The moral theory usually applied to animal ethics is utilitarianism. We are to be concerned exclusively with the suffering and happiness of animals: we must minimize animal disutility and maximize animal utility. That, and only that, is what animals have a right to expect (transposing utilitarianism into a rights-based theory). This seems eminently reasonable: how could a deontological theory apply to animals? How could it be wrong to lie to animals or break promises made to them or be ungrateful to them? They don’t have language or freely perform acts of generosity towards us. We don’t have a duty to carry out their last will and testament, simply because they don’t make wills. The only duties that bind us in relation to animals are related to suffering and (possibly) happiness—the kind that reflect their nature as sentient beings. We don’t have any duty to educate them in world history or respect them as persons or listen to their defenses of their actions or grant them due process in a court of law, since these all presuppose attributes that they don’t possess (even the most intelligent ones). So animals only partially fall under morality as we apply it to humans; they don’t have the kind of across-the-board moral standing we typically accord to each other. They are not full moral beings like our neighbors and friends or people from other lands. They exist a step down the moral ladder, not quite as deserving as fellow humans. True, they fully deserve not to be made to suffer unnecessarily, just like people, but they are not de jure recipients of the full range of duties recognized under deontological ethics. And this shapes our general moral attitude towards them.
But this picture, accepted as self-evident, seems to me fundamentally wrong. Let’s start with lying and promise-breaking (a kind of lying): true, one cannot tell a lie to a dog or break a promise, since dogs don’t understand the relevant speech acts, but performing a false speech act is not what the wrongness of lying and promise-breaking consists in. It isn’t a matter of falsehood as such, or else actors would be liars, but rather of deceiving someone by making false statements. It is the deceiving and misleading that is wrong, not the fact that it is done in language: creating false beliefs and raising dashed expectations is what is wrong. But it is possible to do that non-linguistically: you can act in such a way as intentionally to generate the belief in a dog that it is about to be taken for a walk, but then decline to take it. You can pretend to put food in a bowl for the dog to eat and then hand it an empty bowl. You can regularly give the dog a treat at a certain time each day so as to create an expectation, and then intentionally omit to do so on a particular day, thus disappointing its expectations. These are all examples of deceiving and misleading an animal, even though the means adopted is not a speech act in a shared language. And don’t you normally feel a duty to take the dog for a walk if you have created the expectation that a walk is in the offing? Isn’t it incumbent on you to feed your cat in the morning given that its past experience has led it to expect to be fed then? So the duties that come under the heading of truth-telling and promise-keeping in deontological ethics are applicable to animals, though they are not linguistically mediated. In fact, animals belong with pre-linguistic human infants in this respect: in both cases deception is possible without the mediation of language, and it is as wrong as linguistically mediated deception.
Here we can make a general point about morality, namely that we are apt to be too fixated on language in applying moral principles. We have become familiar with such concepts as racism, sexism, and speciesism as markers of discriminatory types of moral stance; but we can add to this list speech-ism as another type of discrimination, i.e. taking the possession of language as a criterion of moral standing. In fact, language has nothing essentially to do with morality—it is not a morally relevant characteristic. Being capable of being deceived is a morally relevant characteristic, but it is possible to have that and not understand language. If we all lost our linguistic ability tomorrow, that would not affect our moral standing, so long as the rest of our psychology remained intact. It is a failure of imagination to suppose that animals that don’t speak like us don’t have the moral standing that we enjoy; they simply lack a contingent human attribute that evolved some 200,000 years ago while still possessing a psychology much like ours, cognitively and affectively. Language possession is like gender or race or species—not a litmus test for equal moral consideration. Animals can suffer and therefore should not be made to suffer for no good reason, but animals can also be deceived and therefore should not be deceived for no good reason. Suppose you were to poison your dog by giving it tainted food: there would be the wrongful act of causing the dog to suffer (and maybe die) but also the wrongful act of misleading the dog about the food it has been given to eat. It doesn’t matter whether the misleading is done through language or in some other way; it is wrong in either case. Allowing for a bit of poetic license, we can say simply and succinctly that it is wrong to lie to animals and to break promises made to them, i.e. to mislead them intentionally. What makes it wrong to mislead someone linguistically is the same as what makes it wrong to mislead someone in other ways; language is incidental to the wrongness of the action. So we should eschew discriminatory speech-ism (or linguistic-ism): grammar is not the measure of morality any more than gender or race or species is.
I mentioned ingratitude earlier: one of the standard duties listed by deontological ethics is gratitude for good actions performed. It might be thought that this cannot apply to our attitudes to animals for the simple reason that they are not themselves moral agents. I will put aside the question of whether animals are (or can be) moral agents; there is still a question about whether it is appropriate to feel gratitude to an animal. And here I think it is evident that gratitude is an appropriate attitude in our dealings with animals: we can be grateful for their love and also grateful for their existence. We love our pets and our pets love us: this enriches our lives and we feel happier for it. We are, and should be, grateful for this love, even if the animal is not acting from moral motives—as we can feel grateful for the love of our family and friends. Not all actions that occasion gratitude are moral actions, and it would be strange not to feel gratitude for the love of others, animal or human. Also, I think it is appropriate to feel gratitude for the existence of animals—their beauty, their fascination, their affinity to us as well as their difference. Life is better because animals exist as objects of contemplation, aesthetic and scientific. They are a marvel to behold. We are rightly glad to share the planet with them (many of them anyway). Someone who doesn’t feel this is defective in some way—just like someone who fails to appreciate when they have been treated kindly. We call them hard-hearted or blind.
It is difficult to think of a standard moral duty that has no analogue in the case of animals. Beneficence and non-malificence (W.D. Ross’s terms) obviously apply, but it turns out that truth-telling, promise-keeping, and gratitude also apply, suitably extended. What about the duty of making reparations for past wrongs? Even here we can think of cases in which such a duty would be applicable—as with stealing the land of an animal population or removing its source of food. Suppose you decided in your selfishness to buy something fancy to eat instead of buying your dog food and left it to go hungry one day: wouldn’t you owe it to the dog to make up for that the next day? You should give it extra food and pet it penitently or some such thing. You need to compensate for the deprivation you visited upon your dog earlier. Isn’t that what justice requires? The dog may remember your past actions and your present actions will go some way towards making up for what you did earlier; it may think better of you for making amends. This would be even clearer for animals closer to us zoologically such as chimpanzees, though we typically don’t have them for pets. If I accidentally trip over my cat, I feel an obligation to pet him gently and perhaps given him a treat—aren’t I trying to make reparations? The duties we feel towards our intimates do not magically disappear when the species changes. No doubt if Neanderthals still existed and mingled among us we would extend deontological ethics to them too, recognizing the similarities behind the differences; the case of other existing species is not essentially different. Of course, there are animals that really do fall outside the moral circle defined from a deontological point of view: some animals cannot constitutionally be deceived and can feel no love for us humans—starfish, bacteria, flies. These animals cannot be made to have false beliefs or disappointed expectations because of their restricted psychology, so they impose no corresponding duties on us; but many animals clearly do impose such duties. Acknowledging this removes the last remaining gap that separates us morally from other species: now we can see that animals deserve more than what utilitarian ethics can deliver.They are fully within the moral circle defined by deontological ethics. Our duties towards them include more than merely reducing suffering and providing happiness.
 We could assign them to the class of implicit truth-telling and promise-keeping: intentionally creating the impression that certain beliefs and expectations are warranted without explicitly saying so, which we do all the time.
 I am of course assuming that utilitarian ethics is incomplete; there are duties that cannot be subsumed under the utilitarian framework. My point is that if you are a committed deontologist you can still include animals within your official moral theory. Put differently, deontology is not limited in scope to humans, ceding territory to utilitarianism in the case of animals.