This is Clea, an older dog, an alpha Akita. She’s asleep on her dining room rug. On the floor beside her head lies a small plush toy, brushing her cheek. It is a whimsical dragon which belonged to Alexei, her brother, litter-mate, lieutenant and inseparable companion in life, who died about a year and a half ago.
It may seem obvious what is going in the picture, but it is not. Clea cannot share her mind with me. I must be careful not to presume too much about what she is thinking, careful not to make connections between the living Clea and the dead Alexei that may not be in her mind, but only in mine.
That would be projection and unwarranted anthropomorphization. Those are bad. Then, again, so is denial. In any case, there is nobody whom Clea can tell what she feels. In this post, I argue that we should listen anyway.
Can we say with confidence that a dog thinks at all?
Carl Jung, the great psychologist, disappoints me in the following anecdote told by A.I. Allenby (in William McGuire and R.F.C. Hull, C.G. Jung Speaking, Princeton, 1977, p. 157).
Some time ago I went for a swim and then lay on the balcony of the boathouse to sun myself. The level of the lake was so high that the boathouse was surrounded by water. There came my dog in search of me. He could not see me, and was not sure whether I was there. Being of a somewhat cowardly disposition and not very fond of the wet, the dog first put one paw into the water, then withdrew it, and then another paw and withdrew it, too. And this went on for some time. Eventually I made the faintest little noise, and the dog shot through the water and up the steps of the boathouse in one jump.
So far, so good. Here’s what disappoints me, Jung’s comment on the dog’s mental state between meeting the trail’s end and learning where Jung was waiting, safe and sound:
The dog is conditioned by instinct and has no will-power of his own, except when a little noise from his master releases it.
Suppose it had been the other way around. Suppose Jung had followed his dog’s paw-steps to the edge of unusually high water, and could find no sign there of the beloved companion who has clearly entered, but who seems neither to have left the lake nor to be still swimming. Would Jung not feel concern for his friend? Would he not linger? Might he not head off heedless of getting a little wet toward the sound of a familiar bark? Would he lack “will-power” because he responded that way?
Ah, but turning the tables like that may be illegitimate. The dog and the man are different species, and to speak of someone who isn’t human using terms peculiar to human cognition is the dreaded fallacy of anthropomorphization. The name even sounds bad, as if the we were shipping dogs to the Island of Doctor Moreau and its House of Pain.
However, hypothetical table-turning is also the application of an attractive heuristic for uncertain reasoning. Arguably, we should draw similar inferences in similar cases, and in different cases, draw different inferences according to the cases’ relevant differences. There is nothing suspect about comparing what the dog did with what Jung might do, according to that heuristic. We just need to remember all the cautions we always ought to remember about the limits of inference.
Inferring purpose from mute behavior
If interior mental states in dogs aren’t taken for granted, then what behavior counts as evidence for them? Objectively speaking, Jung’s dog walked where Jung had recently walked, stopped at the water’s edge, and then charged through the water to a freshly disclosed destination.
Imputing concern for the missing Jung is conjectural. Tracing Jung’s footsteps isn’t necessarily looking for him. Animals move, that’s what they do. Why not walk to the lake? If Jung took the same route, maybe it’s an easy path. Why not stop at the water’s edge? And when a familiar companion cropped up, why not proceed directly through the shallow water, now that there is someplace specific to go?
Plus, I wasn’t there. Jung was there, and he was unimpressed by his dog’s behavior. Maybe the dog was just doing what it had done other times, when Jung was safely accounted for. Maybe on some occasion the dog had left Jung in his study, went outside to stretch his legs by the lake, and some shore bird inspired the final sprint from water’s edge to boathouse deck. “Similar cases” can cut both ways.
With Clea and the dragon, I was there. Still, all that is happening objectively is that Clea is sleeping with a plush toy. Maybe she likes toys. Her brother obviously isn’t using his anymore, nor will he object if she uses it. Why even think there is anything more to the situation than that?
A parallel inter-specific inferential problem is the evaluation of Neanderthal “burials” by archeologists. The state of that controversy was recently surveyed in Science (News and Analysis, September 22, 2012; 337: 6101, pp. 1443-4 ), and perhaps for a few more weeks the article will be available from
Briefly, some few Neanderthal skeletons appear to most scholars as if careful deliberation preceded the placement of the body. It is hard to tell, however, whether there really was an intention to do more than to dispose of the body hygienically, which can be explained on survival grounds, in whatever way local conditions made expeditious. There is no regular common feature across the different burials, nothing that hints of any widespread set ritual.
In Clea’s case, she is not doing something we routinely see other dogs do. That there even is a plush toy available to her is quirky and unsystematic – her “local conditions.” Shortly before Alexei died, the dogs’ housemates bought a toy for each of them, the dragon and a dinosaur. Alpha Clea chose the dinosaur, and thus it was, without question, hers. She didn’t take both. So, Alexei did get to play with the dragon, as his toy, in those last days before he died.
So, care and deliberation in Clea’s embrace of Alexei’s dragon is not supported by its similarity with what other dogs systematically do. Where we might find support is in the difference within Clea’s behavioral repertoire, whether the dragon is treated differently from her own plush toy, the dinosaur.
In the photo, we can see that the difference in the two toys is dramatic. The dragon is in mint condition. Even six months earlier, the dinosaur was more a rag than a plush toy. That Clea treats the two objects differently is, in my opinion, indisputable.
It is not that Clea refrains from interaction with the dragon. She moves it around in the house, and of course, uses her teeth to hold it when she moves it. She also cleans the dragon. That is, she licks it, and when she does, the surface is wet (confirmed by my touch afterward). Licking is typical behavior for alpha-dogs, so I am unsure of Clea’s exact intention in licking this object. I am confident, however, that the licking is intentional, and that it has the effect of contributing to the clean appearance of the dragon, whether that is part of her intention or not.
I feel very comfortable, then, in my inference that Clea is acting purposefully to preserve the dragon for her gentle use, and has been doing so continually since Alexei died. I also infer that the difference in her treatment of the dragon compared with the dinosaur is because the dragon was Alexei’s.
Inferring specific intention from purposeful action
That opens the question of what the specific meaning being close to Alexei’s dragon holds for Clea.
Clea’s memorial behavior touches on a sphere where human attitudes can be especially complicated. Some humans’ behavior toward their dead reflects elaborate ideas about what happens to the personality of the deceased, with a wide variation among people in such beliefs. I have no reliable idea what Clea believes or how her mental states might be rendered into words. If I said that I did, then I would be liable to charges of projecting my mentality onto hers, an unwarranted anthropomorphization. So, rest assured that I won’t say that here.
If a human being exhibits similar overt behavior to Clea’s without holding complicated beliefs about the aftermath of death, then Clea’s underlying mental state needn’t be elaborate for her behavior to be purposeful like some human behavior. Behavior like Clea’s can be found among human non-believers.
The following testimony is from a recent post on the JREF forum, an atheist-skeptic hangout on the web. The poster was responding to another member’s report of an olfactory encounter with a departed in-law’s ghost. The poster doesn’t believe in ghosts, but her behavior is much like Clea’s:
I had a favorite dog who slept on a particular bed for many years. After she died, I would occasionally comfort myself by burying my face in her bed and breathing warm moist air into it until it released some of her familiar doggie odor. Same thing.
Hypothesizing complicated belief, then, is not required to characterize Clea’s behavior as her having a continuing relationship with her dead brother, mediated by her memory when she is close by things plausibly associated with him. That is good news in that it distances the characterization from the curse of unwarranted anthropomorphization. A similar argument could be made about the Neanderthal burial controversy, or about whether Jung unjustly estimated his dog’s psyche.
It’s also bad news, since we are also distanced from any possibility of learning what Clea is thinking about Alexei by observing her behavior toward his dragon. I would sorely like to know that, but I am grateful for the justified confidence that there is something to know.
Update: Clea died on December 2, 2013, peacefully at her home, surrounded by those she loved. A eulogy is here.