The ability to recognize oneself in a mirror is frequently cited as evidence of sentience. A dog may be frightened or excited by its reflection, but either way it does not appear aware that the reflection is its own. The dog’s behavior suggests that if it recognizes the reflection at all, it believes it to be another dog, appearing through a window.
If the dog did recognize itself, how could we tell? One test, devised in 1970 by Gordon Gallup, involves applying a mark to the animal’s face without its knowledge. If it can recognize itself in the mirror, it will understand that the mark it sees is on its own face. Evidence of this would be surprise, or an effort to investigate the mark, perhaps by turning to view it at different angles, or touching the actual mark with a paw. Gallup found that chimps would readily learn to recognize themselves, but macaque monkeys would not, even after several weeks of exposure to a mirror.
Human children fail the mirror test at one year of age, but pass at two. Until recently, great apes (like chimps) and dolphins were the only animals besides humans who succeeded. Now, it seems African elephants must be added to the list.
Legal scholar Steven Wise has suggested that animals capable of passing sentience tests like the mirror test should be given basic rights, in recognition of their capability for self-awareness. (I’m on a legal theory binge right now and am about to start his book, Drawing The Line: Science and the Case for Animal Rights.) Wise apparently anticipated in his book that elephants would eventually pass the mirror test. However, it’s worth noting that there are intelligent animals that don’t pass, such as dogs and ravens. Some chimps don’t pass. Most gorillas don’t pass (although Koko did). In the elephant study, two of the three elephants tested failed to investigate the mark on their faces. Clearly there are different kinds of intelligence; for individual or species-specific causes, some animals may not find the mark (or the mirror) sufficiently stimulating to provoke a display of self-awareness.
For the record, I have tried the mirror test on my cat. She is very devious, and in order to commit deliberate deceit, one must know that there is a significant difference between one’s own knowledge pool and that of others – basically, you’re self-aware. So I was hopeful. My cat did not pass, although she can recognize me in a mirror and apparently knows what a mirror is (a flat reflective surface, not a window). However, cats are notoriously noncompliant with experimental protocols, so as far as I’m concerned, the jury’s still out.
Update: I just noticed that the October 30 New Yorker cover depicts a cartoon cat spitting at its reflection in a mirror. I can’t tell if the cat is genuinely afraid of its own reflection (failing the mirror test), or taking advantage of the mirror to practice its Halloween performance art (passing the mirror test). Either way, it’s timely.
I once had a dog that passed the mirror test (as far as I’m concerned). Until he was full grown, the mirror on my dresser agitated him when he would sleep on my bed at night, and he would bark or growl at it until he was sure the “other” dog wasn’t going to come any closer. Then, one day I had to pull the dresser out to repaint the wall. This new, if temporary, arrangement upset him, so I carried him over and we observed ourselves in the mirror for a minute. Then, we went behind the dresser, and he was clearly shocked to not find another dog and person standing there. After that, he watched my reflection when I spoke to him (on more than one occasion), and he stopped barking at the mirror at night.