Here are some thoughts on the ethics of animal treatment.
[collapse=Framing the Issue]
First, I think the thread question is curiously framed. Why not ask when is it
wrong to kill animals? Or when it is
expedient? Or when it is
most likely to maximize personal amusement?
The more productive question, I'd say, is whether non-human animals
are deserving of moral consideration -- which would answer, in turn, how we should (or should not) approach the wellbeing of other species. With this in mind, there are already several things we can infer.
Morality concerns
choices -- what we should or should not do, for Situation X. And choices are perpetuated by
agents -- entities that are self-aware, and are capable of reflection and introspection. If you can't reflect and consider a situation and its possible outcomes, then you can't deliberately "choose" one outcome over another.
Humans are capable of these processes, and can thus be considered agents. So we can say, at the very least, that humans are both capable of moral action, and capable of possessing moral consideration. We can shorthand this by saying that a human is a
person (a close synonym with agent, in this context).
[/collapse]
[collapse=On Personhood and Morality]
Are non-human animals agents? Are they people, capable of considering the variables of a scenario and making choices accordingly? Well, this clearly depends on the animal species at hand. Let's take the most intelligent non-human species we know of -- species like cetaceans, great apes, corvids, pachyderms, etc. If we were to find that any or all of these species were indeed self-aware, and thus constitute as agents, then would they be worthy of moral consideration?
I would say that yes, they would be. If, for instance, a Bottlenose Dolphin was an agent (and thus, a person), then they would both be capable of moral action and reasoning, and also be capable of considering we humans under a moral lens (just as we could do the same for them). If morality concerns people, and humans and dolphins classify as people, then morality would concern both, so we'd all need to act accordingly.
Since we're a pro-social species by our nature, we (tend to) value pro-social acts and behaviours, and (tend do) vilify to anti-social acts and behaviours. So we can consider "right" and "wrong" actions and behaviours by the consequences they entail.
The killing of another person (human or otherwise) will tend to be in the wrong, under this social-metric of consequential ethics. All human action is driven by desire. We tend to want what we want, and tend to dislike that which impedes the fulfillment of those desires. And if most people share common desires, then the likelihood of maximizing desire-fulfillment will be greater (especially in the long run) if we all cooperate, helping each other satisfy what we want, rather than everyone going at it alone.
If most people want to survive as long as possible, then we can say that such things as
murder is wrong -- and therefore, that one shouldn't murder non-human persons. If most people don't want to be harmed or injured, then we can say that self-defense is
right, but that initiating aggression or force is
wrong. So don't go punching elephants in their peanuts.
As for killing for sustenance or resources, then we can say that it's right to do so --
only with the consent of the non-human person (i.e. a dolphin consents to have itself killed and processed for food). If you try to kill a person, human or otherwise, for food without their express consent, then it's a violation of their autonomy, and would thus be a morally reprehensible act.
So long as you hold to this metric of pro-social, desire-driven consequential ethics, then you can easily derive edicts to guide how to treat with non-human persons.
[/collapse]
[collapse=On the Treatment of Non-Persons]
What about non-human species that don't possess agency, or personhood? Who cannot reflect on their own choices, acting and reacting purely by dint of instinct? Who cannot engage in moral reflection and reasoning, and thus cannot be subject to moral consideration, both in themselves and for their actions and behaviours?
So would it be right or wrong, for instance, to curb-stomp a chipmunk? By all observations, chipmunks don't appear to possess self-awareness. So it's improbable that they could be classified as agents.
But if that's the case, it would neither be right or wrong to kill a chipmunk -- because they can't possess moral consideration. Morality doesn't apply to non-persons.
So it's not about whether it's right or wrong to kill a chipmunk. You can, if it is your desire. Or you can refrain from doing so, if it is your desire. You can domesticate, rear, and slaughter livestock for nourishment, if you want. Or you can not do that, if you are so inclined.
However.
I would say that it might be better to treat non-persons pro-socially, when you can. So you're going to be kind to a fellow human, but be cruel to a dog? Why would you do this? This kind of
selective empathy seems dangerous, since being selective in your empathy is a habit that could end up bleeding into how you treat actual persons. And being selective in your empathy toward humans is both inconsistent, and something to be discouraged (under the pro-social edicts established thus far).
Ants aren't agents. But in order to reinforce and exercise your sense of empathy, best not to crush them indiscriminately. If you have an ant infestation in your house, and you value not having an infested living space, then you might consider killing them. Killing or not killing ants (or insects generally) is neither right nor wrong; it's about whether you value your desires over the livelihood of these non-agents (insects, plants, reptiles, birds, whatever else).
Plus, mass mistreatment of animals (directly or indirectly) can lead to such things as dwindling of populations, which can affect ecosystems, which can come back to bite us in the human ass. So globally, treating populations in the environment with undue cruelty is a riskier gambit than leaving them be, or encouraging their preservation.
So basically, while non-persons aren't capable of being deserving of moral consideration, is there any reason to treat them poorly (via their killing or otherwise)? I can't think of any.
I will make an aside concerning animal meat, livestock, etc. The mass slaughter of domesticated livestock may be an inconvenience to these animals, but we value their meat more than their life (which is neither right nor wrong). However, I would say that it might be in our best interest to both consider how much meat we might actually need, and to ensure that livestock living conditions are humane (an especially salient point, since poor conditions can lead to things like contaminated meat, wasteful and inefficient use of resources, etc.).
[/collapse]
[collapse=On Aliens]
Since I'm here, I'll throw in musings on how morality might apply to extra-terrestrial persons.
Our morality is a social one (along a pro/anti continuum). The most intelligent species we know of besides us (e.g. apes, cetaceans, pachyderms, etc.) exhibit stronger social wiring as well. So in the event those species classify as persons, it would be much easier to treat with them under a social-moral paradigm.
What of non-human species, from Earth or elsewhere, whose fundamental wiring is not based in sociality? Such as an alien species that doesn't (and can't) care about individual lives, or doesn't understand the difference between order and chaos or life and death (for themselves and others), or a hive mind where individual personhood is a foreign concept? Or a species whose experience is so strange that we can hardly fathom it?
Our pro-social morality is an emergent one -- emergent from our ingrained behaviours, senses, experiences, and instincts. As per the above, however, it is possible that other alien species could exhibit emergent moralities that don't align along a social, mammalian continuum.
So how to treat with such beings will depending on multiple factors, such as whether mutual intelligibility is possible, whether the species in question can be reasoned with, and working out sets of moral edicts for the dealing of distinct species.
But even so, we can still follow our pro-social inclinations in dealing with such persons. Treat them as you would treat a person; and if they violate your autonomy (in what is an anti-social action), then you have cause to discourage or vilify this violation, and react accordingly.
[/collapse]
In summary:
-A pro-social, desire-driven consequential model of ethics is both the most rational and practical model for humans, and is the most apt one for the treatment of non-human animals;
-If the non-human animal is a person, then they are subject to morality. They must thus be treated accordingly. Unless it is with the person's consent, or in self-defense, then the killing of animals should not be pursued;
-If the non-human animal is not a person, then they are not subject to morality. It is therefore neither right nor wrong to treat them one way or another. However, we
might as well treat them as pro-socially as we can, since it would be more beneficial to acquire the habit of consistency in how we apply our empathy, than to be inconsistent in its application.
So there you have it (as finally promised, @
Sucumbio
).
). I'd kill 1000 dogs if it saved one human life. However, I'm liberally a conservationist and want to CONSERVE and treat animals with respect except in the most needed areas (food, legit medical testing, etc). No killing off species for resources. Blablabla. The animal testing probably makes me sound a lot more conservative in this area than I am. 
