I'm going to talk about dogs, for the very important and scholarly reason that I am allergic to cats.

The origins of dogs as "just pets", or companion animals, is...I don't want to say debated, because it's not really, but it's more that it depends on what criteria scholars want to use. In the West, Victorian, 18th century, Renaissance, Rome, and prehistoric are probably the most popular.

To give you an idea of some of the difficulty of identifying what it means to have "just a pet", I'll talk about a couple of examples.

Scholars would all agree that by the 20th century, we're firmly in companion animal territory. However, from 1914 to 1962, the most popular toy dog breed (a reasonable proxy for dog-as-pet) in the UK at that point was the Pekingese. Sarah Cheang argues that the immense popularity of the Pekingese was the result of colonialist nostalgia in England, with a nice twist of Orientalism for 'mystical Old China.' She quotes a 1931 "history" of the Pekingese:

How many people realize, as they walk abroad with that most uncannily clever four-footed friend, that they are in the company of one of an imperial race of the loftiest lineage, a race that for hundreds of years was never set eyes on by any but the most privileged persons of the royal court, a race that less than a hundred years ago was shrouded in mystery, the subject of endless speculation by Chinese and foreigners alike?...We of a modern civilization may be unaware of this, but the proud little Pekingese never forgets.

So is this floofy little lapdog really "just" a pet? Or is it playing some kind of symbolic role? Does that count as a "job"?

Another popular era for the onset of dog as "pet" is the Renaissance. Some solid evidence here is the wild proliferation at this time of toy and miniature dog breeds in Europe. Lapdogs for women are particularly noted and illustrated, and French and other royalty from the 16th century on became particularly linked with the Bichon breeds. But as John Friedman argued, collars and other clothing/accessories for Renaissance-era dogs served as an important sign of the way that dogs helped family construct a self-identity. Is that a "job" for a dog to have?

Similarly, Kathleen Meikle suggests that in the Middle Ages, dogs were a sign of what she calls "ostentation"--basically "We're rich enough to waste food on this dog." (Has she ever had a dog? Mine would probably be A-OK eating whatever.) That's certainly reflected by how Chaucer describes the Prioress and her two little dogs in Canterbury Tales. But on the flip side, we read in the 13th century Ancrenne Wisse, an idealized prescription for the religious life of anchoresses (religious women who vowed to spend the rest of their lives in a single room/set of rooms attached to a church, praying and reading and hearing Mass) that they may not have dogs, but they may have a cat specifically to act as a mouser. The need to specify suggests that people were keeping animals for reasons that had nothing to do with rodent pests--but an anchoress would also have no need of a cat or dog for ostentation.

So I want to finish by mentioning one of my favorite pieces of prehistoric archaeological evidence for dogs as companion animals, at a 12,000 year old site in the upper Jordan River valley (modern-day Israel). The grave contains the skeletons of a human and a canine. The man is clearly elderly--not the kind to be out hunting or herding. The dog is a puppy, four to five months old. Not yet of much use around the household They are buried so that the man is curled up as if lying down in bed or relaxing. His hand is extended and rests on the dog--one wants to say, petting it.

Obviously we don't have the full story here. We don't know he was buried "petting" the puppy; we don't know if the puppy was his. But if those interpretations are correct, that points to pretty solid evidence for a human-animal connection that had little chance of involving more than companionship--he wasn't likely to need much of a working dog, and there was apparently no reason for the pupper by other members of the community once its probable-owner had passed.

So, I would say, the evidence for dogs as only pets in all these eras is ambivalent--dependent more on the criteria we pick than on reconstructable human-pupper relationships. One thing is definitely for sure, though--working breeds or not, humans and dogs have shared a special bond since time immemorial. Medieval hunting and veterinary texts from the Islamic world castigate their readers for using the best silks on dog clothing, and for having dogs jump on their owners' furniture and sleep in their beds.

Not ideal, perhaps, for an orderly household or for the fiercest hunters; wonderful and fitting if your dog is/is also your pet.

Hellenistic tombstones and other evidence from ancient Greece and Rome suggest that the wealthy at least were able to keep animals as pets.

One of my favorite tombstones (which you can see on display at the Getty Villa) depicts a girl named Apollonia gently petting a dove. The tenderness with which she is interacting with the animal implies that it is her pet. The image of children, particularly little girls, with pet doves is nothing new...a 5th century stele shows a girl cuddling with one dove while holding another on her finger. Children's pets were often included on gravestones, especially in the Hellenistic era. Art of that era focused on the delicate and precious, and the gravestones include these pets to reflect the innocence of the children who had died.

Another of my favorites is a stele for the boy Moschion. It depicts him playing with a little dog (and seemingly offering it a bird to eat, which seems cruel, but perhaps this was the ancient equivalent to Milk Bones?). Could he have been training this dog for a useful purpose, rather than just having it as a pet? Perhaps. But I do think it's important that he was so attached to it that it was depicted with him on this tombstone, much the same way people now form attachments to their pets.

Moving onto ancient Rome, Petronius' Satyricon comes to mind when speaking of pet dogs. At a dinner party with the absurdly rich Trimalchio, the main character Encolpius observes a boy, Croesus, with a fat puppy named Pearl. Poor Pearl nearly gets torn apart by a big guard dog named Scylax, and the implication is that he has little use other than the amusement of the boy. It's just one of many illustrations that Trimalchio has more money than he knows what to do with; he can feed not only an army of guests and household slaves but also their pets.

And, of course, every intermediate Latin student who's read Catullus knows about Lesbia's sparrow. While the sparrow is undoubtedly a sexual innuendo, it does remind us that Roman ladies kept birds as pets.

Overall, I'd say there are two factors at work. The first is the attachment to pets that we still feel. The second is undoubtedly status display; someone who has a pet is able to feed both themselves and an otherwise useless animal without a problem.

Anthropology might provide some insight. The Yanamamo in the Orinoco Basin would keep a variety of pets. They had dogs, reptiles, birds. When missionaries brought them chickens they would eat the eggs but not the chickens themselves because any animals that lived in the village were considered part of the family and not to be eaten. These are people that were using stone tools up until a few generations ago. I think pets could be as old as language.

Also, when did people start treating pets as offsprings instead of pets? (Serious question)