Worldbuilding generates a lot of "Can I domesticate [insert real or fictional creature here]" questions.
This is intended to be a canonical version of that question. I'll self-answer, but please don't let that stop anyone from posting their own solution.
So:
What conditions are required for domestication?
I believe my answer to "How does a society domesticate the hippo?" meets your expectations, so I'll duplicate it here.
You are in luck. An experiment to better understand the domestication process was conducted on the Russian Red Fox. The project lead explained:
Belyayev believed that the key factor selected for in the domestication of dogs was not size or fertility, but behavior: specifically, tameability. Since behavior is rooted in biology, selecting for tameness and against aggression means selecting for physiological changes in the systems that govern the body's hormones and neurochemicals.
His process for achieving his goals was, very simply...
The least domesticated foxes, those that flee from experimenters or bite when stroked or handled, are assigned to Class III. Foxes in Class II let themselves be petted and handled but show no emotionally friendly response to experimenters. Foxes in Class I are friendly toward experimenters, wagging their tails and whining. In the sixth generation bred for tameness we had to add an even higher-scoring category. Members of Class IE, the "domesticated elite", are eager to establish human contact, whimpering to attract attention and sniffing and licking experimenters like dogs. They start displaying this kind of behavior before they are one month old. By the tenth generation, 18 percent of fox pups were elite; by the 20th, the figure had reached 35 percent. Today elite foxes make up 70 to 80 percent of our experimentally selected population.
I expect the same rules apply to Hippos ... assuming you can avoid the screaming and the yelling and the, well, death that might occur when you find Class III Hippos. Note that the experimenters had to work through 20 generations of foxes to just to achieve 35% domestication. with 6–8 years needed to reproduce a generation, that's 120–160 years to domesticate Hippos. Bring a sack lunch!
Conclusion to CodeMonkey's Post
The process of domestication is nearly identical for all creatures that can be domesticated (not all can): it's a process of breeding over and over and over the creature in question and for each generation, separating out those with desirable traits and continuing the breeding project with those creatures. Eventually, you get a domesticated species.
To be honest, animal husbandry and directed breeding have actually been around for thousands of years. Humanity's just getting around to writing decent documentation, though.
You didn't just ask about animals, so let's talk about plants.
Now, obviously, the simple answer to "Can I domesticate [plant]?" is "Yes, by selectively breeding it until it has the qualities you want." This is what humans did with wheat, corn (maize), and many other plants besides. Unlike animals, plants are not exactly in a position to run away or attack you, and breeding plants is generally much more straightforward than breeding animals. You also don't have to think about higher-order issues like "behavior" or "social dynamics," as plants lack those to start with.
However, that presumes we're talking about something set in a contemporary or futuristic setting, where society is already established, and where many problems can be solved by a simple application of time and money. In a less developed setting, this may be infeasible, especially if your characters do not know what "breeding" is in the first place, and are just domesticating things by accident (as humans did several times). So let's specifically focus on, not just whether domestication is possible, but whether it is likely to occur for a hypothetical preindustrial species of intelligent whatevers, even when they're not specifically trying to breed anything.
If a plant is to be (accidentally) domesticated, it must serve some purpose. The easy option is "you eat it," but plants can also be used for clothing, as building materials (wood, fibers, etc.), for decoration, or even for religious or spiritual purposes. The point is, there has to be some sort of reason for people to choose to grow more of (the wild version of) that plant. If the wild version's berries are toxic, for example, nobody is going to selectively breed it to make them edible. On the other hand, if they taste mediocre, but supply necessary nutrients, then that's a more reasonable starting point.
At first, anyway. In humanity, the transition to agriculture roughly coincided with the domestication of several species of plant, and critically, these were mostly plants that grew up quickly, largely grains such as wheat and barley. If a plant takes a very long time to grow up, it cannot plausibly be involved in our species' equivalent of the neolithic revolution, because the process of selectively breeding a long-lived plant is incompatible with a nomadic way of life (and a non-nomadic way of life probably requires some form of agriculture). On the other hand, selectively breeding a short-lived plant just requires remembering where you planted it and returning the following season.
Once our species is fully agricultural, however, this becomes much less important. It is absolutely possible to domesticate a long-lived plant once you have a stable, permanent population that stays in one place indefinitely. It just takes a long time.
Different plants use different reproductive strategies. Some of these strategies, such as wind-carried seeds, are difficult for humans (or our species) to intervene in or prevent. This will have the result that the plant reproduces out of control, and little or no artificial selection occurs. Nevertheless, a limited amount of selection may still be possible. For example, if our species regards the plant as a nuisance and tries to kill it, it may evolve to become harder to kill, or to become less annoying.
Jared Diamond has answered this question in Detail in "Guns, Germs and Steel".
He names 6 neccessary conditions for domestication.
Humans have domesticated a large number of diverse creatures - cats, dogs, pigs, horses, cattle, elephants, and killer whales, to name a few.
The factor that unites these disparate creatures is their social structures. They all live in relatively small groups, with complex social interactions. (even cats! look up feral cat colonies if you want to go down that rabbit hole.)
As an example, killer whales and elephants are known for being matriarchal societies, where the extended family looks to a mother or grandmother for guidance. This means the group is interacting in very complex ways to manage itself - which children stay with the group, and which break off to form their own family? At what age? What happens when an individual tries to "marry" into the group?
Compare this with a school of fish or a dazzle of zebra. The animals are grouping together for protection, but they are not necessarily communicating deeply with each other. They are not explicitly attempting to understand the other creatures' intentions or desires. They just react to signals that indicate danger, and ignore everything else.
The complex social interactions of some animals give humanity an "in" with the creature.
All of these social groups have a leader, and with domestication humanity breeds the creature to accept that a person can be the leader of the group.
So when it comes to the question of "can I domesticate [X]?" the answer is:
If it lives in a small(ish) group with a complex social heirachy, yes.
EDIT: I've got a couple comments that orcas and elephants aren't domesticated. I think its arguable, but I also don't think it matters - They are then CANDIDATES for domestication, because their communal behavior is similar to aurochs, wolves, and Eurpoean wild boar, who also lived in small(ish) groups of 5-30 with complex social behavior.
Generational persistence, an animal with some degree of compatible social signaling and behavior, and rational expectations.
You can eventually tame almost anything that you can successfully keep in captivity or convince to cohabitate to some degree of it's own accord.
The only actual pre-requisite is the ability to consistently differentiate desired behavior from undesired behavior, and the ability to breed them selectively.
It's not actually about whether or not an animal can be trained, or bred into an animal that can be trained - almost every organism on this planet can learn and adapt.
It's whether or not human beings are capable of reliably reading desired behaviors and selecting for them. For an example of what happens when we get this wrong, see bears and tigers, and the many many many people who've died trying to keep them as pets.
Past that bare minimum, you have logistical concerns around selecting (or limiting) the animals breeding partners, keeping it protected, fed, and if applicable housed, with at least semi-frequent human contact.
After that your only limiter is persistence - it of course, helps to train a herd animal to be a herd animal that doesn't bolt when humans come around instead of training it to hunt waterfowl, but in principle this is achievable, if you have generations worth of human effort to throw at the problem.
Generally, you don't have generations of human effort to throw at the problem, so you have to start small, with behavioral modifications that aren't too severe, because if you don't elicit some kind of useful result inside of one human lifespan chances are overwhelmingly high that the effort runs out "political goodwill" and folks consider the animal to be 'untameable.'
Creatures are always changing. You are different from your parents in more ways than the ones dictated by your genome. They are different from their parents and so on and so forth. Though many traits (the majority, really) are inherited from ancestors, a few things change every generation.
If conditions on the outside world (called selective pressures) remain the same, these little changes get evened out and the genetic makeup in the overall population remains the same. If, however, selective pressures change from one generation to the next, then the survival advantage given to new mutations can become disproportionately large and the bearers of these mutations will reproduce more quickly than normal folk.
Since the amount of individuals that a particular habitat can sustain is limited (think of these as each generation having a fixed number of spots to fill), individuals who produce a larger number of more survivable offspring will quickly monopolize the spots for the next generation and their particular mutations will spread until it becomes the norm.
If one population of individuals gets split into (say) two for whatever reason and each of the new populations get subjected to different selective pressures then, over time, these changes accumulate and they each become distinct species of life forms.
So to recap. A new species is born from an old one when a population gets split into two and, at least, one of them is subjected to selective pressures which are different from those of the original species. So here is a list:
Now let us go over the list.
The geographical separation is merely the most common way through which genealogical separation is achieved in nature. Animals are often very eager to breed and it is not possible to keep a population of critters in close contact with one another without placing them in the pool of potential partners (or rivals) of one another. Plants are even less capable of keeping to themselves, since a lot of them just dump a load of spores in the air or get their polem attached to insects and what not.
Whatever the mechanism, however, the separation that is important for the speciation process is actually genealogical separation. Two populations which must not be allowed to reproduce with one another. That is, in fact, what characterizes a population, a set of individuals that are able (both physiologically and geographically) to breed with one another.
Theoretically, random mutations can accumulate enough to generate different species even when there are no distinction in the selective pressures experienced by the two populations. Think, for instance, that 1% of the population receives a mutation that makes them 1% better at reproducing while 1% of the population receives a mutation that makes them 1% worse at reproducing. If this goes on, slowly, 1% at a time, the good mutations will spread through the population, gently but surely changing is genes as time goes on. This is a process known as genetic drift.
However, in practice, species as we know them are already pretty well adapted to their habitats. Random mutations that are beneficial to an individual will affect way less than 1% of the population. And, most often, will not confer an advantage that substantially changes the chances they will survive to reproductive age and actually breed. Mutations by themselves are somewhat rare and the majority of them are either damaging or entirely insubstantial to the survivability of their bearers.
The selective pressures eliminate the damaging mutations and allow only the rare beneficial mutations to endure and perpetuate, which is why they are so fundamental to the process of speciation.
The selective pressures must be different, but they can't be too different. If a population of carnivores gets trapped in a island inhabited only by plants, there will be no time for them to adapt. This single generation will perish entirely.
The pressures must be so that they allow for individuals to survive long enough to reproduce. And they must have some accessible source of dietary calories and nutrients, however hard to get, for the population to feed off, which takes us to the next point.
The process of speciation does not happen on a individual lifetime. The genetic makeup of an individual is unchanging, only through mutations and reproduction can the genome of a bloodline and the genetic pool of a population be edited without resorting to direct genetic tinkering.
In order for a new species to emerge from a population, the selective pressures must be such that they allow the population to survive (even if harshly) and reproduce (even if slowly). The harshest the selective pressures, the less likely it is for the population to endure long enough for the adaptions to start piling up in the genome of its members.
So now that we understand what speciation is, we turn to domestication.
Domestication is the process of obtaining a new species of creature that is more amenable to human control and interests than an original (dubbed "wild") species. It is the usual process of speciation where one of the selective pressures is human intent.
What animals can and cannot be domesticated is largely dictated by how well humans can artificially enforce the above constraints of speciation to a population of such animals.
In this video, CPG Grey listed four characteristics that species of animals that have been domesticated by humans across the millennia all have in common. The list is short and self-explanatory and maps well into the ongoing discussion about speciation so I will use it here.
According to Grey, for a species of animal to be a good candidate for domestication by a neolithic tribe of humans, it must be:
Let's go through each of them individually and see how they relate to the process of speciation.
In order to domesticate an animal you will need the resources to keep an entire population of them penned up and alive. Preferably not just alive but healthy and thriving. Since Grey was considering species that the early humans might have domesticated, Carnivores were ruled out. Meat is an expensive resource to produce when you are at the bottom of the tech tree and the meat of a carnivore has the same nutritional value as the meat used to feed it. So why not cut the middle man and just domesticate the meat that feeds the beast?
Hence why most, if not all, of our domesticated animals are herbivores that eat stuff that grows everywhere (like grass and roots) or omnivores that are not picky. These animals could be easily fed and nurtured by early humans without any losses to the human population's own dietary needs.
Even if you could feed them, carnivores are a bad idea anyway. The dangers involved in the continuous exposition to the striking range of carnivores would likely have eliminated the pool of able caretakers faster than it could be replenished. Early humans did not fight infections very well.
So the next item on the list of potential domesticated species are animals that, even in their wild form, can be approached and penned in order to be bred. Bears and Tigers are solitary predators that routinely murder infants of their own species out of territorial instincts. So they are out of question. Hippos are murder machines that routinely comes out in news reports of some african nations for killing people that got too close. Zebras bite and kick until they die of exhaustion, Gazelles are impossible to catch and Buffalos are fuckin' tanks.
As you can see, this Friendly thing really puts a limit on what species a neolithic buddy with some spears and rocks can hopefully approach and pen. Without this prerequisite, early humans could not have realistically contained a population of these creatures in order to genealogically separate them from their wild counterparts, a requirement for domestication.
Domestication is a process that happens across generations. It is handy for humans that the life cycle of the creature they are trying to domesticate by shorter than their own. The would mean that tangible progress can be made in the lifetime of a single dedicated farmer or zoo-technician. By contrast, domesticating animals with a long lifespan requires the careful keeping of records and data across multiple generations of humans and very little to show for at each generation.
Not only that but if a species is notoriously difficult to breed, then it might not do it while in captivity, like some modern species are known for (like cheetahs and pandas).
So a good candidate for domestication must be eager to reproduce, even in captivity.
This requirement is for the ease of domesticating herds of animals. Remember, we need to contain and entire population of individuals. Populations have to be large in order to prevent the spread of genetic diseases and malformations. And, ordinarily, there will be much more domesticated animals than human handlers at given time (think shepherds and sheep) so a handful of humans must be able to control a herd of animals.
This is achieved by exploiting the internal hierarchy of some animals. Creatures with a social hierarchy to their herd usually follow a leader or alpha. Herds of cows and horses have a structure and human can exploit that. By taming and controlling the leading animals, human handlers can control the entire pack. Many of our barnyard animals have this ingrained sense of social strata and are thus easily managed. That must have come in handy back in the neolithic period when tribes were small and isolated.
So once you have successfully captured a population of animals with these particular traits, you just have to keep selecting those animals that have most of the trait you want (fatter, tamer, fluffier, smaller... whatever) and only allow those selected few to reproduce. This is how we simulate the selective pressures that sculpt species in nature. Life cycle after life cycle, generation after generation, and the population will drift further and further from its original species as its gene pool changes. Eventually a new species will be born and the individuals of your domesticated population will become unable to breed with the individuals in the wild, which now belong to a different species. When that happens, the genealogical separation will become absolute and the caretakers will have total control over the gene pool of the captive population.
Now, though Grey's list is useful, it was constructed with neolithic caretakers in mind. Nowadays, technological advancements have allowed processes of domestication to be initialized with animals that have previously been well beyond our capabilities, such as the aforementioned buffalos.
Humans have also been able to expand the roster of creatures that they can coerce into breeding in captivity. So we can see right here that the ability for domestication is not just a trait of the creature to be domesticated, but also of the technological limitations of the caretakers.
Domestication is the sculpting of a bloodline. It is a technique that takes effect in the population level. A new species is born. Taming is different. To tame an animal is to capture it and then train it to be at ease in the presence of humans and even to take commands.
Tamed animals happen in a lot of cultures and they can, and are, placed to good work. However, a tamed animal is not a domesticated animal. Its instincts are still wild, its preferences are still the same as those animals in the wild.
Though boars and wolves have been domesticated into pigs and dogs, elephants and killer whales have not. There are no "domestic elephant" species in the same way that we have domestic wolves (Canis familiaris) as opposed to wild wolves (Canis lupus).
Incidentally, humans have also domesticated themselves. The modern Homo sapiens (the only surviving subspecies of the species that originated around 100 millennia ago) has some abilities that its original "wild" version did not have. For instance, the capacity to digest milk even as an adult, theorized to have been originated due to selective pressures of feeding off the milk produced by domesticated cows and goats.
