Anthropomorphizing Dogs


My son-in-law, Todd, has a 2-year-old mutt named Rico. Rico may be about half Labrador retriever and half German shepherd, but he may also be a Heinz hound (57 varieties). I often tease Todd and Rachel about the way they anthropomorphize their dog.

The other day, I went to their house, and Todd was hanging clothes in the back yard. I sat on the stoop. Rico promptly came over and dropped a tennis ball at my feet. He plainly wanted me to throw it and (I thought) pantomimed turning away from me and running.

I told Todd that I had just read in Science about another dog named Rico. That dog supposedly had a vocabulary of 200 words and would fetch various toys on command from the owner. Without turning from his clothesline, Todd said, “Rico, why don’t you get the rope?”

Rico grabbed the tennis ball he had been tempting me with. Todd said, “No, Rico, get the rope.” Rico immediately got a short, knotted length of rope and began gnawing on it. He did not bring it to Todd.

My dog-in-law Rico has a vocabulary of maybe 40 words. Many are commands: sit, stay, fetch, no bark, you’re free, go outside, load up (get into the car), and more. Others are (to us) nouns: rope, ball, toy. It is possible that Rico interprets these as commands to fetch the items. He clearly interprets kennel, house, and bed, for example, as commands, not nouns. So maybe Rico understands only commands.

I am not about to claim that either of the Ricos understands English or German. They can, however, learn words. Learning words is not learning language, but it is a necessary step, possibly a first step.

Rico in the Science article learned some of his words immediately and by inference, a method the researchers called fast mapping. Briefly, Rico was sent to find a toy he had never heard of. In each of several trials, he found an unknown toy among familiar toys and brought it back with good consistency. A few weeks later, he remembered the names of some of those toys (or the commands to fetch those toys; take your pick). According to the Science article, that is also how children learn new words.

Copernicus and Galileo famously told us that the earth is not at the center of the universe. Evolutionary biology is telling us that we are not unique, and even our vaunted language ability may be only one end of a continuum. The Ricos’ ability to learn words may not be homologous with ours, but it certainly suggests that other animals also have linguistic skills.

I have been wrong to tease Todd and Rachel about anthropomorphizing their dog.

I should have been animalizing humans.


The original Science article is Juliane Kaminski, Josep Call, and Julia Fischer, “Word Learning in a Domestic Dog: Evidence for Fast Mapping,” Science, Vol. 384, pp. 1682-1683, 11 June 2004. See also the Perspectives article, Paul Bloom, “Can a Dog Learn a Word?” Science, Vol. 384, pp. 1605-1606, 11 June 2004.

For press coverage, see James Gorman, “Finally, an Old Dog Can Learn New Tricks,” New York Times, 11 June 2004,[…]e/11dog.html; anonymous, “Dog Prodigy Gives New Meaning to Language,” National Public Radio, 10 June 2004,[…]1952976.html; and anonymous, “Collie Dog’s Word Power Impresses,” BBC News, 11 June 2004,[…]/3794079.stm.

For further information on language in the great apes, see[…]rumbaugh.htm.

For further information on language in parrots, see

For a cry of outrage, see Geoffrey K. Pullum, “So Now It’s Dogs that Understand Language (Sigh),” 10 June 2004, http://itre.cis.upenn/~myl/language[…]39.html#more.


Nice post. I generally try to give animals the benefit of the doubt on cognitive matters, but Rico still surprised me. I’ve read about dolphins learning to generate a novel behavior on command, but I didn’t expect a dog to be able to come up with “I need to bring back the novel object.” Clever border collies.

Your Pullum link is broken, incidentally…should be[…]39.html#more

Some years ago I saw an article about a dog that won a “Dog of The Year” award from some organization. He responded to 180+ commands or vocalizations, walked his family’s child to school each morning, on the way back home retrieved anything the kid had dropped, and went back in the afternoon and walked the kid back home again. I think he was a sheperd mix, but I’m not sure.

Also saw, on a Carson show I think, a dog who’s routine was an absolute mind f##k. He responded to involved commands like,”Go to the cabinet, open the bottom drawer, and bring the folder to Johnnie.” All sorts of potential for flim-flam here but …

Bob Maurus: He responded to involved commands like, “Go to the cabinet, open the bottom drawer, and bring the folder to Johnnie.” All sorts of potential for flim-flam here but …

Yeah, some of those complex “commands” are learned by rote off of a few key words and gestures (for instance, to make sure the dog brings the folder to Carson, the trainer might tap Carson on the shoulder while issuing the command, and the dog registers that as part of the command – still impressive, but not as much to do with actual language). There is a little bit of trickery involved in those really extraordinary feats when it comes to dogs. (I saw a special about training them a few years back.) The impressive stuff is when they can be trained to sniff out cancer (I’ve seen them do it!) and predict epileptic seizures (I know an epileptic who wants to get one of those specially trained dogs).

Even more impressive, I think, are the great apes. Just a few nights ago, I was watching a Jane Goodall special on Animal Planet, and they showed a chimpanzee who had a several hundred word vocabulary which seemed to include a knowledge of syntax, and used none of those tricks I mentioned with the dogs. The chimp’s trainer sat cross-legged, her arms unmoving at her sides, and – get this – wearing a welder’s mask so the chimp couldn’t even see her expressions or read her lips. The chimp was able to comprehend and comply with complex commands, such as “put the key in the refrigerator” or “pick up the red ball.” Just fantastic. I’m a big fan of our closest genetic relatives.

I worked for a year or so in Irene Pepperberg’s lab (while she was still at U. Arizona), and those parrots were simply astonishing. If you show Alex a tray with, say, three red blocks, two green blocks, four red cotton balls, and five green cotton balls, then ask him, “How many green blocks?”, he’ll know. He (and the other, less famous parrots too) can assign multiple labels to an object. Now, I don’t know much about cognition, but that stuff was just freaky.

I loved this article..I have a 2 year old Dobie and she has been able to bring me her toys by name since she was a pup..And needless to say she a lot of toys..When I get her a new one she knows its name by the end of the day and can bring it to me when I ask..She is truely a special little lady..

Hi matt. I’ve read the Science article that all of this revolves around, and in response to pullum’s ‘cry of outrage’ that you link to, what is so exciting about this research is that rico appears to be doing MORE than simply associating sounds to objects-he appears to show the abillity, if given a word he has never heard before, then sent to fetch any of a number of objects, including one which he has never been exposed to before, to reason that the novel word must be a request for the ‘new’ object. In other words, he appears to show ‘reason by exclusion’, and can often still remember the new word, after a single trial, three weeks later. Now, this doesn’t necessarilly indicate that he understands language;he may only associate all of this with fectching(though some intruiging annectdotal evidence suggests otherwise), but if this holds up, with him showing these same abillities in response to other people’s commands, the this certainly shows some amazingly complex thinking for a lil’ old doggie-more than mere associative conditioning!


It is possible that Rico interprets these as commands to fetch the items. He clearly interprets kennel, house, and bed, for example, as commands, not nouns. So maybe Rico understands only commands.

Or maybe we only understand his responses when he interprets them as commands.

About this Entry

This page contains a single entry by Matt Young published on June 15, 2004 5:31 PM.

Icons of ID: No Free Lunch Theorems was the previous entry in this blog.

Three SH’s and one D is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.



Author Archives

Powered by Movable Type 4.381

Site Meter