If They Can't See Me, They Still Might Hear Me: Dogs Think About Sound When Stealing Treats (or Chasing Suspects?)

Canine sensitivity to human social and communicative cues has been the subject of significant research in a number of labs for several decades. Research has also looked at whether dogs are concerned with our attentional states, e.g., whether their behavior changes when we’re looking at them. If a dog is told not to eat a piece of food on the floor, the dog is more likely to obey if the human giving the command is watching the dog (and more likely to disobey if he is not watching). Even if the dog cannot resist the temptation to get to the food when the human is watching, it will do so in a more indirect manner, such as by trying to sneak up from the direction the human is not looking. I reviewed a number of studies to this effect in Service and Therapy Dogs in American Society (section on Eye Contact, beginning at page 22).

Recently, a research team designed an experiment to consider whether a dog’s behavior in taking a piece of forbidden food would differ when the dog could take the food silently from situations where taking the food would require making noise. They found, not surprisingly, that the dogs preferred to take the food in a way that didn’t make noise. The researchers tested dogs living in private homes as well as dogs from a shelter, the idea being to determine whether living with an individual or a family would affect a dog’s ability to use auditory information. There were 22 privately owned dogs used in the study, and 24 shelter dogs.

Food used for the study consisted of hot dog segments, bacon strips, and commercial dog treats. Food was placed in containers. All containers had brass bells hung across the opening, but for half the containers, the ringers had been removed. Owners of private dogs were asked what commands they would use to tell a dog to leave food alone. Dogs from shelters were told, “No.” Trials were videotaped. A dog was put about 1.5 meters from two containers, one of which was silent and one of which was noisy. The experimenter stood between the containers. The dog was held on a leash by someone else. After getting a dog’s attention, the experimenter put a treat inside each container. The experimenter demonstrated the auditory properties of the container by moving the string holding the bells five times. The noisy container rang during this manipulation, while the silent container did not. The dog was told “No,” or “Wait,” or another command designed to discourage it from going to the treat. The experimenter then adopted a Looking or Not Looking position. In the Looking position, she sat between the containers and looked straight ahead. In the Not Looking position, she put her head between her knees and faced the ground. The dog was then released from the leash. Everything was a double blind as possible.

The results indicated that dogs significantly preferred the silent container only in the Not Looking situation. When the experimenter was looking, there appeared no need for silence and they would almost as often approach either container. No significant difference was found between the privately owned dogs and the shelter dogs, so it does not appear that socialization in a family is essential to a dog’s capacity to consider the effect of sound. The authors note that these results could have significance in law enforcement and military contexts. Although they do not speculate further on this suggestion, it is possible to imagine that training a dog in suspect apprehension could be designed to take into account how a dog’s ability to surprise and subdue a suspect will be increased if the dog is sensitive to the suspect’s ability to hear its approach.

Kundey, S.M.A., De Los Reyes, A., Taglang, C., Allen, R., Molina, S., Royer, E., and German, R. (2010) Domesticated Dogs (Canis familiaris) React to What Others Can and Cannot Hear. Applied Animal Behaviour Science, 126, 45-50.

A question that might be worth investigating is whether different breeds show specific patterns to making sounds. A study by researchers in Hungary looked at reactions of dogs selected from three breed groups (Belgian shepherds (Tervuerens and Groenendaels), retrievers (golden and Labrador), and sled dogs (malamutes and huskies)). The sled dog group and the retrievers continued to show friendly, tolerant behavior even when a stranger approached in a threatening manner, but half the Belgian shepherds responded to threatening behavior with their own aggressive and threatening behavior. Even those shepherds that did not respond aggressively did not ignore the threatening behavior. Thus, a threatening suspect might elicit different levels of responses from different breeds of police dogs. Sound factors might be worth investigating in such contexts. Vas, J., Topal, J., Gacsi, M., Miklosi, A., and Csanyi, V. (2005). A Friend or an Enemy? Dogs' reaction to an Unfamiliar Person Showing Behavioural Cues of Threat and Friendliness at Different Times. Applied Animal Behaviour Science, 94, 99-115.

Dogs Misled by Strangers as Well as Owners in Food Choice Experiment. Another recent piece of research considered whether dogs, in a food choice experiment, were misled more easily by their owners than by unfamiliar but friendly humans. Thus, if dogs could choose between a slice of dry sausage and a (presumably) unappetizing pellet, would the pointing of the handler towards the pellet be more persuasive than the pointing of a stranger, despite the dog having seen where the tastier item was placed. The researchers found that dogs were misled almost as much by strangers as by their handlers. Marshall-Pescini, S., Prato-Previde, E., and Valsecchi, P. (2010) Are Dogs (Canis familiaris) Misled More by Their Owners than by Strangers in a Food Choice Task? Animal Cognition (not yet in print). As I’ve said before, they trust us too much.

Studies on Other Species. These types of two-choice tests have been done with other animals besides dogs, including chimpanzees and macaques. The picture shows a wolf in a two-choice test with a human pointing towards one of the choices. Here, researchers found that wolves could be trained to react to human pointing gestures, an ability that may have been significant during the process of domestication. Viranyi, Z, Gacsi, M,, Kubinyi, E., Topal, J., Belenyi, B., Ujfalussy, D., and Miklosi, A. (2008). Comprehension of Human Pointing Gestures in Young Human-Reared Wolves (Canis lupus) and Dogs (Canis familiaris). Animal Cognition, 11, 373-387.

Guide Dog in Pompeian Fresco?

Perhaps the oldest depiction of a dog that might have been functioning as a guide dog is in a fresco from Pompeii (Plate 78. p. 140, Maiuri, A., Roman Painting, Milan, Skira Publishing, 1953). Unfortunately, the fresco is in very poor condition, which probably explains why it is more often referred to than reproduced. (See Fishman, G.A., When Your Eyes Have a Wet Nose: the Evolution of the Use of Guide Dogs and Establishing the Seeing Eye. Survey of Ophthalmology 48, 452-458 (July 2003).)

It is not certain the man is blind, and Amedeo Maiuri's narrative does not speculate on blindness, though others have done so. Maiuri, once curator of the Naples Museum, suggests that the man's posture is that of a beggar. Otto Keller, writing almost 50 years before Maiuri, refers to the man as a blind beggar (blinder Bettler) and the dog as little (Hündlein). Otto Keller. Die Antike Tierwelt. Leipzig (1909).

The man is facing two women who, in my opinion but not in Keller's rendition, are watching but not clearly interacting with him. Their reaction to him may have been the dramatic focus of the painting. The scene appears to be set in a market, stalls of which are visible in the background. The man holds a walking stick far forward in his left hand, a posture that may indicate he was using it as much to find obstacles as to provide support. The leash is brownish red, probably a strip of leather, loose, and attached to what appears to be a wide collar. (The only Photoshop enhancement made to the plate posted here was to color the bottom part of the leash, which fades considerably towards the dog's neck.)

The dog, about knee high, looks back at the man, ears pointed like those of a small terrier. In the caption to the drawing he makes of the scene, Keller labels the dog a pariah (Pariahund), though this is perhaps no more than saying that the dog was a street urchin. If the dog is guiding, it probably means that the two are following a route both knew and navigated together by sharing their senses. As sometimes happens with hearing dogs now, the dog may have gradually learned to substitute his sense for the one his master was losing.

IRS Affirms Deductibility of Psychiatric Service Animals - Comment with Three Addenda (So Far)

An Information Letter issued by the IRS to Representative John Tanner (Dem.-Tenn.) states:

“The costs of buying, training, and maintaining a service animal to assist an individual with mental disabilities may qualify as medical care if the taxpayer can establish that the taxpayer is using the service animal primarily for medical care to alleviate a mental defect or illness and that the taxpayer would not have paid the expenses but for the disease or illness.” Letter of George Blaine, Associate Chief Counsel (Income Tax and Accounting) to Representative John Tanner, INFO 2010-0129 (May 11, 2010, released June 25).

Dr. Esnayra, the president of the Psychiatric Service Dog Society, and I had written to the IRS on July 17, 2009, concerning the deductibility of service dogs for individuals with mental health disabilities. Tax Notes, August 31, 2009. One of the people we had consulted with on the issue was Carling Dinkler, an aide to Representative Tanner. The Representative, Dr. Esnayra, and I were concerned with language from Publication 502, Medical and Dental Expenses, stating:

“You can include in medical expenses the costs of buying, training, and maintaining a guide dog or other service animal to assist a visually-impaired or hearing-impaired person, or a person with other physical disabilities.” (emphasis added). As of this writing, the online posting of Publication 502 retains the adjective “physical.”

The Information Letter does not define service animals for purposes of determining what expenses are deductible. Rather, the Letter refers to two Tax Court decisions that provide parameters that taxpayers must consider in advancing a deduction argument. The first case is Havey v. IRS, 12 TC 409 (1949), in which the taxpayers, a couple, sought to deduct their travel and room expenses at resorts and a dude ranch which were taken by the couple because of the wife’s heart problems. Citing the predecessors of the current Section 213 and its regulations, the court noted that deductible expenses must be incurred primarily for the prevention or alleviation of a physical or mental defect or illness. Personal, living, and family expenses, on the other hand, are not deductible. Thus, athletic club expenses would generally be considered personal or living expenses. The case states:

“In determining allowability, many factors must be considered. Consideration should be accorded the motive or purpose of the taxpayer, but such factor is not alone determinative. To accord it conclusive weight would make nugatory the prohibition against allowing personal, living, or family expenses. Thus also it is important to inquire as to the origin of the expense. Was it incurred at the direction or suggestion of a physician; did the treatment bear directly on the physical condition in question; did the treatment bear such a direct or proximate therapeutic relation to the bodily condition as to justify a reasonable belief the same would be efficacious; was the treatment so proximate in the time to the onset or recurrence of the disease or condition as to make one the true occasion of the other, thus eliminating expense incurred for general, as contrasted with some specific, physical improvement?”

Although a physician had advised the trips the taxpayers sought to deduct in Havey, the court noted that during the trips she was not under the care of a physician. The court did not question that the trips may have been beneficial, but the “record fails to show, however, that the benefit derived by the wife was in any respect different from that enjoyed by any vacationer at the same resorts at the same time.” The court said that a visit to a sanitarium might have received different treatment. Thus, an “incidental benefit is not enough.”

The second case cited in the Information Letter is Jacobs v. Commissioner, 62 TC 813 (1974), where the taxpayer sought to deduct legal fees under a property settlement in a divorce proceeding. The taxpayer’s psychiatrist recommended that he delay marriage plans to a woman to whom he was engaged so that he could resolve his doubts about the marriage. He married anyway, but instead of resolving themselves, the problems of the marriage worsened and the taxpayer began to take anti-depressants. The psychiatrist recommended that the taxpayer get a divorce. When the attorney the taxpayer hired seemed to be going too slow, the psychiatrist recommended that he change attorneys, which he did. On his 1959 return, the taxpayer deducted over $10,000 of expenses relating to the lump sum payment to his wife and the attorneys’ fees.

The Tax Court accepted that the taxpayer had a mental disease for which medical care was appropriate. The court noted that for personal expenses to be deductible, they must be claimed primarily for the prevention or alleviation of disease. The key sentence from the case that the Information Letter paraphrases reads as follows:

“One important condition, which petitioner must satisfy if his claim is to succeed, is whether the expenditure would have been made even if there had been no illness. This ‘but for’ test requires petitioner to prove both that the expenditures were an essential element of the treatment and that they would not have otherwise been incurred for nonmedical reasons.”

The court was not convinced that the taxpayer would have gotten divorced had he not been ill, despite the fact that he began to get better after the divorce:

“The marriage had not worked from the beginning. Petitioner detailed for us how his wife began to attach and abuse him almost immediately after the wedding. Even if petitioner had been emotionally sound, we believe he would have gotten a divorce, if not when he did, then shortly thereafter. Moreover, there is absolutely no evidence that petitioner's wife would not have at some point initiated and procured a divorce. In short, we cannot say that petitioner would not have in any event incurred the expenditures in question.”

Thus, the divorce-related expenses were not deductible. The analogies of the cases to the use of a service dog are clear, though their application may not always be simple. The expense of a dog must be made primarily for the prevention or alleviation of the disease, and the taxpayer must be prepared to show that these expenses would not have been made but for the disease.

The model is easy to apply when considering a guide dog, or a mobility impairment dog, or a dog trained to work with a veteran with PTSD. A medical need is easily established, and the highly expensive training for an appropriate animal would not be undertaken were it not for the way the dog alleviates a condition. In a case like a seizure-alert dog, where a pet spontaneously begins to alert to a handler’s seizures, nothing would appear deductible except subsequent training to enhance this skill. The dog was already a pet and would presumably have been fed and cared for in any case. With an autism service dog, working in a family with an autistic child, the circumstances around the acquisition of the dog might be critical to justifying a deduction.

This is a significant development for the service animal community, and the IRS is to be commended for providing this information to Representative Tanner. Representative Tanner must be thanked as well, and deserves to be considered a new hero of service animal users.

At a panel discussion of the Low Income Taxpayers Committee of the American Bar Association Tax Section in May 2010, I argued that there may be instances, such as with a seizure-alert dog, or a dog that alerts an individual to a substance to which the individual is dangerously allergic, where the all-or-nothing approach to deductibility of service animals may not be appropriate. Even if the animal was obtained primarily for the medical condition, and would not have been obtained but for that condition, the fact is that even a service animal has some of the same companionship needs, as does the handler, that would apply to a pet. As has been true of the Departments of Justice and Transportation, the proliferation of service animal types will likely lead to new issues that require additional nuance in the application of the tax law to taxpayers having these animals.

Addendum in Response to an Email. A reader of this blog asked me to elaborate on a situation where full deduction of a service dog's expenses might not be appropriate. A situation on which I consulted (not on any tax issue) comes to mind. A family acquired a service dog to work with their autistic child. The dog was trained to keep the child from running into traffic and to stop certain behaviors the child sometimes engaged in (such as pica, compulsive eating of non-food items such as soaps and feces), and to bark if the child awoke in the night. The relationship between the child and the dog quickly deteriorated. The child did not sleep well and the parents began to let the dog sleep with themselves or another sibling of the autistic child. The dog shied away from the child in the schoolroom and the school soon advised the parents that they would have to have someone in the school with the dog if it was to continue to come to the school with the child. Soon the only function that the service dog performed for the child on a regular basis was to walk to and from the school with the child and the mother, when it could keep the child from running away.

I do not know precisely how the IRS would view this situation. The dog was highly trained and cost nearly $20,000. (So I was told. My advice was sought by the school, not the family.) The dog would not have been acquired but for the services it was to perform for the autistic child. The acquisition and initial maintenance costs would appear to fit within the parameters of the Information Letter. But after a time, was the dog being used primarily to alleviate a mental defect or illness? Perhaps not. On the other hand, hearing dogs, which have been approved by the IRS (Rev. Rul. 68-295, 1968 2 CB 92), often only signal the presence of a particular sound a few times during a day, though they are available for this work throughout the day. Seizure-alert dogs may not alert to an oncoming seizure for months, though they also are available for their detection ability all the time they are near their handlers. Or would the IRS view this as analogous to a guide dog that goes out of service and becomes a pet when the blind person gets a younger animal? These sorts of issues may take some while to arise, but the Service will have to consider them sooner or later.

Second Addendum in Response to an Email. Another reader of the blog asks if an emotional support animal would be deductible under IRS pronouncements. I’m not sure. The Service does not define “service animal,” “psychiatric service animal,” or “emotional support animal.” Rather, the Information Letter states that if the taxpayer uses the service animal “primarily for medical care to alleviate a mental defect or illness,” and would not have paid the expenses but for the disease or illness, the costs of buying, training, and maintaining the animal “may qualify as medical care,” i.e., may be deductible. Nor does the Service define what it means to “assist” an individual with mental health disabilities.

It is certainly possible that someone with an animal that might be designated an emotional support animal by the Department of Justice or the Department of Transportation could allege that they acquired the animal primarily to alleviate a mental illness and would not have done so but for the illness. Although training is mentioned as a deductible expense, the Service does not specify that the animal must be trained. The Department of Transportation, in recent cruise line access rules, stated that an emotional support animal is “not trained to perform specific physical tasks.” This follows statements in the final air carrier access rules that emotional support animals need not have specific training for their emotional support function, but must be trained to behave appropriately in a public setting.

The Department of Justice, in rules proposed in 2008, indicated that psychiatric service animals, unlike emotional support animals, are “trained to do work or perform tasks.” The distinction between doing work and performing tasks received the following elaboration in those proposed rules:

“[T]he phrase ‘do work’ is slightly broader than ‘perform tasks,’ and adds meaning to the definition. For example, a psychiatric service dog can help some individuals with dissociative identity disorder to remain grounded in time or place. As one service dog user stated, in some cases “critical forms of assistance can’t be construed as physical tasks,” noting that the manifestations of ‘brain-based disabilities,’ such as psychiatric disorders and autism, are as varied as their physical counterparts.” 73 Fed. Reg. 34521.

It would appear that a psychiatric service animal, as described by the Department of Justice, could fit cleanly within the parameters set by the IRS for deductibility of service animals. I suspect that the IRS would be reluctant to treat an untrained emotional support animal in the same way, even if the taxpayer using the animal has a mental defect or illness. Although training is not required under the Service’s parameters, on audit an IRS agent might become concerned that an animal behaving like a pet without any real training related to the taxpayer’s mental illness is, in fact, a pet. It might be hard to argue that a taxpayer is using a pet primarily to alleviate a mental defect or illness.

Also, although the Service does not define “service animal,” it does use the term and might be inclined to look to the Department of Justice, which has considered the significance of these animals for a much longer time, and in much greater depth, than it has. Even though the Department of Transportation includes emotional support animals as a type of service animal (at least in air carrier access rules), such a broad definition of the term might include too many animals that are not sufficiently devoted to service work to justify a deduction. (The IRS might not follow the Department of Justice in restricting service animal status to dogs. In PLR 8033038, 1980 WL 134273, the agency allowed a medical expense for a cat that was "registered with your animal control division as a hearing aid animal." According to the private ruling, the cat was "trained to respond to unusual sounds in an instantaneous and directional manner." This alerted the cat's owner to possible dangers. The ruling analogizes this situation to that described in Rev. Rul. 68-295, which also involved an individual with a hearing impairment.)

On balance, and without being presented with specific facts, I would guess that a taxpayer with an emotional support animal might find it difficult to get the IRS to accept a deduction for such an animal.

Third Addendum in Response to a Phone Call. Perhaps it is because I have spent most of my professional career as a tax lawyer that I receive more comments on tax issues involving service animals than any other type of posting. I got a call from someone who identified herself as an IRS agent asking how she could recognize a taxpayer claiming to have a psychiatric service animal that was really only a pet. Dr. Esnayra and I discussed the verification procedures the IRS might use in such cases in the paper linked to above. We noted that the IRS is in a situation closer to that of the Department of Housing and Urban Development than of the Departments of Justice and Transportation. The latter often involve situations where a public accommodation or an airline must decide, in a very public setting, whether someone’s dog is actually a service dog. The rules of these agencies recognize that an employee’s questions from behind a counter or at the entrance to a restaurant could be intrusive and embarrassing. HUD, however, is often dealing with how a building can verify assistance animal status for someone who has submitted application documents for an apartment. The inquiry can be conducted privately. With the IRS, I think an auditing agent should ask to see any letter from a medical or mental health professional recommending the acquisition of a service animal. The agent may want to get the taxpayer’s permission to speak with the writer of the letter to determine if the professional really stands behind the recommendation. (See my blog of July 6 on a “psychiatric service macaque” where a doctor seems to have written a letter under pressure from the patient, but did not know what psychiatric service animals are supposed to do, or even how service animals are defined.) If the taxpayer presents a certification card for the animal, the agent should verify that the card was issued by a legitimate organization. As Dr. Esnayra and I noted, there are now websites that will sell service animal paraphernalia for several hundred dollars. This may be fraudulent under the laws of some states, but it is occurring and the purchasers may not just be people wanting to take their dogs on vacations. The tax benefits of a service animal can be much greater than having a dependent.

The IRS is certainly familiar with the fact that just because someone says a dog is certified doesn't mean that it has really passed any tests. The agency acquired explosives detection dogs for its Fresno Service Center which were eventually found incapable of detecting much of anything. The vendor was convicted of fraud for selling such dogs to the State Department, the Federal Reserve Bank, and the IRS, and sentenced to 63 months in prison. Presumably some of his business deductions for training the dogs might have gotten a second look. U.S. v. Ebersole, 411 F.3d 517, cert. denied, 126 S.Ct. 1142 (2006), on remand, 2007 WL 219969 (E.D. Va. 2007), aff’d, 189 Fed.Appx. 287 (4th Cir. 2006), motion to vacate denied, 2007 WL 750198 (E.D. Va. 2007).

Cancer Sniffers Perform Better with Some Cancers than Others, but Testing Parameters Need More Standardization

Two scientists, one from the New College of Florida and one from the Pine Street Foundation in San Anselmo, reviewed the current state of canine cancer detection research, finding 531 potentially relevant articles, but focusing their analysis on five articles and one unpublished manuscript. The studies looked at canine detection of the following cancers:

1. Bladder cancer detection from smelling urine of patients, with 41% success rate (compared with 14% expected by chance).1
2. Melanoma detection by smelling lesions on patients, with a success rate between 75 and 85.7%.2
3. Lung and breast cancer by smelling samples of a patient’s exhaled breath,3 with a specificity of 99% and a sensitivity of 99%. The authors of the survey article explain that sensitivity is the proportion of cancer samples which the canines correctly identify, and specificity is the proportion of control samples which the canines correctly indicate as controls.
4. Ovarian cancer by smelling ovarian tumor samples, with a sensitivity of 100% and a specificity of 97.5%.4 The researchers noted that the accuracy was remarkable given that some of the control tissues had been removed from areas adjacent to the tumor within the same patients.
5. Breast and prostate cancer by smelling urine samples in test tubes, with overall success rates that were not statistically significant.5 Another study, not analyzed because as yet unpublished, also failed to find significant results from dogs smelling urine of prostrate cancer patients.
6. Ovarian cancer by smelling exhaled breath condensate, with results not yet available.

The experiment involving breast and prostrate cancer was criticized by the authors of the review study for having different trainers using different methods, so that “inconsistencies in training may have made it nearly impossible for the dogs to perform well.” Storage conditions in the bladder cancer study varied, and dogs did considerably better when the urine sample was relatively fresh. The authors note that this study might have had insufficient controls, which might also have been true of the ovarian, breast, and prostrate cancer studies. The authors suggest also that there may be more cancer biomarkers at the source of the tumor and in exhaled breath than there are in urine.

The authors reaffirm their faith in using dogs as a diagnostic tool, but they emphasize the importance of additional research, as well as repeating studies already done. Mary Elizabeth Thurston suggested many years ago that cancer sniffers might be useful in poor countries (Lost History of the Canine Race).6 Much more research will be needed, however, before regular clinical use of cancer sniffers will become a reality. E. Moser and M. McCulloch (2010), Canine Scent Detection of Human Cancers: A Review of Methods and Accuracy. Journal of Veterinary Behavior, 5, 145-152.

1. Willis, C. M., Church, S. M., Guest, C. M., Cook, W. A., McCarthy, N., Bransbury, A. J., Church, M. R. T., & Church, J. C. T. (2004). Olfactory Detection of Human Bladder Cancer by Dogs: Proof of Principle Study. BMJ [British Medical Journal], 329, 712–715.
2. Pickel, D. P., Manucy, G. P., Walker, B. B., Hall S. B., & Walker, J. C. (2004). Evidence for Canine Olfactory Detection of Melanoma. Applied Animal Behaviour Science, 89, 107–116.
3. McCulloch, M., Jezierski, T., Broffman, M., Hubbard, A., Turner, K., & Janecki, T. (2006). Diagnostic Accuracy of Canine Scent Detection in Early- and Late-Stage Lung and Breast Cancers. Integrative Cancer Therapies, 5(1), 30–9.
4. Horvath, G., Jarverud, G.A., Jarverud, S., Horvath, I., 2008. Human Ovarian Carcinomas Detected by Specific Odor. Integrative Cancer Therapies, 7, 76-80.
5. Gordon, R.T., Schatz, C.B., Myers, L.J., Kosty, M., Gonczy, C., Kroener, J., Tran, M., Kurtzhals, P., Heath, S., Koziol, J.A., Arthur, N., Gabriel, M., Hemping, J., Hemping, G., Nesbitt, S., Tucker-Clark, L., Zaayer, J., 2008. The Use of Canines in the Detection of Human Cancers. Journal of Alternative and Complementary Medicine, 14, 61-67.
6. See my chapter on the Cancer Sniffers in Service and Therapy Dogs in American Society, 112.

Cruise Lines Must Accommodate Service Animals but Department of Transportation Seeks Comment on Emotional Support Animals (with Air Carrier Addendum)

The Department of Transportation has issued final rules on transportation for individuals with disabilities traveling on passenger vessels.[1] The rules are effective November 3, 2010. Comments are sought by October 4.

In 1991, the Department of Transportation stated that the Americans with Disabilities Act covered passenger vessels, including cruise ships, but noted that cruise ships are a unique mode of transportation, consisting of “self-contained floating communities.” The Department described cruise ships as something of a hybrid between a transportation service and a public accommodation, and noted that virtually all cruise ships serving U.S. ports are foreign-flag vessels.[2] In Spector v. Norwegian Cruise Lines,[3] the U.S. Supreme Court held that ADA requirements regarding disabled passengers applied to a foreign-flag vessel. In 2007, the Department issued proposed rules regarding cruise lines and received hundreds of comments.[4] A public hearing was held in April 2008 in which additional views were expressed.

The final rules define “service animal” as follows:

"'Service animal' means any guide dog, signal dog, or other animal individually trained to work or perform tasks for an individual with a disability, including, but not limited to, guiding individuals with impaired vision, alerting individuals with impaired hearing to intruders or sounds, alerting persons with seizure disorders to the onset of a seizure, providing minimal protection or rescue work, pulling a wheelchair, or fetching dropped items."[5]

This definition remains unchanged from the 2007 proposed rules and is identical to the definition under regulations issued by the Department of Justice and contained in 28 CFR 36. 104, though as discussed further below, DOJ has proposed changing this definition.[6]

The section devoted to service animals on ships, 49 CFR 39.91, contains some changes from the 2007 proposals. The proposals stated that a service animal must be able to accompany a passenger in all locations that passengers can use on a vessel, which the final rules confirm, but the latter add the phrase, “including lifeboats.” There can be no overall limitation on the number of service animals that can be bought on a voyage as this would be “tantamount to a number limit on passengers with a disability.”

A new subsection deals with food for the animal:

"You must permit the passenger accompanied by the service animal to bring aboard a reasonable quantity of food for the animal aboard the vessel at no additional charge. If your vessel provides overnight accommodations, you must also provide reasonable refrigeration space for the service animal food."[7]

The preamble specifies that vessels are not required to provide food for service animals, but must allow passengers to bring a reasonable quantity of food aboard at no additional charge. There must be refrigeration space for the animal’s food on ships going overnight, but the requirement does not apply to short-voyage ferries or water taxis.

The verification requirement remains unchanged from the proposal:

"You must accept the following as evidence that an animal is a service animal: Identification cards, other written documentation, presence of harnesses, tags, and/or the credible verbal assurances of a passenger with a disability using the animal."[8]

The proposed rules had provided that if a cruise line decided not to accept an animal as a service animal, the decision had to be explained in writing to the passenger. This requirement has not been included in the final rules.

The final rules contain a new provision dealing with the situation where a foreign government does not allow a service animal to disembark:

"If the legal requirements of a foreign government (e.g., quarantine regulations) do not permit a service animal to disembark at a foreign port, as a PVO you may require the animal to remain on board while its user leaves the vessel. You must work with the animal’s user to ensure that the animal is properly cared for during the user’s absence."[9]

The preamble notes that limitations on the ability of a service animal to leave the ship at a foreign port would be included in the information provided to potential customers inquiring about an upcoming cruise.[10] The vessel can insist that the animal not disembark at such a port, but the passenger should be able to get off without the animal. The vessel should work with the passenger to make sure the animal is cared while the passenger is on a lengthy excursion. Presumably this means that the vessel could charge for care of the animal in such a situation.

The preamble seems to waffle on emotional support animals:

"While this rule does not require it, the Department believes that it is a good idea to permit not only service animals, per se, but also emotional support animals (ESA) to accompany passengers with disabilities who use them. This can be beneficial to individuals who genuinely need the assistance of such an animal to enjoy fully travel and services aboard a vessel. We refer PVOs [passenger vessel operators] and passengers with disabilities to applicable provisions of the Department’s Air Carrier Access Act regulations and appendices (14 CFR part 382) for suggestions on how and in what circumstances it is appropriate to accommodate people using ESAs."

Thus, vessels are not required to accept emotional support animals. The Department acknowledges that this is inconsistent with air carrier access requirements and seeks comment on whether vessels should be required to accommodate emotional support animals. If this were required, the Department seeks comment on what safeguards vessels should be able to require to prevent abuse—such as “passengers attempting to pass off their pets as emotional support animals.”

Neither the new rules nor the preamble mention psychiatric service animals, unlike the 2008 final air carrier access rules,[11] which lump them with emotional support animals in allowing airlines to require advance notice and certain documentation. The preamble to the final air carrier access regulations saw psychiatric service animals and emotional support animals as providing some potential for abuse, but nevertheless referred to the two types of animals as two distinct categories.[12] The distinction in the air carrier access rules appears to be that an emotional support animal accompanies a person with a mental health-related disability listed in the Diagnostic and Statistical Manual of Mental Disorders, that accompanying the individual is necessary for the passenger’s mental health or treatment, and that this necessity and the condition have been assessed by a mental health professional who provides care for the passenger. A psychiatric service animal, for air carrier access, is trained by its owner or a professional trainer to perform tasks, such as fetching medications, reminding the user to take medications, or “acting as a buffer against other people crowding too close.”

The preamble to the final passenger vessel rules states that they are consistent with proposed rules of the Department of Justice amending Title II and Title III, but DOT seeks comment as to whether emotional support animals are an area where different rules might apply to vessels[13] The 2008 proposed regulations of the Department of Justice distinguished between emotional support and psychiatric service, stating:

"The Department is proposing new regulatory text in § 36.104 to formalize its position on emotional support/comfort animals, which is that '[a]nimals whose sole function is to provide emotional support, comfort, therapy, companionship, therapeutic benefits, or promote emotional wellbeing are not service animals.' The Department wishes to state, however, that the exclusion of emotional support animals from ADA coverage does not mean that individuals with psychiatric, cognitive, or mental disabilities cannot use service animals. The Department proposes specific regulatory text in § 36.104 to make this clear: 'The term service animal includes individually trained animals that do work or perform tasks for the benefit of individuals with disabilities, including psychiatric, cognitive, and mental disabilities.' This language simply clarifies the Department’s longstanding position and is not a new position." [14] (emphasis added)

Thus, following the same language as now defines “service animals” for passenger vessels, DOJ has had a “longstanding position” of recognizing service animals with functions related to mental disabilities. However, since there is no mention of psychiatric service animals in the final (nor the proposed) passenger vessel regulations, it is not clear if DOT considers that they are covered. A strong argument could be made that the claim for consistency with the proposed DOJ regulations must mean that they are covered (as distinguished from emotional support animals) but the matter should be clarified. It would be anomalous if DOT, having distinguished psychiatric service animals from emotional support animals for air carrier access purposes were now to be suggesting that they are not to be distinguished for passenger vessel access purposes. It would also mean that DOT and DOJ have a considerable, and probably illogical, rift on this matter, which seems unlikely.

The preamble indicates that subsequent rules addressing accessibility standards will describe relief areas for service animals. This was an area that DOT sought comment on in the 2007 proposed regulations.[15] DOT noted in 2007 that cruise operators typically provide such areas, so the agency may not see this as one of the issues it needs to resolve quickly.

The Department of Transportation is to be commended for clarifying many issues regarding service animals aboard cruise vessels. The failure to specify the treatment of psychiatric service animals is unfortunate, since there are a good many people taking these animals into public accommodations. It would be best to clarify the status of such animals as soon as possible.

Addendum for Federal Register Posting of July 30, 2010. The Department of Transportation posted some corrections to air carrier access regulations on July 30. The only one having to do with service animals is a correction to 14 CFR 382.111(e), which provides that airlines must provide assistance in stowing and retrieving "carry-on items, including mobility aids and other assistive devices stowed in the cabin (see also Sec. 382.91(c))." The parenthetical reference to 14 CFR 382.91(c) was a mistake, since that section refers to relief areas in air terminals. The cross-reference was supposed to be to 14 CFR 382.91(d), which has to do with helping passengers who cannot carry their carry-on luggage get it on the plane. The correction has been made. The mistake might have convinced some people (quite reasonably so) that service animal relief areas were to be provided inside cabins, since the referring provision refers to helping people stow items in the cabin. It would not have been an impossible argument. The Department does not state whether anyone actually made such a case to an airline. Department of Transportation, Correcting Amendments, 75 Fed. Reg. 44885 (July 30, 2010).

Addendum re Comments on Department of Transportation Rules. The Psychiatric Service Dog Society submitted comments regarding these rules on September 21, 2010. I wrote the initial draft of the comments. The comments have been posted by the Department of Transportation.

[1] Department of Transportation, Transportation for Individuals with Disabilities: Passenger Vessels, RIN 2105-AB87, 75 Fed. Reg. 38878 (July 6, 2010).
[2] 56 Fed. Reg. 45599-45560 (September 6, 1991).
[3] 545 U.S. 119 (2005).
[4] 72 Fed. Reg. 2833 (January 23, 2007). The reference to hundreds of comments is contained in the preamble to the final rules (75 Fed. Reg. 38879). A search of the website where comments are posted (www.regulations.gov) did not produce nearly that many, though I acknowledge that I may not have fully understood the operation of the site.
[5] 49 CFR 39.3.
[6] Although the definition of service animal would be different under proposed rules issued by the Department of Justice in 2008, DOJ noted that the inclusion of psychiatric service animals in the definition of service animal was not a change in regulatory position, but rather a clarification of “the Department’s longstanding position” regarding the status of psychiatric service animals. Nondiscrimination on the Basis of Disability by Public Accommodations and in Commercial Facilities, RIN 1190-AA44, 73 Fed. Reg. 34508, 34516 (June 17, 2008). Since the Department of Transportation sees its final regulations as consistent with these proposals, it would appear that psychiatric service animals should also be accepted under the rules now applicable to passenger vessels.
[7] 49 CFR 39.91(c).
[8] 49 CFR 39.91(d).
[9] 49 CFR 39.91(e).
[10] 49 CFR 39.53(e).
[11] Nondiscrimination on the Basis of Disability in Air Travel, RINs 2105-AC97, 2105-AC29, 21050AD41, 73 Fed. Reg. 27614 (May 13, 2008).
[12] 73 Fed. Reg. 27655.
[13] 75 Fed. Reg. 38890.
[14] 73 Fed. Reg. 34508, 34516 (June 17, 2008). On January 21, 2009, the Department of Justice notified the Office of Management and Budget that it was withdrawing draft final rules under Titles II and III, leaving further regulatory modification to the then incoming Obama administration.
[15] 72 Fed. Reg. 2841.

Can a Macaque Be a Psychiatric Service Animal?

A recent case from a federal district court in Missouri highlights some of the issues that users of psychiatric service animals face in establishing their access rights.  The plaintiff, Debby Rose, sued Wal-Mart, an educational institution where she was taking classes (Cox Health Systems), and the county health department where she lived. The defendants moved for summary judgment of dismissal, which was granted. Ms. Rose claimed to have suffered from agoraphobia and anxiety disorder since the 1970s, although she was not diagnosed with agoraphobia until 2006. There was evidence that she had denied having any mental illness on a job application in the early 1990s, and in 1999 she told her treating physician she had no symptoms of anxiety or depression. Even when she was diagnosed, the physician stated that this was based largely on Ms. Rose’s descriptions of herself. She was married three times, held many different jobs and moved often. She worked as a dental assistant and managed the Greene County Humane Society. She also worked as a real estate broker. The court saw these facts as evidence of her ability to operate in the world.

In 2005, Ms. Rose founded Wild Things Exotic Animal Orphanage where she and her sons rescued primates and sought to place them with facilities. In 2004, she had acquired a Bonnet Macaque monkey that was nearly dead and required constant care. She began to take the monkey with her nearly everywhere and began to describe the animal as a service animal and said that it performed various tasks related to her disability, including “breaking the spell,” “breaking off the focus,” “crowd control,” “changing the mood,” and she claimed to notice a change in her heart rate and blood pressure when around the monkey. Her physician described the monkey as sitting with Ms. Rose and comforting her. The monkey performed certain tasks, such as holding her hand, touching her face, sitting on her lap, retrieving a toothbrush to encourage her to get out of bed, bringing her the TV remote control or her cell phone, putting on the turn signal in the car when she reaches the street to her house so that she knows it is time to turn, and opening the car door. The monkey was also described as giving strangers a gentle push to stay away from her.

Although Ms. Rose cited her physician’s testimony as supporting her argument that the animal is a service animal, the doctor actually testified that he had no experience with service animals, did not know the definition of a service animal, and could not state whether the animal would qualify as such under the Americans with Disabilities Act.

The Springfield-Greene County Health Department began to receive complaints regarding Ms. Rose bringing the monkey into restaurants and opened an investigation to determine if the animal qualified as a service animal. The Department contacted various state and federal agencies and concluded that the monkey did not qualify as a service animal and was a threat to public health. The Department sent letters to food services establishments in the area, saying that allowing Ms. Rose to bring in her monkey would be a violation of Missouri health codes. Wal-Mart received the letter and denied Ms. Rose access with the monkey. Cox Health Systems also received the letter and informed Ms. Rose that she could not attend classes at Cox College with the monkey, but could continue without it.

Cox Health Systems also reviewed reports of the Centers for Disease Control, noting that in the context of access to health care facilities, providing access “to exotic animals (e.g., reptiles and non-human primates) that are used as service animals is problematic.” (Guidelines for Environmental Infection Control in Health-Care Facilities (2003), at 110). For a discussion of the zoonotic diseases and their significance in allowing animals into healthcare facilities, see my chapter, Animals in Healthcare Facilities, in Service and Therapy Dogs in American Society: Science, Law and the Evolution of Canine Caregivers (2010).

The federal district court analyzed the situation in light of the Americans with Disabilities Act. The ADA requirements apply in Title II to public entities, which include the county health department, and Title III to places of public accommodation, which includes Wal-Mart and Cox Health Systems. The court first determined that Ms. Rose had not established that she was disabled under the ADA. Other than her “own broad and self-serving statements, the evidentiary record offers no support for Plaintiff’s claimed limitations.” The court found the facts of Ms. Rose’s history actually indicated that she functioned quite well in society. The medical testimony supporting Ms. Rose’s diagnosis of agoraphobia and anxiety disorder came from a physician who seemed to accept his patient’s self-diagnosis. The court concluded that the evidence did “not support a determination that Plaintiff’s disorders limit any of her major life activities, and accordingly, Plaintiff is not disabled under the ADA.”

The court also concluded that the monkey was not a service animal. The court cited the definition of service animal in 28 CFR 36.104. The court noted that the “vast majority” of the tasks performed by the monkey “involve nothing more than the monkey providing comfort.” The court argued that an animal “that simply provides comfort or reassurance is equivalent to a household pet, and does not qualify as a service animal under the ADA.” The court felt that Ms. Rose had failed to explain how the monkey’s tasks relate to her disability. Even where the tasks might relate to the disability, such as keeping people away with an open mouth or a gentle push, “Plaintiff provides no explanation as to the monkey’s training or the specific cues that would trigger the monkey to perform these ‘tasks.’” Although some service animal functions, say an alert of a seizure-alert dog to an oncoming seizure, may begin without formal training, handlers often reinforce this behavior with training. It is not clear whether the court would have accepted the function of the animal had there been more evidence of training procedures.

The court then made an observation about the monkey’s actions in keeping people away in agoraphobic settings: “Such actions are aggressive actions, particularly for a primate, and without evidence of specific training or cues indicating the monkey only performs these tasks in situations where Plaintiff's disorders may require it, these ‘tasks’ may feed into the health and safety concerns Defendants have raised regarding having a primate present in food service establishments.” This would appear to be a variation on the balancing of rights that enters into service animal questions at a number of points. Thus, a guide dog that is aggressive and out of control can be excluded from a public accommodation even if it is trained as a guide dog and works as such most of the time. More on point is the case of Storms v. Fred Meyer Stores, Inc., 120 P.3d 126 (Wash.App. 1 Div. 2005), where a dog put itself between the handler and others in public paces to keep an open space to help reduce the anxiety of a veteran suffering from post-traumatic stress disorder. The Washington appellate court found this function sufficient to label the dog a service dog and allow the matter to go to trial. With the dog, however, the risk of interspecies disease transmission is considerably less.

The court was also impressed with the fact that the defendants did not make knee-jerk decisions in denying access to Ms. Rose with her monkey, and accepted that the health and safety concerns of the defendants were valid.

The case demonstrates that establishing that an animal is a psychiatric service animal is not always easy. The physician’s testimony appeared to be tepid at best, and the court’s description seems to be that of a doctor uncertain of his own diagnosis and unfamiliar with what service animals can do. As a matter of trial strategy, Ms. Rose perhaps should have considered getting the expert opinion of a psychologist or psychiatrist. A diagnosis stating that she could not function in various public environments without the assistance of the monkey would have helped considerably. Since this was a motion for summary judgment, which was granted, it is possible that the defendants would themselves have insisted at trial on an analysis by an independent psychologist or psychiatrist.

People’s preferences in animals deserve some consideration. Take, for instance, the use of guide horses by individuals who prefer miniature horses to dogs. Nevertheless, primates pose a somewhat unique problem because of their evolutionary closeness to humans as a higher primate. That closeness comes with a much higher risk of disease transmission than applies to dogs (or cats or miniature horses). Because this factor was one that the court (properly) put into the balance, Ms. Rose might have succeeded had her psychiatric service animal been a dog.

Was the decision correct? On the facts, yes, in my opinion. Other facts should have changed the result. The Department of Justice proposed updates to its service animal rules in 2008 emphasized that service animals can do work or perform tasks for individuals with psychiatric, cognitive, or mental disabilities. DOJ also said that it has “always required that service animals be individually trained to do work or perform tasks for the benefit of an individual with a disability, but has never imposed any type of formal training requirements or certification process.” Thus, Ms. Rose should have been able, had the matter gone to trial, that she had individually trained her monkey. DOJ noted that “a psychiatric service dog can help some individuals with dissociative identity disorder to remain grounded in time or place.”

DOJ’s proposals also distinguish emotional support from psychiatric service, and exclude the former while including the latter under the coverage of the ADA. It is not clear to me that the federal district court made this distinction, but given the facts this was probably not necessary. DOJ states: “The difference between an emotional support animal and a legitimate psychiatric service animal is the service that is provided (i.e., the actual work or task performed by the service animal).” In the case of Ms. Rose, the court clearly believed that she was trying to make a pet into a service animal, but it must be hoped that had the matter withstood the motion for dismissal, a trial would have considered that the monkey might have been providing psychiatric service beyond mere comfort. This would have required, however, that Ms. Rose establish a psychiatric condition, something she failed to do for purposes of the motion. It is also to be noted that DOJ’s regulatory proposal would exclude “nonhuman primates born in captivity” from the definition of service animal. Proposed 28 CFR 36.104. 73 Fed. Reg. 34508-34557. Consequently, if the regulations are finalized in the current form, Ms. Rose would not be helped by them, though someone with a similarly acting service dog could be.

Can a macaque be a psychiatric service animal? Yes. I believe that Ms. Rose failed to establish a psychiatric condition. Therefore, I believe that she failed to establish that the monkey was doing work or performing tasks for an individual with a disability covered by the ADA. I believe that she chose an animal with a greater threat of zoonotic disease transmission than would be the case with a dog or other non-primate providing similar benefits. It was because the animal was a primate that the county health department got involved. The letters of that department alerted the other entities about the disease transmission potential of the monkey. That probably would not have happened with a dog. Had she established a psychiatric condition, and established that the animal did work or performed tasks that alleviated that condition, I believe she should have been given the opportunity to establish that the macaque was adequately inoculated against relevant transmissible diseases and that it was sufficiently trained not to be a physical threat to people with whom it might come in contact. The disease issue does, however, present a higher hurdle for a primate than for other species with service animal functions.

Rose v. Springfield-Greene County Health Department, 668 F.Supp.2d 1206 (W.D. Mo. 2009). Ms. Rose has also had problems with keeping her orphanage on her farm, as described by an article posted on the website of a local radio station, KSPR.