these, as of the others, Onesicritus does not report their best traits, saying, for instance, that those who have become helpless because of old age or sickness are thrown out alive to dogs kept expressly for this purpose, which in their native tongue are called "undertakers," [ἐνταφιαστάς] and that while the land outside the walls of the metropolis of the Bactrians looks clean, yet most of the land inside the walls is full of human bones, but that Alexander broke up the custom.”

Devotion unto Death

Pliny the Elder in the first century CE (Natural History VIII.lx.143) recorded instances where dogs preferred to die with their masters:

“When Jason of Lycia had been murdered his dog refused to take food and starved to death. But a dog the name of which Duris gives as Hyrcanus when King Lysimachus's pyre was set alight threw itself into the flame, and similarly at the funeral of King Hiero.” 

Stories of dogs protecting their masters’ graves can be found even in places where dogs are not favored animals.  In the 10th century Arabic manuscript, The Book of The Superiority of Dogs over Many of Those Who Wear Clothes, Ibn al-Marzuban says:

“A certain story teller said: al-Rabi' b. Badr had a dog he had reared himself. When al-Rabi' died and was buried, the dog kept on throwing himself against the grave until he himself died. He also said:  ʼAmir b. ʼAntarah had both hounds and guard dogs for his flocks and treated them well when they were with him. When ʼAmir died the dogs remained by his grave until they died there, though his family and relatives had already left him.”

Societies often go through periods where dogs will be sacrificed and buried with their masters.  Rice notes that this was most prevalent in the Old Kingdom in Egypt, with dogs often buried at the threshold of the tomb just as they guarded the master’s house in life.   The practice was known among Native American tribes, as recorded by Pettit (1950) and Gifford (1955). Barnett (1939) writes of the Gulf of Georgia Salish that dogs could be killed at the owner’s death and buried with him.

Dogs and Ghosts

Dogs are said to sense ghosts.  Pythagoras would hold a dog to the mouth of a dead disciple to receive the departing spirit (Ash 1927, at 41).  Dogs might even stand in for those who have died in certain practices.  In describing Russian celebrations of the dead, Georgi (1780) reported:

“On the Thursday in passion-week every father of a family places certain eatables in the yard of his house, with a lighted torch near them, to the memory of each person that has died out of his house.  The dogs, as proxies for the dead, regale themselves on this provision.”

Figure 4. Water ghost with canid figures
In a rock art panel near Dinwoody, Wyoming exaggerated hands stretch from a water ghost, with canid figures below (Francis and Loendorf 2004).  In the Ghost Dance, a Native American religious movement with significant political overtones, as described by duBois (1939), Gayton (1948), and Mooney (1896), Native Americans killed dogs because the ghosts, fearing them, would not come to life if they were present. 

Guides to the Underworld

One of the broadest associations of dogs with death is the belief that they lead the dead to the next world, just as in life they led hunters to game.  How did dogs become psychopompoi, guides for the dead?   Michael Rice, in Swifter than the Arrow, sees this as part of the canine association with graveyards:

“[T]he wild dog or jackal stalking through a graveyard on the edge of a Neolithic or late predynastic settlement could in another dimension of existence become the divine entity which led the justified dead to the Afterlife or, by extension, vigilantly guarded the place of communal burial.” 

Browne (1896) describes an Irish legend:

“On the road between Ballycroy and Bangor, Erris, a phantom dog sometimes appears, as does a white cow, whose appearance is looked on as a death-warning. Several of the lakes are thought to be inhabited by ‘water horses,’ which sometimes come on land and endeavor to coax unwary people to mount them, and then, having got them mounted, carry them off into the water. They are believed to be seen once in every seven years.”

It is with Anubis that the function of guiding souls reached its most beautiful statement.


Figure 5. Jackal-headed dancer, Predynastic Period
Anubis was first represented as a jackal or a therianthrope, a beast-man with the head of a jackal.  Prehistoric rock art from North Africa shows dog-headed men when Libya was fertile enough for the rhinoceros to thrive, though the deistic significance of the therianthropes is uncertain (Coulson and Campbell 2001).  An image on a predynastic palette shows a jackal-headed dancer playing a pipe.  Budge (1920) provides the hieroglyphics for Anubis (a later pronunciation), showing configurations that could be applied to the god from predynastic times into the Coptic period. 

Breasted (1912) describes Anubis as a mortuary god, a god that presided over funeral ceremonies, including mummification. The coffin of Henui from the Middle Kingdom describes the beginning of Anubis’s function on earth as an embalmer:

“Anubis … lord of the Nether World, to whom the westerners (the dead) give praise … him who was in the middle of the mid-heaven, fourth of the sons of Re, who was made to descend from the sky to embalm Osiris, because he was so very worthy in the heart of Re.” 

Anubis, or a priest of the god, was often shown with the pharaoh, whose afterlife would be in the presence of the gods (Borchardt, 1907). Breasted notes that in the Old Kingdom it was not unusual for a coffin to depict the deceased with a jackal’s face, as if jackals were indeed spirits of the dead.

The Book of the Dead, a category of funerary texts, gives additional functions to Anubis in the afterlife, including weighing the heart or soul of the deceased (“weigher of righteousness”). See the vignette from Spell 125, Papyrus London BM EA 9901,3 (Stadler, 2012). That this depiction was standard is demonstrated by extraordinarily similar shroud from the Staatliche Museen in Berlin (Walker 2000, p. 96).

Figure 6. Anubis weighing a soul
Anubis Comes out of Egypt

Plutarch, a Greek philosopher of the second century, sought to make the canine-headed god intelligible to an audience that increasingly required religion and myth to be consistent with philosophy and natural history.  Anubis guarded the gods as dogs do men (τους θεους Φρουρειν ωσπερ οι κυνες τους ανθρωπους). He was a son of Nephthys: 

“And when Nephthys was delivered of Anubis, Isis owned the child. For Nephthys is that part of the world which is below the earth, and invisible to us; and Isis that which is above the earth, and visible. But that which touches upon both these, and is called the horizon (or bounding circle) and is common to them both, is called Anubis, and resembles in shape the dog, because the dog makes use of his sight by night as well as by day (γαρ ό κυων χρηται τη οψει νυκτος τε και ημερας ομοιως). And therefore Anubis seems to me to have a power among the Egyptians much like to that of Hecate among the Grecians, he being as well terrestrial as Olympic. Some again think Anubis to be Saturn; wherefore, they say, because he produces all things out of himself and breeds them in himself. He had the name of Kyon (which signifies in Greek both a dog and a breeder).  Moreover, those that worship the dog have a certain secret meaning that must not be here revealed. And in the more remote and ancient times, the dog had the highest honor paid him in Egypt (παλαι μεν τας μεγιστας εν Αίγυπτω τιμας ό κυων εσχεν); but after that Cambyses had slain the Apis and thrown him away contemptuously like a carrion, no animal came near to him except the dog only (ουδεν προσηλθεν ουδ’ εγευσατο τον σωματος αλλ’ η μονος ό κυων).”

If the worship of Anubis “must not be here revealed,” the authorities of the mysteries were still to be respected, even feared, just as if in recent times the rites of a private club that includes some of the most important men of a community were so protected that political and legal repression could result from writing about them. 

Figure 7. Anubis mask, Ptolemaic Period
The Emperor Tiberius, learning that a woman had been seduced in the temple of Isis in Rome by a man claiming to be Anubis, destroyed the temple and banished the goddess Isis from Rome (Josephus, Antiquities, XVIII. 65-80)  The imposter may have been wearing a mask and it is not certain how deceived the woman actually was.  A number of Anubis masks survive. Yet worship of the Egyptian deities soon returned to Rome.  Anubis could be depicted wearing the uniform of a Roman soldier, as is seen in a statue on the side of a tomb at Kom El Shukafa where Anubis wears a cuirass (dated to the late first or early second century AD by Michalowski, 1968).

In late antiquity, an apocryphal work preserved in Arabic, The Acts of Andrew and Bartholomew, with place names that would be in modern Tunisia and Libya, a fearsome dog-headed man becomes a follower of the two apostles and helps them convert a hostile city to Christianity.  Although named only “Dog’s Head,” the character in the dream-like drama would have been understood in the pre-Moslem and pre-Christian mythology of the area as having divine elements:

“Dog's Head arose, and went to where the disciples were, rejoicing and glad, in the knowledge of the right faith. And his appearance was fearful exceedingly; his height was four cubits; his face was like the face of a large dog, and his eyes like lamps of burning fire, and his back teeth like the tusks of the wild boar; and his [front] teeth like the teeth of a lion ; and the nails of his feet like a curved scythe; and the nails of his hands like the claws of a lion, and his whole appearance frightful, terrifying.” (Lewis 1904)

Seznec (1953) describes depictions of Anubis in the Renaissance, including on a gem worn by Catherine de’ Medici. The last plate shows Anubis as drawn by Petro Stephanonio in his book, published in 1627, on ancient sculptured gems.


Figure 8. Heracles and Cerberus, late 6th century BCE
The great canine archetype of Greek art and poetry was Cerberus (Κερβερος), which Hesiod in the 8th century BCE described as having 50 heads, though later writers usually settled on three, and in vase paintings often just two.  Lilja summarizes early poetic references:

“The original concept of Cerberus … must have been that of a watchdog: a dog was imagined as guarding the palace of Hades in the same way as dogs used to guard houses and palaces in this life.  It was, I think, only gradually that “the devouring aspect of death began to be embodied in Cerberus so that, finally, the watchdog of Hades was identified with death. Not until the final identification with death could Cerberus be imagined as welcoming visitors, the voracity of death being expressed in that the dog devoured those who tried to escape.” 

The twelfth labor of Herakles involved bringing Cerberus up from the underworld. Apollodorus of Athens in the second century CE, says “Cerberus had three heads of dogs, the tail of a dragon, and on his back the heads of all sorts of snakes.”  Herakles, going to the land of the dead, wishes to be initiated into the mysteries at Eleusis first, and this takes some time because of his foreign status and his slaughter of the centaurs. After a few other misadventures, Herakles gets to the underworld.  Pluto allows Herakles to take Cerberus from Hades providing he masters the dog without weapons.  Herakles flings his arms around the dog and holds on despite the bites of the dragon in its tail.  Ascending from Hades at Troezen on the eastern Peloponessus, Herakles leads the beast to Eurystheus, who had commanded the labor, then carries him back to Hades. 

Figure 9. Statue of Cerberus, c. 180 CE
Unlike Anubis, Cerberus remains a dog, though a fearsome and mysterious one with multiple heads and sometimes appendages of other beasts.  Something of a guard dog, Herakles leads him like a war dog, and his functions can be understood as belonging to the more vicious functions dogs can perform. 

Archetypes of the Collective Unconscious

Two great canine archetypes have been created by the western imagination, Anubis and Cerberus.  Each has become eternal, if not as gods or beings of the underworld, at least in myth, poetry, painting, sculpture, recently in film, and always in the unconscious.  One is the majestic guide who leads the procession down the dark path to the river, the other a guardian at the river, on one side or both, keeping the spirits from returning to trouble the living and preventing the living from disturbing the rest of the dead.  Both help us when our earthly presence is beyond help.  Neither is a common dog, one being part man, the other usually having three heads, yet despite or perhaps because of this psychotic shift away from reality we cannot dream without them, cannot depict death without thinking of them and wanting them to be real, regardless of how advanced and abstract our theologies.     

Again I must defer to Michael Rice:

“Anubis is indeed one of the greatest of the archetypes: he is the Night Lord, Grave-Watcher, Soul-Guide, a being who bridges the worlds of the seen and unseen.  In this he is one of the most fully realised of all the animal divinities, brought out of the collective unconscious of the early inhabitants of the Nile Valley, who give meaning and explanation to the entire community of animal-human conflations.  The theriomorphic archetypes of Egypt are the first to be recognised and recorded by any complex society; in this lies their power, for the arcehtype only requires recognition to assume an independent existence.”

As David Gordon White (1991) proved, dog-headed men and gods can be found across civilizations in every part of the globe, often associated with death, whereas Cerberus exists only in those belief systems that have roots in Attic Greek thought and art.  Cerberus is not as generic as Anubis because Cerberus is limited to those peoples who accept a specific geography for the afterlife. Cerberus guards a river of  forgetfulness, or the boundaries of hell but seldom leaves his post, while Anubis is as broad and deep as death itself. 


Figure 10. Anubis Cynocephalus, 1627 CE
Could it be that canids entered the human collective unconscious before they filled any function for our survival?  Did it happen among a group of hunter gatherers in China or Mongolia, in the Middle East, Europe, Africa, or several of these at once? (“Once” is a relative term here since thousands of years could separate events without now being detectible to us looking backward.  It may have even happened in some early societies and been lost where partial domestications began but did not continue.)  The jackal was never domesticated, though some may have been tamed, but if they are the origin of Anubis it is from their habitation of graveyards that this must take its origin, at the place where the dead were separated from the living and left to make their own way to the next world.  Who will lead them if not the denizens of those places?  Is this part of our relationship one with canids, both those that were later domesticated and those that were not?  Their association with death may even precede their association with us as helpers in living.

I am suggesting that our recognition of a species that would share space with us may have begun in part in funerary ritual, stories told around fires about the animals scavenging nearby, in human dream life. Was the fixing of the archetype in fact one of the first steps towards domestication, rather than a later result of it? Is Anubis so strong because the archetype, the dream figure, has never been separated from benefits that dogs provide for us at home and work?

A history of the society of dogs and men cannot be complete without considering the irrational in our relationship.  We must account for what Rudolf Otto called the numinous. (I avoid the term “spiritual” as it connotes belief systems in which I do not participate—worse, while doing so differently for different people—and is in any case unnecessary for my argument.)  Other animals than dogs appear in dreams and visions, and in dreams and visions of death, but those animals tend more often to be figures on the landscape, not nearly so interactive with us as are our dogs.  In dreams and visions dogs look at us, do things for us, understand us, expect and demand things from us, and obey or ignore our commands, just as they do in waking life.  Their thoughts and expressions may come with words, which does not happen in waking life, but given speech their nature is brought closer to ours in the irrational than in the rational.  Our relationship is that of a combined society in which dogs are not completely separate from us by being part of the "other."  They have become part of us as we are part of them, as Anubis combines the canine with the human. We are both children of Anubis.

Was Chloe afraid for herself that morning in Oklahoma City, or for both of us?  A question from an irrational moment cannot receive a rational answer, but there will come that day when I shall know. 
Figure Sources and Permissions

Figure 1: Anubis Embalming an Official, Sennedjem courtesy Dahl and Wikimedia:  for the specific locus in the tomb, see photograph NB 1973.2313 by J.-Fr. Gout on the website of the Institut Francais d'Archeologie Orientale. The original tomb excavation and photographs of the walls were made by Bernard Bruyere (1959). Figure 2: Two Human-Fox Burials, Epipaleolithic Period.  Maher et al.  License for general use. Figure 3: Wall Painting in Tomb of Inerkhau.© Thierry Benderitter.  Permission granted by Dr. Benderitter, Toulon, France, member of the International Association of Egyptologists. For further information as well as photos showing the location of the image in the tomb of Inherkhaouy, see Bruyere (1930). Figure 4: Water Ghost with Canine Figures. Permission to use granted by University of Utah Press, November 24, 2014. Figure 5. Jackal-Headed Dancer, Quibell and Green, 1902, Plate 28, source outside of copyright. Figure 6. Anubis Weighting the Soul of a Decedent, courtesy Jeff Dahl and Wikimedia Commons. Figure 7. Anubis Mask, Roemer-Pelizaeus Museum Hildesheim, courtesy Einsamer Schutze, Wikimedia. Figure 8. Heracles and Cerberus, Louvre F204, courtesy Bibi Saint-Pol and Wikimedia. Figure 9. Statue of Three-Headed Cerberus, Heraklion Museum, courtesy Tom Oates, Wikimedia Commons.  Figure 10. Anubis Cynocephalus, Stephanonio, 1627, source outside of copyright.

  1. Apollodorus. The Library (1921, Frazer, J.G., translator). Loeb Classical Library, vols. 121-2. Harvard University Press, Cambridge.
  2. Ash, E.C. (1927). Dogs: Their History and Development. Ernst Benn Ltd., London.
  3. Barnett, H.G. (1939). Culture Element Distributions; IX: Gulf of Georgia Salish. Berkeley Anthropological Records, 1(5).
  4. Borchardt, L. (1907). Das Grabdenkmal des Königs Ne-User-Re.  Leipzig.
  5. Breasted, J.H. (1912). Development of Religion and Thought in Ancient Egypt. Charles Scribner’s Sons, New York.
  6. Brewer, D.J., Phillips, A.A., and Clark, T. (2002). Dogs in Antiquity: Anubis to Cerberus.  Aris & Phillips. 
  7. Browne, C.R. (1896). The Ethnography of Ballycroy, County Mayo.  Proceedings of the Royal Irish Academy, series 3, vol. 4. Dublin.
  8. Bruyere, B. (1930). Rapport sur les Fouilles de Deir el Medineh, FIFAO 8. Cairo: Institut Francais d’Archeologie Orientale.  
  9. Bruyere, B. (1959). La tombe n° 1 de Den-nedjem à Deir el Médineh. Mémoires publiés par les membres de l'Institut français d'archéologie orientale du Caire 88.  Cairo: Institut Francais d’Archeologie Orientale. 
  10. Budge, E.A.W. (1895). The Book of the Dead: The Papyrus of Ani. The Medici Society, Ltd. New York.
  11. Budge, E.A.W. (1920). Egyptian Hieroglyphic Dictionary, vol. 1. John Murray, London.
    Entry on Anubis, Budge (1920)
  12. Coulson (2001). African Rock Art: Paintings and Engravings on Stone.  Harry N. Abrams, New York.
  13. du Bois, C. (1939). The 1870 Ghost Dance. Berkeley Anthropological Records, 3(1).
  14. Francis, J.E., and Loendorf, L.L. (2004). Ancient Visions: Petroglyphs and Pictographs of the Wind River and Bighorn Country, Wyoming and Montana. University of Utah Press. Salt Lake City.
  15. Gayton, A.H. (1948). Yokuts and Western Mono Ethnography; II: Northern Foothill Yokuts and Western Mono. Berkeley Anthropological Records, 10f(1) and (2).
  16. Georgi, J.G. (1780). Russia: or, A Compleat Historical Account of All the Nations Which Compose That Empire, vol. I. London.
  17. Gifford, E.W. (1955). Central Miwok Ceremonies. Berkeley Anthropological Records, 14(4).
  18. Herodotus (2007 translation). The Landmark Herodotus: The Histories (ed. Strassler, R.K.). Pantheon Books, New York.
  19. Homer. The Iliad vol. 1(1928, Marray, A.T., translator and editor). William Heinemann Ltd., New York.
  20. Jarman, M.R. (1972).  European Deer Economies and the Advent of the Neolithic.  In Papers in Economic Prehistory (Higgs, E.S., ed.). Cambridge University Press. ("[W]e should not overlook the possibility that deer may have been herded and husbanded  in the past.  Although it is usual to consider the deer as essentially a wild animal, there is no reason to believe that this was necessarily always so; indeed, there is much evidence to the contrary.")
  21. Josephus, Jewish Antiquities, vol. 9 (Feldman, L.H., translator). Loeb Classical Library, Cambridge, Massachusetts.
  22. LaFaye, G. (1884).  Histoire du Culte Des Divinites D’Alexandrie.  Paris: Ernest Thorin.
  23. Lewis, A.S. (1904). The Mythological Acts of the Apostles.  Horae Semiticae No. IV. C.J. Clay and Sons, London.
  24. Lilja, N. (1976). Dogs in Ancient Greek Poetry.  Commentationes Humanarum Litterarum. Societas Scientiarum Fennica. Helsinki.
  25. Lupo, K.D. (2011).  A Dog Is for Hunting.  In Ethnozooarchaeology (Albarella, U., and Trentacoste, A., eds.).  Oxford: Oxbow Press.  Karen Lupo, studying the use of dogs for hunting by Central African foragers, noting that “foragers and villagers believe that dogs and other animals possess spirits and can haunt the living.  Dogs are also viewed as a common physical form taken by sorcerers and witches.”  Burials of dogs near humans should not be taken as indicative of a close relationship: “people lack formalized cemeteries and the dead are often interred behind huts or living areas; dogs are often buried in the same general location resulting in close spatial proximity between dog and human graves. However, it is important to note that in this context the proximity of burial location to the house and other family members does not refl ect a similarly close position of the dog within the family.”
  26. Maher, L.A., Stock, J.T., Finney, S., Heywood, J.J.N., Miracle, P.T., and Banning, E.B. (2011).  A Unique Human-Fox Burial from a Pre-Natufian Cemetery in the Levant (Jordan).  PLoS One, 6(1), e15815, doi:10.1371/journal.pone.0015815. 
  27. Mairs, R.M. (2007).  Ethnicity and Funerary Practice in Hellenistic Bactria. In Crossing Frontiers: The Opportunities and Challenges of Interdisciplinary Approaches to Archaeology (Schroeder, H., Bray, P., Gardner, P, Jefferson, V., and Macaulay, E., eds.). Oxford: School of Archaeology, 111- 124.
  28. Marzuban, I. (1978). The Superiority of Dogs Over Many of Those Who Wear Clothes (Smith, G.R., and Haleem, M.A.A., translators). Aris & Phillips, Westminster, England.
  29. Michalowski, J.M.K. (1968).  L'Art de l'ancienne Egypte. L. Mazenod, Paris.
  30. Mooney, J. (1896). The Ghost Dance Religion and the Sioux Outbreak of 1890.  Extract from the 14th Annual Report of the Bureau of Ethnology, Government Printing Office, Washington, D.C.
  31. Morey, D.F. (2006).  Burying Key Evidence: The Social Bond Between Dogs and People.  Journal of Archaeological Science, 33(2), 158-175.
  32. Osborn, D.J., and Osbornova, J. (1998).  The Mammals of Ancient Egypt.  Warminster, England: Aris & Phillips Ltd.
  33. Otto, R. (1961). The Idea of the Holy: An Inquiry into the Non-factor in the Idea of the Divine and Its Relation to the Rational.  Oxford University Press, London.
  34. Pettit, G.A. (1950). The Quileute of La Push, 1775-1945. Berkeley Anthropological Records, 14(1).
  35. Quibell, J.E., and Green, F.W. (1902). Hierakonpolis, Part II. Bernard Quaritch, London.
  36. Rice, M. (2006). Swifter than the Arrow: The Golden Hunting Hounds of Ancient Egypt. I.B. Tauris & Co., London.
  37. Roscher, W.H. (1896). Das Von Der “Kynanthropie” Handelnde Fragment. Leipzig: Bei S. Hirzel.
  38. Russell, N. (2012). Social Zooarchaeology: Humans and Animals in Prehistory.  New York: Cambridge University Press (see particularly the discussion of shamanism at pp. 236-7, and symbolism at pp. 291-4).
  39. Seznec, J. (1953). The Survival of the Pagan Gods.  Princeton University Press.
  40. Stadler, M.A.  (2012).  Thoth, in UCLA Encyclopedia of Egyptology.  Los Angeles. 
  41. Steindorff, G. (1896). Grabfunde des Mittleren Reichs des Königlichen Museen zu Berlin, vol II. Berlin.
  42. Stephanonio, Petro (1627).  Gemmae Antiquitus Sculptae.  Rome:
  43. Walker, S. (2000). Ancient Faces: Mummy Portraits from Roman Egypt. Metropolitan Museum of Art, New York. (In addition to the reference in the text, see the plates on pages 130 and 134.)
  44. White, D.G. (1991). Myths of the Dog-Man.  University of Chicago Press, Chicago.