What are marginal animal cases?
I shall use the term "marginal animal cases" to refer to animals whose immaturity, genetic damage or physical injury precludes them (at least temporarily) from exercising agency or being sentient - e.g. animal embryos, anencephalic animals and decerebrated and permanently comatose animals. I defend the view that these animals have the same moral significance as as mature, sentient animals - not on account of their potential or historical properties such as what they will be (animal embryos), could have been (anencephalic animals), or once were (decerebrated and permanently comatose animals), but simply because they share the same nature as their sentient counterparts.
The sentientist view of marginal animal cases
There are a variety of obligations owed to animals: (i) our prima facie obligations not to kill or injure animals, which are stronger than our comparable obligations to other organisms; (ii) our prima facie obligation to refrain from thwarting animals' telos-promoting pursuits and from causing them pain; and (iii) our unconditional obligation never to inflict cruelty on animals as an end in itself.
For each of our special obligations to animals we can ask: which animals are they owed to? In particular, are they owed to sentient animals or to any animals having the same biological nature as these sentient animals (i.e. the non-sentient con-specifics of sentient beings)? Putting it another way: do sentient animals have a moral significance that their non-sentient con-specifics lack?
If we look at duties, the obligation not to inflict cruelty on animals for its own sake is the least controversial: by definition, it can only be owed to sentient animals. On the other hand, non-sentient animals can be killed or injured, and animals who are not yet sentient (e.g. animal embryos) can have their natural pursuits thwarted, if we prevent them from ever being able to engage in these pursuits (e.g. by decerebrating them). Is the killing, injury or mutilation of sentient animals, more morally reprehensible than inflicting these harms on their non-sentient counterparts? For instance, are there some injuries that we may inflict on rat embryos or decerebrated rats, that we may not inflict on sentient rats?
Traditionally, the debate over the scope of our moral obligations has been couched in terms of actual versus potential properties. The "sentientist" school of thought defines our obligations to animals in terms of their current, actual properties (in particular, their mental states), while the "essentialist" or "naturalist" school takes account of their potential properties as well. I shall argue that an animal's potential properties matter because (and insofar as) they are grounded in the telos of the animal, and I shall also defend the metaphysical claim that an animal (or any other organism) has a nature that defines its telos.
On the "sentientist" side, utilitarians such as Peter Singer (1977) include only animals who are sentient within the scope of their ethical concerns; animal embryos, decerebrated animals and permanently comatose individuals do not count. For instance, Rollin (1992) has proposed that we should render food animals and experimental animals decerebrate (and thus incapable of pain) through genetic engineering. Decerebrated animals, according to Rollin, cannot be said to have a welfare of their own.
If sentientists were consistent, they would exclude not only potentially conscious animal embryos, anencephalic animals that lack the neural wherewithal for mental states, and brain-damaged permanently comatose animals, but also animals in deep sleep or hibernation, from the domain of sentient beings. None of these creatures can be said to "mind" what happens to them, as they are currently incapable of having the mental states with which to do so. Does this mean we can treat them like other organisms such as plants? It seems counter-intuitive to say that an animal's moral status diminishes when it goes into hibernation, but sentientists cannot evade this strange conclusion without compromising their philosophical stance and/or according significance to other non-sentient animals.
First, a sentientist could argue that hibernating animals matter just as much as conscious animals because of their future mental states (which they will be deprived of, if they are killed), but the same argument would also apply to animal embryos. Second, a sentientist could deem hibernating animals to be morally significant because of their past mental states, but that would also include permanently comatose animals. In any case, both options are philosophically unappealing to sentientists, because they invoke properties which are not current. Third, a sentientist could accord a privileged status to hibernating animals because they are capable of being awakened, but this would re-introduce the very "potentialistic" language that sentientists eschew. Finally, a sentientist might say that sentient beings (including hibernating animals) are morally significant because they currently possess a (minimally) functioning brain, but then she cannot explain why the possession of such a brain should be a morally relevant property, even when the brain is not currently able to have mental states, without re-introducing potential properties (e.g. "having a brain matters because a brain is capable of having thoughts").
The exclusion of non-sentient beings from our ethically privileged "inner circle" would also rule out so-called marginal human cases: embryos, patients in a permanent vegetative state, and human beings who lack mental states altogether, because of severe disabilities.
On the other hand, it seems absurd to include not only all embryos that are capable of developing into sentient animals, but also sperm cells and ova, within the ambit of our special obligations to animals, as we would have to do if we respected them on the basis of their potential properties. Even the nucleus of a body cell taken from an animal could be regarded as a potentially sentient being, as it can develop into an animal after it is transplanted into a denucleated ovum and cell division is electrically induced, as part of the cloning process.
Marginal animal cases - why they matter
I maintain that an individual's telos determines its moral status. My argument can be summarised thus:
(1) "Right moral conduct" towards an individual could be defined as respecting the individual's moral status.
(2) The moral status of an individual is determined by some or all of its interests.
(3) Only those interests that are grounded in an individual's telos are morally relevant. (Argued for at the end of chapter 1.)
(4) The moral status of an individual is therefore determined by its telos, which defines those capacities whose realisation is or would be (objectively) good for it.
(5) Hence "right moral conduct" towards an individual can be defined as respecting its telos.
I would also argue that "marginal cases" have the same telos as their "normal" counterparts. Briefly:
(1)(a) An individual's telos can be defined in terms of the various kinds of (biological and non-biological) ends whose realisation is or would be (objectively) good for it.
(b) The telos of an individual animal consists of the entire suite of basic animal goods whose realisation would be good for it.
(2) Only those goods whose realisation is compatible with the individual retaining its identity can be said to be good for it. An end whose realisation cannot be achieved without the individual losing its identity is not part of its telos, as this supposes something absurd: that the individual is not itself.
(3) Individual identity is species-specific. (Strictly speaking, the counterfactual antecedent "If I were a butterfly" makes no sense, because then I would not be "me".)
(4) Only those goods that characterise the species to which an individual belongs can be said to be good for it. (N.B. No individual member of a species can realise all of these goods, owing to the diversity of sexes and phenotypes, which realise the telos of an animal in mutually exclusive ways and instantiate different goods. See chapter 1, section B.7.)
(5) The possibility of an individual animal's realising its telos is grounded in the genetic information which it must possess in order to exercise the basic animal goods that characterise the species to which it belongs.
(6) A "normal" (i.e. fully functional) animal possesses a complete set of intact information which allows it to realise the entire suite of basic animal goods for its species, which are compatible with its sex and phenotype. (The last qualification is important: no individual can simultaneously realise the teloi of both sexes, for instance. Thus any set of basic animal goods realised by an individual will always be a subset of those available for the species.)
(7) The telos of a normal animal consists of all of the basic animal goods that characterise the species to which it belongs, which are available to it, given its sex and phenotype. The counterfactual: "If this individual realised any of the basic animal goods that characterise its species, and are compatible with its sex and phenotype, that would be good for it", is true in a straightforward sense, as no loss of the individual's identity is supposed in its realising any of these goods.
(8) Many "marginal animal cases" (e.g. embryonic and permanently comatose animals) also possess a complete set of intact genetic information which allows them to realise the entire suite of their basic animal goods available to them, given their species, sex and phenotype. They thus have the same telos as normal animals of their species of the same sex and phenotype. We can suppose them to realise any of the goods whose realisation is permitted by their genes.
(9) Individual identity is not DNA-specific: the DNA in our own cells is being continually altered by cell damage, without compromising our individual identity. The counterfactual antecedent "If my DNA were different" or "If my DNA were repaired" therefore makes sense, subject to the limits of (3).
(10) Even for "marginal animal cases" such as anencephalic animals, which possess a genetically damaged copy of the information that grounds the telos of their species, we can still imagine the damage being repaired without loss of the individual's identity. ("Repaired" here means altered in such a way as to permit the full range of functionality possessed by a normal member of the species.) Thus we can, without absurdity, combine the counterfactual: "If this individual's DNA were repaired, that would be good for it", with: "If this individual were to realise any of the goods permitted by its repaired DNA, that would be good for it", to obtain the conclusion: "If this individual were to realise any of the goods that characterise a normal member of its species, of the same sex and phenotype, that would be good for it".
(11) The telos of an individual animal, whether it be "normal" or "marginal", can thus be defined as the range of basic animal goods that typify a member of its species, of the same sex and phenotype.
Since an individual's telos determines its moral status, and "marginal animal cases" have the same telos as their normal counterparts, it follows that "marginal animal cases" have the same moral status as normal animals. "Marginal" animals and their "normal" counterparts have the same moral significance, because the same things are good for them. What they should be is the same, because they share the same nature, which grounds their telos.
The teleological properties of an organism can be deduced from their (actual, not potential) informational properties - that is, the internal program (in its DNA) that contains instructions for its development. Ultimately, it is informational properties - more precisely, its internal program - that constitute the nature of an organism.
It should be stressed that the internal program in an organism's DNA is not dormant but actually executing: it controls its development and supports its bodily structure and functions, during every moment of its waking or sleeping life. This is true for so-called "marginal cases" too. Finally, even in individual cases where the program instructions coding for biological functions have been corrupted by genetic copying mistakes (mutations), the actual function that the corrupted DNA is supposed to code for can be ascertained by comparing it to that of functioning conspecific individuals.
Does the concept of nature make sense, and is it morally relevant?
Talk of "nature" is something of a taboo for present-day evolutionary biologists. I believe this intellectual revulsion stems from a confusion between two concepts of "nature": (i) an unchanging essence, and (ii) the internal unifying principle of an organism, which determines its structure and bodily functions. The former concept is no longer tenable: living things are continually evolving, and even during an individual's lifetime, its DNA undergoes changes. It is the latter concept of nature which I will be defending here.
Despite the fact that lineages of organisms are in continual flux, biologists can use Darwinian thinking to straightforwardly identify the biological functions of any species of organism: look for organs or subsystems which the organism has because they can do something that confers a selective advantage on their possessors. (This is the etiological criterion which we discussed in chapter 1.) Biological functions remain fairly stable over long periods of time: eyes did not evolve overnight, but in a series of steps, over millions of years. We can, then, define an organism's nature, either in terms of its characteristic structure and functions (a layperson's descriptive definition), or in terms of its informational content (an essential definition): the DNA which encodes these functions (unless it has been damaged). In practice, evolution does not undermine "naturalism", because organisms evolve over very long periods of time.
It is thus incorrect to assert, as Clark (1988) does, that the unity of a species consists in its being a common breeding population - a property Clark (rightly)does not consider to be morally significant per se. Rather, the unity of a species consists in its members' shared possession of information that defines the range of basic goods (including biological functions) typical of that species. In certain individuals, the information in a gene may be corrupted, but the damage is identifiable by comparing the gene to that of healthy members of the same species, and it is also capable (at least in principle) of being repaired.
The "marginal animal cases" we have considered possess a complete, actually executing copy of their internal program in every cell of their bodies. In embryonic animals, the only obstacle to the full expression of the internal program is time; in anencephalic animals, DNA damage that has corrupted a few of the program instructions; and in decerebrated animals and animals in a permanently vegetative state, bodily injury. In all these cases, the instructions that code for the possibility of animal agency are already there within the individual's body cells. I contend that since these instructions are what gives animals their specific telos, which determines an animal's moral status, any individual possessing these instructions is entitled to the same respect as a normal animal, regardless of immaturity or impairment. (For a discussion of individuality with reference to early embryos, see Oderberg, 1997.)
The following "thought experiment" (intended for illustrative purposes only) may serve to clarify why it makes sense to regard animals that lack mental states due to genetic damage or bodily injury as the moral equals of "normal" animals. Future medical advances might allow doctors to restore some of these impaired individuals' capacity for animal agency, without altering their individual identity or changing their nature as members of their species. Doctors may one day be able to repair any damage to these animals' DNA, and adult stem cells could even be used to enable them to grow a complete brain, if they have nothing more than a brain stem. (As we saw above, the objection that an individual whose DNA was repaired would no longer be the same individual is groundless.)
What does the foregoing argument imply for the scope of our duties to animals? It implies that "marginal cases" possess the same moral significance as their sentient counterparts, with whom they share a common nature. However, this does not entail that we have the same duties to both (see below).
On the other hand, there is no need to accord any moral significance to sperm cells or ova, or even a cell within an animal's body (which may be cloned). All of these are "potential animals" in some sense, but the information they possess is either too incomplete or inoperative for them to matter as animals do. Animal sperm and ova lack half the information that defines an organism as having the nature of an animal; animal body cells possess the relevant information for making an animal, but it has not been "switched on", or activated, in a suitable environment. Of course, once it is switched on (e.g. in cloning, when the nucleus of an animal's body cell is placed within a denucleated ovum and induced to divide by an electric shock), it acquires the moral status and nature of an animal (Oderberg, 1997).
Is my proposal unacceptably counter-intuitive?
The proposal that sentient animals and their non-sentient con-specifics have a special moral standing simply because they have a certain kind of information in their DNA (i.e. any kind that codes for the possibility of intentional agency), which is currently "switched on", may seem absurd to some people. To make it seem less so, I would like to make three points in reply.
First, I am not arguing that the information contained in an animal's DNA confers moral standing on it per se, but that the range of biological and mental functions coded for in its DNA, the exercise of which comprises the animal's telos, gives the animal moral standing. The logic here is that the actual information contained in an internal program (which is currently executing) describes the animal's telos (which includes the satisfaction of some of its wants). The animal's moral standing is based on its telos.
Second, it has already been argued that other organisms have moral significance, simply because they possess a telos. Why should animals be any different?
Third, the only consistent alternative to a telos-based ethic is to base our obligations to animals on their mental states. I contend that these are too fleeting and evanescent to ground our common-sense intuition that the unjustified killing of an animal while it is asleep or hibernating is just as wrong as performing the same act of killing (without causing pain) while it is awake.
My proposal that all animals sharing the same nature are morally equivalent does not imply that the same actions are morally wrong towards them, or that the same actions are wrong for exactly the same reasons. For instance, if a scientist were performing an experiment that caused pain but no injury to mature sentient animals, then the use of animal embryos in their place would clearly be acceptable, if they were not harmed. On the other hand, if the experiment were fatal, then it would be at least a prima facie wrong in both cases, because it would harm the animals. However, the harm done in each case would not be the same: the experiment would rob the embryo of more years of life than the sentient animal, but it might also cause distress and frustration to the sentient animal. I shall have more to say about animal experimentation in chapter 6.
A test case: Do decerebrated animals matter?
Rollin (1992), who has written extensively on animal suffering (1990), has proposed that we should render food animals and experimental animals "decerebrate" (incapable of conscious experience, and hence incapable of pain), either surgically or through genetic engineering. This is an interesting ethical test case. Classical utilitarians would say that no harm is done as no suffering is caused. Masson and McCarthy might argue that decerebrate animals, while not undergoing pain, forego the funktionslust they would normally enjoy from the normal exercise of their faculties.
Rollin has a reply to this argument, however. Although he believes that every animal should have its telos respected, he argues that decerebration changes an animal's nature, instead of violating it. No funktionslust is foregone, because the animal is now a different kind of being, with a different (non-conscious) kind of functioning.
I contend that Rollin has misconstrued the concept of nature. I have argued that an animal's nature, which defines its telos, is constituted by the information in its DNA that defines its development, internal structure and range of functions. Genetic engineering basically amounts to changing a few of the instructions. Now, it is one thing to change the instructions describing an organ's function, so that one kind of function they express in the animal (e.g. flying in most birds) is replaced by another (e.g. swimming in a penguin). It is another thing to simply impair one of an animal's functions, without replacing it with anything else, e.g. by genetically engineering an animal to make it incapable of learning. This is not a change of an animal's nature but a deliberate corruption of the information that expresses its existing nature - a corruption which stunts the animal forever. Such an action can hardly be described as consistent with respect for an animal's telos.
It is important to realise that even here, the animal retains its nature (and its moral status), as the altered genes which fail to code for its brain development, normally do in other animals of the same biological species.