The Smell of Fear

I think Cecil’s column on an animal’s ability to smell fear was absolutely correct, but I also think the subject could be looked at a little broader. Once, our dog trainer (okay, he was a dog psychologist; we lived in L.A. at the time) said that dogs can hear much better than we and can hear things like faster breathing and even an accelerated heart beat. Also, he said they can smell an increase in sweat coming from you and, like the canine version of a wine connoisseur, smell if adrenaline was one of the components that caused the sweat. Well, writing this down, it suddenly looks a lot more crystals-and-past-life-regression crackpotish than I had previously thought. Still, we did have a dog with very bad eyesight who would always growl when walking past a schizophrenic street person. Maybe it was the tin foil hats, but these people seemed to give off something that the dog did not sense from the rest of the shopping-cart set.
So, while animals may not be able to smell some actual fear-type ooze from special fear pores, they might be able to sense other things that alert their instincts that this person is not a normal part of their pack.

Cecil was a little snappish on the response here. I think the question I’d really like to hear the answer to is, “Can animals SENSE fear?”
I believe the answer is yes, but I’m sure there’s more to it that Cecil could reveal to us if he took the time. Do dogs (or other animals) cue off posture, scent, sound, or other behavior?

This really does not apply directly to the question, but I found Cecil’s evolutionary argument to be quite weak. To argue that people must not emit a noticeable scent when afraid because predators would have weeded out that population group is like saying that people must never emit emit any noticeable odors because predators would have weeded them out. I don’t know about anyone else, but after a couple hours at the gym I know predators (and everyone else) can identify me by scent. Some genes make the cut even if one of their side effects (odor) is less than ideal.

The best lack all conviction
The worst are full of passionate intensity.

Unca Cece wasn’t saying that an animal can’t be smelled by predators (obviously a silly notion). What our dear Uncle was saying (I think) is that it makes no evolutionary sense for an animal to exist that increases its chance of being smelled out by a predator whenever it notices the predator and becomes fearful.

A somewhat more accurate attack on this notion (I think) is to question the assumption inherent in it that being noticed while afraid increases an animal’s chance of being killed by a predator. This, it would seem to me, depends on the animal. Animals that escape predation by hiding probably do better if they can’t be smelled. But herd animals, other predators, and, probably, humans, along with any other animal that escapes predation through methods that make being noticed irrelevant could stink up the place with fear, without it making much difference.

That attack works fine, too. My point, however, is that we are rarely justified in looking at a trait and deciding that “evolution obviously would have eliminated it from teh human species.” Genetics is rarely so straightforward, and the number of factors which may have influenced selection is extremely large. All we can really do is look at traits that do survive and tell interesting (and perhaps acurate) stories about why/how they provided a competetive advantage.
I can think of very few cases where we are justified in denying the existence of a trait simply because natural selection would remove it from the gene pool.

The best lack all conviction
The worst are full of passionate intensity.

Most examples of animals sensing fear are of dogs smelling fear in humans. Extending this discussion beyond predators and prey, many animals have a status heirachy (“pecking order”) based on perceived strength.
During confrontations such animals often bluff, making lots of agressive noises. Being able to sense fear by scent and/or subtle cues would be a useful ability, especially if it allowed you to chase off the alpha male and monopolise all the females without a fight. Conversely, being able to CONCEAL fear would also be useful.
Perhaps domesticated dogs can smell human fear because our own poor sense of smell has prevented us from evolving to conceal it. This doesn’t matter in confrontations with humans, but we can’t bluff dogs if they want to dispute status or territory.

Now in history that we are dominant - individual variants that give off more smell may not be weeded out. Add in agriculture where humans don’t need to stalk prey.

We may begetting smellier in evolutionary terms.

But in looking at evolution - skip the theory - go to the field. Darwin did with much success. Horse breeders don’t dispute micro-evolution - they bank on it!

Hunters know that you must be down wind from a deer to avoid detection. While we may have evolved to smell less previous to now (I doubt it) the deer have evolved to detect better - that’s why they are still here (for the moment).

Many animals process more bits per second (yes this can be calculated, see “The User Illusion” - look it up on Amazon) than humans from each of their sensory inputs. We seem to rank high in visual, but a few animals do more than us in that area. We are relatively poor in smell.

Fitness does not exist in a vacuum, of course. Predators evolve to better sense their prey, then prey evolve to better hide from predators, and back around again. Rather than looking at a static snapshot you need to look at it as a constantly adjusting system. Predators may well have evolved the ability to perceive fear in prey. Over time, the prey that presented these fear cues most strongly got killed off more often, and the ones that presented them less strongly were more likely to pass their genes on to the next generation. The predator might then develop the ability to sense other cues and associate them with fear. The prey then evolves to minimize this… and so on. Evolution is a constant game of catch up.

Dogs smell that big load that some people drop in their pants when they’re scared. They smell what should visually cue a small brown (or grey, if they are actually color-blind) mass on the ground, but what they actually see is an upright figure walking around - that smells like a turd - and it freaks out their sense of reality and order. They must destroy the disorder before their master sees it, lest they get their noses rubbed in it, which is why they attack. They smell ‘fear’, all right…
Link to Do Animals Smell Fear column.

This can be dismissed as anecdotal, but, since we haven’t cited any veterinary biology or medicine regarding animal reactions to specific scents, I’m going to vote for body language.

Notice the original column: the big dog “lunged,” the child reacted in terror, then the dog got excited and began barking.

We used to have a boxer with the world’s sweetest disposition. (She once turned around and licked the face of a 2-year-old who had just dragged her head out of a water dish by the ears on a 93° day.)

The only times I ever saw her look fierce or actually growl at a person was when she encountered someone who displayed fear. If a person locked their knees together, turned their body away, held their arms up to their shoulders, and made “eeeww, ooooh” sounds, her hackles would raise, she would begin to growl, and if we couldn’t get the person to relax, she would begin barking ferociously. I have also seen her wriggle up to a person whom she was “attacking” moments earlier and butt them with her head to encourage petting if we could get them to “act” less afraid.

This does not deny the possibility of associated scents, but I’m going with body language until I see pheromonic evidence.


I have to say that Cecil’s argument against the evolution of the fear-stink is pretty lame here. The fact that it would be stupid or unlikely for that to evolve is hardly a proof that it didn’t. Using that logic, you could argue that we don’t breathe and eat through the same hole, since that contributes to choking.

I’ll go along with body language usually being why a dog–or any animal–can tell when someone is alarmed; we give off all sorts of visual clues. That, however, does not explain why it also works for dogs like mine that were pretty darn close to being blind. She seemed to sense abnormalities in people she most certainly could not have seen. Also, I don’t think I’m ready to subscribe to the load-o-dump theory either.