Funny, how old words morph their meanings, innit? Take “hangdog,” which has been around since the 1670s, adjective & noun, originally meaning “contemptible & sneaking.” [Think Dickens’ passive-aggressive character, Uriah Heep, always presenting himself as ‘umble, while surreptitiously scheming to bring the high & mighty down.] By 2010, The American Heritage Dictionary defined a “hangdog expression” as “looking shamefaced & guilty.”
Consider the word “guilt,” even. As late as 1934, the only definition in The Concise Oxford Dictionary was “culpability.” Guilt wasn’t a psychological construct. None of your subjective, self-referential, conceptual feeling [as in “survivor guilt,” or “Jewish/Catholic/Protestant guilt”]. Just the objective fact of the case: “How does the defendant plead? Guilty or not guilty?” Also, “How does the jury find: guilty or not guilty?” [Their verdict is a subjective opinion, but it’s presumably based on the objective, admissible facts presented.] The notion of remorse doesn’t come into it, until the sentencing phase of the trial, if the erstwhile “not guilty” defendant is found “guilty” [at which point, his lawyer advises him to show how sorry he is by adopting a hangdog expression].
So, here we are in the 21st century, with the burgeoning field of Social Neuroscience and its ugly Iron Maiden, the fMRI [colloquially referred to in the media as The Brain Scanner], claiming to have located the Seat of Guilt in the Brain, no less! Point of order, would that be the seat of objective or subjective guilt, they’ve found? Let me not bore you with my observations on the flawed research designs of such studies [like, having a subject read or watch scenarios of other people behaving badly, in order to light up the Guilt center(s) in the subject’s own brain… What? When I watch Othello snuff Desdamona, I’m the one who feels/is guilty?]. Let me instead quote the brain-imager, Dr. Gregory Miller of the University of Illinois: “Functions do not have a location. Decisions, feelings, perceptions, delusions, memories do not have a spatial location. We image brain events… We do not image, and cannot localise in space, psychological constructs.”
At most, then, the fMRI is currently no better than the 20th Century polygraph at measuring physiological changes correlated with limbic system changes correlated with psychological constructs, such as fear and humiliation. From which I know, having treated several Intelligence Officers who had failed the annual polygraph test because of an exaggerated sense of guilt, over sexual peccadillos, rather than because they were actually guilty of breaching national security. So dedicated were they to their Intel work [from which they were sidelined by the failed polygraph test], that some of them would ask [semi-jokingly], “Is there such a thing as a ‘guilt-ectomy’ that I could have, just so I could pass the polygraph?” Just give those brain-imagers a chance, and they’ll be in there before you can say “knife.”
Back to the title question, addressed on a pet behavior blog, in the form, “Do dogs feel guilt?” Their answer was, “No. Guilt is an abstract concept. Dogs express fear & submission, in response to the owner’s anger, which they sense through body odor, glaring eyes, stance and tone of voice.” The dog “looks hangdog” to avoid or lessen the Alpha Dog’s punishment [just like a defendant who has been “found guilty”].
The wolf looks “sheepish” in the presence of a more powerful wolf. It has correctly read the power subtext of the situation. It endures the humiliation of acting submissive, to avoid the pain & suffering of being put in its place [which might be completely beyond the pale, where chances of survival are slim] by the Big Dog. That’s more useful than guilt. That’s Social Intelligence.